Background: South Africa has a high burden of MDR-TB, and to provide accessible treatment the government has introduced different models of care. We report the most cost-effective model after comparing cost per patient successfully treated across 5 models of care: centralized hospital, district hospitals (2), and community-based care through clinics or mobile injection teams.
Methods: In an observational study five cohorts were followed prospectively. The cost analysis adopted a provider perspective and economic cost per patient successfully treated was calculated based on country protocols and length of treatment per patient per model of care. Logistic regression was used to calculate propensity score weights, to compare pairs of treatment groups, whilst adjusting for baseline imbalances between groups. Propensity score weighted costs and treatment success rates were used in the ICER analysis. Sensitivity analysis focused on varying treatment success and length of hospitalization within each model.
Results: In 1,038 MDR-TB patients 75% were HIV-infected and 56% were successfully treated. The cost per successfully treated patient was 3 to 4.5 times lower in the community-based models with no hospitalization. Overall, the Mobile model was the most cost-effective.
Conclusion: Reducing the length of hospitalization and following community-based models of care improves the affordability of MDR-TB treatment without compromising its effectiveness.
Download full-text PDF |
Source |
---|---|
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC5906004 | PMC |
http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0196003 | PLOS |
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!