Development and Validation of a Sociodemographic and Behavioral Characteristics-Based Risk-Score Algorithm for Targeting HIV Testing Among Adults in Kenya.

To inform targeted HIV testing, we developed and externally validated a risk-score algorithm that incorporated behavioral characteristics. Outpatient data from five health facilities in western Kenya, comprising 19,458 adults ≥ 15 years tested for HIV from September 2017 to May 2018, were included in univariable and multivariable analyses used for algorithm development. Data for 11,330 adults attending one high-volume facility were used for validation. Using the final algorithm, patients were grouped into four risk-score categories: ≤ 9, 10-15, 16-29 and ≥ 30, with increasing HIV prevalence of 0.6% [95% confidence interval (CI) 0.46-0.75], 1.35% (95% CI 0.85-1.84), 2.65% (95% CI 1.8-3.51), and 15.15% (95% CI 9.03-21.27), respectively. The algorithm's discrimination performance was modest, with an area under the receiver-operating-curve of 0.69 (95% CI 0.53-0.84). In settings where universal testing is not feasible, a risk-score algorithm can identify sub-populations with higher HIV-risk to be prioritized for HIV testing.

Mapping geographic clusters of new HIV diagnoses to inform granular-level interventions for HIV epidemic control in western Kenya.

BACKGROUND: As countries make progress towards HIV epidemic control, there is increasing need to identify finer geographic areas to target HIV interventions. We mapped geographic clusters of new HIV diagnoses, and described factors associated with HIV-positive diagnosis, in order to inform targeting of HIV interventions to finer geographic areas and sub-populations. METHODS: We analyzed data for clients aged > 15 years who received home-based HIV testing as part of a routine public health program between May 2016 and July 2017 in Siaya County, western Kenya. Geospatial analysis using Kulldorff's spatial scan statistic was used to detect geographic clusters (radius < 5 kilometers) of new HIV diagnoses. Factors associated with new HIV diagnosis were assessed in a spatially-integrated Bayesian hierarchical model. RESULTS: Of 268,153 clients with HIV test results, 2906 (1.1%) were diagnosed HIV-positive. We found spatial variation in the distribution of new HIV diagnoses, and identified nine clusters in which the number of new HIV diagnoses was significantly (1.56 to 2.64 times) higher than expected. Sub-populations with significantly higher HIV-positive yield identified in the multivariable spatially-integrated Bayesian model included: clients aged 20-24 years [adjusted relative risk (aRR) 3.45, 95% Bayesian Credible Intervals (CI) 2.85-4.20], 25-35 years (aRR 4.76, 95% CI 3.92-5.81) and > 35 years (aRR 2.44, 95% CI 1.99-3.00); those in polygamous marriage (aRR 1.84, 95% CI 1.55-2.16), or separated/divorced (aRR 3.36, 95% CI 2.72-4.08); and clients who reported having never been tested for HIV (aRR 2.35, 95% CI 2.02-2.72), or having been tested > 12 months ago (aRR 1.53, 95% CI 1.41-1.66). CONCLUSION: Our study used routine public health program data to identify granular geographic clusters of higher new HIV diagnoses, and sub-populations with higher HIV-positive yield in the setting of a generalized HIV epidemic. In order to target HIV testing and prevention interventions to finer granular geographic areas for maximal epidemiologic impact, integrating geospatial analysis into routine public health programs can be useful.

Progress with Scale-Up of HIV Viral Load Monitoring - Seven Sub-Saharan African Countries, January 2015-June 2016.

The World Health Organization (WHO) recommends viral load testing as the preferred method for monitoring the clinical response of patients with human immunodeficiency virus (HIV) infection to antiretroviral therapy (ART) (1). Viral load monitoring of patients on ART helps ensure early diagnosis and confirmation of ART failure and enables clinicians to take an appropriate course of action for patient management. When viral suppression is achieved and maintained, HIV transmission is substantially decreased, as is HIV-associated morbidity and mortality (2). CDC and other U.S. government agencies and international partners are supporting multiple countries in sub-Saharan Africa to provide viral load testing of persons with HIV who are on ART. This report examines current capacity for viral load testing based on equipment provided by manufacturers and progress with viral load monitoring of patients on ART in seven sub-Saharan countries (Côte d'Ivoire, Kenya, Malawi, Namibia, South Africa, Tanzania, and Uganda) during January 2015-June 2016. By June 2016, based on the target numbers for viral load testing set by each country, adequate equipment capacity existed in all but one country. During 2015, two countries tested >85% of patients on ART (Namibia [91%] and South Africa [87%]); four countries tested <25% of patients on ART. In 2015, viral suppression was >80% among those patients who received a viral load test in all countries except Côte d'Ivoire. Sustained country commitment and a coordinated global effort is needed to reach the goal for viral load monitoring of all persons with HIV on ART.

Scale-up of HIV Viral Load Monitoring--Seven Sub-Saharan African Countries.

To achieve global targets for universal treatment set forth by the Joint United Nations Programme on human immunodeficiency virus (HIV)/acquired immunodeficiency syndrome (AIDS) (UNAIDS), viral load monitoring for HIV-infected persons receiving antiretroviral therapy (ART) must become the standard of care in low- and middle-income countries (LMIC) (1). CDC and other U.S. government agencies, as part of the President's Emergency Plan for AIDS Relief, are supporting multiple countries in sub-Saharan Africa to change from the use of CD4 cell counts for monitoring of clinical response to ART to the use of viral load monitoring, which is the standard of care in developed countries. Viral load monitoring is the preferred method for immunologic monitoring because it enables earlier and more accurate detection of treatment failure before immunologic decline. This report highlights the initial successes and challenges of viral load monitoring in seven countries that have chosen to scale up viral load testing as a national monitoring strategy for patients on ART in response to World Health Organization (WHO) recommendations. Countries initiating viral load scale-up in 2014 observed increases in coverage after scale-up, and countries initiating in 2015 are anticipating similar trends. However, in six of the seven countries, viral load testing coverage in 2015 remained below target levels. Inefficient specimen transport, need for training, delays in procurement and distribution, and limited financial resources to support scale-up hindered progress. Country commitment and effective partnerships are essential to address the financial, operational, technical, and policy challenges of the rising demand for viral load monitoring.

HIV Risk Factors and Risk Perception Among Adolescent Girls and Young Women: Results From a Population-Based Survey in Western Kenya, 2018.

BACKGROUND: In sub-Saharan Africa, HIV prevalence in adolescent girls and young women (AGYW) is 2-fold to 3-fold higher than that in adolescent boys and young men. Understanding AGYW's perception of HIV risk is essential for HIV prevention efforts. METHODS: We analyzed data from a HIV biobehavioral survey conducted in western Kenya in 2018. Data from AGYW aged 15-24 years who had a documented HIV status were included. We calculated weighted prevalence and evaluated factors associated with outcomes of interest (HIV infection and high risk perception) using generalized linear models to calculate prevalence ratios. RESULTS: A total of 3828 AGYW were included; 63% were aged 15-19 years. HIV prevalence was 4.5% and 14.5% of sexually active AGYW had high risk perception. Over 70% of participants had accessed HIV testing and counseling in the past 12 months. Factors associated with both HIV infection and high risk perception included having an HIV-positive partner or partner with unknown status and having a sexually transmitted infection in the past 12 months. Having an older (by ≥10 years) partner was associated with HIV infection, but not high risk perception. Less than 30% of sexually active AGYW with 3 or more HIV risk factors had high perception of HIV risk. CONCLUSION: Gaps in perceived HIV risk persist among AGYW in Kenya. High access to HIV testing and prevention services in this population highlights platforms through which AGYW may be reached with improved risk counseling, and to increase uptake of HIV prevention strategies.

They are likely to be there: using a family-centered index testing approach to identify children living with HIV in Kenya.

In Kenya, only half of children with a parent living with HIV have been tested for HIV. The effectiveness of family-centered index testing to identify children (0-14 years) living with HIV was examined. A retrospective record review was conducted among adult index patients newly enrolled in HIV care between May and July 2015; family testing, results, and linkage to treatment outcomes were followed through May 2016 at 60 high-volume clinics in Kenya. Chi square test compared yield (percentage of HIV tests positive) among children tested through family-centered index testing, outpatient and inpatient testing. Review of 1937 index client charts led to 3005 eligible children identified for testing. Of 2848 (94.8%) children tested through family-centered index testing, 127 (4.5%) had HIV diagnosed, 100 (78.7%) were linked to care, and 85 of those eligible (91.4%) initiated antiretroviral therapy (ART).Family testing resulted in higher yield compared to inpatient (1.8%, p < 0.001) or outpatient testing (1.6%, p < 0.001). The absolute number of children living with HIV identified was highest with outpatient testing. The relative contribution of testing approach to total children identified with HIV was outpatient testing (69%), family testing (26%), and inpatient testing (5%). The family testing approach demonstrated promise in achieving the first two "90s" (identification and ART initiation) of the 90-90-90 targets for children, with additional effort required to improve linkage from testing to treatment.

High yield of new HIV diagnoses during active case-finding for tuberculosis.

OBJECTIVE: To evaluate the utility of a broad and nonspecific symptom screen for identifying people with undiagnosed HIV infection. DESIGN: Secondary analysis of operational data collected during implementation of a cluster-randomized trial for tuberculosis case detection. METHODS: As part of the trial, adults reporting cough, fever, night sweats, weight loss, or difficulty breathing for any duration in the past month were identified in health facilities and community-based mobile screening units in western Kenya. Adults reporting any symptom were offered HIV testing. We analysed the HIV testing data from this study, using modified Poisson regression, to identify predictors of new HIV diagnoses among adults with symptoms and initially unknown HIV status. RESULTS: We identified 3818 symptomatic adults, referred 1424 (37%) for testing, of whom 1065 (75%) accepted, and 107 (10%) were newly diagnosed with HIV. The prevalence of new HIV diagnoses was 21% [95% confidence interval (CI) 17-25%] among those tested in health facilities and 5% (95% CI 4-7%) among those tested in mobile units. More men were diagnosed with HIV than women, despite fewer men being screened. People who reported 4-5 symptoms were over twice as likely to be diagnosed with HIV compared to those reporting 1-3 symptoms (adjusted prevalence ratio in health facilities = 2.58, 95% CI 1.65-4.05; adjusted prevalence ratio in mobile units = 2.63, 95% CI 1.37-5.03). CONCLUSION: We observed a high yield of new HIV diagnoses among adults identified by active application of a broad symptom screen. Use of integrated tuberculosis and HIV screening could help close the detection gap for both conditions.

Expanded eligibility for HIV testing increases HIV diagnoses-A cross-sectional study in seven health facilities in western Kenya.

Homa Bay, Siaya, and Kisumu counties in western Kenya have the highest estimated HIV prevalence (16.3-21.0%) in the country, and struggle to meet program targets for HIV testing services (HTS). The Kenya Ministry of Health (MOH) recommends annual HIV testing for the general population. We assessed the degree to which reducing the interval for retesting to less than 12 months increased diagnosis of HIV in outpatient departments (OPD) in western Kenya. We conducted a retrospective analysis of routinely collected program data from seven high-volume (>800 monthlyOPD visits) health facilities in March-December, 2017. Data from persons ≥15 years of age seeking medical care (patients) in the OPD and non-care-seekers (non-patients) accompanying patients to the OPD were included. Outcomes were meeting MOH (routine) criteria versus criteria for a reduced retesting interval (RRI) of <12 months, and HIV test result. STATA version 14.2 was used to calculate frequencies and proportions, and to test for differences using bivariate analysis. During the 9-month period, 119,950 clients were screened for HIV testing eligibility, of whom 79% (94,766) were eligible and 97% (92,153) received a test. Among 92,153 clients tested, the median age was 28 years, 57% were female and 40% (36,728) were non-patients. Overall, 20% (18,120) of clients tested met routine eligibility criteria: 4% (3,972) had never been tested, 10% (9,316) reported a negative HIV test in the past >12 months, and 5% (4,832) met other criteria. The remaining 80% (74,033) met criteria for a RRI of < 12 months. In total 1.3% (1,185) of clients had a positive test. Although the percent yield was over 2-fold higher among those meeting routine criteria (2.4% vs. 1.0%; p<0.001), 63% (750) of all HIV infections were found among clients tested less than 12 months ago, the majority (81%) of whom reported having a negative test in the past 3-12 months. Non-patients accounted for 45% (539) of all HIV-positive persons identified. Percent yield was higher among non-patients as compared to patients (1.5% vs. 1.2%; p-value = <0.001) overall and across eligibility criteria and age categories. The majority of HIV diagnoses in the OPD occurred among clients reporting a negative HIV test in the past 12 months, clients ineligible for testing under the current MOH guidelines. Nearly half of all HIV-positive individuals identified in the OPD were non-patients. Our findings suggest that in the setting of a generalized HIV epidemic, retesting persons reporting an HIV-negative test in the past 3-12 months, and routine testing of non-patients accessing the OPD are key strategies for timely diagnosis of persons living with HIV.

Spatial-temporal trend for mother-to-child transmission of HIV up to infancy and during pre-Option B+ in western Kenya, 2007-13.

INTRODUCTION: Using spatial-temporal analyses to understand coverage and trends in elimination of mother-to-child transmission of HIV (e-MTCT) efforts may be helpful in ensuring timely services are delivered to the right place. We present spatial-temporal analysis of seven years of HIV early infant diagnosis (EID) data collected from 12 districts in western Kenya from January 2007 to November 2013, during pre-Option B+ use. METHODS: We included in the analysis infants up to one year old. We performed trend analysis using extended Cochran-Mantel-Haenszel stratified test and logistic regression models to examine trends and associations of infant HIV status at first diagnosis with: early diagnosis (<8 weeks after birth), age at specimen collection, infant ever having breastfed, use of single dose nevirapine, and maternal antiretroviral therapy status. We examined these covariates and fitted spatial and spatial-temporal semiparametric Poisson regression models to explain HIV-infection rates using R-integrated nested Laplace approximation package. We calculated new infections per 100,000 live births and used Quantum GIS to map fitted MTCT estimates for each district in Nyanza region. RESULTS: Median age was two months, interquartile range 1.5-5.8 months. Unadjusted pooled positive rate was 11.8% in the seven-years period and declined from 19.7% in 2007 to 7.0% in 2013, p < 0.01. Uptake of testing ≤8 weeks after birth was under 50% in 2007 and increased to 64.1% by 2013, p < 0.01. By 2013, the overall standardized MTCT rate was 447 infections per 100,000 live births. Based on Bayesian deviance information criterion comparisons, the spatial-temporal model with maternal and infant covariates was best in explaining geographical variation in MTCT. DISCUSSION: Improved EID uptake and reduced MTCT rates are indicators of progress towards e-MTCT. Cojoined analysis of time and covariates in a spatial context provides a robust approach for explaining differences in programmatic impact over time. CONCLUSION: During this pre-Option B+ period, the prevention of mother to child transmission program in this region has not achieved e-MTCT target of ≤50 infections per 100,000 live births. Geographical disparities in program achievements may signify gaps in spatial distribution of e-MTCT efforts and could indicate areas needing further resources and interventions.

Performance of Clinical Screening Algorithms for Tuberculosis Intensified Case Finding among People Living with HIV in Western Kenya.

OBJECTIVE: To assess the performance of symptom-based screening for tuberculosis (TB), alone and with chest radiography among people living with HIV (PLHIV), including pregnant women, in Western Kenya. DESIGN: Prospective cohort study. METHODS: PLHIV from 15 randomly-selected HIV clinics were screened with three clinical algorithms [World Health Organization (WHO), Ministry of Health (MOH), and "Improving Diagnosis of TB in HIV-infected persons" (ID-TB/HIV) study], underwent chest radiography (unless pregnant), and provided two or more sputum specimens for smear microscopy, liquid culture, and Xpert MTB/RIF. Performance of clinical screening was compared to laboratory results, controlling for the complex design of the survey. RESULTS: Overall, 738 (85.6%) of 862 PLHIV enrolled were included in the analysis. Estimated TB prevalence was 11.2% (95% CI, 9.9-12.7). Sensitivity of the three screening algorithms was similar [WHO, 74.1% (95% CI, 64.1-82.2); MOH, 77.5% (95% CI, 68.6-84.5); and ID-TB/HIV, 72.5% (95% CI, 60.9-81.7)]. Sensitivity of the WHO algorithm was significantly lower among HIV-infected pregnant women [28.2% (95% CI, 14.9-46.7)] compared to non-pregnant women [78.3% (95% CI, 67.3-86.4)] and men [77.2% (95% CI, 68.3-84.2)]. Chest radiography increased WHO algorithm sensitivity and negative predictive value to 90.9% (95% CI, 86.4-93.9) and 96.1% (95% CI, 94.4-97.3), respectively, among asymptomatic men and non-pregnant women. CONCLUSIONS: Clinical screening missed approximately 25% of laboratory-confirmed TB cases among all PLHIV and more than 70% among HIV-infected pregnant women. National HIV programs should evaluate the feasibility of laboratory-based screening for TB, such as a single Xpert MTB/RIF test for all PLHIV, especially pregnant women, at enrollment in HIV services.

Positive Predictive Value of the WHO Clinical and Immunologic Criteria to Predict Viral Load Failure among Adults on First, or Second-Line Antiretroviral Therapy in Kenya.

Routine HIV viral load (VL) monitoring is the standard of care for persons receiving antiretroviral therapy (ART) in developed countries. Although the World Health Organization recommends annual VL monitoring of patients on ART, recognizing difficulties in conducting routine VL testing, the WHO continues to recommend targeted VL testing to confirm treatment failure for persons who meet selected immunologic and clinical criteria. Studies have measured positive predictive value (PPV), negative predictive value, sensitivity and specificity of these criteria among patients receiving first-line ART but not specifically among those on second-line or subsequent regimens. Between 2008 and 2011, adult ART patients in Nyanza, Kenya who met national clinical or immunologic criteria for treatment failure received targeted VL testing. We calculated PPV and 95% confidence intervals (CI) of these criteria to detect virologic treatment failure among patients receiving a) first-line ART, b) second/subsequent ART, and c) any regimen. Of 12,134 patient specimens tested, 2,874 (23.7%) were virologically confirmed as treatment failures. The PPV for 2,834 first-line ART patients who met either the clinical or immunologic criteria for treatment failure was 34.4% (95% CI 33.2-35.7), 33.1% (95% CI 24.7-42.3) for the 40 patients on second-line/subsequent regimens, and 33.4% (95% CI 33.1-35.6) for any ART. PPV, regardless of criteria, for first-line ART patients was lowest among patients over 44 years old and highest for patients aged 15 to 34 years. PPV of immunological and clinical criteria for correctly identifying treatment failure was similarly low for adult patients receiving either first-line or second-line/subsequent ART regimens. Our data confirm the inadequacy of clinical and immunologic criteria to correctly identify treatment failure and support the implementation of routine VL testing.

Comparison of trends in tuberculosis incidence among adults living with HIV and adults without HIV--Kenya, 1998-2012.

BACKGROUND: In Kenya, the comparative incidences of tuberculosis among persons with and without HIV have not been described, and the differential impact of public health interventions on tuberculosis incidence in the two groups is unknown. METHODS: We estimated annual tuberculosis incidence stratified by HIV status during 2006-2012 based on the numbers of reported tuberculosis patients with and without HIV infection, the prevalence of HIV infection in the general population, and the total population. We also made crude estimates of annual tuberculosis incidence stratified by HIV status during 1998-2012 by assuming a constant ratio of HIV prevalence among tuberculosis patients compared to the general population. RESULTS: Tuberculosis incidence among both adults with HIV and adults without HIV increased during 1998-2004 then remained relatively stable until 2007. During 2007-2012, tuberculosis incidence declined by 28-44% among adults with HIV and by 11-26% among adults without HIV, concurrent with an increase in antiretroviral therapy uptake. In 2012, tuberculosis incidence among adults with HIV (1,839-1,936 cases/100,000 population) was still eight times as high as among adults without HIV (231-238 cases/100,000 population), and approximately one third of tuberculosis cases were attributable to HIV. CONCLUSIONS: Although tuberculosis incidence has declined among adults with and without HIV, the persistent high incidence of tuberculosis among those with HIV and the disparity between the two groups are concerning. Early diagnosis of HIV, early initiation of antiretroviral therapy, regular screening for tuberculosis, and isoniazid preventive therapy among persons with HIV, as well as tuberculosis control in the general population, are required to address these issues.

The adult population impact of HIV care and antiretroviral therapy in a resource poor setting, 2003-2008.

OBJECTIVE: To describe the population uptake of HIV care including antiretroviral therapy (ART) and its impact on adult mortality in a rural area of western Kenya with high HIV prevalence during a period of rapid HIV services scale-up. DESIGN: Adult medical chart data were abstracted at health facilities providing HIV care/ART to residents of a Health and Demographic Surveillance System (HDSS) and linked with HDSS demographic and mortality data. METHODS: We evaluated secular trends in patient characteristics across enrollment years and estimated proportions of HIV-positive adult residents receiving care. We evaluated adult (18-64 years) population mortality trends using verbal autopsy findings. RESULTS: From 2003 to 2008, 5421 HDSS-resident adults enrolled in HIV care; 61.4% (n=3331) were linked to HDSS follow-up data. As the number of facilities expanded from 1 (2003) to 17 (2008), receipt of HIV services by HIV-positive residents increased from less than 1 to 29.5%, and ART coverage reached 64.0% of adults with CD4 cell count less than 250 cells/µl. The proportion of patients with WHO stage 4 at enrollment decreased from 20.4 to 1.9%, and CD4 cell count testing at enrollment increased from 1.0 to 53.4%. Population-level mortality rates for adults declined 34% for all causes, 26% for AIDS/tuberculosis, and 47% for other infectious diseases; noninfectious disease mortality rates remained constant. CONCLUSION: The initial years of rapid HIV service expansion coincided with a drop in adult mortality by a third. Continued expansion of population access to HIV clinical services, including ART, and program quality improvements will be necessary to achieve further progress in reducing HIV-related morbidity and mortality.

Family model of HIV care and treatment: a retrospective study in Kenya.

BACKGROUND: Nyanza Province, Kenya, had the highest HIV prevalence in the country at 14.9% in 2007, more than twice the national HIV prevalence of 7.1%. Only 16% of HIV-infected adults in the country accurately knew their HIV status. Targeted strategies to reach and test individuals are urgently needed to curb the HIV epidemic. The family unit is one important portal. METHODS: A family model of care was designed to build on the strengths of Kenyan families. Providers use a family information table (FIT) to guide index patients through the steps of identifying family members at HIV risk, address disclosure, facilitate family testing, and work to enrol HIV-positive members and to prevent new infections. Comprehensive family-centred clinical services are built around these steps. To assess the approach, a retrospective study of patients receiving HIV care between September 2007 and September 2009 at Lumumba Health Centre in Kisumu was conducted. A random sample of FITs was examined to assess family reach. RESULTS: Through the family model of care, for each index patient, approximately 2.5 family members at risk were identified and 1.6 family members were tested. The approach was instrumental in reaching children; 61% of family members identified and tested were children. The approach also led to identifying and enrolling a high proportion of HIV- positive partners among those tested: 71% and 89%, respectively. CONCLUSIONS: The family model of care is a feasible approach to broaden HIV case detection and service reach. The approach can be adapted for the local context and should continue to utilize index patient linkages, FIT adaption, and innovative methods to package services for families in a manner that builds on family support and enhances patient care and prevention efforts. Further efforts are needed to increase family member engagement.

Profile: the KEMRI/CDC Health and Demographic Surveillance System--Western Kenya.

The KEMRI/Centers for Disease Control and Prevention (CDC) Health and Demographic Surveillance System (HDSS) is located in Rarieda, Siaya and Gem Districts (Siaya County), lying northeast of Lake Victoria in Nyanza Province, western Kenya. The KEMRI/CDC HDSS, with approximately 220 000 inhabitants, has been the foundation for a variety of studies, including evaluations of insecticide-treated bed nets, burden of diarrhoeal disease and tuberculosis, malaria parasitaemia and anaemia, treatment strategies and immunological correlates of malaria infection, and numerous HIV, tuberculosis, malaria and diarrhoeal disease treatment and vaccine efficacy and effectiveness trials for more than a decade. Current studies include operations research to measure the uptake and effectiveness of the programmatic implementation of integrated malaria control strategies, HIV services, newly introduced vaccines and clinical trials. The HDSS provides general demographic and health information (such as population age structure and density, fertility rates, birth and death rates, in- and out-migrations, patterns of health care access and utilization and the local economics of health care) as well as disease- or intervention-specific information. The HDSS also collects verbal autopsy information on all deaths. Studies take advantage of the sampling frame inherent in the HDSS, whether at individual, household/compound or neighbourhood level.