Background: Mortality rates from hypertension related diseases such as coronary heart disease, hypertensive heart disease, stroke and end stage renal disease show an inverse association with socio-economic status (SES).
Objectives: To review the published literature in order to assess whether (i) there is an association between SES and blood pressure (BP), and if so whether this is explained by (ii) SES differences in treatment rates, or (iii) SES differences in established risk factors for hypertension, or (iv) psycho-social factors associated with SES.
Method: A narrative systematic review of published articles identified from a MEDLINE search from 1966-1996 and manual searching of the retrieved articles' bibliographies.
Results: Lower SES was associated with higher mean BPs in almost all studies in developed countries. This inverse gradient was both stronger and more consistently found in women than in men. The magnitude of the association varied but generally was quite small, with age adjusted mean systolic BP differences of about 2-3 mm Hg between the highest and lowest SES groups. The finding of an SES gradient in BP, despite adjusting for treatment in some studies and the lack of consistent SES differences in hypertension treatment rates, makes differential treatment an unlikely explanation for the SES gradient in BP. A substantial part of the SES gradient was accounted for by the SES gradient in body mass index. Alcohol consumption across SES groups accounted for part of the association in men though few studies examined this issue specifically. In contrast, in undeveloped or developing countries a direct association between SES and BP has often been found which may reflect a higher prevalence of obesity, and higher salt and alcohol intakes among those of higher SES. The SES differences in BP were not detectable in most studies in children. There is little evidence that adverse psycho-social factors associated with low SES cause chronic elevations in BP.
Conclusion: A major challenge in reducing the SES gradient in BP is to understand and prevent the SES differences in obesity, which are particularly large in women. Future research should be directed to this question.