The Impact of Market Forces on
Physician Outcomes
Bruce Landon
We make use of data from the Community Tracking Survey (CTS), a survey of about
12,000 physicians in 60 geographic markets. Measures of competition in these
markets will be used to study the effect of competition and organizational arrangements
on physician perceptions of quality and satisfaction with their practices. Our
measures of competition will include indicators of competition in the markets
for health plans, medical groups and individual medical specialties. A variety
of multivariate statistical models will be used to estimate the impacts of competition
on key outcomes.
Structuring Hospital Networks
and their Effects on Services and Prices
Albert Ma
Providers respond to managed care by doing less in ways that cannot be explained by utilization review or supply-side price incentives. A provider's opportunity to be included within the managed care network (or said another way, the threat to be excluded) works a powerful incentive. We define a "network" as a subset of the providers in a market selling services to a managed care plan on favorable terms (to the plan). In exchange, the plan directs business to the providers in the networks.
The specific aims of this project are:
Consequences for Cardiac
and Other Conditions of Increased Price Competition in Provider Markets
Ed Guadagnoli, John Ayanian, Mary Beth Landrum, Ellen Meara
We will examine whether the quality of care provided to traditional Medicare
patients (fee-for-service) changes with increases in HMO penetration, both overall
penetration and Medicare penetration. We will also examine whether and how provider
composition within a market, the availability of technology, and more conservative
practice patterns might affect quality of care associated with HMO penetration.
Our findings will have important policy implications. If the quality of care
is poorer in areas with higher HMO penetration, stricter measures to monitor
the quality provided by managed care organizations than currently would seem
indicated. Conversely, if quality is better or the same, efforts to increase
managed care enrollment would be indicated.
Carve Outs, Cost
Shifting, and Quality Side Effects
Haiden Huskamp, Margarita Alegria
Carve-out arrangements have become common features of managed care. Separating specific services from the overall plan's risk pool is done by both payers and health plans. Carve-outs are most frequently used in managing mental health and substance abuse care and prescription drugs. More recently carve-outs are being observed in connection with cancer care, cardiac care and radiology services. While these organizational arrangements are not new, they have grown rapidly in recent years. Most research to date has focused on why payers choose to carve-out certain services and what the impacts payer carve-outs have on cost, access and quality of behavioral health (mental health and substance abuse) care. In the research proposed here we extend that body of work by: examining the market forces that influence the choice by health plans to carve-out specific services; studying the impact of the method by which primary care physicians are paid on the use of mental health and pharmacy services under mental health carve-outs; and examining the impact of a behavioral health care carve-out on the use of pharmacy services. This project will draw on national survey data from the AAHP, a unique data set which integrates medical claims, pharmacy claims, specialty mental health claims and administrative data from managed care organizations, and survey data from a quasi-experiment with a carve-out in Puerto Rico.
Selection and Risk Adjustment
in Private Employers' Health Plans
Joseph Newhouse, Ellen Meara
Using a natural experiment, we will estimate the extent of selection among the 170,000 salaried employees and their dependents at the Ford Motor Company. Employees in their first two years of employment have no choice of an indemnity plan and in their third year are given a choice. This enables us to look at utilization of health care services by statistically similar persons who do and do not have a choice of an indemnity plan. Because the employees differ only in their length of tenure, it should be possible to estimate both the effect of the indemnity plan on use, as well as selection. This study will represent one of the few studies of selection among the under 65 which will have access to data on utilization within managed care plans (encounter data).
The literature's usual remedy for selection is risk adjustment (or better risk adjustment). Although Medicare is about to implement health-status based risk adjustment, its lack of use in the private sector is striking. We hypothesize that this may be related to the relationship between health insurance and compensation. For employers that contribute a lump-sum toward multiple health insurance plans, risk adjustment would alter relative compensation in ways that employers may be unable to compensate for with offsetting changes in cash wages. If so, employers may be unwilling to implement changes in their compensation schemes. We wish to explore the plausibility of this hypothesis by calculating in two employer groups how much redistribution would be caused by various risk adjustment methods. Note, that this explanation is consistent with the greater use of risk adjustment by Medicare than by most private employers.
Explaining Managed Care
Penetration in Rural Areas
Joseph Newhouse
The lack of managed care penetration in rural areas, especially in Medicare, is often ascribed to inadequate reimbursement; indeed the Balanced Budget Act of 1997 sharply increased Medicare reimbursement for HMOs in many rural areas (so-called floor counties). But participation by HMOs did not increase as a result. We wish to explore the hypothesis that the lack of penetration has a more fundamental cause, namely that many rural markets are not very competitive for either physicians or hospitals. As a result, managed care plans have less bargaining power than in urban markets and are less able to obtain price discounts or enforce compliance with utilization controls. Specifically, we will test the hypothesis that in rural areas where there are more providers, and hence more competition, entry and penetration of managed care is larger. We will use an 11 state sample, where we know the location of physicians by specialty and hospitals.
Program Project Core Activities
Theoretical Core: Tom McGuire
Statistical Core: Sharon-Lise Normand
Junior Investigator Development: Richard Frank