The Direct Payment Method for Medicare Managed Care Plans

This report was prepared under Contract Number 500-00-0025/T.O.#1 with the Centers for Medicare and Medicaid Services. Any views expressed are those of the authors and do not represent The Urban Institute, its trustees or its sponsors.

The nonpartisan Urban Institute publishes studies, reports, and books on timely topics worthy of public consideration. The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders.

TABLE OF CONTENTS

EXECUTIVE SUMMARY

I. INTRODUCTION AND BACKGROUND
A. THE GOAL OF THE DIRECT PAYMENT METHOD
B. BACKGROUND ON MEDICARE PAYMENTS TO MANAGED CARE PROGRAMS

1. TEFRA Risk Program (1982-1997)
2. Medicare+Choice Program

C. THE DIRECT PAYMENT METHOD

1. Defining the Direct Payment Method
2. Contrast With the Current M+C Payment System

II. CONCEPTUAL FRAMEWORK UNDERLYING THE DIRECT PAYMENT
METHOD: THE MARKET FOR MEDICAL CARE

III. RISK ADJUSTMENT IN MEDICARE MANAGED CARE
A. INTRODUCING RISK ADJUSTED PAYMENTS TO MEDICARE+CHOICE PLANS

1. Risk Adjustment Transition Schedule
2. Interim Risk Adjustment Method
3. Limitations of the Interim Risk Adjuster

B. INTRODUCING AN INPATIENT- AND AMBULATORY-BASED RISK ADJUSTER (2004 AND BEYOND)
1. Data Collection Efforts
2. The "Selected Significant Condition" Risk Adjuster
3. Risk Adjustment under a Direct Payment Method

IV. GEOGRAPHIC INPUT PRICE INDEX
A. What Are The "Inputs" To Medicare+Choice Plans?
B. CURRENT APPROACH
C. RECONSIDERING THE CURRENT APPROACH
D. AN ALTERNATIVE TO THE CURRENT INPUT PRICE INDEX
E. DATA AVAILABILITY AND FEASIBILITY OF NEW DATA COLLECTION

V. ELEMENTS OF THE "OTHER GEOGRAPHIC" AND "OTHER INDIVIDUAL
CHARACTERISTICS" ADJUSTMENT FACTORS
A. PRIOR RESEARCH ON DETERMINANTS OF MEDICARE SPENDING PER BENEFICIARY
B. POTENTIAL CANDIDATES FOR THE INDIVIDUAL CHARACTERISTICS ADJUSTMENT FACTOR
C. POTENTIAL CANDIDATES FOR THE "OTHER" GEOGRAPHIC FACTORS ADJUSTMENT
(MEDICAL AND HEALTH CARE MARKET CHARACTERISTICS)
D. DATA AVAILABILITY

VI. ESTIMATING THE EMPIRICAL MODEL UNDERLYING THE DIRECT
PAYMENT METHOD
A. THE DEPENDENT VARIABLE: PAYMENT VS. COST, OR WHAT'S P AND WHAT'S Q.
B. EMPIRICAL/OPERATIONAL ISSUES

1. Functional Form
2. Statistical Estimation

C. SUMMARY: AN ILLUSTRATIVE EXAMPLE OF IMPLEMENTING THE DIRECT PAYMENT METHOD

VII. THE DIRECT PAYMENT METHOD IN THE POLICY CONTEXT
A. POLICY CHOICES THAT NEED TO BE MADE
B. ISSUES AND IMPLICATIONS FROM OTHER STAKEHOLDER PERSPECTIVES

1. Health Plan Issues
2. Beneficiary Issues
3. Provider Issues
4. Taxpayer Issues

C. CONCLUDING COMMENTS

REFERENCES

Executive Summary

This concept paper extends and develops an idea for an alternative methodology Medicare might use to determine capitation payments to health plans that participate in the Medicare+Choice program. The Direct Payment Method, first suggested by Greenwald et al. (1998), would shift the basis of payment away from the current approach, which builds on county-based Medicare fee-for-service (FFS) expenditures, to a national model that predicts expenditures for individual Medicare enrollees. Ideally, the Direct Payment Method would use detailed encounter data for Medicare+Choice enrollees, information on the cost of the services they received, and a health-based risk adjuster to predict expenditures for individual enrollees. When combined with a geographic input price adjustment, the resulting payment would be independent of the extreme and highly problematic geographic variations in Medicare's FFS payments per enrollee.

In reality, the detailed encounter data and information on the cost of services used by Medicare+Choice enrollees are unlikely to be available any time in the foreseeable future. Therefore, the goal of this concept paper is to elaborate on the basic idea behind the Direct Payment Method and to consider options for implementing it using existing FFS data. Given the necessity of basing Medicare+Choice payment on FFS data, the concept paper characterizes the main difference between the Direct Payment Method and the current payment methodology as one of explicit vs. implicit adjustment for the many factors that influence geographic variations in payments. The current payment methodology explicitly recognizes payment differences due to variations in enrollees' health risk and geographic areas' input prices. All other sources of variation are implicit in the county-based rates, which are blended with a national average county rate in order to reduce variation.

The Direct Payment Method, as developed in this concept paper, would identify and make explicit adjustments, guided by policy objectives, for as many of the other factors, in addition to health risk and input prices, that can be measured with current data. In particular, the Direct Payment Method is conceptualized as the product of a national base rate, which is simply the national average Medicare expenditure per enrollee in the FFS program, and four explicit adjustment factors for health risk, input prices, geographic factors other than input prices, and enrollee characteristics other than the health and demographic measures already incorporated in the health risk adjuster. This formulation is shown to be the equivalent of a comprehensive model of the determinants of health care expenditures, where the elements of the model are based on either the economic theory of the demand for and supply of medical care or the need-predisposing-enabling theory of medical care use (Chapter II).

The subsequent chapters consider each of the elements of the expanded Direct Payment Method. Chapter III summarizes the current state of risk adjustment in Medicare+Choice, recognizing that the Direct Payment Method would use the same risk adjustment factor as the current payment methodology. Chapter IV addresses a variety of issues in the development of a geographic input price index, including the definition of an "input" from the perspective of a health plan that does not provide medical care services directly to enrollees but purchases them from independent physicians, hospitals and other medical care providers.

Chapter V reviews issues in the development of the "other geographic" and "other personal characteristics" adjustment factors. The essence of this chapter is that both theory and prior research show that many factors other than health risk and input prices influence medical care use and expenditures. It identifies a number of such factors that could be incorporated into the Direct Payment Method's formula. This chapter also notes that current Medicare administrative data do not include information on most of the additional personal characteristics, such as income, education, marital status, and living arrangement, that prior research has identified as influencing use and expenditures. As an alternative, it suggests using small-area (5-digit zip codes) population characteristics to represent the effects of individuals' personal characteristics. This chapter also addresses the issue of using flexible definitions of medical care markets for different types of services based on distances between enrollees' residential zip codes and the zip codes of various types of medical care providers.

Chapter VI considers empirical estimation and implementation of the Direct Payment Method. It shows that the essential elements of the payment formula, the various adjustment factors and their weights, could be derived from an empirical model that posits expenditures to be a multiplicative function of the risk adjuster, the prices of the relevant inputs, and variables measuring the other geographic and personal characteristics suggested by theory and prior research. The parameters of this statistical function become the weights assigned to the adjustment factor indices, which are essentially the ratios of the values of a set of characteristics for a particular geographic area to the national average values for those characteristics.

This chapter and the final chapter consider the Direct Payment Method in the policy context. At the technical level, policy can decide whether to include or exclude certain specific factors from the payment formula. If a particular factor, such as income, the availability of hospital beds, or the extent of total HMO enrollment in the area, is included, policy can also decide whether to recognize as desirable or allowable all of the variation in the factor, or whether to set floors and/or ceilings that would constrain the effects of a factor on the resulting payment.

More broadly, in going beyond risk and input price adjustment, the analyses required to develop the Direct Payment Method concept would identify the sources of local variation in Medicare expenditures. It would be up to policymakers to decide which of these factors to explicitly incorporate in plan payments. However, the Direct Payment Method would not tell policymakers what is necessarily the "right price" or the "right degree of variation" to invite enough plan participation to produce the desired level of beneficiary access in all geographic areas. In this sense, the Direct Payment Method is not a panacea.

Moreover, the ability to adjust various components of the Direct Payment Method to meet particular policy goals can be viewed as either strength or a weakness. It opens up the payment process to political influence and maneuvering, particularly as compared to the current payment system which does not attribute variation to specific parameters such as physician supply or beneficiaries' incomes. Thus, the Direct Payment Method could either permit more careful decisions about allowable geographic variations or invite politics to interfere with sound policy decisions. The further development of the Direct Payment Method should be viewed as an alternative to making political judgments that refine the Medicare+Choice system but may not be easily defended. Whether or not having this alternative is worth the required effort will ultimately depend on assessments of how well it might do at correcting problems within the current system.

This report is available in its entirety in the Portable Document Format (PDF).

The nonpartisan Urban Institute publishes studies, reports, and books on timely topics worthy of public consideration. The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders.

Usage, posting and reprint of materials on the UI web site:

Most publications may be downloaded free of charge from the web site in PDF format. This information may be used and copies made for research, academic, policy or other non-commercial purposes. Proper attribution is required.

Copyright of the written materials contained within the Urban Institute website is owned or controlled by the Urban Institute. Posting UI research papers on other websites is permitted subject to prior approval from the Urban Institute—contact paffairs@urban.org.

If you are unable to access or print the PDF document please contact us or call the Publications Office at (202) 261-5687.