Determinants of the population health distribution: an illustration examining body mass index. (13th January 2020)

• Record Type: Journal Article
• Authors: Bann, David; Fitzsimons, Emla; Johnson, William
• Is Part Of: International journal of epidemiology. Volume 49:Number 3(2020)
• Page Start: 731
• Page End: 737
• Publication Date: 2020-01-13
• Subjects: Epidemiological methods -- distributions -- quantile regression -- body mass index -- obesity; Epidemiology -- Periodicals; 614.4
• DOI: 10.1093/ije/dyz245
• Languages: English
• ISSNs: 0300-5771

Abstract

Abstract: Most epidemiological studies examine how risk factors relate to average difference in outcomes (linear regression) or odds of a binary outcome (logistic regression); they do not explicitly examine whether risk factors are associated differentially across the distribution of the health outcome investigated. This paper documents a phenomenon found repeatedly in the minority of epidemiological studies which do this (via quantile regression): associations between a range of established risk factors and body mass index (BMI) are progressively stronger in the upper ends of the BMI distribution. In this paper, we document this finding and provide illustrative evidence of it in the 1958 British birth cohort study. Associations of low childhood socio-economic position, high maternal weight, low childhood general cognition and adult physical inactivity with higher BMI are larger at the upper end of the BMI distribution, on both absolute and relative scales. For example, effect estimates for socio-economic position and childhood cognition were around three times larger at the 90th compared with 10th quantile, while effect estimates for physical inactivity were increasingly larger from the 50th to 90th quantiles, yet null at lower quantiles. We provide potential explanations for these findings and discuss implications. Risk factors may have larger causal effects among those in worse health, and these effects may not be discovered when health is only examined in average terms. In such scenarios, population-based approaches to intervention may have larger benefits than anticipated when assuming equivalent benefit across the population. Further research is needed to understand why effect estimates differ across the BMI outcome distribution and to investigate whether differential effects exist for other physical and mental health outcomes.