The role of methods in maintaining orthodox beliefs in health research
Introduction
Generally accepted views about correct ways of obtaining knowledge arise from socially constructed tenets and beliefs (Kuhn, 1962; Wolpe, 1994; Zito, 1983). Orthodox views about health and correct ways to conduct health research are maintained by traditions that are developed in support of dominant beliefs and by power relationships (Engel, 1977; Gillet, 1994; Lewontin, 1991). The impact of orthodox views on the research methods used to investigate phenomena, and, in turn, the role that methods assume in maintaining the orthodoxy, is less often considered. This paper will summarize developments shaping western orthodox views about human health that are relevant for considering their impact on research methods. A brief overview of the historical roots of the dominant biological paradigm provides background for understanding orthodox beliefs about how to gain knowledge about human health. It will be argued that orthodox methodology arising from the dominant paradigm has impeded innovation in health research. The paper will focus on quantitative research investigations dealing with population health issues.
Section snippets
Knowledge about human health: The dominant orthodoxy
As socially constructed fields of learning, dominant beliefs in a discipline are based on an underlying ideology. An ideology is an organized body of views that seek to monopolize ways of thinking and speaking about the world (Zito, 1983). The dominant tenets and beliefs about how health research should be conducted are derived from the biomedical model of human health (Engel, 1977; Gillet, 1994). The biomedical model functions like a traditional natural science, attempting “to isolate distinct
Documented weaknesses in the dominant orthodoxy
It is well known among biologists (Dubos, 1965; Lewontin, 1991; Lewontin, Rose, & Kamin, 1984; Sparling, 1983; Strohman, 1997), clinical practitioners and researchers (Engel, 1977; Gillet, 1994; Martin, Danner, & Holbrook, 1993; Miller, 1991), public health specialists (Davison, Macintyre, & Davey Smith, 1994; Susser & Susser (1985), Rose (1992); Susser & Susser, 1996), and epidemiologists (Duncan, 1988; Rothman, 1986; Vogt, 1992) that diseases seldom develop from single causes. Yet, as
Maintaining the orthodoxy
How can it be, in the face of extensive scientific and clinical evidence showing that diseases result from complex multicausal processes, that research on human health continues to focus so heavily on single factors that most often are physical or biological agents? The most easily identifiable explanation for the continued dominance of the orthodox biomedical paradigm is control over institutions and resources. In recent times, this dominance has been compounded by the power of pharmaceutical
Research methods in the biomedical model
The 19th century discoveries leading to a general acceptance of the germ theory of disease promoted a science of medicine to replace what was viewed as disorganized, unscientific, anecdotal medical practice with research and knowledge based on biological understanding (Duncan, 1988; Rosen, 1958; Susser & Susser, 1996). Scientific medicine developed parallel with, and was fundamentally affected by, the rapid raise in prestige and power of the positivistic science that dominated western thinking
Accepting the beliefs and models of the orthodoxy
As mentioned above, in research on human health, the biomedical experimental model and/or beliefs about the superiority of the model and the importance of emulating the model were transferred from laboratory research to clinical and population studies. Since experimental design is rarely possible and often unethical in clinical and population health research, the goal became to approximate it with quasi-experimental designs.
In the biomedical model, observational studies are considered a lower
Contemporary research challenges
The deterministic beliefs on which positivism builds are no longer compatible with contemporary science. Causal thinking has moved beyond beliefs about single causes that predict an outcome to the study of dynamic systems. It is now recognized that rules for determining when an event is necessary and sufficient for the occurrence of an outcome over-simplify causal processes. The outcome may have alternate causes, or the event predicting the outcome may occur simultaneously with a causal
Quantitative elaboration of complex relationships
Recognizing that the tools regularly used are not adequate for researching complex problems, Smith and Torrey (1996) emphasized that research for understanding the dynamic systems affecting individuals and societies requires theories of dynamic processes and data, and methods sufficient for testing the theories. They recognized that new methodologies are needed to study nonlinear, dynamic systems, and that quantitative and qualitative methods need to be more systematically integrated to advance
Limited options for research and training in alternatives to the orthodoxy
The problems arising from adopting the beliefs of the orthodoxy are intensified when methods take on a force of their own. The result is that essentially no options become available for many researchers to learn about newer methods even though they may be far more relevant and useful for their research. The occasional options that do arise for learning about the new methods may be taken over by the power of the status quo. Research then becomes driven by orthodox thinking and traditional
Beyond the barriers of orthodox beliefs
In the era dominated by the experimental model and quasi-experimental designs, statistical relationships found in research in this paradigm were inadvertently given a hardness and power that are not warranted. It is easily forgotten that the variables selected and how they are measured determine what is found, that simple changes in the values assigned to variable categories or the removal/addition of an uncertain statistical relationship can change the findings fundamentally. When variables
Summary
The tenets and beliefs of the biomedical model have seriously constrained the knowledge available for promoting and protecting human health. Risk factor research on an endless array of disease agents or possible treatments/interventions has inherent limitations. Testing to see if the statistical effects of some variable remain after controlling for the influence of other variables may seriously underestimate or overestimate the true risk of important causal influences. This means that
References (78)
Causes of change in the health of populationsA biopsychosocial viewpoint
Social Science & Medicine
(1996)Creating knowledge relevant for public health applications in immunology and aging
Mechanisms of aging and development
(1997)Linkage analysis using loglinear models
Computational Statistics and Data Analysis
(1992)- et al.
Understanding the mechanism of age—change of thymic function to promote T cell differentiation
Immunology Letters
(1994) Social psychology of the immune systemA conceptual framework and review of the literature
Social Science & Medicine
(1991)The dynamics of heresy in a profession
Social Science & Medicine
(1994)- et al.
Health psychologyWhy do some people get sick and some stay well
Annual Review of Psychology
(1994) Specification and estimation of latent variable models
Introduction to part threePositivism
- et al.
A comparison of observational studies and randomized controlled trials
New England Journal of Medicine
(2000)
Nutrition and the immune system
Proceedings of the Nutrition Society
Randomized, controlled trials, observational studies and the hierarchy of research designs
New England Journal of Medicine
Response models for mixed binary and quantitative variables
Biometrika
Multivariate dependenciesModels, analysis and interpretations
Markov fields and log linear interaction models for contingency tables
Annals of Statistics
The potential impact of predictive genetic testing for susceptibility to common chronic diseasesA review and proposed research agenda
Sociology of Health and Illness
Integrating theory and methods in population health research
Using theory to guide policy relevant health promotion research
Health Promotion International
Building sound foundationsMeasurement in health promotion research
New directions for healthTowards a knowledge base for public health action
Social Science & Medicine
Researching population healthNew directions
Detecting measurement confounding in epidemiological research: Construct validity in scaling risk behavioursBased on a population sample in Minnesota, USA
Journal of Epidemiology and Community Health
A comparative analysis of graphical interaction and logistical regression modellingSelf-care and coping with a chronic illness in later life
Biometrical Journal
Man adapting
Epidemiology
Introduction to graphical modelling
The need for a new medical modelA challenge for biomedicine
Science
Common themes in microbial pathogenicity
Microbiological Reviews
The immunology of exceptional individualsThe lesson of centenarians
Immunology Today
Beyond the orthodoxHeresy in medicine and social science
Social Science & Medicine
Causal diagrams for epidemiologic research
Epidemiology
Will genetics revolutionize medicine?
New England Journal of Medicine
Symptom reduction and suicide risk in patients treated with placebo in antidepressant clinical trials
Archives of General Psychiatry
Correspondence analysis of genes and tissue types and finding genetic links from microarray data
Genome Informatics
The structure of scientific revolutions
Graphical models for associations between variables, some of which are qualitative and some quantitative
Annals of Statistics
Changes in brain function in patients treated with placebos
American Journal of Psychiatry
Biology as ideology
Cited by (28)
Breast cancer delay: A grounded model of help-seeking behaviour
2011, Social Science and MedicineCitation Excerpt :As we found in our study, contextual factors are fundamental in these instances. It is surprising that the study of BCD has persisted under the dominant biomedical orthodox model (Dean, 2004) without being linked to social theory that could greatly enrich its understanding. The conventional patient-provider classification is actually a time-based taxonomy.
Introduction to: Heresy and orthodoxy in medical theory and research
2004, Social Science and MedicineRe-imagining the 'social' in the nutrition sciences
2012, Public Health NutritionWhy has the Opioid Crisis Remained Unchanged in Canada? The Limits of Bio-Scientific Based Policy Approaches
2024, Journal of Social Work Practice in the AddictionsCanadian government discourses on the overdose death crisis: limitations of a bio-evidenced approach
2022, Drugs, Habits and Social Policy