Department of Epidemiology & Biostatistics, University Medical Centre Nijmegen, The Netherlands.
Gerhard A Zielhuis, Department of Epidemiology & Biostatistics/252, University Medical Centre Nijmegen, PO Box 9101, NL-6500 HB Nijmegen, The Netherlands. E-mail: G.Zielhuis{at}mie.kun.nl
Although research into social and behavioural determinants of health and illness was conducted throughout the 20th century, it has become more common since the 1980s. Because of this, the term social epidemiology is increasingly encountered in the biomedical literature.
Recent commentaries by Kaufman and Cooper1,2 and Muntaner3 on causal explanations in social epidemiology, although of interest to the research field, take the term social epidemiology for granted. With reference to Syme4 and Susser et al.,5 the authors claim that social epidemiology is a distinctly recognised specialisation within epidemiology, and has been for some 30 to 40 years.
This field of research has been defined as the science which studies the link between social environment and the development and distribution of disease in populations,1,6,7 a definition which emphasizes a hybrid area of interest between sociology and epidemiology. Apparently, the term social epidemiology is so established that it is even used to describe the frequency of behaviour, without any reference to disease.8,9
Epidemiology has evolved into many different fields of application, leading to specialities such as occupational epidemiology, cancer epidemiology, and pharmaco-epidemiology. Can social epidemiology be regarded as just another branch? No, we believe the term is a misnomer.
Epidemiology is part of medical science and rests on a human-biological (scientific) background. Biomedical theory about, e.g. carcinogenesis, atherosclerosis, and teratogenesis is the input for scientific discussion and empirical studies in epidemiology. All epidemiological hypotheses are, or at least should be, derived from such theories by deductive reasoning. Similarly, results of epidemiological studies are used to adapt these biomedical theories by inductive reasoning. Therefore, every epidemiologist should have a basic training in biomedicine, comparable to the training that medical practitioners need as a basis for medical practice.
In broader terms: epidemiology, like every other discipline derived from medicine, rests on a human-biological (scientific) background.
Frequency research (the most succinct characterization of epidemiology) is also an important approach in other scientific fields, including psychology and sociology. Unlike epidemiology however, empirical studies to assess the (determinants of) frequency of psychological problems rely predominantly on socio-psychological theories. Should we call this psycho-epidemiology or social epidemiology? An argument in favour of using the term epidemiology outside the field of biomedicine is to stress the similarities in methodology: the statistical approach. But epidemiology is not just defined by its statistical outlook, nor is sociology or psychology. Epidemiology deals with medical knowledge itself. Similarly, sociology and psychology deal primarily with the subject-matter of behaviour (including behaviour related to health and disease) of societies and individuals.
An epidemiologist confronted with a research problem from the field of sociology or psychology is merely a technician who knows how to handle statistics with only a layman's ideas on the topics that are statistically described. Similarly, a sociologist has his common sense only to rely on when confronted with a medical problem. Unless scientists obtain a thorough training in both biomedicine and social sciences, they cannot contribute to scientific development in both areas. In fact, those who use the term social epidemiology acknowledge the failure of this field to identify underlying mechanisms that account for the relation between social environment and health outcomes.3,7
We therefore believe that epidemiologists, sociologists and psychologists should stick to their field of scientific inquiry. This does not imply that epidemiologists cannot use social or psychological determinants (income, stress) or outcomes (e.g. quality of life) in their studies. But even then, biomedical theory should link these items to the core parameters of a study (e.g. income as a determinant for nutritional status; stress as a determinant for hormonal imbalance; quality of life as a sequel of particular diseases). Similarly, sociologists and psychologists may enter medically defined variables (e.g. mortality, infertility) into their studies.
This is not just a statement of purity. We firmly believe that shopping in neighbouring scientific fields, without thorough subject-matter knowledge, will lead to statistical results without relevant meaning. As a consequence, research is reduced to statistically correct procedures without proper inference. Stretching borders between epidemiology as a biomedical discipline and sociology only leads to trivial statements, useless for society.
References
1 Kaufman JS, Cooper RS. Seeking causal explanations in social epidemiology. Am J Epidemiol 1999;150:11320.[Abstract]
2 Cooper RS, Kaufman RS. Is there an absence of theory in social epidemiology? The authors respond to Muntaner. Am J Epidemiol 1999;150:12728.[ISI]
3 Muntaner C. Invited commentary: social mechanisms, race and social epidemiology. Am J Epidemiol 1999;150:12126.[ISI][Medline]
4 Syme SL. Social determinants of disease. In: Last JM, Wallace RB, Gunther R (eds). Maxcy-Rosenan-Last Public Health and Preventive Medicine. 13th Edn. Norwalk, CT: Appleton and Lange, 1992.
5 Susser M, Watson W, Hopper K. Sociology in Medicine. 3rd Edn. New York, NY: Oxford University Press, 1985.
6 Faresjö T. Social environment and health. A social epidemiological frame of reference. Scand J Prim Health Care 1992;10:10510.[Medline]
7 Bloomberg L, Meyers J, Braverman MT. The importance of social interaction: a new perspective on social epidemiology, social risk factors, and health. Health Educ Q 1994;21(43):44763.[ISI][Medline]
8 Grund JP, Adriaans NF, Kaplan CD. Changing cocaine smoking rituals in the Dutch heroin addict population. Br J Addict 1991; 86:43948.[ISI][Medline]
9 Montagne M. The social epidemiology of international drug trafficking: comparison of sources of supply and distribution networks. Int J Addict 1990;25:55777.[ISI][Medline]