Cultural differences in self-rated health in ethnic groups in the Netherlands?

Background: Self-rated health is a good predictor of health, morbidity and use of care. Self-rated health is also a responsive measure to changes in health. It can be measured by one simple question. As such it is included in many surveys in many countries, making measuring of general health comparable between countries. But health measures may be different between different cultures and few studies studied differences in self-rated health between different cultures or interactions in self-rated health between some determinants and cultures. Aim: In a first step towards analysis of validity of self-rated health between different ethnic cultures in the Netherlands, self-rated health and the association with several determinants is compared between native Dutch and several major migrant populations in the Netherlands. Methods: In the Second Dutch National Survey of General Practice self-rated health was measured in 271.329 patients of a national sample of General Practitioners. 6.8% were of non-western origin. Self-rated health was measured by one question on a five-point scale rating from very good to very poor. Determinants as sex, age, education, employment, civil status, and social status were also registered. Association between determinants and ill health, and possible interactions is studied in different migrant groups in the Netherlands and Dutch natives by logistic regression. Results: Fair to very poor health was rated by 35% of Turkish migrants, 29% of Moroccan, 24% of Caribbean, 21% of Surinamese, 21% by other migrants non-Western and Western and 17% by native Dutch. All determinants included in the model predicted significantly fair to very poor health: age, sex, civil status, education, working status and socio-economic status. Corrected for all determinants odds ratios of fair to poor health compared to Dutch natives were 3.8 (95% CI: 3.3-4.5) and 3.7 (3.1-4.4) for Turkish and Moroccan migrants respectively, 2.2 (1.7-2.8) and 1.95 (1.7-2.2) for Caribbean and Surinamese respectively. Other non-Western migrants OR was 2.2 (2.0-2.5) and Western migrants 1.3 (1.2-1.4). Interactions existed between ethnic groups and sex, civil status and working status. Conclusions: Differences in self-rated health still exist between ethnic groups even when corrected for background and socio-demographic variables. Besides differences in morbidity, cultural differences in rating and expressing health may exist. Self-rated health as a measure for morbidity should be cross-cultural validated. (aut. ref.