Should home-based HIV testing and counseling services be offered periodically in programs of ARV treatment as prevention? A case study in Likoma (Malawi)

Abstract

To reduce HIV incidence, prevention programs centered on the use of antiretrovirals require scaling-up HIV testing and counseling (HTC). Home-based HTC services (HBHTC) increase HTC coverage, but HBHTC has only been evaluated during one-off campaigns. Two years after an initial HBHTC campaign ("round 1"), we conducted another HBHTC campaign ("round 2") in Likoma (Malawi). HBHTC participation increased during round 2 among women (from 74 to 83%, P < 0.01). New HBHTC clients were recruited, especially at ages 25 and older. Only 6.9% of women but 15.9% of men remained unreached by HBHTC after round 2. HIV prevalence during round 2 was low among clients who were HIV-negative during round 1 (0.7%), but high among women who received their first ever HIV test during round 2 (42.8%). The costs per newly diagnosed infection increased significantly during round 2. Periodically conducting HBHTC campaigns can further increase HTC, but supplementary interventions to enroll individuals not reached by HBHTC are needed.

Figure 1. Classification of cohort members according to their participation in HBHTC campaigns and use of other HIV testing services

Notes: The dotted black oval represents the entire cohort, whereas the maroon oval repre sents all round 1 HBHTC clients and the light pink oval represents all round 2 HBHTC clients. As a result, the area marked “A” represents “continuing HBHTC clients” (i.e., they participated in both HBHTC rounds). Areas of the pink oval outside of A, i.e., (areas marked “B” or “C”), on the other hand, represent “new HBHTC clients” (i.e., they did not participate in round 1). Areas of the maroon oval outside of “A” repres ent “drop-outs”, i.e., cohort members who participated in round 1 but not in round 2. New HBHTC clients during round 2 are further divided between “repeat testers” (area marked “B”) and “first-time testers” (area marked “C”). Areas outside of A, B, C and D constitute cohort members who were “unreached” through HBHTC.

Figure 2

Flow chart of follow-up and participation in round 2 of the HBHTC campaign on Likoma

Figure 3. Participation in two consecutive HBHTC campaigns on Likoma Island

Notes: proportions depicted in this panel include only cohort members residing on Likoma at the time of both campaigns (i.e., excluding attritors, n=764). Cohort members who tested positive during HBHTC round 1 are included in the participation rates represented in this panel. Error bars represent 95% confidence intervals.

Figure 4. Patterns of repeated uptake of HBHTC after two campaigns on Likoma Island

Notes: drop-outs are cohort members who participated in round 1, but not in round 2; continuing clients participated in both rounds; new clients participated in round 2 only and unreached cohort members did not participate in any HBHTC round. Proportions depicted in this panel include cohort members residing on Likoma at the time of both campaigns (i.e., non-attritors, n=764). Cohort members who tested positive during HBHTC round 1 are included in the participation rates represented in this panel.