Identifying Health Care Quality Measures for Medicaid-Eligible Adults: Background Report
5. Recommendations for Future Core Sets and Measurement Refinement and Development
Table of Contents
During discussions within the workgroups and among the Subcommittee as a whole, members offered ideas for how to identify priorities for future iterations of the Initial Core Set; how to improve current measures; and suggested strategies for collecting, aggregating, and reporting data that may extend the Initial Core Set's usefulness and potentially minimize burden for States.
The Subcommittee made recommendations regarding the process of measure identification, development, and enhancement. First, they encouraged the involvement of stakeholders throughout the measure development and enhancement process so that new measures developed will provide information relevant and meaningful to them. Second, they recommended a focus on measures that will lead to improvement in performance and health outcomes. Finally, they recommended looking more broadly to measures that can be used in other settings (e.g., long-term care services and supports as well as home and community-based services).
Future Content Areas
The Subcommittee highlighted topic areas and particular Medicaid populations for which measures were lacking. The Subcommittee also identified topic or clinical areas where measures currently exist, but potentially could be further honed and included in future Core Sets.
In general, the Subcommittee noted the importance of a continued focus on specific adult Medicaid beneficiaries: women, individuals with mental health/substance use problems, individuals with complex health care needs, frail elderly, and individuals with disabilities. While measures exist for many conditions common across Medicaid adults (e.g., diabetes), they are not typically specified for individuals with severe mental illness and, in fact, often exclude these populations from the denominator. For conditions that are less common in the general population, measures often do not exist altogether (e.g., dementia, viral hepatitis, Alzheimer's disease). The Subcommittee also expressed a need for more measures that assess the quality and content of care received. Many existing measures focus on the number of visits a patient had (e.g., women who had the requisite number of prenatal and postpartum care visits) but not the quality of care received or health outcomes as a result of the care provided. The Subcommittee acknowledged that information on content and quality of care is difficult to collect using current data sources.
The Subcommittee noted the following important areas for priority populations:
Women's Health. Many measures in the women's health domain focus on access to care (particularly for pregnancy). There are existing measures in use that assess secondary prevention (i.e., screenings) in women, and many of these measures were recommended for the Initial Core Set. However, there is a need for measures that promote the full range of women's health needs, such as:
- Measures addressing pre-conception care, including gynecologic care and screenings for sexually transmitted infections (e.g., human papillomavirus, hepatitis).
- Conditions and procedures that impact women (e.g., hypertension, urinary incontinence, hysterectomies).
- Pregnancy and postpartum care measures that go beyond measuring access to care (e.g., ensuring vaccinations and proper screenings during pregnancy, healthy birth outcomes such as progesterone treatment to reduce pre-term births).
Individuals with complex health care needs, frail elderly, and those with disabilities. Due to the complex management and treatment needs for people with multiple chronic and complex health care needs, in these areas, feasible measures are lacking. Refining existing measures and development of additional measures could address topics such as:
- Age bands and exclusions for measures that exclude dually-eligible beneficiaries or those beneficiaries with long-term care needs and social supports (e.g., consider modifying the measures for the AHRQ Prevention Quality Indicators for dehydration, bacterial pneumonia, and urinary tract infection to include institutionalized populations).
- Conditions and topic areas with a particular impact on these populations (e.g., falls/fractures, pneumonia, quality of life, safety, HIV/AIDS screenings).
- Social supports (access to, quality of, etc.).
Mental Health/Substance Use. As raised by the Mental Health and Substance Use workgroup and other Subcommittee members, measures in this area may be challenged by small denominator issues. However, measures are needed in areas that address:
- Brief interventions.
- Medication treatment for opiate use and alcohol.
- Antipsychotic use.
- Ambulatory sensitive conditions or preventable hospitalizations related to mental health conditions.
In addition, the Subcommittee recommended that future measurement activities consider focusing across populations:
Patient-reported outcomes and risk assessments. This area is one in which measures are minimal due to the difficulty of collecting such information using current methods. In particular, effective measures are needed that can address concepts such as quality of life and choice and autonomy in treatment/setting (e.g., advance care planning). Additional topics that warrant attention in this area include:
- Pain assessment and management.
- Functional status.
- Measures that capture the perspective of family/caregivers.
Medication management and safety. Due to the limitations of existing data sources, many existing medication management measures only assess "prescription dispensed" and may not adequately assess "adherence." Future measurement development topics that may be able to leverage EHRs or other clinically enriched data include:
- Safety of psychotropic medication.
- Management of pain medication.
- Integration of physical and behavioral health.
Care coordination. Measures that assess care coordination can be difficult to develop and collect. Measures that assess the following areas were particularly noteworthy for the Subcommittee:
- Medication reconciliation.
- Care transitions.
- Shared accountability (across providers treating the same individual).
Access. While access-to-care measures exist for the general population and for many general services (i.e., screenings), measures that address access to specific services, such as behavioral health care, are less abundant. Additional areas for which to measure access include the following:
- Oral health care.
- Emergency department use.
The Subcommittee acknowledged that many of the gap areas identified lacked usable measures due to the complexities of gathering data from all the necessary data sources or the difficulty of incorporating patient-reported outcomes. Care coordination, for example, requires pulling information from multiple and varied systems that may not normally communicate with one another. Thus, they recommended consideration of new data collection/aggregation and analytic methodologies to support new measures. For example, broader deployment of health information technology in the health care setting may support increased use of patient-generated data and clinically enriched datasets that allow for measures assessing health outcomes. Opportunities to combine Medicare and Medicaid data for people with dual eligibility and integration of data from health information exchanges could be considered for increasing both the meaningfulness and feasibility of measures.
Reporting and Implementation
Cognizant of the data collection burden for States, the Subcommittee suggested several ways to incorporate alternative reporting approaches for the Initial Core Set:
- Stratification of reporting for certain populations, defined by race/ethnicity, Medicaid eligibility category (disabled, dual Medicare/Medicaid eligibility), or condition (e.g., serious mental illness). The Subcommittee noted that providing States with a standardized methodology for stratifying measurement data from the Initial Core measures would be critical for identifying potential disparities in care or unmet needs. For example, general population measures such as Medical Assistance With Smoking and Tobacco Use Cessation or diabetes screening measures could be reported for people with schizophrenia and bipolar disorder or for people dually eligible for Medicare/Medicaid.
- Rotation. Since efforts to improve performance on measures are likely to take more than one year, the Subcommittee recommended rotating measure reporting so that a designated set of measures are reported every other year or some other specified interval (rather than annually). In this way, the burden of reporting could be limited without affecting the usefulness of the information.
- Composite measures. The Subcommittee recommended considering options for creating measure composites that would provide high-level information about care.
- Incorporation of different supplemental sets into the Consumer Assessment of Healthcare Provider and Services (CAHPS®) surveys. The Subcommittee recommended the use of the CAHPS Health Plan survey and the HEDIS Supplemental CAHPS set but also noted that other supplemental sets exist for CAHPS addressing important topics for the Medicaid population (e.g., Cultural Competence).
Page originally created December 2010