INTERNATIONAL MEETING: HEALTH OF PERSONS WITH ID SPONSORED BY THE CDC AND AUCD Anita Yuskauskas, Ph.D. Centers for Medicare & Medicaid Services CMSO Disabled & Elderly Health Programs Group February 24, 2010
MEETING PARTICULARS TOPICS FOR DISCUSSION PROBLEMS TO BE ADDRESSED 1. What data CMS has on health status of persons with disabilities 2. What additional information CMS would like to have 3. Helpfulness of additional surveillance on health status for persons with ID? 1. Health Data on the ID Population 2. Methodological Challenges 3. Use of Resources for Guidance 4. Address an Approach for Accessing #1
#1. CMS DATA ON HEALTH STATUS OF INDIVIDUALS WITH DISABILITIES Medicare health status info on dual eligibles, hospitalization admission, hospitalization discharge Medicaid MSIS claims, encounter data (??); DRA example Medicaid LTC-HCBS - web-based app; MFP
MEDICARE CLAIMS DATABASE PARTS A, B & D Medicare Database is complicated; requires expertise Requires Coordination of benefits Agreement between CMS & SMA - SMD Letter 10-08 Access involves multiple parties : Contractor, two CMS Groups Hospitals provide claims information to Hospital Association (date, primary/secondary payors, diagnosis, gender) States access hospital discharge data from State Hospital Associations
MEDICAID CLAIMS DATABASE: MSIS Medicaid Statistical Information System (MSIS) is a uniform data system that includes Medicaid eligibility and claims records for all Medicaid enrollees from all states. Since 1999, all states have been required to submit MSIS records quarterly.
MEDICAID CLAIMS DATABASE: MEDICAID ANALYTIC EXTRACT (MAX) The quarterly MSIS files are transformed into calendar year person- and claim-level data files. Person-level data includes individual eligibility records, demographic information, eligibility codes and managed care enrollment by month, and total payments for the year Claims records are organized by date of service into four separate files: inpatient, long-term care, prescription drugs, and other services
MEDICAID ANALYTIC EXTRACT (MAX) MSIS and MAX data are subject to extensive data quality checks before they are released for research purposes.
EXAMPLE: MAX FILES USED FOR RESEARCH PURPOSES - DRA MEASURES MANDATE Quality of Care Measures Section 6086(b) of the DRA Directs AHRQ, in consultation with stakeholders to: develop indicators/measures with respect to HCBS offered under State Medicaid programs. Directs the Secretary to: Use measures to assess outcomes (including health & welfare ) and overall system Make available best practices and comparative analysis of the system in each state By September 2010 and for $1 million
DRA HCBS MEASURES MANDATE: PROJECT CONCEPT & PLAN Goal assess the health and welfare of HCBS participants How: Data source: hospital discharge (claims/encounter) data - used as a window into the community Based on tweaking AHRQ developed/nqf endorsed measures Uses the established AHRQ quality indicator (QI) development process
AHRQ S HCBS INDICATORS PROPOSED MEASURE SET 1 Monitoring for the Provision of Evidence-Based Preventive Care to HCBS Medicaid Persons Hospital admissions that evidence suggests could have been avoided, at least in part, through high-quality preventive care Will evaluate existing AHRQ QI measures & modify as needed E.g., exclude from denominator all persons admitted to hospital from LTC facility E.g., evaluate necessity of HCBS-specific risk adjustments Example of indicators: Admissions to hospital for short-term diabetes complications (PQI 1) Admissions to the hospital for dehydration (PQI 10)
AHRQ S HCBS INDICATORS PROPOSED MEASURE SET 1 (CONT D) Monitoring for critical incidents, with hospitals used as data collection agencies for consistent data across states: Intentional injuries caused by others (e.g., abuse, rape, assault, etc.) Unintentional injuries potentially due to neglect Medication errors, pressure ulcers and accidents (fire arms, exposure, serious falls, etc.) Intentional self-inflicted injuries & other behavioral health events Suicide, serious & persistent mental illness, substance abuse
DRA-HCBS DATA SOURCES Medicaid Analytic extract (MAX) 2005 use to: Create the denominator of all types of HCBS populations For Medicaid only HCBS populations, the numerator for the indicators (version 1) (derived from hospital inpatient claims/encounter data) Identify important HCBS sub-population (ID/DD, elderly, etc.) for risk adjustment/stratifications and validation efforts Medicare 2005 For the dually eligible, create the numerator for the indicators (version 1) (derived from hospital inpatient FFS claims)
MEDICAID ANALYTIC EXTRACT (MAX) FOR 2005 Available MAX data files Personal summary record (PS) - Demographics, eligibility, summary utilization and expenditures Inpatient record (IP) - Hospital: Acute, psychiatric, rehabilitation Long-term care record (LT) - Nursing facility (NF) & ICF-MR - Drug record (RX) Other services record (OT) - All services other than IP, LT and RX - Community based LTC - Assisted living and (other non-institutional) residential care services
MEDICAID ANALYTIC EXTRACT (MAX) FOR 2005 New 2005 MAX data elements Medicaid Waiver Group Individual s monthly enrollment in up to three (3) waivers Waiver type 1115, 1915(b), 1915(b/c), 1915(c) 1915(c) population o Aged, aged and disabled, physically disabled, people with brain injuries, HIV/AIDS, intellectually disabled/ developmentally disabled, mental illness / serious emotional disturbance, technology dependent / medically fragile, unknown Waiver identifier Linkable to state MSIS waiver crosswalk data
AHRQ: WHAT HAVE WE LEARNED? Much of the information gathered reflects the application of these indicators to the general population Special consideration should be paid to case mix of HCBS population Patient risk factors, including socioeconomic status and health behaviors may impact rates, but risk adjustment should be carefully considered This indicator set is best viewed as a starting point with the feasible data, but other indicators may prove more useful Some indicators may be too rare to be useful Work underway to learn about variation in indicator rates at the state level, and at the subpopulation level, in order to guide possible appropriate uses for the indicators.
MEDICAID DATABASE: HCBS - MFP QUALITY OF LIFE Longitudinal quality of life survey on all participants tracking QoL changes from the institution through the first two years in the community Domains include living situation, choice and control, access to personal care, respect/dignity, community integration/inclusion, overall life satisfaction, health status.
MEDICAID DATABASE: HCBS WEB-BASED APPLICATION FOR SECTION 1915(C) WAIVERS Helpful when used with other sources of information Population, services, provider qualifications, service delivery models, Measures re assurances (no comparisons): service planning, health and welfare, 50% of approved Waivers are on the web-based format See www.hcbs.net
MEDICAID DATABASE: CHALLENGES IN HCBS Complications of Creating Data Sets in HCBS Wide Variety Of Diagnostic Categories in LTC Wide Range Of Settings Wide Range Of Service Provider Types And Qualifications Wide Range of Measurement Sets: No Standardization No Standard Treatment Intervention, i.e., service definitions & service delivery models
DEFINING POPULATIONS WITH ID Subset of Developmental Disabilities Group Brain Injury Dementia Stroke Subset of Mental Illness Subset of Medically Fragile Conditions Many Are Served Through HCBS
#2 WHAT ADDITIONAL INFORMATION WOULD CMS LIKE TO HAVE ON HEALTH STATUS OF PERSONS WITH ID? Use What s There or Create Anew? Need Better Access & Utilization of Existing CMS Data Expertise with CMS administrative data to know what is possible; Huge databases require sophistication, time, money Implementation of 10-08 SMD letter to access Medicare data: data use agreements follow up Change or Continue Reporting Requirements? Standardize or Vary by Service Delivery Model: Managed Care - Encounter data is limiting; Impact of EHR Comparison Across ID Populations or With Other Groups?
#3 HELPFULNESS OF ADDITIONAL HEALTH SURVEILLANCE ON PID HELPFUL.BUT CHALLENGING Focus beyond Q of L; learn more re health care needs Improved Access to Data Improved Comparisons with More Easily Accessible Data (HCBS, MAX) Recovery Act: EHR or PHR? Incentives for Meaningful Use Measures Factors Influencing Healthcare: HCBS, Medical Home Sustainability Over Time Cost for Maintenance Integration with Other Databases Source of Data: Physicians, Individual/ family? Etc.
THANKS TO: DEB Potter, AHRQ, Fall 2009 Meeting of the HCBS Stakeholders Group for Quality Indicators (QI) Development December 9, 2009,