Tuberculosis (TB) among hospitalized patients is underdiagnosed. This study assessed systematic TB-screening, followed by an enhanced TB-diagnostic package for hospitalized patients implemented by trained lay health workers in KwaZulu-Natal, South Africa. In this before-and-after study we included patients ≥ 18 years. The intervention consisted of systematic clinical screening for TB, HIV and diabetes mellitus by lay health workers and provision of an enhanced TB-diagnostic package including sputum Xpert MTB/Rif Ultra, urine lateral-flow lipoarabinomannan assay (LF-LAM), chest x-ray, and sputum culture. We compared TB case findings with people hospitalized one year preceding the intervention. In the pre-intervention phase, 5217 people were hospitalized. Among 4913 (94.2%) people not on TB treatment, 367 (7.5%) were diagnosed with TB. In the intervention phase, 4015 eligible people were hospitalized. Among 3734 (93.0%) people not on TB treatment, 560 (15.0%) were diagnosed with TB. The proportion of patients diagnosed with TB was higher in the intervention phase (15.0% vs. 7.5%, p < 0.001). Overall in-hospital mortality was lower in the intervention phase [166/3734(4.5%) vs. 336/4913(6.8%), p < 0.001]. Lay health worker-led implementation of systematic TB-screening, coupled with provision of an enhanced TB-diagnostic package significantly improved TB case detection and mortality among hospitalized adults.

BACKGROUND

Viral load (VL) suppression is key to ending the global HIV epidemic, and predicting it is critical for healthcare providers and people living with HIV (PLHIV). Traditional research has focused on statistical analysis, but machine learning (ML) is gradually influencing HIV clinical care. While ML has been used in various settings, there’s a lack of research supporting antiretroviral therapy (ART) programs, especially in resource-limited settings like Guinea. This study aims to identify the most predictive variables of VL suppression and develop ML models for PLHIV in Conakry (Guinea).

METHODS

Anonymized data from HIV patients in eight Conakry health facilities were pre-processed, including variable recoding, record removal, missing value imputation, grouping small categories, creating dummy variables, and oversampling the smallest target class. Support vector machine (SVM), logistic regression (LR), naïve Bayes (NB), random forest (RF), and four stacked models were developed. Optimal parameters were determined through two cross-validation loops using a grid search approach. Sensitivity, specificity, predictive positive value (PPV), predictive negative value (PNV), F-score, and area under the curve (AUC) were computed on unseen data to assess model performance. RF was used to determine the most predictive variables.

RESULTS

RF (94% F-score, 82% AUC) and NB (89% F-score, 82% AUC) were the most optimal models to detect VL suppression and non-suppression when applied to unseen data. The optimal parameters for RF were 1,000 estimators and no maximum depth (Random state = 40), and it identified Regimen schedule_6-Month, Duration on ART (months), Last ART CD4, Regimen schedule_Regular, and Last Pre-ART CD4 as top predictors for VL suppression.

CONCLUSION

This study demonstrated the capability to predict VL suppression but has some limitations. The results are dependent on the quality of the data and are specific to the Guinea context and thus, there may be limitations with generalizability. Future studies may be to conduct a similar study in a different context and develop the most optimal model into an application that can be tested in a clinical context.

BACKGROUND

Bedaquiline (BDQ) resistance presents a critical challenge in the fight against tuberculosis (TB), particularly multidrug-resistant (MDR) strains. The emergence of resistance to BDQ, a key drug in treating MDR-TB, poses significant threats to TB treatment effectiveness.

METHODS

The National Institute of Tuberculosis and Respiratory Diseases in Delhi and the Médecins Sans Frontières clinic in Mumbai provide BDQ, delamanid, and carbapenem-based regimens for patients with suspected or confirmed treatment failure. BDQ phenotypic drug-susceptibility testing (DST) was performed for all BDQ-exposed patients. Treatment regimens were individualized based on exposure history, comorbidities, drug interactions, prior adverse drug reactions, and DST results.

RESULTS

Of 117 BDQ-exposed patients from December 2020–December 2022, 42 (36%) exhibited a BDQ-resistant strain. Median (IQR) age was 24 (22–32) years, with 63 (54%) females and 94% with pulmonary TB. Patients with a BDQ-resistant strain were older (median age: 27 vs 23 years; P = .04), more likely to have lung cavities (risk ratio [RR]: 1.8; 95%-CI: 1.1–3.1; P = .02), and be resistant to clofazimine (RR: 2.3; 95%-CI: 1.5–3.6; P = .001). Overall, 102 patients initiated treatment. Patients with BDQ-resistance had higher risk of unfavorable outcomes compared with BDQ-susceptible patients (RR:2.1; 95%-CI: 1.5–2.8; P &lt; .001). Overall, 87% (33/38) of patients with BDQ-resistance experienced unfavorable treatment outcomes: 15 (40%) died, 15 (40%) had treatment failure, and 3 (8%) were lost-to-follow-up.

CONCLUSIONS

The study highlights a concerning rate of BDQ-resistance among previously treated patients, resulting in poor treatment outcomes. To prevent treatment failure, we recommend implementing BDQ-DST, developing affordable and accurate rapid tests for BDQ-resistance, and intensifying research and development efforts for newer TB drugs.

BACKGROUND

There are few data on the treatment of children and adolescents with multidrug-resistant (MDR) or rifampicin-resistant (RR) tuberculosis, especially with more recently available drugs and regimens. We aimed to describe the clinical and treatment characteristics and their associations with treatment outcomes in this susceptible population.

METHODS

We conducted a systematic review and individual participant data meta-analysis. Databases were searched from Oct 1, 2014, to March 30, 2020. To be eligible, studies must have included more than five children or adolescents (0-19 years of age) treated for microbiologically confirmed or clinically diagnosed MDR or RR tuberculosis within a defined treatment cohort, and reported on regimen composition and treatment outcomes. Abstracts were screened independently by two authors to identify potentially eligible records. Full texts were reviewed by two authors independently to identify studies meeting the eligiblity criteria. For studies meeting eligiblity criteria, anonymised individual patient data was requested and individiual level data included for analysis. The main outcome assessed was treatment outcome defined as treatment success (cure or treatment completed) versus unfavourable outcome (treatment failure or death). Multivariable logistic regression models were used to identify associations between clinical and treatment factors and treatment outcomes. This study is registered with Prospero (CRD42020187230).

FINDINGS

1417 studies were identified through database searching. After removing duplicates and screening for eligibility, the search identified 23 369 individual participants from 42 studies, mostly from India and South Africa. Overall, 16 825 (72·0%) were successfully treated (treatment completed or cured), 2848 died (12·2%), 722 (3·1%) had treatment failure, and 2974 (12·7%) were lost to follow-up. In primary analyses, the median age was 16 (IQR 13-18) years. Of the 17 764 (87·1%) participants with reported HIV status, 2448 (13·8%) were living with HIV. 17 707 (89·6%) had microbiologically confirmed tuberculosis. After adjusting for significant factors associated with treatment outcome, the use of two (adjusted odds ratio [OR] 1·41 [95% CI 1·09-1·82]; p=0·008) or three (2·12 [1·61-2·79]; p<0·0001) WHO-classified group A drugs (bedaquiline, moxifloxacin, levofloxacin, and linezolid) compared with the use of no group A drugs at all was positively associated with treatment success.

INTERPRETATION

Younger and clinically diagnosed children are underrepresented among those treated for MDR and RR tuberculosis and should be a focus for case-finding efforts. Overall treatment outcomes in our analysis were better than in adults but lower than the international targets of 90% or more individuals successfully treated. Treatment with more group A drugs was associated with better treatment outcomes in children and adolescents, highlighting the need for more rapid access to these drugs and improved regimens.

BACKGROUND

Expanding contraceptive options could better meet users’ diverse needs and preferences. Annovera^® is a contraceptive vaginal ring that provides a year of pregnancy prevention while remaining under user control and allowing for regular menstrual cycles. This method may also help to reduce burdens on some health care and supply chain systems. However, knowledge gaps exist regarding initial and ongoing acceptability of contraceptive vaginal rings in African settings.

METHODS

We will undertake an open-label, non-randomized, two-arm, parallel clinical acceptability study with an embedded qualitative component, based in clinics providing contraceptive services in Kenya and Zimbabwe. Women aged 18-45 interested in newly initiating or switching contraception will choose from among all available contraceptive options, including Annovera. We aim to enroll 200 participants selecting Annovera and 200 participants selecting either contraceptive injectables or pills. We will compare method uptake, continuation, and satisfaction over one year. Participants will complete questionnaires administered by study staff during two in-person visits (a screening/enrollment visit, and an end of study visit after 52 weeks of method use or at discontinuation) and four phone appointments (at 4, 12, 24, and 36 weeks of use). We will evaluate used rings for discoloration and residual drug levels. The qualitative component involve in-depth interviews with women in the clinical study, their sexual partners, and their service providers, to further examine drivers of and barriers to interest in and use of contraceptive vaginal rings.

DISCUSSION

This study will explore acceptability of contraceptive vaginal rings in ‘real-world’ contraceptive service settings in two African countries. Findings will be based on actual ring use and contextualized via comparison to two other commonly available methods. As vaginal rings are being considered for multiple reproductive health indications, this work can fill key knowledge gaps and empower decision-makers with information needed to inform future investments in reproductive health.

BACKGROUND

Tuberculosis (TB) remains a significant cause of mortality globally, with India accounting for 27% of the estimated number of people with TB. Multidrug-resistant TB (MDR-TB) and isoniazid (INH) resistance pose additional challenges to effective treatment. We aimed to describe treatment outcomes of INH mono-resistant TB patients under programmatic conditions in Mumbai, India.

METHODS

This retrospective cohort study was conducted at Shatabdi Hospital in Mumbai between 2019-2021.We described the clinical and demographic characteristics, treatment outcomes, and risk factors for unfavourable outcomes among patients with INH mono-resistant TB treated with rifampicin, ethambutol, pyrazinamide, and levofloxacin (LfxREZ) for a duration of 6 months.

RESULTS

Among 3105 patients with drug-resistant TB initiated on treatment, 217 (7 %) had INH mono-resistant TB. Of these, 54 % (117/217) were female, with a median age of 26 years (interquartile range: 20-40). The majority (88 %; 191/217) presented with pulmonary TB, and most (87 %; 188/217) had favourable treatment outcomes, including treatment completion (52 %; 112/217) and cure (35 %; 76/217). Unfavourable outcomes, including treatment failure (2.3 %; 5/217), loss to follow-up (9.2 %; 20/217), or death (1.8 %; 4/217), were observed in 13 % (29/217) of patients. A total of ten (5 %) patients experienced at least one non-severe adverse drug reaction. Factors associated with unfavourable outcomes included severe thinness (p = 0.019) and male gender (p = 0.012).

CONCLUSION

Treating INH mono-resistant patients with LfxREZ resulted in satisfactory outcomes and low toxicity. It is important to rule out drug resistance to INH while determining the treatment regimen.

BACKGROUND

Hepatitis C virus (HCV) antibody self-testing (HCVST) may help expand screening access and support HCV elimination efforts. Despite potential benefits, HCVST is not currently implemented in Pakistan. This study aimed to assess the usability and acceptability of HCVST in a high HCV prevalence informal settlement in Karachi, Pakistan.

METHODS

We performed a cross-sectional study in a hepatitis C clinic from April through June 2023. Participants were invited to perform a saliva-based HCVST (OraSure Technologies, USA) while following pictorial instructions. A study member evaluated test performance using a standardized checklist and provided verbal support if a step could not be completed. Perceived usability and acceptability were assessed using a semi-structured questionnaire. The HCVST was considered successful if the participant was able to complete all steps and correctly interpret test results. Overall concordance and positive and negative agreement were estimated in comparison with the HCVST result read by the study member (inter-reader concordance and agreement) and result of a second rapid HCV test (Abbott Diagnostics Korea Inc, South Korea) performed by a trained user (inter-operator concordance and agreement).

RESULTS

The study included 295 participants of which 97 (32%) were illiterate. In total, 280 (95%, 95% CI 92–97%) HCVSTs were successful. Overall, 38 (13%) people performed the HCVST without verbal assistance, 67 (23%) needed verbal assistance in one step, 190 (64%) in two or more. Assistance was most often needed in managing the test buffer and test reading times. The inter-reader concordance was 96% and inter-operator concordance 93%. Inter-reader and inter-operator positive percent agreement were 84 and 70%, respectively. All participants reported they would use HCVST again and would recommend it to friends and family.

CONCLUSION

Saliva-based HCVST was very well accepted in this clinic-based setting. However, many people requested verbal support in several steps, highlighting the need for clear instructions for use and test devices that are simple to use, particularly in low literacy settings. Moderately low positive percent agreement with the results of a rapid test performed by a trained user highlights potential uncertainty in the accuracy of HCVST in the hands of lay users.