The Public Policy Committee of the Hammond Commission met at the Prince William Community Services Board in the Ferlazzo Government Center in Woodbridge, Virginia on Monday, June 21, 1999. Members present were Dr. Louis F. Rossiter, Professor at Medical College of Virginia, Virginia Commonwealth University and Raymond Burmester, Coalition for the Mentally Disabled Citizens of Virginia. Dr. Rossiter, Chairman, called the meeting to order at 10:10 A.M. and introductions were made. Others in attendance included: Charline Davidson, Director of Planning and Regulations for the DMHMRSAS and liaison to the Committee; Martha Mead, Director of Legislation and Public Relations for the DMHMRSAS; and Lynne Fleming, Office of the Attorney General.

Thomas Geib, Executive Director of the Prince William Community Services Board, welcomed the Committee.

Mr. Lutterman said that the NASMHPD was founded in 1963 and is an affiliate of the National Governors Association. It works with the states and U.S. territories on behalf of their mental health systems. It helps them coordinate federal policy with other national agencies, collect data, and write policies, and facilitates information sharing and technology transfers among states. Mr. Lutterman gave an overview of some of the national initiatives and what states are doing around implementing performance measurement systems for mental health. He said that the public sector is much ahead of the private sector in looking at outcome measures.

• Mr. Lutterman reported that there is currently a lot of interest in performance measurement in mental health. The federal government's Center for Mental Health Services began last year to request that states report on six key indicators as part of their block grant applications. They awarded grants in a pilot study to sixteen states, including Virginia, to help fund the implementation of information systems to collect data and report those measures.

• Accrediting organizations, with the move toward managed care and cost cutting, are looking at performance contracting to monitor the quality of care. Other factors are pressure from consumers and families to provide meaningful choice, quality of services, and continuous monitoring by the system.
• Performance indicators have been proposed to counter the impression that mental health services are not effective and should not be afforded parity in insurance. The National Institute of Mental Health has studies that show certain mental disorders can be treated more effectively than some accepted medical treatments.
• The Health Care Finance Administration (HCFA) has requirements for performance measurement in clinical areas, but not yet in mental health, and The National Committee on Quality Assurance (NCQA) is adding some mental health measures. The Joint Commission on Accreditation of Healthcare Organizations (JCAHO) has an initiative called ORYX, beginning July 1, that requires all accredited organizations to participate in a third party performance measuring system.
• Mr. Lutterman said that there is consensus in the field over the definitions of "domain" and "indicator"; however, there is no uniformity in defining specific measures.
• The NASMHPD President's Task Force on Performance Indicators and Outcome Measures developed in 1997 a standardized framework for the evaluation of public mental health services across the country. The framework consists of five domains: access; quality / appropriateness; outcomes; structure / plan management; and early intervention / prevention. Within each domain, a set of performance measures is also identified. The states were surveyed in the Fall of 1998 to determine the implementation status of these indicators and the associated burden, and every state was implementing some of them. The most common indicator was the consumer survey.
• JCAHO's ORYX initiative integrates performance measures into the accreditation process. Every accredited 24-hour behavioral health organization must collect and submit monthly client-level data to the performance measurement system developed by the NASMHPD Research Institute beginning July 1. The Institute will submit aggregate facility-level data to JCAHO and the hospitals quarterly, and JCAHO will produce performance reports for the states.
• Data must be specifically defined and consistent for reporting, and all states have needed to modify their existing information systems in order to implement the necessary programs. The reports will allow the states to see their own performance over time and to compare their programs with similar programs elsewhere.

Dr. Claiborn said that he was pleased that Virginia is in the forefront on performance measurement. It is a complex field, and the lack of common definitions is a big problem on every level. One of the best benefits from implementation will be to have benchmarks-comparable data for common populations with risk adjustments- within the state or nationally.

• Dr. Claiborn explained that CSBs are the sole publicly funded community service and single point of entry into the system. They are responsible for care management, including assessment, eligibility, authorization and provider monitoring, and this includes contract requirements and licenses. CSBs are accountable for the quality of care, costs, outcomes, and satisfaction to local boards and governments, state government, HCFA, and accrediting bodies.
• Dr. Claiborn said that accountability is currently built around programs or services rather than the clients. He believes that it should move toward more person-centered accountability. He said that accountability programs should be built incrementally, focusing on priorities, clear standards, and economic incentives that have been jointly developed. Dr. Claiborn suggested that the cost benefit for data collection and reporting be considered, and a quality management model be developed that uses periodic sampling rather than studying all of the data.

Ms Garrity explained how Inova Health System redesigned the way the company delivers services to its four hospitals, two long-term care facilities, and a home health care company. The patient care redesign initiative started in 1995 and began implementation in 1998.

• The Conference Model was selected to involve all community stakeholders and build ownership and commitment. Four large group interventions determined the gap between what the customers wanted and what was provided, and designed the blueprint for the new organization.

• Once the design criteria for the delivery system were determined, the following macro outcome measures were adopted: clinical quality, growth, service quality, and cost.
• The new patient care redesign model is outcome-driven and encompasses access management, collaborative care partnership, and quality case management. The number of measures was limited to the most important. The outcome data is collected, analyzed, and used to make recommendations to the care team for practice improvement. There is some reluctance from physicians to share data because of the lack of trust in how the information will be used.
• The nursing care team on the admitting unit handles case management for 90-95% of the patients. The other 5-10% of patients are higher risk and will be managed by advanced practice nurses called clinical practice coordinators. These nurses will also direct any practice changes suggested from the outcome data. INOVA recently hired a medical director for outcome management.
• The redesign model will soon be phased into INOVA’s behavioral health services, and Ms Garrity is interested in the recommendations of the Commission concerning performance measurement.

Mr. Burhop said that working with Virginia Health Information (VHI) has given the ITS Department a different perspective on the data that is being collected, and it would be interesting to compare Virginia’s data with facilities in other states. He said that the initial approach to formatting and reporting the data was to utilize the existing UB-92 billing program that was currently being used in the facilities. The UB-92 is a universal billing code used by the DMHMRSAS PRAIS system. It was found, however, that some of the data values and codes did not match, and there were other elements that did not exist, so DMHMRSAS is revising the system to match the requirements of VHI. Mr. Burhop was asked to confer with ITS and determine if any legislation is needed to fully implement the system and advise the Committee.

Ms Brockwell informed the Committee about the work of Virginia Health Information:

• A pilot project will begin July 1, 1999, when the DMHMRSAS submits state mental health facility patient level data to VHI. VHI will work on edits and corrections with the Department and waive fees during the pilot. She said that VHI hopes to include this data into its Patient Level Database with the other Virginia hospitals by July 30. This Database is distributed quarterly to the Virginia Department of Health and for a fee to vendors and providers, and allows users to compare psychiatric inpatient care across regions and provider types. The system protects the patients’ confidentiality.
• Ms Brockwell said that another dataset would provide readmission and transfer information among psychiatric patients. Users will also be able to compare hospital outcomes, including length of stay, charges, and mortality rates, and with risk adjustments, compare to a norm or benchmark.
• VHI produces consumer guides, and Virginia Hospitals: A Consumer’s Guide is planned for release this summer. This guide will have a description and statistics for every hospital in Virginia and may include the state mental health facilities, if deemed appropriate.
• VHI collects outpatient data from Medicaid and state employees and is conducting some pilot studies, but this area is in the infancy stage.

• The Committee asked that Ms Brockwell and Mr. Burhop work with VHI and the Department of Health to ensure that the DMHMRSAS will receive the quarterly database information.
• The Committee discussed the idea that a group or task force could be formed, possibly by the State Board of Mental Health, Mental Retardation, and Substance Abuse, to provide ongoing guidance in area of data policy.

Dr. Harris and Ms McCall discussed the Virginia Performance and Outcome Measurement System (POMS) pilot project:

• Dr. Harris explained that the purposes of POMS are quality improvement and accountability. The project is a collaborative effort. The POMS Committee includes each disability and consumer advocates, and builds upon national initiatives. The effort has had continuous evaluation and revision as it has progressed.
• POMS was designed to consider cost-benefit ratio for the measures, efficiency, and case-mix adjustment to allow valid comparisons. Initial feedback indicated that time had been underestimated; management, staff, and clinician’s time for collecting and reporting data were high and resulted in increased costs. The project requires standardized assessment instruments, data collection procedures, and software. Dr. Harris said that it has been a challenge to find reliable, valid measures.
• The pilot project of developing and implementing a standardized performance scale is at different stages for each disability population. Phase 1 is completed and involved mental health for adults and children and substance abuse for adults, Phase 2, substance abuse prevention, will be completed soon. Phase 3 involves mental retardation, and the pilot has recently begun. Virginia is a participant in the federal pilot Core Indicators Project for Mental Retardation and is using that work to help develop POMS criteria for MR.
• Some of the challenges that have been encountered are: the lack of resources, especially ITS capacity at the CSBs; the difficulty of obtaining and maintaining consumer involvement; inadequate training; and lack of national benchmarks to compare data.
• Ms McCall explained how the system will be phased in over multiple years, beginning with a limited number of indicators for each population. Phase 1, mental health and substance abuse, will measure the following domains beginning July 1, 2000: access to services; quality and appropriateness of care; consumer outcomes; consumer satisfaction; and seclusion / restraints.

• This year the CSBs are asked to select one of the standardized instruments for tracking consumer treatment outcomes to measure as well as the other indicators. CSBs will be preparing the local operations manuals for POMS and revising the local information systems and programs to collect and transmit the data to DMHMRSAS. Training will be provided by DMHMRSAS.
• The POMS pilot for substance abuse prevention services for school age children is scheduled to be completed in August 1999 and begin implementation in July 2000. The POMS pilot for mental retardation services will begin in July 1999 and is scheduled to be implemented in April 2001.
• Ms McCall said that a lot of work has been done and a lot of work remains to be done before POMS is completely implemented.

The Committee asked Ms McCall and Dr. Harris to estimate the costs involved for the following and inform the Committee:

• Implementation of the MHSIP consumer and family survey with a CSB representative sample; and
• One time start up funding for CSB implementation of POMS.

The Committee also suggested that the proposed funding to CSBs next year for implementation might need to be increased.

Mr. Anderson explained that performance measurement in substance abuse is nationally directed largely because of the Federal Substance Abuse (SA) Block Grant.

• The Block Grant for Substance Abuse is $1.5 billion and accounts for one third of all funds for SA. The Grant given to each state varies from about 10% to 100% of the available funds for SA in the state.
• The federal government has been pressuring the states to adopt Performance Partnership Block Grants based on the Healthy People 2000 format from the Public Health Service. They wanted to set standard objectives based on national baseline data for the states to meet. This initiative would have increased the flexibility the states would have in using these funds, which are currently heavily regulated, but NASADAD opposed it. Mr. Anderson said that there was not adequate data, the methodology was unacceptable for establishing the baselines, and some of the measures were unrealistic and not readily controlled by the state.

• Like mental health, substance abuse treatment deals with a chronic disorder, but unlike mental health, it is treated from an acute care model. This often forces artificial clinical decisions on treatment.

• NASADAD used a state-friendly approach to the development of performance indicators called the Delphi process. A state data advisory group was formed that assisted NASADAD in identifying a list of 49 trial indicators and completed a data dictionary. After state feedback, 19 trial measures were selected. A second Delphi round was used to select 6 measures for the Block Grant application for 2000. These are: criminal justice involvement; employment; homelessness; reduction in alcohol and drug use; discharge status; and length of stay. Compliance will remain voluntary until the data infrastructure is put into place statewide.

• There is currently a treatment outcomes performance pilot study called TOPS2 involving nineteen states. Data is collected at intake, at discharge, and at least once after discharge.
• NASADAD is working with other agencies and the states to use the existing data base (TEDS) and agreed-upon definitions and measures. They are also working closely with the Office of National Drug Control Policy to feed into their broad-ranging performance effectiveness measurement system. NASADAD will advocate with this Office for funding to develop the data infrastructure capabilities within each state. The data infrastructure needs assessment will be conducted through the Delphi process.
• Measures for Substance Abuse Prevention are very fluid at this time given the difficulty in developing data collection processes and measures for achieving a negative. Most measures currently in use are population-based. Some of those proposed are: age at first use; intent to use; perceived risk; and expectations of others.
• Washington State is leading in performance reporting because they have linked databases with mental health, Medicaid, and other agencies' systems. They use seven unique identifiers, including social security numbers, to access data across local and state agencies. NASADAD is assisting other states with similar initiatives. The linking technology is posted on a web site for use by the states. Technical assistance is available for community based organizations through the National Leadership Institute to help providers develop performance measurements for their own use and to participate in the state system.
• Mr. Anderson said that he was very impressed with where Virginia is in the process.

The Committee asked Mr. Anderson to forward information on the relationship between state funding for SA treatment and the spectrum of the criminal justice system. They also asked for information concerning Medicaid coverage for SA services in other states.

After discussion of the topic areas addressed by the Committee, members discussed their ideas for recommendations. Staff was asked to pull together recommendations based on this discussion for the July meeting.

The next meeting will be on July 16 at Central State Hospital in Petersburg and will address care management. Bob Hurley from Virginia Commonwealth University will attend to discuss the approaches and models used in other states. Other presentations will be by the Virginia Hospital Association and representatives from Trigon and Sentara, a panel of advocacy organizations, the Autism Program of Virginia, and Diana Thorpe from DMAS.

The minutes of the May 25, 1999 meeting were approved, and the meeting was adjourned at 4:15 P.M.