Reliability of the questionnaires
The reliability of the life-satisfaction questionnaire had been already examined previously [8]. Therefore, we examined the reliability of the questionnaire questions of self-efficacy. In the present study, the value of Cronbach’s alpha, which is regarded as an index of reliability, was 0.840 for the self-efficacy questions. Principal component analysis of variables was performed in order to confirm the construct validity. Based on the eigenvalue greater than 1, the questionnaire questions included two-factors consisting total adherence and drug taking confidence. The first factor accounts for 42.2% of the total variance. The second one accounts for 14.2%.
Relationship of the life-satisfaction score with nutritional parameters
According to the score of the life-satisfaction questionnaire, we divided into three groups which have low, intermediate and high life-satisfaction. Female sex, non-diabetic patients, and working patients had higher life-satisfaction score significantly. The self-efficacy score, nPCR, CGR were better and body mass index (BMI), body surface area (BSA), height, and dry weight were lower significantly if the life satisfaction score was higher (Table 2).
Table 2 Demographic and clinical characteristics of the patients divided by the life-satisfaction scores
Tertiles of life-satisfaction score (n = 196)
|
|
Low
|
Intermediate
|
High
|
Number of patients
|
64
|
62
|
70
|
Age, years
|
65.6 ± 9.9
|
66.4 ± 10.8
|
69.1 ± 12.1
|
Female, n (%)
|
17 (26.6%)
|
16 (25.8%)
|
32 (45.7%) **
|
Diabetes, n (%)
|
27 (42.2%)
|
16 (25.8%)
|
13 (18.6%) *
|
Hemodialysis duration, years
|
11.8 ± 9.2
|
10.1 ± 8.5
|
11.8 ± 9.3
|
Body mass index, kg/m2
|
22.2 ± 3.8
|
22.7 ± 3.1
|
21.1 ± 3.7 **
|
Body surface area, m2
|
1.63 ± 0.19
|
1.63 ± 0.17
|
1.53 ± 0.19 *
|
Height, cm
|
163.2 ± 8.4
|
162.1 ± 8.1
|
158.8 ± 9.0 *
|
Dry weight, kg
|
59.5 ± 13.0
|
59.9 ± 10.7
|
53.5 ± 12.3 *
|
Life-satisfaction
|
9.0 ± 1.9
|
12.9 ± 0.9
|
17.0 ± 1.6 *
|
Self-efficacy
|
19.1 ± 3.9
|
19.1 ± 2.9
|
21.4 ± 4.1 *
|
Working, n (%)
|
13 (20.3%)
|
17 (27.4%)
|
27 (38.6%) **
|
Live alone, n (%)
|
8 (12.5%)
|
6 (9.7%)
|
5 (7.1%)
|
Emergent dialysis initiation, n (%)
|
14 (21.9%)
|
15 (24.2%)
|
18 (25.7%)
|
Late referral, n (%)
|
27 (42.2%)
|
30 (48.4%)
|
33 (47.1%)
|
Normalized PCR, g/kg/day
|
0.88 ± 0.14
|
0.89 ± 0.30
|
0.92 ± 0.12 **
|
Creatinine generation rate, %
|
95.1 ± 23.1
|
100.4 ± 22.5
|
105.9 ± 22.4*
|
Geriatric nutritional risk index
|
93.7 ± 9.7
|
93.2 ± 7.8
|
90.8 ± 7.7
|
Albumin, g/dL
|
3.48 ± 0.28
|
3.41 ± 0.25
|
3.44 ± 0.21
|
Potassium, mEq/L
|
4.72 ± 0.55
|
4.76 ± 0.63
|
4.88 ± 0.59
|
Phosphate, mg/dL
|
5.15 ± 0.86
|
5.25 ± 0.92
|
5.46 ± 1.00
|
Mean IDWG, % of DW
|
4.06 ± 0.99
|
3.97 ± 1.19
|
3.96 ± 1.11
|
PCR: Protein catabolic rate. IDWG: interdialytic weight gain. DW: Dry weight.
Ordinal variables were analyzed by Cochran-Armitage test.
Interval variables were analyzed by Jonckheere-Terpstra test.
*: p<0.01. **: p<0.05.
Subanalysis of the life-satisfaction score in each clinic
Subanalysis of the life-satisfaction score in each clinic was done to determine whether the relationship between the life-satisfaction score and the patients’ characteristics were common or not. Female sex, non-diabetic patients, and working patients had higher life-satisfaction scores in every clinic. The self-efficacy score was higher if the life-satisfaction scores was higher at all clinics. nPCR and CGR were higher in patients with higher self-satisfaction scores in three clinics (Supplementary file 3: Table S1).
Relationship of self-efficacy on dietary and drug adherence with demographic and clinical characteristics
The enrolled patients were divided into low, intermediate and high self-efficacy groups as well as the analysis of the life-satisfaction score. Elderly and non-working patients had higher self-efficacy scores significantly. The life-satisfaction score was higher and GNRI was lower significantly if the self-efficacy score was higher (Table 3).
Table 3 Demographic and clinical characteristics of the patients divided by the self-efficacy scores
Tertiles of self-efficacy score (n = 196)
|
|
Low
|
Intermediate
|
High
|
Number of patients
|
65
|
69
|
62
|
Age, years
|
63.9 ± 10.2
|
66.7 ± 11.0
|
71.0 ± 10.9 *
|
Female, n (%)
|
24 (36.9%)
|
21 (30.4%)
|
20 (32.3%)
|
Diabetes, n (%)
|
17 (26.2%)
|
19 (27.5%)
|
20 (32.3%)
|
Hemodialysis duration, years
|
12.0 ± 8.7
|
12.0 ± 10.4
|
9.8 ± 7.6
|
Body mass index, kg/m2
|
22.5 ± 3.9
|
22.2 ± 3.9
|
21.1 ± 2.7
|
Body surface area, m2
|
1.63 ± 0.21
|
1.60 ± 0.20
|
1.56 ± 0.16
|
Height, cm
|
162.2 ± 8.4
|
161.3 ± 9.2
|
160.3 ± 8.5
|
Dry weight, kg
|
59.6 ± 13.9
|
58.2 ± 12.9
|
54.5 ± 9.1
|
Life-satisfaction
|
11.9 ± 3.6
|
12.9 ± 3.6
|
14.4 ± 3.5 *
|
Self-efficacy
|
15.9 ± 2.2
|
19.9 ± 0.8
|
24.1 ± 2.4 *
|
Working, n (%)
|
23 (35.4%)
|
22 (31.9%)
|
12 (19.4%)**
|
Live alone, n (%)
|
8 (12.3%)
|
2 (2.9%)
|
9 (14.5%)
|
Emergent dialysis initiation,n(%)
|
16 (24.6%)
|
15 (21.7%)
|
16 (25.8%)
|
Late referral, n (%)
|
33 (50.8%)
|
30 (43.5%)
|
27 (43.5%)
|
Normalized PCR, g/kg/day
|
0.90 ± 0.13
|
0.93 ± 0.28
|
0.86 ± 0.14
|
Creatinine generation rate, %
|
103.3 ± 21.0
|
97.9 ± 24.2
|
101.0 ± 23.7
|
Geriatric nutritional risk index
|
93.3 ± 8.4
|
93.7 ± 9.6
|
90.4 ± 6.7 **
|
Albumin, g/dL
|
3.45 ± 0.24
|
3.46 ± 0.26
|
3.43 ± 0.24
|
Potassium, mEq/L
|
4.86 ± 0.61
|
4.80 ± 0.57
|
4.71 ± 0.60
|
Phosphate, mg/dL
|
5.34 ± 1.06
|
5.28± 0.90
|
5.24 ± 0.85
|
Mean IDWG, % of DW
|
4.18 ± 1.11
|
3.94 ± 1.08
|
3.87 ± 1.08
|
PCR: Protein catabolic rate. IDWG: interdialytic weight gain. DW: Dry weight.
Ordinal variables were analyzed by Cochran-Armitage test.
Interval variables were analyzed by Jonckheere-Terpstra test.
*: p<0.01. **: p<0.05.
Subanalysis of the self-efficacy score in each clinic
Subanalysis of the self-efficacy score in each clinic was done as well. Elderly patients had higher self-efficacy scores in every clinic. The life-satisfaction score was higher in patients with high self-efficacy scores in every clinic. Non-working patients and low GNRI were related to high self-efficacy in three clinics (Supplementary file 4: Table S2). Even when patients were divided to young (67 years old or less) and elderly (68 years old or more) patients by the average age of the participants (Table 1), higher self-efficacy scores were not related to nutritional parameters significantly (Supplementary file 5: Table S3).