Multidrug-resistant tuberculosis (MDR-TB) is an urgent public health challenge in Namibia, with profound socioeconomic consequences. The high burden of both tuberculosis and HIV complicates treatment and underscores the need for optimised drug therapies. Precision medicine, which leverages patient-specific genetic and molecular information, offers promise for improving MDR-TB outcomes. However, its effective application relies on population-specific data, particularly understanding how individuals metabolise tuberculosis drugs and how genetic diversity drives variability in treatment response. Currently, no pharmacokinetic (PK) or pharmacogenetic (PG) data on TB treatment exist for Namibian populations. This gap is particularly concerning, given the country’s genetic diversity, environmental factors and comorbidities that may uniquely influence drug metabolism. This study aims to generate PK and PG data to inform dose optimisation and support personalised treatment strategies for MDR-TB in Namibia. The findings will contribute to improved patient care and inform health system strengthening based on locally relevant evidence.

This cross-sectional study will consist of 100 Namibian participants with matched human DNA and PK data of MDR-TB cases receiving isoniazid, clofazimine, bedaquiline and the fluoroquinolones (levofloxacin or moxifloxacin). PK sampling will be divided as follows: 30 individuals will undergo intensive PK sampling, while the remaining (n=70) will undergo sparse PK sampling. DNA will be extracted at Stellenbosch University (SU), and samples will be genotyped using the H3Africa microarray. Sequences will be aligned to the human reference genome, hg38 (GRCh38p13), using the freely available Burrows-Wheeler Aligner. A subset of the samples (n=20–30) will undergo whole genome sequencing (WGS) to verify imputation results and identify novel genetic variants potentially affecting PK in this population.

Quality control and variant call format file generation will be performed using the Genome Analysis Toolkit best practices (V.3.5). Intensive and sparse PK data will be pooled for the development of a population PK (popPK) model using a non-linear mixed-effects modelling approach. The popPK model will characterise the relationship between TB drug dose and exposure, including quantifying covariates, including genetic variation, explaining PK variability, providing a foundation for dose optimisation and personalised treatment strategies.

Ethics approval was obtained from the University of Namibia Human Research Ethics Committee for Health (Ref. SOM18/2024), the Ministry of Health and Social Services (Ref. 22/4/2/3), the SU Health Research Ethics Committee (Ref. N21/11/136) and the University of Cape Town Human Research Ethics Committee (Ref. 500/2022).

Suicide rates have increased over the last couple of decades globally, particularly in the United States and among populations with lower economic status who present at safety-net healthcare systems. Recently, predictive models for suicide risk have shown promise; however, a model for this specific population does not exist.

To develop a predictive risk model of suicide and intentional self-harm (ISH) for patients presenting at the psychiatric emergency department (ED) of JPS Health Network, a safety net medical and mental healthcare system in Texas.

The study used structured and unstructured electronic medical record (EMR) data (2015–2019) and local medical examiner data (2015–2020) to create predictors and outcome variables. All psychiatric ED notes during calendar years 2018 and 2019 were reviewed using natural language processing to identify presentations for any level of self-harm and subsequent manual review of identified visits to accurately classify ED presentations for treatment of an act of intentional self-harm meeting study criteria. Data from 15 987 patients were used to develop and validate a machine learning-based predictive model that leverages rolling window methodology to predict risk repeatedly across a patient’s trajectory. Feature engineering played a prominent role in defining new predictors.

The best model (XGBoost) achieved the area under the receiver operating characteristic curve of 0.81 for 30-day predictions and demonstrated concentration of ISH and suicide attempt events in high-risk quantiles of risk (65% had events in top 0.1% quantile). The predicted risk can be translated into a propensity of events (80% at the highest predicted risk) to facilitate clinical interpretation.

Machine learning-based models can be used with standard EMRs to identify patients presenting at the psychiatric ED with a high risk of ISH and suicide attempts within the next 30 days.

Integration of a predictive model can significantly aid clinical decision-making in safety-net psychiatric EDs.