group of employers and health plans. These employers, who purchased health care for their employees, were seeking meaningful data to require of their contracting health plans. The health plans wished to reduce costly variations in the data they were required to submit to multiple purchasers. This critical mass of employer-purchasers and health plans ensured the calculation and submission of the HEDIS measures while they were still in a preliminary state, which subsequently attracted other influential supporters. CMS, for example, now requires health plans participating in the Medicare program to submit data on HEDIS measures. Many state Medicaid agencies also require the submission of HEDIS or HEDIS-like measures. In contrast, submission of the Behavioral Healthcare Performance Measurement System inpatient hospital measures to NASMHPD or NRI is not required, but facilities that choose to do so may use those measures to fulfill accreditation reporting requirements.
Reported measures may not accurately represent an individual’s or organization’s performance. Information systems and internal data recording conventions used by individual clinicians and health care organizations vary greatly. Data also may not be collected or stored in ways that facilitate collection of a measure as requested. When measures further require data to be linked across organizations, there may be incompatible data formats. All these factors can introduce error, as can less-than-scrupulous adherence to a measure’s specifications. Because the reporting of quality measures to external bodies for public disclosure to consumers, for use in financial reimbursement strategies to reward best performance, or in response to other quality oversight requirements can have significant consequences for the entity being measured, it is important for the accuracy of the reported measures to be verified. This is typically accomplished through systematic audits of the measures’ calculation. NCQA, for example, has developed standardized auditing procedures for use in verifying the integrity of the calculation of HEDIS measures (NCQA, undated).
Ensuring that quality measures are useful for multiple audiences requires analytic and communication capabilities that can respond to the sometimes differing needs of consumers, health care providers (both individual clinicians and organizations), purchasers, and quality oversight organizations. For example, while clinicians and health care organizations may want numerous, detailed data on their performance on individual