Thursday, November 6, 2014
:
10:55 AM
Picuris (Convention Center)
*Names in bold indicate Presenter
Background: Performance-Based Grants Management (PBGM) programs are increasingly being used by governments of all levels to get more “bang for their buck.” The U.S. Centers for Disease Control and Prevention (CDC) uses PBGM as part of its management of the National Breast and Cervical Cancer Early Detection Program (NBCCEDP), a federally-funded public health program providing cancer screening to low-income, underinsured women. CDC funds all 50 states, the District of Columbia (D.C.), and 16 tribes and territories to implement the NBCCEDP. As a requirement of funding, all grantees report a minimum set of client-level data to CDC semi-annually for monitoring screening delivery and quality. CDC assesses performance on a set of 11 core indicators, two emphasizing priority populations for the program and nine clinical measures promoting complete and timely care and cancer treatment initiation. In 2005, CDC implemented PBGM, integrating performance on the 11 measures into its funding process. The purpose of this study is to assess the effectiveness of CDC’s PBGM program in improving grantee performance.
Methods: We examined grantee performance using CDC’s 11 core indicators before and after implementation of the PBGM. We adopted a panel design for the study, consisting of cross-sectional data for each State grantee and D.C. on the 11 core indicators. The analytic dataset consisted of 23, 6-month performance periods beginning January 2000 through June 2011. We developed weighted least squares regression models for each of the performance indicators, regressing the performance measures on the pre-intervention, intervention, and post-invention variables, while controlling for 50 dummy variables, where each dummy variable represented a particular grantee. We also constructed time-series graphs in order to conduct visual inspection of the data.
Results: The PBGM system had a significant and pronounced effect on both indicators related to NBCCEDP priority populations for screening. This may reflect the greater degree of control that grantees have over performance in these areas as they often conduct outreach and recruitment efforts themselves. This being a high performing program, we could only find three of the 11 indicators where we could separate grantees that were not meeting their targets at the time of the implementation of the program in 2005. The results invariably showed performance improvement for all the indicators that were not meeting the targets.
The indicator Timely Treatment for Invasive Cervical Cancer served as the exception and showed negative trend in performance after the implementation of the PBGM program. We believe that the target was too relaxed and the grantees either started slacking out or diverted their attention towards other indicators which required more attention in terms of performance improvement. For the rest of the indicators the PBGM program showed either marginal improvements or no effect at all.
Conclusion: Greater control by grantees over performance in recruiting priority populations for screening may help explain positive results on the population indicators. In regard to the clinical indicators, the lack of significant findings may reflect consistently high performance by grantees (ceiling effect) and/or benchmarks that may be set too low.
Methods: We examined grantee performance using CDC’s 11 core indicators before and after implementation of the PBGM. We adopted a panel design for the study, consisting of cross-sectional data for each State grantee and D.C. on the 11 core indicators. The analytic dataset consisted of 23, 6-month performance periods beginning January 2000 through June 2011. We developed weighted least squares regression models for each of the performance indicators, regressing the performance measures on the pre-intervention, intervention, and post-invention variables, while controlling for 50 dummy variables, where each dummy variable represented a particular grantee. We also constructed time-series graphs in order to conduct visual inspection of the data.
Results: The PBGM system had a significant and pronounced effect on both indicators related to NBCCEDP priority populations for screening. This may reflect the greater degree of control that grantees have over performance in these areas as they often conduct outreach and recruitment efforts themselves. This being a high performing program, we could only find three of the 11 indicators where we could separate grantees that were not meeting their targets at the time of the implementation of the program in 2005. The results invariably showed performance improvement for all the indicators that were not meeting the targets.
The indicator Timely Treatment for Invasive Cervical Cancer served as the exception and showed negative trend in performance after the implementation of the PBGM program. We believe that the target was too relaxed and the grantees either started slacking out or diverted their attention towards other indicators which required more attention in terms of performance improvement. For the rest of the indicators the PBGM program showed either marginal improvements or no effect at all.
Conclusion: Greater control by grantees over performance in recruiting priority populations for screening may help explain positive results on the population indicators. In regard to the clinical indicators, the lack of significant findings may reflect consistently high performance by grantees (ceiling effect) and/or benchmarks that may be set too low.
Full Paper:
