Development of a NAT2 Testing-integrated Pharmaceutical Care System and Changes in Anti-tuberculosis Drug-induced Hepatitis in a Community Hospital

Abstract

Objective: To develop a N-acetyltransferase 2 (NAT2) testing integrated pharmaceutical care system for newly diagnosed pulmonary tuberculosis patients in a community hospital, and to evaluate changes in anti-tuberculosis drug-induced hepatitis after system implementation. Methods: This quasi-experiment was a system-development study. The subjects comprised new pulmonary tuberculosis patients aged 18 years or older who initiated treatment between October 1, 2023 and December 31, 2025 and received the standard 2HRZE/4HR regimen. Patients with HIV infection, pre-existing liver disease or a known history of hepatitis, a baseline alanine aminotransferase (ALT) level greater than twice the upper limit of normal value before treatment initiation, and patients with drug-resistant tuberculosis were excluded. The development of a pharmaceutical care system with NAT2 genotyping was based on the concept of pharmaceutical care, Donabedian’s structure–process–outcome framework, and two cycles of the Plan–Do–Check–Act (PDCA) framework. The pre-implementation period was from October 1, 2023 to September 30, 2024. The transitional period was from October to December 2024, and the post-implementation period was from January 1, 2025 to December 31, 2025. Data from the transitional period were excluded from the primary analysis. The primary outcome was the monthly incidence of antituberculosis drug-induced hepatitis. The study used a monthly interrupted time-series (ITS) analysis. The main model was segmented Poisson regression, with the monthly number of new pulmonary tuberculosis patients as an offset variable. Results: After full implementation, all patients received NAT2 testing, and the results were used to support individualized isoniazid dose adjustment. All rapid and slow acetylators received dose adjustment, while intermediate acetylators received the standard dose according to the Thai national tuberculosis guideline. Medication adherence assessed by pill count increased from 96.7 ± 1.4% to 97.8 ± 0.9% (P<0.001). Anti-tuberculosis drug-induced hepatitis decreased from 60 of 152 patients (39.5%) to 13 of 165 patients (7.9%). ITS analysis showed a significant immediate reduction in the incidence level after full implementation (incidence rate ratio or IRR 0.503, 95% CI 0.345–0.732, P<0.001), as well as a significant reduction in the post-implementation monthly trend compared with that of the pre-implementation period (IRR 0.837, 95% CI 0.747–0.938, P=0.002). Conclusion: After implementing a pharmaceutical care system with NAT2 testing, the incidence of anti-tuberculosis drug-induced hepatitis showed a decreasing trend, accompanied by improvements in NAT2 testing, dose adjustment, safety monitoring, and medication adherence. These findings should be interpreted as changes associated with an integrated service system rather than the independent effect of NAT2 testing or isoniazid dose adjustment alone. The study cannot conclude its microbiological effectiveness or long-term treatment outcomes.