DESCRIPTIVE EPIDEMIOLOGY OF TYPE 1 DIABETES by zlv64835

VIEWS: 0 PAGES: 38

									  DESCRIPTIVE
EPIDEMIOLOGY OF
 TYPE 1 DIABETES
 Jaakko Tuomilehto
   Department of Public Health
       University of Helsinki
                and
           Diabetes Unit
  National Public Health Institute
        Helsinki, Finland
                 Two Broad Types of
                   Epidemiology

DESCRIPTIVE EPIDEMIOLOGY              ANALYTIC EPIDEMIOLOGY
Examining the distribution of a       Testing a specific hypothesis
disease in a population, and          about the relationship of a
observing the basic features of its   disease to a putative cause,
distribution in terms of time,        by conducting an
place, and person.                    epidemiologic study that
                                      relates the exposure of
Typical study design:                 interest to the disease of
                                      interest.
community health survey
(approximate synonyms - cross-
sectional study, descriptive          Typical study designs: cohort,
study)                                case-control
         The Basic Triad Of
      Descriptive Epidemiology
    THE THREE ESSENTIAL CHARACTERISTICS
    OF DISEASE WE LOOK FOR IN DESCRIPTIVE
    EPIDEMIOLOGY:

   TIME
   PLACE
   PERSON
  Descriptive Epidemiology Is A
    Necessary Antecedent Of
    Analytic Epidemiology

To undertake an analytic
  epidemiologic study you must first:
 Know where to look
 Know what to control for
 Be able to formulate hypotheses
  compatible with laboratory
  evidence
       A COMMON ERROR
IN EPIDEMIOLOGY IS MOVING TO
    ANALYTIC EPIDEMIOLOGY
       WITHOUT HAVING
      A SOLID BASE IN THE
  DESCRIPTIVE EPIDEMIOLOGY
      OF THE CONDITION.
Prevalence of type 1 diabetes in childhood
       Countries with over 10,000
       estimated prevalent cases

Country              Estimated cases (‘000)
India                        88.8
USA                          49.2
Brazil                       23.8
Bangladesh                   14.3
UK                           13.7
Russian Federation           12.4
Egypt                        11.8
Germany                      10.1
DIABETES REGISTRIES –
     MAIN ISSUES
• CASE definition
• POPULATION denominator
• COMPLETENESS of the ascertainment
• CONSISTENCY over time
• STATISTICAL ANALYSIS of the data
      DIABETES INCIDENCE
    REGISTRIES – MAIN ISSUES
• CASE definition
       • changes over time
       • effect of other diseases
• POPULATION denominator
       • dynamic populations
       • population stratification/ethnicity
• COMPLETENESS of the ascertainment
       • capture-recapture estimation
• CONSISTENCY over time
       •. changes in political and health care systems
• STATISTICAL ANALYSIS of the data
       • age standardization
       • meta-analysis to compare
                 WHO
       Diabetes Mondiale




WHO Multinational Project for Childhood Diabetes
                   DiaMond
        Diabetes Mondiale - DiaMond

Establishment of Population Based Registries for 1990-99


         •   Diagnosed as T1DM
         •   On Insulin at first Hospital discharge
         •   Diagnosed between 1/1/90 and 31/12/99
         •   Age at onset 0-14
         •   Member of a defined community
      WHO DiaMond project –
          first 5 years
•   33718 cases of T1DM
•   443 499 882 person years
•   106 centers have provided data
•   189 centers listed
•   71 countries
• 100 Centers included in the first
  publication
                           (Diabetes Care, 2000; 23: 1516-1526.)
     Incidence of Type 1 diabetes mellitus in Europe (per 100,000 /year),
               - high incidence populations, 1990-1994.

       Sardinia ITA                                                                                    36.8
           Finland                                                                                    36.5
           Sweden                                                                 27.5
    Aberdeen GBR                                                           24.0
   8 counties NOR                                                   21.2
   Portalegre POR                                                   21.1
    N.Ireland GBR                                               19.7
      Oxford GBR                                         17.8
     Plymouth GBR                                       17.3
    4 counties DNK                               15.5
Leicestershire GBR                             15.3
       Algarve POR                           14.6
     5 regions NLD                        13.0
     Catalonia ESP                       12.5
          Pavia ITA                    11.7
  Eastern Sicily ITA                   11.7
       Antwerp BEL                    11.6
       Luxembourg                     11.4
 Baden-Wuett. GER
 Baden-Wuett.                      11.0
          Turin ITA                11.0
            Estonia               10.5

                   0.0   5.0   10.0         15.0            20.0           25.0       30.0      35.0          40.0
                                                                                     DIAMOND, Nov. 2000
          Incidence of Type 1 diabetes mellitus in Europe
          (per 100,000/ year), low incidence populations, 1990-1994
West-Bulgaria BUL                                                                     9.9
     Coimbra POR                                                                     9.7
       Attica GRC                                                                    9.7
       Marche ITA                                                                    9.7
           Austria                                                                  9.6
  18 counties HUN                                                             9.1
     4 regions FRA                                                      8.5
          Slovakia                                                     8.5
         Lazio ITA                                                8.1
          Slovenia                                               7.9
         Lithuania                                         7.4
Madeira Island POR                                        7.2
    Lombardia ITA                                         7.2
        Varna BUL                                   6.8
      Cracow POL                              6.1
            Latvia                          5.9
   Bucharest ROM                      5.0
 Wielkopolska POL                     5.0
                                                                        DIAMOND, Nov. 2000
                 0.0   2.0      4.0         6.0                 8.0                  10.0    12.0
      Incidence of Type 1 diabetes mellitus in South America
      (per 100,000/ year), 1990-1994

      Montevideo URY                                                                                8.3

 Tierra del Fuego ARG                                                                          8.0

  Sao Paulo state BRA                                                                          8.0

        Cordoba ARG                                                                     7.0

      Avellaneda ARG                                                             6.5

      Corrientes ARG                                          4.3

Santafè de Bogotà COL                                   3.8

        Santiago CHL                      1.6

            Paraguay                0.9

            Lima PER          0.4

         Caracas VEN      0.1

                        0.0         1.0     2.0   3.0   4.0         5.0   6.0          7.0    8.0         9.0


                                                                                DIAMOND, Nov. 2000
Incidence of Type 1 diabetes mellitus in North America, in
Central America and West Indies (per 100,000/year), 1990-4
Prince Edward Island CAN                                                                        24.5

            Alberta CAN                                                                        24.0

      Allegheny, PA USA                                                          17.8

       Jefferson, AL USA                                                  15.0

         Chicago,IL USA                                         11.7

        Puerto Rico USA                                                          17.4

      Virgin Islands USA                                           13.1

               Dominica                            5.7

                   Cuba                2.9

               Barbados           2.0

           Veracruz MEX          1.5

                           0.0               5.0         10.0          15.0             20.0   25.0    30.0
    Incidence of Type 1 diabetes mellitus in Asia (per 100,000/year), 1990-4
Novosibirsk RUS                                                                             6.0
    Wuhan CHN                                                                  4.6
   Sichuan CHN                                               2.3
  Hokkaido JPN                                             2.2
  Okinawa JPN                                  1.4
      Chiba JPN                                1.4
Hong Kong CHN                                1.3
     Dalian CHN                            1.2
   Huhehot CHN                       0.9
    Beijing CHN                     0.9
    Nanjing CHN                    0.8
ChangChun CHN                      0.8
     Guilin CHN                   0.8
 Wulumuqi CHN                     0.8
  Shanghai CHN                  0.7
    Karachi PAK                 0.7
Zhengzhou CHN                 0.6
   Nanning CHN              0.5
 Shenyang CHN               0.5
      Jinan CHN            0.4
       Jilin CHN           0.4
   Lanzhou CHN             0.4
 Changsha CHN            0.3
     Harbin CHN         0.3
    Hainan CHN         0.2
    TieLing CHN        0.2
      Zunyi CHN      0.1

               0.0                 1.0               2.0           3.0   4.0         5.0   6.0    7.0
            Incidence of Type 1 diabetes mellitus in Oceania, Africa,
                  and Near East (per 100,000/year), 1990-1994.


   Canterbury NZL                                                                        21.9

N. South Wales AUS                                                  14.5

     Auckland NZL                                            12.9

           Kuwait                                                           18.3

        Gafsa TUN                                   8.8

         Beja TUN                             7.8

      Kairoan TUN                      5.7

            Israel                      5.9

         Oran TUN                      5.7

     Monastir TUN                 4.9

       Gezira SDN                 5.0

         Mauritius         1.4

                     0.0         5.0                  10.0          15.0        20.0            25.0
                                                                           DIAMOND, Nov. 2000
                                                                         Incidence 1/100 000
                    07
                         2
                             =
                                 D
                                     om
                                 in
                                    ic
                                       a




                                                                         10
                                                                              15
                                                                                   20
                                                                                        25
                                                                                             30
                                                                                                  35
                                                                                                       40
                                00 n R                                                                      45




                                                                 0
                                                                     5
                                   4           e
                                       = pu
                     07                    Pa b l i
                         1                      r           c
                   08 =                02 a g u
                      3 Sa 9 = ay
               05        =          n
                  3          M t i a g Cu
                     =          on          o            ba
              16 N
                 9        o as (C
                    = v o s tir ( hile
                       Pa i b                 T             )
                           ss i r s uni
             15                o         k             s
                1                 Fu ( R ia)
                   =                  nd u s
                     C                     o sia
                        or                    (             )
                           rie 0 6 B r a
                    02 n                   6           si
                        2 tes                 = l)
                           =                       L
                              C (Ar a t v
                                ra          ge             i
                         0 8 co                  nt a
                            2          w             i
                     04 =                  (P na)
                         0         B            o
                            = e j a lan
                               Va           ( T d)
                                   rn           un
                    01                 a             is
                        1        0 3 (Bu i a )
                           =         5           lg
                              B          =          ar
                      14 e n L i ia)
                          8          g          th
                             = haz ua
                                 G           i ( nia
                 0 2 055 a f s L i b
                    1                    a             y
                       = = O (Tu a )
                          C          ra            ni
                             oi
                                m        n             si
               1                  b r (Al a )
         1 1 27                       a           g
             7 =R                        ( P er
               =                  1 0 o r ia)
                  W e g i 0 tug
                     e s on =                           al
                08 t - B M S l o )
                   1         u         ar           va
                      = l g a ch                         ki
                                                             a
                         Tu r i a e
                             rin                   (
                                         ( B Ita
18                                pr u l g ly)
   5               00
                                     ov            a
     =                                   in r i a
       Ba 0 5          9                     c             )
                          = 07 e (
          de 9               C 8                    I
             n-      =          at         = taly
                W P a alo Es )
                                        n           to
                                                                                                                       DiaMond 1995-1999




                   ue v i
                      rtt a p i a ( n i a
                         em r o S
                              be v i n             pa
    13
       5       11                 rg c e i n )
         =        2                   (G ( I
           C                               er t a l
              hi = P
                ca                  1 8 ma y )
                   go uer               9 = ny
                       IL to R C )
                           (U ic a l g
                               .S o ( a r
                                  .A U                      y
        02                            .)          .S
           0
             =                  1 8 afr .A.)
                                              .
                C                   4
                  an                   = Am
                     te                    P l er
                        rb                    y             .
                           ur 0 6 9 m o
                              y
                                 (N = K u t h
                                     ew u
                                                     w
                     00                    Z            a
                         1 0 7 6 eal i t
                            =                      an
                               Ab = S d
                                   er           w )
                                       de e d
                                           en e n
                                   03
                                       0 (U.K
                                           =
                                               F i .)
                                                   nl
                                                       an
                                                           d
                                                                                                             Incidence of Type 1 diabetes mellitus /100,000/year
                                                                                                    M/F-ratio
                 15
                      1
                          =
                              C
                   16              or
                                        rie
                          9
                              =               nt
                                                   es
                                  Pa
                                        ss              (A




                                                                                      0.00
                                                                                             0.50
                                                                                                    1.00
                                                                                                           1.50
                                                                                                                  2.00
                              01             o               rg
                                                                  en
                                                                                                                         2.50
                                   1               Fu
                                                                       tin
                                       =                nd
                                           Be                o               a)
                                                   ng             (B
                                   05                   ha             ra
                                         5                                  si
                                                                                 l)
                                              =              zi
                                                   O              (L
                                                       ra              ib
                                                            n               ya
                                   18 (Al                      )
                                       4            ge
                      11                    =            ria
                        2                      Pl
                      =                           ym )
                         Pu                            ou
                              er          02
                                 to            9             th
                                     R            =
                                        ic            C
                                            o            ub
                                               (U              a
                                      06 .S.A
                                            9               .)
                                               =
                                      18 Kuw
                                            9=             ai
                                                               t
                       0           03             C
                0 5 71                 5              al
                   3          =             =            ga
                      =          S             Li              ry
                          N a n t thua
                            ov           i                 ni
                      04         o s ago                       a
    13                    0         ib
      5        05            =
                                       i r s (Ch
        =         9
          C                     V a k ( ile)
                                                  R
            hi = P                   rn              us
               ca         a             a                 si
                  go via                     (B               a)
                      IL           pr            ul
                                                    ga
                          (U          ov
                                                         ria
                              .S           in
18                               .A           ce               )
   5                 02
                         2          .)
     =                      =           a          (It
                                                       a
       Ba                      C fr. A ly)
           de                     ra
              n-                      co me
                 W                        w            r.
                   ue                          (P
                       rtt          1             ol
                           e m 00                     an
                                             =              d)
                     00 b e                      Sl
                         9          rg              o
                            =            ( G vak
                               C              e               ia
                                  at
                                     al r m a
                                       on              ny
                                             ia              )
                      00            07 (S
                          1             6             pa
                             =                             in
                                Ab = S                         )
                                    er            w
               12                      de             ed
                  7                          en             en
                     =
                        R
                           e          03 ( U .
                                            0            K.
                   02 gio
                       1          n            =             )
                                                                                                                                        DiaMond 1995-1999




                                                  Fi
                08 = C Ma                             nl
                   1            o i r c h and
                      =            m
                         Tu           br e (I
       02                    rin a ( taly
          0                                    Po              )
            =                      pr               rtu
               C                      ov
                                           in            g
                  an                          ce al)
                     te
                        rb                          (
                           ur            06 I t a
                               y              6            ly
                                  (N             =             )
                                      ew             La
                                                          tv
          11                                  Ze              ia
             7                        07 ala
                =                           8          nd
                   W
                     es           00 = E )
                         t-B          4              st
                              ul          =             on
                                 ga Pa                        ia
                                                  ra
                           08 r i a                    gu
                               2            (B               ay
                        14 = B u l g
                            8          ej            ar
                    08 = G a ( ia)
                        3            af
                                                 Tu
                           =                          ni
                                                         s
                  07          M sa
                     2           on            ( T ia)
                        =             as          un
                           D                           i
                              o m tir ( sia
                                   in            Tu )
                                      ic              ni
                                         an              si
                                                             a)
                                                R
                                                  ep
                                                       ub
                                                             lic
                                                                                                                                Male-female ratio in incidence of Type 1 diabetes
Different methods for summarizing the evidence

Type I     Qualitative summary, the narrative review article

Type II    Quantitative summary of published data
           (usually called meta-analysis)

Type III   Re-analysis of individual data from primary studies.
           This is sometimes called meta-analysis,
           but in epidemiology the term ’pooled analysis’
           is more often used.

Type IV    Prospectively planned, pooled analysis of
           several studies, where pooling is already a part
           of the protocol.
           Data collection procedures, definition of variables,
           questions and hypotheses are standardized
           for the individual studies.
       Main reasons for conducting
       a meta-analysis or a review
1. To assess qualitatively whether a factor has to be
     considered as a risk factor.

2. To provide more precise effect estimates and increased
   statistical power and to analyse dose-response relations.

3. To investigate the heterogeneity between different
      studies.

4. To generalize results of single studies.

5. To investigate rare exposures and interactions.

6. To investigate risks associated with rare diseases.
   The trend in the incidence of Type 1 diabetes in Finnish
                  children aged 14 or under
60.0

50.0

40.0

30.0

20.0

10.0

0.0
   1963   1968   1973   1978      1983   1988        1993         1998
                               Year

                                         Tuomilehto et al. Diabetologia 1999; 42: 655-660.
INCIDENCE TRENDS FOR CHILDHOOD-ONSET
       TYPE 1 DIABETES - EUROPE




                    Onkamo et al. Diabetologia 1999; 42: 1395-1403
Variation and trends in incidence in childhood
diabetes in Europe. Lancet 2000;355:873-6

  EURODIAB ACE Study group findings 1989-94:
  16,362 cases from 44 centres in 29 countries
  Population of children covered : 28 million
  Annual increase (pooled data) : 3.4% (95% CI 2.5-4.4%)

  Age specific increases (pooled data):
  0 - 4 years                6.3% (4.1 - 8.5%)
  5 - 9 years                3.1% (1.5 - 4.8%)
  10 - 14 years              2.4% (1.0 - 3.8%)
INCIDENCE TRENDS FOR CHILDHOOD-ONSET
    TYPE 1 DIABETES - OUTSIDE EUROPE




                     Onkamo et al. Diabetologia 1999; 42: 1395-1403 .
Type 1 Diabetes Incidence in the World

  • Mean incidence during 1990-1999
    – Males       7.8 /100 000
    – Females     7.4 /100 000
    – M/F-ratio     1.06
  • Trend:
    – An increase ~3% / year
   The trend in the incidence of Type 1 diabetes in Finnish
                  children aged 14 or under
60.0

50.0

40.0

30.0

20.0

10.0

0.0
   1963   1968   1973   1978      1983   1988        1993         1998           2001
                               Year

                                         Tuomilehto et al. Diabetologia 1999; 42: 655-660.
          Age-specific incidence of Type 1 diabetes in
       Finnish children aged 1-14 years from 1965 to 1999.
60
       per 100,000/year.
50


40
                                                                1-4
30                                                              5-9
                                                                10-14
20


10


0
     1963   1968   1973    1978     1983   1988   1993   1998
                                  Year
   How To Estimate The
    Penetrance of T1DM
  Susceptibility Genes ?

• Parents of IDDM patients
• Siblings of IDDM patients
• Twin pairs with IDDM
         Type 1 diabetes in siblings of
        probands with Type 1 diabetes
•5126 probands (first diabetic child in the family
     with T1D diagnosed < 18 years
during 1965 – 1979.

• 647 siblings (6.4%) of diagnosed with T1D
      by 2001

• Average duration of diabetes
     • 26,6 + 4,7 years among probands
     • 17,4 + 8,6 years among siblings
Type 1 diabetes in siblings of 5126
  probands with Type 1 diabetes
                               Number of        Number of cases
                               siblings           with T1D
Sex of the sibling
Male                           5,202               367 (7.1%)
Female                         4,966               280 (5.6%)

Family size
1 child     (682 families)              -----
2 children (1,606 families)    1,606               106   (6.6%)
3 children (1,230 families)    2,460               168   (6.8%)
4 children   (678 families)    2,034               133   (6.5%)
≥5 children (772 families)     4,068               240   (5.9%)

Total       (4,968 families)   10,168              647 (6.4%)
     Risk of diabetes in siblings of diabetic
     children by proband's age at diagnosis
 %

12




10
                                                               0-4

 8




 6
                                                               5-9
                                                           10-14
 4

                                                           15-17
 2




 0
     5      10       15            20                25   30
                  Attained age of siblings (years)
                                                                                                                                                                                                ,18
                                                       ,18
                                                                  A                                                                                                                                       B
Cumulative incidence of type 1 diabetes in siblings




                                                                                                                                         Cumulative incidence of type 1 diabetes in siblings
                                                                                                                                                                                                ,16       P<0.0001,
                                                       ,16        P<0.0001,
                                                                                                                                                                                                          log-rank test
                                                                  log-rank test                                      0-4
                                                                                                                                                                                                ,14
                                                       ,14
                                                                                                                                                                                                                               ≥ 1980
                                                       ,12                                                                                                                                      ,12


                                                       ,10                                                                                                                                      ,10
                                                                                                                          0-9
                                                                                                                                                                                                ,08                                     1970-79
                                                       ,08

                                                       ,06                                                               10-                                                                    ,06
                                                                                                                                                                                                                                                     ≤ 1970
                                                                                                                         14
                                                       ,04                                                                                                                                      ,04
                                                                                                                         ≥15
                                                       ,02                                                                                                                                      ,02

                                                      0,00                                                                                                                                     0,00
                                                             0           10           20           30       40      50                                                                                0        10         20       30       40       50

                                                                              Attained age (years)                                                                                                                    Year of birth
                                                                 NUMBER OF SIBLINGS AT RISK                                                                                                           NUMBER OF SIBLINGS AT RISK

       0-4:                                                      1,306        1,242        1,102         822       375           50    ≥ 1980:                                                        208        189       64
       5-9:                                                      2,811        2,745        2,613        2,236    1,233          241    1970-79:                                                        971        933      904      273
     10-14:                                                      4,156        4,120        4,023        3,833    2,601          669     ≤ 1970:                                                       8,989      8,876     8,630    8,425    5,605
                                                                                                                                      1440
      ≥15:                                                       1,895        1,890        1,860        1,807    1,396          488
   The effect of selected predictor variables on the age at
 diagnosis of Type 1 diabetes in siblings of Type 1 diabetes
                            cases

Predictor          Parameter estimate Standard error   p-value

Sex of index              0.86             0.66         0.19
case
Age at                    0.39             0.07        0.0001
diagnosis of
index case
Number of                 1.84             0.73        0.012
siblings
Closest sibling           2.90             0.69        0.0001
by date of birth
Sex                       0.66             0.66         0.32
concordance
Number of concordant and discordant twin pairs for Type 1
   diabetes: Finnish Young Twin Cohort (N=26,000) -
     Born 1968-1986; Follow-up until the end 1998

      Gender        Zygosity   Concordant   Discordant
                                 pairs         pairs
      Male            MZ           7           20
      pairs           DZ           2           46

      Female          MZ           5           12
      pairs           DZ           1           44

      Opposite        DZ           4           85
      sexed pairs


      All             MZ           12          32
      pairs           DZ           7           175
Concordance for Type 1 diabetes
 In Finnish Young Twin Cohort

  Concordance for    MZ twins   DZ twins
  Type 1 diabetes

  Probandwise (P1)     42.9 %     7.4 %

  Pairwise (P2)        27.3 %     3.8 %
Results from fitting threshold models to Type
 1 diabetes data in the Finnish Young Twin
                    Cohort
    COMPONENTS OF VARIANCE                       GOODNESS OF FIT

    MODEL A              C      D       E        !2       df p-value AIC

    ACE         0.877 0.00 -            0.123 4.83        3    0.18      -1.67

    AE          0.877 -         -       0.123 4.83        4    0.30      -3.17

    ADE         0.79     -      0.09 0.12        4.75     3    0.19      -1.25

    CE          -        0.60 -         0.40     30.77 4       0.000     22.8

    A, additive genetic effects;
    C, shared (family) environmental effects;
    D, due to dominance;
    E, unshared environmental effects.
    The best fitting model is generally one with the lowest AIC-value.

								
To top