|Home | About | Journals | Submit | Contact Us | Français|
Studies have largely examined the association between political ideology and health at the aggregate/ecological level. Using individual-level data from 29 European countries, we investigated whether self-reports of political ideology and health are associated. In adjusted models, we found an inverse association between political ideology and self-rated poor health; for a unit increase in the political ideology scale (towards right) the odds ratio (OR) for reporting poor health decreased (OR 0.95, 95% confidence interval 0.94–0.96). Although political ideology per se is unlikely to have a causal link to health, it could be a marker for health-promoting latent attitudes, values and beliefs.
Studies on political ideology and health have largely been assessed at the aggregate or ecological level, and have mostly shown that average health tends to be better in areas where the political orientation of the majority is conservative.1–6 This finding has been interpreted as reflecting the differences in socioeconomic deprivation between areas voting for a party with conservative values and areas voting for a party with liberal values.2 Meanwhile, the emphasis in current studies has been on the contextual influence of political ideology on individual health.7,8 However, it is possible that individuals with conservative values may practice behaviours that could be health promoting. For instance, one's political ideology may reflect religiosity, civic participation or values of individual responsibility, which may have positive health impacts. It is important, therefore, to distinguish between aggregate and individual-level associations of health and ideology because they represent the potentially differing influence of the context and the individual.
It has been shown that at the individual level a relationship exists between political ideology and health status in the USA.9 Individuals who identified themselves as Republicans were 25% less likely to report poor health and 15% less likely to be smokers compared with those who identified themselves as Democrats. The better health status and lower smoking rates were not due to Republicans, on average, exhibiting a higher socioeconomic status (SES) than Democrats.9 In this analysis, we tested for this observed association between political ideology and health status at the individual level in Europe as we could not find a study that had already examined this relationship using disaggregated data.
We pooled and analysed the 2002, 2004 and 2006 European Social Survey (ESS) data, available for 29 European countries,10 which allow an examination of health status and political ideology at a micro-level. The ESS is considered to be an important cross-national data source with strong validity and reliability, with a mean response rate of over 60%. [For further details on the response rates and measurements, see ref. 10, and http://www.europeansocialsurvey.org/index.php?option=com_content&view=article&id=101&Itemid=139 accessed April 19, 2009]. The ESS also included Israel even though it is not typically regarded as part of ‘Europe’. Consequently, our pooled analysis excludes Israel, but we report the results observed for Israel in the text separately. Health status was ascertained from a question, ‘How is your health in general? Would you say it is very good, good, fair, bad or very bad?’ For analysis, we created a binary variable with bad or very bad health (hereafter referred to as poor health) = 1, 0 otherwise. Political ideology was based on a question, ‘In politics people sometimes talk of “left” and “right”. Where would you place yourself on this scale, where zero means the left and 10 means the right?’ In addition to using this linear scale of political ideology, we also grouped the scale into ‘left’ (21.9%) comprising of the first four categories, ‘right’ (24.7%) comprising of the last four categories and ‘middle’ (53.4%) comprising of the three middle categories. We included age, sex, race, years of full-time education, being in paid employment, net household income in the last year prior to the survey and survey year as covariates. After excluding missing values on the responses and independent variables, the final analytic sample consisted of 82 822 individuals for the pooled analysis, and sample size for Israel was 1580. We used binary logistic regression model procedures as implemented in SPSS v15.0 that took account of the survey design. Analyses were conducted both using pooled data (with country fixed effects) as well as separately for every country.
The study was reviewed by Harvard School of Public Health Institutional Review Board and was considered as exempt from full review as the study was based on an anonymous public use data set with no identifiable information on the survey participants.
In age/sex-adjusted models, there was an inverse association between political ideology and self-rated poor health such that for a unit increase in the political ideology scale (towards right) the odds ratios (ORs) for reporting poor health decreased [OR 0.95, 95% confidence interval (CI) 0.94–0.96]. Additional adjustments for an individual's SES did not alter the strength of the association (OR 0.96, 95% CI 0.94–0.97). Figure 1 shows the OR, adjusted for age/sex only and the full model, respectively, and 95% CI for categories of political ideology with the reference representing individuals who identified as ‘left’ in the pooled sample. In the pooled, fully adjusted model, the OR for reporting poor health among those who identified as ‘right’ was 0.73 (95% CI 0.67–0.79), while those who identified as ‘neither left nor right’ were also less likely to report poor health (OR 0.81, 95% CI 0.75–0.86) compared with those who identified as being left. The ORs were virtually identical in models with and without adjustment for individuals’ SES. We also examined the interaction between gender and political ideology and did not find it to be substantial nor statistically significant.
Country-specific models also yielded similar patterns (Supplementary Table 1) even though statistical associations in several countries did not attain conventional levels of statistical precision. In order to gain efficiency, we also tested whether the association between political ideology and poor health varied across countries using a random effects model, and we did not find statistical support for this hypothesis, suggesting that the lower log odds of reporting poor health among individuals who identify as being right are found across all countries.
In the Israeli sample, a positive association between political ideology and poor health was observed; those identifying as ‘right’ were more likely to be in poor health. The association, however, was not statistically significant (Supplementary Table 1).
These findings from Europe confirm the individual-level association between political ideology and health that was observed in the USA.9 Importantly, the usual explanation that these differences in political ideology are due to differences in SES (where individuals who identify with the ‘right’ have a higher SES),2 does not seem to be true. Age/sex-adjusted results and SES-adjusted results yield approximately the same magnitude of effects suggesting that SES may not be a major confounding factor to the association between political ideology and health.
It is possible that the causal link may go from health to political ideology; sick individuals may be influenced by leftist ideology as political parties with a leftist ideology typically tend to support policies that apparently favour disadvantaged populations. Arguably, this reverse association is more likely to occur between voting for a particular party and health, and perhaps less so between political ideologies (in general) and health. For instance, an individual who holds a conservative ideology may vote for a party with progressive ideology if s/he benefits from the latter party being in power. The lack of any ‘objective’ health measure is a limitation as there could be ‘reporting heterogeneity’ to health questions across different countries. However, when we correlated percentage in poor health and average life expectancy for the countries included in this study, we found a strong negative correlation (r = −0.82, P < 0.0001), suggesting a face validity to the self-rated health measure. Despite this, it is still possible that reporting heterogeneity in health could be a function of one's political ideology whereby individuals inclined towards conservative ideology are less likely to report poor health even if they are in poor health. We should note that residual socioeconomic confounding could still explain the association between political ideology and health, even though the effect of political ideology did not alter very much after including the SES variables measured in the ESS.
Political ideology per se is unlikely to have a causal link to health. Past research has shown that people's interpretations of the left–right spectrum used in this analysis incorporate older value perspectives (religious/secular and left-materialist/right-materialist) along with a new value perspective (Materialist/Post-Materialist) making it a comprehensive marker of overall ideology.11 In other words, left–right orientation is not just a marker of political party leaning. Rather, it encompasses general political ideology. Given this pluralist marker, we hypothesize that political ideology is a marker for several latent attitudes, values and beliefs (such as religiosity, social and civic participation, and individual responsibility) that might be health promoting. For example, a study of political ideology in Russia, Belarus and Ukraine from 2001 indicated that favouring restoration of Communism was related to higher odds of poor health and adverse health behaviours.12 Perhaps, those individuals who are anti-Communism are likely to favour pro-market democracy and attitudes that are typically representative of the ‘right’.
There is a clear need to systematically unpack the association between political ideology and health. If political ideology is a marker for aspects related to individual social behaviour, values and attitudes, then research is required to identify and describe these characteristics. Examining the simultaneous importance of political ideology at the individual level and political regimes and ideology at the contextual level is likely to provide a comprehensive assessment of the potential multilevel association between political ideology and health.
No direct funding was available for this study. SVS is supported by the National Institutes of Health Career Development Award (NHLBI 1 K25 HL081275).
Conflicts of interest: None declared.
Supplementary data are available at EURPUB online.