Background: The first Whitehall Study showed an inverse social gradient in mortality from coronary heart disease (CHD) among British civil servants--namely, that there were higher rates in men of lower employment grade. About a quarter of this gradient could be attributed to coronary risk factors. We analysed 5-year CHD incidence rates from the Whitehall II study to assess the contribution to the social gradient of psychosocial work environment, social support, coronary risk factors, and physical height.
Methods: Data were collected in the first three phases of examination of men and women in the Whitehall II study. 7372 people were contacted on all three occasions. Mean length of follow-up was 5.3 years. Characteristics from the baseline, phase 1, questionnaire, and examination were related to newly reported CHD in people without CHD at baseline. Three self-reported CHD outcomes were examined: angina and chest pain from the Rose questionnaire, and doctor-diagnosed ischaemia. The contribution of different factors to the socioeconomic differences in incident CHD was assessed by adjustment of odds ratios.
Findings: Compared with men in the highest grade (administrators), men in the lowest grade (clerical and office-support staff) had an age-adjusted odds ratio of developing any new CHD of 1.50. The largest difference was for doctor-diagnosed ischaemia (odds ratio for the lowest compared with the highest grade 2.27). For women, the odds ratio in the lowest grade was 1.47 for any CHD. Of factors examined, the largest contribution to the socioeconomic gradient in CHD frequency was from low control at work. Height and standard coronary risk factors made smaller contributions. Adjustment for all these factors reduced the odds ratios for newly reported CHD in the lowest grade from 1.5 to 0.95 in men, and from 1.47 to 1.07 in women.
Interpretation: Much of the inverse social gradient in CHD incidence can be attributed to differences in psychosocial work environment. Additional contributions were made by coronary risk factors--mainly smoking--and from factors that act early in life, as represented by physical height.