Article Text


How exactly do politics play a part in determining health? New perspectives on an age old issue
  1. C C Kelleher
  1. Department of Health Promotion, Clinical Sciences Institute, National University of Ireland, Galway, Costello Road, Shantalla, Galway City, Republic of Ireland
  1. Correspondence to:
 Professor C Kelleher;

Statistics from

Voting patterns, suicide, and mortality

This is the latest study to demonstrate an association between type of political pattern and health status, in this case suicide.1 Previous studies have also identified a more general relation between party political voting pattern and mortality.2–5 In this respect, the mechanism through which change in political regime might affect health is of considerable interest. Although the analysis in this paper was also undertaken at ecological rather than at individual level, it is persuasive that there is a temporal relation between the patterns of suicide and the changing governing party. What is particularly notable is the graduated effect seen, with rates highest when both Federal and National Governments were conservative. In the United Kingdom the effect was similar, in that the Liberal Democrat supporting constituencies, occupying an intermediate political position, had less strong relations with all cause standardised mortality ratio than for either Conservative or Labour voting constituencies.2 In the Republic of Ireland it was only in polarised political constituencies that an effect between left wing voting and health status was seen.5

In assessing predictors of suicide in particular, the authors focused on Durkheim’s hypothesis that levels of “anomie” reflect social disorganisation and that therefore this provides a psychosocial mechanism for explaining increased rates of suicide. A similar psychosocial interpretation has been extended more widely in recent years to explain in whole or in part the relativity of health status across social strata, as reflected in measures of income distribution and social inequity.6 Self rated health is a powerful proxy for morbidity and mortality and it can be seen to relate to participation in the voting process in the United States, independently of measures of income distribution.7 In particular the question of how patterns of social capital and social cohesion might have a psychosocial impact on both cause specific and overall mortality patterns has been examined in the inequality literature, with increasingly systematic attempts to conceptualise what these terms mean.8

The converse argument is based on a more structured materialist approach. Page et al1 took careful account of major confounding factors such as war, change in gross domestic product, extreme environmental factors like drought and sedative availability in this study and they also deal persuasively with the means by which variations in economic policy might have a direct effect on community amenities and services. The recent review by Lynch et al on the relation between disease specific outcomes and measures of income distribution across wealthy countries provides support for the view that direct material, rather than primarily psychosocial factors, are at play.9 In that analysis there was no consistent relation between income inequality and measures of mental health. The fact that only in the United States was homicide associated with income distribution suggests that means, motive, and opportunity all have to be considered in looking at specific outcomes and across differing social contexts. As Page et al point out1 it has long been known that rates of suicide are associated with general changes in economic conditions, reflected for instance in rates of unemployment in the population.10–12 There is a large literature on the effects of unemployment on health of both men and women, though this is usually felt to reflect a state of economic uncertainty, particularly for the most disadvantaged, which might have an association with already vulnerable individuals’ decision to take their own lives. Indeed, the literature on the prediction of individual suicide at clinical level is, at best, conflicting.13–16

It does therefore seem clear from the body of evidence in diverse sociopolitical environments, that party political affiliation, as much as participation in the political process, may well be a sensitive barometer to health status. This, taken with the evidence in the paper by Page et al,1 is suggestive that voting pattern could be seen as a sensitive proxy for social and political attitudes, as well as an indicator of the economic effects of policies. Further research should examine intermediate processes more closely to see how these ecological level associations might operate at the individual level. In meso-social settings like the work environment, such psychosocial processes are being explored in relation to health inequalities.6 There is also clearly a need for qualitative or ethnographic approaches to this question that would explore more fully issues of motivation and ideation as they relate to voting intention and participation in the political process and in turn how this is played out at interindividual, community and policy level.


Cecily Kelleher is principal investigator of the Health Research Board (Ireland) funded Research Unit on Health Status and Health Gain and a steering committee member of the European Science Foundation funded Social Variations in Health Expectancy in Europe programme.

Voting patterns, suicide, and mortality


View Abstract


  • Conflicts of interest: none.

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.