Outcomes of HIV partner services for people with HIV and STD coinfection versus new HIV diagnosis: implications for HIV prevention strategies.

To the Editor: W read with great interest the article by Peterman et al.1 and applaud their work in highlighting the value of implementation science and program evaluation. In their 4-year long evaluation of data in nearly 60 health jurisdictions, the implementation of STD performance measures was associated with little improvement in STD program performance. Seen as an intervention, performance measures were initially introduced in the hopes of improving performance. The authors state, “We thought that if we measured and reported programs’ performance on specific activities, then program performance would improve. We expected that the low-performing programs would identify factors that contributed to their low performance and take steps to improve.”1 Although the data presented showed that the STD performance measures were not effective, the authors insist, “...we believe it is too early to abandon performance measures.”1 But if not now, when? What is not considered in the piece by Peterman is the work necessary for the local health department. Data must be collected, cleaned, analyzed, and uploaded to Center for Disease Control and Prevention through the performance measures mechanism. As public health resources continue to be lost, maintaining a basic programmatic infrastructure is challenging.2 Continuing to participate in an “intervention” that has been shown to be unproductive not only adds stress to a fragile system without any identified benefit but also undermines the whole point of program evaluation. If data-based decision making is the goal, why are we ignoring these data?