Adult height and mortality in London: early life, socioeconomic confounding, or shrinkage?

Abstract

Study objective: To examine in detail the cause specific associations between height and mortality.

Design: A prospective cohort study with an 18 year mortality follow up.

Setting and participants: The Whitehall study of 18,403 men in the civil service in London examined between 1967 and 1969 aged 40-64 and followed up for mortality until the end of January 1987.

Main results: There was considerable variation in the strength of height-mortality association by cause. Respiratory disease showed the strongest inverse association, cardiovascular disease a moderate effect, and all neoplasms virtually no effect. Adjustment for age and civil service grade reduced the strength of these associations slightly, but had no impact on the heterogeneous pattern by cause (chi 2 3df p < 0.001). The height-mortality association declined with the length of follow up. By 15+ years, the only appreciable height affect was for respiratory disease mortality.

Conclusions: The attenuation of the height-mortality association with length of follow up might be explained by differential height reduction before entry that was greatest for people who were already ill, and hence at greatest risk of dying. The cause specific variation in the height-mortality association lends little support to the contention that impaired growth in childhood is a marker of general susceptibility to disease in adulthood.