Abstract
This study explores the potential of health care service areas (HCSAs) developed from claims data as the geopolitical units for the assessment of health professional needs in rural areas. HCSAs reflect consumer shopping patterns for health care and alter traditional health professional needs estimates.
Assessing the adequacy of the supply of health care providers in rural areas has been a continuing problem for managers, administrators and policymakers. Despite a myriad of public programs aimed at increasing the number of providers in rural areas-including loan forgiveness, the establishment of rural medical schools, and mentoring programs (among many others)-the number of providers has continued to drop. Although this trend of health providers leaving rural areas is well established, the cause of the trend is widely debated. At the extreme, some argue that providers leave because of a lack of investment by state governments and health plans, while others argue that the lack of rural providers simply reflects market forces and consumer preferences.
While the debate over providers in rural areas has deep political overtones, understanding the true demand for health care providers in rural areas is a key issue facing health planners and managers in rural areas. For example, medical schools need to forecast the demand for physicians, both number and specialty, far into the future because it takes so many years to complete the training of a physician. For clinic managers in rural areas, it is important to understand the potential demand for more practitioners and to understand the impact of providers of different training. If a rural clinic decided to add a new physician, would it be better to hire one trained in family practice or a pediatrician? Also, managed health plans marketing their program to purchasers in rural areas need data to determine whether their existing clinic network is adequate to provide care to new members.
A wide variety of methodologies have been developed to estimate the supply of physicians, nurses, and other health care professionals and the need for their services. All have proven to have serious deficiencies, and estimates of provider shortages or surpluses vary widely depending on which method is used. The major methodological problems center on the definition of the relevant service areas for the analysis, the quantification of need for health services for the populations in those areas, and the estimation of the number and types of providers required to meet those needs. Of these, the most pressing problem, particularly in rural areas, is the definition of service areas that can be used as a basis for all the other analyses.
This article presents a market, or trade area, approach to the development of primary health care service areas (HCSAs) in rural areas. After developing HCSAs, the adequacy of the supply of primary care (PC) providers in rural Minnesota is then assessed. Finally, this article examines the relationship between physician supply and training and the outflow of individuals seeking primary health care from other service areas. Although we developed our model using data from rural Minnesota, the methods are transferable to any state.
The use of HCSAs to assess the supply and demand for health professionals is promising for several reasons. First, it creates a health care service area that is determined by actual consumer behavior rather than by artificial boundaries, such as counties. Second, the HCSAs provide a unit of analysis that can be followed over time to assess the effects of population, policy, and market area changes on the supply of health care practitioners. HCSAs also enable researchers to analyze the factors attracting consumers to specific communities for health care and can help physicians and clinics decide where to locate and how to staff. Finally, because the data necessary to develop these service areas are readily available, the development and assessment of health care service areas will enable managers, researchers, and policymakers to develop accurate models of individual behavior and design policies to reflect this enhanced knowledge.