INTRODUCTION

Two interesting debates in social epidemiology arose in the first decade of the present century between those who defended the idea that political traditions and/or welfare state models are related with the magnitude of socioeconomic inequalities in health status^1-5 and those who maintained that the facts do not support the existence of this relationship.^6-8

Espelt et al.¹ compared the inequalities in self-perceived health in nine European countries grouped into three political traditions: Social Democrats (Sweden, Denmark and Austria), Christian Democrats (The Netherlands, Germany, France and Italy) and late democracies (Portugal and Spain). The authors found greater inequalities in self-perceived health in the group of countries with late democracies. In a comment on this work, Lundberg noted that these findings were contrary to those observed in previous investigations. He also pointed out that grouping countries according to political traditions may be of little help if what we want to know is what specific aspects of the welfare state are important for health.⁶ In Lundberg's opinion, if countries with different forms of government are doing well in terms of population health outcomes despite implementation of different policies, we should consider whether there may be more than one path to achieving success in public health.^6-7 In responding to these criticisms, Espelt et al. defended the idea that governments affect health outcomes through specific policies, but that the type of government is the starting point.² In the opinion of these authors, public health activities include decisions about how to change society, and these decisions involve value judgments about what policies are most appropriate to improve population health.³

For their part, Sekine et al. evaluated whether the pattern of inequalities in physical and mental health differed among civil servants in Great Britain, Finland and Japan.⁴ These authors found lower inequalities in physical health in Finland than in Great Britain or Japan, a finding they attributed to different political traditions and/or welfare state models. In his commentary on this study, Bosma warned that the results obtained were less clear than the authors stated in their article.⁸According to Bosma, the Social Democratic regime may reduce income inequality and poverty, but the findings of this and other studies suggest that, like other regimes, it may not be able to reduce socioeconomic inequalities in health. Bosma considers that the field of research on health inequalities is too politicized, since investigators attribute success to certain measures for the reduction of health inequalities when their effectiveness has not been demonstrated. In their reply, Sekine et al. agreed that political ideology should be avoided in social epidemiology research, but claimed that the types of government influence the distribution of the social determinants of health.⁵

EXTRA-SCIENTIFIC VALUES IN INTERPRETATION OF THE FACTS

This debate is a reminder of others that have emerged in the field of epidemiology about how diseases are produced and the factors that determine their distribution in the population. Whereas for some authors it is impossible to separate ethical and political aspects and professional values in their scientific arguments,^9-11 others consider that epidemiologists and scientists should endeavor to distinguish between scientific and unscientific considerations.^12-14 For those in the first group, the importance of a scientific theory lies in its philosophical and ideological assumptions - such as social justice, human rights, etc. - about why some populations are sicker than others, whereas for the second group, the most important thing from the scientific perspective is whether a theory is valid, not the values it incorporates or the ethical uncertainties it generates.

Vandenbroucke notes that subjectivity is always present in the generation of scientific knowledge due to the different interpretations of theory and facts. And that only the future will tell which side is right.¹⁵ These comments are extraordinarily important for the understanding of scientific practice. Because the presence of subjectivity in science does not contradict scientists' attitude about seeking unequivocal evidence. This is a self-regulating mechanism that eventually corrects subjective judgments in scientific interpretation.¹⁶ Scientists seek objectivity through a continuous process that involves constant testing of the relationships among values, assumptions, hypotheses and empirical work. And scientific practice shows that, faced with indisputable facts, there comes a time when scientists must abandon their values, desires, prejudices and vanities. It is the difficult moment at which they must accept the facts that reveal the deficiencies of their theoretical framework and recognize the contradictory findings that challenge their preconceptions.

For this reason, science remains the ideal procedure against dogma: science does not pretend to provide definitive results. The history of science shows how old theories that were once widely accepted are rejected as inadequate and replaced by new ones. However, these rectifications are not easy due to different interpretation biases on the part of scientists, such as confirmation bias: that is, the tendency to give more support to what confirms one's own convictions and to ignore or discredit what contradicts them. The rectifications will take even longer when empirical, political and ethical considerations live in harmony within the paradigm that, in any line of investigation, predominates in the scientific community at a given moment. One need only look at the theories on the production of socioeconomic inequalities in health¹⁷ or the various reports on the way to address them, such as the report of the Commission on Social Determinants of Health of the World Health Organization,¹⁸ to see the absence of any loud voices that break this harmonious whole.

THE DIFFICULT ALLIANCE BETWEEN SCIENCE, POLITICS AND ETHICS

According to theories on the origin of health inequalities, the socioeconomic distribution of health outcomes is produced by unfair practices and policies that reward certain groups, economically and socially, at the expense of others; therefore, socioeconomic inequalities in health can be reduced by social and economic interventions.¹⁹ For example, the fact that reduction of income inequality has not been a priority issue has been considered one of the possible causes of the failure of the strategy carried out in England to reduce socioeconomic inequalities in health.²⁰ However, to date no one has been able to show that reducing income inequality leads to lower socioeconomic inequalities in health. Several investigations that have analyzed mortality, one of the classic indicators of population health status, have observed increases in inequalities in mortality in both periods of increased²¹ and decreased²² income inequality, as well as in periods of stability.²³ An absence of changes in the magnitude of socioeconomic inequality in mortality has also been observed in periods of reduced income.²⁴ For these reasons, when prominent members of the scientific community in this field of knowledge claim that health inequalities are avoidable through interventions that reduce inequalities in income distribution, they are not referring to a consistent set of facts, but to consistency with a fictional argument that must be superimposed on the facts to avoid seeing them. According to the distinction that Aristotle made between history and literature, these fictional arguments do not represent what has actually happened but rather what might have happened. It is the way to maintain the balance of the narrative discourse between science, politics and ethics, the way to preserve the previously mentioned idea about the origin of the socioeconomic distribution of health.

Indeed, this belief ignores the strident cries that arise when observing the findings of multiple investigations conducted over the last two decades, many of which have been referred to by Lundbert and Bosma.^6-8 The results of these investigations revealed that the Nordic countries, with a long history of applying universal welfare policies, were not among those with the lowest socioeconomic inequalities in mortality in the 1980s, whereas Finland, Norway and Denmark were among those with the highest such inequalities in the 1990s.^25-28 Another relevant finding was that southern European countries like Italy and Spain, with large socioeconomic inequalities, were among those countries with lowest inequalities in mortality in the 1980s, and had the lowest mortality inequalities in the 1990s. The results of studies with data on the first decade of the 21st century have reproduced the previous findings.^29-30 Likewise, in 1990 and 2000, the Nordic countries were among those with the highest socioeconomic inequalities in negative self-perceived health, with the exception of Sweden.^{(25,27, 31-32)} Similarly, in both periods Italy and Spain, together with Germany, were among the countries with the smallest socioeconomic inequalities in negative self-perceived health.^31-32

In most of the aforementioned studies, the objective was to determine the international pattern in the magnitude of socioeconomic inequalities in health in various European countries. And, in the first decade of the 21st century, other investigations were conducted specifically to test the hypothesis that health inequalities were probably lower in Welfare State countries that provided universal services to the population.^33-37 Most of these studies did not find a relationship between the size of the Welfare State and the magnitude of health inequalities.

In 2004 Bartley noted that the first results of the mentioned international comparative studies caused a considerable uproar in the scientific community and in public health professionals.³⁸ It was generally believed that health inequalities would be lower in the Nordic countries with small differences in income between the rich and poor; surprisingly, it was found that health inequalities were lower in countries in southern Europe with high socioeconomic inequalities. Bartley also noted that it was very difficult to establish the policy implications of research on health inequalities based on the results of the investigations conducted. In fact, although different explanations have been proposed for this paradox, the reasons are unknown.³⁹ This does not mean that health inequalities cannot be avoided through welfare policies but that, for now, a good dose of modesty is needed to recognize that it has not been possible to show the relationship between welfare policies and the magnitude of health inequalities.

FALSE SCIENTISTS

Many scientists act as advisors or as authors of reports and action proposals whose objective is to recommend concrete actions to reduce socioeconomic inequalities in health. Commitment to a preconceived truth is common when assuming the role of advisor or author in the proposal of specific actions; thus, when investigators take on such roles, they legitimately base their proposals for action on those facts, beliefs, values, ideologies and interests that they consider to be appropriate. But scientists who aim to advance scientific knowledge in this regard, the same as in any other field of science, should be characterized by just the opposite: by non commitment, by independence from political programs, ideologies or utopias.

There are many political and ethical reasons to recommend reduction of income inequalities and implementation of many other universal welfare policies. But scientists' commitment to ideological or political agendas in explaining socioeconomic inequalities in health should be verified or refuted by tests of the empirical evidence. And in this regard they must assume the possible existence of results contrary to their values. The challenge consists of finding explanations for these findings. Because the scientist who commits to serving a preconceived truth, rather than seeking the explanation hidden behind the facts, renounces the very mission of scientific investigation, which is the search for truth, regardless of the result. This is the opposite of investigators who look for evidence to confirm what they believe is true.⁴⁰ And it matters little whether the preconceived truth is called democracy, welfare policies, human rights or social justice. The scientist who serves a truth other than the one to be discovered is a false scientist.