Birth cohorts and obesity: What do they tell us

Every so many years, national data are gathered on dietary habits and such surveys also provide data on many non-dietary variables, including BMI.  The data thus gathered on weight and height refer only to the date of measurement. It cannot tell us when an individual might have become overweight or obese, given that the sample will be comprised of individuals ranging in age from tens to the nineties. Thus, if a sample is taken in 1960, the extent to which subjects watched TV would vary very considerably. However, if a birth cohort born in 1960 is followed thereafter, then they all experience the same opportunity for TV viewing. 

A recent paper in the Proceedings of the National Academy of Sciences, provides data on birth cohorts and obesity in the US(1). They used data from NIH longitudinal studies: The National Longitudinal Study of Adolescent to Adult Health (Add Health) (2); The Americans' Changing Lives (ACL) survey (3); Mid Life in the United States (4); Health and Retirement Study (5). The main aim of the study was to determine for different birth cohorts, the age at which the mean body mass index (BMI) reached 30, indicative of obesity.  The data are presented in the table below:

Birth cohort                             Age in years at

                                             which mean cohort BMI

                                        is equal or greater than 30 kg/m2

1980-1984                              30-39

1965-1969                              40-49

1955-1959                              50-59

1950-1954                              60-69

These data clearly indicate younger cohorts will endure the adverse effects of obesity for decades longer than their older or siblings or their parents. Cohorts born between 1905 and 1945 never reached a mean BMI indicative of obesity throughout their lives. However, all readily hit the overweight range (BMI 25.0-29.9 kg/m2) from their thirties onwards and some of the cohorts were knocking on the door of obesity. For example, the cohort born between 1940-1945 reached a mean BMI of 29.4 in their seventies. 

The data we obtain from birth cohorts are critically important in understanding the epidemiology of obesity. However, many take the uncomplicated but incorrect view that both overweight and obesity are entirely a recent phenomenon. In my book “Ever seen a fat fox – the biology of human obesity” I cover several earlier birth cohort studies and BMI (6) .

Researchers at the US National Bureau of Economic Research examined the birth cohort changes in BMI using data collected by the National Center for Health Statistics covering subjects born between 1882 and 1986 (7). Their results show that in 1880, average BMI was within the normal weight range (20-25 kg/m2) at about 22. Mean BMI of black females reached 25 by 1900, while for white males, black males and white females the dates that a mean BMI of 25 was achieved was respectively, 1905, 1915 and 1920. Again, black females led the way in achieving a BMI in excess of 30 by the year 1940 with all others achieving this about 1960. Another set of research from the same US Bureau provides data on the BMI of US military recruits from 1864 to 1991(8). These are more robust data because they are based not on models but on measured weight and height. If we take men aged 45 years, the average BMI increased from 23.3 to 26.5 over this time period but the greatest part of this increase (60%) was in the period 1894 to 1961. Only 20% of the rise in BMI was explained post 1961. 

We can also look at more modern data gathered in Denmark and which looks at the rate of change in body weight in two groups (9). The first are 19-year-old military recruits studied since 1945. The second is a set of children aged 7,8,9,10 and 11, all born in 1930. In both cases, the prevalence of obesity was low at the outset of the studies at about 3 obese persons per thousand of the population. That trebled by 1960 but remarkably, it did not change at all between 1960 and 1990. However thereafter obesity again trebled to a level of about 30-40 obese persons per 1000 of the population. This wave-like growth in obesity, which I call the tsunami of lard, has also been recorded by the researchers at the US Bureau of Economic Research in their 1882-1991 modeling of the growth of obesity in the US. The concept that the rise in obesity has not been linear but wave-like is important in evaluating the present crisis of obesity. As ever in science, progress in our knowledge grows from embracing anomalies and exceptions and not from burying awkward data. 

(1) Yang YC, Walsh CE, Johnson MP, Belsky DW, Reason M, Curran P, Aiello AE, Chanti-Ketterl M, Harris KM. Life-course trajectories of body mass index from adolescence to old age: Racial and educational disparities. Proc Natl Acad Sci U S A. 2021 Apr 27;118(17): e2020167118. doi: 10.1073/pnas.2020167118. PMID: 33875595.

(2) https://addhealth.cpc.unc.edu/ 

(3) https://www.icpsr.umich.edu/web/NACDA/studies/4690 

(4) http://midus.wisc.edu/puboverview.php 

(5) https://hrs.isr.umich.edu/about

(6) https://www.amazon.com/Ever-Seen-Fat-Fox-Explored/dp/1910820083 

(7) Komlos J, Brabec M (2010) The trend of mean BMI values of US adults, birth cohorts 1882-1986 indicates that the obesity epidemi

began earlier than hitherto thought. National Bureau of Economic Research. Available from:  http://www.nber.org/papers/w15862 .

(8) Costa D, Steckel RH. Long term trends in health. Welfare and economic growth in the United States.  In RH Steckle and R,Floud eds Health and Welfare during Industrialization, Chicago: university of Chicago Press, 1997.

(9) Olsen LW, Baker JL. Holst C, Sørensen TI. Birth cohort effect on the obesity epidemic in Denmark Epid 2006;17:292-295.