Printer Friendly

Medicaid and third-party liability: using information to achieve program goals.

Medicaid and third-party liability: Using information to achieve program goals


The decade of the eighties produced a number of significant changes in the Medicaid program. Increased numbers of recipients, rising costs, and many other health care issues have had a major impact on the program. As a result, an awareness has grown that increased control over the direction of the program is necessary. A key element of management control is the availability of information that allows the analysis of options for informed decisionmaking. Thus, it is no coincidence that establishment or enhancement of Medicaid Management Information Systems (MMISs) continues to be encouraged by the Health Care Financing Administration (HCFA). Many States have also established eligibility information management systems for more effective administration of the various public assistance programs. As a result, a great deal of information has become available to managers that can be used to shape program goals and to assess trends and progress.


Third-party liability (TPL) has come to play a more prominent role in the drive to better control medical assistance costs. Historically, utilization of third-party payment sources has been shown to be a very cost-effective method of reducing Medicaid costs. The relatively small administrative cost investment in the eligibility determination process and in efficient management of benefit recovery can help control program cost outlays and/or provide revenues to help fund program eligibility and/or benefit expansions.

In New York, the emphasis on TPL began in the early eighties just as statewide implementation of MMIS was being completed. The New York Medicaid program is locally administered, with the State Department of Social Services (DSS) serving as the central administrative agency. Prior to MMIS, Medicaid claims payment was performed by each local district in New York (there are 58, including New York City). The transition to a centralized payment system allowed counties to reassign staff to other areas one of the priorities recommended strongly by the State was the establishment or enhancement of the TPL function. Each county was encouraged to set up a third-party specialist unit, with the State providing training and other support services. There are now approximately 200 full-time equivalent staff assigned to this function at the local level throughout the State.

Use of information

One of the key problems affecting program managers is how to meet new demands using the resources at hand. Not many States have the luxury of sufficient staff and other resources to respond to the growth and complexity of public programs. Even when resources are available, there is always the challenge of using them effectively by selecting the most viable options for resource investment. Clearly, the power of information can guide productive decisionmaking, if existing information is used in new and innovative ways. Two good illustrations of this type of opportunity in New York are: first, the use of information to sensitize local social services officials to the importance and impact of third-party resources and, second, the innovative use of expenditure data to facilitate decisions as to whether it is cost effective for Medicaid to pay for an applicant's health insurance policy or allow it to lapse.

Third-party liability management


Because the New York Medicaid program is locally administered, there is a significant problem in attempting to channel county priorities and resources to achieve statewide goals. In this case, the goal was to maximize Medicaid program savings through the detection and use of other health insurance available to Medicaid clients. The exchange of information between local districts and the establishment of specific performance goals was vitally important to this effort. A natural byproduct of this approach was the introduction of a sense of competition among local districts that served as an indirect incentive to increase the detection and use of health insurance. A recurring management report that utilizes key program information and ranks counties by their performance was established: the Statistical Tracking and Reporting System (STARS). An example of a recent report for a county in upstate New York can be seen in Figure 1.

From the large amount of information available centrally, two indexes were chosen that could be calculated from existing data sources within the New York MMIS. The first, "insurance detection," presents the extent to which local eligibility workers identify other health insurance resources. The index chosen (the detection percentage) is the number of individuals with health insurance divided by all people on assistance. That is, the number of recipients with health insurance divided by caseload equals the detection percentage. Subsets of detection percentages that measure performance by the categorical eligibility status of a client in the cash assistance, medical assistance only, and Supplemental Security Income (SSI) areas were also derived. This was done because the various assistance programs upon which Medicaid eligibility is based usually fall under different managerial authorities (e.g., there are usually separate cash assistance and medical assistance directors) and thus may perform quite differently even within a common or shared administrative structure.

Two other indexes were added to help evaluate local district operations. The first is a production statistic that counts the number of third-party transactions generated by a county and then divides that by the number of workers. This allows the State to measure productivity per worker to determine the proper staffing level for that county's program.

The other performance indicator used is termed the "coverage index." This is a measure of how well a local district is entering third-party information into the eligibility management information system. In New York, insurance coverage is indicated by the entry of a series of codes that define the scope of the policy. For example, the codes cover inpatient hospital, physician, emergency room, clinic, drugs, dental, major medical, etc., for a total of 14 possibilities. Although it is difficult to generalize about insurance policies, most people with health insurance would be expected to have coverage for at least three or four service types, even on a basic policy. As a result, the State would expect an average coverage index per entry of at least three or four. One interesting example of the use of this index occurred when one of the counties being monitored had a coverage index of exactly 1.0. Investigation revealed that the county was entering inpatient hospital coverage only, in order to process the information in as little time as possible. The State was able to intervene and correct the situation, with the result that the county now has a coverage index of 3.2, allowing a much wider variety of Medicaid claims to be edited for cost-avoidance purposes. In Figure 1, one can see examples of a recent report for a county in upstate New York.

The use of this performance-oriented management strategy produced some dramatic results. Using Table 1, one can compare the detection percentage information for three countries from January 1984, when we initiated STARS, to December 1989. As can be seen, the use of local feedback provided a strong incentive to improve.

Health insurance premium purchase

New York has been in the forefront of paying health insurance premiums for Medicaid recipients. The State has implemented a computer software program that is used by the local districts to assist them in arriving at the decision to buy health insurance coverage under certain circumstances. This program takes advantage of the fact that applicants for Medicaid often have health insurance available through a current or former employer. In the latter case, there may be conversion rights available to provide coverage on an individual rather than group basis. Given the opportunity to continue the health insurance policy, it then becomes incumbent on the Medicaid program to determine when it is cost beneficial to do so.

Through the early years of developing this program, a consistent problem was that it was often difficult to determine the benefit side of the cost-benefit ratio. The cost side is quite easy, of course, because this can be determined by calculating the premium amount to be paid. Expected benefits have often been subjectively based on qualitative judgments concerning past and future utilization of medical services. Because New York's program is locally administered, these decisions are made on a case-by-case basis by eligibility workers or third-party specialists in the field, in conjunction with their normal job activities. As a result, there often is not time to thoroughly research each case. In these situations, there may be a tendency not to continue the policy because this represents the path of least resistance.

In 1987, a project was initiated within the New York State DSS to try to bring new technology to this troublesome problem by developing a computer program that would accept a number of variables and produce a decision automatically. To make this available to the greatest number of people, a personal-computer-based concept was developed, although the ultimate application would make this capability available through use of the existing network of terminals in local districts that are connected to the central mainframe. It was determined that the key information necessary to make these decisions was:

* Aid category (Aid to Families with Dependent Children, Supplemental Security Income, or medical assistance only).

* Geographic area (upstate, New York City).

* Premium amount.

* Insurance coverage (inpatient, drugs, durable medical expenditures, major medical, etc.).

* Case size.

* Age.

* Gender.

To calculate the cost-benefit ratio, the cost side of the equation is simply the net amount of the premium that would potentially be paid by the State. The benefit side is much more difficult, because with a new Medicaid applicant, specific expenditure history is lacking. It was determined that the best substitute would be Medicaid expenditure averages for individuals or cases with similar characteristics. Using data available from MMIS, expenditure averages were calculated based on aid category, geographic area, age, and gender. Some of the results of this analysis are shown in Table 2.

When data are entered into the program, the total Medicaid expenditure expectation for each individual in the case is calculated. By comparing the amount of expected average expenditures and which of these services are covered by the insurance policy, the value of the policy can be estimated. By comparing this amount with premium cost, a cost-benefit decision can be made.

The result of this process provides a highly automated way to evaluate some of the most routine aspects of premium purchases. However, it is not infallible. If the future utilization pattern is less than average, then a decision to buy can be wrong. If high-cost medical utilization appears to be imminent, or it can be determined that there is a pre-Medicaid history of high medical costs, then a decision not to buy may be wrong. In any case, the computer software has to be used in conjunction with human judgment to optimize the decision to purchase coverage.

Recently, statewide data on the amount, type, and cost of health insurance purchases were made available. This information will be evaluated to determine how cost effective the program has been and where enhancements might be made. Hopefully, this will allow us to fine tune the computer software and further increase the efficiency of the process.

Further third-party liability savings

Modern Medicaid management information systems provide a wealth of information to program managers. Information about other health insurance resources obtained by caseworkers during the client eligibility intake process, if effectively collected and creatively used, can be a major source of TPL savings.

Evaluation of TPL data collection performance can motivate local offices to maintain high quality and productivity in eliciting and reporting relevant, accurate TPL data. New York uses a results-oriented method of measuring the number of known third-party resources and the amount of savings that are attributable to the information received from each county. Because eligibility intake activities and TPL data collection universally rely on geographically dispersed local contacts with clients, the approach used by New York is potentially applicable in other States, whether local offices are under State or, like New York, county jurisdiction.

More and more States are adopting the practice of paying health insurance premiums when it is cost effective to do so. The automated cost-benefit analysis approach used by New York to determine whether to pay health insurance premiums on behalf of Medicaid clients is a particularly interesting application of data available in the State's information systems.

Although this article focuses on two techniques used by New York for enhancing its TPL program through use of information, there are many other innovative practices utilized by other States. The Health Care Financing Administration has issued a publication entitled Third Party Liability in the Medicaid Program, A Guide to Successful State Agency Practices. This Guide provides information on the two New York practices as well as other exemplary practices that could assist States in improving their TPL efforts. The Guide enables State agencies to identify and assess those practices that have proven to be successful and are transferable to their own State operations. Copies may be obtained by writing to: Health Care Financing Administration Medicaid Bureau Attention: Alfred Czerski, Central Office Coordinator Room 273 East High Rise Building 6325 Security Boulevard Baltimore, Maryland 21207
COPYRIGHT 1990 U.S. Department of Health and Human Services
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1990 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Buzzard, Kenneth
Publication:Health Care Financing Review
Date:Jan 1, 1990
Previous Article:Ambulatory practice variation in Maryland: implications for Medicaid cost management.
Next Article:Improving state Medicaid programs for pregnant women and children.

Related Articles
Publication details Medicaid statistics.
Publication details Medicaid statistics.
Medicaid statistical guide available.
The 1999 revised Medicaid third party liability and estate recovery laws: how to satisfy the liens and preserve the client's public assistance.
Specialty Disease Management Services Applauds CMS for Urging States to Adopt Disease Management Programs.
Medicaid Information Technology Architecture: an overview.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters