Seeking causal explanations in social epidemiology.
(17/1739)
Social factors are associated with a wide variety of health outcomes. Social epidemiology has successfully used the traditional methods of surveillance and description to establish consistent relations between social factors and health status. Epidemiology as an etiologic science, however, has been largely ineffective in moving toward causal explanations for these observed patterns. Using the counterfactual approach to causal inference, the authors describe several fundamental problems that often arise when researchers seek to infer explanatory mechanisms from data on social factors. Contrasts that form standard causal effect estimates require implicit unobserved (counterfactual) quantities, because observational data provide only one exposure state for each individual. Although application of counterfactual arguments has successfully advanced etiologic understanding in other observational settings, the particular nature of social factors often leads to logical contradictions or misleading inferences when investigators fail to clearly articulate the counterfactual contrasts that are implied. For example, because social factors are often attributes of individuals and are components of structured social relations, random assignment is not plausible even as a hypothetical experiment, making typical epidemiologic contrasts inappropriate and the inference equivocal at best. Accordingly, more deliberate and creative approaches to causal inference in social epidemiology are required. Infectious disease epidemiology and systems analysis provide examples of approaches to causal inference that can be used when statistical mimicry of simple experimental designs is not tenable. In an era of increasing social inequality, valid approaches for the study of social factors and health are needed more urgently than ever. (+info)
Relation of probability of causation to relative risk and doubling dose: a methodologic error that has become a social problem.
(18/1739)
Epidemiologists, biostatisticians, and health physicists frequently serve as expert consultants to lawyers, courts, and administrators. One of the most common errors committed by experts is to equate, without qualification, the attributable fraction estimated from epidemiologic data to the probability of causation requested by courts and administrators. This error has become so pervasive that it has been incorporated into judicial precedents and legislation. This commentary provides a brief overview of the error and the context in which it arises. (+info)
Individual causal models and population system models in epidemiology.
(19/1739)
A group of individuals behaves as a population system when patterns of connections among individuals influence population health outcomes. Epidemiology usually treats populations as collections of independent individuals rather than as systems of interacting individuals. An appropriate theoretical structure, which includes the determinants of connections among individuals, is needed to develop a "population system epidemiology." Infection transmission models and sufficient-component cause models provide contrasting templates for the needed theoretical structure. Sufficient-component cause models focus on joint effects of multiple exposures in individuals. They handle time and interactions between individuals in the definition of variables and assume that populations are the sum of their individuals. Transmission models, in contrast, model interactions among individuals over time. Their nonlinear structure means that population risks are not simply the sum of individual risks. The theoretical base for "population system epidemiology" should integrate both approaches. It should model joint effects of multiple exposures in individuals as time related processes while incorporating the determinants and effects of interactions among individuals. Recent advances in G-estimation and discrete individual transmission model formulation provide opportunities for such integration. (+info)
The right answer for the wrong question: consequences of type III error for public health research.
(20/1739)
OBJECTIVES: This study examined the impact of assessing the causes of interindividual variation within a population when the research question of interest is about causes of differences between populations or time periods. This discrepancy between the research focus and the research question is referred to as a type III error, one that provides the right answer for the wrong question. METHODS: Homelessness, obesity, and infant mortality were used to illustrate different consequences of type III errors. These different consequences depend on the relationships between the causes of within- and between-group variation. CONCLUSIONS: The causes of inter-individual variation and the causes of variation between populations and time periods may be distinct. The problem of examining invariant causes deserves attention. (+info)
Health inequalities and social group differences: what should we measure?
(21/1739)
Both health inequalities and social group health differences are important aspects of measuring population health. Despite widespread recognition of their magnitude in many high- and low-income countries, there is considerable debate about the meaning and measurement of health inequalities, social group health differences and inequities. The lack of standard definitions, measurement strategies and indicators has and will continue to limit comparisons--between and within countries, and over time--of health inequalities, and perhaps more importantly comparative analyses of their determinants. Such comparative work, however, will be essential to find effective policies for governments to reduce health inequalities. This article addresses the question of whether we should be measuring health inequalities or social group health differences. To help clarify the strengths and weaknesses of these two approaches, we review some of the major arguments for and against each of them. (+info)
Invited commentary: propensity scores.
(22/1739)
The propensity score is the conditional probability of exposure to a treatment given observed covariates. In a cohort study, matching or stratifying treated and control subjects on a single variable, the propensity score, tends to balance all of the observed covariates; however, unlike random assignment of treatments, the propensity score may not also balance unobserved covariates. The authors review the uses and limitations of propensity scores and provide a brief outline of associated statistical theory. They also present a new result of using propensity scores in case-cohort studies. (+info)
The global burden of mental disorders.
(23/1739)
Recent data on the burden of mental disorders worldwide demonstrates a major public health problem that affects patients, society, and nations as a whole. Research must be done to find effective ways to deal with the increasing burden of mental disorders. Given the growing evidence that mental disorders are disorders of the brain and that they can be treated effectively with both psychosocial counseling and psychotropic medications, intervention packages could be developed to deal with the increasing burden. Such packages should be tested for real-world effectiveness and their cost-effectiveness should be demonstrated to guide policymakers to choose from among many other non-mental health interventions. The transportability and sustainability of intervention packages should be studied in public health research and a link between efficacy, effectiveness, cost-effectiveness, generalizability, and sustainability should be demonstrated. The World Health Organization's initiative on the World Mental Health 2000 Survey will provide the first basic epidemiologic data. Together with other data, the initiative will provide solid evidence for including mental disorders into essential treatment packages. In this way, parity can be achieved for mental disorders and mental health can be mainstreamed into health and public health practice. (+info)
Public conceptions of mental illness: labels, causes, dangerousness, and social distance.
(24/1739)
OBJECTIVES: The authors used nationwide survey data to characterize current public conceptions related to recognition of mental illness and perceived causes, dangerousness, and desired social distance. METHODS: Data were derived from a vignette experiment included in the 1996 General Social Survey. Respondents (n = 1444) were randomly assigned to 1 of 5 vignette conditions. Four vignettes described psychiatric disorders meeting diagnostic criteria, and the fifth depicted a "troubled person" with subclinical problems and worries. RESULTS: Results indicate that the majority of the public identifies schizophrenia (88%) and major depression (69%) as mental illnesses and that most report multicausal explanations combining stressful circumstances with biologic and genetic factors. Results also show, however, that smaller proportions associate alcohol (49%) or drug (44%) abuse with mental illness and that symptoms of mental illness remain strongly connected with public fears about potential violence and with a desire for limited social interaction. CONCLUSIONS: While there is reason for optimism in the public's recognition of mental illness and causal attributions, a strong stereotype of dangerousness and desire for social distance persist. These latter conceptions are likely to negatively affect people with mental illness. (+info)