Establish a Health Care Cost Containment Office

The state should provide purchasers, consumers and providers of health care in Texas with data on the costs and outcomes of patient treatments at different hospitals to reduce health care costs.


Background
Consumers have access to elaborate assessments of which household appliance is most cost effective, but when it comes to buying health care, consumers and major purchasers of health care have little, if any, independent comparative information on which to base their decision. In a market where a patient is separated from choice by his physician, and from price by his insure r, the question of cost-effective treatment is often not considered. Consumer choice is limited and competition not possible as long as comparative information on health care providers is not published for general public review.

Currently, over 30 states offer a variety of models of health care databases as part of their effort to determine the cost and quality of health care in their state. Texas does not have a statewide system for collecting hospital discharge data and by state law cannot publish financial data related to specific hospitals.

Pennsylvania and Iowa each maintain a health care database. In these states, the collection of patient level data to compile databases, which report not only the cost of receiving care for a particular diagnostic-related group but also the outcome of the t reatment adjusted for the severity of the patient s illness, is mandated.

In Pennsylvania, the Pennsylvania Health Care Cost Containment Council collects data on all patients discharged from hospitals in the stat e. In Iowa, it is the Iowa Health Data Commission. All hospitals of 100 or more beds in both states are mandated to provide the data.

In Pennsylvania, consumers and purchasers of health care have access to public information about the costs of procedures and outcomes at hospitals in their area. For example, one could determine that Allegheny General Hospital in Pittsburgh charged an aver age of $95,185 to repair or replace a heart valve during the fourth quarter of 1988 and that, across town, Mercy Hospita l did the same procedure for $48,559. In Southwestern Pennsylvania, the information shows that 14 of 53 patients who suffered a stroke died at a Latrobe Area Hospital, while all 32 similar patients at Armstrong County Hospital survived. 1~

Information from the Pennsylvania database has shown wide variations in the costs of particular treatments, quality and practice patterns within the same localities with no discernible correlation between the cost of the treatment and the outcome of the ca re. The Pennsylvania database has also located areas where residents were experiencing specific illnesses far above average.

Purchasers of health care, such as businesses, labor unions, governmental entities, managed care plans and individual consumers, can use the information to become value purchasers of care and to purchase defined clinical outcomes at affordable prices. In t he absence of such publicly available information, purchasers do not function in a competitive marketplace because they have no systematic way of comparing costs and outcomes of health care services.

Businesses can use the data to constrain their escalating health benefit budgets by purchasing health care value. Patients can receive health care from a more effective health delivery system, better care in the hospital and improved outcomes as compared w ith historical standards. Public health officials can also use it to identify problems.

It can help hospitals improve the quality of their care while reducing costs. For example, data indicated that one Pennsylvania hospital s mortality and major morbidity rates for pneumonia were worse than average. The hospital reviewed its operations and took corrective action. Following these changes, its rates improved. 2~

About 30 hospitals in Texas currently voluntarily contract with a firm to create the same type of database so that they can improve the cost effectiveness and quality of their care.

Patient confidentiality is assured in these models. No individual patient can be identified from the published reports.

Pennsylvania has published Treatment Effectiveness reports for 59 specific types of illness categories. These categories include cholecystectomy (gall bladder), acute myocardial infarction (heart attack), adult pneumonia, cesarean section and coronary by-pass surgery.

Iowa has begun a smaller version of the Pennsylvania program. It has started collecting data on some of these illnesses and will publish its first report soon.

In fiscal 1991, the Pennsylvania Council spent an estimated $2.8 mi llion and had 32 FTEs to coordinate and report on its database. The council contracts with a firm that specializes in this kind of computer-based statistical analysis and also performs some functions in-house.

In addition to the cost incurred by the state, each Pennsylvania and Iowa hospital spent an average of $10 per patient admitted to collect and prepare data, or about four-tenths of 1 percent of total hospital operating costs. Part of this cost was the fee each hospital paid to the contracted firm to process the data they collected, making statistical adjustments for the severity of the illnesses. Some health care purchasers voluntarily agreed to reimburse the hospital for the cost of this charge because they believed this data allowed them to buy a be tter quality of care for less money.

Iowa has a much more modest effort. Iowa s Commission has a smaller budget of $225,000 because they report on fewer diagnostic-related groups.

Texas has over 450 hospitals with a total of 70,656 licensed beds. Administrative costs for a data collection system in Texas would be about $412,000 per year and would cost $7 per bed per year.

The Texas Business Group on Health, representing over 200 major employers, supports the creation of a statewide health care database.


Recommendations
A. The Legislature should create a Health Care Cost Containment Office in the Texas Department of Health to provide data on the costs and outcomes of patient treatments at different hospitals in the state. The Legislature should require hospitals to report the data to the office.

A council composed of purchasers, consumers and providers of health care would provide advice to the office on the type of data and the frequency of reporting needed. The advisory council would provide guidance on such issues as how many and which type of diagnostic-related groups would be reported.

B. To cover the administrative costs of the office, the Legislature should increase the annual or monthly licensing fee for hospitals from $3 per bed to $10 per bed for hospitals with 100 beds or more and remove the cap.

The current hospital licensing fee is based on the number of beds in the hospital and by statute is not to exceed $3 per bed. The total fee cannot be less than $100 or more than $3,000.

C. The state law prohibiting the publication of financial data about individual hospitals should be modified to allow the publication of information about the cost of the treatment of patients in specific diagnostic-related groups to be published.

This would not involve publishing any information about profit or loss, total revenues or spending. The intent is not to provide competing hospitals with information about a hospital s individual financial performance, but to provide consumers and business purchasers with a history of the costs of specific services and outcomes of the treatment adjusted for the severity of the illness of the patient s treatment.


Implications
The State of Texas could use the information discussed above to help in purchasing health c are services for its employees. If the Texas Performance Review recommendation for selective contracting of Medicaid services is implemented, the state could use the information to help select hospitals with which to contract for Medicaid services. Finally , the information could help the state in making decisions related to hospital licensing.

Exempting hospitals with less than 100 beds from the fee increase would help small rural hospitals that have been prone to closure in recent years.


Fiscal Impact
T he current fee collected by the Texas Department of Health for licensing hospitals in the state would be increased from $3 per bed to $10 per bed for hosipitals with 100 or more beds to cover the administrative costs of the new Health Care Cost Containment Office. The fee would cover the cost of an estimated five FTEs and related costs necessary to begin the program. Insufficient data exist to estimate administrative costs to hospitals for reporting data.

There would be no cost to general revenue.


Revenue from Administrative Costs
Fiscal Hospital Out of Hospital Change in
Year License Fees License Fees FTEs

1994 $412,000 $412,000 +5
1995 412,000 412,000 +5
1996 412,000 412,000 +5
1997 412,000 412,000 +5
1998 412,000 412,000 +5




Endnotes
1 Ron Winslow, Data Spur Debate on Hospital Quality, The Wall Street Journal, May 24, 1990.
2 Letter from Barry H. Roth, President, Forbes Health System to Ernest J. Sessa, Executive Director, Pennsylvania Health Care Cost Containment Council, February 22, 1991.