An information system model for negotiating capitation contracts. (Surfing the Information Technology Wave).

THE GROWTH OF MANAGED CARE HAS PROFOUNDLY changed the environment in which health care organizations deliver care. While hospitals have adapted to managed care by accepting discounted fees, many health plans increasingly seek to shift more financial risk to providers by using capitation-based provider payments. Under capitation, hospitals and physicians receive a fixed, per capita monthly payment in exchange for agreeing to provide a specified set of services. It Is critical to understand what services are included In the capitation payment, what (if anything) can be billed separately, and what is not covered as a benefit. Providers are increasingly influenced by capitation-based payment systems.

Understanding whether a capitation contract will allow a hospital to remain financially viable while providing quality care requires estimating the number of patients to be served, the amount and types of services to be offered, and the cost to provide them. The negotiator must consider many issues when bidding on such a capitation contract, including patient characteristics (demographics and lifestyles), costs, and potential revenues.

In order for capitation to be a reasonable risk, the covered population must be large enough to represent a stable statistical sample pool. To predict the volume and mix of services to be covered under a capitation contract, providers need data not only on the current practice but also on the managed care plan's patients, including demographic characteristics (e.g., age, gender, health risks), fertility rates, morbidity rates, utilization history, and pre-existing conditions. If the plan is unable to supply utilization rates for its covered population, providers will need to estimate rates based on their current patients' utilization patterns and the differences between them and the capitated plan's patients. Capitation payments that are based on inaccurate assumptions about the size of the patient population or their utilization rates for covered services can result in financial disaster.

To determine whether to accept a capitation contract, a health care facility must know its costs. Fixed costs include rent, utilities, administrative, professional and staff payroll expenses, liability insurance, and equipment rental or purchase--all expenses that do not change regardless of the volume of service provided. Variable costs change based on the amount of service provided and include items that can be directly linked to the volume and intensity of specific services. The fundamental question of capitation is whether the reimbursement will cover costs for the services provided while supporting the organization's mission and vision.

The utilization patterns of current patients contribute to the revenues of all health care providers. As managed care progresses, revenues for treating patients will come from fixed capitation payments. The health care provider must, therefore, implement a sophisticated information system capable of performing proforma, profit and loss analyses of capitation contracts. Capitation has left most health care facilities unprepared for negotiation. This highly sophisticated system generally has inadequate information systems to determine the costs of the medical care rendered.

The nature of capitation contracts requires organizations to develop more advanced mechanisms to determine cost per patient in order to improve the ability to negotiate with health care brokers. The current system may place the entire health organization in jeopardy due to the potential of underbidding its costs which, in turn, may cause a significant revenue drain; conversely, the organization that overbids for contracts will most likely not obtain the patient group. Thus, an accurate assessment of patient costs must be determined in order to maintain a competitive position in today's fierce market environment.

One hospital's story

Hutzel Hospital, a 331-bed facility within the Detroit Medical Center, is affiliated with Wayne State University School of Medicine. The providers are comprised mainly of physicians and researchers in obstetrics and gynecology, orthopaedic surgery, and ophthalmology. While the hospital is a part of a larger organization, Hutzel Is an independent health care facility with a strong financial basis from many revenue sources, such as direct patient reimbursement, fee-for-service insurance payments, and HMO contracts. While capitation contracts represent less than 10 percent of total revenue, they are rapidly becoming a larger source of income. Administration officials estimate that capitation contracts will comprise more than 40 percent of total revenues within the next five years.

The context level data flow diagram for negotiating capitation contracts at Hutzel Hospital can be seen in Figure 1.

Historically, the hospital operated on a fee-for-service model--the more it did, the greater the reimbursement. Then, Diagnostic Related Groups (DRGs) were introduced, with reimbursement rates determined for specific services rendered. Finally, capitation provided all funds paid up front and all services delivered as costs against fixed revenue from the capitated compensation pool.

A patient visiting Hutzel Hospital typically goes to the admitting department and completes a form; demographic information is entered into a computer database. A history and physical examination are performed and the patient's condition is diagnosed; diagnostic codes and any tests are entered into the medical record. After treatment, procedure codes are entered. If the patient has surgery, the operating room time is included. Labels from supplies that are used during the patient's stay are either manually placed or bar coded on the face sheet of the medical record and charge slips are registered, although the staff sometimes fails to perform these tasks. When the patient is discharged, the number of days spent in the hospital becomes part of the medical record.

The patient identifier, diagnostic and procedure codes, and days spent in the hospital are computerized in the medical records database; any additional information is stored in the medical records department. Supply label codes, diagnostic and procedure codes, operating room time, and room charges are transcribed into the computerized patient billing database, which is part of a larger financial database that includes general ledger and payroll. Hutzel Hospital's admitting, medical records, and financial computer systems are independent, yet a comprehensive information system is being installed.

In addition, because of its outdated information systems, the hospital does not have the capability to directly and accurately aggregate its cost information in the ways needed to analyze quotations for bids. To accurately negotiate for capitation contracts, true patient costs need to be determined.

Problem analysis

The information systems at Hutzel Hospital do not adequately support capitated agreements. The negotiators do not use adequate information to analyze bid quotations. The hospital is at high risk of losing money if the negotiators underestimate the costs for the patient population. The problems include:

1) Relying on procedure and diagnostic codes

2) Assigning prices, not costs, to the patient database

3) Assigning supply prices, not costs, to specific patient databases inconsistently

4) Segmenting the patient population into specific demographic characteristics

5) Aggregating patient data

The alternatives?

The following solutions have been developed to correct the information system problems.

A. Negotiations for capitation contracts typically occur yearly. Thus, the negotiation team does not need updated financial, patient demographic, and medical records data until it is time to analyze quotations for a bid. Alternative A involves annually downloading information from the admitting, medical records, and financial computer systems into a database that can then be queried. As bidding for capitation contracts grows, the number of times information is downloaded will also need to increase.

This system is better than relying strictly on diagnostic and procedure codes. The negotiation team can segment the covered population into potentially low and high-cost patients using financial, demographic, and clinical data. However, it will not allow for an understanding of what actual expenses are for the care received. The financial system can only determine the fees that will be charged to the patients and/or their insurance companies for the services rendered. While this alternative may give a relative value for the procedures and/or medical treatment received by the patient so that the negotiation team is able to perform case-mix analyses, it is inadequate to access the financial requirements of a capitation contract. Also this system does not address the inconsistent allocation of supplies to the patient database.

B. Alternative B involves adapting the hospital's system by annually downloading information from the admitting, medical records, and financial computer systems into a database that can be queried by the negotiators. However, this choice requires implementing a new cost accounting system that is sufficiently detailed at the procedure level to: (1) aggregate cost information in different ways: (2) measure fixed and variable costs separately; and (3) provide accurate cost information. This system would serve the hospital well by providing it not only with accurate cost data to analyze bids, but also with a way to evaluate and better manage its costs so that it can enter into capitation contracts profitably. There are cost accounting packages available on the market.

The ability to aggregate cost information in different ways, such as by grouping individual patient costs by product line (i.e., obstetric care), will assist the negotiators in performing a case-mix analysis to predict what the hospital's costs would be if it entered into a capitated agreement. Also, the negotiators would be able to segment the population into low and high-risk patients. Measuring fixed and variable costs will help the hospital in controlling expenses and in predicting the marginal expenses of providing service to one more patient. The cost information, however, must be accurate indirect costs, such as support and general overhead, must be properly allocated to procedures to ensure that any managed care negotiations are successfully carried out. While this system allows for supply costs to be assigned per patient, it does not address the staff inconsistencies.

C. Alternative C includes a totally integrated hospital information system that combines demographic, clinical, and financial information. The system would consist of a relational, enterprise-wide database and a wide-area network linking workstations. Admitting would be automated; medical records would become electronic; the financial systems would include cost and accounting and general ledger and payroll. The cost accounting system would be similar to the one described in alternative B. Users and/or applications would be allowed to access only that portion of the database defined by a particular external view to ensure security and patient privacy. Information systems personnel would need to assess which parts of the existing systems can be adapted for use in this model.

By extracting detailed demographic, clinical, and financial information from the database and manipulating it to understand what the hospital's costs would be if it entered into a capitated agreement, the negotiators should be able to properly analyze their quotes for bidding. This system also allows for segmenting patients into low and high-cost groups, but does not address the inconsistent assigning of supplies used per patient. This information system is ideal because it is capable of supporting capitation, as well as fee-for-service and other payment methods.

To become more cost competitive in an increasingly managed care environment and to remain a financially viable health care facility, the hospital must have access to detailed cost information to facilitate cost cutting and financial management. This proposed model provides that access. It allows the hospital to aggregate cost information in different ways, such as collecting individual patient expenses by treating physician to evaluate practice patterns. Because the information system allows for growth, systems for clinical pathways, case management, and enterprise-wide scheduling can be added later to manage care more cost effectively and efficiently.

Conclusion

Managed care is producing a highly competitive environment, which means that it is critical that health care organizations not only know their costs, but also manage them to provide the most cost-effective care possible. The analysis shows that alternative C is the best long-term solution for maintaining the hospitals competitive stature. The projected costs to fully implement this system are in the millions of dollars. While this is a substantial investment, implementing this proposal would be completed in a multi-year, phase-in process. By providing an information technology architecture that permits growth, this system will allow the hospital not only to know its costs to better negotiate capitation contracts, but also to eventually manage its expenses more effectively.

Acknowledgment 

The author would like to acknowledge the assistance from Anne Mobley, MBA, and Michael Felberg, MBA, from the University of Michigan Business School, Ann Arbor, and Franklin D. McDonald, MD, who recently passed away and was the past Senior Vice President of Clinical Affairs at Hutzel Hospital and Professor of Medicine at Wayne State University School of Medicine In Detroit.

Scott B. Ransom, DO, MBA, FACPE, FACOG, FACS, CHE, is Associate Professor and Director of the Division of Community Programs and Health Effectiveness in the Department of Obstetrics and Gynecology at Wayne State University School of Medicine. In addition, he serves as the Clinical Director of Clinical Resource Management at the Detroit Medical Center. He can be reached by calling 313/577-1222, via fax at 313/577-2045, or via email at [e-mail].