Measuring the performance of computer vision artificial intelligence to interpret images of HIV self-testing results.

Introduction: HIV self-testing (HIVST) is highly sensitive and specific, addresses known barriers to HIV testing (such as stigma), and is recommended by the World Health Organization as a testing option for the delivery of HIV pre-exposure prophylaxis (PrEP). Nevertheless, HIVST remains underutilized as a diagnostic tool in community-based, differentiated HIV service delivery models, possibly due to concerns about result misinterpretation, which could lead to inadvertent onward transmission of HIV, delays in antiretroviral therapy (ART) initiation, and incorrect initiation on PrEP. Ensuring that HIVST results are accurately interpreted for correct clinical decisions will be critical to maximizing HIVST's potential. Early evidence from a few small pilot studies suggests that artificial intelligence (AI) computer vision and machine learning could potentially assist with this task. As part of a broader study that task-shifted HIV testing to a new setting and cadre of healthcare provider (pharmaceutical technologists at private pharmacies) in Kenya, we sought to understand how well AI technology performed at interpreting HIVST results. Methods: At 20 private pharmacies in Kisumu, Kenya, we offered free blood-based HIVST to clients ≥18 years purchasing products indicative of sexual activity (e.g., condoms). Trained pharmacy providers assisted clients with HIVST (as needed), photographed the completed HIVST, and uploaded the photo to a web-based platform. In real time, each self-test was interpreted independently by the (1) client and (2) pharmacy provider, with the HIVST images subsequently interpreted by (3) an AI algorithm (trained on lab-captured images of HIVST results) and (4) an expert panel of three HIVST readers. Using the expert panel's determination as the ground truth, we calculated the sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) for HIVST result interpretation for the AI algorithm as well as for pharmacy clients and providers, for comparison. Results: From March to June 2022, we screened 1,691 pharmacy clients and enrolled 1,500 in the study. All clients completed HIVST. Among 854 clients whose HIVST images were of sufficient quality to be interpretable by the AI algorithm, 63% (540/854) were female, median age was 26 years (interquartile range: 22-31), and 39% (335/855) reported casual sexual partners. The expert panel identified 94.9% (808/854) of HIVST images as HIV-negative, 5.1% (44/854) as HIV-positive, and 0.2% (2/854) as indeterminant. The AI algorithm demonstrated perfect sensitivity (100%), perfect NPV (100%), and 98.8% specificity, and 81.5% PPV (81.5%) due to seven false-positive results. By comparison, pharmacy clients and providers demonstrated lower sensitivity (93.2% and 97.7% respectively) and NPV (99.6% and 99.9% respectively) but perfect specificity (100%) and perfect PPV (100%). Conclusions: AI computer vision technology shows promise as a tool for providing additional quality assurance of HIV testing, particularly for catching Type II error (false-negative test interpretations) committed by human end-users. We discuss possible use cases for this technology to support differentiated HIV service delivery and identify areas for future research that is needed to assess the potential impacts-both positive and negative-of deploying this technology in real-world HIV service delivery settings.

Measuring the performance of HIV self-testing at private pharmacies in Kenya: a cross-sectional study.

INTRODUCTION: HIV self-testing (HIVST) has the potential to support daily oral pre-exposure prophylaxis (PrEP) delivery in private pharmacies, but many national guidelines have not approved HIVST for PrEP dispensing. In Kenya, pharmacy providers are permitted to deliver HIVST, but often do not have the required certification to deliver rapid diagnostic testing (RDT). We estimated the performance of provider-delivered HIVST compared to RDT, the standard of care for PrEP delivery, at private pharmacies in Kenya to inform decisions on the use of HIVST for PrEP scale-up. METHODS: At 20 pharmacies in Kisumu County, we trained pharmacy providers (pharmacists and pharmaceutical technologists) on blood-based HIVST use and client assistance (if requested). We recruited pharmacy clients purchasing sexual and reproductive health-related products (e.g. condoms) and enrolled those ≥18 years with self-reported behaviours associated with HIV risk. Enrolled clients received HIVST with associated provider counselling, followed by RDT by a certified HIV testing services (HTS) counsellor. Pharmacy providers and clients independently interpreted HIVST results prior to RDT (results interpreted only by the HTS counsellor). We calculated the sensitivity and specificity of pharmacy provider-delivered HIVST compared to HTS counsellor-administered RDT. RESULTS: Between March and June 2022, we screened 1691 clients and enrolled 1500; 64% (954/1500) were female and the median age was 26 years (IQR 22-31). We additionally enrolled 40 providers; 42% (17/40) were pharmacy owners and their median years of experience was 6 (IQR 4-10). The majority (79%, 1190/1500) of clients requested provider assistance with HIVST and providers spent a median of 20 minutes (IQR 15-43) with each HIVST client. The sensitivity of provider-delivered HIVST at the pharmacy was high when interpreted by providers (98.5%, 95% CI 97.8%, 99.1%) and clients (98.8%, 95% CI 98.0%, 99.3%), as was the specificity of HIVST in this setting (provider-interpretation: 96.9%, 95% CI 89.2%, 99.6%; client-interpretation: 93.8%, 95% CI 84.8%, 98.3%). CONCLUSIONS: When compared to the national HIV testing algorithm, provider-delivered blood-based HIVST at private pharmacies in Kenya performed well. These findings suggest that blood-based HIVST may be a useful tool to support PrEP initiation and continuation at private pharmacies and potentially other community-based delivery settings.

Provider perspectives on service delivery modifications to maintain access to HIV pre-exposure prophylaxis during the COVID-19 pandemic: qualitative results from a PrEP implementation project in Kenya.

INTRODUCTION: HIV pre-exposure prophylaxis (PrEP) is an essential prevention strategy being scaled up for priority populations in Kenya, including for HIV serodiscordant couples. The COVID-19 pandemic posed challenges to PrEP rollout. We conducted a qualitative study of PrEP providers to understand how clinics adjusted PrEP delivery during the COVID-19 pandemic. METHODS: Since 2017, the Partners Scale-Up Project has integrated PrEP into 25 HIV clinics in Central and Western Kenya. We conducted qualitative interviews with 40 purposively sampled clinic personnel. We interviewed personnel once during the first pandemic wave (May-Aug 2020) and again after some decline in COVID-19 rates (Nov-Jan 2021). We analysed data using inductive memo-writing and summarized data by themes along the PrEP delivery cascade, guided by the Framework for Reporting Adaptation and Modifications (FRAME). RESULTS: We interviewed 27 clinical officers, five nurses, four health records and information officers, and four counsellors from Central (n = 20) and Western (n = 20) Kenya. About half (n = 19) were female, with a median age of 32 (IQR: 29-34) and 2.3 years of experience delivering PrEP (IQR: 2-3). All participants reported clinic changes in PrEP demand creation and service delivery during the pandemic. Modifications occurred during PrEP implementation and sustainment phases, were partly reactive to the pandemic and also facilitated by interim Ministry of Health guidance on PrEP delivery during COVID, and were made by PrEP delivery teams, clients and clinic managers. Commonly reported modifications included dispensing multiple-month PrEP refills, intensifying phone-based client engagement and collaborating with other HIV clinics to ensure that clients with prolonged stays in other regions could continue to access PrEP. Some clinics also adopted practices to streamline visits, such as within clinical-room PrEP dispensing, pre-packing PrEP and task-shifting. Most providers liked these changes and hoped they would continue after the pandemic subsides. CONCLUSIONS: COVID-19 served as a catalyst for PrEP delivery innovations in Kenya. HIV clinics successfully and rapidly adapted their PrEP demand creation, refill and retention strategies to promote PrEP uptake and effective use. These modified implementation strategies highlight opportunities to streamline the delivery of PrEP, as well as other HIV and chronic care services, and strengthen engagement with populations post-pandemic.

The Acceptability of Pharmacy-Based HPV Vaccination in Western Kenya among Pharmacy Clients and Providers.

Vaccine coverage for the human papillomavirus (HPV) remains low globally, and differentiated models of vaccine delivery are needed to expand access. Pharmacy-based models of the HPV vaccination may engage women who could benefit. We assessed the acceptability of such a model among pharmacy clients and providers at 20 private pharmacies in Kisumu County, Kenya. In questionnaires, participants (≥18 years) were asked the extent they agreed (5-point scale) with statements that assessed different acceptability component constructs outlined in the Theoretical Framework of Acceptability (TFA). From March to June 2022, 1500 pharmacy clients and 40 providers were enrolled and completed questionnaires. Most clients liked the intervention (TFA: affective attitude; 96%, 1435/1500) and did not think it would be hard to obtain (TFA: burden; 93%, 1399/1500). All providers agreed the intervention could reduce HPV infection (TFA: perceived effectiveness) and felt confident they could deliver it (TFA: self-efficacy). Among the clients who had received or were planning to receive the HPV vaccine in the future, half (50%, 178/358) preferred a pharmacy-based HPV vaccination. In this study, most Kenyan pharmacy clients and providers perceived a pharmacy-delivered HPV vaccination as highly acceptable; however, more research is needed to test the feasibility and effectiveness of this novel vaccine delivery model in Africa.

Using an on-site modular training approach to amplify prep service delivery in public health facilities in Kenya.

Delivery of oral PrEP, a potent HIV prevention intervention, has begun within public health systems in many countries in Africa. Training as many health providers as possible expeditiously is necessary to efficiently and rapidly scale up PrEP delivery among at risk populations and thereby realize the greatest impact of PrEP. We designed and implemented an innovative on-site modular training approach delivered in five two-hour modules. The modules could be covered in two consecutive days or be broken across several days enabling flexibility to accommodate health provider work schedules. We assessed knowledge gain comparing pre-and post-training test scores and determined monthly PrEP uptake for six months following the training intervention. We also evaluated the cost of this training approach and conducted key informant interviews to explore acceptability among health providers. Between January 2019 and December 2020, 2111 health providers from 104 health facilities were trained on PrEP. Of 1821 (83%) providers who completed both pre- and post-tests, 505 (28%) were nurses, 333 (18%) were HIV counsellors, 276 (15%) were clinical officers and 255 (14%) were lay providers. The mean score prior to and after training was 58% and 82% respectively (p <0.001). On average, health facilities initiated an average of 2.7 (SD 4.7) people on PrEP each month after the training, a number that did not decline over six months post-training (p = 0.62). Assuming Ministry of Health costs, the costs per provider trained was $16.27. Health providers expressed satisfaction with this training approach because it enabled many providers within a facility receive training. On-site modular training is an effective approach for improving PrEP education for health workers in public health facilities, It is also acceptable and low-cost. This method of training can be scaled up to rapidly amplify the number of health workers able to offer PrEP services.

A one-stop shop model for improved efficiency of pre-exposure prophylaxis delivery in public clinics in western Kenya: a mixed methods implementation science study.

INTRODUCTION: In public clinics in Kenya, separate, sequential delivery of the component services of pre-exposure prophylaxis (PrEP) (e.g. HIV testing, counselling, and dispensing) creates long wait times that hinder clients' ability and desire to access and continue PrEP. We conducted a mixed methods study in four public clinics in western Kenya to identify strategies for operationalizing a one-stop shop (OSS) model and evaluate whether this model could improve client wait time and care acceptability among clients and providers without negatively impacting uptake or continuation. METHODS: From January 2020 through November 2020, we collected and analysed 47 time-and-motion observations using Mann-Whitney U tests, 29 provider and client interviews, 68 technical assistance reports, and clinic flow maps from intervention clinics. We used controlled interrupted time series (cITS) to compare trends in PrEP initiation and on-time returns from a 12-month pre-intervention period (January-December 2019) to an 8-month post-period (January-November 2020, excluding a 3-month COVID-19 wash-out period) at intervention and control clinics. RESULTS: From the pre- to post-period, median client wait time at intervention clinics dropped significantly from 31 to 6 minutes (p = 0.02), while median provider contact time remained around 23 minutes (p = 0.4). Intervention clinics achieved efficiency gains by moving PrEP delivery to lower volume departments, moving steps closer together (e.g. relocating supplies; cross-training and task-shifting), and differentiating clients based on the subset of services needed. Clients and providers found the OSS model highly acceptable and additionally identified increased privacy, reduced stigma, and higher quality client-provider interactions as benefits of the model. From the pre- to post-period, average monthly initiations at intervention and control clinics increased by 6 and 2.3, respectively, and percent of expected follow-up visits occurring on time decreased by 18% and 26%, respectively; cITS analysis of PrEP initiations (n = 1227) and follow-up visits (n = 2696) revealed no significant difference between intervention and control clinics in terms of trends in PrEP initiation and on-time returns (all p>0.05). CONCLUSIONS: An OSS model significantly improved client wait time and care acceptability without negatively impacting initiations or continuations, thus highlighting opportunities to improve the efficiency of PrEP delivery efficiency and client-centredness.

HIV risk and pre-exposure prophylaxis interest among women seeking post-abortion care in Kenya: a cross-sectional study.

INTRODUCTION: Post-abortion clinics located in regions with high HIV burden may ideal locations to integrate counselling and delivery of HIV pre-exposure prophylaxis (PrEP), aligning with normative goals for integrated delivery of HIV and reproductive health care. The objective of this study was to gauge the degree to which Kenyan women seeking care for a pregnancy loss, including induced abortion, are at risk for HIV and whether women would welcome an introduction to PrEP prior to discharge from post-abortion care. METHODS: We conducted a mixed-methods study from August 2019 to February 2020 with women ages 15 to 30 recruited sequentially as they were accessing post-abortion care at public and private facilities in Thika and Kisumu, Kenya. Data collection was through a cross-sectional survey and laboratory testing for common sexually transmitted infections (N = 200), and in-depth interviews (N = 30). Descriptive statistics summarize PrEP knowledge and referrals and a multivariable log-link binomial model estimated correlates of receiving a referral for PrEP. Qualitative data were analysed using inductive and deductive approaches. RESULTS: Among 200 HIV-negative women (median age 21.0, interquartile range 19.0 to 22.0), the prevalence of Chlamydia trachomatis was 18.2% and Neisseria gonorrhoeae was 2.0%. Half of the women scored ≥5 on a validated tool that would correspond to an expected HIV incidence of 9.5% per year. Approximately half (55.8%) of women were familiar with PrEP prior to the study and 33.3% received a referral from study staff to a clinic offering PrEP. In qualitative interviews, women expressed interest in accessing PrEP from the gynaecology ward that provided post-abortion care but they preferred alternative locations for PrEP refills. CONCLUSIONS: Kenyan women accessing post-abortion care have substantial HIV risk and were favourable about the idea of receiving support to initiate PrEP as part of care offered during post-abortion care. These settings can be integrated into national PrEP programmes as locations providing PrEP referrals and initiation.