|
|
|
|
|
|
|
-
Reader Opinion - |
|
|
Author:
Dr. Kiumarss Nasseri Modern Epidemiology was originally defined as “the study of the distribution and determinants of diseases in man” in the middle of the last century 1. Expansion of disease causation from single infectious agents to a complex mix of risk factors has significantly modified the nature and procedures of this science. Evaluation of over a hundred variant definitions between 1987 and 2017 reveals that despite some notable differences, the principles and approaches has practically remained focused on identification of the problem, search for causation and suggestion of probable remedial interventions 2. Notwithstanding the excellence of academic training, an important aspect of this discipline. i.e. the “art of epidemiology”, which is active advocacy for public health action is not effectively covered by the current training programs. Epidemiologists do the research, define the issue and suggest intervention methods, but are not sufficiently trained on the practical aspects of their profession, i.e., consultation and hands-on involvement with administrative decision making for public health actions. The significance of the works of forerunners like John Snow on Cholera 3, James Lind on Scurvy 4, and Takaki Kanehiro on Beriberi 5, is that their intervention suggestions was followed by immediate and effective remedial actions, something that currently is not possible with the multiple risk factor nature of chronic diseases and the expanded complex of multifactorial social and administrative interventions. To address this issue and following the highly successful Epidemiologic Intelligence Services (EIS) program at the Center for Disease Control and Prevention (CDC) in 1951 that offered community training for epidemiologists 6, few academic centers have developed programs to provide field training and active participation for health professionals on the global health issues 7, 8, 9. Effective control and prevention of diseases at community level is generally based on complex administrative decision making and actions is heavily based on epidemiological information, that is best provided and explained by epidemiologists present at the meetings. This kind of involvement requires a preparation that is currently not included in the classic training programs. It might be beneficial to develop a regular academic course on the “art of epidemiology”, explaining and providing examples and basic tools for effective involvement of epidemiologists in administrative decision making processes. Another point to consider is the suitability of the principles and methods of epidemiology to expand from public health to other issues of mass phenomena in the community. Currently a large number of research is published on various social events like domestic violence 10, marriage, family structure and fertility 11, 12, poverty and deprivation 13, commercial trends 14, and various polling results, that are generally focused on defining these phenomena and presenting statistical associations with some demographic and temporal variables which, can and occasionally has been misinterpreted as causal relations. A good example of this situation is Freakonomics, which was developed by economists to study the incentives and motivations behind people's choices 15, with detailed presentations of various statistical associations. It has been characterized as “… a science with excellent tools for gaining answers but a serious shortage of interesting questions.” 16, and is also criticized for attempting to imply causation from correlation 17. The main concern with these kinds of approaches is the possibility of making interventional decisions based on studies that lack the rigorous standards of analytical studies for determination of causation and inadequate control for issues that are generally identified as “bias” in epidemiology. With the rapid expansion and easy access to mega digital databases and development of limitless artificial intelligence (AI) generated algorithms, the possibility of first obtaining an answer and subsequently crafting a plausible question for it is on the rise, resulting in strengthening the misinterpretation of statistical significance as evidence of causation. Time may be right for combining the teachings of statistics in the contexts of population-based sciences with organized training in an introductory course of basic principles and methods of epidemiological and analytical study designs. This exposure may facilitate the development of robust research designs with less bias and better chance for detecting causation and identification of workable interventions for a wide range of mass phenomena in various communities. Like statistics that originated in mathematics and expanded to other fields of population-based studies, epidemiology that started in public health, can and should expand to other fields of population studies. While statistics determines the “distribution and association”, epidemiology looks for “causation” and joining them will provide a basic framework for finding the most probable reason(s) for common mass event and suggesting the best workable intervention to achieve the desired outcome, regardless of the nature of mass phenomena.
|
|
|
|
|
