6RAKAI FINAL PUNE Presentation 27 OCT 2009 Tom Lutalo by xiaohuicaicai


									           A community based trial of
           enhanced family planning outreach
           in Rakai, Uganda

Tom Lutalo1, Edward Kimera1
Godfrey Kigozi1, Fred Nalugoda1,
David Serwadda1, Maria Wawer2, Laurie Schwab Zabin2,
Fred Wabwire-Mangen1, Nelson Sewankambo1, Ron Gray2

1RakaiHealth Sciences Program
2Johns Hopkins Bloomberg School of Public Health

                      9th INDEPTH AGM, 27th OCT
                            2009, PUNE INDIA

•   What RAKAI does
•   A review of previous findings
•   The objectives of the intervention
•   Community selection
•   The Intervention activities
•   Results
•   Public Health Challenges
          The Rakai Health Sciences Program (RHSP)

• Since 1994 a cohort of ~12-14,000 adults aged 15-49
  yrs has been under annual surveillance

• RHSP provides health education about
      • STDs & HIV
      • Family planning
      • Abstinence and condom promotion for prevention of
        STDs and HIV
      • Condoms, which are also provided free of charge or at
        subsidized prices
                1995-99 Cohort and FGD findings

• Only 13% of women who wanted no more children were
  using contraceptives

• 15% of women who wanted to delay pregnancy by >2 yrs
  were practicing family planning

• FGD participants were more vocal about perceived health
  risks of FP than about potential benefits

• Impediments included women’s access to FP services,
  costs of contraception and men’s negative attitudes
  towards FP

• Identified a need to introduce and evaluate interventions to
  correct misconceptions, and to improve access to quality
              Objectives of Intervention

• Establish and maintain community based
  outreach programs to improve awareness and
  utilization of hormonal contraceptives
  (the pill and injections which are female
  controlled methods)

• Evaluate impact of intervention using cohort data
   – Changes and differences in hormonal contraceptive use
     and pregnancy rates by study arm
          Community selection and study period

• Convenient sampling of 6 clusters nested in 11
  clusters (initially) randomized for the RAKAI STD

• The 11 clusters form the RAKAI cohort

• The intervention was staggered over a three year
  period (1999 - 2002)
                Intervention activities
• Community members identified FP agents and Service
  Providers who accepted to volunteer

• RAKAI trained, followed up and continued to retrain

• Used HIV counselors at time of HIV result giving

• Used DRAMA shows with FP themes

• Programs for in-school, out of school and TBAs
• Service Delivery
   – Supplied and equipped volunteers to provide condoms, oral
     contraceptives and injectables

• Continued sharing of experiences
   – Amongst volunteers
   – Village meetings->Certified Satisfied Users of hormonal
     contraceptives and condoms testified
            Non intervention clusters

• Standard national (and NGO) programs of promoting
  FP services using selected Health Units

• Condoms supplied by RHSP in both arms
             Evaluation of intervention

• Used RAKAI cohort survey data

  – Compared promoted contraceptive use at baseline (1999-
    2000; RND 6) and after over 30 months follow up (2003-04;
    RND 9) in FP and non-FP clusters

  – Pregnancy rates by arm

  – Chi-square tests of statistical significance for absolute
    change in contraceptive use by arm

  – Multivariate adjusted analyses included baseline hormonal
    contraceptive use, socio-demographic and behavioral
    characteristics found to differ at baseline and variables
    thought to be related to hormonal contraceptive use
        Socio-demographic and behavioral characteristics in the
          intervention and non intervention arms at baseline
                                         Baseline (1999/2000)
                                        Intervention  Control
                                          (6720)       (3574)
Characteristics                             %            %                          p-value
   Female                                     53.9                  51.0
   Male                                      46.1                    49.0             0.006
   Rural                                      80.3                  83.4
   Peri-Urban                                19.7                   16.6              <0.0001
Age (yrs)
   15-19                                      14.7                  12.4
   20-29                                      46.6                  48.6
   30-39                                      25.8                  28.3
   40-49                                      12.9                  10.7              <0.0001
Marital Status
   Not Married                               29.0                    24.5
   Married                                   71.0                    75.5             <0.0001
Number of children
   0-1                                       32.2                    32.0
   2-4                                       41.0                    44.6
   5+                                        26.8                    23.4             0.010
Other characteristics compared
Religion p<0.0001, Education level p=0.246, Sex partners past yr p=0.086
HIV status p=0.158
                       Use of Hormonal Contraceptives
           between study arms at baseline and follow up

• Use of hormonal
  contraceptives was similar
  at baseline
  (13.4% Vs 13.8%, p=0.69)
• At follow up, hormonal
  contraceptive use was            10
  significantly higher in the FP
  clusters compared to non FP      5
  (23.2% Vs 19.9%, p=0.009)        0
                                    1999/2000         2003/4

                                        FP clusters     Non FP
         Use of oral contraceptives in rural and
         peri-urban communities at baseline and
                      FUP by study arm
• At baseline, oral                             Rural
  contraceptive prevalence         6

  was slightly higher in the       5
  intervention arm in both         3                       FP clusters
  rural and peri-urban areas       2
                                                           Non FP

• In the rural areas there was a    1999/2000     2003/4

  significant increase in use
  (p<0.001) in the intervention
  arm with no change in the
  non intervention arm                      Peri-urban

• Slight increase in use were      6
  observed in the peri-urban       4                       FP clusters
  communities in both study        2
                                                           Non FP

  arms (ns)                        1
                                    1999/2000     2003/4
                  Use of Injectables in rural and
                 peri-Urban Communities by study arm
                at baseline and FUP
• There was no change in         25

  use in non-intervention        20

  peri-urban clusters            15
                                                               FP clusters
                                 10                            Non FP


• At FUP there was a                 0
                                         1999/2000   2003/4
  significant increase in
  use observed in the
  intervention arm              20
  (p<0.001)                     15

                                10                                     FP clusters
                                                                       Non FP

• In the rural                   5

  communities there was          0
                                  1999/2000           2003/4

  an increase in both
          Pregnancy rates between study arms at
                 baseline and follow up

• Pregnancy rates at baseline
  were comparable                    18
  (16.5% Vs 17.6%, p=0.276)          16
• At follow up pregnancy rates       12
  were significantly lower in the
  intervention clusters compared     10
  to the non intervention clusters   8
  (12.5% Vs 14.8%, p=0.024)          6
                                      1999/2000         2003/4

                                          FP clusters     Non FP
         Multivariate logistic regression

• Odds of hormonal contraceptive use at
  follow up were increased in
   – intervention versus control communities
     (adj OR 1.33, 95% CI 1.03 – 1.72)
   – Intervention communities with earlier FP (adj
     OR=1.36, 95% CI 1.02-1.81)

Duration of intervention, education and number of
  children were strong determinants of hormonal
  contraceptive use
         Condom use and adjusted analysis

• Condom use by males in the two study arms
  increased significantly between baseline and follow
  up in the intervention arm (p<0.01) but the increase
  was not statistically significant in the control arm
   (Intervention: 30.5% to 33.9%)
   (Control:     30.1% to 33.0%)

• The intervention resulted in
   –  modest but statistically significant increased use of
     hormonal contraceptives (pill & injections)
   – Lower pregnancy rates in intervention arm
         Public Health challenges for such an
• Intensive supervision of the volunteers

• High turnover rate of volunteers requiring
  identifying and training of new ones

• Increase in counseling time by resident HIV
  counselors for quality FP counseling

• High demand for modern contraceptives

• Sustainability
   – The intervention activities cost 300,000 $ over the three
        RHSP appreciates the efforts of;

The Gates Institute at Johns Hopkins Bloomberg
  University (funders of intervention)

The RHSP Family Planning Team
The PIs and all other staff of RHSP
The district teams involved
The study volunteers
INDEPTH for accepting RAKAI to share experience

RHSP Research Center

To top