DECISION ANALYSIS

Decision Analysis

Decision Analysis

Introduction

It is often believed that quantitative methods are insufficient to explore fully the qualitative elements of important decisions, particularly when one is concerned with such ethical considerations as individual rights, interests of multiple stakeholders, and non-financial societal concerns. Indeed, in their famous book Decisions with Multiple Objectives: Preferences and Value Tradeoffs, Keeney and Raiffa write,

It is almost a categorical truism that decision problems in the public domain are very complex. They almost universally involve multiple conflicting objectives, nebulous types of non-repeatable uncertainties, costs and benefits accruing to various individuals, businesses, groups, and other organizations - some of these being non-identifiable at the time of the decision - and effects that linger over time and reverberate throughout the whole societal superstructure. (Keeney and Raiffa, 1976: 12)

The fundamental objections to formal quantitative methods are (1) all models, whether qualitative or quantitative, necessarily abstract away some of the richness of particular situations, and (2) complex problems require subjective evaluations, and it is exactly these subjective evaluations that are often missed by the analysis. While it is tautologically true that “bad” models leave much to be desired, the trouble with formal analysis is not that subjective evaluations cannot be incorporated, but that too often, too few decision-makers are willing to formalize their personal preferences and subjective assessments.

Decision Analysis in the field of Medicine

Until the 1960s, more than 50% of all medical doctors in the United States were family physicians, pediatricians, or general internists.¹ Today, about one-third of all U.S. physicians practice primary care medicine (internal medicine, pediatrics, family practice).² Although opinions vary on the optimal ratio of primary care to specialty physicians, in the mid-1990s, the consensus among leaders in medicine was that more primary care physicians would be needed to ensure adequate health care[sup3-6] and that the future output of graduates should include at least 50% primary care physicians.[sup2,7] Medical schools responded with a renewed emphasis on primary care. The Robert Wood Johnson Generalist Physician Foundation supported the effort.[sup8] As schools receiving Generalist Initiative funds began their work in 1994, their admission committees struggled to find objective predictors to assess the likelihood of any particular candidate choosing a primary care medical career.

The present study examines career decision making in primary care medicine. Studies related to student characteristics are reviewed for residency choice. An InfoTrac One File and MEDLINE (1966-September 2002) searches of the literature related to student characteristics were conducted. We excluded studies that contained samples of medical students outside the United States, failed to address instrument response bias, were limited to student characteristics that influence rural practice choice, focused on a specific nonprimary care specialty choice (e.g., anesthesiology, surgery), were limited to only one specialty within primary care, or did not include internal medicine, pediatrics, and family medicine in their definition of primary care.

First, we analyze studies that considered predictor variables individually using descriptive and univariate statistics. Next, we explore studies that used multivariate statistics to generate predictive models. Then we identify knowledge gaps regarding primary care specialty choice, discussing ...