Tracking Health Care Costs

Health Affairs

Volumn 15, Issue 3, 1996, Pages 140-149

  Ginsburg, Paul B ^a   Pickreign, Jeremy D ^a  

^a CENTER FOR STUDYING HEALTH SYSTEM CHANGE   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

ARTICLE; ECONOMICS; EPIDEMIOLOGY; HEALTH CARE COST; HEALTH CARE DELIVERY; HEALTH CARE POLICY; HEALTH INSURANCE; HEALTH SERVICES RESEARCH; HUMAN; INFORMATION PROCESSING; UNITED STATES;

DATA COLLECTION; DELIVERY OF HEALTH CARE; HEALTH BENEFIT PLANS, EMPLOYEE; HEALTH CARE COSTS; HEALTH EXPENDITURES; HEALTH POLICY; HEALTH SERVICES RESEARCH; HUMANS; INSURANCE, HEALTH; SAMPLING STUDIES; UNITED STATES;

EID: 0344554749     PISSN: 02782715     EISSN: None     Source Type: Journal    
DOI: 10.1377/hlthaff.15.3.140     Document Type: Article

References (23)

1. For example, see K.R. Levit, H.C. Lazenby, and L. Sivarajan, "Health Care Spending in 1994: Slowest in Decades," Health Affairs (Summer 1996): 130-144. In 1994 rates of growth were 12 percent for Medicare and 4 percent for private insurance. If the question was how burdensome are increases in health care costs to public and private purchasers, then the numbers do not need adjustment. If the question concerns efficacy in cost control, then adjustments must be made for factors such as enrollment growth, types of services covered, and the lag structure of Medicare volume performance standards. These adjustments reduced the differential from eight percentage points to two percentage points.
2. Often the terms costs and expenditures are used interchangeably. Conceptually, the primary interest is in costs, which reflect the resources devoted to health care that are not available to produce other goods and services. Practically, most available data reflect expenditures, or what is paid for health services by those who purchase them (or received by the providers of health services). Costs and expenditures differ when the payment is greater or less than the resources that go into providing the services.
3. If a slowing of cost increases comes mainly from changes in prices paid for services, this scenario raises questions about the potential for continued reductions in the rate of cost growth. There is a limit to how far prices can decline before it becomes impossible to keep labor and capital producing those services. Economists also make a distinction between price changes, which reflect a redistribution of resources between buyers and sellers, and changes in the quantity of health services, which represent a net change in resources going to health care. See M.V. Pauly, "When Does Curbing Health Care Costs Really Help the Economy?" Health Affairs (Summer 1995): 68-82; and P.B. Ginsburg, "The Role of Society in Making Distributive Judgments," Health Affairs (Fall 1995): 283.
4. If a slowing of cost increases comes mainly from changes in prices paid for services, this scenario raises questions about the potential for continued reductions in the rate of cost growth. There is a limit to how far prices can decline before it becomes impossible to keep labor and capital producing those services. Economists also make a distinction between price changes, which reflect a redistribution of resources between buyers and sellers, and changes in the quantity of health services, which represent a net change in resources going to health care. See M.V. Pauly, "When Does Curbing Health Care Costs Really Help the Economy?" Health Affairs (Summer 1995): 68-82; and P.B. Ginsburg, "The Role of Society in Making Distributive Judgments," Health Affairs (Fall 1995): 283.
5. Many believe that a shift from inpatient hospital services toward outpatient services and a shift from specialty physician services toward primary care services are desirable, although hard evidence about the efficacy of such changes in the aggregate is limited. Data showing such shifts would lead some to have a more positive appraisal of reduced rates of cost increases.
6. Foster Higgins National Survey of Employer-Sponsored Health Plans (Press release, 30 January 1996).
7. HCFA Office of National Cost Estimates, "Revisions to the National Health Accounts and Methodology," Health Care Financing Review (Summer 1990): 42-54.
8. See J.P. Cookson and P. Reilly, "Modeling and Forecasting Health Care Consumption," in Milliman and Robertson Research Report (Radnor, Pa.: Milliman and Robertson, August 1994).
9. The HCI simulates premium trends for a standard health insurance policy with a specified deductible by applying actuarial adjustments to the trend data from providers. Since we did not want to reflect the effects of an unchanging deductible, we obtained a version for a zero-deductible policy-essentially no actuarial adjustments.
10. We are grateful to Mark Pauly for suggesting this use of the series.
11. G.A. Jensen et al., "Employer-Sponsored Health Insurance in 1995" (Unpublished manuscript, March 1996).
12. Under COBRA 1995, firms are, required to offer continuation coverage at 102 percent of its cost to insured employees who leave the firm. The COBRA rate is a useful data element because Labor Department regulations prescribe its calculation, and employers must file the rate.
13. All of the means and variances here are geometric rather than arithmetic. See T.B.L. Kirkwood, "Geometric Means and Measures of Dispersion," Biometrics 35 (December 1979): 908-909.
14. 2 = 0.91). If the EHCI had been a perfect leading indicator of national health expenditures, then the elasticity would have been 1.0, and the intercept would have been 0.0. The elasticity and the intercept are significantly different from these two benchmarks.
15. Applying the adjustment to the EHCI for 1995 increases it from 3.2 percent to 5.6 percent. But this adjustment may be too large. In this very simple adjustment model, the very low rates of increase in costs over the past few years are interpreted as random, whereas the sum of the evidence presented here suggests a secular change. For this reason, we did not include the adjusted value in Exhibit 1.
16. 2 = 0.86). Neither the elasticity nor the intercept was significantly different from the benchmarks described above. Adjustment based on this model did not reduce the mean absolute difference.
17. We used the recently developed gross domestic product (GDP) chain-type price index. See Levit et al., "Health Care Spending in 1994."
18. Note that we calculated changes for the premiums for a standardized mix of plans. This is likely to be a conservative measure of the decline in premium growth, since premium savings from shifts to managed care are not counted.
19. Note that this comparison is with the HCI, which is more comparable to private health insurance premiums than is the expanded version developed here, which includes Medicare.
20. R.H. Miller, "Competition in the Health System: Good News and Bad News," Health Affairs (Summer 1996): 107-120.
21. It is difficult to separate overall spending increases into price and quantity (the latter including service mix and intensity) because of limitations of the CPI in capturing discounts, especially for medical services. The CPI methodology is being changed to account for discounts. New components of the Producer Price Index (PPI) for health services, which are designed to reflect discounts, have recently been developed. Although the PPI differs from the CPI in important ways, the hospital component does show a lower rate of increase in hospital prices for the past two years, but the difference from the figure underlying Exhibit 3 is only about one percentage point per year.
22. Prospective Payment Assessment Commission, Medicare and the American Health System: Report to Congress (Washington: ProPAC, June 1996), Figure 3-2.
23. M.A. Morrisey, Cost Shifting in Health Care: Separating the Evidence from Rhetoric (Washington: American Enterprise Institute Press, 1994).