Study Population
Figure 1 shows the flow of the study participants. A total of 2,442 patients aged ≥40 years participated in J-DOPPS Phase V. Among the 1,593 patients who answered the questionnaire at baseline, 148 were excluded because they were not independent in more than three of the five BADLs at baseline. In the follow-up survey, 1,200 patients answered the questionnaire. After excluding 18 patients with an evaluation period for the change in ADLs of less 90 days, 849 patients who had no missing data were included in the statistical analysis.
Table 1 depicts the baseline characteristics of the 849 participants included in the statistical analysis. Their mean age was 64.9 years (SD, 10.7), and 64.4% of the participants were men. The proportion of patients with diabetes was 36.3%. During the follow-up period (median, 359 days; IQR, 337–375 days), 228 (26.9%) patients experienced at least one hospitalization with a stay of more than two days. Figure 2 depicts the distribution of the cumulative length of hospital stay among patients who had been hospitalized. The median of the cumulative length of hospital stay was 11 days (range 2–187). A decline in BADLs and IADLs occurred in 143 (16.8%) and 198 (23.3%) patients from the baseline and follow-up, respectively.
Table 1. Baseline characteristics of study participants
|
Total
(n=849)
|
Non-hospitalized Participants
(n=621)
|
Hospitalized Participants
(n=228)
|
Age (years), mean [SD]
|
64.9 [10.7]
|
64.3 [10.8]
|
66.7 [10.1]
|
Male, n (%)
|
547 (64.4)
|
397 (63.9)
|
150 (65.8)
|
Dialysis vintage (years), median [IQR]
|
5.6 [2.2-11.6]
|
5.5 [2.1-11.4]
|
5.6 [2.3-12.6]
|
BMI (kg/m2), n (%)
|
|
|
|
<18.5
|
176 (20.7)
|
120 (19.3)
|
56 (24.6)
|
≥18.5 to <25
|
559 (65.8)
|
416 (67.0)
|
143 (62.7)
|
≥25
|
114 (13.4)
|
85 (13.7)
|
29 (12.7)
|
Functional status score, n (%)
|
|
|
|
< 8
|
49 (5.8)
|
34 (5.5)
|
15 (6.6)
|
8 to < 11
|
77 (9.1)
|
55 (8.9)
|
22 (9.6)
|
11 to < 13
|
171 (20.1)
|
108 (17.4)
|
63 (27.6)
|
13
|
552 (65.0)
|
424 (68.3)
|
128 (56.1)
|
Comorbidities, n (%)
|
|
|
|
Diabetes
|
308 (36.3)
|
212 (34.1)
|
96 (42.1)
|
Cerebrovascular diseases
|
71 (8.4)
|
49 (7.9)
|
22 (9.6)
|
Coronary heart disease
|
202 (23.8)
|
138 (22.2)
|
64 (28.1)
|
Other cardiovascular diseases
|
175 (20.6)
|
118 (19.0)
|
57 (25.0)
|
Congestive heart disease
|
131 (15.4)
|
89 (14.3)
|
42 (18.4)
|
Cancer other than skin cancer
|
84 (9.9)
|
61 (9.8)
|
23 (10.1)
|
Neurological disease
|
35 (4.1)
|
25 (4.0)
|
10 (4.4)
|
Peripheral vascular disease
|
110 (13.0)
|
65 (10.5)
|
45 (19.7)
|
Dementia
|
3 (0.4)
|
2 (0.3)
|
1 (0.4)
|
Psychiatric disorder
|
32 (3.8)
|
21 (3.4)
|
11 (4.8)
|
Albumin (g/dL), mean [SD]
|
3.7 [0.4]
|
3.7 [0.4]
|
3.6 [0.4]
|
Phosphorus (mg/dL), mean [SD]
|
5.3 [1.3]
|
5.3 [1.3]
|
5.3 [1.4]
|
Creatinine (mg/dL), mean [SD]
|
10.7 [2.5]
|
10.9 [2.5]
|
10.4 [2.5]
|
Blood urea nitrogen (mg/dL), mean [SD]
|
63.8 [14.1]
|
64.1 [14.1]
|
63.1 [14.1]
|
Hemoglobin (g/L), mean [SD]
|
10.6 [1.1]
|
10.6 [1.1]
|
10.5 [1.3]
|
Single-pool Kt/V, mean [SD]
|
1.4 [0.3]
|
1.4 [0.3]
|
1.4 [0.3]
|
SD, standard deviation; IQR, interquartile range;
Proportion of Decline for each BADL and IADL
Figure 3 depicts the proportion of BADLs and IADLs, showing a decline according to patients’ hospitalization status. Among the BADLs, a decline in bathing was most frequently observed (4%), whereas a decline in using the toilet was least frequently observed (1%) in the hospitalized patients. Among the IADLs, 20% of the hospitalized patients showed a decline in getting to places beyond walking distance. Even the IADLs showing the lowest proportions of decline (i.e., taking medications and using the telephone) had higher proportions than bathing (a BADL). For all BADLs and IADLs, the proportion of decline was higher among hospitalized than non-hospitalized patients.
Association between Cumulative Length of Hospital Stay and Declines in BADLs and IADLs
The cumulative length of hospital stay was associated with an increased risk of decline in BADLs (adjusted RR (per day), 1.012, 95% CI [1.005 to 1.020], P = 0.002) and IADLs (adjusted RR (per day), 1.011, 95% CI [1.004 to 1.019], P=0.003). For age (per 10-year increment), the RRs [95% CI] of declines in BADLs and IADLs were 1.20, 95% CI [0.96–1.50] (P=0.176) and 1.21, 95% CI [1.00 to 1.47] (P=0.046), respectively. Regarding diabetes, the RRs [95% CI] for declines were 1.36, 95% CI [0.97 to 1.91] (P=0.072) for BADLs and 1.38, 95% CI [1.04 to 1.84] (P=0.027) for IADLs (Table 2). A restricted cubic spline curve analysis showed the linear relationships between the cumulative length of hospital stay and predicted probabilities of decline in BADLs and IADLs using the restricted cubic spline curve (Figure 4).
Table 2. Association between cumulative length of hospital stay and decline in ADLs
|
BADLs
|
IADLs
|
|
Crude RR
[95% CI]
|
Adjusted RRa
[95% CI]
|
Crude RR
[95% CI]
|
Adjusted RRa
[95% CI]
|
Cumulative length of hospital stay
|
|
|
|
|
Per day
|
1.014
[1.006 to 1.022]
|
1.012
[1.005 to 1.020]
|
1.015
[1.007 to 1.023]
|
1.011
[1.004 to 1.019]
|
Per 7 days
|
1.10
[1.04 to 1.16]
|
1.12
[1.04 to 1.19]
|
1.11
[1.05 to 1.17]
|
1.08
[1.03 to 1.14]
|
Per 30 days
|
1.47
[1.18 to 1.83]
|
1.42
[1.15 to 1.75]
|
1.51
[1.23 to 1.85]
|
1.38
[1.13 to 1.68]
|
Age (per 10 years)
|
|
1.20
[0.96 to 1.50]
|
|
1.21
[1.00 to 1.47]
|
Diabetes
|
|
1.36
[0.97 to 1.91]
|
|
1.38
[1.04 to 1.84]
|
ADL, activities of daily living; BADL, basic activities of daily living; IADL, instrumental activities of daily living; RR, risk ratio; CI, confidence interval.
a Length of hospital stay, age and diabetes were included simultaneously in the model after adjusting for sex, dialysis vintage, body mass index (< 18.5, ≥ 18.5 to < 25, ≥ 25), functional status score, comorbidities (cerebrovascular diseases, coronary heart disease, other cardiovascular diseases, congestive heart disease, cancer other than skin cancer, neurologic disease, peripheral vascular disease, dementia, and psychiatric disorder), albumin, phosphorus, creatinine and single-pool Kt/V.
Association between Number of Hospitalizations and Decline in BADLs and IADLs
The incidences of decline in BADLs according to the number of hospitalizations categorized as 0, 1, and ≥2 were 13.0% (n=81/621), 22.9% (n=32/140), and 34.1% (n=30/88), respectively, whereas the incidences of decline in IADLs were 20.0% (n=124/621), 30.7% (n=43/140), and 35.2% (n=31/88), respectively. The adjusted RRs for declines in BADLs among patients with one hospitalization and for those with more than two hospitalizations were 1.69, 95% CI [1.17 to 2.45] (P=0.006) and 2.27, 95% CI [1.58 to 3.26] (P<0.001), respectively, compared to patients without hospitalization. The adjusted RRs [95% CI] for declines in IADLs in patients with one and more than two hospitalizations were 1.41, 95% CI [1.09 to 1.83] (P=0.008) and 1.39, 95% CI [0.97 to 1.99] (P=0.074), respectively (Table 3).
Table 3. Association between number of hospitalizations and decline in ADLs
|
BADLs
|
IADLs
|
|
Incidence
|
Crude RR
[95% CI]
|
Adjusted RRa
[95% CI]
|
Incidence
|
Crude RR
[95% CI]
|
Adjusted RRa
[95% CI]
|
Number of hospitalizations
|
|
|
|
|
|
|
0
|
13.0%
(81/621)
|
ref
|
ref
|
20.0%
(124/621)
|
ref
|
ref
|
1
|
22.9%
(32/140)
|
1.75
[1.18 to 2.61]
|
1.69
[1.17 to 2.45]
|
30.7%
(43/140)
|
1.54
[1.15 to 2.06]
|
1.41
[1.09 to 1.83]
|
≥2
|
34.1%
(30/88)
|
2.61
[1.82 to 3.75]
|
2.27
[1.58 to 3.26]
|
35.2%
(31/88)
|
1.76
[1.29 to 2.41]
|
1.39
[0.97 to 1.99]
|
BADL, basic activities of daily living; IADL, instrumental activities of daily living; RR, risk ratio; CI, confidence interval; ref, reference.
aAdjusted for age, sex, dialysis vintage, body mass index (<18.5, ≥18.5 to <25, ≥25), functional status score, comorbidities (diabetes, cerebrovascular diseases, coronary heart disease, other cardiovascular diseases, congestive heart disease, cancer other than skin cancer, neurologic disease, peripheral vascular disease, dementia, and psychiatric disorder), albumin, phosphorus, creatinine and single-pool Kt/V.
Association between Hospitalization and Decline in BADLs and IADLs by Age Group
The incidences of decline in BADLs and IADLs were 13.7% and 18.7% among patients aged under 65 years and 19.6% and 27.5% among patients aged 65 years or older, respectively. The adjusted RRs for declines in BADLs and IADLs per one day increment in the cumulative length of hospital stay were 1.013, 95% CI [1.001 to 1.025] (P=0.034) and 1.017, 95% CI [1.006 to 1.028] (P=0.003), respectively, among patients aged below 65 years, whereas they were 1.014, 95% CI [1.003 to 1.026] (P=0.017) and 1.007, 95% CI [0.999 to 1.015] (P=0.112), respectively, among patients aged 65 years or older (p for interaction = 0.52 for BADLs and 0.43 for IADLs). Among patients aged below 65 years, the adjusted RRs [95% CI] for declines in BADLs and IADLs were 1.83, 95% CI [1.00 to 3.35] (P=0.512) and 1.53, 95% CI [0.96 to 2.42] (P=0.072) among patients with one hospitalization and 2.65, 95% CI [1.49 to 4.70] (P<0.001) and 1.38, 95% CI [0.82 to 2.32] (P=0.23) among those with more than two hospitalizations, respectively, compared to patients without a hospitalization. Among patients aged 65 years or older, the adjusted RRs for declines in BADLs and IADLs were 1.56, 95% CI [1.00 to 2.43] (P=0.051) and 1.30, 95% CI [0.96 to 1.77] (P=0.087) among patients with one hospitalization and 2.04, 95% CI [1.35 to 3.07] (P<0.001) and 1.32, 95% CI [0.85 to 2.06] (P=0.22) among those with more than two hospitalizations, respectively. The interactions were not statistically significant (P for interaction = 0.82 for BADLs and 0.87 for IADLs; Table 4). The results from the other secondary analyses by dialysis vintage group and cause-specific hospitalizations are shown in Additional file 1:Table S1, Table S2.
Table 4. Association between hospitalization and decline in ADLs by age group
|
BADLs
|
IADLs
|
|
<65 years
(n=401)
|
≥65 years
(n=448)
|
<65 years
(n=401)
|
≥65 years
(n=448)
|
|
Adjusted RRa [95% CI]
|
Cumulative length of hospital stay (per day)
|
1.013
[1.001 to 1.025]
|
1.014
[1.003 to 1.026]
|
1.017
[1.006 to 1.028]
|
1.007
[0.999 to 1.015]
|
|
P for interactionb = 0.52
|
P for interactionb = 0.43
|
|
|
BADLs
|
IADLs
|
|
<65 years
(n=401)
|
≥65 years
(n=448)
|
<65 years
(n=401)
|
≥65 years
(n=448)
|
Number of hospitalizations
|
|
|
|
|
0
|
ref
|
ref
|
ref
|
ref
|
1
|
1.83
[1.00 to 3.35]
|
1.56
[1.00 to 2.43]
|
1.53
[0.96 to 2.42]
|
1.30
[0.96 to 1.77]
|
≥2
|
2.65
[1.49 to 4.70]
|
2.04
[1.35 to 3.07]
|
1.38
[0.82 to 2.32]
|
1.32
[0.85 to 2.06]
|
|
P for interactionc =0.82
|
P for interactionc =0.87
|
BADL, basic activities of daily living; IADL, instrumental activities of daily living; RR, risk ratio; CI, confidence interval; ref, reference.
aAdjusted for age, sex, dialysis vintage, body mass index (<18.5, ≥18.5 to <25, ≥25), functional status score, comorbidities (diabetes, cerebrovascular diseases, coronary heart disease, other cardiovascular diseases, congestive heart disease, cancer other than skin cancer, neurologic disease, peripheral vascular disease, dementia, and psychiatric disorder), albumin, phosphorus, creatinine and single-pool Kt/V.
bTesting the statistical significance of the product terms in the cumulative length of hospitalization and age group using a Wald test.
cTesting the statistical significance of product terms in the number of hospitalizations and age group using a Wald test.
Sensitivity Analysis
On defining the outcomes as a decline in two of the five BADLs and eight IADLs, the adjusted RRs for declines in BADLs and IADLs per 1-day increment in cumulative length of hospital stay were 1.010, 95% CI [1.000 to 1.020] (P=0.051) and 1.007, 95% CI [0.999 to 1.015] (P=0.077), respectively. Further, on analyzing only those patients whose evaluation periods for ADL changes were within the IQR (337–375 days), the adjusted RRs [95% CI] for declines in BADLs and IADLs per one day increment in cumulative length of hospital stay were 1.027, 95% CI [1.012 to 1.042] (P<0.001) and 1.007, 95% CI [0.999 to 1.015] (P=0.076), respectively. Finally, in the analysis including 148 patients who were excluded due to dependence in at least three of the five BADLs in primary analysis, the adjusted RRs for declines in BADLs and IADLs per one day increment in cumulative length of hospital stay were 1.012, 95% CI [1.004 to 1.019] (P=0.002) and 1.012, 95% CI [1.004 to 1.019] (P=0.003), respectively.