The consistency of self-rated health in comparative perspective.

Self-rated health (SRH) is a widely used measure of individual health and has been shown to be a strong predictor of morbidity and mortality. However, surprisingly little attention has been paid to the consistency of SRH across populations. Studies that have addressed this issue explicitly have tended to focus on a few countries4,5 or onmultiple populations within a country.6 However, despite this narrow knowledge base, studies have started to be published that use SRH in a broadly comparative context without directly addressing the issue of international consistency. Such studies are likely to become more common as more and more broadly cross-national data become available. Consequently, this article explores patterns of SRH across 67 developed and developing counties using data from the World Values Survey to take a first glimpse at the comparability of SRH across populations. Individual-level data on SRH for use in this study were taken fromWaves 3 and 4 of theWorld Values Survey. SRH was not asked in Waves 1 and 2. The SRH instrument was given to a total of 130,287 respondents in 67 countries in Waves 3 and 4. Interviews for these waves were conducted between 1994 and 2004. In all countries, SRH response rates were over 98%. Unlike in many surveys using SRH, the response options in the World Values Survey are symmetrical, ranging from ‘very poor’ to ‘very good’; the additional response option ‘excellent’, often found elsewhere, is not offered. The global frequency of answers was: 1.1% very poor, 7.2% poor, 28.9% fair, 40.7% good, and 22.1% very good. Frequencies varied dramatically by country, as reported in Table 1. Also shown in Table 1 are the raw country means for SRH and the adjusted country means after a regression-based adjustment for age and gender based on the average impact of these two variables on SRH in the full global sample. The genderand age-adjusted country means are benchmarked to the SRH of a 40-year-old male.