Cut-off Points of Treatment Delay to Predict Poor Outcomes Among New Pulmonary Tuberculosis Cases in Dalian, China: A Cohort Study

Figures & data

Figure 1 The flow chart of the inclusion of new pulmonary TB cases.

Table 1 Characteristics of New Pulmonary TB Cases Base on the Time of Treatment Initiation

Characteristics	N (%)	Time of Treatment Initiation, M (IQR)	z	P value
Total	18,100(100)	30(14–59)
Age			529.779	<0.001
Young	9063(50.1)	25(12–44)
Middle	6170(34.1)	32(17–66)
Elderly	2867(15.9)	34(20–70)
Gender			11.426	<0.001
Female	5747(31.8)	31(15–62)
Male	12,353(68.2)	30(14–57)
Residence			116.402	<0.001
Rural	10,293(56.9)	31(15–64)
Urban	7807(43.1)	29(14–46)
Employment status			38.357	<0.001
Employed	6989(38.6)	31(15–64)
Unemployed	11,111(61.4)	30(14–53)
Drug resistance			16.063	<0.001
Yes	563(3.1)	32(18–68)
No	17,537(96.9)	30(14–59)
Diabetes mellitus			17.161	<0.001
Yes	897(5)	31(16–66)
No	17,203(95)	30(14–59)
Severe case			0.346	0.557
Yes	1709(9.4)	31(14–61)
No	16,391(90.6)	30(14–59)
Lung cavitation			130.066	<0.001
Yes	3800(21)	34(16–71)
No	14,305(79)	29(14–53)
Treatment options			25.823	<0.001
Standard therapy	17,247(95.3)	30(14–59)
Experiential therapy	853(4.7)	32(19–61)

Abbreviations: M, median; IQR, interquartile range.

Table 2 The Association Between Incident Poor Outcomes and Risk Factors in Cox Regression Models

Independent Variables	Poor Treatment Outcomes			Death
	Incidence Rate, Per 1000	AHR	95% CI	Incidence Rate, Per 1000	AHR	95% CI
Time of treatment initiation
Q1 (0–14 days)	28.3	Ref.	Ref.	9.8	Ref.	Ref.
Q2 (15–30 days)	34.1	1.07	0.85–1.35	12.5	0.99	0.67–1.47
Q3 (31–59 days)	39.9	1.15	0.91–1.44	13.3	0.95	0.64–1.41
Q4 (60 days and above)	54.3	1.43	1.15–1.78	22.6	1.49	1.05–2.13
Gender
Female	21.4	Ref.	Ref.	7.7	Ref.	Ref.
Male	47.3	1.78	1.46–2.17	17.7	1.94	1.40–2.69
Age
Young	19.6	Ref.	Ref.	3.0	Ref.	Ref.
Middle	47.5	1.85	1.53–2.24	12.6	3.59	2.30–5.59
Elderly	82.3	3.57	2.92–4.36	55.1	15.64	10.32–23.71
Residence
Rural	37	Ref.	Ref.	15.4	Ref.	Ref.
Urban	41.8	1.37	1.15–1.63	13.4	1.07	0.80–1.43
Employment status
Unemployed	35.5	Ref.	Ref.	13.3	Ref.	Ref.
Employed	41.3	1.21	1.02–1.45	15.3	1.25	0.93–1.68
Drug resistance
No	32	Ref.	Ref.	13.6	Ref.	Ref.
Yes	259.3	6.98	5.76–8.44	42.6	2.72	1.77–4.17
Treatment options
Standard therapy	37.2	Ref.	Ref.	13.1	Ref.	Ref.
Experiential therapy	76.2	1.34	1.03–1.74	43.4	1.93	1.35–2.75
Severe case
No	38.1	Ref.	Ref.	14.9	Ref.	Ref.
Yes	48	0.93	0.72–1.21	11.1	0.66	0.39–1.11
Diabetes mellitus
No	36.6	Ref.	Ref.	14.0	Ref.	Ref.
Yes	87	1.28	0.99–1.65	24.5	0.98	0.62–1.56
Lung cavitation
No	30	Ref.	Ref.	13.4	Ref.	Ref.
Yes	73.2	1.91	1.60–2.28	18.7	1.36	1.00–1.85

Abbreviations: AHR, adjusted hazard ratio; CI, confidence interval; Q, quartile; Ref, reference;

Figure 2 Adjusted hazard ratios (AHRs) of treatment outcomes event risk according to the time of treatment initiation.

Notes: In RCS fitted for Cox regression models, we adjusted for age, gender, resident area, occupation, drug resistance, severe case, lung cavitation, treatment options and lung cavitation.

Figure 3 The moderating effects of age, gender and diabetes mellitus.

Notes: To assess the moderating effects of age, gender, and diabetes mellitus, we adjusted for age, gender, resident area, occupation, drug resistance, severe case, lung cavitation, treatment options and lung cavitation in Cox regression models.