A total of 1,947 individuals were included, of whom 1,710 (87.8%) were assigned to the control group and 237 (12.2%) were assigned to the experimental group. Regarding the demographic information of these individuals, the percentages of individuals who were eligible for social welfare, had disabilities, and had severe disabilities were significantly greater in the experimental group than those in the control group. This indicates that individuals in the experimental group had greater complexity in their long-term care needs (Table 1).
Table 1
Demographic information (N = 1947)
|
Experimental group
|
Control group
|
p-Value
|
Item
|
(n = 237)
|
(n = 1710)
|
Age, years, M (SD)
|
77.0 (14.7)
|
77.8 (14.1)
|
0.419
|
Gender N (%)
|
|
|
|
Female
|
128 (54)
|
1010 (59.1)
|
0.136
|
Male
|
109 (46)
|
699 (40.9)
|
|
Marital status N (%)
|
|
|
|
Divorced or single
|
129 (54.4)
|
994 (58.2)
|
0.276
|
Married
|
108 (45.6)
|
715 (41.8)
|
|
Social welfare status N (%)
|
|
|
|
Eligible*
|
59 (24.9)
|
236 (13.8)
|
< 0.001*
|
Not eligible
|
178 (75.1)
|
1473 (86.2)
|
|
Disability card
|
|
|
|
Possessed
|
159 (67.1)
|
994 (58.2)
|
0.009*
|
No
|
78 (32.9)
|
715 (41.8)
|
|
Living alone N (%)
|
|
|
|
No
|
202 (85.2)
|
1432 (83.8)
|
0.571
|
Yes
|
35 (14.8)
|
277 (16.2)
|
|
CMS rating**
|
|
|
|
Mild
|
26
|
372
|
< 0.001*
|
Moderate
|
98
|
756
|
|
Severe
|
113
|
582
|
|
* mid or low-income household, disability living allowance for non-listed |
** Long-Term Care Case-Mix System |
Regarding the changes in disability postponement and depression improvement (Table 2), the percentages in the experimental group were greater than those in the control group for CMS improvement (p = 0.004), ADL score maintenance or improvement (p = 0.013), and CESD score maintenance or improvement (p = 0.012).
Table 2
Disability postponement, depression, and caregiver load after intervention
|
Experimental group
|
Control group
|
p-Value
|
Item
|
(n = 237)
|
(n = 1710)
|
CMS N (%)
|
|
|
|
Improved
|
55 (23.2)
|
283 (16.6)
|
0.004*
|
Unchanged
|
108 (45.6)
|
965 (56.4)
|
|
Deteriorated
|
74 (31.2)
|
462 (27.0)
|
|
ADL N (%)
|
|
|
|
Improved
|
43 (18.5)
|
367 (22.1)
|
0.013*
|
Unchanged
|
136 (58.6)
|
804 (48.4)
|
|
Deteriorated
|
53 (22.8)
|
491 (29.5)
|
|
missing = 53
|
|
|
|
CESD N (%)
|
|
|
|
Improved
|
8 (11.4)
|
48 (18.5)
|
0.012*
|
Unchanged
|
59 (84.3)
|
173 (66.8)
|
|
Deteriorated
|
3 (4.3)
|
38 (14.7)
|
|
missing = 1618
|
|
|
|
After controlling for other variable, the risks of deterioration in the CMS ratings, ADL scores, and CESD scores of the two groups were compared by the logistic regression in Table 3. For CMS rating deterioration, the odds ratio of the experimental group was 2.09 times that of the control group (95% CI = 1.03–4.24). Regarding the collinearity between the CMS rating and ADL score, when the ADL score was excluded from the control variables, no significant difference was observed in the deterioration risks of the two groups. For ADL scores, the two groups did not exhibit a significant difference (OR = 0.97, 95% CI = 0.50–1.88). For CESD scores, the odds ratio of the experimental group was 0.20 (95% CI = 0.05–0.73) compared with that of the control group, suggesting that the experimental group faced a significantly lower risk of CESD score deterioration than did the control group.
Table 3
CMS rating, ADL score, and CESD score after intervention
Item
|
Group
|
OR
|
95% C.I.
|
P
|
Risk of deterioration, CMS rating
|
Control group
|
1
|
|
|
|
Experimental group
|
2.09
|
1.03–4.24
|
0.041*
|
Risk of deterioration, ADL score
|
Control group
|
1
|
|
|
|
Experimental group
|
0.97
|
0.50–1.88
|
0.929
|
Risk of deterioration, CESD score
|
Control group
|
1
|
|
|
|
Experimental group
|
0.20
|
0.05–0.73
|
0.015*
|
Regarding the expended medical resources (according to the medical records kept by the hospital) before and after the intervention, for the first 6 months, the control group demonstrated a reduction in hospitalisation expenses (by 35.3%), hospitalisations (p = 0.004), and days of hospitalisation (p = 0.001), which indicates an effective reduction in the days of hospitalisation. Emergency care expenses (reduced by 35.6%, p = 0.003) and instances of emergency care (p < 0.001) were lower, indicating an effective reduction of emergency care expenses in the experimental group (Table 4). A comparison of the two groups reveals that they exhibited no significant difference in either emergency care expenses or instances of emergency care (Table 5).
Table 4 Expended medical resources before and after intervention in the experimental group in the first 6 months (N= 1947)
Table 5
Changes in expended medical resources after intervention
|
Experimental group
|
Control group
|
p-Value
|
Item
|
(n = 237)
|
(n = 1710)
|
Outpatient visits M (SD)
|
|
|
|
|
|
|
|
Times
|
0.42 (7.94)
|
0.41 (9.17)
|
0.648
|
Expenses (NTD)
|
1318.15 (16670.90)
|
1369.88 (25683.48)
|
0.978
|
Hospitalisation M (SD)
|
|
|
|
|
|
|
|
Times
|
0.01 (0.32)
|
-0.01 (0.31)
|
0.111
|
Expenses (NTD)
|
-5429.88 (84713.21)
|
-70666.84 (79246.60)
|
0.624
|
The costs of the experimental group receiving long-term care with advanced case management were analysed to assess the adequacy of the management based on actual costs. The total annual cost for four stations would be NT$5.84 million. The four stations served a total of 517 individuals annually. Regarding the reduction in the expended medical resources, in the experimental group, National Health Insurance expenses valued at NT$480,003 were saved annually, calculated based on the reduction in instances of emergency care before and after intervention in the first 6 months and in the emergency care expenses of NT$2,579 per person per use. Additional National Health Insurance expenses valued at NT$10,513,236 were saved annually, calculated based on the reduction in days of hospitalisation and the hospitalisation expense of NT$5,241 per person per day. In summary, community-based advanced case management, with which it served 517 individuals per year and saved NT$10,993,239 by reducing emergency care and hospitalisation expenses; what it achieved far outweighed the required investment. For intangible benefits, the proactive and rapid provision of human-centred, family-oriented, community-based medical and care services that were customised to the care receiver enabled the integration of long-term care with the National Health Insurance system, thereby reducing the waste of health insurance resources and lowering the costs incurred on the taxpayers and patients’ families.