We
tested the hypothesis that maternal smoking during pregnancy increases
both the risk of early onset type 2 diabetes and non-diabetic
obesity in
offspring.
We used data from the British National Child
Development Study (NCDS), based on the Perinatal Mortality Survey (PMS)
of about 17000 births from 3 to 9 March 1958. The first sweep of
the study in 1965 had 15396 responses. The cohort remained
generally representative at age 33 years (n=11359). A
total of 5998 cohort members were followed. Missing data reduced the
proportion in social class V from 6.4% in 1965 to 5.3%.
Ethics committee approval was obtained for research involving medical
examinations, and cohort members signed consent forms at age 33
allowing access to medical
records.
Medical examinations and
record reviews by local authority medical officers were conducted at
ages 7 and 16. Children with incomplete or equivocal information on
diabetes or with a recorded onset before the age of 16 were not
included in the main analysis as they are unlikely to have type 2
diabetes. A personal interview at age 33 asked about diabetes. Those
with only gestational diabetes were also excluded; 15 men and 13 women
with an onset of diabetes between 16 and 33 years were
identified.
At the birth midwives
recorded information on the child's sex, birth weight,
mother's age, mother's age on leaving full time education,
family social class, and smoking during pregnancy (after the fourth
month) divided into non-smokers, medium (1-9
cigarettes/day) heavy (>10), and variable (a balance of medium
and heavy). Details of maternal smoking were again recorded in 1974, as
non-smoking and <1, 1-5, 6-10, 11-20,
21-30, or >30
cigarettes/day.
Cohort
members' own smoking behaviour was recorded during an interview at
age 16 and they were classified as non-smokers or as smoking
<1, 1-9, 10-19, 20-29, >29
cigarettes/week. Interviewers measured height in centimetres and
weight in kilograms using stadiometers and electronic balances at age
33. Multiple logistic regression analysis was used for two outcomes:
diabetes and body mass index (BMI) >30, independent of diabetes.
Where obesity was the outcome, people with diabetes were excluded and
we adjusted for sex, own smoking at age 16, and all the maternal
factors at birth.
After we excluded
the diabetic cohort members, some 10% (493) were obese
(BMI>30) at age 33. The adjusted odds ratios (and 95%
confidence intervals) for obesity associated with maternal smoking
during pregnancy were 1.34 (1.07 to 1.69), 1.35 (0.95 to 1.92), and
1.38 (1.06 to 1.79), with a significant trend (P=0.003) for
medium, variable, and heavy smokers, respectively
(table). Non-diabetic
cohort members who smoked at age 16 did not have an increased risk of
obesity.
Comment
The
association of diabetes with maternal smoking during pregnancy
(independent of finer-grain measures of mothers' smoking in
1974, own smoking at age 16, and other potential confounding factors)
suggests that it is atrue risk factor for early adult onset diabetes.
Cigarette smoking as a young adult was also independently associated
with an increased risk of subsequent
diabetes.
In utero exposures due to
smoking during pregnancy may increase the risk of both type 2 diabetes
and obesity through programming, resulting in lifelong metabolic
dysregulation, possibly due to fetal malnutrition or toxicity. The odds
ratios for obesity without type 2 diabetes were more modest than those
for diabetes and the scope for confounding may be greater. Smoking
during pregnancy may represent another important determinant of
metabolic dysregulation and type 2 diabetes in offspring. Smoking
during pregnancy should always be strongly
discouraged.
Maternal
smoking in pregnancy, own smoking at age 16 years, and risk of diabetes
among young adults. Odds ratio with 95% confidence intervals
using logistic regression with a diagnosis of diabetes after age 16 as
the dependent
variable
| No
(%) without
diabetes (n=4917) | No
(%) with
diabetes (n=28) | Unadjusted | Adjusted* | | | Adjusted† |
| | | Odds
ratio (95%
CI) | P
value | Odds
ratio (95%
CI) | P
value | Odds
ratio (95%
CI) | P
value |
Mother
smoking while pregnant (after fourth
month) |
Non-smoker | 3430
(69.8) | 12
(42.9) | 1.00 | | 1.00 | | 1.00 | |
Medium
smoker | 689
(14.0) | 3
(10.7) | 1.24
(0.35 to
4.42) | 0.735 | 1.11
(0.31 to
4.04) | 0.872 | 1.01
(0.23 to
4.53) | 0.990 |
Varies
between medium and
heavy | 277
(5.6) | 4
(14.3) | 4.13
(1.32 to
12.88) | 0.015 | 4.13
(1.27 to
13.40) | 0.018 | 3.55
(0.88 to
14.38) | 0.076 |
Heavy
smoker | 521
(10.6) | 9
(32.1) | 4.94
(2.07 to
11.77) | <0.001 | 4.55
(1.82 to
11.36) | 0.001 | 4.02
(1.14 to
14.14) | 0.030 |
Own
smoking at age 16 years
(cigarettes/week) |
Non-smoker | 3251
(66.1) | 13
(46.4) | 1.00 | | 1.00 | | 1.00 | |
<1 | 147
(3.0) | 1
(3.6) | 1.70
(0.22 to
13.09) | 0.610 | 2.09
(0.254 to
17.23) | 0.492 | 2.07
(0.25 to
17.19) | 0.502 |
1-9 | 510
(10.4) | 3
(10.7) | 1.47
(0.42 to
5.18) | 0.548 | 1.76
(0.48 to
6.46) | 0.394 | 1.92
(0.52 to
7.10) | 0.332 |
10-19 | 176
(3.6) | 2
(7.1) | 2.84
(0.64 to
12.69) | 0.171 | 2.41
(0.51 to
11.41) | 0.268 | 2.48
(0.52 to
11.97) | 0.257 |
20-29 | 189
(3.8) | 1
(3.6) | 1.32
(0.17 to
10.17) | 0.788 | 1.42
(0.177 to
11.43) | 0.741 | 1.61
(0.20 to
12.96) | 0.653 |
≥30 | 644
(13.1) | 8
(28.6) | 3.11
(1.28 to
7.52) | 0.012 | 3.53
(1.40 to
8.91) | 0.008 | 3.62
(1.42 to
9.24) | 0.007 |
*Adjusted
for maternal smoking during pregnancy, sex, mother's age at birth
of cohort member, age mother left school, family social class at birth,
birth weight, own smoking at age 16 years, and body mass index at age
33
years.
†Adjusted
for all of the above and maternal smoking in
1974.
Enheten för Klinisk Epidemiologi, Karolinska Sjukhuset L1:00, SE-171 76 Stockholm, Sweden
Anders Ekbom, professor
Scott M Montgomery, principal research
Email: Scott.Montgomery@medks.ki.se
studentBMJ 2002;10:45-88 March ISSN 0966-6494
