This information is for reference purposes only. It was current when produced and may now be outdated. Archive material is no longer maintained, and some links may not work. Persons with disabilities having difficulty accessing this information should contact us at: https://info.ahrq.gov. Let us know the nature of the problem, the Web address of what you want, and your contact information.
Please go to www.ahrq.gov for current information.
Realizing the Promise of Value-based Purchasing
Using Data Effectively
Michael Bailit, M.B.A., President, Bailit Health Purchasing, LLP, Wellesley, MA.
Angela Dombrowicki, Bureau Director, Managed Health Care Programs, Division of Health
Care Financing, Department of Health and Family Services, Madison, WI.
State programs often do not know how to use data from health plans and providers to
effectively design, implement, monitor, and evaluate their purchasing efforts. This session
examined the efforts of purchasers that sought to make better use of the data they collected to
inform their health care purchasing activities.
Many States have improved the amount and quality of data they collect. Yet Michael Bailit
observed that some States have been unable to ensure that these data are being used to support
quality improvement initiatives. Several steps are necessary to maximize the
usability of data:
- Define data across a broad array of program performance priorities (and the collection
strategy should support the higher of these priorities).
- Collect data both directly and indirectly.
- Confirm data quality through mechanisms such as audits.
- Analyze data for management use.
- Apply analyzed data directly to designing programs, identifying problems, designing
and tracking interventions, and reporting to stakeholders.
Failure to address any one of these steps can undermine an entire value-based purchasing
strategy. Purchasers should follow these general "rules of thumb" to avoid
- Define purchasing priorities first and ensure that the data strategy directly supports
them—do not allow the reporting and collection requirements to address only lower
- Do not collect more data than you can use, and do not collect data if you have
insufficient resources to use them.
- Conduct timely and valid data analyses and do not delay reporting, which causes data to
- Require the development of action plans based on data findings, rather than shelving
and not using data.
- Schedule regular agency, staff, and contractor evaluations and use qualitative and not
quantitative information in these.
- Make data management part of your culture—you must have an ongoing commitment
to maintaining data in the purchasing effort.
MassHealth, the Massachusetts Medicaid program, has used data successfully
to examine and improve health plan performance. MassHealth collects a range of
data—Health Plan Employer Data and Information
Set (HEDIS®) measures, primary care case management (PCCM) primary care physicians (PCPs) quality indicators, Consumer Assessment of Health Plans (CAHPS®), encounter data, and State-defined and plan-reported data—for many purposes:
- MassHealth uses data to set annual Agency priorities and goals, with each goal having
attendant measures defined at the beginning of the year and checked semi-annually.
- A subset of HEDIS® measures helps to assess the performance of managed care
organizations (MCOs) and the PCCM plan.
- Baseline data on HEDIS® measures are collected to work with contracted providers to
implement quality improvement processes.
- An annual HEDIS® report provides detailed information on health plan performance
compared to national, State, and Medicaid HEDIS® benchmarks.
- HEDIS® and other data are used to identify opportunities for improvement and
goal setting for MCOs (including tracking results).
- Quality indicators (HEDIS® and non-HEDIS®) are highlighted in quarterly reports that
profile PCP practices.
- Encounter data are used to analyze differences in plan enrollment and use, and in
case mix across categories and plans.
MassHealth has also used data analysis results to develop financial incentives and disincentives
in its contracting with plans—incentives are attached to goals to make significant improvement
in performance, while disincentives are attached to failure to improve.
Angela Dombrowicki from the Wisconsin Medicaid program offered another
example of successfully using data, describing the State's use of encounter data as part of a
multi-pronged, evolving quality improvement strategy. Achieving uniform encounter data—accurate records of service provided to all Medicaid recipients enrolled in the managed care
system—is beneficial because it:
- Permits comparison across health maintenance organizations (HMOs).
- Enables systematic monitoring to identify "best practices" and those areas needing
- Streamlines HMO reporting.
- Reduces administrative effort by the State and HMOs by using only one data set for
quality improvement activities.
- Permits data analysis for any chosen indicator/process/outcome for which there are
Wisconsin collected and published survey data prior to gathering encounter data, which were
collected annually from HMOs by asking them specific questions about a segment of their
population. These data tended to be incomplete, difficult to validate without extensive chart
audits, difficult to use for trending, untimely, and inaccurate. Encounter data are submitted
monthly and are complete, easily verifiable, flexible, and timely.
Some of the reasons the encounter data system has been successful in Wisconsin are:
- The State made the new data system a priority and worked intensively with HMOs over
18 months to get feedback on a sample data set, and designed this system around what
HMOs could handle.
- The contract language was strengthened to include strict compliance penalties for data
- Other reporting requirements were relaxed somewhat to focus primarily on developing
- The system continued building upon past successes of data validity audits by reviewing
two areas on-site with HMOs: the capability of the HMO data system, and sample chart
reviews of selected data.
Wisconsin's experience can inform others about how to use data for quality improvement. In
Medicaid managed care, a data system like Wisconsin's can assist HMOs improving
management of certain conditions, such as childhood asthma. State Medicaid programs and
other payers may also learn from Wisconsin's current initiative, the Minimal Operational Data
Set (MODS). MODS aims to provide a master database/Web site for consumers, policymakers,
researchers, legislators, and the general public that will inform them about HMO performance
in common quality areas.
Fossett JW, Goggin M, Hall JS, et al. Managing Medicaid Managed Care: Are States
Becoming Prudent Purchasers? Health Affairs 2000 19(4):36-49.
Previous Section Contents Next Section