Multilevel models that integrate associations between proximal and distant determinants of health have significantly improved our understanding of how health inequalities emerge. For example, it has been argued that the likelihood of being exposed to individual material, psychosocial and behavioural risk factors (micro-level) is strongly influenced by determinants at the macro-level (societal level) such as macroeconomic characteristics (income distribution, national wealth, and welfare), health care policy, or societal norms (CSDH 2008). Area-based measures of poverty and deprivation have also been found to be associated with health outcomes after adjustment for individual-level factors. The effects of these group-level variables on individual-level outcomes have been referred to as contextual effects (Sauzet and Leyland 2017).

However, with regard to socioeconomic inequalities in health, surprisingly, little is known about contextual factors located at the (intermediate) meso-level. In general, meso-level entities can be understood as smaller scale, lower level social arrangements or units with a different set of guidelines for societal organization. Key terms to define these arrangements are “institutions” or “institutional structures” which can be broadly described as systems of established and prevalent social rules that shape social interactions. Although the meso-level is explicitly mentioned in some of the integrative models to explain health inequalities, an explicit institutional approach including compositional and contextual characteristics of the major institutions of society has rarely been the focus of either theoretical reasoning or empirical research.

Coming of age is a good example to illustrate how institutional contexts produce and reproduce health inequalities. Girls and boys participate in different institutional contexts and develop through involvement in institutionalized forms of practice that are characterized by institution-specific communication, activities, environments, and learning objectives. These forms of practice initiate but also restrict young people’s activities and thereby become important conditions for their development (Silverstein and Giarrusso 2011). Family/home, kindergarten, preschool, primary and secondary school, higher education, vocational schools, and training, the workplace and local health care system are such important institutional contexts, as they are relevant in most children’s and adolescents’ lives (Blum et al. 2012). Although these different institutional contexts have varying importance in childhood, adolescence, and young adulthood, they do have a strong reciprocal effect on each other.

It is likely that characteristics of these institutions have an independent contextual effect on health above and beyond the individual level. Here, the meso-level generally includes group-level characteristics which can be separated into “compositional” factors (which people are found in an institution) and “contextual” ones (characteristics of an institution). For example, the school provides students with differential learning and developmental opportunities that are determined by group composition and schools’ contextual characteristics. Compositional features, for example, refer to student characteristics in classes and schools measured by aggregating information from students to the class or school level, such as the average level of the psychosocial learning environment or the proportion of students with specific socioeconomic background characteristics. Contextual factors, in contrast, focus on the shared organizational, cultural, social, and physical factors of the institution, meaning the built environment, quality of sanitary facilities, the number of teachers and their qualification, and even the availability of healthy food at school.

In terms of socioeconomic inequalities in health, characteristics of different institutional contexts were often measured only at the individual level, while the compositional and contextual characteristics of the respective setting itself were seldom studied. However, it is likely that health inequalities can either be aggravated or attenuated by institutional determinants. A healthy, positive institutional environment in a school can, for instance, help children from low social classes to compensate risks from other domains. In turn, negative compositional characteristics (e.g., low average SES or class climate) and contextual factors (e.g., low teaching quality, bad infrastructure, or working conditions) may add additional health risks to those already present. Thus, there is a striking lack of theoretical and empirical knowledge about the role of institutions, i.e., those societal entities which form a central link between the individual and the macro-level, in producing patterns of health inequalities in the different developmental stages from early childhood to early adulthood. This lack of evidence hampers the effective design of institutional interventions in the form of policies or preventive measures taken by the institutions themselves.

An institutional view is also important from a life course perspective (Kelly-Irving et al. 2015). Prior studies paid little attention to changes in the importance of different determinants of health inequalities—either individual or contextual features—as girls and boys grow older. Thus, a comprehensive life course approach from early childhood to early adulthood that takes the specific interrelations between the institutional domains into account is needed. Accordingly, research should combine explanatory factors and mechanisms at different levels and take the specific explanations at the different life stages and institutions into account. Interdisciplinary life course research with a strong emphasis on the meso-level of societal institutions can substantially contribute to a better understanding of health inequalities in early life.