Appendix 9: National Committee for Quality Assurance Presentation to the Subcommittee
Sarah Hudson Scholle and Sepheen Byron made the following slide presentation to the Subcommittee entitled "Preparing for Measurement Selection: Landscape of Measures." Select to access the PowerPoint® presentation (700 KB). Free PowerPoint® Viewer (Plugin Software Help).
Preparing for Measurement Selection: Landscape of Measures
Sarah Hudson Scholle
- Performance Measurement in Medicaid
- Quality of Care for Adults in Medicaid
- Measures Inventory
- Key Challenges
Performance Measurement in Medicaid
State of Measurement in Medicaid
- There is no national reporting of Medicaid quality data representing all different populations enrolled (that's why we are here...)
- Two new reports shed light on current efforts:
- Managed care: NCQA's Medicaid Benchmarking Project Report
- FFS: CHCS' Performance Measurement in Fee-for-Service Medicaid: Emerging Best Practices
NCQA Medicaid Benchmarking Rep
- Purpose: Test the feasibility of collecting comparable performance measure results from state Medicaid agencies and combining these data with existing HEDIS data in NCQA's database to develop robust benchmarks for Medicaid.
- Why Focus on Managed Care and HEDIS?
- 71% of the Medicaid population in states that use managed care arrangements including PCCM and MCOs.
- 37 states contract with MCOs.
- Nearly 90 percent of state Medicaid programs reported using HEDIS measures for evaluate quality of children's care:
- No comparable data available for adults.
How Do States Use HEDIS?
- States use HEDIS measures to meet the federal requirements for performance measurement in Medicaid.
- States may use the HEDIS data plans have submitted to NCQA, require plans to submit data directly to the state or the EQRO, or calculate performance rates themselves.
- Twenty-five Medicaid programs use or require NCQA Accreditation.
Medicaid Programs & HEDIS
|States without Medicaid managed care plans (excluded from study)
||Alabama, Alaska, Arkansas, Guam, Idaho, Iowa, Louisiana, Maine, Mississippi, Montana, New Hampshire, North Carolina, North Dakota, Oklahoma, South Dakota, Virgin Islands and Wyoming
|States where all health plans submitted HEDIS data to NCQA
||California, Colorado, District of Columbia, Kentucky, Maryland, Michigan, Nebraska, New Mexico, Tennessee, Virginia and Washington
|States where some health plans submitted HEDIS data to NCQA
||Indiana, Minnesota, New Jersey, Pennsylvania, Puerto Rico, Rhode Island and West Virginia
|States where no health plans submitted HEDIS data to NCQA
||Oregon, South Carolina, Vermont
|Total Medicaid Programs
*Includes the District of Columbia, Puerto Rico, Guam and the Virgin Islands.
Most Commonly Used Measures
- Well-Child Visits in the First 15 Months
- Well Child Visits in the Third, Fourth, Fifth and Sixth Years of Life
- Adolescent Well-Care Visits
- Childhood Immunization Status
- Cervical Cancer Screening
- Prenatal and Postpartum Care
- Use of Appropriate Medications for People With Asthma
- Comprehensive Diabetes Care (CDC) - HbA1c Testing
- CDC - Eye Exam (Retinal) Performed
- CDC - LDL - C Screening
- Follow-Up After Hospitalization for Metal Illness
Most Common Differences Between State Measures and NCQA HEDIS data
- Specification changes
- Continuous enrollment
- Measurement year
- Data source
- Numerator changes
- Data collection process
CHCS Report: Performance Measurement in FFS Medicaid
- "Just do it"
- Key Themes:
- Involve providers and other relevant stakeholders
- Clarify the purpose of measurement:
- Reporting and comparisons among delivery systems
- Quality improvement
- Set clear goals for public reporting
- Value the role of leadership in the process
Measures and Data Sources
- Consider measures that rely on administrative data for ease of capture.
- Consider business case with focus on overuse measures, such as hospital readmissions.
- Adapt HEDIS measures to fit the FFS environment:
- Look outside HEDIS for special populations like mental health.
- Consider other data sources:
- Patient/family surveys, Registries, Lab test results, Chronic disease and obstetrics assessment forms, Health information technology.
Resources and Time
- Consider resources needed for development and implementation of the measurement system:
- Many variations exist depending on structures and resources available within states:
- Be patient:
- Expect it to take a year from the start of developing a new measure to reporting it, depending on the complexity of the measure and the availability of analytic capacity.
- Devote resources to auditing measures.
Performance in Medicaid
Status of Health Care Quality in Medicaid
- 2009 HEDIS provides window on national performance among managed care organizations (MCOs).
- HEDIS performance rates for Medicaid MCOs are often lower than for Commercial and/or Medicare MCOs.
- There are a few exceptions...
2009 HEDIS Performance for Medicaid vs Other MCOs
|Adult BMI Assessment
|Breast Cancer Screening
|Cervical Cancer Screening
|Initiation of Alcohol/Drug Treatment
|Follow After Mental Health Hosp (30 days)
|Persistent Beta Blocker Use After Heart Attack
|Diabetes: A1c Screening
|Diabetes: Poor A1c Control (>9.0%) (lower=better)
|Diabetes: Cholesterol Screening
|Diabetes: Cholesterol <100
|Hypertension: Blood Pressure <140/90
|Asthma: Appropriate Medications
Childhood Immunization Retreat in Private Plans, But Not in Medicaid
Image: A line chart shows the rise in childhood immunizations for both private insurance and Medicaid from 1999-2009, until 2008-2009 when the percentage of immunizations for private insurance drops of by several percentage points.
Notes: The most surprising result was a disturbing 4% drop in childhood immunization rates among private plans that primarily serve kids in middle-class families. This did not happen in Medicaid plans that serve low-income kids, where we saw a 3% rise, continuing the steady improvement they've been making.
Big Gains in Chlamydia Screening—with Medicaid Plans Leading
Image: A line chart shows a large increase in chlamydia screening from the period 2001-2009 for both private insurance and Medicaid plans. Medicaid plans have a consistently higher percentage of screening than private insurance.
We are also seeing big gains in chlamydia screening, especially in Medicaid plans. This is a common and often silent STD that, left untreated, can cause irreversible damage, including infertility. We're also seeing better scores for diabetic care, prenatal and postpartum care, controlling high blood pressure and other measures.
Using the Iinventory to Identify Potential Measures
Creating Measures Inventory
- Sources of measures:
- Measures endorsed by National Quality Forum
- Measures nominated by CMS
- Measures submitted by 15 Medicaid medical directors
- Measures suggested by Panel co-chairs and members
- We attempted to "de-duplicate" the list..."
Contents of Measure Inventory
- Pivot Table (allows identifying groups of measures)
- Definitions of descriptors
- Sorted by Measure Steward
- All measures have unique "ID number" for searching (NQF ID provided if relevant)
- Measures characteristics
- Information on current use
- Category in Revised IOM framework
- Population of interest
- Excel makes sorting and filtering of measures possible
Characteristics of Each Measure
- Measure ID
- NQF ID
- Measure owner/steward
- Measure name
- Measure description
- Specific conditions
- Condition type
- Measure type
- Data sources
- Unit of measurement
From the Legislation....
The Secretary shall identify and publish a recommended core set of adult health quality measures for Medicaid eligible adults in the same manner as the Secretary identifies and publishes a core set of child health quality measures under section 1139A, including with respect to identifying and publishing existing adult health quality measures that are in use under public and privately sponsored health care coverage arrangements, or that are part of reporting systems that measure both the presence and duration of health insurance coverage over time, that may be applicable to Medicaid eligible adults.
Descriptors of Current Use
- Medicaid: in use by Medicaid programs or health plans in the state
- Other: in use by other federal programs (VA, Medicare Advantage, PQRI, etc)
- Any: in use in either Medicaid or other program
- States: list of states in which the measures are used
||Adult <65 yrs
||Complex Health Needs
||Mental Health and Substance Abuse
|Patient & Family Centered
|Health Systems Infrastructure
Information for Each Measure
- Patient & Family centered
- Care coordination
- Health systems infrastructure
- Condition Type
- Female Only
- Reproductive Health
- Adults <65
- MH & Sub Abuse In Use
- Functional status
Starting Lists for Each Workgroup
- Maternal/reproductive health:
- Female only and in use at all
- Adult health:
- Adults <65 and in use in Medicaid
- Mental Health/Substance Use:
- MH&SA and in use anywhere
- Complex conditions:
- Cross cutting measures that are in use at all: functional status, care coordination, health system infrastructure, avoidable hospitalizations
Key Challenges in Measuring Quality for Adult Medicaid Populations
- Current measures do not address needs of complex populations
- Examples of measure concepts and issues in selecting measures for the core set:
- Avoidable hospitalizations
- Care Coordination
- Functional status
- Avoidable hospitalization measures, including hospital readmissions and admissions for ambulatory care-sensitive conditions (ACSC), are important markers of waste.
Examples of Avoidable Hospitalization Measures
||Plan All-Cause Readmission (new for 2011)
||30-Day All-Cause Risk Standardized Readmission Rate Following Heart Failure Hospitalization (risk adjusted)
|State of CO
||Number of admissions for Ambulatory Care Sensitive Conditions for waiver and Medicaid clients
||Diabetes Short-Term Complications Admission Rate/100,000
- Key issues for Re-admission measures:
- Population: all age groups, adults only, etc
- Hospitalizations: specific-cause discharges versus all-cause discharges.
- Counting of readmissions: all-cause readmissions or specific-cause readmissions
- Readmission timeframe: 30 days versus 3, 7, 14, 90, ...
- Risk adjustment
- Continuous enrollment
- Key issues for ACSC Admissions:
- Eligible population
- Risk adjustment
- Continuous enrollment
- Outcome measures include mortality and functional status.
- Patients/families value these measures in particular.
- These measures may reflect the net result of care for multiple conditions and care received from multiple providers and settings.
||Improvement in bathing among home-based care recipients
|Focus on Therapeutic Outcomes, Inc
||Functional status change for patients with knee impairments
||Change in Basic Mobility as Measured by the AM-PAC
||Medicare Health Outcomes Survey (HOS)
||IQI 11: Abdominal Aortic Artery (AAA) Repair Mortality Rate (risk adjusted)
||IQI 17: Acute Stroke Mortality Rate
- Key Issues:
- Population/Site of care/Population size
- Cross section versus longitudinal assessment
- Risk adjustment
- Data source and completeness
- NQF (2006) identified dimensions of care coordination including:
- The need for a medical home,
- Proactive plan of care and follow-up
- Strategy for communication,
- Availability of information systems to support care
- Process for transitions or "hand-offs" (across providers and settings)
||30-Day Post-Hospital AMI Discharge Evaluation and Management Service Measure (In proposed IPPS rule May 2010)
||Care for Older Adults (COA): Functional Status Assessment
||Medication Reconciliation Post-Discharge (MRP)
||Management plan for people with asthma
||Advance Care Plan
||Documentation and Verification of Current Medications in the Medical Record
- Key issues:
- Availability and use of measures
- Data sources/completeness
- Feasibility and cost of measurement
Return to Contents
Proceed to Next Section