Proportion of bacteriologically confirmed pulmonary tuberculosis cases diagnosed at the 14 study treatment units with data collection both in the baseline and intervention phase.

$n = 81,231; 562 patients with missing values for history of anti-TB treatment were excluded.
*Robust standard errors to adjust for clustering at TU level
# Adjusted for clustering at TU level, AND adjusted for age group, sex and past TB history.
Abbreviations: Suspects = number of presumptive pulmonary TB patients tested; All PTB = All diagnosed cases of pulmonary tuberculosis; Bact+PTB = bacteriologically confirmed pulmonary tuberculosis; RR = adjusted relative risk ratio; 95% CI = 95% confidence interval
Proportion of bacteriologically confirmed pulmonary tuberculosis cases diagnosed at the 14 study treatment units with data collection both in the baseline and intervention phase.

样本量$n=81231$；排除562例抗结核治疗史信息缺失的患者。
* 采用稳健标准误校正治疗单元（Treatment Unit）层面的聚类效应
# 校正了治疗单元层面的聚类效应，同时校正了年龄组、性别与既往结核病史。
缩写对照表：
Suspects（疑似筛查肺结核患者数）：接受检测的疑似肺结核患者人数；
All PTB（全部确诊肺结核病例）：所有确诊的肺结核病例；
Bact+PTB（细菌学确诊肺结核）：经细菌学证实的肺结核病例；
RR（校正相对风险比）：校正后的相对风险比；
95% CI（95%置信区间）：95%置信区间。
本研究14个治疗单元均在基线阶段与干预阶段完成数据收集，其中经细菌学证实的肺结核确诊病例占比。