Publication date: Jul 01, 2026
The long-term impact of COVID-19 on tuberculosis (TB) is concerning. This study assessed whether COVID-19 was associated with an increased risk of active TB treatment risk in Japan using a nationwide propensity score-matched cohort design. A retrospective cohort study used Japan’s National Insurance Database from January 2020 to December 2022. COVID-19 patients were matched 1:1 with controls using propensity scores based on age, sex, immunosuppressants use, Charlson comorbidity index, related medical conditions, healthcare utilization and prior TB history. The primary outcome was initiation of isoniazid-based therapy plus rifampin with a TB-related ICD-10 diagnosis code. Cox proportional hazards models estimated hazard ratios (HRs) for COVID-19’s effect on TB treatment initiation. After matching, 3,097,422 individuals were included per group. During 8-month median follow-up, 886 participants initiated TB treatment. The HR among COVID-19 infected individuals was 4. 14 (95% CI: 3. 51-4. 89) versus matched controls. Subgroup analyses showed interaction between prior TB history and COVID-19, with HR of 14. 7 (5. 35-40. 2) in those with prior TB history, versus those without (HR 3. 84, 95% CI 3. 24-4. 54). Risk remained elevated in participants without hospitalization (HR 3. 96, 95% CI 3. 35-4. 68). This study provides robust population-based evidence demonstrating a moderate association between prior COVID-19 and subsequent initiation of active tuberculosis treatment, particularly in those with prior TB history.

Open Access PDF
| Concepts | Keywords |
|---|---|
| December | COVID‐19 |
| Hrs | National Insurance Database |
| Immunosuppressants | propensity score matching |
| Japanese | tuberculosis |
| Tuberculosis |
Semantics
| Type | Source | Name |
|---|---|---|
| disease | MESH | COVID-19 |
| disease | MESH | Tuberculosis |
| pathway | KEGG | Tuberculosis |
| drug | DRUGBANK | Isoniazid |
| drug | DRUGBANK | Rifampicin |
| disease | MESH | ICD |
| disease | MESH | HRs |
| disease | MESH | included |
| disease | MESH | Long Covid |