THE IMPACT OF HIVAIDS ON UNDER-FIVE MORTALITY IN MALAWI by ive16829

VIEWS: 0 PAGES: 151

									    The Impact of HIV/AIDS on Under-Five

                Mortality in Malawi




                          By:

       Chodziwadziwa Whiteson Kabudula




A thesis submitted in fulllment of the requirements for
 the degree of Magister Scientiae in the Department of
      Statistics, University of the Western Cape.


                      Supervisor:

             Dr. Henry Victor Doctor



                    November 2007
„o mumD un™leD —untD sister —nd ™ousins for their ™ontinued loveD en™our—gementD —nd
                                     supportF




                                         i
Keywords


eshƒ smp—™t wodel


hem€roj


pertility


rs†Geshƒ


w—l—wi


wort—lity


wotherEtoEghild „r—nsmission


€roje™tions


ƒu˜Eƒ—h—r—n efri™—


…nderE(ve mort—lity




                               ii
Abstract


elthough the underE(ve mort—lity r—te in w—l—wi h—s ˜een de™lining sin™e IWTHD it still
rem—ins one of the highest in the worldF sn order to —ppropri—tely t—rget interventions to
—™hieve su˜st—nti—l redu™tions in de—ths —mong ™hildren under the —ge of (ve ye—rs in w—l—wiD
there is —n ongoing need for ˜etter knowledge of the proportion of ™—useEspe™i(™ underE(ve
mort—lity in the ™ountryF ‚esponding to this needD this study h—s estim—ted the proportion
of underE(ve mort—lity dire™tly —ttri˜ut—˜le to rs†Geshƒ in w—l—wi during the period PHHH
to PHHRF „he study used some of the existing indire™t methods —nd m—them—ti™—l modelsF „o
— l—rger extentD the study —dopted the methods —nd —ppro—™hes re™ommended ˜y …xeshƒ
—nd ‡ryD p—rti™ul—rly those implemented in hem€roj —nd eshƒ smp—™t wodel ˜oth of
whi™h —re p—rt of the ƒpe™trum p—™k—geF rs†Geshƒ is estim—ted to h—ve dire™tly ™—used
—˜out IPFT per™ent of underE(ve mort—lity in w—l—wi ˜etween PHHH —nd PHHR —nd to h—ve
signi(™—ntly h—mpered —™hievement of the ™ountry9s over—ll underE(ve mort—lity redu™tion
go—lsF elthough exp—nsion of ™over—ge of tre—tment with xevir—pine @whi™h is the ™urrently
—v—il—˜le prevention of motherEtoE™hild tr—nsmission @€w„g„A servi™e in w—l—wiA would
help redu™e rs†Geshƒ —ttri˜ut—˜le underE(ve mort—lityD the ™on™lusion dr—wn from the
(ndings of this study is th—t in order to m—ke su˜st—nti—l ™ontri˜utions to the ™ountry9s
over—ll underE(ve mort—lity redu™tion go—lsD w—l—wi needs to intensify its e'orts in —ll the
four key ™omponents of — ™omprehensive set of €w„g„ servi™esF purthermoreD w—l—wi
—lso needs to strengthen its e'orts in —ddressing other m—jor ™—uses of underE(ve mort—lity
˜esides putting in pl—™e — ™omprehensive set of €w„g„ servi™es in order to —™hieve further
su˜st—nti—l redu™tions in de—ths —mong ™hildren less th—n (ve ye—rs of —geF



                                             iii
Declaration


s de™l—re th—t The Impact of HIV/AIDS on Under-Five Mortality in Malawi is my own workD
th—t it h—s not ˜een su˜mitted ˜efore for —ny degree or ex—min—tion in —ny other universityD
—nd th—t —ll the sour™es s h—ve used or quoted h—ve ˜een indi™—ted —nd —™knowledged —s
™omplete referen™esF




ghodziw—dziw— ‡hiteson u—˜udul—                                              xovem˜er PHHU




ƒignedX. . . . . . . . . . . . . . . . . . . . . . . . . . .




                                                               iv
Acknowledgments


s would like to th—nk my supervisorD hr renry †i™tor ho™tor for guiding —nd en™our—ging
me throughout this study —nd ™—refully re—ding my work —nd o'ering useful suggestions on
how it ™ould ˜e improvedF


s —lso wish to th—nk wrF qeorge fello from the ipidemiology hep—rtmentD winistry of
re—lth ‘w—l—wi“ for providing me with the —nten—t—l ™lini™ rs† surveill—n™e survey d—t—
th—t w—s used in estim—ting rs† prev—len™e —mong women of reprodu™tive —ge in w—l—wi
from IWWW to PHHSD wrF iv—n™e u—dongol— from the x—tion—l ƒt—tisti™—l y0™e ‘w—l—wi“
for providing me with reports on v—rious demogr—phi™ issues in w—l—wiD —nd wrF ƒimon
w—kom˜e —nd wrF wi™h—el iliy— from the rs† …nitD winistry of re—lth ‘w—l—wi“ for
providing me with inform—tion on the st—tus of —ntiretrovir—l tre—tment —nd prevention of
motherEtoE™hild tr—nsmission progr—mmes in w—l—wiF


ƒpe™i—l th—nks to €rofessor ‚enette flign—ut —nd the hep—rtment of ƒt—tisti™sD …niversity
of the ‡estern g—pe —nd €rofessor pritz r—hne —nd the efri™—n snstitute for w—them—ti™—l
ƒ™ien™es for the (n—n™i—l support tow—rds my studiesF


„he (n—n™i—l —ssist—n™e of the hep—rtment of ƒ™ien™e —nd „e™hnology @hƒ„A ‘ƒouth efri™—“
tow—rds this rese—r™h is here˜y —™knowledgedF ypinions expressed —nd ™on™lusions —rrived
—tD —re those of the —uthor —nd —re not ne™ess—rily to ˜e —ttri˜uted to the hƒ„F


pin—llyD words —lone ™—n not express my gr—titude to the elmighty qod who m—de it possi˜le
for me to ™omplete this study —nd for the in(nite ˜lessingsF

                                             v
Contents


Keywords                                                                                ii

Abstract                                                                               iii

Declaration                                                                            iv

Acknowledgments                                                                         v

List of Tables                                                                         xi

List of Figures                                                                       xvi

Acronyms and abbreviations                                                       xviii

1   Introduction                                                                        1
    IFI   Background to the study F   F F F F F F F F F F F F F F F F F F F F F F F     I


    IFP   Statement of the problem    F F F F F F F F F F F F F F F F F F F F F F F     Q


    IFQ   Aim and objectives of the study       F F F F F F F F F F F F F F F F F F     Q


                                      vi
    IFR   Research question           F F F F F F F F F F F F F F F F F F F F F F F F F F F F F    R


    IFS   Rationale and signicance of this study                       F F F F F F F F F F F F    R


    IFT   Conceptual framework                F F F F F F F F F F F F F F F F F F F F F F F F F    S


    IFU   Research methodology                F F F F F F F F F F F F F F F F F F F F F F F F F    S


    IFV   Denitions of key concepts                F F F F F F F F F F F F F F F F F F F F F F    T


    IFW   Structure of the thesis F           F F F F F F F F F F F F F F F F F F F F F F F F F    U



2   Literature review                                                                              9
    PFI   Brief natural history of HIV/AIDS                     F F F F F F F F F F F F F F F F    W


    PFP   Global HIV/AIDS situation                   F F F F F F F F F F F F F F F F F F F F F   IQ


    PFQ   HIV/AIDS situation in Malawi                    F F F F F F F F F F F F F F F F F F F   IT


    PFR   HIV/AIDS and child mortality                    F F F F F F F F F F F F F F F F F F F   IU


          PFRFI   e™quisition of rs† infe™tion in ™hildren F F F F F F F F F F F F F F F F F      IV


          PFRFP   ƒurviv—l of rs† infe™ted ™hildren F F F F F F F F F F F F F F F F F F F F       IW


          PFRFQ   ‚edu™ing underE(ve mort—lity —ttri˜ut—˜le to rs†Geshƒ F F F F F F F             PI


    PFS   Malawi's response to the HIV/AIDS epidemic and eorts
          to reduce under-ve mortality attributable to HIV/AIDS                                  PP



3   Research design and methodology                                                               26
    QFI   Methodology F         F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F   PT
          QFIFI   istim—ting rs† prev—len™e —mong women of reprodu™tive —ge F F F F               PU


          QFIFP   istim—ting the num˜er of women of reprodu™tive —ge F F F F F F F F F            QP


          QFIFQ   istim—ting the num˜er of rs† infe™ted women           F F F F F F F F F F F F   RI


          QFIFR   istim—ting the num˜er of ™hildren       F F F F F F F F F F F F F F F F F F F   RI


          QFIFS   istim—ting the num˜er of ™hildren infe™ted with rs† through motherE
                  toE™hild tr—nsmission F F F F F F F F F F F F F F F F F F F F F F F F F F F     RP


          QFIFT   istim—ting the num˜er of rs† infe™ted ™hildren dying ˜efore the —ge
                  of (ve ye—rs F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F    RT


          QFIFU   istim—ting the proportion of underE(ve mort—lity in w—l—wi during
                  the period PHHH to PHHR th—t is dire™tly —ttri˜ut—˜le to rs†Geshƒ F F           RV


    QFP   Data sources        F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F   RW


          QFPFI   rs† prev—len™e d—t— F F F F F F F F F F F F F F F F F F F F F F F F F F F       SH


          QFPFP   hemogr—phi™ d—t— F F F F F F F F F F F F F F F F F F F F F F F F F F F F        SH


    QFQ   Software      F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F   SH



4   Results                                                                                       52
    RFI   Estimates of HIV prevalence among women of
          reproductive age F F F F F F F F F F F F F F F F F F F F F F F          F F F F F F F   SQ


          RFIFI   exg sentinel surveill—n™e sites in w—l—wiD IWWW to PHHS F F F F F F F F         SQ


          RFIFP   ege p—tterns of women s—mpled in exg surveill—n™e surveys in
                  w—l—wiD IWWW to PHHS F F F F F F F F F F F F F F F F F F F F F F F F F F F      ST
          RFIFQ   ege spe™i(™ rs† prev—len™e —mong women s—mpled in exg sentinel
                  surveill—n™e surveys F F F F F F F F F F F F F F F F F F F F F F F F F F F F   ST


          RFIFR   edjusted —ge spe™i(™ rs† prev—len™e for nonEur˜—n sites F F F F F F F          TI


          RFIFS   x—tion—l estim—tes of —ge spe™i(™ rs† prev—len™e —mong women ˜—sed
                  on exg surveill—n™e d—t— F F F F F F F F F F F F F F F F F F F F F F F F       TP


          RFIFT   pin—l n—tion—l estim—tes of —ge spe™i(™ rs† prev—len™e —mong women
                  ˜—sed on exg surveill—n™e d—t— —nd PHHR w—l—wi hrƒ results F F F F             TS


    RFP   Estimates of women of reproductive age                       F F F F F F F F F F F F   TW


          RFPFI   gomp—rison of estim—tes of women of reprodu™tive —ge in w—l—wi from
                  IWWW to PHHR F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F   UR


    RFQ   Estimates of the number of HIV infected women                              F F F F F   UU


    RFR   Estimates of the number of children                    F F F F F F F F F F F F F F F   VQ


    RFS   Estimates of the number of children infected with HIV
          through mother-to-child transmission F F F F F F F F F F F F F F                       VW


    RFT   Estimates of the number of HIV infected children dying
          before the age of ve years during the period 2000 to 2004                             WQ


    RFU   Estimates of under-ve mortality directly attributable
          to HIV/AIDS in Malawi during the period 2000 to 2004                                   WS


    RFV   HIV/AIDS, PMTCT programs and U5MR reduction
          goals in Malawi F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F          WU



5   Discussion and conclusions                                                                   99
Appendix       105

Bibliography   122
List of Tables

 QFI   …r˜—n —nd rur—l fem—le popul—tions in w—l—wi F F F F F F F F F F F F F F F F F         QH


 QFP   €er™ent of ur˜—n —nd rur—l fem—le popul—tions in w—l—wi F F F F F F F F F F F          QH


 QFQ   ‡ry wv„ for fem—les in w—l—wiD PHHHEPHHR F F F F F F F F F F F F F F F F F F           QT


 QFR   ƒurviv—l r—tios of fem—les from go—leEhemeny north model life t—˜le F F F F F          QU


 QFS   ƒurviv—l r—tios of fem—les from go—leEhemeny south model life t—˜le F F F F F          QV


 QFT   ƒurviv—l r—tios of fem—les from …x p—r i—st model life t—˜le F F F F F F F F F         QW


 QFU   €roje™ted life expe™t—n™y —t ˜irth v—lues for fem—les in w—l—wi F F F F F F F F        RI


 QFV   €„‚ for rs† infe™ted pregn—nt women re™eiving xevir—pineF F F F F F F F F F            RT


 RFI   gl—ssi(™—tion of exg sentinel surveill—n™e sites in w—l—wiD IWWW to PHHSF F F          SS


 RFP   „rends in the num˜er of women s—mpled in exg sentinel surveill—n™e surveys
       in w—l—wi from IWWW to PHHS ˜y site ™—tegoryF F F F F F F F F F F F F F F F F F        ST


 RFQ   ege p—tterns of women s—mpled in exg surveill—n™e surveys in w—l—wiD IWWW
       to PHHSF F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F   SU



                                              xi
RFR   ege spe™i(™ rs† prev—len™e for women s—mpled in exg surveill—n™e surveys
      in —ggreg—ted ur˜—n sites in w—l—wiD IWWW to PHHSF F F F F F F F F F F F F F F F    SV


RFS   ege spe™i(™ rs† prev—len™e for women s—mpled in exg surveill—n™e surveys
      in —ggreg—ted nonEur˜—n sites in w—l—wiD IWWW to PHHSF F F F F F F F F F F F F      SW


RFT   ege spe™i(™ rs† prev—len™e for —ggreg—ted nonEur˜—n sites in w—l—wi from
      IWWW to PHHS —djusted for represent—tivenessF F F F F F F F F F F F F F F F F F F   TP


RFU   €relimin—ry estim—tes of the n—tion—l —ge spe™i(™ rs† prev—len™e —mong
      women in w—l—wi from IWWW to PHHS ˜—sed on exg surveill—n™e d—t—F F F F F           TQ


RFV   pin—l estim—tes of the n—tion—l —ge spe™i(™ rs† prev—len™e —mong women in
      w—l—wi from IWWW to PHHR ˜—sed on exg surveill—n™e d—t— F F F F F F F F F F         TQ


RFW   istim—tes of —ge spe™i(™ rs† prev—len™e —mong fem—les —ged ISERW in w—l—wi
      derived from PHHR w—l—wi hrƒ F F F F F F F F F F F F F F F F F F F F F F F F F      TS


RFIH istim—tes of r—tios of exg rs† prev—len™e to gener—l popul—tion rs† prev—len™e TV


RFII pin—l estim—tes of —ge spe™i(™ n—tion—l rs† prev—len™e —mong women of
      reprodu™tive —ge in w—l—wiD IWWW to PHHS F F F F F F F F F F F F F F F F F F F F    TV


RFIP ege smoothed popul—tion of fem—les in w—l—wi in IWWVF F F F F F F F F F F F F        TW


RFIQ istim—tes of women of reprodu™tive —ge in w—l—wi from IWWW to PHHI ˜—sed
      on ‡ry wv„ systemF F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F      UH


RFIR istim—tes of women of reprodu™tive —ge in w—l—wi from PHHP to PHHR ˜—sed
      on ‡ry wv„ systemF F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F      UH


RFIS istim—tes of women of reprodu™tive —ge in w—l—wi from IWWW to PHHI ˜—sed
      on …x p—r i—st wv„ systemF F F F F F F F F F F F F F F F F F F F F F F F F F F      UI
RFIT istim—tes of women of reprodu™tive —ge in w—l—wi from PHHP to PHHR ˜—sed
     on …x p—r i—st wv„ systemF F F F F F F F F F F F F F F F F F F F F F F F F F F        UI


RFIU istim—tes of women of reprodu™tive —ge in w—l—wi from IWWW to PHHI ˜—sed
     on go—leEhemeny north wv„ systemF F F F F F F F F F F F F F F F F F F F F F F         UP


RFIV istim—tes of women of reprodu™tive —ge in w—l—wi from PHHP to PHHR ˜—sed
     on go—leEhemeny north wv„ systemF F F F F F F F F F F F F F F F F F F F F F F         UP


RFIW istim—tes of women of reprodu™tive —ge in w—l—wi from IWWW to PHHI ˜—sed
     on go—leEhemeny ƒouth wv„ systemF         F F F F F F F F F F F F F F F F F F F F F   UQ


RFPH istim—tes of women of reprodu™tive —ge in w—l—wi from PHHP to PHHR ˜—sed
     on go—leEhemeny ƒouth wv„ systemF         F F F F F F F F F F F F F F F F F F F F F   UQ


RFPI istim—tes of the num˜er of rs† infe™ted women in w—l—wi for estim—tes of
     women ˜—sed on ‡ry wv„ systemF F F F F F F F F F F F F F F F F F F F F F F            UV


RFPP istim—tes of the num˜er of rs† infe™ted women in w—l—wi for estim—tes of
     women ˜—sed on …x p—r i—st wv„ systemF F F F F F F F F F F F F F F F F F F            UW


RFPQ istim—tes of the num˜er of rs† infe™ted women in w—l—wi for estim—tes of
     women ˜—sed on go—leEhemeny xorth wv„ systemF F F F F F F F F F F F F F F             VH


RFPR istim—tes of the num˜er of rs† infe™ted women in w—l—wi for estim—tes of
     women ˜—sed on go—leEhemeny ƒouth wv„ systemF F F F F F F F F F F F F F F             VI


RFPS istim—tes of n—tion—l rs† prev—len™e —mong women of reprodu™tive —ge in
     w—l—wi from IWWW to PHHS ˜y xeg@PHHSA F F F F F F F F F F F F F F F F F F F F         VI


RFPT egeEspe™i(™ fertility r—tes @per I HHH womenA in w—l—wi in PHHH —nd PHHRF         F   VQ


RFPU istim—tes of —geEspe™i(™ fertility r—tes @per I HHH womenA for —ll women in
     w—l—wi from IWWW to PHHRF F F F F F F F F F F F F F F F F F F F F F F F F F F F F     VS
RFPV istim—tes of the num˜er of ˜irths to —ll women in w—l—wi from IWWW to PHHR
     for estim—tes of women ˜—sed on ‡ry wv„ systemF F F F F F F F F F F F F F           VT


RFPW istim—tes of the num˜er of ˜irths to —ll women in w—l—wi from IWWW to PHHR
     for estim—tes of women ˜—sed on …x p—r i—st wv„ systemF F F F F F F F F F           VT


RFQH istim—tes of the num˜er of ˜irths to —ll women in w—l—wi from IWWW to PHHR
     for estim—tes of women ˜—sed on go—leEhemeny xorth wv„ systemF F F F F F            VU


RFQI istim—tes of the num˜er of ˜irths to —ll women in w—l—wi from IWWW to PHHR
     for estim—tes of women ˜—sed on go—leEhemeny ƒouth wv„ systemF F F F F F            VU


RFQP istim—tes of the num˜er of ˜irths to rs†Einfe™ted women in w—l—wi from
     IWWW to PHHR for estim—tes of women ˜—sed on ‡ry wv„ systemF F F F F F F            VW


RFQQ istim—tes of the num˜er of ˜irths to rs†Einfe™ted women in w—l—wi from
     IWWW to PHHR for estim—tes of women ˜—sed on …x p—r i—st wv„ systemF F F            WH


RFQR istim—tes of the num˜er of ˜irths to rs†Einfe™ted women in w—l—wi from IWWW
     to PHHR for estim—tes of women ˜—sed on go—leEhemeny xorth wv„ systemF              WH


RFQS istim—tes of the num˜er of ˜irths to rs†Einfe™ted women in w—l—wi from IWWW
     to PHHR for estim—tes of women ˜—sed on go—leEhemeny ƒouth wv„ systemF              WI


RFQT istim—tes of the num˜er of ™hildren infe™ted with rs† through w„g„ in
     w—l—wi from IWWW to PHHR for estim—tes of women ˜—sed on ‡ry wv„ systemF WP


RFQU istim—tes of the num˜er of ™hildren infe™ted with rs† through w„g„ in
     w—l—wi from IWWW to PHHR for estim—tes of women ˜—sed on …x p—r i—st
     wv„ systemF F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F   WP
RFQV istim—tes of the num˜er of ™hildren infe™ted with rs† through w„g„ in
    w—l—wi from IWWW to PHHR for estim—tes of women ˜—sed on go—leEhemeny
    xorth wv„ systemF F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F   WQ


RFQW istim—tes of the num˜er of ™hildren infe™ted with rs† through w„g„ in
    w—l—wi from IWWW to PHHR for estim—tes of women ˜—sed on go—leEhemeny
    ƒouth wv„ systemF F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F   WQ


RFRH istim—tes of the tot—l num˜er of ™hildren infe™ted with rs† through w„g„
    dying ˜efore the —ge of (ve ye—rs during the period PHHH to PHHR in w—l—wiF         WR


RFRI istim—tes of …Sw‚ due to rs†Geshƒ —nd proportion of underE(ve mort—lity
    —ttri˜ut—˜le to rs†Geshƒ in w—l—wi during the period PHHH to PHHRF F F F F          WT
List of Figures

 PFI   ƒt—ges of rs† infe™tionF F F F F F F F F F F F F F F F F F F F F F F F F F F F F F    IQ


 QFI   ƒurviv—l distri˜ution of rs†Einfe™ted ™hildrenF F F F F F F F F F F F F F F F F F     RV


 RFI   exg sentinel surveill—n™e sites in w—l—wiD IWWW to PHHS F F F F F F F F F F F F       SR


 RFP   ege spe™i(™ rs† prev—len™e in —ggreg—ted ur˜—n sitesF F F F F F F F F F F F F         TH


 RFQ   ege spe™i(™ rs† prev—len™e in —ggreg—ted nonEur˜—n sitesF F F F F F F F F F F         TI


 RFR   x—tion—l estim—tes of —ge spe™i(™ rs† prev—len™e —mong women ˜—sed on
       exg d—t—F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F   TR


 RFS   gomp—rison ˜etween estim—tes of —ge spe™i(™ rs† prev—len™e —mong women
       —ged ISERW in w—l—wi in PHHR ˜—sed on PHHR w—l—wi hrƒ —nd exg d—t—F F F               TU


 RFT   gomp—rison of estim—tes of tot—l women popul—tion —ged ISERW in w—l—wi
       from IWWW to PHHRF F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F    US


 RFU   gomp—rison of estim—tes of women —ged ISERW in w—l—wi from IWWWEPHHR F F              UT


 RFV   gomp—rison of this study9s estim—tes of rs† prev—len™e for —dult fem—les in
       w—l—wi with xeg @PHHSA9s estim—tes from PHHH to PHHR F F F F F F F F F F F F          VP

                                             xvi
RFW   gomp—rison of —geEspe™i(™ fertility r—tes @per I HHH womenA in w—l—wi in PHHH
      —nd PHHRF F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F   VR


RFIH gomp—rison of estim—tes of the num˜er of ˜irths to —ll women in w—l—wi from
      IWWW to PHHRF F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F   VV


RFII rs†GeshƒD €w„g„ progr—ms —nd …Sw‚ redu™tion go—ls in w—l—wi —ssumE
      ing — €„‚ of QH7F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F F     WV
Acronyms and abbreviations


ANC     enten—t—l ™lini™

AIDS    e™quired smmunode(™ien™y ƒyndrome

AIM     eshƒ smp—™t wodel

DHS     hemogr—phi™ —nd re—lth ƒurvey

HIV     rum—n smmunode(™ien™y †irus

IDU     snje™tion drug users

MDG     willennium hevelopment qo—ls

MLT     wodel vife „—˜le

MSM     wen who h—ve sex with men

MTCT    wotherEtoE™hild tr—nsmission

NAC     x—tion—l eshƒ gommission ‘w—l—wi“

NSO     x—tion—l ƒt—tisti™—l y0™e ‘w—l—wi“

PMTCT   €revention of motherEtoE™hild tr—nsmission

PTR     €erin—t—l tr—nsmission r—te

SSA     ƒu˜Eƒ—h—r—n efri™—

U5MR    …nderE(ve mort—lity r—te



                                       xviii
Chapter 1


Introduction



1.1 Background to the study

wost ™ountries in the world registered su˜st—nti—l redu™tions in mort—lity —mong inf—nts —nd
™hildren during the l—te PHth ™entury @ehm—d et —lFD PHHHAF roweverD underE(ve mort—lity
still rem—ins one of the le—ding ™h—llenges for pu˜li™ he—lth espe™i—lly in the developing worldF
e˜out II million ™hildren still die every ye—r ˜efore their (fth ˜irthd—y worldwide —nd more
th—n IH million of these de—ths o™™ur in the developing world @rill —nd emouzouD PHHTY fl—™k
et —lFD PHHQAF ‡ithin the developing worldD ƒu˜Eƒ—h—r—n efri™— @ƒƒeA is the most severely
—'e™ted region —nd —™™ounts for more th—n oneEthird of glo˜—l underE(ve de—ths @rill —nd
emouzouD PHHTAF


p—™tors responsi˜le for the high levels of underE(ve mort—lity in the developing world —nd
ƒƒe in p—rti™ul—r —re m—ny —nd v—riedF „he high mort—lity levels —re either the e'e™t of
poor so™i—lD e™onomi™D ™ultur—lD —nd he—lth system ™onditions th—t oper—te through — set
of proxim—te determin—nts th—t dire™tly in)uen™e the risk of dise—se —nd the out™ome of
dise—se pro™esses proposed ˜y wosley —nd ghen @IWVRA or neg—tive ™h—nges in the proxim—te


                                               I
determin—nts themselvesF e study ™ondu™ted ˜y ‚utstein @PHHHA illustr—tes the —sso™i—tion
˜etween high mort—lity —mong ™hildren less th—n (ve ye—rs of —ge —nd neg—tive ™h—nges in
some of the proxim—te determin—nts of ™hild mort—lity in developing ™ountriesF p—™tors
ex—mined ˜y ‚utstein @PHHHA in™lude fertility ˜eh—vioursD nutrition—l st—tus of ™hildren
—nd p—tterns of ˜re—stfeeding —nd inf—nt feedingD m—tern—l —nd ™hild he—lth st—tus —nd
the use of he—lth servi™esD environment—l he—lth ™onditions —nd so™ioe™onomi™ f—™torsF
elthough these f—™tors in)uen™e the o™™urren™e of — wide spe™trum of ™hildhood dise—sesD
only (ve ™ommuni™—˜le dise—ses @pneumoni—D di—rrhe—D m—l—ri—D me—sles —nd rs†GeshƒA
—re responsi˜le for more th—n h—lf of de—ths in ™hildren under the —ge of (ve ye—rs in
developing ™ountries @fry™e et —lD PHHSY ‡ryD PHHS—AF feing the world9s most severely
rs†Geshƒ —'e™ted region @…xeshƒ PHHTAD the underE(ve mort—lity ˜urden of rs†Geshƒ
is most pronoun™ed in ƒƒeF „here is —n indi™—tion th—t p—edi—tri™ rs† infe™tion ™ontr—™ted
through motherEtoE™hild tr—nsmission @w„g„A is signi(™—ntly ™ontri˜uting to the o˜served
level of underE(ve mort—lity in ƒƒe ™ountries with high —dult rs† prev—len™e @edetunjiD
PHHHY ‡—lkerD PHHPY q—renne —nd q—kusiD PHHTAF


elthough the underE(ve mort—lity r—te @…Sw‚A in w—l—wi h—s ˜een de™lining sin™e IWTH @rill
et —lD IWWVY ehm—d et —lFD PHHHD …xsgipD IWWV —nd PHHSAD w—l—wi9s underE(ve mort—lity
rem—ins one of the highest in the worldF es of PHHRD w—l—wi9s underE(ve mort—lity w—s r—nked
IW in the world in terms of severity @…xsgipD PHHSAF ‡ith respe™t to —dult rs†Geshƒ
infe™tionD w—l—wi —lso h—s one of the highest prev—len™e r—tes in the worldF e™™ording to
…xeshƒ @PHHTAD rs† prev—len™e r—te —mong —dults —ged IS to RW st—nds —t IRFI7F ‡ithout
dou˜tD the high —dult rs† prev—len™e in w—l—wi h—s —n e'e™t on the levels of underE(ve
mort—lity in the ™ountryF roweverD the m—gnitude of the ™ontri˜ution of rs†Geshƒ to the
most re™ent estim—tes of underE(ve mort—lity in w—l—wi is not ™le—rly knownF sn order to
—ppropri—tely t—rget interventions to redu™e de—ths —mong ™hildren under the —ge of (ve ye—rs
—nd to monitor progress there rem—ins —n ongoing need to underst—nd ˜etter the proportion
of ™—useEspe™i(™ underE(ve mort—lityF „hereforeD this study —ims —t estim—ting the dire™t
™ontri˜ution of rs†Geshƒ to the o˜served level of underE(ve mort—lity in w—l—wi during the
period PHHH to PHHRF


                                             P
1.2 Statement of the problem

„he most re™ent estim—tes of …Sw‚ in w—l—wi —v—il—˜le —t the time of ™ondu™ting this
study —re ˜—sed on the PHHR w—l—wi hemogr—phi™ —nd re—lth ƒurvey @hrƒAF „he estim—tes
indi™—te th—t the num˜er of de—ths —mong ™hildren —ged ˜elow (ve ye—rs stood —t IQQ de—ths
per I HHH live ˜irths for the period PHHH to PHHR @xƒy —nd y‚g w—™roD PHHSAF roweverD the
proportion of underE(ve mort—lity th—t ™—n ˜e dire™tly —ttri˜uted to rs†Geshƒ ™ontr—™ted
through w„g„ during this period is not ™le—rly knownF „hereforeD there is need for estim—tes
of the underE(ve mort—lity dire™tly —ttri˜ut—˜le to rs†Geshƒ during this period to —ssess
the m—gnitude ˜y whi™h prevention of motherEtoE™hild tr—nsmission @€w„g„A progr—mmes
™urrently ˜eing implemented in w—l—wi would potenti—lly h—ve redu™ed this o˜served level
of underE(ve mort—lityF




1.3 Aim and objectives of the study

„he —im of this study is to estim—te the dire™t ™ontri˜ution of rs†Geshƒ to the o˜served
level of underE(ve mort—lity in w—l—wi during the period PHHH to PHHRF elthough there —re
m—ny w—ys in whi™h —dult rs†Geshƒ ™ould —'e™t the level of underE(ve mort—lity —s outlined
˜y edetunji @PHHHA —nd ‡—lker et —lF @PHHPAD this study will only estim—te the ™ontri˜ution
of rs†Geshƒ to underE(ve mort—lity th—t resulted through verti™—l or perin—t—l tr—nsmission
of rs† from infe™ted mothers to their ™hildrenF „husD estim—tes of the indire™t e'e™ts of
—dult rs†Geshƒ on underE(ve mort—lity —re outside the s™ope of this studyF ƒpe™i(™—llyD
the o˜je™tive of this study is to estim—te how mu™h of the underE(ve mort—lity o˜served in
w—l—wi for the period PHHH to PHHR is dire™tly —ttri˜ut—˜le to rs†GeshƒF




                                             Q
1.4 Research question

„his study —ttempts to —nswer the following questionX What proportion of the under-ve
mortality observed in Malawi for the period 2000 to 2004 would be directly attributed to
HIV/AIDS?




1.5 Rationale and signicance of this study

vike m—jority of ™ountries in the worldD w—l—wi signed the willennium he™l—r—tion th—t
w—s —dopted in ƒeptem˜er PHHH @httpXGGwwwFoh™hrForgGenglishGl—wGmillenniumFhtmA —nd
therefore ˜e™—me p—rty to the willennium hevelopment qo—l @whqA of redu™ing ˜y two
thirds the …Sw‚ o˜served in IWWH ˜y PHIS @httpXGGwwwFunForgGmillenniumgo—lsGY rill —nd
emouzouD PHHTY ‡—lker et —lFD PHHPD w—l—wi millennium development go—ls report PHHQAF
„his redu™tion tr—nsl—tes into —n —nnu—l —ver—ge redu™tion r—te of 4.3%F


w—l—wi9s …Sw‚ w—s one of the highest in ƒƒe in IWWHF istim—tes ˜—sed on the IWWP w—l—wi
hrƒ indi™—te th—t the num˜er of de—ths —mong ™hildren —ged ˜elow (ve ye—rs for the period
IWVV to IWWP w—s PQR de—ths per I HHH live ˜irths @xƒy —nd y‚g w—™roD IWWRAF „his me—ns
th—t w—l—wi t—rgets to redu™e its …Sw‚ to UV de—ths per I HHH live ˜irths ˜y the ye—r PHISD
whi™h is still high ˜y world st—nd—rdsF


elthough w—l—wi experien™ed — de™line in underE(ve mort—lity from PQR de—ths per I HHH
live ˜irths in IWWH to IQQ de—ths per I HHH live ˜irths in PHHR @xƒy —nd y‚g w—™roD IWWR
—nd PHHSAD the —nnu—l —ver—ge r—te of redu™tion @3.96%A during this period is lower th—n
the expe™ted t—rget of 4.3%F „husD w—l—wi is slightly o' tr—™k in meeting the ™hildEsurviv—l
whqF „hereforeD in order to properly t—rget interventions to —™hieve further su˜st—nti—l
redu™tions in …Sw‚D the w—l—wi—n government —nd other st—keholders need ™—useEspe™i(™
estim—tes of the ™ountry9s underE(ve mort—lityF w—l—wi ˜eing — ™ountry with one of the
highest rs† prev—len™e r—tes in the worldD it is of interest to estim—te the proportion of underE

                                               R
(ve mort—lity th—t is dire™tly —ttri˜ut—˜le to rs†Geshƒ to gu—ge the potenti—l ™ontri˜ution
of €w„g„ progr—mmes tow—rds meeting the over—ll ™hildEsurviv—l whqF purtherD for the
(rst time w—l—wi now h—s n—tion—l rs† prev—len™e estim—tes for men —nd women derived
from the PHHR w—l—wi hrƒ whi™h in™luded rs† testing @xƒy —nd y‚g w—™roD PHHSAF es
estim—tes of rs† prev—len™e —re —n integr—l p—rt of the pro™ess of estim—ting the dire™t e'e™t
of rs†Geshƒ on underE(ve mort—lityD it is inevit—˜le to in™orpor—te the inform—tion on rs†
prev—len™e from this gener—l popul—tionE˜—sed survey into the existing d—t— —nd m—ke new
estim—tes of the dire™t imp—™t of rs†Geshƒ on underE(ve mort—lity in w—l—wiF ‚esults of
this study —re expe™ted to ˜e of poli™y relev—n™e to w—l—wi —nd other ™ountries experien™ing
simil—r rs† prev—len™e —nd mort—lity p—tternsF




1.6 Conceptual framework

„his study is ˜—sed on the premise th—t the num˜er of ™hildren under the —ge of (ve ye—rs who
die of rs†Geshƒ is dire™tly rel—ted to the num˜er of ™hildren who —™quired rs† infe™tion
from their mothersF „hereforeD the proportion of underE(ve ye—r old ™hildren dying from
rs†Geshƒ is — fun™tion of the proportion of rs†Einfe™ted mothersD the r—te of w„g„ —nd
the proportion of infe™ted ™hildren who die ˜efore their (fth ˜irthd—y @edetunjiD PHHHAF




1.7 Research methodology

‡hile popul—tionE˜—sed d—t— on tot—l underE(ve mort—lity —re in™re—singly —v—il—˜le
through the hrƒ @httpXGGwwwFme—suredhsF™omGAD popul—tionE˜—sed epidemiologi™—l d—t—
—re extremely limited in ƒƒe ™ountriesF sn most ™ountries reli—˜le d—t— on ™—use of de—th —re
un™ommon ˜e™—use vit—l registr—tion systems h—ve extremely limited ™over—ge @edetunjiD
PHHHY ƒ—lomon —nd wurr—yD PHHIY ‡—lker et —lF PHHPAF por this re—sonD demogr—phers



                                              S
—nd he—lth rese—r™hers h—ve tended to resort to indire™t methodsD in™luding the use of
m—them—ti™—l models in estim—ting the proportion of ™—useEspe™i(™ underE(ve mort—lityF
‡ith respe™t to rs†GeshƒD some of the indire™t methods —nd m—them—ti™—l models used
to estim—te its ™ontri˜ution tow—rds underE(ve mort—lity in™lude those reported ˜y edetunji
@PHHHAD ‡—lker et —lF @PHHPAD —˜— et —lF @PHHQAD ƒtover @PHHSA —nd tohnson —nd horrington
@PHHTAF „his study uses some of the existing indire™t methods —nd m—them—ti™—l models
to derive estim—tes of underE(ve mort—lity dire™tly —ttri˜ut—˜le to rs†Geshƒ in w—l—wi
for the period PHHH to PHHRF     „o — l—rger extentD the study —dopted the methods —nd
—ppro—™hes re™ommended ˜y …xeshƒ —nd ‡ry th—t —re implemented in hem€roj —nd
eshƒ smp—™t wodel ™omputer p—™k—ges @„he …xeshƒ ‚eferen™e qroup on istim—tesD
wodels —nd €roje™tionsD PHHQY ƒtover —nd uirmeyerD PHHRY ƒtover PHHSAF xo new d—t—
were ™olle™tedD inste—d the study used se™ond—ry d—t— —nd pu˜lished reportsF




1.8 Denitions of key concepts

„he following de(nitions of the key ™on™epts or terms —re used in this studyF


Acquired Immunodeciency Syndrome (AIDS):                 e dise—se of the ˜ody9s immune
system ™—used ˜y the hum—n immunode(™ien™y virus @rs†AF st is ™h—r—™terized ˜y the de—th
of ghR ™ells @—n import—nt p—rt of the ˜ody9s immune systemAD whi™h le—ves the ˜ody
vulner—˜le to lifeEthre—tening ™onditions su™h —s infe™tions —nd ™—n™ersF


Death:   €erm—nent dis—ppe—r—n™e of —ll eviden™e of life —t —ny time —fter ˜irth h—s t—ken
pl—™eF


Epidemic:   e dise—se th—t h—s spre—d r—pidly through — segment of the hum—n popul—tion
in — given geogr—phi™ —re—F


Human Immunodeciency Virus (HIV):                „he virus th—t ™—uses eshƒF rs† ˜elongs
to the retrovirus f—milyD —nd h—s two su˜typesX rs†EID whi™h is responsi˜le for most rs†

                                              T
infe™tions throughout the worldD —nd rs†EP whi™h is found prim—rily in ‡est efri™—F


HIV prevalence:     „he num˜er of rs†Einfe™ted people in — popul—tion —t — given timeF


Mortality:   e gener—l term for in™iden™e of de—ths in — popul—tion or groupF


Mother-to-Child Transmission (MTCT):             „he p—ss—ge of rs† from —n rs†Einfe™ted
mother to her inf—ntF „he inf—nt m—y ˜e™ome infe™ted while in the mother9s wom˜D during
l—˜our —nd deliveryD or through ˜re—stfeedingF


Sub-Saharan Africa:      „he term used to des™ri˜e the —re— of the efri™—n ™ontinent whi™h
lies south of the ƒ—h—r— desertF


Under-ve mortality:      „he pro˜—˜ility of dying ˜etween ˜irth —nd the (fth ˜irthd—yF


Women of reproductive age:         ‡omen —ged from IS to RW ye—rsF




1.9 Structure of the thesis

„his ™h—pter h—s outlined the m—in rese—r™h pro˜lem under investig—tion in this studyF „he
™h—pter h—s presented the m—in —ims —nd o˜je™tives of this study —nd provided the r—tion—le
for ™—rrying out the studyF „his ™h—pter h—s —lso ˜rie)y des™ri˜ed the ™on™eptu—l fr—mework
for this study —s well —s the methodology usedF gh—pter two presents — review of the
liter—ture on — num˜er of —spe™ts pert—ining to this studyD whi™h in™ludeX — ˜rief history of
rs†GeshƒY —n overview of the glo˜—l rs†Geshƒ situ—tionY —n overview of the rs†Geshƒ
situ—tion in w—l—wiY me™h—nisms through whi™h rs†Geshƒ ™ontri˜utes to ™hild mort—lityY
—nd w—ys of redu™ing rs†Geshƒ —ttri˜ut—˜le mort—lity in ™hildrenF gh—pter three provides
— det—iled des™ription of the methods —nd —ppro—™hes used in this studyF gh—pter four
™ont—ins the results of the studyF gh—pter (ve presents — dis™ussion of the results of the
study —nd ™on™lusions dr—wn from the (ndingsF v—stlyD the eppendix ™ont—ins sele™ted


                                             U
progr—ms written in ‚ th—t were used to o˜t—in the results of this studyF




                                             V
Chapter 2


Literature review


„his ™h—pter presents — review of the liter—ture on — num˜er of —spe™ts pert—ining to this
studyF „he ™h—pter ™onsists of six se™tionsF „he (rst se™tion presents — ˜rief n—tur—l history of
rs†GeshƒF „he se™ond se™tion presents —n overview of the glo˜—l rs†Geshƒ situ—tionF „he
third se™tion presents —n overview of the rs†Geshƒ situ—tion in w—l—wiF „he fourth se™tion
des™ri˜es the me™h—nisms through whi™h rs†Geshƒ ™ontri˜utes to ™hild mort—lityF „he (fth
se™tion presents w—ys of redu™ing rs†Geshƒ —ttri˜ut—˜le mort—lity in ™hildrenF pin—llyD
the sixth se™tion presents inform—tion pert—ining to w—l—wi9s response to the rs†Geshƒ
epidemi™ —nd e'orts to to redu™e underE(ve mort—lity —ttri˜ut—˜le to rs†GeshƒF




2.1 Brief natural history of HIV/AIDS

eshƒ initi—lly emerged —s two r—re illnesses th—t —'e™ted — few homosexu—l men in g—liforni—
—nd xew ‰ork gity in the …nited ƒt—tes of emeri™— in IWVIF „hese were (ve ™—ses of
€neumo™ystis ™—rinii pneumoni— @ghgD IWVI—A —nd PT ™—ses of — v—s™ul—r skin tumor known
—s u—posi9s s—r™om— —mong young —nd otherwise he—lth homosexu—l men @ghgD IWVI˜AF


                                               W
st w—s dis™overed th—t ˜oth of these r—re dise—ses were —sso™i—ted with — de(™ien™y in the
immune system eviden™ed ˜y — de™line in the num˜er of helper „ lympho™ytes @ghR C „
vympho™ytesA @frookmeyer —nd q—ilD IWWRXQAF roweverD ™—se reports of —ll these p—tients
indi™—ted th—t none of them su'ered from —ny underlying illness th—t would ˜e responsi˜le
for the development of these often f—t—l —nd un™ommon dise—ses @ghgD IWVI—Y ghgD IWVI˜Y
frookmeyer —nd q—ilD IWWRXQY uelly —nd ƒtF v—wren™eD IWVVXPAF „he genters for hise—se
gontrol —nd €revention @ghgA then n—med the syndrome —™™ounting for these few ™—ses of
mysterious illnesses —s —™quired immune de(™ien™y syndrome @eshƒA in IWVP @frookmeyer
—nd q—ilD IWWRXQAF


„he ghg de(ned eshƒ —s — dise—se th—t w—s 4—t le—st moder—tely predi™tive of — defe™t
in ™ellEmedi—ted immunityD o™™urring in — person with no known ™—use for diminished
resist—n™e to th—t dise—se4 su™h —s €neumo™ystis ™—rinii pneumoni—D u—posi9s s—r™om—
—nd serious opportunisti™ infe™tions @ghgD IWVPAF prom tune ID IWVI to ƒeptem˜er ISD
IWVPD the ghg re™eived reports of SWQ eshƒ ™—ses out of whi™h PRQ @41%A ™—ses h—d died
@ghgD IWVPAF „here—fterD the num˜er of eshƒ ™—ses —nd de—ths ˜eg—n to riseD ˜ut the
underlying ™—uses of the immune de(™ien™ies seen in the eshƒ ™—ses were unknownF sn
responseD medi™—l investig—tors s™r—m˜led to (nd — ™—use —nd more import—ntly — ™ure for
this new dise—seF „hen in IWVQD investig—tors —t the snstitut €—steur in pr—n™e identi(ed
unusu—l serum —nti˜odies in — m—jority of eshƒ p—tients —nd isol—ted — retrovirus th—t they
˜elieve to ˜e responsi˜le for those —nti˜odies @gi™ho™kiD PHHTAF „hey n—med this virus —s
lymph—denop—thyE—sso™i—ted virus @ve†A @uelly —nd ƒtF v—wren™eD IWVVXPAF e ye—r l—ter the
…nited ƒt—tes government —nnoun™ed th—t their s™ientistD hrF ‚o˜ert q—llo of the x—tion—l
g—n™er snstitute isol—ted — retrovirus ™—lled hum—n „E™ell lymphotropi™ virus type sss @r„v†E
sssAD th—t he too ™l—imed w—s responsi˜le for the eshƒ out˜re—kF „his resulted in — dispute
˜etween te—ms from the two rese—r™h institutions @f—rnett —nd fl—ikieD IWWPXPAF „wo ye—rs
l—ter it w—s ™on(rmed th—t r„v†Esss —nd ve† were —™tu—lly the s—me virus @gi™ho™ki PHHTY
uelly —nd ƒtF v—wren™eD IWVVXPAD —nd q—llo w—s still ™redited with its dis™overy @gi™ho™kiD
PHHTAF pollowing this ™on(rm—tionD —n intern—tion—l ™ommittee of s™ientists ren—med the
virus —s rum—n smmunode(™ien™y †irus @rs†A @gi™ho™kiD PHHTY uelly —nd ƒtF v—wren™eD


                                             IH
IWVVXPY frookmeyer —nd q—ilD IWWRXRAF


„he dis™overy of rs† set the st—ge for —dv—n™ing knowledge —˜out the dyn—mi™—l fe—tures
of rs† —nd eshƒF e we—lth of inform—tion is now —v—il—˜le on the tr—nsmission me™h—nism
of rs† —nd the progression from rs† infe™tion through eshƒF st is now known th—t rs†
tr—nsmission ™—n o™™ur when there is ™ont—™t ˜etween spe™i(™ ˜ody )uidsD p—rti™ul—rly ˜loodD
v—gin—l se™retionsD semenD preEej—™ul—tory )uid —nd ˜re—st milk of —n rs†Einfe™ted person —nd
the ˜lood —ndGor mu™ous mem˜r—nes of —n uninfe™ted person @vewthw—ite —nd ‡ilkinsD PHHSY
ƒ™hoe˜erleinD PHHIY vevyD IWWQAF gurrently there is no eviden™e th—t rs† ™—n ˜e tr—nsmitted
through pure te—rsD mu™ous from the noseD s—liv—D swe—tD urine —nd fe™es @ƒ™hoe˜erleinD PHHIY
vevyD IWWQAF


st is —lso now known th—t in —dultsD —fter —™quisition of rs†D there —re —t le—st three ph—ses
to the ™ourse of rs† infe™tionD whi™h in™lude prim—ry rs† infe™tion ph—seD l—tent ph—seD
—nd symptom—ti™ rs† infe™tion ph—se @ƒ—n pr—n™is™o eshƒ pound—tion PHHTY vewthw—ite
—nd ‡ilkinsD PHHSY windel —nd „en—ntEplowersD PHHIY f—nnister et —lFD PHHHAF €rim—ry rs†
infe™tionD whi™h is —lso ™—lled —™ute rs† infe™tion or sero™onversion illness represents the (rst
st—ge of rs† infe™tion —fter the —™quisition of rs† when —nti˜odies —re developing @windel
—nd „en—ntEplowersD PHHIAF huring the prim—ry or —™ute rs† infe™tion ph—seD rs† m—kes
its w—y to the lymph nodes —nd —™tively repli™—tes —nd rele—ses new virus p—rti™les into the
˜loodstre—m @ƒ—n pr—n™is™o eshƒ pound—tion PHHTAF €eople —t this st—ge often h—ve — very
high rs† vir—l lo—d —nd the ™on™entr—tion of ghRC „ ™ells me—sured in their ˜lood de™lines
@windel —nd „en—ntEplowersD PHHIY f—nnister et —lD PHHHY €erelson —nd xelsonD IWWWY vevyD
IWWQAF elso during the prim—ry or —™ute rs† infe™tion ph—se some people experien™e )uE
like symptomsD su™h —s feversD ™hillsD night swe—tsD —nd r—shes whi™h usu—lly l—st ˜etween
two to six weeks —nd resolve without spe™i(™ tre—tment @ƒ—n pr—n™is™o eshƒ pound—tion
PHHTY windel —nd „en—ntEplowersD PHHIY f—nnister et —lD PHHHY €erelson —nd xelsonD IWWWAF
roweverD people with —™ute rs† infe™tion usu—lly do not test positive for rs† —nti˜odies until
sero™onversionD whi™h ™ommonly o™™urs ˜etween T to IP weeks —fter infe™tion is ™omplete
@ƒ—n pr—n™is™o eshƒ pound—tion PHHTY €iot —nd gole˜undersD IWVUAF „he l—tent ph—se of
rs† infe™tion su™™eeds sero™onversionF efter sero™onversionD the ™on™entr—tion of ghRC

                                               II
„ ™ells in the ˜lood usu—lly in™re—ses —g—in —nd de™lines gr—du—lly during the l—tent ph—se
@vewthw—ite —nd ‡ilkinsD PHHSY f—nnister et —lD PHHHY €erelson —nd xelsonD IWWWAF elso —fter
sero™onversionD the vir—l lo—d in the ˜lood f—lls to — mu™h lower level ™—lled the –set point9
—nd rem—ins rel—tively ™onst—nt during the l—tent ph—se of rs† infe™tion @vewthw—ite —nd
‡ilkinsD PHHSY windel —nd „en—ntEplowersD PHHIY f—nnister et —lD PHHHY €erelson —nd xelsonD
IWWWAF elthoughD rs† —nti˜odies ™ontinue to ˜e dete™t—˜le in the ˜loodD for mu™h of the
l—tent ph—se the infe™ted person looks —nd feels ™ompletely well —nd is not unduly sus™epti˜le
to infe™tions @ƒ—n pr—n™is™o eshƒ pound—tion PHHTY f—nnister et —lFD PHHHY €erelson —nd
xelsonD IWWWAF sn the —˜sen™e of tre—tmentD it h—s ˜een reported th—t the l—tent ph—se of
rs† infe™tion l—sts from IV months to IS ye—rs or more @f—nnister et —lD PHHHY g—ll—w—y et —lFD
IWWWAF e su˜st—nti—l redu™tion in the ™on™entr—tion of ghRC „ ™ells —nd in™re—se in the vir—l
lo—d m—rks the end of the l—tent ph—se —nd the ˜eginning of the symptom—ti™ rs† infe™tion
ph—se @f—nnister et —lD PHHHAF huring the e—rly —nd medium st—ges of the symptom—ti™ rs†
infe™tion ph—seD m—ny people ˜egin to experien™e some mild rs† infe™tion symptomsD su™h
—s skin r—shesD f—tigueD night swe—tsD slight weight lossD mouth ul™ersD —nd fung—l skin —nd
n—il infe™tions @ƒ—n pr—n™is™o eshƒ pound—tion PHHTY €iot —nd gole˜undersD IWVUAF sn the
l—te st—ge of the symptom—ti™ rs† infe™tion ph—se ™—lled eshƒD the d—m—ge to the immune
system is more severeF ghRC „ ™ell ™ount f—lls ˜elow PHH ™ells per mi™ro liter @g—ll—w—y
et —lFD IWWWA —nd vir—l lo—d in the ˜lood re—™hes —˜ove IH HHH genome ™opies per milliliter
@f—nnister et —lD PHHHAF „he over—ll we—kness of the immune system during the l—te st—ge
of the symptom—ti™ rs† infe™tion ph—se —llows the est—˜lishment of opportunisti™ infe™tions
su™h —s €neumo™ystis ™—rinii pneumoni—D wy™o˜—™terium —vium ™omplex dise—seD u—posi9s
s—r™om—D ™ytomeg—lovirusD toxopl—smosisD —nd ™—ndidi—sis whi™h eventu—lly le—d to de—th
@ƒ—n pr—n™is™o eshƒ pound—tion PHHTY f—nnister et —lD PHHHY g—ll—w—y et —lFD IWWWY €iot —nd
gole˜undersD IWVUAF „he ™ourse of rs† infe™tion des™ri˜ed —˜ove is summ—rized in pigure
PFIF




                                              IP
                            pigure PFIX ƒt—ges of rs† infe™tionF

   ƒour™eX httpXGGmslF™sFuiu™FeduG∼yershov—G˜™˜RWSG˜™˜€roje™tEQ•(lesGim—geHISFgifF

                                 e™™essed on PH tune PHHUF

2.2 Global HIV/AIDS situation

ƒin™e the (rst eshƒ ™—ses were reported —mong — few g—y men in the …nited ƒt—tes of
emeri™— qu—rter — ™entury —goD eshƒ h—s grown into — glo˜—l epidemi™ in ne—rly every
™ountry worldwideF e™™ording to …xeshƒ @PHHTAD rs† h—s glo˜—lly now infe™ted more th—n
TS million people of whom —n estim—ted PS million people h—ve diedF sn PHHS —loneD —n
estim—ted PFV million people lost their lives to eshƒD RFI million people were newly infe™ted
with rs† —nd QVFT million people were living with rs† worldwideF


elthough the rs†Geshƒ epidemi™ h—s emerged into — glo˜—l epidemi™D it is in—™™ur—te to

                                             IQ
spe—k of — single glo˜—l rs†Geshƒ epidemi™F „he rs†Geshƒ epidemi™ h—s disproportionE
—tely —'e™ted ™ert—in geogr—phi™ regions @for ex—mpleD ƒƒe —nd the g—ri˜˜e—nA —nd su˜E
popul—tions @for ex—mpleD women in ƒƒeD men who h—ve sex with men @wƒwAD inje™ting drug
users @sh…sAD —nd sex workersAF „hereforeD when —ssessing the glo˜—l rs†Geshƒ situ—tionD
it is import—nt to t—ke this heterogeneity into —™™ountF sn the next p—r—gr—phs we present
— summ—ry of ™urrent rs†Geshƒ trends in sele™ted regions of the world ˜—sed on the PHHT
…xeshƒ ‚eport on the qlo˜—l eshƒ ipidemi™ —nd — report pu˜lished on the eve of the ITth
sntern—tion—l eshƒ gonferen™e @eugust IQEIVD PHHTD in „orontoD g—n—d—A in the wor˜idity
—nd wort—lity ‡eekly ‚eport ˜y ghgF


gurrentlyD ƒƒe is r—ted —s the world9s most severely rs†Geshƒ —'e™ted region ˜e™—use
it h—s just over 10% of the tot—l world9s popul—tion ˜ut it is home to —˜out 64% of —ll
the people living with rs† in the worldF „he rs†Geshƒ epidemi™ is suspe™ted to h—ve
pro˜—˜ly ˜eg—n in ƒƒe —˜out US ye—rs —go —fter some people who —te wild ™himp—nzees in
g—meroon ˜e™—me infe™ted with — lowEvirulen™e progenitor of the virus th—t ™—uses eshƒ
@‡illD PHHTAF „he prim—rily mode through whi™h rs† is tr—nsmitted —mong —dults in this
region is heterosexu—l ™ont—™t @fuvé et —lFD PHHPA —nd —ll the ™ountries ex™ept engol— h—ve
—n estim—ted rs† prev—len™e r—te of more th—n IH per™ent —mong —dults —ged IS to RW ye—rsF
sn PHHSD —n estim—ted QFP million people in the ƒƒe region ˜e™—me newly infe™ted with rs†D
while PFR million people died of eshƒF „he region —lso h—s more rs† infe™ted women th—n
menF sn PHHSD it w—s estim—ted th—t 4.6% of fem—le youth —ged ISEPR ye—rs were living with
rs† ™omp—red to 1.7% of m—les of the s—me —ge groupF


sn ƒƒeD the rs†Geshƒ ™risis is more severe in southern efri™—E the epi™enter of the
rs†Geshƒ epidemi™F sn fotsw—n—D vesothoD ƒw—zil—nd —nd im˜—˜weD the estim—ted —dult
rs† prev—len™e r—tes —re —˜ove 20%F sn ƒouth efri™—D the rs†Geshƒ epidemi™ shows no
eviden™e of —˜—tingF sn PHHSD ƒouth efri™— h—d —n estim—ted rs† prev—len™e r—te of 18.8%
in —dults —ged ISERW ye—rsF st h—s —lso ˜een estim—ted th—t there were SFS million people
living with rs† in ƒouth efri™— in PHHSF „his (gure positioned ƒouth efri™— —s the ™ountry
with the l—rgest num˜er of people living with rs† in the world —long with sndi— in esi—F



                                            IR
hespite rem—ining the world9s severely rs†Geshƒ —'e™ted regionD de™lines in —dult rs†
prev—len™e h—ve re™ently ˜een o˜served in some ™ountries in ƒƒeF „hese ™ountries —re ueny—D
…g—nd—D im˜—˜we —nd ur˜—n —re—s of furkin— p—soF xeverthelessD eshƒ rel—ted de—ths
™ontinue to rise in these ™ountriesF


sn esi—D the rs†Geshƒ epidemi™ is driven ˜y sex workersD sh…s —nd wƒwF istim—tes for
PHHS indi™—te th—t VFQ million people of whom PFR million were —dult women were living
with rs† in esi—F purtherD in esi—D —n estim—ted IVH HHH ™hildren were living with rs†
—nd —pproxim—tely WQH HHH people were newly infe™ted with rs† in PHHSF ƒo f—rD eshƒ h—s
™l—imed —pproxim—tely THH HHH lives in esi—F


„he rs†Geshƒ estim—tes from esi— —lso indi™—te th—t more th—n twoEthirds @—˜out SFU
million peopleA of —ll the people living with rs† in esi— live in sndi—F ‡hile —pproxim—tely
80% of the rs† infe™tions in sndi— —re —™quired through heterosexu—l ™ont—™tD in ghin—
TSH HHH sh…s —™™ount for —˜out h—lf of the people living with rs†D —nd in „h—il—nd —nd
g—m˜odi—D the epidemi™ h—s l—rgely ˜een driven ˜y ™ommer™i—l sex workF sn „h—il—ndD there
h—s ˜een — de™line in rs† prev—len™e in pregn—nt women from 2.4% in IWWS to 1.2% in PHHQF
sn ™ontr—stD rs† prev—len™e —mong wƒw in™re—sed from 17% in PHHQ to 28% in PHHSF


h—t— from i—stern iurope —nd gentr—l esi— indi™—te th—t the rs†Geshƒ epidemi™ there is
—lso ™ontinuing to spre—dF sn PHHS it w—s estim—ted th—t —˜out qu—rter of — million people
were newly infe™ted with rs†F „his (gure ˜rought the tot—l num˜er of people living with
rs† in this region to IFS millionF


sn emeri™—D most of the rs† infe™tions o™™ur —mong wƒwD sh…s —nd sex workersF fr—zilD
the most populous ™ountry in the emeri™—n ™ontinent —fter the …nited ƒt—tes h—d —n
estim—ted —dult rs† prev—len™e of 0.5% —nd w—s home to —pproxim—tely 30% of —ll the
people living with rs† in ƒouth —nd gentr—l emeri™— —nd the g—ri˜˜e—n in PHHSF „he
g—ri˜˜e—n is the se™ond most severely rs†Geshƒ —'e™ted region —fter ƒƒe —nd the epidemi™
there is —lso l—rgely spre—d through heterosexu—l inter™ourseF ‚e™ent d—t— indi™—te th—t
n—tion—l rs† prev—len™e r—tes in —ll the —re—s in the g—ri˜˜e—n ex™ept r—iti h—ve rem—ined


                                            IS
™onst—ntF‚e™entlyD there h—s —lso ˜een —n in™re—se in eviden™e of resurgent epidemi™s in the
…nited ƒt—tes of emeri™— —nd in some ™ountries in iurope —mong wƒwF


„hese st—tisti™s support the rem—rk m—de ˜y ili—s eF erhouni —nd his ™olle—gues from
the x—tion—l snstitute of ellergy —nd snfe™tious hise—ses @xseshA in their st—tement on the
PSth ennivers—ry of the pirst €u˜lished ‚eports of eshƒ on tune SD PHHT th—t eshƒ is the
de—dliest p—ndemi™ of our gener—tion —nd one of the worst in hum—n history @„he fodyD
PHHTAF „he rs†Geshƒ epidemi™ h—s not sp—red —ny regionD r—™eD or so™i—l ™l—ss of people
worldwideF foth the powerful —nd powerless in every so™iety worldwide —re —'e™ted with
this leth—l epidemi™F




2.3 HIV/AIDS situation in Malawi

„he (rst eshƒ ™—ses in w—l—wi were di—gnosed in IWVS @evertD PHHUY wieƒ…‚i iv—lu—tionD
PHHRY ‰oder —nd w—ting—D PHHRY xegD PHHQ—AF roweverD it h—s ˜een —rgued th—t the P7
rs† prev—len™e r—te dis™overed —mong —nten—t—l ™lini™ —ttendees —t ueen iliz—˜eth gentr—l
rospit—l in fl—ntyre in the southern region in th—t ye—r suggests th—t the epidemi™ m—y h—ve
st—rted in w—l—wi —s e—rlier —s IWUU @‰oder —nd w—ting—D PHHRAF


vike elsewhere in ƒƒeD in w—l—wi rs† is predomin—ntly spre—d through heterosexu—l
inter™ourse —mong the —dult popul—tion —nd ˜y verti™—l tr—nsmission to inf—nts from rs†E
infe™ted mothers @xegD PHHQ˜AF huring the (rst de™—de of the rs†Geshƒ epidemi™ in
w—l—wiD prev—len™e —mong women —ttending —nten—t—l ™lini™s @exgA grew r—pidlyF fy IWWVD
PT7 of —nten—t—l ™lini™ —ttendees in ur˜—n —re—s tested rs† positive @wieƒ…‚i iv—lu—tionD
PHHRAF ‚e™ent trends indi™—te th—t the medi—n rs† prev—len™e for —ll —nten—t—l —ttendees
de™lined from PPFV7 in IWWW to ITFW7 in PHHID IUFH7 in PHHQ —nd ISFH7 in PHHS @xegD
PHHSAF elthough rs† prev—len™e r—tes rem—in high in ur˜—n —re—s ™omp—red to rur—l —re—s
@evertD PHHUY ‡ryD PHHS˜AD re™ent trends —lso suggest th—t rs† prev—len™e is de™lining
in m—ny ur˜—n —re—s —nd in™re—sing in m—ny rur—l ones @evertD PHHUY …xeshƒG‡ryD


                                            IT
PHHSAF roweverD n—tion—l estim—tes of rs† prev—len™e in the —dult popul—tion —re reported
to h—ve rem—ined ™onst—nt sin™e IWWT @xegD PHHSAF purthermoreD rs† prev—len™e r—tes
—re p—rti™ul—rly highest —mong women —ged ˜etween IQ —nd PR ye—rs @evertD PHHUY ‡ryD
PHHS˜Y …xeshƒG‡ryD PHHSAF es high levels of movement ˜etween ur˜—nD rur—l —nd mining
—re—s —re import—nt ™—t—lysts for the spre—d of rs† infe™tionD the m—in drivers of the rs†
epidemi™ in w—l—wi —re mo˜ile groupsD whi™h in™lude tru™k driversD sex workersD (shermenD
(sh tr—dersD migr—nt —nd se—son—l workersD milit—ry personnelD prisoners —nd refugees @evertD
PHHUY ‡ryD PHHS˜AF




2.4 HIV/AIDS and child mortality

rs†Geshƒ ™ontri˜utes to ™hild mort—lity ˜oth dire™tly —nd indire™tlyF hire™tlyD rs†Geshƒ
™ontri˜utes to ™hild mort—lity in the sense th—t ™hildren infe™ted with rs† through
w„g„ themselves die of eshƒ @fr—hm˜h—ttD PHHTY re™ht et —lFD PHHTY xewell et —lFD
PHHRY xg9weshemi et —lFD PHHQY edetunjiD PHHHY €e™kh—m —nd qi˜˜D IWWSAF           sndire™tlyD
rs†Geshƒ ™ontri˜utes to ™hild mort—lity in the sense th—t f—milies —nd ™ommunities
we—kened ˜y rs†Geshƒ render ™hildren more sus™epti˜le to illness —nd de—th from other
™—uses @sntern—tion—l eshƒ †—™™ine sniti—tiveD PHHSAF „here —re m—ny possi˜le w—ys through
whi™h rs†Geshƒ indire™tly le—ds to neg—tive he—lth out™omes —nd ™onsequently in™re—sed
mort—lity —mong ™hildren reg—rdless of the ™hildren9s rs† st—tusF ƒome of these w—ys in™lude
neg—tive he—lth imp—™ts em—n—ting from the de—th of or frequent illness of p—rents —nd ™—reE
giversD loss of he—lth workers either through illness —nd de—th from rs†Geshƒ or —ttrition
of new workers from fe—r of infe™tion —nd diversion of resour™es needed for provision of
he—lth ™—re servi™es for ™hildren to pur™h—se the expensive drugs needed for ™—ring those
living with rs†Geshƒ @re™ht et —lFD PHHTY xg9weshemi et —lFD PHHQY edetunjiD PHHHAF ƒin™e
the fo™us of this study is not on the indire™t e'e™ts of rs†Geshƒ on ™hild mort—lityD we
do not present — det—iled review of the m—ny possi˜le s™en—rios through whi™h rs†Geshƒ
indire™tly ™ontri˜utes to ™hild mort—lityF sn the su˜se™tions th—t followD we review v—rious


                                             IU
issues pert—ining to how rs†Geshƒ dire™tly ™ontri˜utes to ™hild mort—lity whi™h in™lude
—v—il—˜le knowledge on —™quisition of rs† infe™tion in ™hildrenD surviv—l of rs† infe™ted
™hildren —nd w—ys of redu™ing rs†Geshƒ —ttri˜ut—˜le ™hild mort—lityF




2.4.1 Acquisition of HIV infection in children

e™™ording to xewell et —lF @PHHRA —nd he go™k et —lF @PHHHAD the epidemiology of rs†Geshƒ
in ™hildren mirrors th—t —mong women of reprodu™tive —geF ghildren get infe™ted with rs†
ne—rly —lw—ys through w„g„ whi™h ™—n o™™ur during pregn—™y @intr—uterine tr—nsmissionAD
delivery @intr—p—rtum tr—nsmissionA —nd the ˜re—stfeeding period @postp—rtum tr—nsmissionA
@xewell et —lFD PHHRY h—˜is et —lFD PHHHY xewellD IWWVAF


„here —re v—rious w—ys in whi™h intr—uterine rs† tr—nsmission ™—n o™™urF e™™ording to
xewell @IWWVAD intr—uterine rs† tr—nsmission m—y o™™ur through pl—™ent—l te—rs th—t would
result in tr—nsfusion of rs† infe™ted ˜lood from the mother to the fet—l ™ir™ul—tion through
either tr—nspl—™ent—l ™ellul—r tr—0™ or ˜y progressive infe™tion of di'erent pl—™ent—l l—yers
until the virus re—™hes the fetopl—™ent—l ™ir™ul—tionF f—sed on reviews ˜y xewell @IWWVA —nd
xewell et —lF @PHHRAD the —v—il—˜le eviden™e th—t supports intr—uterine rs† tr—nsmission
indi™—tes th—t infe™tion is r—re in e—rly pregn—n™y ˜ut more ™ommon in l—te pregn—n™yF
enother review ˜y he go™k et —lF @PHHHA indi™—te th—t in the —˜sen™e of ˜re—stfeedingD
intr—uterine rs† tr—nsmission —™™ounts for 30% of pedi—tri™ rs† infe™tions —nd the rem—ining
70% of inf—nt rs† infe™tions o™™ur during l—˜our —nd deliveryF


„he —v—il—˜le liter—ture on intr—p—rtum tr—nsmission indi™—tes th—t during l—˜our —nd
deliveryD tr—nsmission of rs† infe™tion ™ould o™™ur through dire™t ™ont—™t of the fetusGinf—nt
with rs† infe™ted m—tern—l ˜lood —nd genit—l se™retions during p—ss—ge through the ˜irth
™—n—lD through —s™ending infe™tion from the v—gin— or ™ervix to the fet—l mem˜r—nes
—nd —mnioti™ )uidD through —˜sorption in the fet—lEneon—t—l digestive tr—™tD —nd through
m—tern—lEfet—l mi™rotr—nsfusion during uterine ™ontr—™tions in l—˜our @xewellD IWWVAF



                                              IV
ƒome of the risk f—™tors for intr—uterine —nd intr—p—rtum tr—nsmission identi(ed in the
liter—ture in™lude high m—tern—l vir—l lo—dD low m—tern—l ghRC „ ™ell ™ountsD —dv—n™ed
m—tern—l immune de(™ien™yD long l—˜our —nd prolonged mem˜r—ne rupture @xewell et —lFD
PHHRY he go™k et —lFD PHHHY xewellD IWWVY €e™kh—m —nd qi˜˜D IWWSAF e™™ording to — review
˜y he go™k et —lF @PHHHAD in untre—ted non˜re—stfeeding popul—tionsD ™umul—tive r—tes of
rs† tr—nsmission during the intr—uterine —nd intr—p—rtum periods r—nge from 14% to 32%F


xewell et —lF   @PHHRA report th—t ˜re—stfeeding —pproxim—tely dou˜les the risk of rs†
tr—nsmissionF xewell @IWWVA st—tes th—t tr—nsmission of rs† through ˜re—st milk ™—n o™™ur
in situ—tions where the mother ˜e™omes infe™ted with rs† shortly —fter delivery —s well —s
in est—˜lished m—tern—l rs† infe™tionsF elthough rs† tr—nsmission through ˜re—st milk
™—n o™™ur —t —ny point during l—™t—tionD the empiri™—l eviden™e from epidemiologi™ ™ohort
studies —nd ™lini™—l tri—ls reviewed ˜y powler —nd xewell @PHHPA suggest higher tr—nsmission
risk in the (rst six months of life ™omp—red with low r—tes in the se™ond ye—r of lifeF „he
risk f—™tors for postp—rtum tr—nsmission of rs† infe™tion through ˜re—stfeeding in™lude
m—tern—l vir—l lo—d in pl—sm— —nd ˜re—st milkD rs†Erel—ted m—tern—l immune st—tusD ˜reeding
nipplesD su˜™lini™—l —nd ™lini™—l m—stitis —nd ˜re—st —˜s™esses @…xsgipD …xeshƒD ‡ry —nd
…xp€eD PHHRY powler —nd xewellD PHHPAF st h—s —lso ˜een suggested th—t sero™onversion of
the mother during l—™t—tion in™re—ses the risk of postp—rtum tr—nsmission of rs† infe™tion
through ˜re—stfeeding @he go™k et —lD PHHHAF sn their reviewD he go™k et —lF @PHHHA found
th—t —mong ˜re—stfeeding popul—tions in resour™e poor settings ™umul—tive tr—nsmission r—tes
of rs† infe™tion r—nge from 25% to 48%F




2.4.2 Survival of HIV infected children

„here is limited inform—tion reg—rding the surviv—l st—tus of ™hildren who —™quire rs†
infe™tion from their mothers through w„g„F †ery few ™ommunityE˜—sed studies h—ve
me—sured the rs† st—tus of inf—nts —nd tr—™ked their surviv—l ˜e™—use the persisten™e of
m—tern—l —nti˜odies in the ˜lood of inf—nts m—kes the st—nd—rd ivsƒe rs† tests whi™h —re


                                            IW
used for —dults in—ppropri—te for testing ™hildren —nd —ltern—tive polymer—se ™h—in re—™tion
@€g‚A tests th—t ™—n dete™t the presen™e of the virus @—s opposed to rs† —nti˜odiesA —re
very expensive —nd di0™ult to —dminister outside of — hospit—l setting @—˜— et —lFD PHHQAF


es reported ˜y „he …xeshƒ ‚eferen™e qroup on istim—tesD wodelling —nd €roje™tions
@PHHPAD there is no inform—tion on long term surviv—l of ™hildren infe™ted with rs† vi—
w„g„F wost of the few studies th—t h—ve me—sured the rs† st—tus of inf—nts @for inst—n™eD
— study in fl—ntyreD w—l—wi ˜y „—h— et —lF @PHHHA —nd —nother study in the q—m˜i— ˜y yt—
et —lF @PHHHAA h—ve only tr—™ked their surviv—l up to —round 24 to 36 months of life @„he
…xeshƒ ‚eferen™e qroup on istim—tesD wodelling —nd €roje™tionsD PHHPAF ynly one study
™ondu™ted in uig—liD ‚w—nd— followed rs†Einfe™ted ™hildren until (ve ye—rs of —ge @ƒpir— et
—lFD IWWWAF


efter reviewing pu˜lished —nd unpu˜lished d—t— of some of the studies on surviv—l out™omes
of rs† infe™ted ™hildren whi™h in™luded studies ˜y h—˜is et —lF @PHHIA in gote d9svoire —nd
furkin— p—soD ƒpir— et —lF @IWWWA in ‚w—nd—D vep—ge et —lF @IWWVA in …g—nd—D te—n et —lF
@IWWWA in r—iti —nd fo˜—t et —lF @IWWWA in ƒouth efri™—D „he …xeshƒ ‚eferen™e qroup
on istim—tesD wodelling —nd €roje™tions @PHHPA re™ommended th—t surviv—l times for rs†
infe™ted ™hildren up to 15 ye—rs of life ˜e gener—ted ˜y — dou˜le ‡ei˜ull distri˜ution whi™h
t—kes the formD

              S(y) = 1 − [p(1 − exp(−(β1 y)α1 )) + (1 − p)(1 − exp(−(β2 y)α2 ))]         @PFIA

where S(y) represents the fr—™tion of ™hildren who h—ve not died from rs†Erel—ted ™—uses y
ye—rs —fter ˜irthD p is the proportion of ™hildren who progress to de—th r—pidlyD α1 D β1 D α2
—nd β2 D —re p—r—meters of the ™omponent ‡ei˜ull ™urves des™ri˜ing mort—lity —mongst the
group of ™hildren th—t progress to de—th r—pidly —nd the group of ™hildren th—t progress to
de—th slowly respe™tively @—˜— et —lFD PHHQAF


„he —˜ove formul—tion of the surviv—l ™urve for rs† infe™ted ™hildren —llows for two periods
of very high mort—lityX in inf—n™y when rs† frequently overwhelms the imm—ture immune
systemD —nd —fter —ge nineD when it is —ssumed th—t the few rem—ining survivors will su™™um˜


                                             PH
to opportunisti™ infe™tions —nd progress to eshƒ in the s—me w—y —s —dults @„he …xeshƒ
‚eferen™e qroup on istim—tesD wodelling —nd €roje™tionsD PHHPAF pitting of this ™urve to
the reviewed d—t— suggested th—t ˜y (ve ye—rs —fter ˜irth 60% of rs† infe™ted ™hildren would
˜e de—dF




2.4.3 Reducing under-ve mortality attributable to HIV/AIDS

e™™ording to ‡—lker et —lF @PHHPAD rs†Geshƒ —ttri˜ut—˜le underE(ve mort—lity ™—n ˜e
redu™ed in three m—in w—ysF pirst is redu™tion of rs† prev—len™e in women of ™hild˜e—ring
—geD se™ond is prevention of unw—nted pregn—n™ies in rs†Einfe™ted women —nd third is dire™t
prevention of w„g„ of rs† infe™tion from infe™ted mothers to their inf—ntsF snterventions
designed to redu™e rs† prev—len™e in women of ™hild˜e—ring —ge —re those th—t —re —lso
intended to redu™e the spre—d of rs† in the —dult popul—tion —nd they in™lude m—ss
medi— ™—mp—ignsD promotion of volunt—ry ™ounseling —nd testingD ™ondom so™i—l m—rketingD
improved tre—tment for sexu—lly tr—nsmitted infe™tions —nd pu˜li™ se™tor ™ondom promotion
—nd distri˜ution @tohnson —nd horringtonD PHHTY ƒtover et —lFD PHHPAF €resentlyD the —v—il—˜le
—ppro—™hes to interventions to prevent w„g„ of rs† infe™tion only t—rget the perip—rtum
periodF snterventions demonstr—ted to ˜e e'e™tive in™lude the redu™tion of m—tern—l vir—l
lo—d through —ntiretrovir—l ther—py in pregn—n™y —nd during deliveryD the —void—n™e of
exposure to ™ont—min—ted m—tern—l se™retions through ele™tive ™—es—re—n se™tion deliveryD
—nd the —void—n™e of ˜re—stfeeding @xewell et —lFD PHHRAF „he —v—il—˜le eviden™e indi™—te
th—t w„g„ ™—n ˜e gre—tly redu™ed ˜y the —dministr—tion of zidovudine @h†A or—lly to
the wom—n initi—ted ˜etween IR —nd QR weeks of gest—tionD intr—venously during l—˜our
—nd or—lly to the neon—te in the (rst six weeks of life @h—˜is et —lFD PHHHY he go™k et —lFD
PHHHAF ‚edu™tions in tr—nsmission of one h—lf to one third h—ve ˜een demonstr—ted in tri—ls of
simpli(ed h† regimens in ƒouthe—st esi— —nd ƒƒe @ he go™k et —lFD PHHHAF sntr—p—rtum —nd
neon—t—l singleEdose of xevir—pine h—s —lso ˜een demonstr—ted to —™hieve simil—r redu™tions
in w„g„ —s h† @qu—y et —lFD IWWWAF



                                             PI
2.5 Malawi's response to the HIV/AIDS epidemic and

        eorts to reduce under-ve mortality attributable to

        HIV/AIDS

pollowing the identi(™—tion of the (rst eshƒ ™—sesD the w—l—wi—n government implemented —
num˜er of poli™iesD institution—l —nd oper—tion—l str—tegies —imed —t ™ont—ining the epidemi™
—nd preventing new infe™tionsF „hese in™luded the ˜irth of the x—tion—l eshƒ gontrol
€rogr—mme @xeg€A in IWVW —nd the form—tion of the g—˜inet gommittee on re—lth —nd
rs†Geshƒ to provide poli™y —nd politi™—l dire™tion to the winistry of re—lth @w—l—wi rs†
—nd eshƒ wonitoring —nd iv—lu—tion ‚eport PHHSAF


fetween IWVW —nd IWWQD only two m—jor interventions were implemented in —n e'ort to
stop the spre—d of rs†Geshƒ in w—l—wiF        „hese interventions were — ˜lood s™reening
poli™y —nd — str—tegy for he—lth edu™—tion —˜out rs†Geshƒ rel—ted risks —nd prevention
@wieƒ…‚i iv—lu—tionD PHHRAF es the toll of rs†Geshƒ st—rted to growD w—l—wi re—lized
the need to intensify its e'orts in (ghting the epidemi™F fy IWWQD — multidis™iplin—ry
—ppro—™h th—t in™orpor—ted so™i—lD psy™hologi™—lD —nd e™onomi™ dimensions into the response
to the epidemi™ w—s under developmentF            „he prevention —nd ™ontrol str—tegies of
xeg€ were ˜ro—dened —nd en™omp—ssed —™tivities in surveill—n™eD volunt—ry ™ounseling
—nd testingD homeE˜—sed ™—reD inform—tion —nd edu™—tion ™—mp—ignsD ˜eh—viour—l ™h—nge
™ommuni™—tionsD ™ontrol of sexu—lly tr—nsmitted dise—ses —nd rese—r™h @i˜idAF         xeg€
—lso st—rted to work with other st—keholders su™h —s the priv—te se™torD nongovernment—l
org—niz—tionsD donorsD religious org—niz—tionsD ™ommunityE˜—sed org—niz—tions —nd people
living with rs†Geshƒ @i˜idAF


sn IWWTD —n ev—lu—tion of the n—tion—l response to the rs†Geshƒ epidemi™ ˜y the w—l—wi—n
government —nd its ™oll—˜or—ting p—rtners in the (ght —g—inst the epidemi™ found th—t
—lthough there w—s high —w—reness of rs† —nd eshƒD there w—s little ˜eh—viour ™h—nge


                                             PP
—nd the epidemi™ ™ontinued to grow @w—l—wi rs† —nd eshƒ wonitoring —nd iv—lu—tion
‚eport PHHSAF st w—s therefore re™ommended th—t w—l—wi should develop — ™omprehensive
(veEye—r pl—n to guide rs† —nd eshƒ preventionD tre—tment —nd imp—™t mitig—tionF „his
resulted in the development of the w—l—wi x—tion—l rs†Geshƒ ƒtr—tegi™ pr—mework @xƒpA
in IWWWF „he xƒp ™overed the period PHHH to PHHR —nd fo™used on preventionD —dvo™—™yD —nd
˜eh—vior ™h—ngeY tre—tmentD ™—reD —nd supportY se™tor—l m—instre—mingY imp—™t mitig—tionY
—nd surveill—n™e —nd monitoringF st —lso pl—™ed emph—sis on the need for v—rious st—keholders
in™luding the priv—te se™torD —nd te—™hing institutions to t—ke —n —™tive role in designingD
implementingD —nd monitoring the multiEse™tor—l —nd multidis™iplin—ry rs† —nd eshƒ
interventions in the ™ountry @i˜idAF


„he IWWT ev—lu—tion of the n—tion—l response to the rs†Geshƒ epidemi™ in w—l—wi —lso
reve—led th—t there w—s insu0™ient ™oordin—tion of pl—nningD implement—tionD monitoring
—nd ev—lu—tion of the —™tivities of v—rious —gen™ies involved in the (ght —g—inst the epidemi™
—nd highlighted l—™k of —dequ—te support to xeg€ —nd overEreli—n™e on the he—lth se™tor for
the n—tion—l response @i˜idAF sn responseD with —n —im to improve the multise™tor—l n—tion—l
responseD the w—l—wi—n governmentD together with the st—keholders est—˜lished the x—tion—l
eshƒ gommission in tuly PHHI within the y0™e of the €resident —nd g—˜inet to repl—™e
xeg€F ƒin™e its est—˜lishmentD xeg h—s ˜een ™oordin—ting the multise™tor—l implement—tion
of the n—tion—l response —nd its oper—tions —re monitored ˜y — ˜o—rd of ™ommissioners dr—wn
from governmentD nongovernment—l org—niz—tionsD religious org—niz—tions —nd the priv—te
se™tor @i˜idAF


enother import—nt step in reorg—nising the n—tion—l e'orts to (ght the rs†Geshƒ epidemi™
in w—l—wi w—s the development the n—tion—l eshƒ poli™y in PHHQF „he poli™y stipul—tes the
guiding prin™iples for —ll n—tion—l rs†Geshƒ progr—mmes —nd interventions @‡ryD PHHS˜Y
w—l—wi rs† —nd eshƒ wonitoring —nd iv—lu—tion ‚eport PHHSY xeg PHHQ—AF


es — ™onsequen™e of the multiEse™tor—l —nd multidis™iplin—ry —ppro—™h to the (ght —g—inst
the rs† epidemi™D w—l—wi h—s sever—l pu˜li™D priv—teD ™ommunityE˜—sed —nd f—ithE˜—sed
org—nis—tions involved in v—rious rs†Geshƒ progr—mmes —nd interventions th—t either —im


                                             PQ
—t preventing the further spre—d of rs† infe™tion or mitig—ting the imp—™t of rs†Geshƒ on
the so™ioe™onomi™ st—tus of individu—lsD f—miliesD ™ommunities —nd the n—tion —s stipul—ted in
the x—tion—l rs†Geshƒ poli™yF „hese org—nis—tions h—ve h—d — positive imp—™t in r—ising the
levels of —w—reness —˜out rs†Geshƒ in w—l—wiF pindings from the PHHR w—l—wi hrƒ @xƒy
—nd y‚g w—™roD PHHSA —nd the (rst ˜eh—viour—l surveill—n™e survey ™ondu™ted in w—l—wi in
PHHR @xeg et —lFD PHHRA indi™—te th—t —lmost every one —ged IS ye—rs —nd —˜ove in w—l—wi h—s
he—rd —˜out rs†GeshƒF hespite the univers—l knowledge —˜out the existen™e of rs†Geshƒ
these two n—tion—l surveys found th—t mis™on™eptions —˜out how rs† is tr—nsmitted —nd
how rs† infe™tion ™—n ˜e prevented still exist in w—l—wiF „he (ndings from the two surveys
further indi™—te th—t the v—rious org—nis—tions involved in rs†Geshƒ prevention progr—mmes
h—ve not su˜st—nti—lly imp—™ted on ˜eh—viour ™h—ngeF „he progress —™hieved in —doption of
riskEredu™ing ˜eh—viour su™h —s prem—rit—l sexu—l —˜stinen™eD limiting the num˜er of sexu—l
p—rtners —nd ™onsistent ™ondom use with ™—us—l p—rtners is not very impressiveF


€—rt of the e'orts to mitig—te the imp—™t of rs†Geshƒ in w—l—wi h—s ˜een the provision
of free —ntiretrovir—l tre—tment to rs†Einfe™ted individu—ls in the pu˜li™ se™tor sin™e PHHQ
@‡ryD PHHS˜AF huring the intervening ye—rs huge progress h—s ˜een —™hieved in s™—ling
up tre—tmentF „here h—s ˜een — dr—m—ti™ rise in the num˜er of rs†Einfe™ted individu—ls
re™eiving —ntiretrovir—l tre—tment from R HHH individu—ls ˜y he™em˜erD PHHQ to VI HHH
individu—ls ˜y he™em˜erD PHHT @evertD PHHUY ‡ry et —lD PHHUY ‡ryD PHHS˜AF „he num˜er
of sites providing free —ntiretrovir—l tre—tment in the pu˜li™ se™tor h—s —lso ˜een growing
exponenti—llyF   es of he™em˜erD PHHT there were IHR sites providing free —ntiretrovir—l
tre—tment to eligi˜le rs†Einfe™ted p—tientsF


ep—rt from the rs†Geshƒ prevention progr—mmes —imed —t prim—ry prevention of rs†
infe™tion —mong womenD government —nd some priv—te —nd nonEgovernment—l org—niz—tions
h—ve re™ently ˜e™ome involved in the provision of servi™es —imed —t preventing w„g„ of rs†
infe™tion —s p—rt of e'orts to redu™e underE(ve mort—lity dire™tly —ttri˜ut—˜le to rs†Geshƒ
in w—l—wiF feginning PHHPD rs†Einfe™ted pregn—nt women in w—l—wi h—ve ˜een re™eiving
singleEdose xevir—pine to prevent them from p—ssing the infe™tion to their inf—ntsF e™™ording
to the w—l—wi rs† —nd eshƒ wonitoring —nd iv—lu—tion ‚eport PHHSD the num˜er of sites

                                               PR
providing €w„g„ servi™es in™re—sed from W sites in PHHP to QT sites ˜y end of PHHRF „he
num˜er of rs†Einfe™ted pregn—nt women re™eiving x†€ —lso in™re—sed from P IWV in PHHQ to
P UIW in PHHR —nd S HSR in PHHS @w—l—wi rs† —nd eshƒ wonitoring —nd iv—lu—tion ‚eport
PHHSD winistry of re—lthD vighthouse „rust @vilongweA —nd ghg @w—l—wiAD PHHTAF


w—l—wi9s e'orts to (ght the rs†Geshƒ epidemi™ —re supported ˜y (n—n™i—l —ssist—n™e from
— num˜er of intern—tion—l donors some of whi™h in™lude the ‡orld re—lth yrg—nis—tionD
…xeshƒD the €resident9s imergen™y €l—n for eshƒ ‚elief @€i€pe‚AD the qlo˜—l pund
to pight eshƒD „u˜er™ulosis —nd w—l—ri—D the ‡orld f—nk wultiEgountry rs†Geshƒ
€rogr—mme @we€AD the …nited ƒt—tes egen™y for sntern—tion—l hevelopmentD the …nited
ƒt—tes genters for hise—se gontrol —nd €reventionD the …nited uingdom hep—rtment for
sntern—tion—l hevelopment —nd wede™ins ƒ—n prontiersF


sn summ—ryD w—l—wi h—s ™ome — long w—y in putting in pl—™e str—tegies to ™om˜—t the
rs†Geshƒ epidemi™F ƒever—l interventions —nd progr—mmes h—ve ˜een put in pl—™e to h—lt
the spre—d of the epidemi™ —nd mitig—te its imp—™t in™luding rs†Geshƒ —ttri˜ut—˜le underE
(ve mort—lityF st is o˜vious th—t the imp—™t of the str—tegies —nd interventions re™ently
in™orpor—ted in the response to the rs†Geshƒ epidemi™ on the tr—je™tory of the epidemi™
—nd its imp—™t on the w—l—wi—n popul—tion would ˜e seen in the ye—rs to ™omeF roweverD
there is need to estim—te the extent to whi™h these str—tegies —nd interventions would redu™e
the rs†Geshƒ ˜urden in w—l—wiF st is for this re—son th—t this study —ttempts to ex—mine
the extent to whi™h the ™urrent €w„g„ progr—mmes in w—l—wi would help in re—™hing the
™ountry9s over—ll underE(ve mort—lity redu™tion go—lsF




                                             PS
Chapter 3


Research design and methodology


„his study did not involve ™olle™tion of new d—t— ˜ut used some of the existing indire™t
methods —nd pro™eduresD in™luding m—them—ti™—l models to derive estim—tes of the
proportion of underE(ve mort—lity dire™tly —ttri˜ut—˜le to rs†Geshƒ in w—l—wi during the
period PHHH to PHHRF „o — l—rger extentD the study —dopted the methods —nd pro™edures th—t
h—ve ˜een used ˜y — num˜er of rese—r™hers worldwide —nd re™ommended ˜y …xeshƒ —nd
‡ryF sn this ™h—pter we present det—ils of the methods —nd pro™edures used in this studyF




3.1 Methodology

„his study is su˜divided into seven m—jor p—rtsF „he (rst p—rt involves estim—ting the
prev—len™e of rs† —mong women —ged ISERW in w—l—wi for the period IWWW to PHHRF „he
se™ond p—rt involves estim—ting the num˜er of women in the reprodu™tive —ge @ISERW ye—rsA
in w—l—wi for the period IWWW to PHHRF „he third p—rt involves —pplying the estim—ted
rs† prev—len™e (gures to the estim—ted num˜er of women of reprodu™tive —ge to estim—te
the num˜er of rs† infe™ted womenF „he fourth p—rt involves —pplying estim—tes of —geE


                                           PT
spe™i(™ fertility r—tes to the estim—ted num˜er of women of reprodu™tive —ge to estim—te
the num˜er of ™hildren ˜orn in e—™h ye—rF „he (fth p—rt involves —pplying motherEtoE™hild
tr—nsmission @w„g„A r—tes to the estim—ted num˜er of ™hildren ˜orn to rs† infe™ted women
to estim—te the num˜er of ™hildren infe™ted with rs† through w„g„F „he sixth p—rt involves
—pplying ™hildren9s rs† surviv—l distri˜utions to estim—te the num˜er of ™hildren dying due
to rs†Geshƒ ˜efore their (fth ˜irthd—y in w—l—wi during the period PHHH to PHHRF sn
the l—st p—rtD the estim—ted num˜er of ™hildren dying due to rs†Geshƒ ˜efore their (fth
˜irthd—y during the period PHHH to PHHR is used together with the tot—l num˜er of estim—ted
˜irth from PHHH to PHHR to estim—te the proportion of underE(ve mort—lity th—t is dire™tly
—ttri˜ut—˜le to rs†Geshƒ in w—l—wi for the period PHHH to PHHRF ‡h—t follows —re det—iled
des™riptions of the methods —nd pro™edures th—t h—ve ˜een used in e—™h of the st—ted p—rtsF




3.1.1 Estimating HIV prevalence among women of reproductive age

istim—tes of rs† prev—len™e —mong women of reprodu™tive —ge were l—rgely ˜—sed on exg
sentinel surveill—n™e d—t— ™olle™ted in w—l—wi from IWWW to PHHSF sn developing ™ountries
with gener—lised epidemi™s @de(ned —s — prev—len™e of —t le—st 1% —mong pregn—nt women
—ttending —nten—t—l ™lini™sA like w—l—wiD rs† prev—len™e —mong pregn—nt women is —ssumed
to ˜e — good —pproxim—tion of rs† prev—len™e —mong the gener—l —dult popul—tion @ISERW
ye—rsA @‡ry —nd …xeshƒ PHHQD „he …xeshƒ ‚eferen™e qroup on istim—tesD wodels
—nd €roje™tionsD PHHQAF roweverD exg surveill—n™e systems —re —sso™i—ted with — num˜er
of limit—tionsF   pirstD there is high level of diversity in s—mple sizes —mong the sites
whi™h ™—n result in underEor overErepresent—tion of prev—len™e @‚hu™h—roenpornp—ni™h —nd
gh—mr—trithirongD PHHIAF ƒe™ondD sites ™l—ssi(ed —s rur—l —re lo™—ted in sm—ll towns —nd
l—rgest rur—l settlements with higher levels of e™onomi™ —™tivity —nd mo˜ility —nd pro˜—˜ly
—lso —sso™i—ted with higher rs† prev—len™eD therefore m—y overEestim—te prev—len™e in most
remote —re—s @ƒ™hw—rtl—nder et —lFD IWWWY ‡ry —nd …xeshƒD PHHQY „he …xeshƒ ‚eferen™e
qroup on istim—tesD wodels —nd €roje™tionsD PHHQAF „hirdD sin™e testing is only done to
those women in the gener—l popul—tion who ˜e™ome pregn—ntD the prev—len™e from exg

                                            PU
surveill—n™e m—y overEestim—te the prev—len™e in the gener—l popul—tion —s pregn—nt women
m—y ˜e h—ving unprote™ted sex —t — gre—ter r—te th—n women in the gener—l popul—tionF sn
—dditionD the prev—len™e in exg —ttendees m—y underEestim—te the prev—len™e in the gener—l
popul—tion ˜e™—use women with rs† —sso™i—ted infertility —re not ™—ptured @xƒy —nd y‚g
w—™roD PHHSAF


sn order to minimise some of the st—ted sour™es of ˜i—s th—t —re inherent in the exg
sentinel surveill—n™e systemD some —djustments —re m—de to the exg surveill—n™e d—t— when
estim—ting rs† prev—len™e for —ll —dultsF sn this se™tionD we des™ri˜e the pro™edure th—t w—s
used in this study to —djust the existing exg surveill—n™e d—t— in estim—ting rs† prev—len™e
—mong women of reprodu™tive —ge in w—l—wi from IWWW to PHHSF


e num˜er of steps were involved in estim—ting the rs† prev—len™e —mong women of
reprodu™tive —geF pirstD we —n—lysed —v—il—˜le exg sentinel surveill—n™e d—t— for the ye—rs
IWWW to PHHS to o˜t—in un—djusted —geEspe™i(™ rs† prev—len™e (gures —mong women visiting
exgs in w—l—wi for the ye—rs IWWW to PHHSF elthough the exg sentinel sites in w—l—wi —re
™l—ssi(ed into three ™—tegories @ur˜—nD semiEur˜—n —nd rur—lAD in this study they were only
split into two ™—tegoriesX ur˜—n —nd nonEur˜—n sites @semiEur˜—n —nd rur—l sitesAF „husD
from the —n—lysis we o˜t—ined sep—r—te —ge spe™i(™ rs† prev—len™e (gures for —ggreg—ted
ur˜—n sites —nd —ggreg—ted nonEur˜—n sitesF „hese rs† prev—len™e r—tes were used —s initi—l
estim—tes of rs† prev—len™e in e—™h of the two exg site ™—tegoriesF


xinety (ve per™ent ™on(den™e interv—ls were ™—l™ul—ted for the —ge spe™i(™ rs† prev—len™e
estim—tes for e—™h site ™—tegory o˜t—ined —˜ove using the —djusted ‡—ld method proposed
˜y egresti —nd goull @IWWVAF „he ™hoi™e of this method w—s motiv—ted ˜y the f—™t th—t it is
rel—tively simple to ™ompute —nd h—s ˜een demonstr—ted to provide the ˜est —ver—ge ™over—ge
even for very sm—ll s—mple sizes th—n the other methods of ™—l™ul—ting ˜inomi—l ™on(den™e
interv—ls @egresti —nd goullD IWWVY ƒ—uro —nd vewisD PHHSAF „he —djusted ‡—ld method
uses the usu—l simple ‡—ld formul— ˜ut is 4—djusted4 in th—t it —dds h—lf of the squ—red
Z E™riti™—l v—lue to the numer—tor —nd the entire squ—red ™riti™—l v—lue to the denomin—tor
˜efore ™omputing the interv—lF „husD the —djusted ‡—ld method ™omputes the 100(1 − α)%


                                             PV
™on(den™e interv—l for p —s
                                                       p(1 − p)
                                                       ˆ     ˆ
                                           p ± zα/2
                                           ˆ                                                   @QFIA
                                                          N
                 z2
                  α/2
            x+
where p =
      ˆ          N
                   2
                        D N = n + zα/2 D x denotes — ˜inomi—l v—ri—teD n denotes the s—mple size —nd
                                   2


zc denotes the 1 − c qu—ntile of the st—nd—rd norm—l distri˜ution @egresti —nd goullD IWWVY
ƒ—uro —nd vewisD PHHSAF


ƒe™ondD the initi—l rs† prev—len™e estim—tes were then —djusted for represent—tiveness ˜y
lowering the estim—ted prev—len™e in the —ggreg—ted nonEur˜—n ™lini™ sitesF gonsistent with
the ‡ryG…xeshƒ —ppro—™hD the —djusted estim—te of rs† prev—len™e in the nonEur˜—n
sites —re equ—l to 80% of the initi—l rs† prev—len™e estim—tes o˜t—ined in the (rst step —˜ove
@„he …xeshƒ ‚eferen™e qroup on istim—tesD wodels —nd €roje™tionsD PHHQAF


„hirdD we ™—l™ul—ted n—tion—l estim—tes of —ge spe™i(™ rs† prev—len™e —mong women ˜—sed
on exg surveill—n™e d—t— ˜y —pplying the —ge spe™i(™ fem—le popul—tion size ˜y ur˜—n —nd
rur—l residen™eF „—˜le QFI shows the popul—tion of fem—les of reprodu™tive —ge in ur˜—n —nd
rur—l —re—s ˜y (ve ye—r —ge groups in w—l—wi —s o˜served in the two most re™ent ™ensusesF
 f—sed on „—˜le QFI the per™ent—ge of the fem—le popul—tion th—t w—s in the ur˜—n —re—s in
IWVU —nd IWWV —nd the —nnu—l in™re—se in the ur˜—n popul—tion —ssuming — ™onst—nt —nnu—l
in™re—se during this period is shown in „—˜le QFPF




                                                      PW
            „—˜le QFIX …r˜—n —nd rur—l fem—le popul—tions in w—l—wi

                                IWVU                             IWWV
ege group            …r˜—n             ‚ur—l            …r˜—n           ‚ur—l
ISEIW                RR TQU            QSUWHH           WIIWT           RTVVUS
PHEPR                RT UQH            QPQ QVW          WT HSI          RRU VUI
PSEPW                QV PIV            PUH SVR          TT RHW          QQP IRQ
QHEQR                PR UPT            PHR RWR          RR QVT          PSQ UUS
QSEQW                PH QWV            PHI QHQ          QP QQQ          PIQ RSI
RHERR                II VHQ            IQW UUV          PH IRW          ITH QWQ
RSERW                V VWR             IPU PVU          IS QVP          ISI IIT


                             ƒour™eX xƒy @IWWQY PHHHA




    „—˜le QFPX €er™ent of ur˜—n —nd rur—l fem—le popul—tions in w—l—wi

                                 7 …r˜—n
   ege group             IWVU                IWWV        ennu—l 7 in™re—se
   ISEIW                 IIFI                ITFQ        HFS
   PHEPR                 IPFT                IUFU        HFS
   PSEPW                 IPFR                ITFU        HFR
   QHEQR                 IHFV                IRFW        HFR
   QSEQW                 WFP                 IQFP        HFR
   RHERW                 UFP                 IHFP        HFQ


                         ƒour™eX gomputed ˜y —uthorF




                                        QH
eg—in —ssuming ™onst—nt —nnu—l in™re—sesD the —nnu—l in™re—ses in —ge spe™i(™ ur˜—n fem—le
popul—tion presented in „—˜le QFP were used to ™—l™ul—te estim—tes of the per™ent—ge of
fem—le popul—tion in the ur˜—n —nd the rur—l —re—s in w—l—wi for the ye—rs IWWW to PHHSF
„he ™—l™ul—ted estim—tes of the per™ent—ge of ur˜—n —nd rur—l popul—tions were —pplied to
the rs† prev—len™e (gures in —ggreg—ted ur˜—n sites —nd the —djusted rs† prev—len™e (gures
in —ggreg—ted nonEur˜—n sites to o˜t—in n—tion—l estim—tes of rs† prev—len™e —mong fem—lesF
„he n—tion—l estim—te of rs† prev—len™e for fem—les —ged a in ye—r y were ™—l™ul—ted using
the following formul—X

HIV •N tnf,a,y = (HIV •U bnf,a,y × P rop•U bnf,a,y ) + (HIV •Rl•Adjf,a,y × P rop•Rlf,a,y )
                                                                                        @QFPA
where HIV •N tnf,a,y denotes the estim—ted n—tion—l rs† prev—len™e for fem—les in —ge group
a in ye—r y ˜—sed on exg d—t—D HIV •U bnf,a,y denotes the estim—ted rs† prev—len™e for
fem—les in —ge group a in ye—r y in —ggreg—ted ur˜—n sitesD P rop•U bnf,a,y denotes the
estim—ted per™ent—ge of fem—les in ur˜—n —re—s in —ge group a in ye—r y D HIV •Rl•Adjf,a,y
denotes the —djusted estim—te of rs† prev—len™e for fem—les in —ge group a in ye—r y
in —ggreg—ted nonEur˜—n sites —nd P rop•Rlf,a,y = (100% − P rop•U bnf,a,y ) denotes the
estim—ted per™ent of fem—les in nonEur˜—n —re—s in —ge group a in ye—r y F


ƒin™e IWWWD exg sentinel surveill—n™e surveys were only ™ondu™ted in w—l—wi every two ye—rsF
„hereforeD —fter o˜t—ining the n—tion—l —ge spe™i(™ estim—tes of rs† prev—len™e for fem—les
˜—sed on exg surveill—n™e d—t—D prev—len™e (gures for the ye—rs in whi™h the surveys were
not ™ondu™ted were interpol—ted from these resultsF ege spe™i(™ estim—tes of rs† prev—len™e
for e—™h ye—r in whi™h the survey w—s not done were ™—l™ul—ted —s the —ver—ge of the n—tion—l
—ge spe™i(™ estim—tes of rs† prev—len™e in the pre™eding —nd su™™eeding ye—rF „husD the
formul—X
                                 1
                 HIV •N tnf,a,y = (HIV •N tnf,a,y−1 + HIV •N tnf,a,y+1 )                @QFQA
                                 2
w—s used to estim—te n—tion—l —ge spe™i(™ rs† prev—len™e (gures for fem—les in w—l—wi for
the ye—rs in whi™h exg sentinel surveill—n™e surveys were not ™ondu™tedF


„here—fterD the n—tion—l —ge spe™i(™ estim—tes of rs† prev—len™e for fem—les for PHHR ˜—sed

                                             QI
on the exg sentinel surveill—n™e d—t— were ™omp—red with the —ge spe™i(™ estim—te of
rs† prev—len™e from the PHHR whrƒ —nd —ge spe™i(™ r—tio of exg to gener—l popul—tion
rs† prev—len™e for w—l—wi w—s o˜t—inedF v—stlyD the —ge spe™i(™ r—tio of exg to gener—l
popul—tion rs† prev—len™e w—s used to further —djust the n—tion—l —ge spe™i(™ estim—tes of
rs† prev—len™e for fem—les ˜—sed on the exg surveill—n™e d—t— to o˜t—in (n—l estim—tes of
n—tion—l rs† prev—len™e for fem—les in —ge group a in ye—r y for the ye—rs from IWWW to PHHS
denoted —s HIV •prevalencef ,a,y F




3.1.2 Estimating the number of women of reproductive age

ƒin™e our interest w—s in estim—ting the num˜er of ™hildren th—t died of rs†Geshƒ ˜efore
the —ge of (ve ye—rs during the period PHHH to PHHRD we required the num˜er of ™hildren
˜orn during the period PHHH to PHHR th—t —™quired rs† infe™tion from theirs mothersF „his
ent—iled th—t we —lso needed the num˜er of women of reprodu™tive —ge in those ye—rsF „he
num˜er of women in the ISERW —ge group for the ye—rs in question w—s proje™ted from the
results of the IWWV ™ensus whi™h w—s ™ondu™ted from 1st to 21st ƒeptem˜er IWWVF e™™ording to
the —n—lyti™ report of the IWWV w—l—wi ™ensus @xƒyD PHHPAD —n ev—lu—tion of —ge d—t— showed
th—t the —ge distri˜ution were distorted ˜y —ge misreportingF „hereforeD in this study we
took the —geEsmoothed popul—tion distri˜ution of fem—les —s the ˜—se ye—r popul—tion —nd
then proje™ted this popul—tion using the methodology th—t is used in the m—them—ti™—l
model th—t is implemented in the hem€roj ™omputer progr—mF „his model is ˜—sed on the
st—nd—rd ™ohort ™omponent proje™tion model th—t h—s ˜een modi(ed to produ™e single ye—r
proje™tions @ƒtover —nd uirmeyerD PHHRAF


gonsistent with the methodology in the hem€roj ™omputer progr—mD we st—rted the
proje™tion ˜y sep—r—ting the —geEsmoothed ˜—se ye—r popul—tion in (veEye—r —ge groups
from SEW to RSERW into single ye—rs of —ge using the feers formul—e @i˜idAF „he SEW —ge group
w—s splitted using the following formul—eX

        a5 = 0.0404 ∗ p1 + 0.2000 ∗ p2 − 0.0344 ∗ p3 − 0.0128 ∗ p4 + 0.0068 ∗ p5        @QFRA

                                             QP
         a6 = 0.0093 ∗ p1 + 0.2268 ∗ p2 − 0.0402 ∗ p3 + 0.0028 ∗ p4 + 0.0013 ∗ p5           @QFSA

         a7 = −0.0108 ∗ p1 + 0.2272 ∗ p2 − 0.0248 ∗ p3 + 0.0112 ∗ p4 − 0.0028 ∗ p5          @QFTA

         a8 = −0.0198 ∗ p1 + 0.1992 ∗ p2 + 0.0172 ∗ p3 + 0.0072 ∗ p4 − 0.0038 ∗ p5          @QFUA

         a9 = −0.0191 ∗ p1 + 0.1468 ∗ p2 + 0.0822 ∗ p3 − 0.0084 ∗ p4 − 0.0015 ∗ p5          @QFVA

where p1 D p2 D p3 D p4 —nd p5 refer to the popul—tion —ged HERD SEWD IHEIRD ISEIWD —nd PHEPR
respe™tively —nd a5 D a6 D a7 D a8 —nd a9 refer to the popul—tion —t single —ges SD TD UD VD —nd W
respe™tivelyF


„he —ge groups IHEIR to RSERW were splitted using the following formul—eX

    a1 = −0.0117 ∗ pa−2 + 0.0804 ∗ pa−1 + 0.1570 ∗ pa − 0.0284 ∗ pa+1 + 0.0027 ∗ pa+2@QFWA

                                                                                    @QFIHA
    a2 = −0.0020 ∗ pa−2 + 0.0160 ∗ pa−1 + 0.2200 ∗ pa − 0.0400 ∗ pa+1 + 0.0060 ∗ pa+2

    a3 = 0.0050 ∗ pa−2 − 0.0280 ∗ pa−1 + 0.2460 ∗ pa − 0.0280 ∗ pa+1 + 0.0050 ∗ pa+2 @QFIIA

    a4 = 0.0060 ∗ pa−2 − 0.0400 ∗ pa−1 + 0.2200 ∗ pa + 0.0160 ∗ pa+1 − 0.0020 ∗ pa+2 @QFIPA

    a5 = 0.0027 ∗ pa−2 − 0.0284 ∗ pa−1 + 0.1570 ∗ pa + 0.0804 ∗ pa+1 − 0.0117 ∗ pa+2 @QFIQA

where a1 D a2 D a3 D a4 —nd a5 denote the (rstD se™ondD thirdD fourth —nd (fth —ges in the
p—rti™ul—r —ge group —nd pa+i D (i = {−2, ..., 2}) is the popul—tion of —ge group i groups older
th—n the referen™e groupF


por e—™h ye—rD the methodology in the hem€roj ™omputer progr—m ™—l™ul—tes the popul—tion
of fem—les —ged a using the formul—X
                                                    M igrf,a−1,y−1
                    P opf,a,y =   P opf,a−1,y−1 +                    × Sf,a−1,y−1          @QFIRA
                                                         2
where P opf,a,y is the popul—tion of fem—les —ged a in ye—r y D M igrf,a−1,y−1 is the net num˜er
of migr—nts th—t were —ged a − 1 in ye—r y − 1 —nd Sf,a−1,y−1 is the surviv—l r—tio or proportion
of the fem—le popul—tion —ged a − 1 in ye—r y − 1 th—t survive to —ge a in ye—r y F


roweverD migr—tion d—t— were not ™olle™ted in the IWWV ™ensusF ƒoD like in the proje™tions
done ˜y the w—l—wi x—tion—l ƒt—tisti™—l y0™e for the period IWWW to PHPQD in this study net
migr—tion w—s —ssumed to ˜e zeroF „hereforeD the popul—tion of fem—les —ged a in ye—r y

                                                QQ
w—s ™—l™ul—ted —sX
                             P opf,a,y = P opf,a−1,y−1 × Sf,a−1,y−1                       @QFISA


„he surviv—l r—tios in the model implemented in the hem€roj ™omputer progr—m —re derived
from model life t—˜les @wv„sA whi™hD for —ges older th—n (ve ye—rsD provide (veEye—r surviv—l
r—tiosF e™™ording to ƒtover —nd uirmeyer @PHHRAD these (veEye—r surviv—l r—tios —re ˜y
de(nition the proportion of the popul—tion of — p—rti™ul—r (veEye—r —ge group th—t survives
to the next (veEye—r —ge group (ve ye—rs l—terF sn the model these (veEye—r surviv—l r—tios —re
™onverted to singleEye—r surviv—l r—tios ˜y t—king the (fth root of the (veEye—r surviv—l r—tiosF
„here—fterD the result is used —s the surviv—l r—tio for —ll the (ve —ges in the ™orresponding
—ge groupF „he wv„s used in hem€roj ™omputer progr—m —re the four f—milies of go—leE
hemeny wv„ system @xorthD ƒouthD i—stD —nd ‡estA —nd the (ve f—milies of …nited x—tions
wv„ system @v—tin emeri™—D ghileD ƒouth esi—D i—st esi— —nd qener—lAF


sn the ™—se of w—l—wiD ™hoosing the —ppropri—te wv„ from whi™h to derive the surviv—l r—tios
required in equ—tion QFIS —mong the existing wv„s is — ™h—llenge ˜e™—use there is more th—n
one wv„ th—t is —sserted to (t the w—l—wi—n mort—lity experien™eF xƒy @PHHPA indi™—tes
th—t ™omp—rison —n—lyses ˜etween the reported pro˜—˜ility of dying ˜etween ex—™t —ges x
—nd x + nD n qx D v—lues in the IWWV ™ensus with the ™orresponding v—lues in the (ve f—milies of
the …nited x—tions wv„ system —s well —s in the four f—milies of go—leEhenemy wv„ system
suggested th—t the north f—mily of go—leEhenemy wv„ system w—s more suit—˜le to represent
mort—lity ™onditions in w—l—wiF yn the ™ontr—ryD in prep—ring the n—tion—l estim—tes of
rs†Geshƒ for w—l—wi in PHHQD the w—l—wi x—tion—l eshƒ gommission used the ƒouth f—mily
of go—leEhenemy wv„ system @xegD PHHQ˜AF „he hem€roj ™omputer p—™k—ge —lso provides
the ƒouth f—mily of go—leEhenemy wv„ system —s the def—ult wv„ for w—l—wiF sn —dditionD
there is —lso — set of wv„s for w—l—wi th—t h—ve ˜een ™onstru™ted ˜y ‡ry —nd —re —v—il—˜le
online —t httpXGGwwwFwhoFintGwhosisGd—t—˜—seGlife•t—˜lesGlife•t—˜lesF™fmF purthermoreD
ho™tor @PHHRA reports th—t ™omp—rison —n—lyses of the p—ttern in —ge spe™i(™ de—th r—tes in
w—l—wi during the IWVUEIWWV inter™ens—l period with the ™orresponding —ge spe™i(™ de—th
r—tes from IU model p—tterns of mort—lityX four p—tterns from the region—l wv„s of go—le


                                               QR
—nd hemenyD (ve p—tterns ˜—sed upon the …nited x—tions wv„sD seven p—tterns ˜—sed upon
the sxhi€„r mort—lity p—tterns —nd one p—ttern ˜—sed upon the ‡ry wv„ for PHHH using
d—t— from IWVU —nd IWWV w—l—wi ™ensuses reve—led th—t the ˜estE(tting model for fem—les in
w—l—wi w—s €—ttern Q of sxhi€„r mort—lity p—tternsF


p—™ed with the ™h—llenge of ™hoosing the ˜est wv„ from whi™h to derive the surviv—l r—tios
required in equ—tion QFISD the —ppro—™h —dopted in this study w—s to use surviv—l r—tios
derived from —ll the four wv„s —sserted to (t the w—l—wi—n mort—lity experien™eF ƒin™eD
—™™ording to sxhi€„r xetwork @PHHPAD €—ttern Q of sxhi€„r mort—lity p—tterns is most
simil—r to the p—r i—st f—mily of the …nited x—tions wv„ systemD the …nited x—tions p—r
i—st wv„ w—s used inste—d of €—ttern Q of sxhi€„r mort—lity p—tternsF „he surviv—l
r—tios used in this study were therefore derived from go—leEhemeny northD go—leEhemeny
southD …nited x—tions p—r i—st —nd ‡ry wv„sF ‡hile the ‡ry wv„ is ˜—sed on d—t—
th—t in™lude the in)uen™e of rs†Geshƒ on mort—lityD —s pointed out ˜y sxhi€„r xetwork
@PHHPAD the go—leEhemeny —nd the …nited x—tions f—milies of wv„ ƒystems do not ™ont—in
—n eshƒ p—ttern of mort—lityD therefore go—leEhemeny north —nd go—leEhemeny south wv„s
m—y overestim—te the —nnu—l num˜er of women of reprodu™tive —geF ren™eD the de™ision to
use the four wv„s w—s m—de with —n intention to produ™e —ltern—tive estim—tes of the num˜er
of women of reprodu™tive —ge in w—l—wi th—t in™lude —ll possi˜le s™en—riosF


„he ‡ry wv„ for fem—les in w—l—wi for the period PHHHEPHHR is shown in „—˜le QFQ —nd
the surviv—l r—tios ™orresponding to life expe™t—n™y v—lues of QS to TH for fem—les for the
other three wv„s —re presented in „—˜les QFR to QFTF „hese surviv—l r—tios were o˜t—ined
from (les ™—lled 4Cdnorth.f 4D 4Cdsouth.f 4 —nd 4Unea.f 4 respe™tively th—t —re ™ont—ined in
— folder n—med 4DP4 th—t ™omes —long with the ƒpe™trum €roje™tion €—™k—geF




                                            QS
          „—˜le QFQX ‡ry wv„ for fem—les in w—l—wiD PHHHEPHHR

ege r—nge     n Mx      n qx      lx      n dx    n Lx        Tx      ex
   `I       HFIISVQ   HFIHUIR   IHHHHH   IHUIR   WPSHH      RIRWUIQ   RIFS
   IER      HFHIVWT   HFHUPSS   VWPVT    TRUU    QRISWU     RHSUPIQ   RSFR
   SEW      HFHHTHU   HFHPWVV   VPVHV    PRUR    RHUVST     QUISTIT   RRFW
  IHEIR     HFHHPUR   HFHIQTQ   VHQQR    IHWS    QWVWQS     QQHUUTH   RIFP
  ISEIW     HFHHPVV   HFHIRQ    UWPRH    IIQQ    QWQQTR     PWHVVPS   QTFU
  PHEPR     HFHHVTU   HFHRPRQ   UVIHT    QQIR    QVPPRT     PSISRTI   QPFP
  PSEPW     HFHIVUR   HFHVWSP   URUWP    TTWS    QSUPPQ     PIQQPIT   PVFS
  QHEQR     HFHQHHQ   HFIQWTV   TVHWU    WSIP    QITUHR     IUUSWWQ   PTFI
  QSEQW     HFHQRQV   HFISVPW   SVSVS    WPUQ    PTWURI     IRSWPVW   PRFW
  RHERR     HFHQPHI   HFIRVP    RWQII    UQHV    PPVPVV     IIVWSRV   PRFI
  RSERW     HFHPVVV   HFIQRTU   RPHHR    STSU    IWSVUU     WTIPTH    PPFW
  SHESR     HFHPIQP   HFIHIPP   QTQRU    QTUW    IUPSQV     UTSQVQ    PIFI
  SSESW     HFHPTHT   HFIPPQI   QPTTV    QWWT    ISQQSI     SWPVRS    IVFI
  THETR     HFHPWRP   HFIQUHR   PVTUP    QWPW    IQQSQW     RQWRWR    ISFQ
  TSETW     HFHRPQT   HFIWISQ   PRURQ    RUQW    IIIVTV     QHSWST    IPFR
  UHEUR     HFHTQVR   HFPUSPV   PHHHR    SSHU    VTPSQ      IWRHVV    WFU
  USEUW     HFHWUIP   HFQWHUR   IRRWU    STTS    SVQPS      IHUVQS    UFR
  VHEVR     HFIRTHU   HFSQRWV    VVQQ    RUPS    QPQSH       RWSIH    SFT
  VSEVW     HFPIRSQ   HFTWVP     RIHU    PVTV    IQQTU      IUITH     RFP
  WHEWR     HFQHTSS   HFUWVRS    IPRH     WWH     QPPW       QUWQ     QFI
  WSEWW      HFRPT    HFVSSHV    PSH      PIR         SHI    STR      PFQ
  IHHC      HFSUSUI      I        QT      QT          TQ      TQ      IFU

              ƒour™eX vife „—˜les for ‡ry wem˜er ƒt—tes
    httpXGGwwwFwhoFintGwhosisGd—t—˜—seGlife•t—˜lesGlife•t—˜lesF™fm
                      e™™essed on IH y™to˜erD PHHUF




                                   QT
„—˜le QFRX ƒurviv—l r—tios of fem—les from go—leEhemeny north model life t—˜le

 QSFHHH       RHFHHH        RSFHHH          SHFHHH      SSFHHH        THFHHH
HFVIIWTH     HFVRQHVH      HFVUHPPH        HFVWQWVH    HFWIRSQH      HFWQQUPH
HFWQHVWT     HFWRQRTW      HFWSRRPS        HFWTRUPH    HFWURWUI      HFWVQIVT
HFWSRPWH     HFWTQIIR      HFWUHTIH        HFWUURVP    HFWVQUUI      HFWVVUWI
HFWTQVSU     HFWUIIHH      HFWUUIQV        HFWVPTPP    HFWVUQTV      HFWWIITS
HFWUISPH     HFWUUQWI      HFWVPPRH        HFWVTSTR    HFWWHHWQ      HFWWPWQI
HFUSVUUH     HFUWUUIH      HFVQPRIH        HFVTRIHH    HFVWPQQH      HFWIUVRH
HFVSRUQH     HFVVIVPH      HFWHSIHH        HFWPTSHH    HFWRTTVH      HFWTQPQH
HFWRVQHH     HFWSUUHH      HFWTSVSH        HFWUQPVH    HFWVHPRH      HFWVTHQH
HFWTHWPH     HFWTUTIH      HFWUQRRH        HFWUVSIH    HFWVQQTH      HFWVUTRH
HFWSURSH     HFWTRQWH      HFWUHRRH        HFWUSSQH    HFWVHRWH      HFWVSHHH
HFWSIHPH     HFWSVVTH      HFWTSUHH        HFWUIRPH    HFWUTWTH      HFWVPHTH
HFWRQHPH     HFWSPPRH      HFWTHPVH        HFWTUHTH    HFWUQRUH      HFWUWRHH
HFWQRPIH     HFWRRWRH      HFWSRQHH        HFWTPPTH    HFWTWTHH      HFWUTRHH
HFWPTVVH     HFWQVRWH      HFWRVTPH        HFWSUIUH    HFWTRVWH      HFWUIUHH
HFWPIIWH     HFWQQIPH      HFWRQSQH        HFWSPPSH    HFWSWVSH      HFWTUPQH
HFWIHQHH     HFWPQHTH      HFWQRIWH        HFWRQSPH    HFWSIQQH      HFWSWHWH
HFVVSWQH     HFWHIRWH      HFWISHVH        HFWPTSUH    HFWQSVTH      HFWRSPPH
HFVRQRRH     HFVTRHWH      HFVVPISH        HFVWUSQH    HFWHWTQH      HFWPIWRH
HFUUVHIH     HFVHSIWH      HFVPVWVH        HFVRWHWH    HFVTRVQH      HFVVIHRH
HFTVRPIH     HFUIWQQH      HFUSHIIH        HFUUSWWH    HFUWTHUH      HFVIUHHH
HFSSVSWH     HFTHIPIH      HFTQVTSH        HFTUHIPH    HFTWRHWH      HFUIWQWH
HFRIIISH     HFRSVUTH      HFSHIIUH        HFSQUPQH    HFSTRTTH      HFSWRHRH
HFPQVRQH     HFPTUUUH      HFPWSHIH        HFQIVWTH    HFQQUTWH      HFQSVPQH

                   ƒour™eX ƒpe™trum softw—re version PFQWF
    httpXGGwwwFun—idsForgGenGrs†•d—t—GipidemiologyGepisoftw—reF—spF
                        e™™essed on IS eugustD PHHUF



                                      QU
„—˜le QFSX ƒurviv—l r—tios of fem—les from go—leEhemeny south model life t—˜le

 QSFHHH       RHFHHH        RSFHHH          SHFHHH      SSFHHH        THFHHH
HFVHIQTH     HFVPURVH      HFVRWQUH        HFVTWHQH    HFVVUTRH      HFWHSHWH
HFVVVSTR     HFWHWSTW      HFWPVVHT        HFWRSTWV    HFWTHPHW      HFWUQHQH
HFWRPPSP     HFWSRPPV      HFWTRUIH        HFWUQSSW    HFWVHWPQ      HFWVUIRT
HFWTUTRP     HFWURTUQ      HFWVHTVS        HFWVSTSP    HFWVWUQV      HFWWQHUS
HFWVHWWQ     HFWVSPRQ      HFWVVVIP        HFWWIUQP    HFWWRIHQ      HFWWSWTW
HFUPRRSH     HFUTQRTH      HFUWVRPH        HFVQHPHH    HFVSWSWH      HFVVTSRH
HFVSSQHH     HFVVQUSH      HFWHWHTH        HFWQHSIH    HFWRVTVH      HFWTRSQH
HFWTRSSH     HFWUISSH      HFWUVHWH        HFWVQTHH    HFWVVPVH      HFWWPQQH
HFWTVIPH     HFWURHSH      HFWUWQPH        HFWVRIVH    HFWVVQSH      HFWWIWVH
HFWSVPRH     HFWTSVPH      HFWUPSHH        HFWUVUTH    HFWVRIUH      HFWVVVVH
HFWSIRUH     HFWTHIRH      HFWTUUQH        HFWURVPH    HFWVIHHH      HFWVTRIH
HFWRVRHH     HFWSURWH      HFWTSRPH        HFWUPUSH    HFWUWIWH      HFWVRVWH
HFWRSTUH     HFWSSHVH      HFWTQPQH        HFWUHTUH    HFWUUPWH      HFWVQPIH
HFWRPVPH     HFWSPPSH      HFWTHQHH        HFWTUTTH    HFWURQIH      HFWVHQPH
HFWQWQSH     HFWRVTQH      HFWSTRRH        HFWTQSTH    HFWUHIPH      HFWUTIRH
HFWPUTPH     HFWQUWTH      HFWRTTUH        HFWSRRQH    HFWTIUHH      HFWTVRVH
HFWHQPSH     HFWITRUH      HFWPUTIH        HFWQURQH    HFWRTUPH      HFWSSRQH
HFVSTVUH     HFVUSVUH      HFVWIVWH        HFWHSWUH    HFWIWQUH      HFWQPHHH
HFUVPHPH     HFVHWQRH      HFVQPPRH        HFVSPRPH    HFVUIUSH      HFVWHHWH
HFTUUSIH     HFUIQHTH      HFURPQSH        HFUTVTQH    HFUWRIIH      HFVIVSQH
HFSQHVTH     HFSUQQTH      HFTHUWRH        HFTQWITH    HFTTWVQH      HFTWWTIH
HFQTSSTH     HFRIPPPH      HFRSHSPH        HFRVSQUH    HFSPHPTH      HFSSRURH
HFPIISPH     HFPQVTIH      HFPTPSVH        HFPVRUQH    HFQHUTHH      HFQQHVUH

                   ƒour™eX ƒpe™trum softw—re version PFQWF
    httpXGGwwwFun—idsForgGenGrs†•d—t—GipidemiologyGepisoftw—reF—spF
                        e™™essed on IS eugustD PHHUF



                                      QV
   „—˜le QFTX ƒurviv—l r—tios of fem—les from …x p—r i—st model life t—˜le

 QSFHHH       RHFHHH        RSFHHH          SHFHHH      SSFHHH        THFHHH
HFVSRHUH     HFVUQVIH      HFVWPHHH        HFWHWHHH    HFWPSHPH      HFWRHISH
HFWRTUUQ     HFWSVURR      HFWTVTPI        HFWUTUWW    HFWVQSHQ      HFWVVVWS
HFWTWIPH     HFWUTSTV      HFWVPSRT        HFWVUQQH    HFWWIISP      HFWWRIRW
HFWUWPTQ     HFWVRRHQ      HFWVVRST        HFWWITVR    HFWWRPPP      HFWWTPHP
HFWVRWVV     HFWVVUQV      HFWWITWR        HFWWRHIW    HFWWSVSI      HFWWUPUR
HFVIPRTH     HFVRIURH      HFVTUWVH        HFVWIQSH    HFWIPRHH      HFWQITQH
HFWIPTVH     HFWQPRUH      HFWRVSRH        HFWTPHSH    HFWUQPTH      HFWVPISH
HFWTTTHH     HFWURUVH      HFWVIPSH        HFWVTRHH    HFWWHSHH      HFWWQUHH
HFWSVIUH     HFWTVRHH      HFWUTSPH        HFWVQHHH    HFWVVISH      HFWWPIUH
HFWPTTSH     HFWRRSUH      HFWSVVWH        HFWUHQSH    HFWUWRSH      HFWVTSSH
HFWHVWPH     HFWQHQVH      HFWRUUVH        HFWTIWHH    HFWUQPVH      HFWVPPUH
HFWHHQSH     HFWPPRWH      HFWRHUVH        HFWSSWQH    HFWTVRIH      HFWUVSIH
HFVWRRQH     HFWITHVH      HFWQRQUH        HFWRWVWH    HFWTQHPH      HFWUQWWH
HFVWHHRH     HFWHWWHH      HFWPUIVH        HFWRPQSH    HFWSSTVH      HFWTUQPH
HFVVHIHH     HFVWVTWH      HFWISQVH        HFWQHSRH    HFWRRRHH      HFWSUHSH
HFVSSVUH     HFVUSPPH      HFVWQHUH        HFWHWUTH    HFWPSSIH      HFWRHRHH
HFVIURTH     HFVQVVTH      HFVSWHTH        HFVUVRRH    HFVWUPQH      HFWISSQH
HFUTTUUH     HFUWHQWH      HFVIQPHH        HFVQSTIH    HFVSUWIH      HFVVHPUH
HFUHSRUH     HFUQHQSH      HFUSRWSH        HFUUWUQH    HFVHSHUH      HFVQIPVH
HFTPVPQH     HFTSQURH      HFTUWSUH        HFUHTPSH    HFUQRQIH      HFUTRPRH
HFSQHHRH     HFSSTQRH      HFSVQSWH        HFTIPRRH    HFTRQTHH      HFTUUVRH
HFRHUURH     HFRPWUTH      HFRSPTHH        HFRUTPPH    HFSHITIH      HFSPWUWH
HFPQTSSH     HFPSHSSH      HFPTQUSH        HFPVIIQH    HFPWUUPH      HFQITQRH

                  ƒour™eX ƒpe™trum softw—re version PFQWF
    httpXGGwwwFun—idsForgGenGrs†•d—t—GipidemiologyGepisoftw—reF—spF
                        e™™essed on IS eugustD PHHUF



                                      QW
es des™ri˜ed ˜y ƒtover —nd uirmeyer @PHHSAD the form—t in „—˜les QFR to QFT is —s followsX


   • ‚ow IX life expe™t—n™y —t ˜irth

   • ‚ow PX oneEye—r surviv—l r—tios for new˜orns

   • ‚ow QX oneEye—r surviv—l r—tios for —ges one to two

   • ‚ows RD SD TX oneEye—r surviv—l r—tios for —ges two to threeD three to fourD —nd four to
        (veF

   • ‚ow UX the proportion of ˜irths during — (veEye—r period th—t will survive within the
        HER —ge group to the end of the periodF

   • ‚ow VX the proportion of those —ged HER who will survive into the SEW —ge group (ve
        ye—rs l—terF

   • ‚ows WEPQY the proportion of the —ge group @SEWD IHEIRD· · · D USEVHA th—t will survive
        into the next —ge group (ve ye—rs l—terF

   • ‚ow PRX the proportion of the —ge group 80+ th—t will survive (ve ye—rs l—terF


sn the ™—se of the ‡ry wv„D the surviv—l r—tios used in this study were ™omputed from the

n Lx   ™olumn of „—˜le QFQ where—s for the other three wv„s the surviv—l r—tio v—lues used in
the study were determined ˜—sed on proje™ted inform—tion —˜out expe™t—tion of life —t ˜irth
done ˜y the w—l—wi x—tion—l ƒt—tisti™—l y0™eF „he proje™ted life expe™t—n™y —t ˜irth v—lues
for fem—les for the period under ™onsider—tion in this study —re presented in „—˜le QFUF ƒin™e
the ex—™t life expe™t—n™y —t ˜irth v—lues presented in „—˜le QFU —re not —v—il—˜le in „—˜les
QFR to QFTD line—r interpol—tion w—s used to estim—te the surviv—l r—tios ™orresponding to the
given life expe™t—n™y —t ˜irth v—luesF




                                               RH
         „—˜le QFUX €roje™ted life expe™t—n™y —t ˜irth v—lues for fem—les in w—l—wi


              ‰e—r               IWWV     IWWW      PHHH      PHHI     PHHP    PHHQ    PHHR

   vife expe™t—n™y —t ˜irth     RQFPQ     RQFUU     RRFQQ    RRFWH     RSFRU   RTFHQ   RTFTH

            ƒour™eX httpXGGwwwFnsoFm—l—wiFnetGF e™™essed on IS eugustD PHHUF




3.1.3 Estimating the number of HIV infected women

„he num˜er of women of — p—rti™ul—r —geD aD who —re infe™ted with rs† in ye—r y D w—s
™—l™ul—ted using — formul— simil—r to the one used in the m—them—ti™—l model for m—king
rs†Geshƒ proje™tions whi™h is implemented in — ™omputer progr—m ™—lled eshƒ smp—™t
wodel @eswA @ƒtoverD PHHSAF felow is the formul— th—t w—s used for this t—skF

                         HIVf ,a,y = P opf,a,y × HIV •prevalencef ,a,y                    @QFITA

whereD HIVf ,a,y is the num˜er of rs† infe™ted women in —ge group a in ye—r y D P opf,a,y is the
popul—tion of fem—les —ged a in ye—r y o˜t—ined from equ—tion QFIS —nd HIV •prevalencef ,a,y
is the estim—ted prev—len™e of rs† for fem—les —ged a in ye—r y o˜t—ined in se™tion QFIFIF




3.1.4 Estimating the number of children

„he methodology in the hem€roj ™omputer progr—m w—s —g—in —dopted in ™—l™ul—ting the
num˜er of ™hildren ˜orn to —ll women in e—™h ye—rF por e—™h ye—rD the num˜er of ™hildren
˜orn to fem—les in —ge group a w—s ™—l™ul—ted —sX

                              birthsf ,a,y = P opf,a,y × ASF Rf ,a,y                      @QFIUA

where birthsf ,a,y is the num˜er of ™hildren ˜orn in ye—r y to women of —ge aD P opf,a,y is
the popul—tion of fem—les —ged a in ye—r y o˜t—ined from equ—tion QFIS —nd ASF Rf ,a,y

                                               RI
is the —ge spe™i(™ fertility r—te @eƒp‚A in ye—r y for women of —ge aF rereD sin™e the
˜—se ye—r popul—tion we used w—s re™orded in ƒeptem˜er we —ssume th—t P opf,a,y would ˜e
—pproxim—tely equ—l to the midEye—r popul—tion of fem—les —ged a in ye—r y F


„he eƒp‚s required in equ—tion QFIU were derived from the PHHH —nd PHHR w—l—wi hrƒs
@xƒy —nd y‚g w—™roD PHHI —nd PHHSAF ‡e —ssumed — ™onst—nt —nnu—l ™h—nge in the eƒp‚s
˜etween PHHH —nd PHHR —nd estim—ted the eƒp‚s for the ye—rs from IWWW to PHHR ˜y line—r
interpol—tionF


por e—™h ye—rD —fter ™—l™ul—ting the num˜er of ™hildren ˜orn to women in — p—rti™ul—r —ge
groupD the tot—l num˜er of ™hildren ˜orn to —ll women of reprodu™tive —ge w—s ™—l™ul—ted
using the formul—X
                                             45−49
                                birthsy =             birthsf,a,y                     @QFIVA
                                            a=15−19




3.1.5 Estimating the number of children infected with HIV through

         mother-to-child transmission

‡e —dopted the —ppro—™h ˜y …xeshƒG‡ry in estim—ting the num˜er of ™hildren —™quiring
rs† infe™tion from their mothersF por e—™h ye—rD we st—rted ˜y ™—l™ul—ting the num˜er of
™hildren ˜orn to rs† infe™ted womenF


„wo studies ™ondu™ted in …g—nd— found th—t rs†Einfe™ted women h—d lower fertility r—tes
th—n rs†Eneg—tive womenF yne of the studies w—s done in rur—l ‚—k—i distri™t —nd found
th—t rs† infe™ted women h—d eƒp‚s whi™h were 55% less th—n those of nonErs† infe™ted
women @qr—y et —lFDIWWVAF „he other study w—s ™ondu™ted —mong rur—l women in w—s—k— —nd
found th—t the fertility r—te in rs†Einfe™ted women w—s 26% lower th—n th—t of uninfe™ted
women @g—rpenter et —lFD IWWUAF yn the ˜—sis of the (ndings from these studies it is —ssumed
th—t fertility might ˜e 20 to 50 per™ent lower —mong rs† infe™ted women ™omp—red to
uninfe™ted women @„he …xeshƒ ‚eferen™e qroup on istim—tesD wodels —nd €roje™tionsD


                                              RP
PHHQAF roweverD „he …xeshƒ ‚eferen™e qroup on istim—tesD wodels —nd €roje™tions
@PHHQA —rgues th—t fertility —mong young women who —re rs† positive is likely to ˜e higher
th—n th—t of —ll women sin™e —ll rs† positive women —re sexu—lly —™tive ˜ut not —ll young
women —re sexu—lly —™tiveF „hereforeD in the …xeshƒG‡ry —ppro—™h it is —ssumed th—t
fertility —mong ISEIW ye—r old women is 50% higher —mong rs† positive women th—n rs†
neg—tive women —nd th—t fertility —mong women —ged PHERW is 20% lower —mong rs† positive
women th—n rs† neg—tive women @„he …xeshƒ ‚eferen™e qroup on istim—tesD wodels —nd
€roje™tionsD PHHQAF


vet Bf,a,y denote the num˜er of ™hildren ˜orn to rs†Epositive women —ged a ye—rs in ye—r
     +


y D B − f,a,y denote the num˜er of ™hildren ˜orn to rs†Eneg—tive women —ged a ye—rs in ye—r
y —nd N egf,a,y denote the midEye—r popul—tion of rs†Eneg—tive women —ged a ye—rs in ye—r
y F „hen using equ—tions QFISD QFIT —nd QFIU we h—veX

                                        +        −
                        birthsf ,a,y = Bf,a,y + Bf,a,y                               @QFIWA

                         P opf,a,y = HIVf ,a,y + N egf,a,y                           @QFPHA
                                                      +       −
                                      Bf,a,y        Bf,a,y + Bf,a,y
                       ASF Rf ,a,y =           =                                     @QFPIA
                                     P opf,a,y   HIVf ,a,y + N egf,a,y
prom equ—tion QFPI we o˜t—in

                             +        −
                            Bf,a,y + Bf,a,y = P opf,a,y × ASF Rf ,a,y                @QFPPA

hividing ˜oth sides of equ—tion QFPP ˜y N egf,a,y yields
                           +              −
                          Bf,a,y         Bf,a,y         P opf,a,y
                                    +               =             × ASF Rf ,a,y      @QFPQA
                        N egf,a,y       N egf,a,y       N egf,a,y


xow —ssuming th—t fertility —mong ISEIW ye—rs old women is 50% higher —mong rs†Epositive
th—n rs†Eneg—tive women we h—ve
                                      +                           −
                                     Bf,a,y                      Bf,a,y
                                                    = 1.5 ×                          @QFPRA
                                    HIVf ,a,y                  N egf,a,y
                                      −                        +
                                     Bf,a,y                   Bf,a,y
                               ⇒                    =                                @QFPSA
                                    N egf,a,y             1.5 × HIVf ,a,y
where a = 15 − 19F

                                                     RQ
ƒu˜stituting equ—tion QFPS in equ—tion QFPQ yields
                        +                   +
                       Bf,a,y              Bf,a,y                   P opf,a,y
                                  +                        =                  × ASF Rf ,a,y                 @QFPTA
                      N egf,a,y  1.5 × HIVf ,a,y                    N egf,a,y
               +
              Bf,a,y (1.5 × HIVf ,a,y + N egf,a,y )          P opf,a,y
       ⇒                                                   =           × ASF Rf ,a,y                   @QFPUA
                  1.5 × HIVf ,a,y × N egf,a,y                N egf,a,y
                                                             1.5 × HIVf ,a,y × P opf,a,y × ASF Rf ,a,y
                                                 +
                                              ⇒ Bf,a,y     =                                           @QFPVA
                                                                     1.5 × HIVf ,a,y + N egf,a,y

ƒimil—rlyD —ssuming th—t fertility —mong women PHERW ye—rs old is 20% lower —mong rs†E
positive women th—n rs†Eneg—tive women we h—ve
                                                +                           −
                                               Bf,a,y                      Bf,a,y
                                                           = 0.8 ×                                          @QFPWA
                                              HIVf ,a,y                   N egf,a,y
                                                −                            +
                                               Bf,a,y               1.25 × Bf,a,y
                                          ⇒                =                                                @QFQHA
                                              N egf,a,y               HIVf ,a,y
where a = {20 − 24, ..., 44 − 49}F


ƒu˜stituting equ—tion QFQH in equ—tion QFPQ gives
                                 +                   +
                                Bf,a,y       1.25 × Bf,a,y             P opf,a,y
                                         +                      =                × ASF Rf ,a,y              @QFQIA
                          N egf,a,y      HIVf ,a,y                     N egf,a,y
                 +
                Bf,a,y (HIVf ,a,y + 1.25 × N egf,a,y )            P opf,a,y
          ⇒                                                     =           × ASF Rf ,a,y                   @QFQPA
                        HIVf ,a,y × N egf,a,y                     N egf,a,y
                                                                  HIVf ,a,y × P opf,a,y × ASF Rf ,a,y
                                                       +
                                                    ⇒ Bf,a,y    =                                           @QFQQA
                                                                     HIVf ,a,y + 1.25 × N egf,a,y

gom˜ining the inform—tion in equ—tions QFPV —nd QFQQD the num˜er of ™hildren ˜orn to rs†
infe™ted mothers w—s ™omputed using the formul—X
                       1.5×HIV ×P op ×ASF R
                                f ,a,y    f,a,y       f ,a,y
                      
                             1.5×HIVf ,a,y +N egf,a,y
                                                                                      a = {15 − 19}
   HIV •birthsf,a,y =                                                                                       @QFQRA
                       HIVf ,a,y ×P opf,a,y ×ASF Rf ,a,y                     a = {20 − 24, ..., 45 − 49}
                                         HIVf ,a,y +1.25×N egf,a,y

where HIV •birthsf,a,y denotes the num˜er of ™hildren ˜orn to rs†Einfe™ted women —ged a
in ye—r y F


efter ™—l™ul—ting the num˜er of ™hildren ˜orn to rs† infe™ted women in e—™h —ge groupD the
formul—
                                                            45−49
                                   HIV •birthsy =                     HIV •birthsf,a,y                      @QFQSA
                                                          a=15−19


                                                               RR
w—s employed to ™—l™ul—te the tot—l num˜er of ™hildren ˜orn to —ll rs† infe™ted women in —
given ye—r y F


pin—llyD the num˜er of ™hildren infe™ted with rs† through w„g„ in —ny given ye—r w—s
™—l™ul—ted using the formul—

                         HIV + •birthsy = HIV •birthsy × P T R                         @QFQTA

where P T R denotes the perin—t—l tr—nsmission r—teD whi™h is de(ned —s the per™ent—ge of
™hildren ˜orn to rs†Einfe™ted mothers who —re infe™ted themselvesF


st h—s ˜een do™umented th—t ™hildren —™quire rs† infe™tions from their mothers during
gest—tion or ˜irth or —fter ˜irth through ˜re—stfeeding @„he …xeshƒ ‚eferen™e qroup on
istim—tesD wodels —nd €roje™tionsD PHHQAF f—sed on — review of studies on w„g„D the
…xeshƒ ‚eferen™e qroup on istim—tesD wodels —nd €roje™tions reports th—t in developing
™ountries where — signi(™—nt —mount of tr—nsmission m—y o™™ur through ˜re—stfeedingD the
€„‚ lies ˜etween 25% —nd 48% in the —˜sen™e of progr—ms to prevent w„g„ su™h —s
tre—tment with nevir—pine @x†€AD or zidovudine @h†A —nd x†€D or some other tre—tment or
inf—nt feeding options @„he …xeshƒ ‚eferen™e qroup on istim—tesD wodels —nd €roje™tionsD
PHHQY ƒtoverD PHHSAF „he …xeshƒG‡ry —ppro—™h is to use — v—lue of 32% for the €„‚
in ƒƒe ™ountries @„he …xeshƒ ‚eferen™e qroup on istim—tesD wodels —nd €roje™tionsD
PHHQY ƒtoverD PHHSAF yther studies h—ve used — r—te of 30% @edetunjiD PHHHA or 35% @—˜—
et —lFD PHHQA for ™ountries in ƒƒeF sn this study we used —ll the three proposed tr—nsmission
r—tes @iFeFD 30%D 32% —nd 35%A inorder to gener—te —ltern—tive estim—tes of the num˜er of
™hildren infe™ted with rs† vi— w„g„ in the —˜s™en™e of interventions to prevent w„g„F
yur —ssumption is th—t these three tr—nsmission r—tes —re equ—lly likely to —rise in w—l—wiF


feginning PHHPD €w„g„ servi™es h—ve ˜een provided to — ™ert—in proportion of rs† infe™ted
pregn—nt women in w—l—wiF e™™ording to the w—l—wi rs† —nd eshƒ wonitoring —nd
iv—lu—tion ‚eport PHHSD PIWV —nd PUIW rs† positive pregn—nt women re™eived x†€ in
PHHQ —nd PHHR respe™tivelyF ‡e did not (nd the ex—™t num˜er of rs† infe™ted women th—t
re™eived x†€ in PHHPF roweverD we —ssume it to ˜e —round VRHD whi™h is the num˜er of


                                             RS
pregn—nt women th—t tested positive in f—™ilities providing €w„g„ servi™es @winistry of
re—lthD vighthouse „rust @vilongweA —nd ghg @w—l—wiAD PHHTAF sn our (n—l estim—tes of
the num˜er of ™hildren infe™ted with rs† vi— w„g„ in w—l—wi we took into —™™ount the
rs† infe™ted pregn—nt women th—t re™eived x†€ —nd used the def—ult tr—nsmission r—tes
from esw for women re™eiving x†€ presented in „—˜le QFV ˜elowF

          „—˜le QFVX €„‚ for rs† infe™ted pregn—nt women re™eiving xevir—pineF


        f—se tr—nsmission r—te @no interventionA            30%        32%         35%

        „r—nsmission r—te with intervention @x†€A           16%        24%         28%


                          ƒour™eX ƒpe™trum softw—re version PFQWF

           httpXGGwwwFun—idsForgGenGrs†•d—t—GipidemiologyGepisoftw—reF—spF

                                e™™essed on IS eugustD PHHUF




3.1.6 Estimating the number of HIV infected children dying before

         the age of ve years

st h—s ˜een reported th—t ™hildren who —re infe™ted with rs† vi— w„g„ gener—lly progress
to eshƒ f—ster th—n —dults @„he …xeshƒ ‚eferen™e qroup on istim—tesD wodels —nd
€roje™tionsD PHHQAF „he medi—n time from ˜irth to eshƒ h—ve ˜een reported to r—nge from
one ye—r to 6.3 ye—rs ˜—sed on — num˜er of studies ™ited ˜y „he …xeshƒ ‚eferen™e qroup
on istim—tesD wodels —nd €roje™tions @PHHQAF st h—s ˜een reported th—t sever—l of those
studies found th—t some ™hildren @—˜out 40%A progress to eshƒ within — few months while
the rest t—ke ™onsider—˜ly longerF e …xeshƒ review of the —v—il—˜le eviden™e suggests th—t
the surviv—l of the infe™ted ™hildren is ˜est des™ri˜ed ˜y — r—pid progression from infe™tion to
de—th for some ™hildren —nd mu™h slower progression for others @…xeshƒD PHHI ™ited ˜y „he

                                              RT
…xeshƒ ‚eferen™e qroup on istim—tesD wodels —nd €roje™tionsD PHHQAF yn the ˜—sis of
this —v—il—˜le eviden™eD „he …xeshƒ ‚eferen™e qroup on istim—tesD wodels —nd €roje™tions
@PHHPA h—s developed pro™edures for estim—ting the proportion of ™hildren who —re infe™ted
with rs† vi— w„g„ who will die from rs†Erel—ted ™—uses y ye—rs —fter ˜irthF „his study
—dopted those pro™edures to estim—te the num˜er of ™hildren dying due to rs†Geshƒ ˜efore
their (fth ˜irthd—y in w—l—wi during the period PHHH to PHHRF


e™™ording to „he …xeshƒ ‚eferen™e qroup on istim—tesD wodels —nd €roje™tions9
pro™eduresD the sh—pe of the surviv—l ™urve for rs†Einfe™ted ™hildren up to the —ge of 15
is spe™i(ed ˜y the following dou˜le ‡ei˜ull fun™tionX

              S(y) = 1 − [p(1 − exp(−(β1 y)α1 )) + (1 − p)(1 − exp(−(β2 y)α2 ))]           @QFQUA

where S(y) represents the fr—™tion of ™hildren who h—ve not died from rs†Erel—ted ™—uses y
ye—rs —fter ˜irthD p is the proportion of ™hildren who progress to de—th r—pidlyD α1 D β1 D α2 —nd
β2 D —re p—r—meters of the ™omponent ‡ei˜ull ™urves des™ri˜ing mort—lity —mongst the group
of ™hildren th—t progress to de—th r—pidly —nd the group of ™hildren th—t progress to de—th
slowly respe™tively @—˜— et —lFD PHHQAF en —n—lysis ™ondu™ted ˜y „he …xeshƒ ‚eferen™e
qroup on istim—tes wodels —nd €roje™tions @PHHPA found th—t the p—r—meter v—lues th—t
g—ve the ˜est (t to — wide r—nge of net surviv—l d—t— me—sured in ™lini™ —nd ™ommunity
˜—sed ™ohorts were p = 0.6D α1 = 0.9D β1 = 0.9D α2 = 10.0 —nd β2 = 0.1F „hese p—r—meter
v—lues —re the ones th—t were used in this study in estim—ting the num˜er of rs†Einfe™ted
™hildren who would die ˜efore the —ge of (ve ye—rs in w—l—wi during the period in questionF
pigure QFI shows the ™urve of the ™umul—tive proportion of the rs†Einfe™ted ™hildren th—t
progress to eshƒ de—th y ye—rs —fter ˜irth using the st—ted p—r—metersF „he ™urve indi™—tes
th—t over — third of infe™ted ™hildren would die ˜efore their (rst ˜irthd—yD ne—rly two thirds
˜y —ge (ve —nd none of them would survive p—st the —ge of ISF




                                               RU
                              90 100
                              80
                              70
       Percentage surviving

                              60
                              50
                              40
                              30
                              20
                              10
                              0




                                         0    1    2    3    4    5        6   7     8   9   10   11   12   13

                                                                      Age of child




                                       pigure QFIX ƒurviv—l distri˜ution of rs†Einfe™ted ™hildrenF

  ƒour™eX „he …xeshƒ ‚eferen™e qroup on istim—tes wodels —nd €roje™tions @PHHPAF

3.1.7 Estimating the proportion of under-ve mortality in Malawi

                         during the period 2000 to 2004 that is directly attributable

                         to HIV/AIDS

‡e —dopted the —ppro—™h ˜y ‡—lker et —lF @PHHPA to move from the estim—tes of the tot—l
num˜er of rs† infe™ted ™hildren dying ˜efore the —ge of (ve ye—rs during the period PHHH
to PHHR ™omputed ˜y the pro™edures des™ri˜ed —˜ove to estim—tes of the proportion of
underE(ve mort—lity dire™tly —ttri˜ut—˜le to rs†Geshƒ in w—l—wi from PHHH to PHHRF pirstD
we ™omputed estim—tes of …Sw‚ in rs†Einfe™ted ™hildren ˜y dividing the estim—ted tot—l
num˜er of rs† infe™ted ™hildren dying ˜efore the —ge of (ve ye—rs from PHHH to PHHR ˜y
the estim—ted tot—l num˜er of ™hildren infe™ted with rs† through w„g„ from PHHH to


                                                                      RV
PHHRF ƒe™ondD we ™omputed estim—tes of …Sw‚ for rs†Eneg—tive ™hildren ˜y su˜tr—™ting the
estim—tes of …Sw‚ in rs†Einfe™ted ™hildren from the PHHR w—l—wi hrƒ estim—te of …Sw‚
from —ll ™—usesF „hirdD we used the ™omputed estim—tes of …Sw‚ without rs†Geshƒ to
™ompute estim—tes of the tot—l num˜er of rs† infe™ted ™hildren dying ˜efore the —ge of (ve
ye—rs from nonErs† ™—uses from PHHH to PHHR sin™e rs† infe™ted ™hildren ™—n —lso die from
nonErs† rel—ted ™—uses @‡—lker et —lFD PHHPAF pourthD we su˜tr—™ted estim—tes of the tot—l
num˜er of rs† infe™ted ™hildren dying ˜efore the —ge of (ve ye—rs from nonErs† ™—uses
from PHHH to PHHR from the estim—tes of tot—l num˜er of rs† infe™ted ™hildren dying ˜efore
the —ge of (ve ye—rs from PHHH to PHHR to o˜t—in estim—tes of the tot—l num˜er of rs†
infe™ted ™hildren dying ˜efore the —ge of (ve ye—rs from rs† rel—ted ™—uses from PHHH to
PHHRF pifthD we ™omputed estim—tes of …Sw‚ due to rs†Geshƒ from PHHH to PHHR ˜y
dividing the estim—tes of the tot—l num˜er of rs† infe™ted ™hildren dying ˜efore the —ge
of (ve ye—rs from rs† rel—ted ™—uses from PHHH to PHHR ˜y the estim—ted tot—l num˜er
of ™hildren ˜orn to —ll women from PHHH to PHHRF v—stlyD we ™omputed estim—tes of the
proportion of …Sw‚ —ttri˜ut—˜le to rs†Geshƒ in w—l—wi during the period PHHH to PHHR
˜y dividing the estim—tes of …Sw‚ due to rs†Geshƒ from PHHH to PHHR ˜y the PHHR w—l—wi
hrƒ estim—tes of …Sw‚F




3.2 Data sources

„he study m—inly used se™ond—ry d—t— ™olle™ted ˜y government dep—rtments —nd other
institutions working on rs†Geshƒ rel—ted issues in w—l—wi —s well —s d—t— from relev—nt
pu˜lished studies —nd reportsF felow we present the sour™es of d—t— whi™h formed the ˜—sis
of the rs†Geshƒ —nd demogr—phi™ estim—tes —nd proje™tions done in this studyF




                                           RW
3.2.1 HIV prevalence data

…ntil re™entlyD the most widely —v—il—˜le sour™e of epidemiologi™—l d—t— on rs† prev—len™e
in w—l—wi h—s ˜een — routine exg surveill—n™e systemF ƒin™e IWWRD IW sentinel exg sites
lo™—ted in the ur˜—n —nd the rur—l —re—s in —ll the three regions @xorthD gentr—lD —nd ƒouthA
h—ve ˜een used to ™olle™t d—t— on rs† prev—len™e every one or two ye—rs using ™onsistent
methodology in the s—me popul—tion group @wieƒ…‚i iv—lu—tionD PHHRY xƒy —nd w—™ro
sntern—tion—l sn™FD PHHSAF „hereforeD d—t— for setting —ssumptions —˜out the prev—len™e of
rs† —mong women in e—™h —ge group from —ge IS to —ge RW from IWWW to PHHS were o˜t—ined
from the ipidemiology hep—rtment of the w—l—wi winistry of re—lth —nd €opul—tionF




3.2.2 Demographic data

„he IWWV w—l—wi €opul—tion —nd rousing gensus en—lyti™ ‚eport @xƒyD PHHPA w—s the
sour™e of the ˜—se ye—r popul—tion for estim—ting the num˜er of women of reprodu™tive —ge
for e—™h ye—r from IWWW to PHHRF snform—tion on eƒp‚s w—s o˜t—ined from the PHHH —nd
PHHR w—l—wi hrƒsF




3.3 Software

wost of the ™omput—tions needed in this study were done using ‚ version PFRFIF ‚ is — freely
—v—il—˜le l—ngu—ge —nd environment for st—tisti™—l ™omputing —nd gr—phi™s whi™h provides
— wide v—riety of st—tisti™—l —nd gr—phi™—l te™hniquesX line—r —nd nonline—r modellingD
st—tisti™—l testsD time series —n—lysisD ™l—ssi(™—tionD ™lusteringD —nd m—ny more othersF
wore inform—tion —˜out ‚ —nd instru™tion for downlo—ding — ™opy is —v—il—˜le on the ‚
home p—ge @httpXGGwwwFrEproje™tForgAF      en—lysis of the rs† prev—len™e d—t— w—s done
                                                           A
using ƒ€ƒƒ version IRF „his thesis do™ument w—s typeset in v „iˆ using wiu„eˆ PFS


                                             SH
inst—ll—tion downlo—ded from httpXGGsour™eforgeFnetGproje™tsGmiktexG —nd „eˆni™genter
editor downlo—ded from httpXGGsour™eforgeFnetGproje™tsGtexni™™enterGF




                                           SI
Chapter 4


Results


„his ™h—pter presents estim—tes of the proportion of underE(ve mort—lity dire™tly —ttri˜uted
to rs†Geshƒ in w—l—wiF pirstD the ™h—pter presents estim—tes of rs† prev—len™e —mong
women of reprodu™tive —ge from IWWW to PHHSF ƒe™ondD it presents estim—tes of the num˜er
of women of reprodu™tive —ge in w—l—wi from IWWW to PHHRF sn the third p—rt the ™h—pter
presents estim—tes of the num˜er of women of reprodu™tive —ge infe™ted with rs†F „he fourth
—nd (fth p—rts present estim—tes of the num˜er of ™hildren ˜orn to —ll women —nd estim—tes
of the num˜er of ™hildren infe™ted with rs† through w„g„ respe™tivelyF „he sixth p—rt
presents estim—tes of the num˜er of rs† infe™ted ™hildren dying ˜efore the —ge of (ve ye—rsF
et the endD the ™h—pter presents estim—tes of the proportion of underE(ve mort—lity dire™tly
—ttri˜uted to rs†Geshƒ in w—l—wi during the period PHHH to PHHRF




                                            SP
4.1 Estimates of HIV prevalence among women of

         reproductive age

sn this se™tion we present estim—tes of rs† prev—len™e —mong women of reprodu™tive —ge in
w—l—wi from IWWW to PHHSF ‡e present exg sentinel surveill—n™e results (rst —nd end the
se™tion with (n—l estim—tes of rs† prev—len™e —mong women of reprodu™tive —ge in w—l—wiF




4.1.1 ANC sentinel surveillance sites in Malawi, 1999 to 2005

ƒin™e IWWWD exg rs† surveill—n™e in w—l—wi h—s ˜een ™ondu™ted every two ye—rs in IW sites
in —ll the three regions of the ™ountry shown in pigure RFIF iight of the sites —re ™l—ssi(ed
—s rur—l ™lini™sD iight —s semiEur˜—n ™lini™s —nd three —s ur˜—n ™lini™s @see „—˜le RFIAF es
st—ted e—rlierD in this study we re™l—ssi(ed the exg sentinel sites into ur˜—n —nd nonEur˜—n
sitesF f—sed on our ™l—ssi(™—tion of the sentinel sitesD the num˜ers of women s—mpled in the
exg surveill—n™e surveys ˜y site ™—tegory in w—l—wi from IWWW to PHHS —re shown in „—˜le
RFPF elthough „—˜le RFP shows th—t the tot—l num˜ers of women s—mpled in nonEur˜—n sites
were ™onsistently more th—n the tot—l num˜ers of women s—mpled in ur˜—n sitesD the women
s—mpled in the former were predomin—ntly from semiEur˜—n sitesF por inst—n™e in PHHQ the
69.2% of women s—mpled in the nonEur˜—n sites ™omprised 48.9% of women from semiEur˜—n
sitesF




                                             SQ
          pigure RFIX exg sentinel surveill—n™e sites in w—l—wiD IWWW to PHHS


                                ƒour™eX xeg @PHHQ˜A

xoteX vim˜e re—lth gentre —nd wzuzu re—lth gentre repl—™ed ueen iliz—˜eth gentr—l

rospit—l —nd ƒtF tohn9s wission rospit—l respe™tively whi™h were used in IWWW —nd PHHIF

                                          SR
„—˜le RFIX gl—ssi(™—tion of exg sentinel surveill—n™e sites in w—l—wiD IWWW to PHHSF


      ƒentinel site                            ‚egion            g—tegory

      u—poro re—lth gentre                     xorth             ‚ur—l

      w˜—l—™h—nd— re—lth ™entre                xorth             ‚ur—l

      ‚umphi histri™t rospit—l                 xorth             ƒemiEur˜—n

      xkh—t—Ef—y histri™t rospit—l             xorth             ƒemiEur˜—n

      wzuzu re—lth gentre                      xorth             …r˜—n

      u—m˜oni re—lth gentre                    gentre            ‚ur—l

      „honje re—lth gentre                     gentre            ‚ur—l

      u—sin— re—lth gentre                     gentre            ‚ur—l

      w™hinji histri™t rospit—l                gentre            ƒemiEur˜—n

      ƒtF enne9s wission rospit—l              gentre            ƒemiEur˜—n

      xt™heu histri™t rospit—l                 gentre            ƒemiEur˜—n

      vilongwe gentr—l rospit—l                gentre            …r˜—n

      q—w—n—ni re—lth gentre                   ƒouth             ‚ur—l

      wilep— re—lth gentre                     ƒouth             ‚ur—l

      wi—ng— re—lth gentre                     ƒouth             ‚ur—l

      w—ngo™hi histri™t rospit—l               ƒouth             ƒemiEur˜—n

      wul—nje wission rospit—l                 ƒouth             ƒemiEur˜—n

      xs—nje histri™t rospit—l                 ƒouth             ƒemiEur˜—n

      vim˜e re—lth gentre                      ƒouth             …r˜—n

                                 ƒour™eX xeg @PHHQ™A


                                         SS
„—˜le RFPX „rends in the num˜er of women s—mpled in exg sentinel surveill—n™e surveys in

w—l—wi from IWWW to PHHS ˜y site ™—tegoryF


  ƒentinel site           IWWW                  PHHI                  PHHQ                  PHHS

  ™—tegory        xum˜er         7      xum˜er         7      xum˜er         7      xum˜er         7

  …r˜—n           I WHI          PUFT   I VSQ          PSFI   P RTH          QHFV   P STR          PVFT

  xonEur˜—n       R WVR          UPFR   S SPI          URFW   S SIU          TWFP   T QWH          UIFR

  „ot—l           T VVS          IHH    U QUR          IHH    U WUU          IHH    V WSR          IHH




4.1.2 Age patterns of women sampled in ANC surveillance surveys

          in Malawi, 1999 to 2005

es evident in „—˜le RFQD the —ge p—ttern of women re™ruited in —ll the surveys in ur˜—n —nd
nonEur˜—n sites from IWWW to PHHS w—s very simil—rF sn ˜oth ur˜—n —nd nonEur˜—n sitesD the
s—mples ™omprised of rel—tively young women in their prime ye—rs of sexu—l —™tivityF ‡omen
—ged less th—n 30 ye—rs ™onstituted over 75% of the s—mples in ˜oth ™—tegories with —t le—st
55% of the women ˜eing —ged ˜etween 15 —nd 24 ye—rsF




4.1.3 Age specic HIV prevalence among women sampled in ANC

          sentinel surveillance surveys

ege spe™i(™ rs† prev—len™e —mong women —ged ISERW s—mpled in exg surveill—n™e surveys
in w—l—wi from IWWW to PHHS for —ggreg—ted ur˜—n —nd nonEur˜—n sites —re presented in
„—˜les RFR —nd RFS respe™tivelyF es ™—n ˜e noted from these t—˜lesD rs† prev—len™e rem—ined

                                                  ST
„—˜le RFQX ege p—tterns of women s—mpled in exg surveill—n™e surveys in w—l—wiD IWWW to
PHHSF

                                        €—nel eX…r˜—n ƒites
                        IWWW                   PHHI                  PHHQ                  PHHS
  ege group     xum˜er         7       xum˜er         7      xum˜er         7      xum˜er         7
  `IS           R              HFP     S              HFQ    P              HFI    V              HFV
  ISEIW         QUV            IWFW    QSR            IWFP   RVI            IWFT   RVU            IWFI
  PHEPR         VIT            RQFI    UVI            RPFS   IHUW           RRFI   IHSU           RIFS
  PSEPW         RPH            PPFP    RRH            PQFW   STS            PQFI   TPV            PRFU
  QHEQR         IWI            IHFI    IWH            IHFQ   PSI            IHFQ   PUW            IIFH
  QSEQW         UT             RFH     TP             QFR    TH             PFS    UP             PFV
  RHERW         W              HFS     U              HFR    W              HFR    IR             HFS
  SHC           I              HFI     E              E      I              HFH    I              HFH
  „ot—l         IVWS           IHH     IVQW           IHH    PRRV           IHH    PSRT           IHH

                                      €—nel fXxonEur˜—n ƒites
                        IWWW                   PHHI                  PHHQ                  PHHS
  ege group     xum˜er         7       xum˜er         7      xum˜er         7      xum˜er         7
  `IS           V              HFP     IQ             HFP    IQ             HFP    S              HFI
  ISEIW         I ISQ          PQFS    I PHQ          PPFI   I IQQ          PHFU   I PWR          PHFR
  PHEPR         I WIQ          QWFH    I WUU          QTFQ   P IPR          QVFV   P QPR          QTFT
  PSEPW         WWH            PHFP    I PHR          PPFI   I ISU          PIFI   I RHV          PPFP
  QHEQR         RWW            IHFP    TRU            IIFW   TRI            IIFU   VSR            IQFR
  QSEQW         PSV            SFQ     QIQ            SFU    QIW            SFV    QPT            SFI
  RHERW         VR             IFU     WP             IFU    VR             IFS    IQR            PFI
  SHC           P              HFH     P              HFH    U              HFI    IH             HFP
  „ot—l         R WHU          IHH     S RSI          IHH    S RUV          IHH    T QSS          IHH



—˜ove 10% —™ross —ll —ge groups in ˜oth —ggreg—ted ur˜—n —nd nonEur˜—n sites during the
entire period from 1999 to 2005F „he ™on(den™e interv—ls in these t—˜les suggest th—t rs†

                                                  SU
prev—len™e in the youngest women @ISEIW —ge groupA rem—ined signi(™—ntly lower th—n th—t
in the older women throughout the period 1999 to 2005F purthermoreD for ˜oth —ggreg—ted
ur˜—n —nd nonEur˜—n sitesD rs† prev—len™e (gures pe—ked in the PSEPW —nd QHEQR —ge groups
throughout the period IWWW to PHHS —s the ™on(den™e interv—ls for rs† prev—len™e (gures in
the two —ge groups indi™—te no st—tisti™—l signi(™—nt di'eren™esF

„—˜le RFRX ege spe™i(™ rs† prev—len™e for women s—mpled in exg surveill—n™e surveys in
—ggreg—ted ur˜—n sites in w—l—wiD IWWW to PHHSF



                            IWWW                                         PHHI
 Age      „ot—l     xum˜er      7        95%         „ot—l      xum˜er     7      95%
 group    sampled   HIV+        HIV+     CI          sampled    HIV+       HIV+   CI
 ISEIW    QUV       UU          PHFR     ITFTEPRFU   QSR        RI         IIFT   VFTEISFR
 PHEPR    VIT       PHQ         PRFW     PPFHEPUFW   UVI        IVU        PQFW   PIFIEPUFI
 PSEPW    RPH       IIV         PVFI     PRFHEQPFT   RRH        IIV        PTFV   PPFWEQIFP
 QHEQR    IWI       SW          QHFW     PRFVEQUFV   IWH        RV         PSFQ   IWFTEQIFW
 QSEQW    UT        PP          PVFW     IWFWERHFI   TP         IV         PWFH   IWFPERIFR
 RHERW    W         P           PPFP     SFQESSFU    U          I          IRFQ   IFIRESVFPQ

                            PHHQ                                         PHHS
 Age      „ot—l     xum˜er      7        95%         „ot—l      xum˜er     7      95%
 group    sampled   HIV+        HIV+     CI          sampled    HIV+       HIV+   CI
 ISEIW    RVI       UV          ITFP     IQFPEIWFV   RVU        TP         IPFU   IHFIEITFH
 PHEPR    I HUW     PPS         PHFW     IVFSEPQFR   I HSU      IWV        IVFU   ITFSEPIFP
 PSEPW    STS       IQV         PRFR     PIFIEPVFI   TPV        IST        PRFV   PIFTEPVFR
 QHEQR    PSI       US          PWFW     PRFSEQSFV   PUW        VP         PWFR   PRFQEQSFH
 QSEQW    TH        IP          PHFH     IIFUEQIFW   UP         IU         PQFT   ISFPEQRFU
 RHERW    W         Q           QQFQ     IIFUETRFW   IR         I          UFI    `HFHHIEQQFS




                                              SV
„—˜le RFSX ege spe™i(™ rs† prev—len™e for women s—mpled in exg surveill—n™e surveys in
—ggreg—ted nonEur˜—n sites in w—l—wiD IWWW to PHHSF



                           IWWW                                        PHHI
 Age      „ot—l     xum˜er     7       95%            „ot—l     xum˜er    7      95%
 group    sampled   HIV+       HIV+    CI             sampled   HIV+      HIV+   CI
 ISEIW    I ISQ     PIH        IVFP    ITFIEPHFT      I PHQ     IRI       IIFU   IHFHEIQFU
 PHEPR    I WIQ     RWS        PSFW    PRFHEPUFW      I WUU     QUH       IVFU   IUFIEPHFS
 PSEPW    WWH       PVI        PVFR    PSFUEQIFQ      I PHR     PVU       PQFV   PHFVEPSFT
 QHEQR    RWW       IIQ        PPFT    IWFPEPTFS      TRU       IQU       PIFP   IVFPEPRFS
 QSEQW    PSV       RU         IVFP    IRFHEPQFR      QIQ       SW        IVFV   IRFWEPQFT
 RHERW    VR        IQ         ISFS    WFIEPRFV       WP        IP        IQFH   UFSEPIFT

                           PHHQ                                        PHHS
 Age      „ot—l     xum˜er     7       95%            „ot—l     xum˜er    7      95%
 group    sampled   HIV+       HIV+    CI             sampled   HIV+      HIV+   CI
 ISEIW    I IQQ     ITU        IRFU    IPFVEITFW      I PWR     IPI       WFR    UFWEIIFI
 PHEPR    P IPR     RIH        IWFQ    IUFUEPIFH      P QPR     QST       ISFQ   IQFWEITFV
 PSEPW    I ISU     PRT        PIFQ    IWFHEPQFU      I RHV     PVQ       PHFI   IVFIEPPFQ
 QHEQR    TRI       IRR        PPFS    IWFREPSFW      VSR       ISI       IUFU   ISFQEPHFR
 QSEQW    QIW       SQ         ITFT    IPFWEPIFI      QPT       SQ        ITFQ   IPFTEPHFU
 RHERW    VR        IT         IWFH    IPFHEPVFV      IQR       PI        ISFU   IHFREPPFV



en ex—min—tion of trends in the —ge spe™i(™ rs† prev—len™e —mong the women s—mpled
in exg surveill—n™e surveys in w—l—wi during the period IWWW to PHHS indi™—ted th—t rs†
prev—len™e in younger —ge groups de™lined ˜ut rem—ined —lmost st—˜le in old womenF es
depi™ted in pigures RFP —nd RFQ there w—s — gener—l downw—rd trend in rs† prev—len™e
˜etween IWWW —nd PHHS in the ISEIWD PHEPR —nd PSEPW —ge groups in ˜oth the —ggreg—ted
ur˜—n —nd nonEur˜—n sitesF roweverD in ˜oth sentinel site ™—tegories the de™line w—s more
pronoun™ed in the ISEIW —nd PHEPR —ge groupsF es „—˜le RFR showsD the num˜ers of women

                                            SW
—ged RHERW s—mpled in the exg surveill—n™e surveys in the ur˜—n sites were very sm—llD
therefore the trends in rs† prev—len™e in this —ge group should ˜e interpreted with ™—utionF
„he trendline of rs† prev—len™e for women —ged RHERW in ur˜—n sites shown in pigure RFP
might not ˜e — true represent—tion of re—lityF „husD the pe—k shown in pigure RFP does not
ne™ess—rily me—n rs† prev—len™e in ur˜—n —re—s in w—l—wi w—s highest —mong women —ged
RHERW ™omp—red to women in the younger —ge groups in PHHQF

                                                                                        Age group
                     50




                                                                                        q   15−19
                                                                                            20−24
                                                                                            25−29
                                                                                        q   30−34
                     40




                                                                                            35−39
                                                                                            40−49
  HIV positive (%)




                             q
                     30




                                                                           q                   q


                                                    q
                     20




                             q


                                                                           q


                                                                                               q
                                                    q
                     10
                     0




                           1999                   2001                   2003                 2005

                                                              Year



                          pigure RFPX ege spe™i(™ rs† prev—len™e in —ggreg—ted ur˜—n sitesF




                                                         TH
                                                                                      Age group
                     40                                                               q   15−19
                                                                                          20−24
                                                                                          25−29
                                                                                      q   30−34
                                                                                          35−39
                     30



                                                                                          40−49
  HIV positive (%)




                           q                                             q
                                                  q
                     20




                           q
                                                                                               q


                                                                         q


                                                  q
                     10




                                                                                               q
                     0




                          1999                  2001                  2003                    2005

                                                            Year



                      pigure RFQX ege spe™i(™ rs† prev—len™e in —ggreg—ted nonEur˜—n sitesF




4.1.4 Adjusted age specic HIV prevalence for non-urban sites

fe™—use the sites we ™l—ssi(ed —s nonEur˜—n —re lo™—ted in sm—ll towns —nd l—rge rur—l
settlements with higher levels of e™onomi™ —™tivity —nd mo˜ility —nd pro˜—˜ly —sso™i—ted
with higher rs† prev—len™e —nd m—y therefore overEestim—te rs† prev—len™e in most remote
—re—sD ™onsistent with the …xeshƒG‡ry —ppro—™hD we —djusted down the —ge spe™i(™ rs†
prev—len™e (gures in nonEur˜—n sites presented in „—˜le RFS ˜y 20%F „he —djusted —ge spe™i(™
rs† prev—len™e (gures for the —ggreg—ted nonEur˜—n sites —re presented in „—˜le RFTF




                                                       TI
„—˜le RFTX ege spe™i(™ rs† prev—len™e for —ggreg—ted nonEur˜—n sites in w—l—wi from IWWW

to PHHS —djusted for represent—tivenessF


                                              rs† prev—len™e @7A

         ege group             IWWW           PHHI            PHHQ            PHHS

         ISEIW                 IRFT           WFR             IIFV            UFS

         PHEPR                 PHFU           ISFH            ISFR            IPFP

         PSEPW                 PPFU           IWFH            IUFH            ITFI

         QHEQR                 IVFI           IUFH            IVFH            IRFP

         QSEQW                 IRFT           ISFH            IQFQ            IQFH

         RHERW                 IPFR           IHFR            ISFP            IPFT



4.1.5 National estimates of age specic HIV prevalence among

         women based on ANC surveillance data

epplying the estim—tes of the per™ent—ges of fem—le popul—tions in ur˜—n —nd rur—l —re—s to
the estim—tes of —ge spe™i(™ rs† prev—len™e for the —ggreg—ted ur˜—n —nd nonEur˜—n sites
presented in „—˜les RFR —nd RFT respe™tivelyD yielded prelimin—ry estim—tes of n—tion—l —geE
spe™i(™ rs† prev—len™e for women in w—l—wi from IWWW to PHHS ˜—sed on exg surveill—n™e
d—t— presented in „—˜le RFUF


istim—tes of n—tion—l —ge spe™i(™ rs† prev—len™e —mong women ˜—sed on the exg
surveill—n™e d—t— for the ye—rs PHHHD PHHP —nd PHHR were line—rly interpol—ted from the
rs† prev—len™e (gures presented in „—˜le RFUF „he interpol—ted estim—tes of —ge spe™i(™
rs† prev—len™e together with the —ge spe™i(™ rs† prev—len™e (gures in „—˜le RFU formed
the (n—l n—tion—l estim—tes of —ge spe™i(™ rs† prev—len™e —mong women ˜—sed on exg



                                            TP
„—˜le RFUX €relimin—ry estim—tes of the n—tion—l —ge spe™i(™ rs† prev—len™e —mong women

in w—l—wi from IWWW to PHHS ˜—sed on exg surveill—n™e d—t—F

                                                   rs† prev—len™e @7A

         ege group             IWWW                PHHI          PHHQ          PHHS

         ISEIW                 ISFS                WFV           IPFT          VFS

         PHEPR                 PIFS                ITFU          ITFS          IQFT

         PSEPW                 PQFT                PHFR          IVFR          IUFV

         QHEQR                 PHFH                IVFQ          PHFH          ITFV

         QSEQW                 ITFS                IUFH          IRFQ          IRFU

         RHERW                 IQFR                IHFV          IUFQ          IIFW



surveill—n™e d—t— presented in „—˜le RFVF

„—˜le RFVX pin—l estim—tes of the n—tion—l —ge spe™i(™ rs† prev—len™e —mong women in
w—l—wi from IWWW to PHHR ˜—sed on exg surveill—n™e d—t—

                                              rs† prev—len™e @7A
     ege group       IWWW      PHHH         PHHI          PHHP   PHHQ   PHHR          PHHS

     ISEIW           ISFS      IPFT         WFV           IIFP   IPFT   IHFT          VFS
     PHEPR           PIFS      IWFI         ITFU          ITFT   ITFS   ISFI          IQFT
     PSEPW           PQFT      PPFH         PHFR          IWFR   IVFR   IVFI          IUFV
     QHEQR           PHFH      IWFP         IVFQ          IWFI   PHFH   IVFR          ITFV
     QSEQW           ITFS      ITFV         IUFH          ISFU   IRFQ   IRFS          IRFU
     RHERW           IQFR      IPFI         IHFV          IRFI   IUFQ   IRFT          IIFW




                                                  TQ
es depi™ted in pigure RFRD the (n—l estim—tes of n—tion—l —ge spe™i(™ rs† prev—len™e —mong
women in w—l—wi from IWWW to PHHS ˜—sed on the exg surveill—n™e d—t— suggest th—t the
tr—je™tory of rs† prev—len™e —™ross —ge groups w—s —lmost simil—r in —ll the ye—rs ex™ept
PHHQF ‡ith the ex™eption of the RHERW —ge group whose prev—len™e (gures were —'e™ted
˜y sm—ll s—mple sizes —s indi™—ted e—rlierD rs† prev—len™e r—tes were lowest —mong women
—ged ISEIW ye—rs —nd pe—ked —mong women —ged PSEPW ye—rsF pigure RFR —lso reve—ls th—t
there w—s — su˜st—nti—l de™line in n—tion—l rs† prev—len™e —mong women in the younger —ge
groups @ISEIWD PHEPR —nd PSEPWAF
                     30




                                                                            q
                                                                                 1999
                                                                            q    2000
                                                                                 2001
                                                                                 2002
                     25




                                                                                 2003
                                            q                                    2004
                                            q
                                                                            q
                                                                                 2005
                                    q
                     20




                                                            q
                                    q                       q
  HIV positive (%)




                                            q
                                                            q         q
                                                                      q
                            q
                     15




                                                                      q
                                    q                                               q
                            q
                                                                                    q
                                                                                    q
                     10




                            q
                     5
                     0




                          15−19   20−24   25−29        30−34       35−39         40−49

                                                Age group



pigure RFRX x—tion—l estim—tes of —ge spe™i(™ rs† prev—len™e —mong women ˜—sed on exg

d—t—F




                                            TR
4.1.6 Final national estimates of age specic HIV prevalence among

         women based on ANC surveillance data and 2004 Malawi

         DHS results

por the (rst time w—l—wi h—s n—tion—l popul—tionE˜—sed rs† prev—len™e estim—tes for women
—nd men derived from the PHHR w—l—wi hrƒ whi™h in™luded rs† testingF „he —ge spe™i(™
n—tion—l rs† prev—len™e estim—tes for women of reprodu™tive —ge derived from the PHHR
w—l—wi hrƒ —re presented in „—˜le RFWF

„—˜le RFWX istim—tes of —ge spe™i(™ rs† prev—len™e —mong fem—les —ged ISERW in w—l—wi

derived from PHHR w—l—wi hrƒ

         ege group              xum˜er tested               rs† prev—len™e @7A

         ISEIW                  SHH                         QFU

         PHEPR                  TTI                         IQFP

         PSEPW                  RUU                         ISFS

         QHEQR                  QVP                         IVFI

         QSEQW                  PSU                         IUFH

         RHERW                  RHV                         ISFW


                         ƒour™eX xƒy —nd y‚g w—™ro @PHHSAF

 xoteX xum˜ers for the RHERW —ge group were ™omputed ˜y —uthor ˜—sed on the origin—l

       num˜ers in the RHERR —nd RSERW —ge groups in the PHHR w—l—wi hrƒ reportF



e ™omp—rison w—s m—de ˜etween the n—tion—l estim—tes of —ge spe™i(™ rs† prev—len™e —mong
fem—les —ged ISERW in w—l—wi for the ye—r PHHR derived from the PHHR w—l—wi hrƒ —nd


                                           TS
those ˜—sed on the exg surveill—n™e d—t—F e™™ording to xƒy —nd y‚g w—™ro @PHHSAD the
PHHR w—l—wi hrƒ w—s ™ondu™ted from y™to˜er PHHR to t—nu—ry PHHS while the PHHS exg
surveill—n™e w—s ™ondu™ted over — period of eight weeks st—rting on I eugust PHHS @xeg
PHHSAF „hereforeD —ppropri—te rs† prev—len™e estim—tes from the exg surveill—n™e d—t— th—t
™ould ˜e ™omp—red to the rs† prev—len™e estim—tes derived from the PHHR w—l—wi hrƒ were
™omputed —s the —ver—ge of the estim—tes of rs† prev—len™e ˜—sed on exg surveill—n™e d—t—
for the ye—rs PHHR —nd PHHS presented in „—˜le RFVF pigure RFS illustr—tes the ™omp—rison
—n—lysis of the ™omputed rs† prev—len™e estim—tes from the exg surveill—n™e d—t— —nd
the rs† prev—len™e estim—tes from the PHHR w—l—wi hrƒF es ™—n ˜e noted from pigure RFSD
estim—tes of rs† prev—len™e ˜y —ge ˜—sed on the exg surveill—n™e d—t— di'er from estim—tes
of rs† prev—len™e for the gener—l popul—tion derived from the PHHR w—l—wi hrƒF por the ISE
IWD PHEPR —nd PSEPW —ge groupsD estim—tes of rs† prev—len™e ˜—sed on the exg surveill—n™e
d—t— —re higher th—n those o˜t—ined from the PHHR w—l—wi hrƒ —nd there is — reversed
rel—tionship when ™omp—ring the QHEQRD QSEQW —nd RHERW —ge groupsF „his ™rossEover e'e™tD
—lthough with di'erent m—gnitudesD h—s ˜een o˜served in most studies ™omp—ring estim—tes
of rs† prev—len™e from exg surveill—n™e d—t— —nd rs† prev—len™e in the gener—l popul—tion
@pylkesnes et —lFD IWWVY ‡ry —nd …xeshƒ PHHQAF por this studyD the m—gnitudes of the
di'eren™es ˜etween the estim—tes of rs† prev—len™e ˜y —ge ˜—sed on exg surveill—n™e d—t—
—nd those derived from the PHHR w—l—wi hrƒ —re presented in „—˜le RFIH —s r—tios of exg
rs† prev—len™e to gener—l popul—tion rs† prev—len™eF


„he estim—tes of —ge spe™i(™ r—tios of exg rs† prev—len™e to gener—l popul—tion rs†
prev—len™e presented in „—˜le RFIH were used to further —djust the estim—tes of rs†
prev—len™e ˜y —ge ˜—sed on exg surveill—n™e d—t— presented in „—˜le RFV to o˜t—in (n—l
estim—tes of n—tion—l rs† prev—len™e ˜y —ge for —ll women of reprodu™tive —ge in w—l—wi for
the ye—rs IWWW to PHHSF „his yielded estim—tes of n—tion—l rs† prev—len™e ˜y —ge for women
of reprodu™tive —ge in w—l—wi for the ye—rs IWWW to PHHS presented in „—˜le RFIIF




                                            TT
                                                               q       2004 MDHS
                                                               q       ANC
                    20




                                           q               q
                                                           q
                                                                   q

                                                                              q
                                           q
                    15




                                                                   q
                                   q
 HIV positive (%)




                                   q                                          q
                    10




                           q
                    5




                           q
                    0




                         15−19   20−24   25−29        30−34    35−39        40−49

                                               Age group



pigure RFSX gomp—rison ˜etween estim—tes of —ge spe™i(™ rs† prev—len™e —mong women

—ged ISERW in w—l—wi in PHHR ˜—sed on PHHR w—l—wi hrƒ —nd exg d—t—F




                                           TU
„—˜le RFIHX istim—tes of r—tios of exg rs† prev—len™e to gener—l popul—tion rs† prev—len™e


   ege group       ‚—tio of exg rs† prev—len™e to gener—l popul—tion rs† prev—len™e

   ISEIW           PFSV

   PHEPR           IFHW

   PSEPW           IFIT

   QHEQR           HFWU

   QSEQW           HFVT

   RHERW           HFVQ




„—˜le RFIIX pin—l estim—tes of —ge spe™i(™ n—tion—l rs† prev—len™e —mong women of

reprodu™tive —ge in w—l—wiD IWWW to PHHS


                                           rs† €rev—len™e @7A

   ege group       IWWW      PHHH       PHHI       PHHP      PHHQ      PHHR       PHHS

   ISEIW           TFH       RFW        QFV        RFQ       RFW       RFI        QFQ

   PHEPR           IWFV      IUFT       ISFR       ISFQ      ISFP      IQFW       IPFS

   PSEPW           PHFR      IWFH       IUFU       ITFV      ISFW      ISFT       ISFR

   QHEQR           PHFT      IWFU       IVFV       IWFU      PHFS      IVFW       IUFQ

   QSEQW           IWFP      IWFS       IWFV       IVFP      ITFT      ITFW       IUFI

   RHERW           ITFP      IRFT       IQFH       ITFW      PHFV      IUFT       IRFQ




                                              TV
4.2 Estimates of women of reproductive age

pour di'erent sets of estim—tes of women of reprodu™tive —ge in w—l—wi for the period IWWW
to PHHR were m—deF „he estim—tes were o˜t—ined ˜y proje™ting the —ge smoothed popul—tion
of fem—les presented in „—˜le RFIP whi™h w—s o˜t—ined from the —n—lyti™ report of the IWWV
™ensus @xƒyD PHHPA using four di'erent wv„ systems —s outlined in the methodology se™tion
@se™tion QFIFPAF „he (rst set of estim—tes w—s done using surviv—l r—tios derived from the
‡ry wv„ systemD the se™ond using surviv—l r—tios derived from the …nited x—tions p—r i—st
wv„ systemD the third using surviv—l r—tios derived from the north f—mily of go—leEhemeny
wv„ system —nd the fourth using surviv—l r—tios derived from the south f—mily of go—leE
hemeny wv„ systemF „—˜les RFIQ to RFPH present the estim—tes of women of reprodu™tive
—ge in w—l—wi for the ye—rs IWWW to PHHR ˜—sed on these four wv„ systemsF

              „—˜le RFIPX ege smoothed popul—tion of fem—les in w—l—wi in IWWVF


        ege group           €opul—tion size        ege group       €opul—tion size

        HER                 VQS UVH                RHERR           IWQ TRU

        SEW                 UPT SPS                RSERW           ISQ QWQ

        IHEIR               TPQ TPS                SHESR           IIP HTW

        ISEIW               SSP SHI                SSESW           WI WHI

        PHEPR               SIR IVQ                THETR           VH VST

        PSEPW               RPV PWI                TSETW           UQ TRP

        QHEQR               QHR WVU                UHEUR           TW HHR

        QSEQW               PQV WSV                US & over       TT WRQ


                                      ƒour™eX xƒyD PHHPF




                                              TW
„—˜le RFIQX istim—tes of women of reprodu™tive —ge in w—l—wi from IWWW to PHHI ˜—sed on
‡ry wv„ systemF

                       IWWW                        PHHH                      PHHI
   ege       €opul—tion       7 of     €opul—tion         7 of     €opul—tion       7 of
   group     size             tot—l    size               tot—l    size             tot—l
   ISEIW     STH PQW          PQFHT    SUH PPI            PQFHS    SVP QHU          PQFII
   PHEPR     SIU UHQ          PIFQI    SIW SQR            PIFHH    SPH WTI          PHFTV
   PSEPW     RRH QPR          IVFIP    RSH THT            IVFPI    RSW HTV          IVFPP
   QHEQR     QIU HST          IQFHS    QQI TIR            IQFRH    QRU IIQ          IQFUV
   QSEQW     PRI IIQ          WFWP     PRQ QWR            WFVR     PRT THV          WFUW
   RHERR     IWS UTR          VFHT     IWU URH            UFWW     IWW TPQ          UFWP
   RSERW     ISU RSW          TFRV     ITH VHU            TFSH     ITQ SRS          TFRW
   „ot—l     P RPW TSV        IHHFHH   P RUQ WIT          IHHFHH   P SIW PPS        IHHFHH



„—˜le RFIRX istim—tes of women of reprodu™tive —ge in w—l—wi from PHHP to PHHR ˜—sed on
‡ry wv„ systemF

                       PHHP                        PHHQ                      PHHR
   ege       €opul—tion       7 of     €opul—tion         7 of     €opul—tion       7 of
   group     size             tot—l    size               tot—l    size             tot—l
   ISEIW     SWT IRQ          PQFPQ    TII RRI            PQFQV    TPU UHP          PQFSR
   PHEPR     SPQ TQT          PHFRI    SPV URP            PHFPP    SQT IVW          PHFII
   PSEPW     RTS UQU          IVFIS    RUH UTV            IVFHH    RUQ WQW          IUFUU
   QHEQR     QTI TTI          IRFHW    QUR HII            IRFQH    QVR RUV          IRFRP
   QSEQW     PSI UIR          WFVI     PSW IRT            WFWI     PTW RHW          IHFIH
   RHERR     PHI RSI          UFVS     PHQ QHR            UFUU     PHS IQV          UFTW
   RSERW     ITS VRP          TFRT     ITU VQV            TFRP     ITW TUT          TFQT
   „ot—l     P STT IVR        IHHFHH   P TIS PSH          IHHFHH   P TTT SQI        IHHFHH



                                              UH
„—˜le RFISX istim—tes of women of reprodu™tive —ge in w—l—wi from IWWW to PHHI ˜—sed on
…x p—r i—st wv„ systemF

                       IWWW                        PHHH                      PHHI
   ege       €opul—tion       7 of     €opul—tion         7 of     €opul—tion       7 of
   group     size             tot—l    size               tot—l    size             tot—l
   ISEIW     SSV PWV          PPFVR    STT SVP            PPFTQ    SUU IWS          PPFSI
   PHEPR     SIU WPV          PIFIW    SIW TQS            PHFUT    SPH SVT          PHFQH
   PSEPW     RRR HWU          IVFIU    RSU UHS            IVFPW    RTV WWS          IVFPW
   QHEQR     QPP HQU          IQFIV    QRI UTV            IQFTS    QTP TWI          IRFIR
   QSEQW     PRS PRH          IHFHQ    PSI WII            IHFHT    PSW VRS          IHFIQ
   RHERR     IWV PVQ          VFII     PHQ ITU            VFIP     PHV QTH          VFIQ
   RSERW     ISV HPT          TFRU     ITP QVR            TFRW     ITT SVP          TFSH
   „ot—l     P RRQ WHW        IHHFHH   P SHQ ISP          IHHFHH   P STR PSR        IHHFHH



„—˜le RFITX istim—tes of women of reprodu™tive —ge in w—l—wi from PHHP to PHHR ˜—sed on
…x p—r i—st wv„ systemF

                       PHHP                        PHHQ                      PHHR
   ege       €opul—tion       7 of     €opul—tion         7 of     €opul—tion       7 of
   group     size             tot—l    size               tot—l    size             tot—l
   ISEIW     SVW UTV          PPFRR    THQ WUP            PPFRP    TIW SIU          PPFRP
   PHEPR     SPP RPI          IWFVV    SPT PVI            IWFSQ    SQP UIS          IWFPV
   PSEPW     RUU WTQ          IVFIW    RVR UHV            IUFWW    RVW IWI          IUFUH
   QHEQR     QVP VRV          IRFSU    RHH VQH            IRFVV    RIT RUH          ISFHU
   QSEQW     PUH IRS          IHFPV    PVQ RHQ            IHFSP    PWW VUH          IHFVS
   RHERR     PIQ WIT          VFIR     PIW WPR            VFIT     PPT ISU          VFIV
   RSERW     IUH UWT          TFSH     IUS ITU            TFSH     IUW UQH          TFSH
   „ot—l     P TPU VSU        IHHFHH   P TWR PVS          IHHFHH   P UTQ TSH        IHHFHH



                                              UI
„—˜le RFIUX istim—tes of women of reprodu™tive —ge in w—l—wi from IWWW to PHHI ˜—sed on
go—leEhemeny north wv„ systemF

                       IWWW                        PHHH                      PHHI
   ege       €opul—tion       7 of     €opul—tion         7 of     €opul—tion       7 of
   group     size             tot—l    size               tot—l    size             tot—l
   ISEIW     SSW SUR          PPFUW    STV TUR            PPFSP    SUW TSR          PPFQQ
   PHEPR     SPH IPS          PIFIW    SPQ UTW            PHFUS    SPT QVR          PHFPV
   PSEPW     RRT PRI          IVFIV    RTI WWI            IVFQH    RUS QUQ          IVFQI
   QHEQR     QPQ TPP          IQFIV    QRS HTR            IQFTU    QTU VPH          IRFIU
   QSEQW     PRT RWP          IHFHR    PSR RQI            IHFHV    PTQ TSW          IHFIT
   RHERR     IWW SPW          VFIQ     PHS TQP            VFIR     PIP HIV          VFIU
   RSERW     ISW RRI          TFRW     ITS IUQ            TFSR     IUH TWQ          TFSV
   „ot—l     P RSS HPR        IHHFHH   P SPR UQR          IHHFHH   P SWS THI        IHHFHH



„—˜le RFIVX istim—tes of women of reprodu™tive —ge in w—l—wi from PHHP to PHHR ˜—sed on
go—leEhemeny north wv„ systemF

                       PHHP                        PHHQ                      PHHR
   ege       €opul—tion       7 of     €opul—tion         7 of     €opul—tion       7 of
   group     size             tot—l    size               tot—l    size             tot—l
   ISEIW     SWP IRU          PPFIW    THS VRQ            PPFHW    TPH RUQ          PPFHH
   PHEPR     SPW TPR          IWFVS    SQR TVQ            IWFRW    SRI WHU          IWFPP
   PSEPW     RVT QQW          IVFPQ    RWR WUS            IVFHS    SHI HVU          IUFUU
   QHEQR     QVW VWU          IRFTI    RHW VRH            IRFWR    RPU RHW          ISFIT
   QSEQW     PUS QHW          IHFQP    PWH HIR            IHFSU    QHV HUI          IHFWQ
   RHERR     PIV UQW          VFPH     PPS VVU            VFPR     PQQ PUR          VFPU
   RSERW     IUT IUH          TFTH     IVI URR            TFTQ     IVU RVV          TFTS
   „ot—l     P TTV PPS        IHHFHH   P URP WVT          IHHFHH   P VIW UHW        IHHFHH



                                              UP
„—˜le RFIWX istim—tes of women of reprodu™tive —ge in w—l—wi from IWWW to PHHI ˜—sed on
go—leEhemeny ƒouth wv„ systemF

                       IWWW                        PHHH                      PHHI
   ege       €opul—tion       7 of     €opul—tion         7 of     €opul—tion       7 of
   group     size             tot—l    size               tot—l    size             tot—l
   ISEIW     SSW VVV          PPFUV    STW RPH            PPFSH    SVH WSI          PPFQH
   PHEPR     SPH QIU          PIFIU    SPR ITS            PHFUI    SPT WWT          PHFPQ
   PSEPW     RRT TSQ          IVFIU    RTP UUV            IVFPV    RUT RWI          IVFPW
   QHEQR     QPR IWH          IQFIW    QRT PHQ            IQFTV    QTW SQS          IRFIV
   QSEQW     PRU IHI          IHFHS    PSS TRR            IHFIH    PTS RUT          IHFIW
   RHERR     PHH HWW          VFIR     PHT UVR            VFIU     PIQ UTP          VFPH
   RSERW     ISW VWW          TFSH     ITT IIS            TFST     IUP IRP          TFTI
   „ot—l     P RSV IRU        IHHFHH   P SQI IHW          IHHFHH   P THS QSQ        IHHFHH



„—˜le RFPHX istim—tes of women of reprodu™tive —ge in w—l—wi from PHHP to PHHR ˜—sed on
go—leEhemeny ƒouth wv„ systemF

                       PHHP                        PHHQ                      PHHR
   ege       €opul—tion       7 of     €opul—tion         7 of     €opul—tion       7 of
   group     size             tot—l    size               tot—l    size             tot—l
   ISEIW     SWR IPP          PPFIT    THV TPW            PPFHS    TPR PWT          PIFWV
   PHEPR     SQH RTS          IWFUV    SQS UUS            IWFRI    SRQ QUQ          IWFIQ
   PSEPW     RVU UQT          IVFIW    RWT THQ            IUFWW    SHP WTP          IUFUI
   QHEQR     QWP IVI          IRFTQ    RIP TUI            IRFWS    RQH URW          ISFIT
   QSEQW     PUU URS          IHFQT    PWQ IHU            IHFTP    QII VSH          IHFWV
   RHERR     PPI HVV          VFPR     PPV VTI            VFPW     PQT VUQ          VFQR
   RSERW     IUV ISH          TFTR     IVR PVP            TFTV     IWH SWU          TFUI
   „ot—l     P TVI RVU        IHHFHH   P USW WPV          IHHFHH   P VRH UHH        IHHFHH



                                              UQ
4.2.1 Comparison of estimates of women of reproductive age in

         Malawi from 1999 to 2004

es ™—n ˜e noted from „—˜les RFIQ to RFPHD throughout the ye—rs the estim—tes ˜—sed on
—ll the four wv„ systems indi™—te th—t —˜out 75% of the women in the ISERW —ge group
in w—l—wi were less th—n QS ye—rs oldF purthermoreD —s shown in pigure RFTD the ‡ry
wv„ system produ™ed the lowest estim—tes of the tot—l popul—tion of women —ged ISERW in
w—l—wi from IWWW to PHHR —nd the highest estim—tes were o˜t—ined from the south f—mily
of go—leEhemeny wv„ systemF sn —dditionD ™omp—rison ˜etween the four sets of estim—tes
of the tot—l popul—tion of women —ged ISERW in w—l—wi from IWWW to PHHR derived from
the four wv„ systems —nd estim—tes done ˜y the w—l—wi x—tion—l ƒt—tisti™—l y0™e —nd
the x—tion—l eshƒ gommission reve—led th—t there were no huge di'eren™es ˜etween the
estim—tes derived from the north f—mily of go—leEhemeny wv„ system —nd those done ˜y
the x—tion—l eshƒ gommission @see pigure RFUAF pigure RFU further indi™—tes th—t there were
—lso no huge di'eren™es ˜etween estim—tes derived from the south f—mily of go—leEhemeny
wv„ system —nd those done ˜y the w—l—wi x—tion—l ƒt—tisti™—l y0™eF




                                            UR
                        3500000
                                                                            WHO
                                                                            UN Far East
                                                                            Coale-Demeny North
                        3000000                                             Coale-Demeny South




                        2500000
  Population of women




                        2000000




                        1500000




                        1000000




                        500000




                             0
                                  1999   2000   2001          2002   2003         2004
                                                       Year




pigure RFTX gomp—rison of estim—tes of tot—l women popul—tion —ged ISERW in w—l—wi from

IWWW to PHHRF




                                                  US
                      3500000
                                                                                        Coale-Demeny South
                                                                                        NSO
                                                                                        NAC 2005
                      3000000                                                           Coale-Demeny North




                      2500000
Population of women




                      2000000




                      1500000




                      1000000




                      500000




                           0
                                1999    2000         2001          2002        2003          2004
                                                            Year




                pigure RFUX gomp—rison of estim—tes of women —ged ISERW in w—l—wi from IWWWEPHHR




                                                       UT
4.3 Estimates of the number of HIV infected women

istim—tes of the num˜er of women —ged ISERW infe™ted with rs† were m—de ˜y —pplying
the —ge spe™i(™ estim—tes of rs† prev—len™e presented in „—˜le RFII to the estim—tes of the
tot—l popul—tion of women —ged ISERW in w—l—wi presented in „—˜les RFIQ to RFPHF e r—nge
of the o˜t—ined estim—tes —re presented in „—˜les RFPI to RFPRF snterestinglyD the estim—tes
of the num˜er of women —ged ISERW infe™ted with rs† presented in „—˜les RFPI to RFPR
yield estim—tes of rs† prev—len™e for —dult fem—les @—ge ISERWA in w—l—wi for the ye—rs PHHH
to PHHR th—t diverge only very n—rrowly from those presented in „—˜le RFPSD whi™h were
produ™ed ˜y xeg @PHHSA using the ƒpe™trum p—™k—ge in ™om˜in—tion with the istim—tion
—nd €roje™tion €—™k—geF „he di'eren™es ˜etween the estim—tes of rs† prev—len™e for —dult
fem—les in w—l—wi from PHHH to PHHR o˜t—ined in this study —nd xeg @PHHSA9s estim—tes
r—nge from 0.6 to 2.1 per™ent—ge points with the estim—tes o˜t—ined in this study ˜eing
lower th—n xeg @PHHSA9s estim—tes in —ll the ye—rs —s shown in pigure RFVF




                                            UU
„—˜le RFPIX istim—tes of the num˜er of rs† infe™ted women in w—l—wi for estim—tes of

women ˜—sed on ‡ry wv„ systemF


   ege group            IWWW      PHHH           PHHI      PHHP      PHHQ      PHHR

   ISEIW                QQ UQR    PU WRU         PP HIT    PS VHW    PW VPT    PS TWH

   PHEPR                IHP RUH   WI QHR         UW WWR    VH HVS    VH SRR    UR RII

   PSEPW                VW WUQ    VS VIR         VI HRW    UV IUQ    UR WIW    UR IIS

   QHEQR                TS PTR    TS PWR         TS PRI    UI IQH    UT VPQ    UP UIV

   QSEQW                RT QRP    RU SQH         RV WIV    RS WIH    RQ IPT    RS RUV

   RHERR                QI TIW    PV VRV         PT HHR    QR HVI    RP QHU    QT HIU

   RSERW                PS RQP    PQ RTH         PI QHR    PV HSU    QR WPU    PW UWI

   „ot—l rs†Einfe™ted

   women @ISERWA        QWR VQR   QUH IWU        QRR SPT   QTQ PRS   QVP RUP   QSV PPH

   7 rs†Einfe™ted

   women @ISERWA        ITFQ      ISFH           IQFU      IRFP      IRFT      IQFR




                                            UV
„—˜le RFPPX istim—tes of the num˜er of rs† infe™ted women in w—l—wi for estim—tes of

women ˜—sed on …x p—r i—st wv„ systemF


   ege group            IWWW      PHHH           PHHI      PHHP      PHHQ      PHHR

   ISEIW                QQ TIV    PU UTW         PI VPQ    PS SQQ    PW RTP    PS QSS

   PHEPR                IHP SIS   WI QPP         UW WQT    UW VWW    VH ITW    UQ WPW

   PSEPW                WH URQ    VU ITT         VP VHP    VH PPS    UU IQV    UT SHH

   QHEQR                TT PVW    TU PWQ         TV ITW    US PWV    VP QQI    UV UTW

   QSEQW                RU IQS    RW IWR         SI SRR    RW PUP    RU ITQ    SH TPH

   RHERR                QP HPT    PW TRH         PU IRP    QT IWH    RS UTS    QW UHU

   RSERW                PS SPR    PQ TWH         PI UHH    PV VWS    QT RSP    QI SST

   „ot—l rs†Einfe™ted

   women @ISERWA        QWU VSH   QUT HUR        QSQ IIT   QUS QIP   QWV RVH   QUT RQT

   7 rs†Einfe™ted

   women @ISERWA        ITFQ      ISFH           IQFV      IRFQ      IRFV      IQFT




                                            UW
„—˜le RFPQX istim—tes of the num˜er of rs† infe™ted women in w—l—wi for estim—tes of

women ˜—sed on go—leEhemeny xorth wv„ systemF


   ege qroup            IWWW      PHHH           PHHI      PHHP      PHHQ      PHHR

   ISEIW                QQ TWR    PU VUI         PI WIT    PS TQT    PW SSQ    PS QWR

   PHEPR                IHP WSH   WP HRV         VH VPU    VI HHI    VI RRW    US PHR

   PSEPW                WI IVP    VU WVQ         VQ WPV    VI TQI    UV UUP    UV QTH

   QHEQR                TT TIS    TU WRP         TW IQQ    UT TVR    VR IVP    VH VQV

   QSEQW                RU QUT    RW TVT         SP QHH    SH PIR    RV PTQ    SP HHR

   RHERR                QP PPU    QH HHH         PU TIV    QU HHT    RU HHT    RH WSU

   RSERW                PS USP    PR HWU         PP PQS    PW VHR    QU VPH    QP WIV

   „ot—l rs† infe™ted

   women @ISERWA        QWW UWT   QUW TPU        QSU WSU   QVI WUT   RHU HRS   QVS TUS

   7 rs†Einfe™ted

   women @ISERWA        ITFQ      ISFH           IQFV      IRFQ      IRFV      IQFU




                                            VH
„—˜le RFPRX istim—tes of the num˜er of rs† infe™ted women in w—l—wi for estim—tes of

women ˜—sed on go—leEhemeny ƒouth wv„ systemF


   ege qroup              IWWW       PHHH           PHHI      PHHP      PHHQ      PHHR

   ISEIW                  QQ UIQ     PU WHV         PI WTS    PS UPP    PW TVW    PS SSH

   PHEPR                  IHP WVV    WP IIV         VH WPH    VI IQH    VI TIS    US RHV

   PSEPW                  WI PTT     VV IQQ         VR IPS    VI VTS    UW HQI    UV TSR

   QHEQR                  TT UQP     TV ITU         TW RSS    UU IQQ    VR UTQ    VI RTW

   QSEQW                  RU RWQ     RW WPQ         SP TTI    SH TSV    RV UUV    SP TRP

   RHERR                  QP QIW     QH ITV         PU VRS    QU RHR    RU TPS    RI SVW

   RSERW                  PS VPT     PR PQS         PP RPR    QH IQW    QV QRW    QQ RTR

   „ot—l rs† infe™ted

   women @ISERWA          RHH QQU    QVH TSP        QSW QWS   QVR HSI   RHW VSH   QVV UUT

   7 rs†Einfe™ted

   women @ISERWA          ITFQ       ISFH           IQFV      IRFQ      IRFW      IQFU




„—˜le RFPSX istim—tes of n—tion—l rs† prev—len™e —mong women of reprodu™tive —ge in

w—l—wi from IWWW to PHHS ˜y xeg@PHHSA



                                               rs† €rev—len™e @7A

   ege qroup            PHHH        PHHI            PHHP       PHHQ       PHHR       PHHS

   ISERW                ITFI        ISFW            ISFU       ISFS       ISFR       ISFR

                ƒour™eX gomputed ˜y —uthor from xeg@PHHSA9s reportF



                                               VI
                   25




                                                                   q
                                                                       NAC (2005)
                                                                   q
                                                                       This study
                   20




                         q
                                q
HIV positive (%)




                                                q
                                                               q                q
                   15




                         q                                     q
                                                q
                                q                                               q
                   10
                   5
                   0




                        2000   2001            2002          2003            2004

                                               Year


pigure RFVX gomp—rison of this study9s estim—tes of rs† prev—len™e for —dult fem—les in

w—l—wi with xeg @PHHSA9s estim—tes from PHHH to PHHR




                                          VP
4.4 Estimates of the number of children

istim—tes of the num˜er of ™hildren ˜orn to —ll women of reprodu™tive —ge in w—l—wi from
IWWW to PHHR were ˜—sed on eƒp‚s derived from the PHHH —nd PHHR w—l—wi hrƒs presented
in „—˜le RFPTF en ex—min—tion of the eƒp‚s presented in „—˜le RFPT reve—ls th—t fertility
slightly de™lined ˜etween PHHH —nd PHHR in —ll —ge groups ex™ept the QHEQR —ge —s shown in
pigure RFWF


  „—˜le RFPTX egeEspe™i(™ fertility r—tes @per I HHH womenA in w—l—wi in PHHH —nd PHHRF




                                                 ege qroup

                    ISEIW     PHEPR      PSEPW      QHEQR     QSEQW      RHERR      RSERW

  PHHH whrƒ         IUP       QHS        PUP        PIW       ITU        WR         RI

  PHHR whrƒ         ITP       PWQ        PSR        PPP       ITQ        VH         QS


                       ƒour™eX xƒy —nd y‚g w—™ro @PHHIY PHHSAF



es we required estim—tes of eƒp‚s for —ll the ye—rs st—rting from IWWW up to PHHR in order
to ™ompute estim—tes of the num˜er of ™hildren ˜orn to —ll women during this periodD we
—ssumed — ™onst—nt —nnu—l r—te of ™h—nge in eƒp‚s ˜etween PHHH —nd PHHR —nd —pplied line—r
interpol—tion to the eƒp‚s presented in „—˜le RFPT —˜ove to derive estim—tes of eƒp‚s for
the missing ye—rsF „—˜le RFPU shows the ™omputed estim—tes of eƒp‚s together with those
in „—˜le RFPTF „he set of eƒp‚s presented in „—˜le RFPU —re the ones th—t were used in
™omputing estim—tes of the num˜er of ™hildren ˜orn to —ll women in w—l—wi during the
period from IWWW to PHHRF




                                           VQ
                             350



                                                                                      2000 MDHS
                                                                                q     2004 MDHS
                             300




                                             q
                             250




                                                     q
    Births per 1 000 women




                                                                  q
                             200




                                     q                                    q
                             150
                             100




                                                                                      q
                             50




                                                                                              q
                             0




                                   15−19   20−24   25−29        30−34   35−39       40−44   45−49

                                                           Age group


pigure RFWX gomp—rison of —geEspe™i(™ fertility r—tes @per I HHH womenA in w—l—wi in PHHH

—nd PHHRF




                                                           VR
„—˜le RFPUX istim—tes of —geEspe™i(™ fertility r—tes @per I HHH womenA for —ll women in

w—l—wi from IWWW to PHHRF




   ege group          IWWW          PHHH       PHHI        PHHP         PHHQ         PHHR

   ISEIW              IUR           IUP        IUH         ITU          ITR          ITP

   PHEPR              QHV           QHS        QHP         PWW          PWT          PWQ

   PSEPW              PUT           PUP        PTV         PTQ          PSV          PSR

   QHEQR              PIV           PIW        PPH         PPH          PPI          PPP

   QSEQW              ITV           ITU        ITT         ITS          ITR          ITQ

   RHERR              WV            WR         WH          VU           VR           VH

   RSERW              RP            RI         RH          QV           QT           QS



istim—tes of eƒp‚s presented in „—˜le RFPU were —pplied to the estim—tes of the tot—l
popul—tion of women —ged ISERW in w—l—wi presented in „—˜les RFIQ to RFPH —nd yielded —
r—nge of estim—tes of num˜er of ™hildren ˜orn to —ll women in w—l—wi from IWWW to PHHR
presented in „—˜les RFPV to RFQIF


istim—tes of the num˜er of ™hildren ˜orn to —ll women followed — simil—r p—ttern —s estim—tes
of the tot—l num˜er of women of reprodu™tive —geF „he highest num˜ers of ˜irths were those
˜—sed on estim—tes of women using the south f—mily of go—leEhemeny wv„ system —nd the
lowest were those ˜—sed on the …x p—r i—st wv„ systemF es evident in pigure RFIHD the
estim—tes of the —nnu—l num˜er of ˜irths to —ll women in w—l—wi ˜—sed on the three wv„s
—re —lmost ™omp—r—˜le to the estim—tes produ™ed ˜y …xsgip @ PHHSD PHHRD PHHQ —nd PHHPAF




                                             VS
„—˜le RFPVX istim—tes of the num˜er of ˜irths to —ll women in w—l—wi from IWWW to PHHR for

estim—tes of women ˜—sed on ‡ry wv„ systemF

   ege group       IWWW        PHHH        PHHI        PHHP        PHHQ       PHHR

   ISEIW           WU RVP      WV HUV      WV WWP      WW SST      IHH PUT    IHI TVV

   PHEPR           ISW RSQ     ISV RSV     ISU QQH     IST STU     IST SHV    ISU IHQ

   PSEPW           IPI SPW     IPP STS     IPQ HQH     IPP RVW     IPI RSV    IPH QVI

   QHEQR           TW IIV      UP TPQ      UT QTS      UW STS      VP TST     VS QSR

   QSEQW           RH SHU      RH TRU      RH WQU      RI SQQ      RP SHH     RQ WIR

   RHERR           IW IVS      IV SVV      IU WTT      IU SPT      IU HUV     IT RII

   RSERW           T TIQ       T SWQ       T SRP       T QHP       T HRP      S WQW

   „ot—l ˜irths    SIQ VVU     SIU SSP     SPI ITP     SPQ SQV     SPT SIV    SQH UWH



„—˜le RFPWX istim—tes of the num˜er of ˜irths to —ll women in w—l—wi from IWWW to PHHR for

estim—tes of women ˜—sed on …x p—r i—st wv„ systemF

   ege group       IWWW        PHHH        PHHI        PHHP        PHHQ       PHHR

   ISEIW           WU IRR      WU RSP      WV IPQ      WV RWI      WW HSI     IHH QTP

   PHEPR           ISW SPP     ISV RVW     ISU PIU     IST PHR     ISS UUW    IST HVS

   PSEPW           IPP SUI     IPR RWT     IPS TWI     IPS UHR     IPS HSS    IPR PSS

   QHEQR           UH PHR      UR VRU      UW UWP      VR PPU      VV SVQ     WP RST

   QSEQW           RI PHH      RP HTW      RQ IQR      RR SUR      RT RUV     RV VUW

   RHERR           IW RQP      IW HWV      IV USP      IV TII      IV RUR     IV HWQ

   RSERW           T TQU       T TSV       T TTQ       T RWH       T QHT      T PWI

   „ot—l ˜irths    SIT UIH     SPQ IHW     SPW QUP     SQR QHI     SQW UPT    SRT RPI




                                           VT
„—˜le RFQHX istim—tes of the num˜er of ˜irths to —ll women in w—l—wi from IWWW to PHHR for

estim—tes of women ˜—sed on go—leEhemeny xorth wv„ systemF

   ege group       IWWW        PHHH        PHHI        PHHP        PHHQ       PHHR

   ISEIW           WU QTT      WU VIP      WV SRI      WV VVW      WW QSV     IHH SIU

   PHEPR           ITH IWV     ISW USH     ISV WTV     ISV QSV     ISV PTT    ISV UUW

   PSEPW           IPQ ITQ     IPS TTP     IPU RHH     IPU WHU     IPU UHR    IPU PUT

   QHEQR           UH SSH      US STW      VH WPH      VS UUU      WH SUS     WR VVS

   QSEQW           RI RII      RP RWH      RQ UTU      RS RPT      RU STP     SH PIT

   RHERR           IW SSR      IW QPW      IW HVP      IW HQH      IV WUS     IV TTP

   RSERW           T TWU       TU UP       T VPV       T TWR       T SRQ      T STP

   „ot—l ˜irths    SIV WQW     SPU QVR     SQS SHT     SRP HVI     SRV WVQ    SST VWU



„—˜le RFQIX istim—tes of the num˜er of ˜irths to —ll women in w—l—wi from IWWW to PHHR for

estim—tes of women ˜—sed on go—leEhemeny ƒouth wv„ systemF


   ege group       IWWW        PHHH        PHHI        PHHP        PHHQ       PHHR

   ISEIW           WU RPI      WU WRH      WV UTP      WW PIV      WW VIS     IHI IQT

   PHEPR           ITH PSV     ISW VUH     ISW ISQ     ISV THW     ISV SVW    ISW PHV

   PSEPW           IPQ PUT     IPS VUT     IPU UHH     IPV PUS     IPV IPR    IPU USP

   QHEQR           UH TUQ      US VIV      VI PWV      VT PVH      WI PHH     WS TPT

   QSEQW           RI SIQ      RP TWQ      RR HTW      RS VPV      RV HUH     SH VQP

   RHERR           IW TIH      IW RQV      IW PQW      IW PQS      IW PPR     IV WSH

   RSERW           T UIT       T VII       T VVT       T UUH       T TQR      T TUI

   „ot—l ˜irths    SIW RTU     SPV RRT     SQU IHU     SRR PIS     SSI TST    STH IUS



                                           VU
                     700000
                                                                             WHO
                                                                             UNICEF
                                                                             UN Far East
                     600000                                                  Coale-Demeny North
                                                                             Coale-Demeny South



                     500000
  Number of births




                     400000




                     300000




                     200000




                     100000




                         0
                              1999   2000   2001          2002     2003           2004
                                                   Year




pigure RFIHX gomp—rison of estim—tes of the num˜er of ˜irths to —ll women in w—l—wi from

IWWW to PHHRF




                                              VV
4.5 Estimates of the number of children infected with

        HIV through mother-to-child transmission

istim—tes of the num˜er of ™hildren infe™ted with rs† through w„g„ st—rted with estim—tes
of the num˜er of ™hildren ˜orn to —ll rs† infe™ted womenF essuming th—t fertility —mong
ISEIW ye—rs old women is 50% higher —mong rs†Epositive th—n rs†Eneg—tive women —nd
th—t fertility —mong women PHERW ye—rs old is 20% lower —mong rs†Epositive women th—n
rs†Eneg—tive women we —pplied equ—tion QFQR outlined in the methodology se™tion @se™tion
QFIFSA —nd estim—tes of eƒp‚s for —ll women in w—l—wi from IWWW to PHHR presented in „—˜le
RFPU to estim—tes of rs†Einfe™ted women presented in „—˜les RFPI to RFPR —nd estim—tes of
rs†Eneg—tive women to o˜t—in estim—tes of num˜er of ˜irths to rs†Einfe™ted womenF „he
estim—ted num˜ers of ™hildren ˜orn to rs†Einfe™ted women in w—l—wi from IWWW to PHHR —re
presented in „—˜les RFQP to RFQSF

„—˜le RFQPX istim—tes of the num˜er of ˜irths to rs†Einfe™ted women in w—l—wi from IWWW

to PHHR for estim—tes of women ˜—sed on ‡ry wv„ systemF

   ege group         IWWW           PHHH     PHHI       PHHP        PHHQ       PHHR

   ISEIW             V SRU          U HQV    S SIH      T QPV       U ITQ      T IIU

   PHEPR             PT PVW         PQ HWH   IW WQW     IW UTI      IW TUP     IU WRH

   PSEPW             PH UIP         IW RIQ   IV HIQ     IU HIW      IS WUP     IS SRT

   QHEQR             II VUI         II WHV   II WQI     IQ HQI      IR ITR     IQ RPP

   QSEQW             T RUU          T THV    T UTS      T PWH       S VSQ      T IQV

   RHERR             P STP          P PQS    I WPP      P RSS       P WTT      P QVW

   RSERW             VVQ            UWQ      UHH        VVQ         I HSH      VTS

   „ot—l ˜irths      UU QRI         UI HVS   TR UVH     TS UTU      TT VRH     TP RIU




                                             VW
„—˜le RFQQX istim—tes of the num˜er of ˜irths to rs†Einfe™ted women in w—l—wi from IWWW

to PHHR for estim—tes of women ˜—sed on …x p—r i—st wv„ systemF

   ege group        IWWW       PHHH       PHHI        PHHP        PHHQ       PHHR

   ISEIW            V SIV      T WWQ      S RTP       T PTH       U HUS      T HQV

   PHEPR            PT QHI     PQ HWR     IW WPR      IW UIS      IW SVI     IU VPR

   PSEPW            PH VWH     IW UIV     IV RHQ      IU RTT      IT RRS     IT HRU

   QHEQR            IP HSU     IP PUQ     IP RTT      IQ UWS      IS IVH     IR SQW

   QSEQW            T SVV      T VQW      U IPV       T USH       T RHI      T VQI

   RHERR            P SWS      P PWT      P HHT       P THU       Q PHW      P TQR

   RSERW            VVT        VHH        UIQ         WHW         I HWS      WIT

   „ot—l ˜irths     UU VQS     UP HIQ     TT IHP      TU SHP      TV WVT     TR VPW



„—˜le RFQRX istim—tes of the num˜er of ˜irths to rs†Einfe™ted women in w—l—wi from IWWW

to PHHR for estim—tes of women ˜—sed on go—leEhemeny xorth wv„ systemF

   ege group        IWWW       PHHH       PHHI        PHHP        PHHQ       PHHR

   ISEIW            V SQU      U HIW      S RVS       T PVT       U HWU      T HRU

   PHEPR            PT RIP     PQ PUV     PH IRU      IW WVU      IW VWQ     IV IQI

   PSEPW            PH WWI     IW WHQ     IV TSQ      IU UUP      IT UWQ     IT RQU

   QHEQR            IP IIT     IP QWI     IP TRQ      IR HRW      IS SPI     IR WPI

   QSEQW            T TPP      T WHV      U PQP       T VUW       T SSH      U HIV

   RHERR            P TII      P QPR      P HRP       P TTT       QP WT      P UIU

   RSERW            VWR        VIR        UQI         WQV         I IQU      WSS

   „ot—l ˜irths     UV IVQ     UP TQU     TT WQQ      TV SUU      UH PVU     TT PPT




                                          WH
„—˜le RFQSX istim—tes of the num˜er of ˜irths to rs†Einfe™ted women in w—l—wi from IWWW

to PHHR for estim—tes of women ˜—sed on go—leEhemeny ƒouth wv„ systemF

   ege group        IWWW       PHHH       PHHI        PHHP       PHHQ        PHHR

   ISEIW            V SRP      UH PV      S RWU       T QHU      U IQH       T HVR

   PHEPR            PT RPP     PQ PWT     PH IUH      PH HIW     IW WQR      IV IVH

   PSEPW            PI HIH     IW WQU     IV TWU      IU VPQ     IT VRV      IT RWW

   QHEQR            IP IQV     IP RQP     IP UHP      IR IQI     IS TPV      IS HQV

   QSEQW            T TQV      T WRI      U PVP       T WRH      T TPH       U IHR

   RHERR            P TIV      P QQU      P HSV       P TWR      Q QQW       P USW

   RSERW            VWU        VIW        UQU         WRV        I ISP       WUI

   „ot—l ˜irths     UV PTS     UP UWH     TU IRQ      TV VTP     UH TSI      TT TQS




                                          WI
istim—tes of the num˜er of ™hildren infe™ted with rs† through w„g„ were then ™omputed
˜y —pplying €„‚s to the estim—tes of the tot—l num˜ers of ™hildren ˜orn to rs†Einfe™ted
women in w—l—wi from IWWW to PHHR presented in „—˜les RFQP to RFQSF „hree s™en—rios of
€„‚s in the —˜sen™e of —ny intervention @€„‚ equ—l to 30%D €„‚ equ—l to 32% —nd €„‚
equ—l to 30%A were ™onsidered together with their ™orresponding def—ult v—lues from esw
for rs† infe™ted pregn—nt women who re™eived xevir—pine presented in „—˜le QFVF „he (n—l
estim—tes of the num˜er of ™hildren infe™ted with rs† through w„g„ in w—l—wi from IWWW
to PHHR under the three s™en—rios —re presented in „—˜les RFQT to RFQWF

„—˜le RFQTX istim—tes of the num˜er of ™hildren infe™ted with rs† through w„g„ in w—l—wi

from IWWW to PHHR for estim—tes of women ˜—sed on ‡ry wv„ systemF


   f—se tr—nsmission r—te     IWWW      PHHH       PHHI      PHHP         PHHQ     PHHR

   30%                        PQ PHP    PI QPT     IW RQR    IW TIP       IW URR   IV QRR

   32%                        PR URW    PP URU     PH UQH    PH WUV       PI PIQ   IW UST

   35%                        PU HTW    PR VVH     PP TUQ    PP WTH       PQ PRH   PI TST



„—˜le RFQUX istim—tes of the num˜er of ™hildren infe™ted with rs† through w„g„ in w—l—wi

from IWWW to PHHR for estim—tes of women ˜—sed on …x p—r i—st wv„ systemF


   f—se tr—nsmission r—te     IWWW      PHHH       PHHI      PHHP         PHHQ     PHHR

   30%                        PQ QSH    PI THR     IW VQI    PH IQQ       PH QVV   IW HTV

   32%                        PR WHU    PQ HRR     PI ISQ    PI SQQ       PI WHH   PH SPV

   35%                        PU PRP    PS PHS     PQ IQT    PQ STU       PQ WWI   PP SHH




                                             WP
„—˜le RFQVX istim—tes of the num˜er of ™hildren infe™ted with rs† through w„g„ in w—l—wi

from IWWW to PHHR for estim—tes of women ˜—sed on go—leEhemeny xorth wv„ systemF


   f—se tr—nsmission r—te   IWWW      PHHH      PHHI      PHHP      PHHQ      PHHR

   30%                      PQ RSS    PI UWI    PH HVH    PH RST    PH UUV    IW RVU

   32%                      PS HIW    PQ PRR    PI RIW    PI VUU    PP QIT    PH WUS

   35%                      PU QTR    PS RPQ    PQ RPU    PQ WRQ    PR RRU    PP WVW



„—˜le RFQWX istim—tes of the num˜er of ™hildren infe™ted with rs† through w„g„ in w—l—wi

from IWWW to PHHR for estim—tes of women ˜—sed on go—leEhemeny ƒouth wv„ systemF


   f—se tr—nsmission r—te   IWWW      PHHH      PHHI      PHHP      PHHQ      PHHR

   30%                      PQ RVH    PI VQU    PH IRQ    PH SRI    PH VVV    IW TIH

   32%                      PS HRS    PQ PWQ    PI RVT    PI WTW    PP RQP    PI IHT

   35%                      PU QWQ    PS RUT    PQ SHH    PR HRQ    PR SUR    PQ IQP



4.6 Estimates of the number of HIV infected children
    dying before the age of ve years during the period
    2000 to 2004

‡e ™omputed estim—tes of the tot—l num˜er of ™hildren infe™ted with rs† through w„g„
th—t died ˜efore —tt—ining the —ge of (ve ye—rs during the period PHHH to PHHR in w—l—wi
˜y —pplying the surviv—l ™urve of ™hildren infe™ted with rs† through w„g„ presented ˜y
equ—tion QFQU in the methodology se™tion @se™tion QFIFTA to the estim—tes of the num˜ers
of ™hildren infe™ted with rs† through w„g„ during the period PHHH to PHHR presented in
„—˜les RFQT to RFQWF ‡e —ssumed —n even distri˜ution in the num˜er of ™hildren infe™ted

                                          WQ
with rs† through w„g„ in e—™h ye—r —nd st—rted ˜y ™omputing estim—tes of the —ver—ge
num˜er of ™hildren infe™ted with rs† through w„g„ per month in e—™h ye—r from PHHH to
PHHRF „here—fterD —pplying the surviv—l ™urve of ™hildren infe™ted with rs† through w„g„
presented ˜y equ—tion QFQU to the monthly estim—tes of the num˜er of ™hildren infe™ted with
rs† through w„g„ we o˜t—ined estim—tes of the tot—l num˜er of ™hildren infe™ted with
rs† through w„g„ th—t died ˜efore their (fth ˜irthd—y in w—l—wi during the period PHHH
to PHHR presented in „—˜le RFRHF es shown in „—˜le RFRHD for —ll the three s™en—rios of €„‚sD
the south f—mily of go—leEhemeny wv„ system yielded the highest estim—tes of the tot—l
num˜er of underE(ve de—ths —mong rs†Einfe™ted ™hildren in w—l—wi during the period PHHH
to PHHRF

„—˜le RFRHX istim—tes of the tot—l num˜er of ™hildren infe™ted with rs† through w„g„

dying ˜efore the —ge of (ve ye—rs during the period PHHH to PHHR in w—l—wiF


                                                      wv„

 f—se                ‡ry      …x p—r i—st    go—leEhemeny xorth      go—leEhemeny ƒouth

 tr—nsmission r—te

 30%                 RS IWQ   RT QSW         RU HUT                  RU PTU

 32%                 RV QVV   RW TPR         SH QVS                  SH SVT

 35%                 SP WTQ   SR QPI         SS ISP                  SS QVT




                                             WR
4.7 Estimates of under-ve mortality directly attributable
    to HIV/AIDS in Malawi during the period 2000 to
    2004

„he pro™ess of estim—ting the proportion of underE(ve mort—lity th—t is dire™tly —ttri˜ut—˜le
to rs†Geshƒ in w—l—wi during the period PHHH to PHHR involved ™omputing estim—tes of
— ™orre™tion for ™ompeting ™—uses of mort—lity —mong rs†Einfe™ted ™hildren —nd estim—tes
of …Sw‚ due to rs†Geshƒ —s des™ri˜ed in the methodology se™tion @se™tion QFIFUAF …sing
the PHHR w—l—wi hrƒ …Sw‚ estim—te of IQQ de—ths per IHHH live ˜irths —nd the results
presented in „—˜les RFPW to RFQI —nd RFRH we o˜t—ined estim—tes of …Sw‚ due to rs†Geshƒ
—nd the proportion of underE(ve mort—lity dire™tly —ttri˜ut—˜le to rs†Geshƒ in w—l—wi
during the period PHHH to PHHR presented in „—˜le RFRIF es „—˜le RFRI showsD —˜out IIFU7
to IQFV7 of underE(ve de—ths th—t o™™ured in w—l—wi during the period PHHH to PHHR were
due to p—edi—tri™ rs†Einfe™tion ™ontr—™ted through w„g„F „husD on —ver—ge rs†Geshƒ
dire™tly ™—used —˜out IPFU7 of underE(ve mort—lity in w—l—wi during the period PHHH to
PHHRF


yn the one h—ndD the results presented in „—˜le RFRI indi™—te th—t the …Sw‚ due to
rs†Geshƒ —nd the proportion of underE(ve mort—lity dire™tly —ttri˜ut—˜le to rs†Geshƒ
depends on the €„‚F e low €„‚ yields — low …Sw‚ due to rs†Geshƒ —s well —s proportion
of underE(ve mort—lity dire™tly —ttri˜ut—˜le to rs†GeshƒF yn the other h—ndD the results
presented in „—˜le RFRI indi™—te th—t the type of wv„ used h—s no e'e™t on the estim—tes
of …Sw‚ due to rs†Geshƒ —nd the proportion of underE(ve mort—lity dire™tly —ttri˜ut—˜le
to rs†GeshƒF „he minor di'eren™es in the estim—tes of …Sw‚ due to rs†Geshƒ —nd the
proportion of underE(ve mort—lity dire™tly —ttri˜ut—˜le to rs†Geshƒ for the s—me €„‚
—™ross the wv„s —re due to roundingF




                                             WS
„—˜le RFRIX istim—tes of …Sw‚ due to rs†Geshƒ —nd proportion of underE(ve mort—lity

—ttri˜ut—˜le to rs†Geshƒ in w—l—wi during the period PHHH to PHHRF


      wv„             f—se           …Sw‚ @per IHHHA   €roportion of

                      tr—nsmission   due to rs†Geshƒ   underE(ve mort—lity

                      r—te                             —ttri˜ut—˜le to rs†Geshƒ

      ‡ry             30%            ISFPT             IIFRU7

                      32%            ITFQU             IPFQH7

                      35%            IUFWR             IQFRW7



      …x p—r          30%            ISFQR             IIFSQ7

      i—st            32%            ITFRR             IPFQT7

                      35%            IVFHQ             IQFST7



      go—leEhemeny    QH7            ISFQT             IIFSS7

      xorth           QP7            ITFRT             IPFQV7

                      QS7            IVFHS             IQFSU7



      go—leEhemeny    QH7            ISFQT             IIFSS7

      ƒouth           QP7            ITFRT             IPFQV7

                      QS7            IVFHT             IQFSV7




                                             WT
4.8 HIV/AIDS, PMTCT programs and U5MR reduction

        goals in Malawi

sn this se™tion we present results on how rs†Geshƒ dire™tly —'e™ted progress tow—rds
re—™hing over—ll underE(ve mort—lity redu™tion go—ls in w—l—wiF „he se™tion —lso presents
results on the potenti—l ™ontri˜ution of €w„g„ progr—ms ™urrently ˜eing implemented in
w—l—wi tow—rds re—™hing the ™ountry9s over—ll underE(ve mort—lity redu™tion go—lsF


‡e —ssessed how rs†Geshƒ dire™tly —'e™ted progress tow—rds re—™hing over—ll underE(ve
mort—lity redu™tion go—ls in w—l—wi ˜y ™omp—ring estim—tes of …Sw‚ due to nonErs†Geshƒ
™—uses during the period PHHH to PHHR with —n estim—te of the level of …Sw‚ needed to ˜e
—™hieved ˜y PHHR if w—l—wi w—s on tr—™k tow—rds re—™hing the underE(ve mort—lity redu™tion
whqF es shown in pigure RFIID —fter dis™ounting the dire™t e'e™ts of rs†GeshƒD …Sw‚ in
w—l—wi during the period PHHH to PHHR is ˜elow the level it w—s required to h—ve re—™hed
h—d the ™ountry ˜een on tr—™k tow—rds re—™hing the over—ll underE(ve mort—lity redu™tion
whq depi™ted ˜y the dotted horizont—l lineF


„o —ssess the potenti—l ™ontri˜ution €w„g„ progr—ms ™urrently ˜eing implemented in
w—l—wi would m—ke tow—rds re—™hing w—l—wi9s over—ll underE(ve mort—lity redu™tion go—ls
we dis™ounted —n estim—ted num˜er of underE(ve de—ths th—t would ˜e —verted in w—l—wi
during the period PHHH to PHHR if there w—s IHH7 ™over—ge of €w„g„ servi™es from the
estim—ted num˜er of underE(ve de—ths from —ll ™—uses ™omputed using the PHHR w—l—wi
hrƒ estim—te of …Sw‚ from —ll ™—uses in w—l—wi during the period PHHH to PHHRF ‡e then
™omputed —n estim—te of …Sw‚ with IHH7 €w„g„ ™over—ge whi™h we —g—in ™omp—red with
the level of …Sw‚ needed to ˜e —™hieved ˜y PHHR if w—l—wi w—s on tr—™k tow—rds re—™hing the
underE(ve mort—lity redu™tion whqF sn our —n—lysis we only ™onsidered — IHH7 ™over—ge
of tre—tment with x†€ —s it is the only est—˜lished €w„g„ servi™e —v—il—˜le in w—l—wi
—t the time of ™ondu™ting this study @w—l—wi rs† —nd eshƒ wonitoring —nd iv—lu—tion
‚eport PHHS AF es pigure RFII further showsD — IHH7 ™over—ge of tre—tment with xevir—pine


                                            WU
for €w„g„ ™ould h—ve redu™ed the …Sw‚ in w—l—wi during the period PHHH to PHHR to
—lmost the level needed for ˜eing on tr—™k tow—rds re—™hing the over—ll underE(ve mort—lity
redu™tion whq only if the €„‚ w—s QH7F



                                                                             PTR=30%
                                                                             PTR=32%
                  140




                                                                             PTR=35%
                                                                             2004 MDHS
                  120
                  100
U5MR (per 1000)

                  80
                  60
                  40
                  20
                  0




                        2004 MDHS       100% NVP coverage                  No HIV


pigure RFIIX rs†GeshƒD €w„g„ progr—ms —nd …Sw‚ redu™tion go—ls in w—l—wi —ssuming

— €„‚ of QH7F



                                            WV
Chapter 5


Discussion and conclusions


„his study h—s estim—ted the proportion of underE(ve mort—lity th—t is dire™tly —ttri˜ut—˜le
to rs†Geshƒ in w—l—wi during the period PHHH to PHHRF „he steps involved in deriving the
estim—tes in™luded estim—ting rs† prev—len™e —mong women of reprodu™tive —ge using exg
surveill—n™e survey d—t— —nd rs† prev—len™e results from the PHHR w—l—wi hrƒD estim—ting
the num˜er of women of reprodu™tive —geD estim—ting the num˜er of rs† infe™ted womenD
estim—ting the tot—l num˜er of ˜irths in e—™h ye—rD estim—ting the tot—l num˜er of ™hildren
infe™ted with rs† through w„g„D —nd estim—ting the …Sw‚ due to rs†GeshƒF „o —
l—rge extentD the study —dopted the methods —nd pro™edures re™ommended ˜y …xeshƒ —nd
‡ry th—t —re implemented in hem€roj —nd eshƒ smp—™t wodel ™omputer p—™k—gesF „o
—™™ount for un™ert—inties in the estim—tesD the study used surviv—l r—tios from four sets of
wv„s deemed to (t the w—l—wi—n mort—lity experien™e —nd three sets of €„‚ —ssumed to
—rise in ƒƒe ™ountries ˜y other rese—r™hers @ƒtoverD PHHSY „he …xeshƒ ‚eferen™e qroup on
istim—tesD wodels —nd €roje™tionsD PHHQY —˜— et —lFD PHHQY ‡—lker et —lFD PHHPY edetunjiD
PHHHAF


fe™—use there were only few d—t— points for rs† prev—len™e from the exg surveill—n™e
surveys we did not derive smoothed n—tion—l estim—tes of rs† prev—len™e for women of



                                            WW
reprodu™tive —ge in w—l—wi during the period PHHH to PHHRF xeverthelessD using widely used
—nd re™ommended pro™eduresD ™onsider—˜le e'orts were m—de to —djust the rs† prev—len™e
d—t— from the exg surveill—n™e surveys to derive pl—usi˜le estim—tes of rs† prev—len™e in
women of reprodu™tive —ge in w—l—wiF


„he (nding th—t estim—tes of n—tion—l rs† prev—len™e for women —ged IS to RW derived
in this study —re lower th—n those produ™ed ˜y xeg @PHHSA suggests th—t estim—tes of
the proportion of underE(ve mort—lity th—t is dire™tly —ttri˜ut—˜le to rs†Geshƒ in w—l—wi
during the period PHHH to PHHR o˜t—ined in this study m—y ˜e slightly lower th—n those th—t
™ould ˜e ˜—sed on xeg @PHHSA9s n—tion—l rs† prev—len™e estim—tesF roweverD sin™e the
di'eren™es ˜etween this study9s estim—tes of n—tion—l rs† prev—len™e —nd those produ™ed ˜y
xeg @PHHSA —re very sm—ll the ™on™lusions dr—wn from the (ndings of this study would not
˜e signi(™—ntly —ltered if estim—tes of n—tion—l rs† prev—len™e for women of reprodu™tive
—ge in w—l—wi during the period PHHH to PHHR were derived from those produ™ed ˜y xeg
@PHHSAF


„he (nding th—t the estim—tes of the num˜er of women of reprodu™tive —ge o˜t—ined in
this study —re very ™lose to those produ™ed ˜y xeg @PHHSA —nd xƒy @PHHHA —lso indi™—tes
th—t the estim—tes o˜t—ined in this study —re within pl—usi˜le ˜ounds —s f—r —s the num˜er
of women of reprodu™tive —ge in w—l—wi —re ™on™ernedF „he sm—ll di'eren™e ˜etween the
estim—tes of the num˜er of ˜irths in e—™h ye—r o˜t—ined in this study —nd those produ™ed
˜y …xsgip @PHHSD PHHRD PHHQ —nd PHHPA is — further indi™—tion th—t the estim—tes of the
num˜er of ˜irths in w—l—wi during the period PHHH to PHHR o˜t—ined in this study —re within
pl—usi˜le ˜oundsF


elthough the estim—tes of n—tion—l rs† prev—len™e for women of reprodu™tive —ge m—y
˜e —'e™ted ˜y the sm—ll num˜ers of women s—mpled in —ges RH ye—rs —nd —˜ove in exg
surveill—n™e surveysD owing to low fertility in su™h women @xƒy —nd y‚g w—™roD PHHSA we
expe™t very little e'e™t on the estim—tes of the proportion of underE(ve mort—lity th—t is
dire™tly —ttri˜ut—˜le to rs†GeshƒF purthermoreD —lthough our estim—tes of the proportion
of underE(ve mort—lity th—t is dire™tly —ttri˜ut—˜le to rs†Geshƒ in w—l—wi during the


                                           IHH
period PHHH to PHHR m—y di'er from those th—t m—y ˜e produ™ed ˜y other rese—r™hers —nd
s™ientists owing to v—rying demogr—phi™ —nd epidemiologi™—l —ssumptionsD we ˜elieve th—t the
results o˜t—ined in this study ™—nnot gre—tly diverge from those th—t ™—n ˜e produ™ed using
the ‡ryG…xeshƒ re™ommended methods —nd pro™eduresD p—rti™ul—rly those implemented
in the ƒpe™trum p—™k—ge @qhys et —lFD PHHRAF „his ™l—im is supported ˜y f—™t th—t to —
l—rge extent this study used methods —nd pro™edures simil—r to those implemented in the
ƒpe™trum p—™k—ge —nd there —re sm—ll di'eren™es ˜etween the estim—tes of some of the
p—r—meters needed to ™ompute estim—tes of the proportion of underE(ve mort—lity th—t is
dire™tly —ttri˜ut—˜le to rs†Geshƒ o˜t—ined in this study —nd those produ™ed ˜y other
rese—r™hers —nd experts using simil—r methods —nd pro™edures —s illustr—ted in the results
se™tionF


„—king the v—lue IQQ per I HHH from PHHR w—l—wi hrƒ —s —n estim—te of …Sw‚ from —ll
™—uses in w—l—wi during the period PHHH to PHHRD the m—in (nding of this study is th—t
the proportion of underE(ve mort—lity th—t is dire™tly —ttri˜ut—˜le to rs†Geshƒ in w—l—wi
during this period r—nges from IIFSQ7 to IQFSV7F „hese proportions ™orrespond to €„‚s
r—nging from QH7 to QS7 respe™tivelyF „hese estim—tes —re slightly —˜ove the estim—te of
VFW7 produ™ed ˜y ‡—lker et —lF @PHHPA for IWWW —nd IIFR7 produ™ed ˜y —˜— et —lF @PHHQA
for PHHIF ƒin™e this study used —ppro—™hes th—t —re to — gre—t extent simil—r to those used
˜y ‡—lker et —lF @PHHPA in estim—ting the proportion of underE(ve mort—lity th—t is dire™tly
—ttri˜ut—˜le to rs†GeshƒD —lthough ‡—lker et —lF @PHHHA9s estim—tes ™over single ye—rsD this
study9s (nding suggests — possi˜le slight in™re—se in the dire™t ™ontri˜ution of rs†Geshƒ
tow—rds underE(ve mort—lity in w—l—wi during the period PHHH to PHHR ™omp—red to previous
ye—rsF elthough —˜— et —lF @PHHQA used — di'erent —ppro—™h from the one used in this studyD
this study9s results ™—n —lso ˜e ™omp—red to their results sin™e it h—s ˜een found th—t wv„
—ssumptions do not —'e™t estim—tes of the proportion of underE(ve mort—lity —ttri˜ut—˜le
to rs†GeshƒF ƒin™e —˜— et —lF @PHHQAD —lthough —g—in their estim—tes —re for single ye—rsD
used — €„‚ of QS7 whi™h ™orresponds to —n estim—ted proportion of underE(ve mort—lity
dirre™tly —ttri˜ut—˜le to rs†Geshƒ of IQFSU7 from this studyD the s—me ™on™lusion th—t
the proportion of underE(ve mort—lity dire™tly —ttri˜ut—˜le to rs†Geshƒ in w—l—wi slightly


                                            IHI
in™re—sed during the period PHHH to PHHR ™omp—red to the previous ye—rs ™—n —lso ˜e dr—wnF


prom this study9s —ssessment of how rs†Geshƒ dire™tly —'e™ted progress tow—rds re—™hing
over—ll underE(ve mort—lity redu™tion go—ls in w—l—wi it h—s ˜een noted th—t rs†Geshƒ
signi(™—ntly h—mpered the —™hievement of the level of …Sw‚ needed for w—l—wi to ˜e on
tr—™k tow—rds re—™hing the over—ll underE(ve mort—lity redu™tion whqF „his (nding indi™—tes
th—t —verting rs†Geshƒ rel—ted de—ths —mong ™hildren ˜elow the —ge of (ve ye—rs ™ould help
in ˜ringing down the over—ll underE(ve mort—lity in w—l—wi to t—rgeted levelsF


e distin™t fe—ture of this study is th—t —p—rt from providing insights on how rs†Geshƒ
neg—tively —'e™ted the —™hievement of over—ll underE(ve mort—lity redu™tion go—ls in w—l—wi
it h—s —lso —ttempted to —ssess the potenti—l ™ontri˜ution of s™—ling up ™over—ge of €w„g„
servi™es ™urrently ˜eing provided in w—l—wi tow—rds —™hieving the ™ountry9s over—ll underE
(ve mort—lity redu™tion go—lsF „he (nding th—t IHH7 ™over—ge of tre—tment with xevir—pine
to prevent w„g„ of rs† infe™tion would not h—ve gre—tely ™ontri˜uted tow—rds re—™hing
over—ll underE(ve mort—lity redu™tion go—ls in w—l—wi during the period PHHH to PHHR ™on(rms
th—t only — ™omprehensive set of €w„g„ servi™es whi™h ™omprise prim—ry prevention of
rs† infe™tion —mong women of reprodu™tive —geD prevention of unintended pregn—n™ies
—mong rs†Einfe™ted womenD prevention of w„g„ tr—nsmission of rs† infe™tion from rs†E
infe™ted mothers to their inf—nts —nd the provision of —ppropri—te tre—tmentD ™—re —nd
support to rs†Einfe™ted mothers —nd their inf—nts —nd f—milies would signi(™—ntly ™ontri˜ute
tow—rds re—™hing over—ll underE(ve mort—lity redu™tion go—ls @…xsgipD …xeshƒD ‡ry —nd
…xp€eD PHHRAF „husD in order to m—ke su˜st—nti—l ™ontri˜utions to the ™ountry9s over—ll
underE(ve mort—lity redu™tion go—lsD w—l—wi should intensify its e'orts in —ll the four key
™omponents of — ™omprehensive €w„g„ progr—mF elthough w—l—wi h—s m—de su˜st—nti—l
progress in putting in pl—™e v—rious rs† prevention progr—mmesD —s —rgued ˜y the qlo˜—l
rs† €revention ‡orking qroup @PHHUAD (ndings from the PHHR w—l—wi hrƒ @xƒy —nd y‚g
w—™roD PHHSA —nd the (rst ˜eh—viour—l surveill—n™e survey ™ondu™ted in w—l—wi in PHHR @xeg
et —lFD PHHRA th—t — num˜er of people in w—l—wi still do not h—ve —™™ur—te —nd ™omprehensive
knowledge of rs† —nd others still eng—ge in high risk ˜eh—viourD me—ns th—t individu—ls
—nd ™ommunities in w—l—wi h—ve not yet ˜een re—™hed with the level of prevention ™over—ge

                                            IHP
needed to h—ve — m—jor imp—™tF „he s—me —rgument holds for the m—jority of people who still
rem—in un—w—re of their rs† st—tusF w—l—wi —lso needs to intensify ™over—ge —nd —v—il—˜ility
of rs† testing —nd ™ounselling servi™esF purthermoreD w—l—wi —lso needs to improve on
™over—ge —nd —v—il—˜ility of ™ontr—™eption servi™es espe™i—lly —mong rs†Einfe™ted women —s
— w—y of preventing unintended pregn—n™ies —mong rs†Einfe™ted womenF


e™™ording to xƒy —nd y‚g w—™ro @PHHSAD the other four ™ommuni™—˜le dise—sesX
pneumoni—D di—rrhe—D m—l—ri— —nd me—sles reported to ˜e the le—ding ™—uses of de—ths in
™hildren under the —ge of (ve ye—rs in developing ™ountries ˜y fry™e et —lF @PHHSA —nd ‡ry
@PHHS—A —lso m—ke — su˜st—nti—l ™ontri˜ution to w—l—wi9s underE(ve mort—lityF „hereforeD
—tt—inment of underE(ve mort—lity redu™tion t—rgets in w—l—wi requires ˜oth exp—nsion of
™over—ge —nd —v—il—˜ility of — ™omprehensive set of €w„g„ servi™es —s well —s strengthening
of e'orts to —ddress these other m—jor ™—uses of underE(ve mort—lityF


vike other studies th—t h—ve —ttempted to estim—te the dire™t imp—™t of rs†Geshƒ on ™hild
mort—lity this study does not ™—pture the m—ny indire™t e'e™ts of rs†Geshƒ on underE(ve
mort—lity in w—l—wi @‡—lker et —lFD PHHPY edetunjiD PHHHAF „hereforeD due to the —ddition—l
indire™t e'e™tsD the over—ll ™ontri˜ution of rs†Geshƒ tow—rds underE(ve mort—lity in w—l—wi
during the period PHHH to PHHR is without dou˜t gre—ter th—n o˜t—ined in this studyF


sn summ—ryD there —re two m—in ™on™lusions dr—wn from the (ndings of this studyF yne
m—in ™on™lusion is th—t rs†Geshƒ h—s signi(™—ntly h—mpered —™hievement of w—l—wi9s
over—ll underE(ve mort—lity redu™tion t—rgetsF „he other m—in ™on™lusion is th—t —tt—inment
of further redu™tions in w—l—wi9s underE(ve mort—lity requires ˜oth exp—nsion of ™over—ge
—nd —v—il—˜ility of — ™omprehensive set of €w„g„ servi™es —nd strengthening of e'orts to
—ddress the other m—jor ™—uses of underE(ve mort—lityF roweverD sin™e putting in pl—™e —
™omprehensive set of €w„g„ servi™es —longside strengthening e'orts to —ddress the other
m—jor ™—uses of underE(ve mort—lity would require huge (n—n™i—l resour™esD t—king into
—™™ount w—l—wi9s e™onomi™ situ—tionD we —rgue th—t the —tt—inment of further su˜st—nti—l
redu™tions in w—l—wi9s underE(ve mort—lity in the ye—rs to ™ome depend on (n—n™i—l support
from the ™ountry9s intern—tion—l donorsF …nless more (n—n™i—l support ˜e™ome —v—il—˜le to


                                            IHQ
e'e™tively —ddress rs†Geshƒ —nd the other m—jor ™—uses of underE(ve mort—lityD w—l—wi9s
underE(ve mort—lity will ™ontinue to ˜e one of the highest in the world in the ye—rs to ™omeF




                                            IHR
Appendix



R Code

„his se™tion presents ‚ ™ode used to ™ompute the estim—tes presented in this thesisF „he
™ode w—s written in — num˜er of text (lesF „he ™ode in one (le is linked to the ™ode in —nother
(le ˜y the 4sour™e4 st—tementF ell st—tements ˜eginning with 5 —re ™omments expl—ining
wh—t the ™ode is doingF


 code for estimating proportion of urban population code le name:             …•‚•popnFtxt
     5re—ding the t—˜le ™ont—ining the ur˜—n —nd rur—l popul—tions —s re™orded in IWVU —nd
     IWWV ™ensus reportsF
     ur˜•rur—l`Ere—dFt—˜le@4ur˜—n•rur—l•propFtxt4D he—derapevƒiA
     5™om˜ine —ge groups RHERR —nd RSERW
     ur˜•rur—l•RH•RW`E ur˜•rur—l‘TD“Cur˜•rur—l‘UD“
     5™re—ting — d—t— t—˜le ™ont—ining the ur˜—n —nd rur—l popul—tions with —ge groups
     RHERR —nd RSERW ™om˜ined
     ur˜•rur—l•IS•RW`Er˜ind@ur˜•rur—l‘IXSD“Dur˜•rur—l•RH•RWA
     n—mes@ur˜•rur—l•IS•RWA`E™@4ur˜—n•IWVU4D 4rur—l•IWVU4D
     4ur˜—n•IWWV4D 4rur—l•IWWV4A
     ege`E™@4ISEIW4D 4PHEPR4D 4PSEPW4D 4QHEQR4D 4QSEQW4D 4RHERW4A
     rowFn—mes@ur˜•rur—l•IS•RWA`Eege


                                             IHS
5™omputing proportion ur˜—n in IWVU —nd IWWV
prop•ur˜—nVU`Eur˜•rur—l•IS•RW6ur˜—n•IWVUG@ur˜•rur—l•IS•RW6ur˜—n•IWVU
Cur˜•rur—l•IS•RW6rur—l•IWVUA
prop•ur˜—nWV`Eur˜•rur—l•IS•RW6ur˜—n•IWWVG@ur˜•rur—l•IS•RW6ur˜—n•IWWV
Cur˜•rur—l•IS•RW6rur—l•IWWVA
prop•ur˜—nVU•WV`E™˜ind@prop•ur˜—nVUDprop•ur˜—nWVA
rowFn—mes@prop•ur˜—nVU•WVA`Eege
5™omputing —nnu—l proportion ur˜—n ™h—nge IWVUEIWWV
prop•ur˜—n•VU•WV•—™`E™˜ind@@prop•ur˜—nVU•WV‘DP“Eprop•ur˜—nVU•WV‘DI“AG@IWWVE
IWVUAA
5™re—te — fun™tion for estim—ting —ge spe™i(™ proportion ur˜—n popul—tion for —ny ye—r
from IWVU
prop•ur˜—n`Efun™tion@yA 5 y is —ny ye—r `aIWVU
{interv—l`E@yEIWVUA
prop•ur˜—n`Eprop•ur˜—nVU•WV‘DI“ Cinterv—lBprop•ur˜—n•VU•WV•—™
™˜ind@prop•ur˜—nA
}
5™re—ting — fun™tion for estim—ting —ge spe™i(™ proportion rur—l popul—tion for —ny
ye—r from IWVU
prop•rur—l`Efun™tion@yA 5 y —ny ye—r `aIWVU
{prop•rur—l`EIEprop•ur˜—n@yA
™˜ind@prop•rur—lA
}




                                      IHT
code for adjusting HIV prevalence in non-urban sites for representativeness
   code le name: exg•xyx•…‚fex•VHFtxt
   5 „he ™ode in this (le w—s used to gener—te the results in „—˜le RFT on p—ge TP
   5re—ding the (le ™ont—ining the un—djusted rs† prev—len™e in —ggreg—ted nonEur˜—n
   sites
   xyx•…‚fex`Ere—dFt—˜le@4exg•xyx•…‚fexFtxt4D he—derapevƒiA
   ege`E™@ 4ISEIW4D4PHEPR4D4PSEPW4D4QHEQR4D4QSEQW4D4RHERW4A
   ye—r`E™@IWWWDPHHIDPHHQDPHHSA
   xyx•…‚fex•HFV`Exyx•…‚fexBHFV
   n—mes@xyx•…‚fex•HFVA`Eye—r
   rowFn—mes@xyx•…‚fex•HFVA`Eege



code for estimating unadjusted national HIV prevalence
   code le name: rs†•xe„syxFtxt
   5re—ding in the ™ode ™ont—ining the —djusted —ge spe™i(™ rs† prev—len™e r—tes for
   —ggreg—ted rur—l sitesF
   sour™e@4exg•xyx•…‚fex•VHFtxt4A
   5 re—ding the —ge spe™i(™ rs† r—tes for —ggreg—ted ur˜—n sites
   …‚fex`Ere—dFt—˜le@4exg•…‚fexFtxt4D he—derapevƒiA
   5re—ding the ™ode for ™—l™ul—ting proportion ur˜—n —nd rur—lF
   sour™e@4…•‚•popnFtxt4A
   5™re—ting — d—t— fr—me ™ont—ining —ge spe™i(™ proportion ur˜—n
   for the ye—rs @IWWWDPHHIDPHHQDPHHSA
   prop•ur˜—n•WW•HS`E™˜ind@A
   for @i in seq@IWWWDPHHSDPAA
   {ur˜—n•prop`Eprop•ur˜—n@iA
   prop•ur˜—n•WW•HS`E™˜ind@prop•ur˜—n•WW•HSDur˜—n•propA
   }
   prop•ur˜—n•WW•HS`Ed—t—Ffr—me@prop•ur˜—n•WW•HSA



                                         IHU
n—mes@prop•ur˜—n•WW•HSA`Eye—r
5™re—ting — d—t— fr—me ™ont—ining —ge spe™i(™ proportion rur—l for
the ye—rs IWWWDPHHIDPHHQDPHHS
prop•rur—l•WW•HS`EIEprop•ur˜—n•WW•HS
prop•rur—l•WW•HS`Ed—t—Ffr—me@prop•rur—l•WW•HSA
n—mes@prop•rur—l•WW•HSA`Eye—r
5™—l™ul—ting estim—tes of n—tion—l rs† prev—len™e for the ye—rs
IWWWDPHHIDPHHQDPHHS presented in „—˜le RFU on p—ge TQ
rs†•xe„syx`E@prop•ur˜—n•WW•HSB…‚fexA
C@prop•rur—l•WW•HSBxyx•…‚fex•HFVA
5™—l™ul—ting estim—tes of rs† prev—len™e for the ye—rs PHHHD PHHP —nd PHHR
rs†•xe„syx•PHHH`EHFSB@rs†•xe„syx‘I“Crs†•xe„syx‘P“A
rs†•xe„syx•PHHP`EHFSB@rs†•xe„syx‘P“Crs†•xe„syx‘Q“A
rs†•xe„syx•PHHR`EHFSB@rs†•xe„syx‘Q“Crs†•xe„syx‘R“A
5™re—ting — d—t— fr—me ™ont—ining estim—tes of rs† prev—len™e for PHHHD PHHP —nd PHHR
rs†•xe„syxP`E™˜ind@rs†•xe„syx•PHHHDrs†•xe„syx•PHHPDrs†•xe„syx•PHHRA
n—mes@rs†•xe„syxPA`E™@4PHHH4D4PHHP4D 4PHHR4A
5™re—ting — d—t— fr—me ™ont—ining —ll the estim—tes of rs† prev—len™e from IWWW to
PHHS presented in „—˜le RFV on p—ge TQ
rs†•xe„syx•evv`E—rr—y@Ddima™@TDUAA
rs†•xe„syx•evv`Ed—t—Ffr—me@rs†•xe„syx•evvA
iaseq@IDUDPA
jaseq@PDTDPA
rs†•xe„syx•evv‘Di“`Ers†•xe„syx
rs†•xe„syx•evv‘Dj“`Ers†•xe„syxP
n—mes@rs†•xe„syx•evvA`EIWWWXPHHS
rowFn—mes@rs†•xe„syx•evvA`Eege
rs†•exg•IWWW`Ers†•xe„syx•evv‘DI“
n—mes@rs†•exg•IWWWA`Eege
rs†•exg•IWWWFt—˜le`E—sFt—˜le@rs†•exg•IWWWA


                                     IHV
code for combining HIV prevalence rates for age groups 40-44 and 45-49 from
   the 2004 Malawi DHS report
   code le name: rs†•hrƒFtxt
   5„he ™ode in this (le w—s used to gener—te the results presented in „—˜le RFW on p—ge
   TSF 5™re—ting — ve™tor for num˜er tested
   num˜er•tstd `E™@SHHDTTIDRUUDQVPDPSUDPQSDIUQA
   ege`E™@ 4ISEIW4D4PHEPR4D4PSEPW4D4QHEQR4D4QSEQW4D4RHERW4A
   5™re—ting — ve™tor ™ont—ining the —ge spe™i(™ prev—len™e r—tes for PHHR w—l—wi hrƒ
   report
   rs†•PHHRwhrƒ`E™@QFUDIQFPDISFSDIVFIDIUFHDIUFWDIQFQA
   5™omputing num˜ers rs† positive in e—™h —ge group
   num˜er•postv`E@num˜er•tstdBrs†•PHHRwhrƒAGIHH
   5 rounding the elements in num˜er•postv to whole num˜ers
   num˜er•postvFrounded`Eround@num˜er•postvA
   5—dding the num˜ers positive in —ge groups RHERR —nd RSERW
   num˜er•postv•RH•RW`Enum˜er•postvFrounded‘length@num˜er•postvFroundedA“ C
   num˜er•postvFrounded‘length@num˜er•postvFroundedAEI“
   5—dding the num˜ers tested in —ge groups RHERR —nd RSERW
   num˜er•tstd•RH•RW`Enum˜er•tstd‘length@num˜er•tstdA“
   C num˜er•tstd‘length@num˜er•tstdAEI“
   5 ™—l™ul—ting per™ent—ge positive in RHERW —ge group
   per™nt•postv•RH•RW`E@num˜er•postv•RH•RWGnum˜er•tstd•RH•RWABIHH
   5™re—ting — ve™tor ™ont—ining the —ge spe™i(™ rs† prev—len™e r—tes for PHHR w—l—wi
   hrƒ with —ge groups RHERR —nd RSERW ™om˜ined
   rs†•PHHRwhrƒ•™om˜`E™@rs†•PHHRwhrƒ‘E™@@length@rs†•PHHRwhrƒAEIAD
   length@rs†•PHHRwhrƒAA“Dper™nt•postv•RH•RWA
   5—ssigning n—mes to the elemnts in rs†•PHHRwhrƒ•™om˜
   n—mes@rs†•PHHRwhrƒ•™om˜A`Eege



                                        IHW
   5™onverting rs†•PHHRwhrƒ•™om˜ into — t—˜leF
   5„his m—kes it possi˜le to plot the rs† prev—len™e r—tes with the —ge groups on the
   horizont—l —xis
   rs†•PHHRwhrƒ•™om˜Ft—˜le`E—sFt—˜le@rs†•PHHRwhrƒ•™om˜A

code for estimating nal national HIV prevalence
   code le name: exg•†ƒ•hrƒFtxt
   5„he ™ode in this (le w—s used to produ™e the gr—ph in pigure RFS on p—ge TU —nd
   to ™ompute estim—tes of r—tios of exg rs† prev—len™e to gener—l popul—tion rs†
   prev—len™e presented in „—˜le RFIH on p—ge TV —nd (n—l n—tion—l estim—tes of rs†
   prev—len™e —djusted using PHHR w—l—wi hrƒ results presented in „—˜le RFII on p—ge
   TV
   5sour™ing the ™ode in rs†•xe„syxFtxt —nd rs†•†ƒ•hrƒFtxt
   sour™e@4rs†•xe„syxFtxt4A
   sour™e@4rs†•†ƒ•hrƒFtxt4A
   5™ompute se™ond rs† prev—len™e estim—tes for exg for PHHR
   rs†•PHHR•exgP`EIGPB@rs†•xe„syx•evv‘DT“Crs†•xe„syx•evv‘DU“A
   5™omputing the r—tios of exg to hrƒ estim—tes
   exg•P•hrƒ `E™˜ind@rs†•PHHR•exgPGrs†•PHHRwhrƒ•™om˜Ft—˜leA
   5™ompute (n—l estim—tes of n—tion—l rs† prev—len™e ˜y —ge for the ye—rs IWWW to PHHS
   rs†•xe„syx•psxev`Ers†•xe„syx•evvGexg•P•hrƒ



code for computing survival ratios corresponding to the projected life ex-
   pectancy at birth values presented in Table 3.7 on page 41
   code le name: surviv—l•™—lFtxt
   5re—ding the (le ™ont—ining surviv—l r—tios for — p—rti™ul—r model life t—˜le
   surviv—l•t—˜le`Ere—dFt—˜le@4™dnorthFtxt4D he—dera„‚…iA
   5™dnorthFtxt ™ont—ins fem—le surviv—l r—tios from the north f—mily of go—leEhemeny
   model life t—˜le system
   5™onverting surviv—l•t—˜le into — d—t—fr—me


                                          IIH
surviv—l•t—˜leFfr—me`Ed—t—Ffr—me@surviv—l•t—˜leA
5™—l™ul—ting surviv—l r—tios ™orresponding to life expe™t—n™y —t ˜irth v—lues of RQFPQD
RQFUUD RRFQQD RRFWH
5st—rt ˜y (nding the di'eren™e ˜etween surviv—l r—tios for life expe™t—n™y v—lues of RH
—nd RS
life•expRH•RS`Esurviv—l•t—˜leFfr—me6ˆRSEsurviv—l•t—˜leFfr—me6ˆRH
5divide the di'eren™e into SHH equ—l units
unit•life•expRH•RS`Elife•expRH•RSGSHH
5xow ™—l™ul—te surviv—l r—tios for life expe™t—n™y —t ˜irth v—lue of RQFPQD RQFUUD RRFQQ
—nd RRFWH
life•expRQFPQ`Esurviv—l•t—˜leFfr—me6ˆRHC@QPQBunit•life•expRH•RSA
life•expRQFUU`Esurviv—l•t—˜leFfr—me6ˆRHC@QUUBunit•life•expRH•RSA
life•expRRFQQ`Esurviv—l•t—˜leFfr—me6ˆRHC@RQQBunit•life•expRH•RSA
life•expRRFWH`Esurviv—l•t—˜leFfr—me6ˆRHC@RWHBunit•life•expRH•RSA
5™—l™ul—ting surviv—l r—tios ™orresponding to life expe™t—n™y —t ˜irth v—lues of RSFRUD
RTFHQD RTFTH in — simil—r m—nner —s —˜ove
life•expRS•SH`Esurviv—l•t—˜leFfr—me6ˆSHEsurviv—l•t—˜leFfr—me6ˆRS
unit•life•expRS•SH`Elife•expRS•SHGSHH
life•expRSFRU`Esurviv—l•t—˜leFfr—me6ˆRSC@RUBunit•life•expRS•SHA
life•expRTFHQ`Esurviv—l•t—˜leFfr—me6ˆRSC@IHQBunit•life•expRS•SHA
life•expRTFTH`Esurviv—l•t—˜leFfr—me6ˆRSC@ITHBunit•life•expRS•SHA
5putting the U sets of surviv—l r—tios together
surviv—l•r—tios`E™˜ind@life•expRQFPQDlife•expRQFUUDlife•expRRFQQDlife•expRRFWHD
life•expRSFRUDlife•expRTFHQDlife•expRTFTHA
5extr—™ting from surviv—l•r—tios the surviv—l r—tios of the —ge group@SEWD IHEIRDFFFDRSE
RWA
surviv—l•r—tiosS•RW`Esurviv—l•r—tios‘VXITD“
5m—king surviv—l•r—tiosS•RW — d—t—fr—me
surviv—l•r—tiosS•RW`Ed—t—Ffr—me@surviv—l•r—tiosS•RWA
ye—r`EIWWVXPHHR


                                       III
   n—mes@surviv—l•r—tiosS•RWA`Eye—r



code for estimating the number of women of reproductive age in Malawi from
   1999 to 2004
   code le name: proj•popFtxt
   5„he ™ode in this (le w—s used to o˜t—in estim—tes of women of reprodu™tive —ge in
   w—l—wi from IWWW to PHHR presented in „—˜les RFIS to RFPH on p—ges UI to UQF „he ™ode
   presented here is for estim—tes ˜—sed on go—leEhemeny xorth model life t—˜le system



   ye—rs`EIWWVXPHHR
   dur—tion`Elength@ye—rsA
   —ges`ESXRW
   —ge•width`Elength@—gesA
   5™re—ting — m—trix with zeros
   proj•m—trix`Erep@HDdur—tionB—ge•widthA
   dim@proj•m—trixA`E™@—ge•widthDdur—tionA
   5™onverting the zero m—trix into — d—t— fr—me
   proj•m—trixFfr—me`Ed—t—Ffr—me@proj•m—trixA
   n—mes@proj•m—trixFfr—meA`Eye—rs
   rowFn—mes@proj•m—trixFfr—meA`E—ges
   5re—ding ˜—se line popul—tion d—t— from ˜—se•popF™sv
   ˜—se•pop`Ere—dF™sv@4˜—se•popPF™sv4D he—dera„‚…iA
   5˜—se•popF™sv ™ont—ins the —geEsmoothed fem—le popul—tion —ged S to RW in w—l—wi
   in IWWV whi™h h—s ˜een sep—r—ted into single ye—rs using the feers formul—eF
   5™re—ting — fun™tion for (nding (ve ye—r pop groups from —ge S to —ge RW
   popS—ges`Efun™tion@xA
   {(ve`E™@A
   —aI
   ˜aS


                                        IIP
for @i in IXWA
{tot—lS`Esum@x‘—X˜“A
(ve`E™@(veDtot—lSA
—a—CS
˜a˜CS
}
(ve}
5™re—ting — d—t— fr—me ™ont—ining popul—tion in (ve ye—r —ge groups
popSy•grps`Efun™tion@xA
{(ve•y•grps`E™˜ind@A
for @i in xA
{sumS—ges`EpopS—ges@iA
(ve•y•grps`E™˜ind@(ve•y•grpsDsumS—gesA
} (ve•y•grps
}
5sour™ing (le ™ont—ining surviv—l r—tios
sour™e@4surviv—l•™—lFtxt4A
5™onverting (ve ye—r surviv—l r—tios into single ye—r surviv—l r—tios
5™re—ting — fun™tion th—t ™onverts (ve ye—r surviv—l r—tios into single ye—r surviv—l
r—tios ˜y t—king the (fth root
(ve•ye—rPsingle•ye—r•surviv—l`Efun™tion@xA
{x ¢ @IGSA}
5™onverting the surviv—l r—tios in surviv—l•™—lFtxt into single ye—rs
surviv—l•r—tiosS•RW•singlesI`E(ve•ye—rPsingle•ye—r•surviv—l@surviv—l•r—tiosS•RWA
5™re—ting — fun™tion th—t repli™—tes S times e—™h S ye—r surviv—l r—tio th—t h—s ˜een
™onverted to single ye—r surviv—l r—tio
(ve•ye—rPsingle•ye—r•surviv—l•ˆS`Efun™tion@xA
{—alength@xA
˜arep@SD—A
garep@xD˜A


                                          IIQ
g}
surviv—l•r—tiosS•RW•singlesP
`E(ve•ye—rPsingle•ye—r•surviv—l•ˆS@surviv—l•r—tiosS•RW•singlesIA
dim@surviv—l•r—tiosS•RW•singlesPA`Edim@proj•m—trixA
surviv—l•r—tiosS•RW•singlesPFfr—me`Ed—t—Ffr—me@surviv—l•r—tiosS•RW•singlesPA
n—mes@surviv—l•r—tiosS•RW•singlesPFfr—meA`Eye—rs
rowFn—mes@surviv—l•r—tiosS•RW•singlesPFfr—meA`E—ges
for @i in IXUA
{if @iaaIA
{proj•m—trixFfr—me‘iXRSDi“a˜—se•pop‘DP“ }
else {proj•m—trixFfr—me‘iXRSDi“aproj•m—trixFfr—me‘@iEIAX@RSEIAD
@iEIA“Bsurviv—l•r—tiosS•RW•singlesP‘@iEIAX@RSEIAD@iEIA“}
}
5o˜t—ining popul—tion —ge ISERW in S ye—r groups
popIS•RW`EpopSy•grps@proj•m—trixFfr—meA‘QXWD“
popIS•RWFfr—me`Ed—t—Ffr—me@popIS•RWA
n—mes@popIS•RWFfr—meA`Eye—rs
5rounding to whole num˜ers
popIS•RW•(n—l`Eround@popIS•RWFfr—meA
5 ™re—ting — fun™tion for ™—l™ul—ting tot—l popul—tion —ged ISERW
tot—l•pop`Efun™tion@xA
{tot—l`E™@A
for @i in xA
{—asum@iA
tot—l`E™@tot—lD—A}
n—mes@tot—lA`En—mes@xA
tot—l
}
popIS•RW•(n—l•tot—l`Etot—l•pop@popIS•RW•(n—lA




                                       IIR
code for estimating the number of HIV-infected women
   code le name: €y€•sxpFtxt
   5 „he ™ode in this (le w—s used to o˜t—in estim—tes of the num˜er of rs†Einfe™ted
   women in w—l—wi from IWWW to PHHR presented in „—˜les RFPP to RFPS on p—ges UW to
   VIF „he ™ode presented here is for estim—tes ˜—sed on go—leEhemeny xorth model life
   t—˜le system
   5sour™ing ™ode in proj•popFtxt
   sour™e@4proj•popFtxt4A
   5extr—™ting popul—tion in IWWW to PHHR
   popWW•HR`EpopIS•RW•(n—l‘DEI“
   5 sour™ing ™ode for (n—l estim—tes of n—tion—l rs† prev—len™e r—tes
   sour™e@4exg•†ƒ•hrƒFtxt4A
   rs†•r—tes`Ers†•xe„syx•psxev‘IXT“
   5this ™ode gives rs† r—tes with the r—te for RHERW —pplied to —ge groups RHERR —nd
   RSERW
   rs†•r—tesP`E™@A
   for @i in rs†•r—tesA
   {
   r—te`Ei
   r—teP`E™@iDi‘length@iA“A
   rs†•r—tesP`E™˜ind@rs†•r—tesPDr—tePA
   } rs†•r—tesPFfr—me`Ed—t—Ffr—me@rs†•r—tesPA
   n—mes@rs†•r—tesPFfr—meA`E™@4IWWW4D4PHHH4D4PHHI4D4PHHP4D4PHHQ4D4PHHR4A
   rowFn—mes@rs†•r—tesPFfr—meA`E™@4ISEIW4D4PHEPR4D4PSEPW4D4QHEQR4D4QSEQW4D4RHERR4D4RSE
   RW4A
   5™omputing num˜er of women —ged ISERW infe™ted with rs†
   inf•w`Eround@@popWW•HRBrs†•r—tesPFfr—meAGIHHA
   n—mes@inf•wA`E™@4infw•IWWW4D4infw•PHHH4D4infw•PHHI4D
   4infw•PHHP4D4infw•PHHQ4D4infw•PHHR4A
   rowFn—mes@inf•wA`ErowFn—mes@rs†•r—tesPFfr—meA


                                        IIS
   5™omputing tot—l num˜er of rs† infe™ted women
   inf•w•tot—l`Etot—l•pop@inf•wA
   5™omputing rs† prev—len™e in ISERW —ge group in IWWW to PHHR
   rs†•prev•—dult`Einf•w•tot—lGpopIS•RW•(n—l•tot—l‘EI“BIHH
   5™omp—ring estim—tes of rs† prev for fem—le —dults with spe™trum estim—tes ˜y xeg
   xeg•fem—le•rs†•IS•RW`E™@RHWIUIDRIRIWHDRIWPHRDRPRHURDRQIUPQDRRQRTQA
   xeg•fem—le•IS•RW•tot—l`E™@PSRPPHPDPTHTTRVDPTUIQUIDPUQUQISDPVHUIRRDPVVIIHIA
   xeg•prev•—dult`E@xeg•fem—le•rs†•IS•RWGxeg•fem—le•IS•RW•tot—lABIHH
   n—mes@xeg•prev•—dultA`En—mes@rs†•xe„syx•psxevA‘EI“
   xeg•†ƒ•gry`Exeg•prev•—dult‘ET“Ers†•prev•—dult‘EI“




code for estimating age-specic fertility rates in Malawi from 1999 to 2004
   code le name: eƒp‚Ftxt
   5 „he ™ode in this (le w—s used to produ™e the gr—ph in pigure RFW on p—ge VR —nd to
   o˜t—in estim—tes of —geEspe™i(™ fertility r—tes in w—l—wi from IWWW to PHHR presented
   in „—˜le RFPU on p—ge VSF
   eges`E™@4ISEIW4D4PHEPR4D4PSEPW4D4QHEQR4D4QSEQW4D4RHERR4D4RSERW4A
   eƒp‚•PHHH`E™@IUPDQHSDPUPDPIWDITUDWRDRIA
   n—mes@eƒp‚•PHHHA`Eeges
   eƒp‚•PHHR`E™@ITPDPWQDPSRDPPPDITQDVHDQSA
   n—mes@eƒp‚•PHHRA`Eeges
   eƒp‚•PHHHFt—˜le`E—sFt—˜le@eƒp‚•PHHHA
   eƒp‚•PHHRFt—˜le`E—sFt—˜le@eƒp‚•PHHRA
   5™—l™ul—ting estim—tes of eƒp‚ —nnu—l r—te of ™h—nge ˜etween PHHH —nd PHHRF
   eƒp‚•PHHH•PHHR`Eeƒp‚•PHHHFt—˜leEeƒp‚•PHHRFt—˜le
   eƒp‚•PHHH•PHHR•ye—r•™h—nge`Eeƒp‚•PHHH•PHHRGR
   eƒp‚HH•HR`E™@eƒp‚•PHHHFt—˜leA
   for @i in IXRA



                                        IIT
   {
   eƒp‚`Eround@eƒp‚•PHHHFt—˜leE@iBeƒp‚•PHHH•PHHR•ye—r•™h—ngeAA
   eƒp‚HH•HR`E™˜ind@eƒp‚HH•HRDeƒp‚A
   }
   5estim—ting eƒp‚ for IWWW
   eƒp‚WW`Eround@eƒp‚•PHHHFt—˜leCeƒp‚•PHHH•PHHR•ye—r•™h—ngeA
   5 produ™ing estim—tes of eƒp‚ from IWWW to PHHR
   eƒp‚WW•HR`E™˜ind@eƒp‚WWDeƒp‚HH•HRA
   eƒp‚WW•HRFfr—me`Ed—t—Ffr—me@eƒp‚WW•HRA
   n—mes@eƒp‚WW•HRFfr—meA`E™@4eƒp‚IWWW4D 4eƒp‚•PHHH4D4eƒp‚•PHHI4D
   4eƒp‚•PHHP4D4eƒp‚•PHHQ4D4eƒp‚•PHHR4A
   5rowFn—mes@rs†•r—tesPFfr—meA`E™@4ISEIW4D4PHEPR4D4PSEPW4D4QHEQR4D4QSEQW4D4RHERR4D4RSE
   RW4A




code for estimating number of births
   code le name: ˜irthsFtxt
   5 „he ™ode in this (le w—s used to o˜t—in estim—tes of the num˜er of ˜irths to ˜oth
   —ll women of reprodu™tive —ge —nd rs†Einfe™ted womenF
   5sour™ing ™ode in €y€•sxpFtxt
   sour™e@4€y€•sxpFtxt4A
   5sour™ing ™ode in eƒp‚Ftxt
   sour™e@4eƒp‚Ftxt4A
   5™omputing num˜er of ˜irths to —ll women of reprodu™tive —ge from IWWW to PHHR
   ˜irthsWW•HR`Eround@@popWW•HRGIHHHABeƒp‚WW•HRFfr—meA
   5™omputing tot—l num˜er of ˜irths to —ll women
   ˜irthsWW•HR•tot—l`Etot—l•pop@˜irthsWW•HRA
   5™omputing ˜irths to rs† positive women
   ˜irth•infIS•IW`E@IFSB@popWW•HR‘ID“ABeƒp‚WW•HRFfr—me‘ID“B@inf•w‘ID“AAG@IFSBinf•w‘ID“



                                        IIU
C @popWW•HR‘ID“Einf•w‘ID“AA
˜irth•infPH•RW`E@@popWW•HR‘EID“ABeƒp‚WW•HRFfr—me‘EID“B@inf•w‘EID“AAG@inf•w‘EID“ C
IFPSB@popWW•HR‘EID“Einf•w‘EID“AA
inf•w•˜irths`Eround@r˜ind@˜irth•infIS•IWD˜irth•infPH•RWAGIHHHA
inf•w•˜irths•tot—l`Etot—l•pop@inf•w•˜irthsA
5 istim—ting the num˜er of ™hildren infe™ted with rs† through w„g„
5 ™re—ting — ve™tor for num˜er of women re™eiving x†€ from IWWW to PHHR
nvp`E™@HDHDHDVRHDPIWVDPUIWA
inf•w•˜irths•nvp`Envp
n—mes@inf•w•˜irths•nvpA`En—mes@inf•w•˜irthsA
5™omputing num˜er of ™hildren whose mothers did not re™eive x†€
inf•w•˜irths•nvp•free`Einf•w•˜irths•tot—lEinf•w•˜irths•nvp
5™omputing num˜er of ™hildren infe™ted vi— w„g„ —ssuming €„‚ of QH7 for no x†€
—nd IT7 for x†€
inf•˜irthsQH`Eround@@inf•w•˜irths•nvp•freeBHFQAC@inf•w•˜irths•nvpBHFITAA
5™omputing num˜er of ™hildren infe™ted vi— w„g„ —ssuming €„‚ of QP7 for no x†€
—nd PR7 for x†€
inf•˜irthsQP`Eround@@inf•w•˜irths•nvp•freeBHFQPAC@inf•w•˜irths•nvpBHFPRAA
5™omputing num˜er of ™hildren infe™ted vi— w„g„ —ssuming €„‚ of QS7 for no x†€
—nd PV7 for x†€
inf•˜irthsQS`Eround@@inf•w•˜irths•nvp•freeBHFQSAC@inf•w•˜irths•nvpBHFPVAA
5istim—tes with IHH7 nvp ™over—ge
inf•˜irthsQH•nvp`Eround@inf•w•˜irths•tot—lBHFITA
inf•˜irthsQP•nvp`Eround@inf•w•˜irths•tot—lBHFPRA
inf•˜irthsQS•nvp`Eround@inf•w•˜irths•tot—lBHFPVA




                                    IIV
code for survival curve of HIV-infected children
   code le name: ™hild•rs†•surviv—lFtxt
   5 „he ™ode in this (le w—s used to gener—te the surviv—l ™urve of rs†Einfe™ted ™hildren
   —nd produ™e the gr—ph in pigure QFI on p—ge RVF
   5™re—ting — fun™tion for ™—l™ul—ting the proportion of ™hildren dying due to rs† —fter
   t ye—rs sin™e ˜irth
   ™hildren•dying`Efun™tion@tA
   {paHFT
   —•IaHFW
   —•PaIH
   ˜•IaHFW
   ˜•PaHFI
   d•t`E@pB@IEexp@E@˜•IBtA ¢ —•IAAC@IEpAB@IEexp@E@˜•PBtA ¢ —•PAAA
   d•t}
   5™re—ting — fun™tion for ™—l™ul—ting the proportion of ™hildren not dying due to rs†
   —fter t ye—rs sin™e ˜irth
   ™hildren•surving`Efun™tion@tA
   {IE™hildren•dying@tA
   }
   taseq@HDIQDHFIA
   survivors`E™hildren•surving@tA
   surviv—l•t—˜lea™˜ind@tDsurvivorsA
   5plotting the surviv—l distri˜ution
   plot@tDsurvivorsBIHHD typea4l4D ™olaRD xl—˜a4ege of ™hild4D lwdaID yl—˜a4€er™ent—ge
   ƒurviving4A
   —xis@ID x—xpa™@HD IQD IQAA 5spe™i(es the interv—l in the x —xis
   —xis@PD y—xpa™@HDIHHDIHAA 5spe™i(es the interv—l in the y —xis
   grid@lwdaIA




                                         IIW
code for estimating the number of HIV-infected children dying before their fth
   birthday during the period 2000 to 2004
   code le name: ™ummul—tive•dyingFtxt
    5 „he ™ode in this (le w—s used to ™ompute estim—tes of the num˜er of rs†Einfe™ted
    ™hildren dying ˜efore their (fth ˜irthd—y during the period PHHH to PHHR presented in
    „—˜le RFRH on p—ge WRF „he ™ode presented here is for €„‚aQH7
    5™—l™ul—ting tot—l num˜er of ™hildren under (ve ye—rs dying due to rs†
    5sour™ing ™ode for ˜irths
    sour™e@4˜irthsFtxt4A
    5sour™ing ™ode for surviv—l ™urve of rs†Einfe™ted ™hildren
    sour™e@4™hild•rs†•surviv—lFtxt4A
    5™omputing —ver—ge num˜er of ™hildren infe™ted with rs† thru w„g„ per month in
    e—™h ye—r from PHHH to PHHR
    —ver—ge•inf•˜irths`Eround@inf•˜irthsQH‘EI“GIPA
    5for e—™h ye—rD repli™—ting the —ver—ge num˜er of infe™ted ™hildren IP times so th—t
    the tot—l should ˜e equ—l to the tot—l for th—t ye—r
    inf•˜irths•monthly`Erep@—ver—ge•inf•˜irthsD rep@IPDlength@—ver—ge•inf•˜irthsAAA
    time`ESWXH
    prop•dying`E™hildren•dying@timeGIPA
    rs†•de—ths`Eround@—ver—ge•inf•˜irthsBprop•dyingA
    tot—l•rs†•de—ths•HH•HR`Esum@rs†•de—thsA




code for estimating the proportion of under-ve mortality directly attributable
   to HIV/AIDS in Malawi during the period 2000 to 2004
   code le name: ™ummul—tive•dyingFtxt
    5 „his ™ode is the se™ond p—rt of the ™ode in ™ummul—tive•dyingFtxt —nd w—s used to
    produ™e estim—tes of …Sw‚ due to rs†Geshƒ —nd proportion of underE(ve mort—lity
    —ttri˜ut—˜le to rs†Geshƒ in w—l—wi during the period PHHH to PHHR presented in „—˜le



                                           IPH
RFRI on p—ge WTF „he ™ode presented here is for €„‚aQH7
tot—l•rs†•inf•˜irths`Esum@inf•˜irthsQH‘EI“A
tot—l•˜irths•HH•HR`Esum@˜irthsWW•HR•tot—l‘EI“A
tot—l•uSm•—ll•™—use•HH•HR`Eround@tot—l•˜irths•HH•HRBIQQGIHHHA
tot—l•uSm•rs†•neg•HH•HR`Etot—l•uSm•—ll•™—use•HH•HREtot—l•rs†•de—ths•HH•HR
uSmr•non•rs†`Etot—l•uSm•rs†•neg•HH•HRGtot—l•˜irths•HH•HRBIHHH
5estim—te of mort—lity in rs†Einfe™ted ™hildren —djusted for ™ompeting ™—uses
tot—l•rs†•de—ths•—dj`E
tot—l•rs†•de—ths•HH•HREround@uSmr•non•rs†Btot—l•rs†•de—ths•HH•HRGIHHHA
rs†•uSmr•—dj`E@tot—l•rs†•de—ths•—djGtot—l•˜irths•HH•HRABIHHH
5estim—te of rs† —ttri˜ut—˜le underE(ve mort—lity
rs†•mort•€ep`E@rs†•uSmr•—djGIQQABIHH
5estim—tes wth IHH7 nvp ™ove—r—ge
—ver—ge•inf•˜irths•nvp`Eround@inf•˜irthsQH•nvp‘EI“GIPA
5for e—™h ye—rD repli™—ting the —ver—ge num˜er of infe™ted ™hildren IP times so th—t
the tot—l should ˜e equ—l to the tot—l for th—t ye—r
inf•˜irths•nvp•monthly`Erep@—ver—ge•inf•˜irths•nvpD
rep@IPDlength@—ver—ge•inf•˜irths•nvpAAA
rs†•de—ths•nvp`Eround@—ver—ge•inf•˜irths•nvpBprop•dyingA
tot—l•rs†•de—ths•nvp•HH•HR`Esum@rs†•de—ths•nvpA
nonvp•nvp`Etot—l•rs†•de—ths•HH•HREtot—l•rs†•de—ths•nvp•HH•HR
tot—l•uSm•—ll•™—use•nvp•HH•HR`Etot—l•uSm•—ll•™—use•HH•HREnonvp•nvp
uSmr•—ll•™—use•nvp•HH•HR`E
@tot—l•uSm•—ll•™—use•nvp•HH•HRGtot—l•˜irths•HH•HRABIHHH




                                       IPI
Bibliography

 edetunjiD tF @PHHHAF „rends in underES mort—lity r—tes —nd the rs†Geshƒ epidemi™F
 Bulletin of the World Health OrganisationD UV@IHAXIPHHEIPHTF

 egrestiD eF —nd goullD fF @IWWVAF epproxim—te is ˜etter th—n 4ex—™t4 for interv—l
 estim—tion of ˜inomi—l proportionsF The American StatisticianD SPXIIWEIPTF

 ehm—dD yFfF et —lF @PHHHAF „he de™line in ™hild mort—lityX — re—ppr—is—lF Bulletin of
 the World Health OrganisationD UV@IHAX IIUSEIIWIF

 evert @PHHUAF rs† 8 eshƒ in w—l—wiF …‚vX http://www.avert.org/aids-malawi.htm.
 e™™essed on IS eugustD PHHUF

 f—nnisterD fFeF et —lF @PHHHAF Infectious Disease (2nd ed.)F yxfordX fl—™kwell ƒ™ien™eF

 fl—™kD ‚FiF et —lF @PHHQAF ‡here —nd why —re IH million ™hildren dying every ye—rc
 The LacentD QTIXPPPTEPPQRF

 f—rnettD eF —nd fl—ikieD €F @IWWPAF AIDS in Africa: Its present and future impactF xew
 ‰orkX tohn ‡iley 8 ƒonsD sn™F

 fo˜—tD ‚FD goov—di—D rFD woodleyD hFD et —lF @IWWWAF wort—lity in — ™ohort of ™hildren
 ˜orn to rs†EI infe™ted women from hur˜—nD ƒouth efri™—F SAMJD VWXTRTETRVF

 fr—hm˜h—ttD rF et —lF @PHHTAF wort—lity in rs†Einfe™ted —nd uninfe™ted ™hildren of
 rs†Einfe™ted —nd uninfe™ted mothers in rur—l …g—nd—F Journal of Acquired Immune


                                        IPP
Deciency SyndromesD RI@RAXSHRESHVF

frookmeyerD ‚F —nd q—ilD wFrF @IWWRAF AIDS epidemiology: A quantitative approachF
yxfordX yxford …niversity €ressF

fry™eD tF et —lF @PHHSAF ‡ry estim—tes of the ™—use of de—th in ™hildrenF The LancetD
QTSXIIRUEIISPF

fuvéD eF et —lFD @PHHPAF „he spre—d —nd e'e™t of rs†EI infe™tion in su˜Eƒ—h—r—n efri™—F
The LancetD QSWXPHIIEPHIUF

g—ll—w—yD hFƒF et —lF @IWWWAF †irus phenotype swit™hing —nd dise—se progression in
rs†EI infe™tionF Proceedings: Biological SciencesD PTT@IRQUAXPSPQEPSQHF

g—rpenterD vFwFD x—kiyingiD tFƒFD ‚u˜er—ntw—riD eFD w—l—m˜—D ƒFƒFD u—m—liD eF —nd
‡hitworthD tFeFqF @IWWUAF istim—tes of the imp—™t of rs† infe™tion on fertility in —
rur—l …g—nd—n popul—tion ™ohortF Health Transition ReviewD U @ƒupplement PAXIIQEIPTF

genters     for   hise—se     gontrol     @IWVI—AF     €neumo™y™stis     €neumoni—      E    vos
engelesF     wor˜idity       —nd      wort—lity    ‡eekly      ‚eportD    QHXPSHEPSPF       …‚vX
http://www.cdc.gov/hiv/resources/reports/mmwr/pdf/mmwr05jun81.pdfF                   e™™essed
on IS eugustD PHHUF

          @IWVI˜AF u—posi9s s—r™om— —nd €neumo™y™stis €neumoni— —mong homosexu—l
men E xew ‰ork gity —nd g—liforni—F wor˜idity —nd wort—lity ‡eekly ‚eportD QHXQHSE
QHVF …‚vX http://www.cdc.gov/hiv/resources/reports/mmwr/pdf/mmwr04jul81.pdfF IS
eugustD PHHUF

          @IWVPAF gurrent trends upd—te on e™quired smmune he(™ien™y ƒyndrome
@eshƒA !…nited ƒt—tes F wor˜idity —nd wort—lity ‡eekly ‚eportD QIXSHUESHVF
…‚vX http://www.cdc.gov/mmwr/preview/mmwrhtml/00001163.htm. e™™essed on T
xovem˜erD PHHTF

gi™ho™kiD         wF        @PHHTAF      „he         history    of       rs†GeshƒF          …‚vX



                                             IPQ
http://aids.about.com/od/newlydiagnosed/a/hivtimeline.htm. e™™essed on IS eugustD
PHHUF

h—˜isD pFD ileng—D xFD wed—D xFD et —lF @PHHIAF IVEmonth mort—lity —nd perin—t—l
exposure to zidovudine in ‡est efri™—F AIDSD ISXUUIEUUWF

h—˜isD pFD xewellD wFvFD pr—nsenD vFD et —lF@PHHHAF €revention of motherEtoE™hild
tr—nsmission of rs† in developing ™ountriesX re™ommend—tions for pr—™ti™eF Health
Policy PlanningD ISXQRERPF

he go™kD uFD powlerD wFqFD wer™ierD iFD et —lF @PHHHAF €revention of motherEtoE™hild
rs† tr—nsmission in resour™eE poor ™ountriesEtr—nsl—ting rese—r™h into poli™y —nd
pr—™ti™eF JAMAD PVQXIIUSEIIVPF

ho™torD rF†F @PHHRAF xew mort—lity p—tterns for w—l—wi during the l—te twentieth
™enturyF …npu˜lished m—nus™riptF

powlerD wFqFD xewellD wFvF @PHHPAF fre—stEfeeding —nd rs†EI tr—nsmission in resour™eE
limited settingsF Journal of Acquired Immune Deciency SyndromesD QHXPQHEPQWF

pylkesnesD uF et —lF @IWWVAF ƒtudying dyn—mi™s of the rs† epidemi™X popul—tionE˜—sed
d—t— ™omp—red with sentinel surveill—n™e in —m˜i—F AIDSD IPXIPPUEIPQRF

q—renneD wF —nd q—kusiDiF @PHHTAF re—lth tr—nsitions in su˜Eƒ—h—r—n efri™—X overview
of mort—lity trends in ™hildren under S ye—rs old @IWSHEPHHHAF Bulletin of the World
Health OrganisationD VR@TAXRPSESHRF

qhysD €FhF et —lF @PHHRAF „he …xeshƒ estim—tion —nd proje™tion p—™k—geX — softw—re
p—™k—ge to estim—te —nd proje™t n—tion—l rs† epidemi™sF Sexually Transmitted
InfectionsD VHXSEWF

qlo˜—l rs† €revention ‡orking qroup @PHHUAF fringing rs† prevention to s™—leX
en      urgent   glo˜—l   priorityF   …‚vXhttp://www.globalhivprevention.org/pdfs/PWG-
HIV_prevention_report_FINAL.pdf. e™™essed on IW eugustD PHHUF


                                          IPR
qr—yD ‚FD ‡—werD wFD ƒerw—dd—D hF et —lF@IWWVAF €opul—tionE˜—sed study of fertility in
women with rs†EI infe™tion in …g—nd—FThe LancetD QSI@WHWTAXWVEIHQF

qu—yD vFeFD wusokeD €FD plemingD „F et —lF @IWWWAF sntr—p—rtum —nd neon—t—l singleEdose
nevir—pine ™omp—red with zidovudine for prevention of motherEtoE™hild tr—nsmission of
rs†EI in u—mp—l—D …g—nd—X rs†xi„ HIP r—ndomised tri—lF The LancetD QSRXUWSEVHPF

re™htD ‚F et —lF @PHHTAF €utting st „ogetherX eshƒ —nd the willennium hevelopment
qo—lsF PLoS MedicineD Q@IIAXIWWPEIWWVF

rillD uF —nd emouzouD eF @PHHTAF „rends in ghild wort—lityD IWTH to PHHHF t—misonD
hF„F et —lF @edsFAD Disease and Mortality in Sub-Saharan Africa (2nd edition)F ‡—shington
hFgFD ‡orld f—nkXISEQHF

rillD uF et —lF @IWWVAF Trends in child mortality in the developing world: 1960 to 1996F
…‚vX http://www.childinfo.org/cmr/kh98meth.html. e™™essed on IS eugust PHHUF

sxhi€„r xetwork @PHHPAF Population and health in developing countries. Volume
1: Population, health, and survival at INDEPTH SitesF ytt—w—X              sntern—tion—l
hevelopment ‚ese—r™h genterF

sntern—tion—l   eshƒ     †—™™ine   sniti—tive   @PHHSAF   Putting it together:     AIDS
and   the   Millennium     Development     GoalsF   €oli™y   his™ussion   €—perF   …‚vX
http://www.iavi.org/viewle.cfm?d=33078F e™™essed on IS eugustD PHHUF

te—nD ƒFƒFD €—peD tF‡FD †erdierD ‚FsFD et —lF @IWWWAF „he n—tur—l history of hum—n
immunode(™ien™y virus I infe™tion in r—iti—n inf—ntsF Pediatr Infect Dis JD IVXSVETQF

tohnsonD vFpF —nd horringtonD ‚FiF @PHHTAF wodelling the demogr—phi™ imp—™t of
rs†Geshƒ in ƒouth efri™— —nd the likely imp—™t of interventionsF Demographic
ResearchD IRXSRIESURF

uellyD tFeF —nd ƒtF v—wren™eD tFƒF @IWVVAF The AIDS health crisis: Psychological and
social interventionsF xew ‰orkX €lenum €ressF


                                         IPS
vep—geD €FD ƒpir—D ‚FD u—li˜—l—D ƒFD et —l @IWWVAF ‚eport of — workshop for ™lini™—l
rese—r™hX     g—re of hum—n immunode(™ien™y virusEinfe™ted ™hildren in developing
™ountriesF Pediatr Infect Dis JD IUXSVIESVTF

vevyDFtFeF @IWWQAF €—thogenesis of rum—n smmunode(™ien™y †irus snfe™tionF
Microbiological ReviewsD SU@IAXIVQEPVWF

vewthw—iteD €F —nd ‡ilkinsD iF @PHHSAF x—tur—l history of rs†GeshƒF MedicineD
QQ@TAXIHEIQ

w—l—wi      rs†   —nd   eshƒ     wonitoring     —nd     iv—lu—tion    ‚eport   PHHSF   …‚vX
http://data.unaids.org/pub/Report/2006/2006_country_progress_report_malawi_en.pdf.
e™™essed on IS eugustD PHHUF

w—l—wi willennium hevelopment qo—ls ‚eportD PHHQF
…‚vX http://www.sarpn.org.za/documents/d0001131/Malawi_MDG_report_2003.pdf.
e™™essed on IS eugustD PHHUF

wieƒ…‚i iv—lu—tion @PHHRAF AIDS in Africa during the Nineties: Malawi. A review
and analysis of survey and research resultsF g—rolin— €opul—tion genterD …niversity of
xorth g—rolin— —t gh—pel rillF

windelD eF —nd „en—ntEplowersD wF @PHHIAF efg of eshƒX x—tur—l history —nd
m—n—gement of e—rly rs† infe™tionF BMJD QPPXIPWHEIPWQF

winistry of re—lthD vighthouse „rust @vilongweA —nd ghg @w—l—wiA @PHHTAF ‚eport of
— ™ountryEwide survey of rs†Geshƒ servi™es in w—l—wi for the ye—r PHHSF …npu˜lished
m—nus™riptF

wosleyD ‡FrF —nd ghenD vFgF @IWVRAF en —n—lyti™ fr—mework for the study of ™hild
surviv—l in developing ™ountriesF Population and Development ReviewD IHXPSERSF

x—tion—l       eshƒ     gommission     ‘w—l—wi“         et    —lF    @PHHRAF    feh—viour—l
surveill—n™e       survey      @fƒƒA          reportD        w—l—wi        PHHRF       …‚vX



                                          IPT
http://www.fhi.org/en/HIVAIDS/pub/survreports/res_MalawiBSS.htm.             e™™essed
on IS eugustD PHHUF

x—tion—l eshƒ gommission ‘w—l—wi“ @PHHQ—AF x—tion—l rs†Geshƒ €oli™yX e ™—ll for
renewed —™tionF qovernment of w—l—wiF …‚vX http://data.unaids.org/Topics/Human-
Rights/malawi_national_policy_en.pdf. e™™essed on V eugustD PHHUF

       @PHHQ˜AF rs†Geshƒ in w—l—wiX istim—tes of the prev—len™e of infe™tion —nd the
impli™—tionsF …‚vX http://www.synergyaids.com/documents/MAL_AIDS.pdf e™™essed
on IS eugustD PHHUF

       @PHHQ™AF rs† ƒentinel ƒurveill—n™e ‚eport PHHQF

       @PHHSAFrs† —nd ƒyphilis ƒeroEƒurvey —nd x—tion—l rs† €rev—len™e istim—tes
‚eport PHHSF

xewellD wFvFD fr—hm˜h—ttD rFD —nd qhysD €FhF @PHHRAF ghild mort—lity —nd rs†
infe™tion in efri™—X — reviewF AIDSD IV@suppl PAXƒPUEƒQRF

xewellD wFvF@IWWVAF we™h—nisms —nd timing of motherEtoE™hild tr—nsmission of rs†EIF
AIDSD IPXVQIEVQUF

xg9weshemiD tF et —lF @PHHQAF rs† smp—™t on wother —nd ghild wort—lity in ‚ur—l
„—nz—ni—F Journal of Acquired Immune Deciency SyndromesD QQ@QAXQWQERHRF

xƒy‘w—l—wi“ @IWWQAF Malawi population and housing census 1987 Volume II: Population
characteristicsF om˜—X x—tion—l ƒt—tisti™—l y0™eF

        @PHHHAF 1998 Malawi population and housing census report of nal census
results: Population densityF om˜—X x—tion—l ƒt—tisti™—l y0™eF

       @PHHPAF 1998 Malawi population and housing census analytical reportF om˜—X
x—tion—l ƒt—tisti™—l y0™eF

      F w—l—wi €opul—tion €roje™tions IWWWEPHHSX         w—l—wi popul—tion proje™tion



                                       IPU
pro™edure …‚vX http://www.nso.malawi.net. e™™essed on IS eugustD PHHUF

xƒy‘w—l—wi“D —nd y‚g w—™roF @IWWRAFMalawi Demographic and Health Survey 1992F
g—lvertonD w—ryl—ndX xƒy —nd y‚g w—™roF

      @PHHIAFMalawi Demographic and Health Survey 2000F g—lvertonD w—ryl—ndX
xƒy —nd y‚g w—™roF

      @PHHSAFMalawi Demographic and Health Survey 2004F g—lvertonD w—ryl—ndX
xƒy —nd y‚g w—™roF

yt—D wFyFgFD y9honov—nD hFD el—˜iD ƒFeF @PHHHAF w—tern—l rs†EI —nd rs†EP infe™tion
—nd ™hild surviv—l in „he q—m˜i—F AIDSD IRXRQSERQWF

€e™kh—mD gF —nd qi˜˜D hF @IWWSAF wotherEtoE™hild tr—nsmission of the hum—n
immunode(™ien™y virusF The New England Journal of MedicineD QQQ@SAXPWVEQHPF

€erelsonD eFƒ —nd xelsonD €F‡F @IWWWAF w—them—ti™—l —n—lysis of rs†EI dyn—mi™s in
vivoF SIAM ReviewD RI@IAXQERRF

€iotD €F —nd gole˜undersD ‚F @IWVUAF glini™—l m—nifest—tions —nd the n—tur—l history
of rs† infe™tion in —dultsD sn eshƒEe glo˜—l perspe™tive ‘ƒpe™i—l sssue“F West J
MedDIRUXUHWEUIPF

‚hu™h—roenpornp—ni™hD yF —nd gh—mr—trithirongD eF @PHHIAF hemogr—phi™ smp—™t of
eshƒ on the „h—i popul—tionFAsia-Pacic Population JournalD ITXUIEVVF

‚utsteinD ƒFyF @PHHHAF p—™tors —sso™i—ted with trends in inf—nt —nd ™hild mort—lity
in developing ™ountries during the IWWHsF Bulletin of the World Health OrganizationD
UVXIPSTEIPUHF

ƒ—lomonD tFeF —nd wurr—yD gFtFvF @PHHIAF wodelling rs†Geshƒ epidemi™s in su˜E
ƒ—h—r—n efri™— using seroprev—len™e d—t— from —nten—t—l ™lini™sF Bulletin of the World
Health OrganisationD UW@UAX SWTETHUF



                                       IPV
ƒ—n pr—n™is™o eshƒ pound—tion @PHHTAF „he ƒt—ges of rs† hise—seF …‚vX
http://www.sfaf.org/aids101/hiv_disease.html. e™™essed on IS eugustD PHHUF

ƒ—uroD tF —nd vewisD ‚FtF @PHHSAF 4istim—ting ™ompletion r—tes from sm—ll s—mples
using ˜inomi—l ™on(den™e interv—lsX ™omp—risons —nd reommend—tions4 in €ro™eedings
of the hum—n f—™tors —nd ergonomi™s so™iety 49th —nnu—l meetingXPIHHEPIHRF …‚vX
http://www.measuringusability.com/sauro-lewisHFES.pdfF e™™essed on IS eugustD
PHHUF

ƒ™hoe˜erleinD hF @PHHIAF Everybody: Preventing HIV and Other Sexually Transmitted
Diseases Among Young Teens (Revised Edition). g—r˜ond—leD gyX ‚eh idu™—tion—l
€rogr—msF

ƒ™hw—rtl—nderD fFD ƒt—ne™kiD uFeFD frownD „F et —lF @IWWWAF gountryEspe™i(™ estim—tes
—nd models of rs† —nd eshƒX methods —nd limit—tionsF AIDSD IQXPRRSEPRSVF

ƒpe™trum softw—re version PFQWF €rep—red ˜y the €oli™y €roje™t @PHHTAF …‚vX
http://www.unaids.org/en/HIV_data/Epidemiology/episoftware.asp. e™™essed on IS
eugustD PHHUF

ƒpir—D ‚FD vep—geD €FD wsell—tiD €F et —lF @IWWWAF x—tur—l history of rum—n
smmunode(™ien™y †irus „ype I infe™tion in ™hildrenX              e (veEye—r prospe™tive
study in ‚w—nd—F PediatricsD IHRXeSTF

ƒtoverD tF @PHHSAF esw †ersion RX        e ™omputer progr—m for m—king rs†Geshƒ
proje™tions —nd ix—mining the so™i—l —nd e™onomi™ imp—™ts of eshƒF …‚vX
http://www.unaids.org/en/HIV_data/Epidemiology/episoftware.aspF           e™™essed       on
IS eugustD PHHUF

ƒtoverD     tF    —nd      uirmeyerD    ƒF     @PHHRAF   hem€roj    †ersion       RX     e
™omputer         progr—m       for     m—king       popul—tion     proje™tionsF        …‚vX
http://www.unaids.org/en/HIV_data/Epidemiology/episoftware.asp.           e™™essed       on
IS eugustD PHHUF



                                         IPW
ƒtoverD tF et —lF @PHHPAF g—n we reverse the rs†Geshƒ p—ndemi™ with —n exp—nded
responseF The LancetD QTHXUQEUUF

„—h—D iF„F et —lF @PHHHAF wor˜idity —mong rum—n smmunode(™ien™y †irusEIEinfe™ted
—nd Euninfe™ted efri™—n ™hildrenF PediatricsD IHT@TAXeUUF

„he fody @PHHTAF ƒt—tement of ili—s eF erhouniD t—™k ‡hites™—rverD —nd enthony
ƒF p—u™iD x—tion—l snstitutes of re—lth on the PSth ennivers—ry of the pirst €u˜lished
‚eports of eshƒF …‚vX http://www.thebody.com/niaid/2006/aids_25.htmlF e™™essed
on IS eugustD PHHUF

„he …xeshƒ ‚eferen™e qroup on istim—tesD wodels —nd €roje™tions @PHHQAF
istim—ting —nd proje™ting n—tion—l rs† epidemi™sX „he models —nd methodology
of the …xeshƒG‡ry —ppro—™h to estim—ting —nd proje™ting n—tion—l rs†Geshƒ
epidemi™sF     …‚vX      http://www.constellafutures.com/software/EPP/epp_man.pdfF
e™™essed on IS eugustD PHHUF

        @PHHPAF smproved methods —nd —ssumptions for estim—tion of the rs†Geshƒ
epidemi™ —nd its imp—™tX ‚e™ommend—tions of the …xeshƒ ‚eferen™e qroup on
istim—tesD wodelling nd €roje™tionsF AIDSD ITX‡IE‡IRF

…xeshƒ @PHHTAF …xeshƒ PHHT report on the glo˜—l eshƒ epidemi™F qenev—D ƒwitzerE
l—ndX      …‚vX   http://www.unaids.org/en/HIV_data/2006GlobalReport/default.asp.
e™™essed on IS eugustD PHHUF

…xeshƒG‡ry @PHHSAF AIDS Epidemic Update: Sub-Saharan AfricaF he™em˜er PHHSF

…xsgip @PHHSAF The state of the world's children 2006F xew ‰orkX …xsgipF

        @PHHRAF The state of the world's children 2005F xew ‰orkX …xsgipF

        @PHHQAF The state of the world's children 2004F xew ‰orkX …xsgipF

        @PHHPAF The state of the world's children 2003F xew ‰orkX …xsgipF




                                       IQH
         @IWWVAF The state of the world's children 1998F xew ‰orkX yxford …niversity
€ressF

…xsgipD …xeshƒD ‡ry —nd …xp€e @PHHRAF HIV transmission through breastfeeding:
A review of available evidenceF qenev—X‡ry

‡—lkerD xFD ƒ™hw—rtl—nderD fF —nd fry™eD tF @PHHPAF weeting intern—tion—l go—ls in
™hild surviv—l —nd rs†GeshƒF The LacentD QTHXPVREPVWF

‡illD    qFpF @PHHTAF PS ‰e—rs of de—dly lessonsF ‡—shington postD         „uesd—yD
tune      TD    PHHTX     €—ge   eISF    …‚vX    http://www.washingtonpost.com/wp-
dyn/content/article/2006/06/05/AR2006060501285.html. e™™essed on IH xovem˜erD
PHHTF

‡ry @PHHS—AF The world health report 2005:make every mother and child
countFqenev—X‡ry

         @PHHS˜AF ƒumm—ry gountry pro(le for rs†Geshƒ „re—tment ƒ™—leE…pX w—l—wiF
…‚vX http://www.who.int/hiv/HIVCP_MWI.pdf. e™™essed on IS eugustD PHHUF

‡ry —nd …xeshƒ @PHHQAF ‚e™on™iling —nten—t—l ™lini™E˜—sed surveill—n™e —nd
popul—tionE˜—sed survey estim—tes of rs† prev—len™e in su˜Eƒ—h—r—n efri™—F …‚vF
http://data.unaids.org/UNA-docs/ANC-Population_surveys_report_en.pdf. e™™essed
on IS eugustD PHHUF

‡ry et —lFD @PHHUAF Towards universal access :       Scaling up priority HIV/AIDS
interventions in the health sector : Progress report, April 2007. qenev—X ‡ryF …‚vX
http://www.who.int/hiv/mediacentre/univeral_access_progress_report_en.pdf.e™™essed
on IS eugustD PHHUF

‰onderD €F —nd w—ting—D €F @PHHRAF Voluntary counseling and testing for HIV in
Malawi: Public perspectives and recent VCT experienceF g—lvertonX y‚g w—™roF
…‚vX : http://www.measuredhs.com/pubs/pdf/QRS8/MalawiVCT.pdf. e™™essed on IS
eugustD PHHUF


                                        IQI
—˜—D    fFD    w—rstonD    wFD    —nd     ploydD   ƒF   @PHHQAF   „he   e'e™t   of
rs†     on     ™hild   mort—lity   trends      in   su˜Eƒ—h—r—n    efri™—F    …‚vX
http://www.un.org/esa/population/publications/adultmort/Adultmortality.htmF
e™™essed on IP pe˜ru—ryD PHHU




                                     IQP

								
To top