Skip to main content
. 2020 Apr 29;15(4):e0232337. doi: 10.1371/journal.pone.0232337

Table 2. Factors associated with not using EMM^ within first month of diagnosis among eligible notified people with TB in the 30 selected counties of China between July and December, 2018^^.

Factors Total Not using EMM^ PR (95% CI) aPR@ (95% CI)
n (%)
Total 1810 763 (42.2) - - - -
Age in years
<15 8 5 (62.5) 1.6 (0.90–2.69) 1.9 (1.01–3.39)*
15–44 629 252 (40.1) ref ref
45–64 667 274 (41.1) 1.0 (0.90–1.17) 1.1 (0.91–1.19)
> = 65 506 232 (45.9) 1.1 (1.00–1.31)* 1.3 (1.13–1.54)*
Sex
Male 1221 505 (41.4) ref ref
Female 589 258 (43.8) 1.1 (0.95–1.19) 1.1 (0.95–1.19)
Occupation
Farmers and herdsmen 816 309 (37.9) ref ref
Semi-skilled employee 168 71 (42.3) 1.1 (0.92–1.36) 1.2 (1.01–1.52)*
Salary employee 194 93 (47.9) 1.3 (1.07–1.50)* 1.1 (0.97–1.37)
Unemployed 450 217 (48.2) 1.3 (1.12–1.45)* 1.2 (1.02–1.35)*
Studying 100 38 (38.0) 1.0 (0.77–1.31) 0.9 (0.70–1.27)
Others 82 35 (42.7) 1.1 (0.86–1.47) 1.2 (0.91–1.57)
Migrant**
No 1733 736 (42.5) 1.2 (0.89–1.65) 1.2 (0.86–1.61)
Yes 77 27 (35.1) ref ref
Classification
Bacteriologically confirmed PTB 980 396 (40.4) ref ref
Clinically diagnosed PTB 672 278 (41.4) 1.0 (0.91–1.15) 1.1 (0.97–1.23)
Pleurisy 158 89 (56.3) 1.4 (1.19–1.63)* 1.6 (1.34–1.83)*
Category
New 1716 708 (41.3) ref ref
Previously treated 94 55 (58.5) 1.4 (1.19–1.70)* 1.6 (1.32–1.88)*
Treatment initiation interval (days)#
Zero 1731 729 (42.1) ref ref
≥ One 79 34 (43.0) 1.0 (0.79–1.32) 1.2 (0.89–1.52)

Row percentages

TB–tuberculosis; PTB–pulmonary TB;EMM- Electronic Medication Monitor; PR–prevalence ratio; aPR–adjusted prevalence ratio; CI–confidence interval

^ Includes those who used EMM later during treatment or never used EMM during treatment

^^ Of 2294 notified people with TB, 2227 has EMM eligibility data recorded. Of 2227, 1810 were eligible to use EMM at notification. Of 1810, 1047 used EMM within first month of diagnosis.

@ Adjusted prevalence ratios (95% CI) calculated using log binomial regression. We included all the variables including county in the multivariable analysis (county not shown in the table)

* Statistically significant

** Migrant defined as patients staying in the same prefecture for less than six months

# Time interval from diagnosis to treatment, the median interval was zero days and used as cut-off to categorize