Comparison of logistic regression with regularized machine learning methods for the prediction of tuberculosis disease in people living with HIV: cross-sectional hospital-based study in Kisumu County, Kenya.

Background: Tuberculosis (TB) is a major public health concern, particularly among people living with the Human immunodeficiency Virus (PLWH). Accurate prediction of TB disease in this population is crucial for early diagnosis and effective treatment. Logistic regression and regularized machine learning methods have been used to predict TB, but their comparative performance in HIV patients remains unclear. The study aims to compare the predictive performance of logistic regression with that of regularized machine learning methods for TB disease in HIV patients. Methods: Retrospective analysis of data from HIV patients diagnosed with TB in three hospitals in Kisumu County (JOOTRH, Kisumu sub-county hospital, Lumumba health center) between [dates]. Logistic regression, Lasso, Ridge, Elastic net regression were used to develop predictive models for TB disease. Model performance was evaluated using accuracy, and area under the receiver operating characteristic curve (AUC-ROC). Results: Of the 927 PLWH included in the study, 107 (12.6%) were diagnosed with TB. Being in WHO disease stage III/IV (aOR: 7.13; 95%CI: 3.86-13.33) and having a cough in the last 4 weeks (aOR: 2.34;95%CI: 1.43-3.89) were significant associated with the TB. Logistic regression achieved accuracy of 0.868, and AUC-ROC of 0.744. Elastic net regression also showed good predictive performance with accuracy, and AUC-ROC values of 0.874 and 0.762, respectively. Conclusions: Our results suggest that logistic regression, Lasso, Ridge regression, and Elastic net can all be effective methods for predicting TB disease in HIV patients. These findings may have important implications for the development of accurate and reliable models for TB prediction in HIV patients.

Hypertension and Associated Inflammatory Markers Among HIV-Infected Patients in Tanzania.

There remains a dearth of data regarding the association between chronic inflammation and hypertension (HTN) in sub-Saharan Africa, a region that accounts for >70% of the global burden of HIV infection. Therefore, we assessed the levels of biomarkers among HIV+ individuals and its associations with HTN in Tanzania. A cross-sectional study was conducted at one of the largest clinics in Tanzania and data from 261 HIV+ patients were analyzed. Standardized tools were used to collect data. Blood pressure was measured using Omron® M2 blood pressure monitor. Enzyme-linked immunosorbent assay was used to test for inflammatory markers [C-reactive protein (CRP), interleukin (IL)-6, IL-18, soluble tumor necrosis factor receptor type I (sTNFRI), sTNFRII]. Bivariate and multivariable analysis was conducted to examine association between the biomarkers and HTN. We further conducted age-sex-alcohol-adjusted models to control for any confounders. The prevalence of HTN was 43% with a high prevalence reported in female (70%) participants and those older than 55 years of age (77%). Being women, older than 55 years of age, married, and being overweight was associated with HTN. The highest correlations were observed between TNR2 and CRP (ɤ = 0.13, P = 0.044), and TNR2 and IL-18 (ɤ = 0.13, P = 0.034). Participants who had elevated CRP levels were 2 times more likely to experience HTN in the age-adjusted model [odds ratio (OR) = 3.5, 95% confidence interval (CI) = 1.1-11.3], age-sex-adjusted model (OR = 3.3, 95% CI = 1.0-10.9), and the full model (OR = 2.9, 95% CI = 0.8-10.0). Our study shows that high CRP levels are significantly associated with the higher prevalence of HTN notwithstanding all other markers, which showed a positive association with HTN despite not being significant. These findings point to the importance of creating awareness, education, and screening for HTN among HIV patients in high epidemic countries. More rigorous studies are needed to know the exact pathway mechanisms of inflammation in HIV patients.

Cigarette Smoking Behaviors and Beliefs in Persons Living With HIV in Nairobi, Kenya.

INTRODUCTION: Persons living with HIV (PLWH) use tobacco at higher rates than the general population in both high-income countries and low- and middle-income countries. Tobacco use rates are increasing in sub-Saharan Africa, the home to most of the world's PLWH. As the reach of antiretroviral therapy (ART) expands and HIV-related morbidity and mortality wanes, tobacco use is emerging as a leading cause of disease and death in PLWH. A better understanding of tobacco use behaviors in various settings will be crucial to designing optimal tobacco control strategies. METHODS: In late 2019, we enrolled 50 PLWH cigarette smokers from 6 clinical sites in Nairobi, Kenya (4 HIV care clinics and 2 methadone maintenance programs) for one-on-one interviews focusing on their behaviors and beliefs related to tobacco use. RESULTS: Fifty PLWH smokers completed the interviews. The mean age was 38.5 ± 9.7 years (range 20-57 years) and 68% were male. All were currently receiving ART. They smoked a mean of 14.9 ± 12.4 cigarettes per day, and 82% reported smoking every day. Only 6% reported dual use of smokeless tobacco products. Nicotine dependence was moderate or high in 74%. More than a third (36%) reported a prior history of tuberculosis. In our sample, use of other substances was common, especially alcohol, marijuana, and methadone. On the motivation to quit scale, 90% were at least in the contemplation stage, but only 2% had ever received behavioral cessation counseling, and only 8% had ever used pharmacotherapy (exclusively nicotine replacement therapy). Participants reported significant concern about developing smoking-related illness, exposing others to secondary smoke, and the financial burden associated with their tobacco use. Measures of intrinsic and extrinsic motivation to quit, smoker and abstainer self-concept, and social support yielded encouraging results regarding the possibility of successful quitting. CONCLUSIONS: Tobacco use is an important health concern in PLWH in Kenya. A more thorough understanding of their tobacco use behaviors and beliefs will provide critical information for providers, public health officials, and policy makers as they redouble their efforts to confront this urgent health challenge.