OBJECTIVE: To examine the incident rate of tuberculosis (TB) and its associates among adults with type 2 diabetes in Shanghai, China. METHODS: We conducted a retrospective cohort study among 170 399 patients with type 2 diabetes aged ≥18 years who were registered in Shanghai community-based diabetes management system between 2004 and 2009. Their TB status was tracked until 31 December 2014. Cox regression was performed to identify the risk factors for TB. RESULTS: We documented 785 new TB cases during 654 977 person-years of follow-up. The incident rate of TB was 224.20 (206.69, 243.16) per 100 000 person-years among men and 51.34 (44.75, 58.92) per 100 000 person-years among women. A 1-unit increase of BMI was associated with a risk reduction in 16% (P < 0.01) for men and a 14% (P < 0.01) reduction for women. TB cases were more likely to be insulin-dependent [men: hazard ratio = 2.13 (1.29, 3.53); women: 3.28 (1.28, 8.39)] and had a poor glucose level initially [men: 1.21 (1.15, 1.27); women: 1.27 (1.18, 1.37)]. The risk factor for TB specific to men was a young age at diagnosis of diabetes, and the protective factor specific to women was actively engaging in physical activity. CONCLUSIONS: TB incident rate among patients with type 2 diabetes was substantially higher among men than among women. The risk of TB was reversely associated with initial BMI. The severity of poor glucose control among patients with diabetes was also linearly associated with the risk of TB.
OBJECTIVE: To examine the incident rate of tuberculosis (TB) and its associates among adults with type 2 diabetes in Shanghai, China. METHODS: We conducted a retrospective cohort study among 170 399 patients with type 2 diabetes aged ≥18 years who were registered in Shanghai community-based diabetes management system between 2004 and 2009. Their TB status was tracked until 31 December 2014. Cox regression was performed to identify the risk factors for TB. RESULTS: We documented 785 new TB cases during 654 977 person-years of follow-up. The incident rate of TB was 224.20 (206.69, 243.16) per 100 000 person-years among men and 51.34 (44.75, 58.92) per 100 000 person-years among women. A 1-unit increase of BMI was associated with a risk reduction in 16% (P < 0.01) for men and a 14% (P < 0.01) reduction for women. TB cases were more likely to be insulin-dependent [men: hazard ratio = 2.13 (1.29, 3.53); women: 3.28 (1.28, 8.39)] and had a poor glucose level initially [men: 1.21 (1.15, 1.27); women: 1.27 (1.18, 1.37)]. The risk factor for TB specific to men was a young age at diagnosis of diabetes, and the protective factor specific to women was actively engaging in physical activity. CONCLUSIONS: TB incident rate among patients with type 2 diabetes was substantially higher among men than among women. The risk of TB was reversely associated with initial BMI. The severity of poor glucose control among patients with diabetes was also linearly associated with the risk of TB.