Prevalence of overweight and obesity in British children: cohort study

Dear sir,

We were surprised by Reilly’s implication [1] that a discrepancy between
the observed prevalence of overweight in children measured between 1995-7
and the 1990 UK growth standards was evidence of a rapid secular trend to
increased obesity in early childhood. To make any inferences about change
over time on the basis of measurements at a single point in time seems
unwise, particularly when compared to new and largely untested standards.

Thirty years ago an epidemic of childhood obesity was predicted when
upward centile crossing was seen in groups of infants [2]. This was later
found to be an artefact of the standards, when the same growth pattern was
found in malnourished Gambian children [3].

We hold comparable height and weight data on two birth cohorts of
Newcastle children, from the Thousand families study [4], born in 1947 and
the Growth and Development Study [5], born 1987-8. In the 1947 cohort, 688
(89%) children were measured at mean (standard deviation) age 9.0 (0.12)
years while in the 1987-8 cohort 448 children (78% of a 20% representative
sample) were measured aged 8.5 (0.3) years. Both sets of data were
compared to the UK 1990 growth standards[6]. As would be expected the
1947 cohort had markedly lower mean standard deviation scores (SDS) for
height and weight (see table).

However, mean body mass index (BMI) SDS were remarkably similar, although
there was an increase in the proportion of children above the 85th and
95th percentiles in the 1987-8 cohort. Surprisingly, the UK 1990 standard
fitted the distribution of the 1947 BMI values better than the 1987-8
cohort. This suggests that the UK 1990 upper centiles have in fact been
set too low. Thus we can confirm a 5% increase in the prevalence of
overweight, but would suggest that this has taken place over 40 rather
than 5-7 years. This is of importance for those individuals with BMI above
the 95th centile. In the 1947 cohort 39% of this 5% went on to be obese
(BMI>30 kg/m2) aged 50, a threefold increase in risk compared to their
slimmer peers. However, the public health importance of such an increase
is slight, since most fat adults were not fat children: 85% of those
obese at 50 had not been above the 95th percentile as children.

For the majority of children, however, BMI has changed little over forty
years, despite changing patterns of exercise and diet. Rona and Chinn
found a similar stability in weight for height over a 20 year period, but
did find an increase in triceps skinfold thickness [7]. Are today’s
children replacing muscle with fat ? The epidemic of childhood obesity
is much talked about, but objective evidence for it is surprisingly slim.
Indeed, so are the great majority of 9 year olds: 80% of the 1987-8 cohort
had BMI <18 kg/m2. In contrast, the epidemic in adulthood is
inescapable: by the age of fifty, two thirds of the 1947 cohort were
overweight (BMI >24 kg/m2) and 15% obese (BMI >30 kg/m2).
In childhood growth will be compromised if energy intake drops too low.

In the 1987-8 as in the 1947 cohort, significant social class gradients
for height were apparent, even after adjustment for birth-weight, and no
better explanation for these than undernutrition has yet been found. In
the absence of convincing evidence of a whole population shift in BMI, we
would suggest that it is inappropriate to advocate energy intake reduction
in childhood, where undernutrition is still prevalent, to prevent what is
still predominantly an adult problem. We hope to report shortly on further
analyses which explore the influence of obesity in childhood on adult
health.

Table:
Growth data at age 8-9 years from the 1947 and 1987-8 birth cohorts,
compared to UK 1990 standards

____________________________________________________________
                     1947        1987-8        P 1947 v
                                                1987-8  
____________________________________________________________

Height SDS          -0.85 (1.1)  -0.12 (1.0)   <_0.0001 mean="mean" sd="sd" _-0.50="_-0.50" _1.0="_1.0" _-0.02="_-0.02" _1.1="_1.1" _0.0001="_0.0001" weight="weight" sds="sds" _0.07="_0.07" _="_"/>85th BMI
percentile           11.0         18.8           0.02++
% >95th BMI
percentile            3.7          8.5           0.008++
____________________________________________________________

+ Standard normal test
++ Chi Squared

References

1. Reilly J, Dorosty A, Emmett P. Prevalence of overweight and obesity in
british children: cohort study. BMJ 1999;319:1039.

2. Taitz LS. Infantile overnutrition among artificially fed infants in the
Sheffield region. BMJ 1971;1:315-316.

3. Whitehead RG, Paul AA, Cole TJ. Diet and the growth of healthy infants
. J Hum Nut Diet 1989:273-84 .

4. Lamont D, Parker L, Cohen M, White M, Bennett S, Unwin N, et al. Early
life and later determinants of adult disease: a 50-year follow-up study of
the Newcastle Thousand Families cohort. Public Health 1998;112:85-93.

5. Wright C, Cheetham T. The strengths and limitations of parental heights
as a predictor of attained height. Arch Dis Child 1999;81(2):257-60.

6. Freeman JV, Cole TJ, Chinn S, Jones PRM, White EM, Preece MA. Cross
sectional stature and weight reference curves for the UK, 1990. Arch Dis
Child 1995;73:17-24.

7. Rona R, Chinn S. The national study of health and growth. Oxford:
Oxford University Press, 1999.