Abstract:

Objective To evaluate the implementation effectiveness of community-based pulmonary tuberculosis (PTB) active case-finding(ACF) interventions among key populations. Methods A multi-center prospective cohort study was conducted in 2013-2015 among five key populations of tuberculosis (1.Residents aged 65 and above; 2.Residents with diabetes mellitus; 3.Residents with HIV/AIDS; 4.Residents with a history of TB; 5.Close contacts of active PTB patients) in selected 27 townships/communities in 10 provinces located in eastern, central and western regions of China. TB health promotion and education, TB symptom screening, and chest X-ray examination were carried out annually. The sensitivity and specificity of different screening approaches, the effectiveness of ACF among different key populations, and factors that influenced ACF contributions were analyzed. Results During 2013-2015, the intervention coverage of health education and TB symptom screening were 97.8% (42684/43654), 91.2% (41732/45768), and 88.1% (42441/48178) respectively; the intervention coverage of chest X-ray examination were 86.0% (37538/43654), 81.0% (37070/45768), and 75.7% (36483/48178), respectively. The TB key message awareness rate increased from 34.2% (73066/213420) in 2013 to 67.2% (142629/212205) in 2015, showing an upward trend year by year (Z_trend=215.568, P<0.01). The sensitivities of symptom screening only were low which could find 11.6% (69/596)of TB cases by using national TB control program defining symptoms and 11.7% (70/596) by using symptoms defined by this study, while specificities were quite high (99.1% (125123/126261) and 99.0% (125008/126261) respectively). The sensitivities of only using chest X-ray examination and study-defined symptom screening plus chest X-ray were both 90.7% (594/655),the specificities were 94.2% (104004/110436) and 93.3% (103 062/110436), respectively. Among all diagnosed PTB cases, 84.9% (556/655) were detected through ACF. The contribution rates of ACF in 2013, 2014, and 2015 were 95.7% (244/255), 81.0% (200/247), and 73.2% (112/153), showing a downward trend year by year (Z_trend=-6.403, P<0.01). The results of multi-variable analysis showed that compared with age group 15-24, the contribution rates of ACF in age group 55-64 (OR=7.18; 95%CI: 1.59-32.39) and age group ≥65 (OR=13.52; 95%CI: 3.31-55.16) were higher; compared with eastern regions, the contribution rate of ACF in western regions (OR=2.44; 95%CI: 1.38-4.29) was higher. Conclusion Carrying out an annual community-based ACF activity among key populations could significantly increase PTB case detection. The contribution rates of ACF were significantly higher in older people and western regions. However, they showed decreasing trends year by year in the three years’ implementation phase. Therefore, whether to take ACF activities every year for the same group of people should be considered under this group’s specific ACF contribution rate.

Key words: Tuberculosis,pulmonary, Population surveillance, Community-based participatory research, Intervention studies