160594 A Cluster Analysis with Repeated Measure to Assess Modification in Screening Behavior of a Diabetic Population

Monday, November 5, 2007

Owen Johnson, DrPH , Health Care Informatics, Scan Health Plan, San Diego, CA
Purpose: Assess the effectiveness of a reminder and incentive program designed to increase the screening rates for Cholesterol (LDL), Kidney Nephropathy, and Hemoglobin (HBA1C) among the diabetic population of the HMO from March 2005 to March 2006.

Methods: By HEDIS criteria a total of 24,591 eligible diabetic members were identified. These members were randomized into 2 groups depending on whether their primary care provider was in a controlled (n=4983) or intervention (n=6016) group. Each cohort of the provider group was subsequently randomized into 3 mutually exclusive clusters for each (LDL, Kidney, and HBA1C) screening examination. Each screening modalities had a 3 clusters of member groups, a control (n=3,683, received standard monthly Health News Letter), a reminder group (n=3,666, received a tailored reminder to get their LDL screening done), and a gift group (n=3,650, received the tailored reminder and a incentive on completion their screening). The LDL screening rates pre and post the intervention were analyzed using SAS's Mixed Model to evaluate the statistical differences in the screening rates.

Results: The LDL screening rates of 63.84% in the pre-intervention year was assessed to be significantly different from the screening rates of 69.48% during of the intervention year. The screening rates increased by 8.83% from the pre to post intervention period. The change of rates however could not associated with the reminder and incentive interventions. Members in the reminder group however, who did not screened in the pre-intervention period but did in the post-intervention period showed a marginally (p-value of 0.06) significant difference in their screening rates compared to the control group.

Conclusion: The LDL screening rates increased from baseline but the change could not be attributed to the reminder and gift interventions. Targeting non-compliant members for the intervention may have generated more meaningful results.

Learning Objectives:
The emphasis of how important the informatic study design is critical in generating the desired results.

Keywords: Diabetes, Medicare

Presenting author's disclosure statement:

Any relevant financial relationships? No
Any institutionally-contracted trials related to this submission?

I agree to comply with the American Public Health Association Conflict of Interest and Commercial Support Guidelines, and to disclose to the participants any off-label or experimental uses of a commercial product or service discussed in my presentation.