The Curious Economist  |   August 2020
The Demand for Medical Care
Article Information
Practice Management / The Curious Economist
The Curious Economist   |   August 2020
The Demand for Medical Care
ASA Monitor 8 2020, Vol.84, 22-23.
ASA Monitor 8 2020, Vol.84, 22-23.
Thomas R. Miller, PhD, MBA, ASA Director of Analytics and Research Services, and Director of the ASA Center for Anesthesia Workforce Studies (CAWS).
Thomas R. Miller, PhD, MBA, ASA Director of Analytics and Research Services, and Director of the ASA Center for Anesthesia Workforce Studies (CAWS).
Thomas R. Miller, PhD, MBA, ASA Director of Analytics and Research Services, and Director of the ASA Center for Anesthesia Workforce Studies (CAWS).
“Recent years have witnessed a sharp upsurge of interest in the economics of health”(N Engl J Med 1968;279:190-5). In the spirit of transparency, “recent years” refers to the period around the start of Medicare. This statement was made by Victor Fuchs, Professor of Economics, at Stanford University in 1968 as he examined the growing demand for medical care. Textbooks on health economics typically begin with a discussion on the demand for health care, and despite different perspectives, economists and policymakers share a strong interest in understanding the factors behind the demand for health and medical care.
In this first issue of “The Curious Economist.”, I discuss the demand for medical care based on early and foundational health economics literature. I end with a brief discussion of the demand for anesthesia services.
Defining demand
In economics, demand refers to the willingness and ability to pay. It is the relationship between price and quantity purchased (demanded). Demand is different from, and may not represent, “need,” “want,” or “desire,” despite these terms often being used interchangeably (N Engl J Med 1968;279:190-5).
The demand for medical services is different from the demand for other products and services such as clothes or food. It is irregular and unpredictable, a function of illness or injury. Additionally, there is substantial imbalance of material knowledge between the patients using the services, the clinicians who provide the services, and the insurers that pay for the services (Am Econ Rev 1963;53:941-73). Similar to other products, however, medical care behaves as a “normal” good in economic terms; there is an increase in the demand for medical care when consumer income rises.
There is another important difference between medical care and a typical consumer product. There is likely no direct utility or pleasure from going to the physician's office or undergoing a procedure. Therefore, the demand for medical care is referred to as a derived demand reflecting individuals' demand for health and well-being. Economists treated health as some combination of a consumption good, an outcome of interest, an explanatory variable, and an investment in human capital.
The human capital model of the demand for health
Grossman introduced the human capital model of the demand for health in 1972 in a published article from his dissertation (J Political Econ 1972;80:223-55). The central proposition of the model is that “good health” can be viewed as a durable capital stock that produces an output of healthy time. Individuals inherit an initial stock of health.
Grossman views “aging,” and the associated loss of biological reserve, as depreciation of our “health capital” over our life span. With this perspective, he brought tools of economic analysis to understanding the demand for health and to examining how much people are willing to pay to have good health. If you are old (in Grossman's terms, your “health capital has depreciated”), then the price of good health goes up as it costs more to keep older patients healthy.
On the other hand, Grossman views educated people as “efficient producers of good health” because educated people are likely to invest in their health with preventive care, exercise, and avoidance of injury. These efficient methods of maintaining your “health capital” make the price of health relatively inexpensive. For example, it costs less to maintain a healthy diet than to manage the medical complications of obesity.
In this framework, the estimated price of good health depends on age, education, and many other variables besides the price of medical care. An interesting conclusion of Grossman's model is that, under certain conditions, an increase in the price of good health may simultaneously reduce the quantity of health demanded and increase the quantity of medical care demanded (J Political Econ 1972;80:223-55).
Grossman's model has been cited approximately 7,600 times and has been a foundational framework for many health services research articles. For example, Dardanoni used the Grossman model and examined the effects of uncertainty on the demand for medical care (J Health Econ 1990;9:23-38). Burggraf and colleagues analyzed the Russian demand for medical care to highlight the practical relevance of Grossman's health investment model (J Public Health 2016;24:41-56).
Demand and health insurance
How do we know if there is a demand curve for health, and what does it look like? Although beyond the scope of this column, the demand for health, health care, and health insurance are interrelated (Oxford Rev Econ 1989;5:21-33). Two early and important experiments in health insurance have shed light on the demand for medical care.
The RAND Health Insurance Experiment (HIE), 1974-1982, was the largest health policy study in U.S. history of how cost-sharing arrangements affect people's use of health services and outcomes (Am Econ Rev 1987;77:251-77). This complex randomized controlled design enrolled families in six cities in one of 14 different fee-for-service or prepaid group practice plans. Each plan offered different coverage and coinsurance for medical and dental services. Although conducted more than 40 years ago, this study is still the gold standard for understanding consumer responsiveness to out-of-pocket price.
The results of the HIE should not seem surprising. There was clear evidence of a demand curve for medical care. The catastrophic insurance plan reduced expenditures 31% relative to the zero out-of-pocket price plan. Price elasticity of demand is the percentage change in the quantity demanded divided by the percentage change in the price. The estimated price elasticity of demand in the HIE was approximately –0.2. In other words, a 1% increase in price resulted in a 0.2% decrease in demand. In 2016, Dunn found a similar price elasticity of demand using health insurance claims data and an instrumental variable approach (J Health Econ 2016;48:74-88).∗
The 2008 Medicaid expansion in Oregon based on lottery drawings was another randomized controlled study, albeit on a much smaller scale than the HIE. In the first two years there was substantially higher health care utilization, including primary and preventive care and hospitalizations, among those who enrolled in the Medicaid program (Q J Econ 2012;127:1057-1106; N Engl J Med 2013;368:1713-22).
Physician-induced demand for medical services
A key question in the research on the demand for medical services is whether physicians (supply) influence demand for medical services. In a 2019 international scoping review, the authors identified nine supply and three demand (patient-related) factors that may influence physician-induced demand (J Prev Med Public Health 2019;52:72-81). The supply factors are 1) physician income, 2) physician incentive for pecuniary profit, 3) type and size of hospital, 4) payment methods, 5) physician density, 6) type of employment, 7) consultation time per service, 8) change in the price of services, and 9) demographic and other observable characteristics of the physician. The patient-related factors were health status, observable non-clinical characteristics, and insurance coverage.
Much of the literature has focused on the influence of physician density. A 2009 systematic review of 25 studies found that higher geographic density of physicians was associated with more health care use (Health Policy 2009;91:121-34). Surgery and surgeons were examined in nine of the articles. In one of the first studies, based on data between 1963 and 1970, Fuchs found considerable support for the hypothesis that surgeons shift the demand for operations (J Hum Resour 1978;13:35-56). Not surprisingly, I could find no studies on physician-induced demand that focused on anesthesiologists.
There is still much debate on the topic. Does the association between physician density and medical care use reflect physician-induced demand or simply the increased availability to support pent-up demand? Reviewers of studies on physician-induced demand found many methodological inconsistencies and potential flaws. An important criticism is that the studies lack results for policy implications, and most economists argue that payment methods and insurance coverage are more important than physician density (Milbank Mem Fund Q Health Soc 1983;61:252-77).
The demand for anesthesia services
The demand for anesthesia services depends, to some extent, on the demand for surgery and other procedures that require anesthesia. Reviewing the theory and evidence concerning the demand for health care can provide a useful framework. In the context of procedural demand, the health economics literature supports several common sense hypotheses:
  • Price matters, whether direct to the consumer or indirect via insurance (e.g., an input cost factor to a bundled or global payment).

  • Quality and value matter as they reflect investment in human capital and support the shadow price of health.

  • An individual's current health status will affect the level of investment (per Grossman's human capital model) required and the willingness to make such investment.

  • The supply of the procedural workforce may influence demand.

The ASA Center for Anesthesia Workforce Studies (CAWS) uses age- and sex-specific population forecasts as the basis for projecting the demand for procedures and anesthesia services. Additionally, CAWS conducts literature reviews and seeks expert opinion from physician anesthesiologists to inform assumptions about changes in factors that influence procedural use rates (i.e., the number of per capita procedures). These factors include 1) health status of the population, 2) pharmaceutical and medical technology, 3) supply of surgeons and proceduralists, 4) payment levels and methods, and 5) insurance coverage and costs to patients. The work by CAWS is in its early development. To date, the COVID-19 pandemic has not been incorporated into the anesthesia demand model and forecasts are not available.
I appreciate the opportunity to contribute brief articles on health economics to the Monitor. Although these topics may not be of everyday concern to anesthesia professionals, I hope that others are as curious as I am. I welcome your suggestions for future issues at
Instrumental variables are an econometrics technique to statistically adjust for bias, which is avoided by design in randomized controlled studies but may occur with the use of retrospective secondary datasets (JAMA-J Am Med Assoc 1994;272:859-66; J Econ Perspect 2001;15:69-85; Epidemiology 2017;28:396-8)
Thomas R. Miller, PhD, MBA, ASA Director of Analytics and Research Services, and Director of the ASA Center for Anesthesia Workforce Studies (CAWS).
Thomas R. Miller, PhD, MBA, ASA Director of Analytics and Research Services, and Director of the ASA Center for Anesthesia Workforce Studies (CAWS).
Thomas R. Miller, PhD, MBA, ASA Director of Analytics and Research Services, and Director of the ASA Center for Anesthesia Workforce Studies (CAWS).
Submit a Comment
Submit A Comment

Contributors must reveal any conflict of interest. Comments are moderated.

Affiliation & Institution
I have a potential conflict of interest
Comment Title

This feature is available to Subscribers Only
ASA Member Login or Create an Account ×