On the whole, public health research, and epidemiology more specifically, is largely devoid of its own theoretical developments and frameworks. In epidemiology, for example, hypotheses are tested by applying quantitative research methods and statistics. Little time is spent on debating the premise of the hypothesis or questioning the methods. Virtually no theoretical constructs are acknowledged.
Ok, so I may be exaggerating, but I’m not wrong …
The advantage to this lack of specific disciplinary anchors, is that public health research can be flexible and forge new approaches. Public health research is increasingly interdisciplinary these days. Sociologists, psychologists, statisticians, epidemiologists, biologists… and even *sharp intake of breath* medical doctors, actually do get together and produce some great research. The best public health research departments are places where people from different disciplines can safely interact. But does this mean that public health research is a Jack of all trades, trying out different approaches, dipping in and out of disciplines, but mastering none? I think not.
Health research does pick out the theoretical and methodological practices that most suit it from different sources – a magpie attracted to shiny ideas. The key is that researchers are brought together as experts from different disciplines, yet open to others. Often, it becomes obvious that the same concept is being developed under a different name in several disciplines.
My research tends to take a lifecourse approach to conceptualising health. The lifecourse approach to health is a nice example of a conceptual framework that has grown out of theoretical developments from many disciplines. It is a useful way to understand the mechanisms that may lead to the all-pervasive socioeconomic gradient in health.
Taking a lifecourse approach to health – where did it all start?
The early origins of health hypothesis
In the 1970s Anders Forsdahl showed associations between infant mortality and subsequent adult mortality from arteriosclerotic disease in Norway. He conducted analyses on the geographical differences in mortality rate within Norway. His results showed that the province of Finnmark, with the worst mortality rate, had experienced high infant mortality rate in the past.
Forsdahl’s findings and the concept of critical periods in early growth are reflected in the work of Barker et al on historical birth cohorts in the UK, and previous work from natural experiments. In their article, Barker and Osmond described strong correlations between infant mortality rates between 1921-1925 and adult mortality from ischaemic heart disease from 1968-1978 in the UK. They noted that while the increase in mortality from diseases such as ischaemic heart disease was associated with rising prosperity, these diseases were more common in individuals from poorer social groups. The authors hypothesised that “adverse influences in childhood associated with poor living standards increase susceptibility to other influences, associated with affluence, encountered in later life” (Barker and Osmond 1986, p.1080). They showed that high rates of infant mortality were strongly correlated with subsequent rates of coronary heart disease mortality in the same generation. They suggested that exposure to adverse conditions in utero and in infancy was associated with developing coronary heart disease in adulthood.
Much evidence stemming from animal studies supports the proposal that a biological event occurring during a critical period of animal development can permanently program the organism to take on specific behaviours, known as biological programming (Lucas 1998; Lucas, Fewtrell et al. 1999).
The lifecourse approach to health
The lifecourse approach to health is a conceptual model merging social science and epidemiological methods. In this framework, susceptibility to disease is an interaction between biological and social phenomena. Using a lifecourse approach encapsulates both the ‘objective’ measuring of health, wellbeing and the social environment, but also the subjective ideas about the experience of illness or poor social circumstances. Characteristically, individuals determine the trajectory of their lifecourse, but are passively subjected to external insults (Giele and Elder Jr 1998).
The lifecourse approach originated in the social sciences, where there was a primary interest in assessing the “social organisation of an individual’s passage through life” (Backett and Davison 1995). It was an approach used within a social constructionist theory explaining the way in which people age physically and they “seem to pass into and out of a series of age-based social categories and identities” (Hockey and James 2003, p.23). Though applying a lifecourse approach to epidemiology initially stemmed from the development of a biological hypothesis it quickly incorporated the concepts of social organisation and categorisation from sociology and anthropology.
By showing that the strong associations found between early life deprivation and later adult disease could be cancelled by statistical adjustment for adult socio-economic disadvantage, Davey Smith and Ben-Shlomo highlighted the importance of considering the interactions and cumulative effects of factors occurring along the lifecourse (Ben-Shlomo and Davey Smith 1991).
The lifecourse approach is based on ‘social and biological pathways’, and ‘social and biological chains of risk’ (Kuh, Power et al. 1997). These concepts were developed in Kuh and Ben-Shlomo’s (1997) book A life course approach to chronic disease epidemiology, which helped to establish lifecourse epidemiology as a bona fide theoretical and methodological approach in health research.
Lifecourse mechanisms and concepts
Three broad mechanisms are commonly outlined in the lifecourse approach which link exposures occurring at different points across the lifecourse to health and disease outcomes in later life: critical/ sensitive periods; accumulation; and trajectories (Blane 2001, p.65; Kuh and Ben-Shlomo 2004). They are delt with extensively in most textbooks and papers on the topic. Below, I describe the notions that I find most useful…
Time & timing
Time and timing are crucial factors in the lifecourse approach (Lynch and Davey Smith 2005). Obviously, time is also an essential component in the development of chronic diseases, which have long induction and latency periods. In the social construction of the lifecourse, there is a tendency to place “particular emphasis on the existence of a set of structurally defined but culturally experienced life stages” (Backett and Davison 1995, p.629). The timing of exposures, in other words, the stage along the lifecourse at which an exposure occurs, can be important in understanding its later effects (Lynch and Davey Smith 2005, p.2). Exposures, or accumulations of exposures may also occur between generations thereby extending the temporal plane of investigation (Davey Smith, Harding et al. 2000).
At always comes back to “Embodiment”
A helpful way of conceptualizing all these factors together is through the notion of embodiment or biological embedding (Hertzman 2012) “a concept referring to how we literally incorporate, biologically, the material and social world in which we live” (Krieger 2001). Embedding occurs when experiences alter human development and biology, the way in which this happens is influenced by systematic differences in social environments that endure over time and have the capacity to affect individuals over their lifecourse.
Backett, K. C. and C. Davison (1995). “Lifecourse and lifestyle: The social and cultural location of health behaviours.” Social science and medicine 40(5): 629-636.
Barker, D. J. P. and C. Osmond (1986). “Diet and coronary heart disease in England and Wales during and after the Second World War.” Journal of epidemiology and community health 40.
Barker, D. J. P. and C. Osmond (1986). “Infant mortality, childhood nutrition, and ischaemic heart disease in England and Wales.” The Lancet: 1077-1081.
Ben-Shlomo, Y. and G. Davey Smith (1991). “Deprivation in infancy or in adult life: which is more important for mortality risk?” Lancet 337: 530-534.
Blane, D. (2001). The life course, the social gradient and health. Social determinants of health. M. Marmot and R. G. Wilkinson. Oxford, Oxford University Press.
Davey Smith, G., S. Harding, et al. (2000). “Relation between infants’ birth weight and mothers’ mortality – prospective observational study.” British Medical Journal 320: 839-840.
Giele, J. Z. and G. H. Elder Jr, Eds. (1998). Methods of life course research: qualitative and quantitative approaches. London, Sage.
Hertzman, C. (2012). “Putting the concept of biological embedding in historical perspective.” Proceedings of the National Academy of Sciences of the United States of America 109: 17160-17167.
Hockey, J. and A. James (2003). Social identities across the life course. Basingstoke, Palgrave Macmillan.
Krieger, N. (2001). “A glossary for social epidemiology.” Journal of epidemiology and community health 55: 693-700.
Kuh, D. and Y. Ben-Shlomo (2004). Introduction. A life course approach to chronic disease epidemiology. D. Kuh and Y. Ben-Shlomo. Oxford, Oxford University Press: 3-14.
Kuh, D., C. Power, et al. (1997). Social pathways between childhood and adult health. A life course approach to chronic disease epidemiology. D. Kuh and Y. Ben-Shlomo. Oxford, Oxford University Press.
Lucas, A. (1998). “Programming by early nutrition: an experimental approach.” American society for nutritional sciences 128(Symposium – The effects of childhood diet on adult health and disease): 401s-406s.
Lucas, A., M. S. Fewtrell, et al. (1999). “Foetal origins of adult disease – the hypothesis revisited.” British medical journal 319: 245-249.
Lynch, J. and G. Davey Smith (2005). “A life course approach to chronic disease epidemiology.” Annual review of public health 26: 1-35.
Life gets under your skin edited by Prof Mel Bartley (University College London)