5032.0: Wednesday, November 15, 2000 - Table 5

Abstract #6411

Validating Medicaid recipients' self-reports of health service use and managed care plan membership using claims data

Rosemary V. Chaudry, PhD, MHA, RN1, Nancy B. Schoeps, PhD2, William P. Brandon, PhD, MPH3, and Laure D. Debow, BS2. (1) University of North Carolina Charlotte, 1414 Willowood Way, Marion, OH 43302, 740-389-4940, chau5@aol.com, (2) Department of Statistics & Applied Mathematics, University of North Carolina Charlotte, 9201 University City Boulevard, Charlotte, NC 28223, (3) Department of Political Science, University of North Carolina Charlotte, 9201 University City Boulevard, Charlotte, NC 28223

Surveys are the primary means of collecting consumer data for evaluating health care programs. This paper reports on analyses to validate Medicaid recipients' self-reported health service use and HMO or primary care case management (PCCM) enrollment using Medicaid paid claims data. Data were collected through cross-sectional surveys conducted in 1996 and 1998 to evaluate a mandatory Medicaid managed care demonstration program in North Carolina, using a pre-post, quasi-experimental design. Both the demonstration and control counties had fee-for-service Medicaid arrangements in 1996. The demonstration county began enrolling recipients into HMOs in mid-1996, and the control county began enrolling recipients into PCCM in 1998.

Using a 6-month recall period, survey respondents were queried regarding health service use, health status, quality, and satisfaction with care. One set of analyses (two samples, n=200 each) compared self-reported service use (having had any medical care, having visited a doctor's office or clinic, having visited an emergency room) with paid claims data over the recall period. A second set of analyses used claims data to validate self report of HMO membership (plan name, length of enrollment) in the demonstration county (n=900) and enrollment in the PCCM program in the comparison county (n=200).

Managed care plans rely on member surveys to collect data on access, quality, and satisfaction for both commercial and Medicaid enrolled populations. Study results have implications for the design and application of measures to assess the validity of enrollees' reports of actual and potential access to care.

Learning Objectives: Describe two methods of assessing the validity of health survey data. Name two techniques to improve the validity of health survey data

Keywords: Data Collection, Medicaid

Presenting author's disclosure statement:
Organization/institution whose products or services will be discussed: None
I have a significant financial interest/arrangement or affiliation with any organization/institution whose products or services are being discussed in this session.
Relationship: Evaluation research funded by NC Dept.of Health & Human Services, Division of Medical Assistance

The 128th Annual Meeting of APHA