3.1 The level of non-adherence to anti-TB treatment
A total of 481 newly diagnosed TB patients were included in this study. The mean score of the medication adherence scale was 6.53±1.85. The 220 (45.7%) patients included in our study were classified as having good adherence, and 132 (27.4%) and 129 (26.8%) had moderate and low adherence, respectively.
3.2 Socio-demographic characteristics
In our study, patients ranged in age from 18 to 88 years, with an average age of 44.10±17.85 years. More than half of TB patients (62.0%) entering the study were male, whereas only 183(38%) were female. Nearly three-quarters of TB patients (71.5%) were married, compared with 18(3.7%) who were divorced or widowed. TB patients with an education level of secondary school or below (41.0%) accounted for the most percentage. Most TB patients have low monthly incomes, with only 84(17.5%) of TB patients earning more than 5,000 yuan a month. Nearly one third of TB patients (32.0%) arrived at the facility more than an hour and 136(28.3%) less than 30 minutes. The results of the univariate analysis showed that education level and the time required to reach the medical institution were significantly correlated with the medication adherence of TB patients(P<0.05), and the adherence of TB patients of different genders, ages, marital status, and monthly income in our study could not be considered to be different(P>0.05). (Table 1).
Table 1. Medication adherence level by sociodemographic factors
Variables
|
Total n (%)
|
Adherence Level n (%)
|
P
|
Low
|
Medium
|
High
|
Sex
|
|
|
|
|
|
Male
|
298(62.0)
|
78(26.2)
|
84(28.2)
|
136(45.6)
|
0.869
|
Female
|
183(38.0)
|
51(27.9)
|
48(26.2)
|
84(45.9)
|
|
Age (years)
|
|
|
|
|
|
<30
|
128(26.6)
|
36(28.1)
|
37(28.9)
|
55(43.0)
|
0.942
|
30-60
|
244(50.7)
|
65(26.6)
|
67(27.5)
|
112(45.9)
|
|
>60
|
109(22.7)
|
28(25.7)
|
28(25.7)
|
53(48.6)
|
|
Marital status
|
|
|
|
|
|
Unmarried
|
119(24.7)
|
35(29.4)
|
42(35.3)
|
42(35.3)
|
0.096
|
Married
|
344(71.5)
|
90(26.2)
|
85(24.7)
|
169(49.1)
|
|
Divorced or widowed
|
18(3.7)
|
4(22.2)
|
5(27.8)
|
9(50.0)
|
|
Education
|
|
|
|
|
|
Junior high school or below
|
197(41.0)
|
52(26.4)
|
54(27.4)
|
91(46.2)
|
0.025
|
High school or technical secondary school
|
117(24.3)
|
29(24.8)
|
22(18.8)
|
66(56.4)
|
|
College degree or above
|
167(34.7)
|
48(28.7)
|
56(33.5)
|
63(37.7)
|
|
Monthly income (Yuan)
|
|
|
|
|
|
<1000
|
133(27.7)
|
31(23.3)
|
42(31.6)
|
60(45.1)
|
0.210
|
1000-3000
|
115(23.9)
|
34(29.6)
|
33(28.7)
|
48(41.7)
|
|
3001-5000
|
149(31.0)
|
40(26.8)
|
30(20.1)
|
79(53.0)
|
|
>5000
|
84(17.5)
|
24(28.6)
|
27(32.1)
|
33(39.3)
|
|
Time to arrive at the medical facility (min)
|
|
|
|
|
<31
|
136(28.3)
|
36(26.5)
|
26(19.1)
|
74(54.4)
|
0.033
|
31-60
|
191(39.7)
|
45(23.6)
|
62(32.5)
|
84(44.0)
|
|
>60
|
154(32.0)
|
48(31.2)
|
44(28.6)
|
62(40.3)
|
|
3.3 Family support
Table 2 shows that family supervising medication, family spiritual encouragement, and family member relationships were significantly different in different groups by univariate analysis(P<0.05). A high proportion of low adherence (48.6%) was found in TB patients whose family members sometimes supervised their medication, while a high proportion of high adherence (47.7% and 54.5%, respectively) was found in the two groups of patients whose family members frequently supervised their medication and those who did not. Patients with frequent spiritual encouragement from family members had a higher proportion of high adherence (47.0%). The relationship between family members of most patients (95.6%) was good, and there was a little proportion of poor adherence (26.3%). A significantly higher proportion of patients (91.3%) had family member who were able to regularly help solve problems in daily life (Table 2).
Table 2. Medication adherence level by family support factors
Variables
|
Total n (%)
|
Adherence Level n (%)
|
P
|
Low
|
Medium
|
High
|
Family supervision for medication
|
|
|
|
Often
|
411(85.4)
|
105(25.5)
|
110(26.8)
|
196(47.7)
|
0.002
|
Sometimes
|
37(7.7)
|
18(48.6)
|
13(35.1)
|
6(16.2)
|
|
Never
|
33(6.9)
|
6(18.2)
|
9(27.3)
|
18(54.5)
|
|
Family spirit encouragement
|
|
|
|
Often
|
451(93.8)
|
116(25.7)
|
123(27.3)
|
212(47.0)
|
0.012*
|
Sometimes
|
20(4.2)
|
10(50.0)
|
3(15.0)
|
7(35.0)
|
|
Never
|
10(2.1)
|
3(30.0)
|
6(60.0)
|
1(10.0)
|
|
Family relationship
|
|
|
Good
|
460(95.6)
|
121(26.3)
|
124(27.0)
|
215(46.7)
|
0.012*
|
General
|
17(3.5)
|
8(47.1)
|
4(23.5)
|
5(29.4)
|
|
Poor
|
4(0.8)
|
0(0.0)
|
4(100.0)
|
0(0.0)
|
|
Family members help solve problems
|
|
|
Often
|
439(91.3)
|
112(25.5)
|
122(27.8)
|
205(46.7)
|
0.134
|
Sometimes
|
24(5.0)
|
12(50.0)
|
5(20.8)
|
7(29.2)
|
|
Never
|
18(3.7)
|
5(27.8)
|
5(27.8)
|
8(44.4)
|
|
*Means that the theoretical number is too small and adopt the Fisher’s exact test method.
3.4 Society support
TB patients who had one or two close friends accounted for 46.6% of the patients included in the study, but 51 (10.6%) had no close or supportive friends. A small number of TB patients (12.3%) lived alone, while the rest lived with family, friends or colleagues. Nearly 30% of TB patients reported poor relationships with neighbors and co-workers (29.7% and 27.0%, respectively). More than half of TB patients (56.5%) reported that they never participated in group activities, while only 41 (8.5%) were regularly or actively involved in group activities. Patients generally had a good doctor-patient relationship, with a small minority (19.3%) reporting a poor relationship with the medical staff. The mean score of knowledge about TB of the included TB patients was 4.72±1.20, with 146(30.4%) getting full marks, but 24(5%) of the patients still scored less than 3 points. Univariate analysis found that doctor-patient relationship, acquired knowledge of TB and participation in group activities were correlated with medication adherence of TB patients (P<0.05). (Table3).
Table 3. Medication adherence level by society support factors
Variables
|
Total n (%)
|
Adherence Level n (%)
|
P
|
Low
|
Medium
|
High
|
The number of close friends
|
|
|
|
|
0
|
51(10.6)
|
17(33.3)
|
12(23.5)
|
22(43.1)
|
0.729
|
1-2
|
223(46.4)
|
62(27.8)
|
61(27.4)
|
100(44.8)
|
|
≥3
|
207(43.0)
|
50(24.2)
|
59(28.5)
|
98(47.3)
|
|
Living condition
|
|
|
|
|
|
Living alone
|
59(12.3)
|
12(20.3)
|
18(30.5)
|
29(49.2)
|
0.483
|
Living with other
|
422(87.7)
|
117(27.7)
|
114(27.0)
|
191(45.3)
|
|
Relationships with neighbors
|
|
|
|
|
Poor
|
143(29.7)
|
36(25.2)
|
45(31.5)
|
62(43.4)
|
0.394
|
General
|
220(45.7)
|
66(30.0)
|
57(25.9)
|
97(44.1)
|
|
Good
|
118(24.5)
|
27(22.9)
|
30(25.4)
|
61(51.7)
|
|
Relationships with colleagues
|
|
|
|
|
Poor
|
130(27.0)
|
32(24.6)
|
31(23.8)
|
67(51.5)
|
0.566
|
General
|
225(46.8)
|
64(28.4)
|
66(29.3)
|
95(42.2)
|
|
Good
|
126(26.2)
|
33(26.2)
|
35(27.8)
|
58(46.0)
|
|
Relationships with doctors
|
|
|
|
|
Poor
|
93(19.3)
|
30(32.3)
|
29(31.2)
|
34(36.6)
|
0.001
|
General
|
223(46.4)
|
66(29.6)
|
69(30.9)
|
88(39.5)
|
|
Good
|
165(34.3)
|
33(20.0)
|
34(20.6)
|
98(59.4)
|
|
Acquired knowledge of TB
|
|
|
|
|
Poor
|
24(5.0)
|
7(29.2)
|
7(29.2)
|
10(41.7)
|
0.016
|
General
|
149(31.0)
|
55(36.9)
|
37(24.8)
|
57(38.3)
|
|
Good
|
308(64.0)
|
67(21.8)
|
88(28.6)
|
153(49.7)
|
|
Participation in group activities
|
|
|
|
|
Often
|
41(8.5)
|
11(26.8)
|
14(34.1)
|
16(39.0)
|
0.008
|
Sometimes
|
168(34.9)
|
55(32.7)
|
53(31.5)
|
60(35.7)
|
|
Never
|
272(56.5)
|
63(23.2)
|
65(23.9)
|
144(52.9)
|
|
3.5 National policy support
The number of patients (59%) who knew about the country's treatment policies for TB was relatively small. There was no significant association with medication adherence (P>0.05), although unknown patients had a higher percentage of low adherence. Satisfaction with the national medical security policy for TB treatment and the need to increase policy support for TB treatment were proved to be relevant factors (P<0.05). The proportion of TB patients who were satisfied with the national medical security policy for TB treatment was 50.7%, while the proportion who were generally satisfied and less satisfied with the medical security policy was 39.3% and 10.0%, respectively. Most patients (96.7%) believed that the government still needed to increase support for TB treatment (Table 4).
Table 4. Medication adherence level by national policy support factors
Variables
|
Total n (%)
|
Adherence Level n (%)
|
P
|
Low
|
Medium
|
High
|
National TB treatment policy
|
|
|
|
|
|
Know
|
284(59.0)
|
74(26.1)
|
69(24.3)
|
141(49.6)
|
0.083
|
Unknow
|
197(41.0)
|
55(27.9)
|
63(32.0)
|
79(40.1)
|
|
Medical security policy satisfaction
|
|
|
|
|
|
Satisfaction
|
244(50.7)
|
66(27.0)
|
52(21.3)
|
126(51.6)
|
0.004
|
General satisfaction
|
189(39.3)
|
54(28.6)
|
58(30.7)
|
77(40.7)
|
|
Not too satisfaction
|
48(10.0)
|
9(18.8)
|
22(45.8)
|
17(35.4)
|
|
Increase support for TB treatment policies
|
|
|
|
|
Need
|
408(84.8)
|
98(24.0)
|
113(27.7)
|
197(48.3)
|
0.004
|
General need
|
57(11.9)
|
27(47.4)
|
15(26.3)
|
15(26.3)
|
|
Not too need
|
16(3.3)
|
4(25.0)
|
4(25.0)
|
8(50.0)
|
|
3.6 Ordinal logistic regression analysis of factors independently associated with medication adherence among patients
Ordinal logistic regression analysis showed that patients whose family members regularly supervised medication (OR:0.34, 95% CI:0.16-0.71) and whose family members often encouraged them mentally (OR:0.14, 95% CI:0.03-0.80) were more likely to have a high medication adherence. Patients with better doctor-patient relationship (OR:0.58, 95% CI:0.38-0.88) and more TB-related knowledge (OR:0.49, 95% CI:0.33-0.72) were more likely to stay away from low medication adherence. Patients who want greater support for TB treatment policies (OR:0.39, 95% CI:0.22-0.67) were more likely to have a good medication adherence. However, patients with a college degree or higher (OR:1.78, 95% CI:1.10-2.88) were more likely to have a low medication adherence (Table 5).
Table 5. Ordinal logistic regression analysis the predictors of medication adherence
Variables
|
OR
|
95% CI
|
P
|
Education (Ref: College degree or above)
|
|
|
|
Junior high school or below
|
1.17
|
0.76-1.81
|
0.471
|
Technical secondary school or high school
|
1.78
|
1.10-2.88
|
0.018
|
Time to arrive at the medical facility (Ref:>60mins)
|
|
|
|
<31
|
1.21
|
0.76-1.92
|
0.434
|
31-60
|
1.20
|
0.79-1.82
|
0.403
|
Family supervision for medication (Ref: Often)
|
|
|
|
Never
|
2.79
|
1.19-6.53
|
0.018
|
Sometimes
|
0.34
|
0.16-0.71
|
0.004
|
Family spirit encouragement (Ref: Often)
|
|
|
|
Never
|
0.14
|
0.03-0.80
|
0.027
|
Sometimes
|
0.75
|
0.28-2.03
|
0.572
|
Family relationship (Ref: Good)
|
|
|
|
Poor
|
2.62
|
0.22-30.60
|
0.443
|
General
|
0.83
|
0.30-2.28
|
0.720
|
Relationships with doctors (Ref: Good)
|
|
|
|
Poor
|
0.72
|
0.42-1.23
|
0.229
|
General
|
0.58
|
0.38-0.88
|
0.010
|
Acquired knowledge of TB (Ref: Good)
|
|
|
|
Poor
|
0.79
|
0.34-1.82
|
0.577
|
General
|
0.49
|
0.33-0.72
|
<0.001
|
Participation in group activities (Ref: Often)
|
|
|
|
Never
|
1.28
|
0.66-2.47
|
0.461
|
Sometimes
|
0.87
|
0.45-1.68
|
0.678
|
Medical security policy satisfaction (Ref: Satisfaction)
|
|
|
|
Not too satisfaction
|
0.88
|
0.47-1.65
|
0.698
|
General satisfaction
|
0.80
|
0.55-1.17
|
0.254
|
Increase support for TB treatment policies (Ref: Need)
|
|
|
|
Not too need
|
0.85
|
0.31-2.31
|
0.749
|
General need
|
0.39
|
0.22-0.67
|
0.001
|
Ref is reference
3.7 Patients' advice for increased policy support in an open question
Some TB patients stated one or more advice for greater policy support, and advice was received from 109 TB patients. More economic support (87.2%) was most frequently proposed, and include increased reimbursement rates (33.9 %), increased subsidies for nutrition, transportation, and other expenses (19.3 %), increased free drug coverage and time (14.7 %), free treatment (11.9%), and reduced testing costs (3.7%), especially among poor patients and patients with first-line drugs resistance. The next most common advice was to increase TB prevention and patient management (especially isolation and education of infectious patients) (9.2%), to propagandize knowledge of TB (4.6%), to increase psychological counselling (4.6%), to provide nutritious meals to inpatients or to improve the quality of food and beverage (2.8%).There are also a few patients who proposed to strengthen infrastructure construction, improve out-of-town reimbursement, provide job security and medication supply security for patients, optimize the problems of hospital transfer, reduce drug side effects, improve the success rate of treatment, and increase the special protection for patients with comorbidities and students(Fig. 1).