Avi Dor

End-stage renal disease and economic incentives: the International Study of Health Care Organization and Financing (ISHCOF)

End-stage renal disease (ESRD) is a debilitating, costly, and increasingly common condition. Little is known about how different financing approaches affect ESRD outcomes and delivery of care. This paper presents results from a comparative review of 12 countries with alternative models of incentives and benefits, collected under the International Study of Health Care Organization and Financing, a substudy within the Dialysis Outcomes and Practice Patterns Study. Variation in spending per ESRD patient is relatively small, but correlated with overall per capita health care spending. Remaining differences in costs and outcomes do not seem strongly linked to differences in incentives.

Loss of health insurance and the risk for a decline in self-reported health and physical functioning.

Background. Millions of Americans are intermittently uninsured. The health consequences of this are not known. Setting. National survey. Participants. Six thousand seventy-two participants in the Health and Retirement Study (HRS) age 51 to 61 years old with private insurance in 1992. Measurements. Loss of insurance coverage between 1992 and 1992 and development of a major decline in overall health or a new physical difficulty between 1994 and 1996. Results. In 1994, 5768 (95.0%) people continued to have private insurance, 229 (3.8%) reported having lost all insurance, and 75 (1.2%) converted to having only public insurance. Over the subsequent 2 years (1994–1996), the risk for a major decline in overall health was 15.6% for those who lost all insurance versus 7.2% for those with continuous private insurance (P <0.001). After adjusting for baseline sociodemographics, health behaviors, and health status, the adjusted relative risk for a major decline in health for those who lost coverage was 1.82 (95% CI, 1.25–2.59) compared with those with continuous private insurance. Those who lost insurance also had a higher risk for developing a new mobility difficulty compared with those with continuous private insurance (28.5% vs. 20.4%, respectively;P = 0.02), but this was not significant in multivariate analysis (adjusted RR, 1.26; 95% CI, 0.90–1.68). Conclusions. Loss of insurance has adverse health consequences even within 2 years after becoming uninsured. Studies of insurance coverage should routinely measure the number of Americans uninsured at any time over the preceding 2 years as a more accurate measure of the population at risk from being uninsured.

Non-minimum cost functions and the stochastic frontier: On applications to health care providers

By allowing firms to deviate from their optimal path for systematic reasons, frontier estimators in the general literature have helped to inject more realism into econometric modelling of firms, and now, of hospitals and nursing homes as well. Unfortunately, one such estimator, the DEA technique (employed by Kooreman) does not allow for a stochastic error term. In practice, all random noise in the DEA is lumped together with the true inefficiency, making the resulting inefficiency scores suspect. The stochastic frontier model has the advantage of disentangling the two sources of error. For this reason I will address most of my comments to the papers in this issue by Vitaliano and Toren and by Zuckerman et al. that rely on maximum likelihood estimators (MLE). Another way to estimate the stochastic frontier is by using panel data techniques. Although the latter approach is not represented in this compendium it has certain advantages over MLE, which I will touch upon briefly. I will also discuss other approaches that are derived more directly from economic theory that have not yet received much attention.

The costs of Medicare patients in nursing homes in the United States : A multiple output analysis

The reluctance of nursing homes to admit Medicare beneficiaries is well recognised. A possible explanation is that under the current Medicare reimbursement policy, nursing homes do not have adequate incentives to increase admissions. This paper examines that hypothesis. Using a national sample of nursing homes, Medicare-specific marginal costs were estimated from a flexible form of the cost function. In most cases, Medicare-specific marginal costs were well above Medicare reimbursement rates. From an investigation of economies of scale and economies of scope, it appears that reimbursements may be sufficient only in nursing homes that specialize in Medicare beneficiaries. Unless Medicare reimbursements are brought in line with costs, access of Medicare beneficiaries to the nursing homes will remain limited.

Does Prescription Drug Adherence Reduce Hospitalizations and Costs? The Case of Diabetes

PURPOSE To estimate the impact of diabetic drug adherence on hospitalizations, emergency room (ER) visits, and hospital costs. METHODS It is often difficult to measure the impact of drug adherence on hospitalizations since both adherence and hospitalizations may be correlated with unobservable patient severity. We control for such unobservables using propensity score methods and instrumental variables for adherence such as drug coinsurance levels and direct-to-consumer advertising. FINDINGS We find a significant bias due to unobservable severity in that patients with more severe health are more apt to comply with medications. Thus, the relationship between adherence and hospitalization will be underestimated if one does not control for unobservable severity. Overall, we find that increasing diabetic drug adherence from 50% to 100% reduces the hospitalization rate by 23.3% from 15% to 11.5%. ER visits reduce by 46.2% from 17.3% to 9.3%. Although such an increase in adherence increases diabetic drug spending by

Use of preventive services by medicare fee-for-service beneficiaries: Does spillover from managed care matter?

776 a year per diabetic, the cost savings for averted hospitalizations and ER visits are

Hospital Transaction Prices and Managed-Care Discounting for Selected Medical Technologies

886 per diabetic, a cost offset of

THE MEDICARE COST OF RENAL DIALYSIS : EVIDENCE FROM A STATISTICAL COST FUNCTION

1.14 per

Pricing of Surgeries for Colon Cancer: Patient Severity and Market Factors

1.00 spent on diabetic drugs. ORIGINALITY Most of the drug cost-offset literature focuses only on the impact of cost-sharing and drug spending on cost-offsets, making it impossible to back-out the empirical impact of actual drug adherence on cost-offsets. In this chapter, we estimate the direct impact of adherence on hospitalizations and costs.

The hospital-physician interaction in U.S. hospitals: Evolving payment schemes and their incentives

Background:Health care delivery varies with the level of managed care activity (MCA) in an area, potentially affecting health care for those not participating in managed care programs. However, the extent to which MCA is associated with the use of cancer screening by fee-for-service beneficiaries (FFS) is unclear. Objective:We sought to study colorectal cancer screening among Medicare FFS beneficiaries in relation to levels of Medicare MCA. Research Design:This study linked 1999 Medicare denominator and Part B claims data with the 1998 Area Resource File. After categorizing MCA as low (<10%), moderate (10–29.99%), or high (≥30%), we assessed the association between colorectal cancer screening among FFS beneficiaries and MCA, controlling for individual demographic variables and county-level attributes of socioeconomic status and physician resources. Subjects:We included Medicare FFS beneficiaries 65 years of age or older with both Part A and Part B coverage for the entire calendar year from large counties in the study. Measures:We measured the likelihood of undergoing fecal occult blood testing (FOBT), flexible sigmoidoscopy (FLEX), or colonoscopy (COL). Results:Compared with Medicare FFS beneficiaries residing in counties with low MCA, those in high MCA counties were significantly more likely to undergo FOBT (adjusted odds ratio [AOR] 1.10, 95% confidence interval [CI] 1.04–1.16), FLEX (AOR 1.11, 95% CI 1.04–1.18), or colonoscopy, after receiving FOBT/FLEX (AOR 1.07, 95% CI 1.02–1.13). Conclusions:From a public health perspective, an association between higher levels of MCA and colorectal cancer screening among those not enrolled in managed care may translate into modest increases in use of colorectal cancer screening and possibly earlier detection.