Failure to return pillbox is a predictor of being lost to follow-up among people living with HIV on antiretroviral therapy in rural Tanzania.

OBJECTIVES: Pill count is used to assess drug adherence in people living with HIV (PLHIV). Carrying a pillbox is associated with fear of concealment and stigma and might indicate poor adherence and predict someone who will be lost to follow-up (LTFU). We therefore assessed the association between pillbox return and being LTFU in rural Tanzania. METHODS: This is a nested study of the Kilombero and Ulanga Antiretroviral Cohort (KIULARCO). We included PLHIV aged ≥ 18 years enrolled in KIULARCO between January 2013 and March 2019 with follow-up through January 2020, who were on antiretroviral treatment (ART) for ≥ 6 months. Baseline was defined as the latest ART initiation or KIULARCO enrolment. We determined the association between time-dependent failed pillbox return updated at every visit and LTFU using Kaplan-Meier estimation and Cox models. RESULTS: Among 2552 PLHIV included in the study, 1735 (68.0%) were female, 959 (40.3%) had a WHO stage III/IV and 1487 (66.4%) had a CD4 cell count < 350 cells/µL. The median age was 38.4 years [interquartile range (IQR): 31.7-46.2]. During a median follow-up of 33.1 months (IQR: 17.5-52.4), 909 (35.6%) participants were LTFU, 43 (1.7%) died and 194 (7.6%) had transferred to another clinic. The probability of being LTFU was higher among PLHIV with failed pillbox return than among those who returned their pillbox [30.0%, 95% confidence interval (CI): 26.8-33.2% vs. 19.4%, 95% CI: 17.4-21.6%, respectively, at 24 months (hazard ratio = 1.67, 95% CI: 1.46-1.90; p < 0.001)]. CONCLUSIONS: Failed pillbox return was associated with a higher risk of being LTFU and could be used as a simple tool to identify PLHIV for appropriate interventions to reduce their chance of being LTFU.

Machine learning with routine electronic medical record data to identify people at high risk of disengagement from HIV care in Tanzania.

Machine learning methods for health care delivery optimization have the potential to improve retention in HIV care, a critical target of global efforts to end the epidemic. However, these methods have not been widely applied to medical record data in low- and middle-income countries. We used an ensemble decision tree approach to predict risk of disengagement from HIV care (missing an appointment by ≥28 days) in Tanzania. Our approach used routine electronic medical records (EMR) from the time of antiretroviral therapy (ART) initiation through 24 months of follow-up for 178 adults (63% female). We compared prediction accuracy when using EMR-based predictors alone and in combination with sociodemographic survey data collected by a research study. Models that included only EMR-based indicators and incorporated changes across past clinical visits achieved a mean accuracy of 75.2% for predicting risk of disengagement in the next 6 months, with a mean sensitivity of 54.7% for targeting the 30% highest-risk individuals. Additionally including survey-based predictors only modestly improved model performance. The most important variables for prediction were time-varying EMR indicators including changes in treatment status, body weight, and WHO clinical stage. Machine learning methods applied to existing EMR data in resource-constrained settings can predict individuals' future risk of disengagement from HIV care, potentially enabling better targeting and efficiency of interventions to promote retention in care.

Duration on ART, Alcohol Use and HIV Stage May Predict Risky Sexual Behavior in a Resource-limited Environment: A Cross-sectional Study.

BACKGROUND: The intention of antiretroviral therapy (ART) and regular clinic visits are to engender safe sex attitudes among HIV-infected individuals. However, this may not be the case due to the perceived therapeutic benefits of ART and may result in exposure to drug-resistant HIV strains. OBJECTIVES: We aimed to determine the prevalence and predict the factors associated with risky sexual behaviors among ART users in a resource-limited environment. METHODS: Two hundred and ninety-one sexually active ART-users aged 18-50 years and seeking care at the HIV clinic in Dodoma, Tanzania, participated in this study. The outcome variables modeled in a logistic regression were condom use, multiple sex partners, casual sex partners, and payment for sex. The predictors included in the models were the patients' socio-demographic characteristics. In addition, a new variable, sexual risk scores, was generated by culminating all the outcome variables. Finally, a multiple Poisson regression with the socio-demographic variables of the participants was used to model the sexual risk scores. RESULTS: Patients reported inconsistent/no condom use (44%), payment for sex (4%), casual sex encounters (13%), multiple sex partners (21%), and STD symptoms (15%). While having a casual sexual partner was significantly associated with age group in a Pearson Chi-square (p=0.0147), participants ≤35 years old were less likely to have single-sex partners than older participants (AOR: 0.188, 95 C.I: 0.042-0.849). Meanwhile, the likelihood of condom use was higher among participants with no HIV-infected family members (AOR= 2.409, 95% C.I:1.236,4.697) and among participants who had single-sex partners (AOR= 2.721, 95% C.I.: 1.115,6.640); these participants were less likely to report STD symptoms (AOR=0.265, 95% C.I.: 0.081-0.865). Adjusted analysis showed that estimated mean sexual risk scores significantly increased (mean, λ=1.61, 95% C.I:1.0817-2.4063) for recent ART recipients (within 1-3 years vs. ≥8 years). However, sexual risk scores of participants with HIV stage 3 were 38.8% lower than participants at stage 4 (95% C.I.: 0.3910-0.9558), and non-alcohol drinkers had an adjusted mean sexual risk score of 29% lower than participants who were alcohol drinkers (95% C.I.: 0.5102-0.9879). CONCLUSION: Researchers should prioritize patients at HIV CTC for education concerning safe sexual practices for those characterized by alcohol consumption, younger age (less than 35 years old), HIV stage 4, or commencement of ART within three years.

Cervical Cancer Screening among Women Receiving Antiretroviral Therapy in a Resource-Limited Environment.

BACKGROUND: Cervical cancer is among the most prevalent cancer among women worldwide and women living with HIV are at increased risk, especially in a resource-limited environment. OBJECTIVE: This study aimed to determine levels of awareness, knowledge, uptake, and willingness to screen for cervical cancer among women receiving care in an HIV clinic at Dodoma Regional Referral Hospital (DRRH), Tanzania. METHODS: Data were collected for a period of three weeks from July 21 to August 11, 2017 using a mobile phone data collection App. A total of 421 Women aged 18-50 years old were included in the study. RESULTS: Majority of the women interviewed (n=306, 73%) were aware of cervical cancer. Among those who were aware, 84% (n=257) did not recall ever being screened for cervical cancer, and majority had a poor knowledge of cervical cancer. Educational level completed (p=0.01), income per month (p=0.02), age group (p<0.0001), and area of residence (p<0.0001) were all significantly associated to awareness of cervical cancer. Most of the women who have never screened (n=231, 91%) expressed willingness to be screened. Prior uptake of cervical cancer screening was associated with number of live births (p=0.001) and area of residence (p=0.04). And Willingness to screen was significantly associated with age groups (p=0.03) and the number of live births (p=0.03). Moreover, we found that younger age and urban residence was positively associated with awareness and uptake of cervical cancer screening. Willingness was found to decrease as age increased. CONCLUSION: The study found that despite older women's higher risk of cervical cancer, those who indicated willingness to screen were younger. Additional education, health promotion, and integration of cervical cancer screening services is needed to improve cervical cancer awareness and screening uptake at the HIV clinic.
.