Central implementation strategies outperform local ones in improving HIV testing in Veterans Healthcare Administration facilities.

Background: Pilot data suggest that a multifaceted approach may increase HIV testing rates, but the scalability of this approach and the level of support needed for successful implementation remain unknown.

Objective: To evaluate the effectiveness of a scaled-up multi-component intervention in increasing the rate of risk-based and routine HIV diagnostic testing in primary care clinics and the impact of differing levels of program support.

Design: Three arm, quasi-experimental implementation research study.

Setting: Veterans Health Administration (VHA) facilities.

Patients: Persons receiving primary care between June 2009 and September 2011 INTERVENTION: A multimodal program, including a real-time electronic clinical reminder to facilitate HIV testing, provider feedback reports and provider education, was implemented in Central and Local Arm Sites; sites in the Central Arm also received ongoing programmatic support. Control Arm sites had no intervention

Main Measures: Frequency of performing HIV testing during the 6 months before and after implementation of a risk-based clinical reminder (phase I) or routine clinical reminder (phase II).

Key Results: The adjusted rate of risk-based testing increased by 0.4 %, 5.6 % and 10.1 % in the Control, Local and Central Arms, respectively (all comparisons, p < 0.01). During phase II, the adjusted rate of routine testing increased by 1.1 %, 6.3 % and 9.2 % in the Control, Local and Central Arms, respectively (all comparisons, p < 0.01). At study end, 70-80 % of patients had been offered an HIV test.

Conclusions: Use of clinical reminders, provider feedback, education and social marketing significantly increased the frequency at which HIV testing is offered and performed in VHA facilities. These findings support a multimodal approach toward achieving the goal of having every American know their HIV status as a matter of routine clinical practice.