Study population
During the study period, 73,116 drug-susceptible pulmonary TB and 1,673 drug-resistant pulmonary TB patients were extracted. Among drug-susceptible pulmonary TB patients, 39,540 and 33,576 were newly diagnosed and treated before and after introducing the total cost exemption of medical service copayment. In the case of drug-resistant TB, newly diagnosed patients during pre- and post-cost exemption policy were 950 and 723 for each (Figure 1). As demographic data described in Table 1, male and older patients were frequent in both groups. In the case of drug-susceptible pulmonary TB, the long-term treatment discontinuation rate during the post-policy period was lower than before the policy change (30.1% vs. 25.3%, P < 0.0001). However, the long-term treatment discontinuation rate of anti-TB medications in drug-resistant pulmonary TB showed no difference between the pre-and post-cost exemption period (68.3% vs. 69.8%, P = 0.502).
Long-term treatment interruption by treatment phase in the pulmonary TB groups
Among the drug-susceptible TB patients, compared to the periods before and after the cost exemption policy, the number of long-term discontinuation cases per 100,000 patients during the intensive phase showed no significant difference (8,841 vs. 8,422, P = 0.074). Furthermore, there was no significant change in the slope of the long-term discontinuation rate (slope change: 0.015, P = 0.747). On the other hand, in the continuation phase, the number of long-term discontinuation cases per 100,000 patients and trend slope significantly decreased after the cost exemption policy (21,940 vs. 17,319, P < 0.0001; slope change: −0.097, P = 0.011) (Figure 2). Regarding the intensive phase of drug-resistant TB patients, although the number of long-term discontinuation cases per 100,000 patients decreased (34,182 vs. 31,784, P = 0.501), the slope showed an increasing tendency after policy change during the intensive phase (slope change: 0.733, P = 0.001). However, there was no significant change during the continuation phase (44,348 vs. 44,962, P = 0.782; slope change: −0.049, P = 0.803) (Figure 3).
Risk factors for long-term TB treatment interruption
Old age (aOR, 1.15; 95% CI, 1.11–1.19), multiple comorbidities (CCI ≥ 3) (aOR, 1.17; 95% CI, 1.13–1.21), and drug resistance for first-line TB drugs (aOR, 6.04; 95% CI, 5.43–6.71) were revealed as risk factors for predicting the long-term discontinuation of anti-TB medications among the total study population (Table 2). They were also identified as significant risk factors in the analysis among drug-susceptible and drug-resistant TB patients (Supplementary Tables S2 and S3).
Mortalities according to the history of long-term TB treatment interruption
The mean follow-up periods were 47.9 and 63.8 months for drug-susceptible and drug-resistant TB for each. In the case of drug-susceptible pulmonary TB, patients with long-term treatment interruption showed a higher overall mortality rate than those without (32.2% vs. 15.0%, P < 0.0001) (Figure 4). In the case of drug-resistant TB, mortality was also higher in patients with a history of long-term treatment discontinuation (12.1% vs. 8.3%, P = 0.008).
Risk factors associated with 1-year and overall mortality
Old age (aHR 5.96; 95% CI, 5.23–6.78 [1-year mortality], aHR, 5.83; 95% CI, 5.49–6.20 [overall mortality]), male sex (aHR 1.47; 95% CI, 1.35–1.60 [1-year mortality], aHR, 1.59; 95% CI, 1.52–1.66 [overall mortality]), high CCI (aHR 2.40; 95% CI, 2.18–2.65 [1-year mortality], aHR, 1.97; 95% CI, 1.88–2.06 [overall mortality]), and history of long-term discontinuation of anti-TB drugs (aHR 2.01; 95% CI, 1.86–2.18 [1-year mortality], aHR, 1.77; 95% CI, 1.70–1.84 [overall mortality]) were revealed as risk factors for both 1-year and overall mortality. However, drug-resistant for first-line TB drugs was a negative risk factor for mortality (aHR, 0.02; 95% CI, 0.01–0.15 [1-year mortality], aHR, 0.57; 95% CI, 0.48–0.68 [overall mortality]) (Table 3).