«10.1 Health Care and the Cost of Living The standard analytic framework for constructing a cost-of-living index compares the change in expenditure ...»
Irving Shapiro, Matthew D. Shapiro, and David W. Wilcox.
“Measuring the Value of Cataract Surgery”
In David M. Cutler and Ernst R. Berndt, eds.
Medical Care Output and Productivity.
Chicago: University of Chicago Press, 2001.
(National Bureau of Economic Research/Conference on Research in Income
and Wealth. Studies in Income and Wealth, Volume 62)
Irving Shapiro, Matthew D. Shapiro,
and David W. Wilcox
10.1 Health Care and the Cost of Living The standard analytic framework for constructing a cost-of-living index compares the change in expenditure between a base and a reference period needed to deliver a fixed level of utility.' This framework, which relies on a stable, well-defined function relating per period expenditure to prices and utility, has serious limitations for measuring how health-care expendi- tures affect the cost of living.
Health expenditure on life-extending therapies can not only increase 0 current-period utility, but have durable effects on utility. Durability per se is not special to health care. The purchase of a refrigerator or car also provides for a flow of utility into the future. This aspect of durability can be handled in a cost-of-living index by taking a service- flow approach.2 By making the number of periods of life endogenous, Irving Shapiro is medical director at the Phillips Eye Institute, Minneapolis, and clinical professor of ophthalmology at the University of Minnesota. Matthew D. Shapiro is professor of economics and senior research scientist at the Survey Research Center at the University of Michigan, and a research associate of the National Bureau of Economic Research. David W. Wilcox is assistant secretary for economic policy at the U.S. Department of the Treasury.
The authors are grateful to Praveen Kache and Laura Marburger for research assistance.
Matthew Shapiro gratefully acknowledges the financial support of the National Institute on Aging through the Michigan Exploratory Center on the Demography of Aging and program project 2-PO1 AG 10170. The authors gratefully acknowledge the very helpful discussions with and comments of Andrew Abel, Zvi Griliches, Richard Suzman, and participants in the NBER Summer Institute. The views expressed in this paper are not necessarily those of the U.S. Department of the Treasury.
Copyright is not claimed for this chapter.
I. See Pollak (1989) or Diewert (1987).
2. For the U.S. Consumer Price Index, housing purchases, but not purchases of other durables. are accounted for on a service-flow basis, Hence, while the data requirements and 41 1 412 Irving Shapiro, Matthew D. Shapiro, and David W. Wilcox health care expenditures can cause interesting, and even perverse, im- plications for cost-of-living measurement. Specifically, a life-saving expenditure can substantially increase the annuity value of expenditure needed to maintain a fixed level of per-period utility over a longer lifetime.3 Traditional cost-of-living measurement, which takes a perperiod rather than a lifetime perspective, does not account for this effect.
Most health care expenditure is driven by adverse shocks to health.
Again, standard cost-of-living analysis, which compares a stable utility function across time, does not account for this aspect of health expenditure. These shocks will have direct effects on the demand for health care expenditures. They will also have cross-effects on demand for other expenditures even after compensation for the wealth effects of the health shock.
Many medical treatment decisions are binary, with little scope for varying either the quantity of treatment or its quality. Especially in the United States, only treatments at or close to the state of the art are ~ f f e r e dHence, health expenditures appear to be lumpy and exogenous, especially given the importance of third-party payment. Consequently, health care is not easily modeled in the marginalist framework that underlies the theory of cost-of-living indexes.
These points about health care-their potential life-and-death nature, their state-contingency, and their exogeneity-might drive one to the conclusion that economic analysis of the choice to undertake treatment is inappropriate. In this paper we argue, however, that economic decision making by patients is important for understanding the demand for certain medical procedures. Shocks to health need not require acute treatment.
Many conditions are chronic and progressive. The patient may face a slow, variable, and unpredictable progression of disease. In such cases, the timing and nature of the treatment might be highly uncertain and variable across patients. Moreover, over time as treatment regimes change, the medical intervention may take place at a different point in the course of the other conceptual problems for measuring health remain highly problematic, durability in itself is not a unique complication.
3. We are grateful to Zvi Griliches for emphasizing this point to us.
Indeed, for a person with fixed lifetime resources, this life-extending intervention could lead to an impoverishment of nonhealth expenditures as fixed resources are more thinly spread to maintain consumption over a longer lifespan. The annuity features of Social Security and Medicare, however, provide some insurance against these consequences of health expendi ture.
4. The state of the art does diffuse slowly, so there may be variation by the setting where the health care is delivered, by region, and so on. Moreover, patients with different access to health care (owing to insurance coverage, locale, education, or income) may receive different treatment. Yet, for a particular patient, there is typically little economic trade-off in the choice of treatment.
Measuring the Value of Cataract Surgery 413 disease. Even if the medical intervention once it is indicated is exogenously determined by the state of the art, with little or no scope for varying the quantity or quality of the intervention, the timing of the intervention may be highly endogenous.
Some medical procedures have declining costs over time, possibility in pecuniary terms, but especially when quality of outcomes and reduced morbidity are taken into account. This declining cost has important consequences for the demand for the procedure to the extent that it relaxes the medical criteria for receiving treatment. Heterogeneity in the course of the disease makes it important to distinguish between these two margins of adjustment. For some patients, relaxed criteria for receiving a medical intervention will affect the timing of treatment, with treatment being received earlier in the course of the disease as criteria become more relaxed;
for others with less serious disease, or with disease that is slower to progress, they might never become candidates for a treatment under tighter criteria, but will receive it under relaxed criteria, perhaps quite early in the course of the disease.
The demand for the procedure will increase as the effective price falls and the equilibrium moves down the demand curve. The movement down the demand curve means that the marginal valuation of the procedure is lower. Hedonic regression or survey assessments of the quality of outcomes from medical procedures will reveal declining marginal benefit over time. It would be incorrect, however, to mechanically apply such results in a cost-of-living analysis. In particular, to the extent that patients receive the intervention earlier in the course of the disease, the main benefits of the procedure might come many periods after the procedure. While the benefits of having the new procedure might be quite small in the period the procedure is carried out, substantial benefit accrues in subsequent periods where the patient avoids suffering the progressively worsening symptoms and disability while waiting to become a candidate for the former procedure. A cost-of-living index that takes into account only the benefit in the period the procedure is carried out will have a potentially large upward bias.
This paper will present a case study of cataract surgery. Dramatic changes in the technology for cataract surgery make it an excellent illustration of the importance of accounting for the timing of procedure in the course of a disease. We will argue in the conclusion, however, that similar considerations apply to the treatment of various medical conditions.
The organization of the remainder of the paper is as follows. Section
10.2 outlines developments in the techniques of cataract surgery since midcentury. It then discusses how these improvements in technique have reduced the degree of visual impairment of patients receiving cataract surgery, thereby dramatically increasing the rate of surgery. Section 10.3 discusses how the benefits of surgery should be valued across time given the changing visual function at time of surgery. Section 10.4 discusses the 414 Irving Shapiro, Matthew D. Shapiro, and David W. Wilcox resource and monetary costs of cataract surgery. It presents a cost index for cataract surgery and contrasts the results with the current Bureau of Labor Statistics (BLS) procedures for measuring health cost. Section 10.5 makes recommendations for measuring prices in the health care sector based on the findings about cataract surgery. Section 10.6 offers conclusions.
10.2 Treatment of Cataract Section 10.2.1 gives the chronology of treatment for cataract since World War 11. Section 10.2.2 describes how changes in the techniques for cataract surgery and for postoperative optical correction changed the criteria for cataract surgery over time. Section 10.2.3 describes how these improvements have led to relaxed criteria for extraction of cataracts and dramatically increased rates of cataract extraction.
10.2.1 Techniques of Cataract Surgery: A Chronology The lens focuses light coming into the eye onto the retina. A cataract is a cloudy lens, which can impair vision. Cataracts are removed surgically.
U p to the late 1970s, no other lens was inserted into the eye, so anyone whose cataracts had been removed required thick glasses or contact lenses to provide focus. In the late 197Os, however, surgeons in the United States started inserting an intraocular lens (IOL) as a replacement for the cloudy, natural lens. IOLs eliminate the need for thick glasses or contact lenses.
They leave the patient with much better postoperative vision than they could have obtained with the cataract glasses and eliminate the need for inserting, removing, and caring for contact lenses.
There have been dramatic changes in the technique of cataract surgery-how the incision is made, how the cataract is extracted, and how the incision is closed. In the immediate post-World War I1 period, extracapsular extraction was the standard technique. This technique did not necessarily remove all the cataract. In the early 1950s, the technique switched to intracapsular extraction, in which the entire cataract and its enclosing envelope (capsule) were removed by suction or freezing. Because these techniques required a large incision, standard postoperative care included hospitalization often as long as a week. Through the 1960s, techniques of extraction and suturing gradually improved. These improvements were facilitated by the routine use of an operating microscope.
Hospital stays were reduced to a typical stay of three days.
The modern era of extracapsular extraction opened in the early 1970s.
This technique was pioneered with phacoemulsification, a technique where the cataract was broken into tiny pieces and removed from the eye by controlled suction. The smaller incisions allowed by phacoemulsification made outpatient treatment increasingly prevalent. Yet, the typical extractions remained intracapsular. Improvements in sutures and suture techMeasuring the Value of Cataract Surgery 415 nique, giving more secure wound closure, allowed hospital stays with intracapsular extraction to fall to a single night. At the end of the 1970s, there was an increasing trend toward the use of phacoemulsification with its smaller incision. The 1980s saw an increased use of phacoemulsification because of its small incision, complete removal of the cataract, and the reduced postoperative complications allowed by leaving the posterior capsule intact. Leaving the posterior capsule intact reduced postoperative complication~.~ By 1990, phacoemulsification was common for extraction of the cataract. Improvements in techniques in the 1990s included further reduction in the size of the incision. Smaller incisions can be closed with fewer sutures, resulting in better and faster healing of the wound. IOLs were designed to fit through the small incision. Indeed, it is now possible to make incisions that heal without suturing. With reduced or no time needed for suturing, the operation can be completed quickly, sometimes in less than ten minutes. This improvement in surgical technique has allowed for innovations in the delivery of anesthesia. The standard technique has been to inject the anesthetic agents beside the eye and behind it. With a fast and highly controlled operation, anesthesia can now be in the form of topical drops on the eye and anesthetic agent in the irrigating solutions within the eye. New developments in IOLs are improving postoperative vision.
Standard IOLs are focused at a fixed distance. Multifocal lenses, which are recently becoming common, allow focus at several distances.
See table 10.1 for a summary of the evolution of cataract treatment, and an estimate of the number of days in hospital each treatment required for a typical patient with no other complications.6 The outpatient surgery includes both surgery done in a hospital and surgery done in outpatient clinics. which tends to cost less.
10.2.2 Interaction of Improvement in Surgical Techniques and the Threshold for Surgery Throughout the period being studied, the criteria for surgery have been based on the extent to which the cataract impairs activities of daily life, such as work, reading, driving, and leisure activities. There are objective tests of visual acuity (e.g., Snellen visual acuity),’ which are indicative of whether a patient is a candidate for surgery, but there are no hard and fast rules for assessing whether a patient is a candidate for surgery based on these measurements alone. The physician must assess other underlying medical conditions-both of the eye and generally. Moreover, patients
5. The YAG laser could treat a clouded posterior capsule without invasive surgery.