Peng Lu1, Yu Zhang2, Qiao Liu1, Xianyan Ding1, Wen Kong3, Limei Zhu1, Wei Lu4. 1. Department of Chronic Communicable Disease, Center for Disease Control and Prevention of Jiangsu Province, Nanjing, Jiangsu Province, People's Republic of China. 2. Southeast University, School of Public Health, Nanjing, Jiangsu Province, People's Republic of China. 3. Center for Disease Control and Prevention of Qingpu District, Shanghai City, People's Republic of China. 4. Department of Chronic Communicable Disease, Center for Disease Control and Prevention of Jiangsu Province, Nanjing, Jiangsu Province, People's Republic of China. Electronic address: weiluxx@sina.cn.
Abstract
OBJECTIVE: To explore whether all individuals with diabetes are at an increased risk of tuberculosis or whether there are high-risk groups of diabetes patients. METHODS: A population-based census was conducted in China, including 27 807 individuals. Risk factors for incident tuberculosis were investigated and participants included in the study were linked to the Tuberculosis Management Information System. Cox proportional hazards models adjusting for several variables were constructed in two steps. First, the incident rate of tuberculosis in individuals with/without diabetes was compared without any stratification by high body mass index (BMI). Second, two groups stratified by BMI were compared. RESULTS: Over 7 years of follow-up, 108 individuals progressed to tuberculosis. In the multivariable Cox regression analysis, participants with diabetes had a 2.26 higher hazard (P = 0.017) of tuberculosis when compared to participants without diabetes. However, participants with a BMI >24 kg/m2 and diagnosed with diabetes had a similar risk as overweight participants without diabetes (P = 0.953), while diabetes with a BMI <24 kg/m2 (adjusted hazard ratio 2.68; P = 0.006) was a risk factor for developing active tuberculosis (compared with non-diabetes with a BMI <24 kg/m2). Among individuals with a low BMI, participants with a Bacille Calmette-Guérin scar showed a decreased tuberculosis risk (adjusted hazard ratio 0.48; P = 0.028). CONCLUSIONS: Individuals with diabetes were found to have a higher risk of developing tuberculosis, but only when they were not overweight, suggesting targeted screening should be considered.
OBJECTIVE: To explore whether all individuals with diabetes are at an increased risk of tuberculosis or whether there are high-risk groups of diabetespatients. METHODS: A population-based census was conducted in China, including 27 807 individuals. Risk factors for incident tuberculosis were investigated and participants included in the study were linked to the Tuberculosis Management Information System. Cox proportional hazards models adjusting for several variables were constructed in two steps. First, the incident rate of tuberculosis in individuals with/without diabetes was compared without any stratification by high body mass index (BMI). Second, two groups stratified by BMI were compared. RESULTS: Over 7 years of follow-up, 108 individuals progressed to tuberculosis. In the multivariable Cox regression analysis, participants with diabetes had a 2.26 higher hazard (P = 0.017) of tuberculosis when compared to participants without diabetes. However, participants with a BMI >24 kg/m2 and diagnosed with diabetes had a similar risk as overweight participants without diabetes (P = 0.953), while diabetes with a BMI <24 kg/m2 (adjusted hazard ratio 2.68; P = 0.006) was a risk factor for developing active tuberculosis (compared with non-diabetes with a BMI <24 kg/m2). Among individuals with a low BMI, participants with a Bacille Calmette-Guérin scar showed a decreased tuberculosis risk (adjusted hazard ratio 0.48; P = 0.028). CONCLUSIONS: Individuals with diabetes were found to have a higher risk of developing tuberculosis, but only when they were not overweight, suggesting targeted screening should be considered.