Publication date: Aug 01, 2026
Observational studies investigating causal (aetiological) questions often address confounding bias using data-driven predictive models where variables are selected regardless of their causal role, challenging interpretation. We compared the estimated effect of HIV co-infection on end-of-treatment outcomes among people with multidrug/rifampicin-resistant TB using a causal framework model and a data-driven predictive model. The causal framework guided confounder adjustment. Results were compared to those from models that applied alternative variable selection strategies (based on P values and change-in-estimate), which do not draw on domain knowledge to identify relevant variables. The model informed by a causal diagram indicated that people living with HIV had a 31% lower probability of achieving a successful outcome compared to those without HIV (adjusted relative risk [aRR] 0. 69, 95% confidence interval [CI]: 0. 41-0. 98). In contrast, data-driven models produced attenuated associations (aRR 0. 78, 95% CI: 0. 50-1. 06 and aRR 0. 80, 95% CI: 0. 5-1. 09 for the P value and change-in-estimate strategies, respectively), reflecting omission of key confounders, and inappropriate inclusion of mediators and colliders. When the research question is aetiological, using a causal approach to guide variable selection ensures proper adjustment, improves interpretability, and establishes a stronger foundation for future observational research.
Open Access PDF
| Concepts | Keywords |
|---|---|
| causal inference | |
| causal structure | |
| epidemiologic methods | |
| HIV | |
| MDR/RR-TB | |
| observational studies | |
| tuberculosis |
Semantics
| Type | Source | Name |
|---|---|---|
| disease | MESH | co-infection |
| drug | DRUGBANK | Rifampicin |
| disease | MESH | tuberculosis |
| pathway | KEGG | Tuberculosis |