What is the level of uptake of Partner Notification Services in HIV testing in Selected Health Facilities in Muranga County – Kenya; a retrospective Study

Background: Identification of people living with HIV is key in HIV prevention and control. Partner Notification service is a World Health Organization backed strategy of reaching out to sexual partners of people diagnosed with HIV for HIV testing. However, its adoption and success rate in Kenya remains unknown. Methods: A cross sectional facility based study was undertaken in five purposively selected health facilities in Muranga County, Kenya. A retrospective review of patient medical records data for HIV positive index clients and their Sexual Partners conducted. Census approach applied to extract data for study subjects from Partner Notification Service registers for the period covering January 2017 to August 2018. Epi Info software was used for data analysis. Results: A total of 183 index clients offered Partner notification services. The mean age of the indexed clients studied was 39(SD ±13.1). Females comprised 64 % of clients studied. Of the 183 indexed clients, 89% accepted the services and elicited 216 sexual partners for tracing. The ratio of elicited sexual partners to index client was 1.3:1. Out of the 216 sexual partners, 77% were reached and tested. A total of 46 [32%] of the sexual partners elicited and traced, tested HIV positive. The most preferred approaches were provider referral (51%) and contract referral (45%). Dual referral (4%) was the least preferred approach. Conclusions: Partner notification services is acceptable and an effective strategy of increasing HIV case identification and raising awareness to exposed sexual partners in low resource countries.

Of the 183 indexed clients, 89% accepted the services and elicited 216 sexual partners for tracing. The ratio of elicited sexual partners to index client was 1.3:1. Out of the 216 sexual partners, 77% were reached and tested. A total of 46 [32%] of the sexual partners elicited and traced, tested HIV positive. The most preferred approaches were provider referral (51%) and contract referral (45%). Dual referral (4%) was the least preferred approach.
Conclusions: Partner notification services is acceptable and an effective strategy of increasing HIV case identification and raising awareness to exposed sexual partners in low Partner services are underutilized in Kenya therefore more research on the coverage and implementation gaps for HIV Partner Notification services is required. In Murang'a County, Partner Notification Services strategy has been adopted as part of routine HIV testing services(HTS) and is being implemented in all health care facilities offering HTS. There is limited evidence on uptake and effectiveness of use of PNS in sexual partner elicitation, notification and testing in Kenya. The aim of this study findings was to support policy makers and health managers on improving implementation of PNS services aimed at achieving increasing HIV case identification and coverage of care and treatment.

Methods
The study was a retrospective; health facility based study-involving review of secondary data from PNS registers. The registers documented HIV positive index clients who were offered PNS and the sexual partners elicited. The review period covered January 2017 to August 2018. Five health facilities in Gatanga sub county, Murang'a County were purposively selected for review based on the high volume workload, adoption of PNS and availability of care and treatment services. The selected facilities included one sub-county hospital, two health centres and two dispensaries. In this study, study subjects were included based on the following criteria: (i) indexed clients from the selected facilities who were offered PNS, (i) sexual partners elicited by indexed clients, (ii) availability of subject details in the PNS registers for the period between January 2017 to August 2018. Census approach was used to list and enrol 183 index study subjects who met the study criteria.
Data was abstracted using two data abstraction tools and entered into an excel sheet from the PNS registers; the first checklist contained details of all the index clients and the second checklist contained details on all the sexual partners elicited by the index client. Epi-Info, a data analytic software was used to clean the data and perform descriptive statistics for the study. Study authorization and permission sought from relevant institutions. Dummy codes [anonymised] assigned to facilities and study subjects.
Password protected databases were also applied as part of confidentiality measures.

Results
Records of 183 index clients in the period between January 2017 to August 2018 reviewed.
Mean age for the index clients was 39 [SD ± 13.1] and most were female 118 [65%].  Table 2 below. For index clients aged below 25 years 9[43%] elicited 2 or more sexual partners while among index clients with formal employment 56% elicited 2 or more sexual partners.  The ratio of sexual partners to index clients varied across the health facilities where Gatura had the highest ratio of 2.10:1. Overall ratio of sexual partners to index clients was 1.33:1 as shown in Table 4

Discussion
These study findings demonstrate that PNS is generally acceptable and effective in increasing HIV case identification. The findings of this study on PNS acceptance shows a higher rate of PNS uptake than that of another RCT done in Kenya where acceptance rate was 67% (6). Routine implementation of the PNS as part of HTS in health facilities would contribute to increased HIV case finding. The ratio of sexual partners to index clients varied significantly across gender and health facilities. The strategy was most effective in eliciting male SPs. This is similar to a study done in Tanzania where male index clients were found to be 6.2 times more likely to list more than one sexual partner (7). There were more female index clients than male hence the high rate of male sexual partners' elicitation. Further, uptake of voluntary HTS is higher among females compared to males.
PNS shows potential benefits for increasing HIV case identification as well as increasing awareness of possible exposure to HIV among SPs (6). Undertaking more than one follow up to reach the sexual partners increases the potential for successful elicitation.
HIV testing uptake among partners elicited varied as well across the various facilities with lowest uptake being 62% and highest being 100%. This is attributed to the effectiveness of the different methods of reaching out to the partners as preferred by index clients as well as the level of elicitation skills of the HTS provider. Contract referral approach is the most effective approach in reaching sexual partners. However, provider referral approach was the most preferred approach by the index clients. These findings are contrary to a study on outcomes of PNS done in Tanzania where almost all index clients chose passive referral approach (7).
The study established first time contact acceptance rate of PNS among clients to be higher compared to that of index clients previously identified. This is similar to findings in a study in Tanzania  contract referral approach had the highest success rate of reaching and testing the sexual partners.
PNS is most effective when reach out messages are provided in more than one attempt; Only half of sexual partners are successfully reached in subsequent attempts after initial contact is made

Limitations Of The Study
This study results have some shortcomings. Considering the study used secondary data, there were some incomplete data such as on marital status for sexual partners, which was not documented.

Ethics approval and consent to participate
Researchers received permission from Murang'a County Government, Department of Health to conduct the study. The study used secondary data extracted from registers used in health facilities with anonymous identification details of the clients.

Conflict of interest
Authors have no conflict of interest in this study.

Consent to Publish
Not applicable

We acknowledge Field Epidemiology & Laboratory Training Program-Kenya [FELTP-Kenya]
for partial financing of this study.

Authors Contributions
Rewel Kariuki conceptualized this study. All the other authors provided technical review in developing, designing and implementation of the study. All the authors participated in the drafting and review of the paper.