Permanent employment or public assistance may increase tuberculosis survival among working-age patients in Japan.

Objectives: To measure the effect of employment and health insurance status on the survival of working age tuberculosis (TB) patients in Japan.
Methods: Retrospective cohort analysis of new smear-positive pulmonary TB patients aged 15-59 years registered in the Japanese national TB surveillance system between 2007 and 2010. We performed univariate and multivariate Cox proportional hazard model analysis. The survival curves for employment and health insurance status were calculated using Kaplan-Meier analysis.
Results: Of 9097 patients studied, 267 (2.9%) died of TB within 12 months. After adjustment with a multivariate model, employment and health insurance status were independently associated with increased risk of TB death: unemployment (HR 2.80, 95%CI 2.11-3.72), absence of insurance (HR 1.48, 95%CI 1.02-2.15). The analysis of survival curves indicated that those with public assistance had almost the same survival rate as insured patients in the unemployed group. Permanent workers (employed >30 days) had the highest survival rates, followed by casual workers (employed <30 days) and the unemployed in the insured group.
Conclusion: Patients with permanent jobs had better survival rates than unemployed patients and casual workers. Despite being unemployed, receiving public assistance could improve survival. Health measures are required for the unemployed and casual workers.