Male Caregiver Barriers to HIV Index Case Testing of Untested Children

To the Editor: INTRODUCTION Children lag behind adults in attaining the first 95 target of the UNAIDS 95-95-95 goals.1 Index case testing (ICT)—testing the children of adults in HIV care—has a relatively high diagnostic yield and has been scaled to identify undiagnosed children of parents in HIV care in countries with high HIV prevalence.2–4 Despite the high yield, ICT uptake remains suboptimal; it is important to identify limitations of ICT reach. In assisted partner testing services (APS)—a case-finding model similar to ICT where sexual partners of adults in HIV care are tested—female participants had higher rates of partner HIV testing and case-finding rates compared with male participants.5 Understanding differences in uptake of ICT and challenges to ICT by caregiver sex could inform efforts to improve pediatric HIV case-finding strategies to identify undiagnosed children. During recruitment for the Financial Incentives to Increase Uptake of Pediatric HIV Testing randomized controlled trial (RCT) (NCT03049917) testing effectiveness of financial incentives to increase pediatric HIV testing uptake, we compared reasons children could not be tested through ICT by caregiver sex. METHODS Ethics Statement The study was approved by the University of Washington Institutional Review Board and the Kenyatta National Hospital Ethics and Research Committee. All participants provided oral consent for screening questions used in this analysis. The study protocol6 and trial results7 are presented elsewhere. Recruitment and ICT Offer Sequential clients attending 19 HIV care clinics in Kisumu and Siaya counties in Kenya were screened for eligibility. The study team attempted to screen all adults in HIV care. Clients were asked whether they had children aged 0–12 years of unknown HIV status. Female index clients were eligible for ICT if they had any children who had never been tested for HIV or had not completed testing after the cessation of breastfeeding. Male index clients were eligible for ICT if they met the same requirements and knew that the mother of the untested child(ren) was HIV-positive or deceased. Caregivers were ineligible if they could not access to their children. Among potential index clients, we characterized the reasons children could not be tested through ICT. Differences in proportions were compared between male and female index clients using the χ2 tests. RESULTS Of 22,275 HIV-positive adults screened, most were female [15,008 (67%)]. A higher proportion of male clients [7% (483/7267)] reported untested children aged 12 years or younger compared with female clients [2% (358/15,008); P < 0.001; Table 1]. Among caregivers with untested children aged 12 years or younger, 34% (166/483) of male and 89% (320/358) of female caregivers were eligible for ICT (P < 0.001). TABLE 1. - Comparing Male and Female Adults Approached for Participation in ICT Trial Adult Sex, n (%) P Male Female Adults approached 7267 (33) 15,008 (67) Adults reporting untested children aged 12 years or younger* 483 (7) 358 (2) <0.001 Unknown maternal HIV status 54 (11) 0 (0) <0.001 Negative maternal HIV status 120 (25) 2 (1) <0.001 No access to child 141 (29) 31 (9) <0.001 Separated or divorced 116 (82) 16 (52) <0.001 Distance from child (not separated or divorced) 25 (18) 15 (48) <0.001 Eligible for ICT and met study criteria 166 (34) 320 (89) <0.001 Declined to participate in trial 19 (11) 15 (5) 0.006 Not interested/in a rush 8 (42) 12 (80) 0.026 Wanted to consult with partner 9 (47) 1 (7) 0.010 Planned to test child later (outside study) 2 (11) 2 (13) 0.801 Agreed to participate in trial 147 (89) 305 (95) 0.006 Children tested through ICT 63 (43) 148 (49) 0.258 Children tested HIV-positive through ICT 2 (3) 5 (3) 0.940 *Two male and 5 female caregivers with children aged 12 years or younger were not eligible for ICT because of reasons other than not having access to the child or maternal HIV status of the child is unknown or negative. Among caregivers with untested children aged 12 years or younger, a higher proportion of male caregivers were ineligible for ICT compared with female caregivers because of inability to physically access their children [29% (141/483) vs. 9% (31/358); P < 0.001]. Among caregivers who did not have access to their children, a higher proportion of male than female caregivers did not have access to their children because of separation or divorce [82% (116/141) vs. 52% (16/31)] rather than distance alone [18% (25/141) vs. 48% (15/31)] (P < 0.001 for both comparisons). Male caregivers were more likely to decline participation in the RCT than female caregivers [11% (19/166) vs. 5% (15/320); P = 0.006]. Among those who declined to participate, a higher proportion of male caregivers declined because they wanted to consult with a partner before participating compared with female caregivers [47% (9/19) vs. 7% (1/15), P = 0.010]. Overall, 89% (147/166) of male and 95% (305/320) of female caregivers who were eligible for ICT and could access their children agreed to participate in the trial (P = 0.006). Of these, there was no difference in male [43% (63/147)] and female [49% (148/305)] caregivers whose children were tested for HIV in the RCT (P = 0.258) (previously reported).7 Among those with children who were tested through ICT, 3% (2/63) of male caregivers and 3% (5/148) of female caregivers had children who tested HIV-positive (P = 0.940). DISCUSSION We found that many untested children could not be accessed for testing through ICT because they were inaccessible to adult clients. Overall, male caregivers were more likely to have untested children but less likely to be eligible for ICT. In this study, we screened caregivers attending HIV care services, reflecting successful engagement in care. Male caregiver reasons for not electing pediatric ICT included separation from their child, unknown maternal HIV status, and desire to involve mothers in the decision to participate in pediatric ICT. Partner communication and disclosure have been associated with male involvement in prevention of mother-to-child transmission of HIV programs,8 which results in improved maternal and infant health outcomes,9–12 and early initiation of pediatric ART,13 respectively. Promoting partner communication and disclosure after prevention of mother-to-child transmission of HIV programs could facilitate increased male involvement in ICT. In this analysis, male caregivers faced unique barriers to participating in pediatric ICT. A study in Cameroon similarly found that female caregivers were 1.6 times more likely to test their children through ICT compared with male caregivers.14 In addition, a study in Kenya found that children were 3.2 times more likely to be tested if more information was available about maternal HIV status.15 Knowledge of maternal HIV status remains a barrier to engaging male caregivers in pediatric ICT. To increase the effectiveness of ICT, the barriers of male index clients must be addressed. Although there were no differences between the proportion of children testing HIV-positive by caregiver sex in our cohort, a study in Tanzania found that orphans and vulnerable children with male caregivers were 1.4 times more likely to test HIV-positive compared with those with female caregivers, underlining the importance of engaging male caregivers in pediatric ICT strategies in sub-Saharan Africa.16 Novel approaches are needed to efficiently and systematically identify, assess, and test the children of HIV-positive male index clients not yet in care.11–14 Although scale-up of prevention of mother-to-child transmission of HIV and ICT programs have dramatically reduced the proportion of HIV-positive caregivers in care with untested children, gaps still exist in pediatric testing. Our Kenyan ICT study in 2013–2014 observed that 7% of caregivers in HIV care had untested children aged 12 years or younger, whereas, in this 2017–2018 trial, that proportion dropped to 3%.17 As programs in sub-Saharan Africa scale up ICT, it will become more difficult to identify the decreasing number of untested children. Family HIV disclosure, for both previous and current relationships, may provide additional benefit to identify untested children. Use of APS to reach children of HIV-positive indexes has been tested in South Africa and Lesotho and merits expansion.18,19 Reaching children through APS could potentially increase male index clients' access to children by fostering communication and disclosure among partners; however, the South African study noted that sex dynamics remain important barriers to APS.18 In our study, 11% of men reported that mothers of their children had an unknown HIV status, indicating a need for APS among women of unknown HIV status to increase partner disclosure and pediatric HIV testing. Incorporating counseling on pediatric HIV testing and communication with partners about pediatric HIV testing into APS could increase HIV testing of untested children of male caregivers. We screened clients in 19 HIV clinics in Western Kenya and sequentially screened every adult presenting for care, increasing generalizability and external validity. Due to the limitations of screening data for this trial, we do not have more specific maternal mortality data. Further studies to understand eligibility and additional barriers to ICT among male caregivers are warranted. CONCLUSIONS As programs scale up ICT and intensify efforts to reach the first 95 of the 95-95-95 UNAIDS targets for pediatric HIV testing, male caregiver barriers to ICT—including lack of access to their child, unknown maternal HIV status, and desire to involve mothers in the testing decision—must be addressed to effectively reach untested children.