Impact of Elderly Acute Care Discharge Services on Prevention of Rehospitalisation: A Retrospective Cohort Study Using National Health Data from Kita Ward, Tokyo

Masumi Takei; Satoshi Miyata; Mariko Inoue; Kenzo Takahashi

doi:10.5334/ijic.8913

Figures & Tables

Table 1

Patient characteristics and outcomes with and without discharge services.*

	TOTAL (n = 6,681)				WITHOUT DISCHARGE SERVICES (n = 2,775)				WITH DISCHARGE SERVICES: (n = 3,906)				p-VALUE
	NO.	(%)	MEAN (SD)	MEDIAN (IQR)	NO.	(%)	MEAN (SD)	MEDIAN (IQR)	NO.	(%)	MEAN (SD)	MEDIAN (IQR)	p-VALUE
Age, years			81.3 (7.83)				81.1 (7.82)				81.4 (7.84)		0.13
65–74	1,535	(23.0)			682	(24.6)			853	(21.9)			0.03
75–84	2,874	(43.1)			1,183	(42.7)			1,691	(43.3)
≥85	2,265	(33.9)			908	(32.7)			1,357	(34.8)
Sex
Male	2,900	(43.4)			1,195	(43.1)			1,705	(43.7)			0.63
Female	3,781	(56.6)			1,580	(56.9)			2,201	(56.3)
Number of complications
>5	3,378	(50.6)			1,185	(42.7)			2,185	(55.9)			<0.001
Length of stay, days			11.4 (8.61)	9 (4, 17)			10.5 (8.74)	8(3, 16)			12.1 (8.46)	10 (5, 18)	<0.001
<7	2,519	(37.7)			1,229	(44.3)			1,290	(33.0)			<0.001
7–14	2,026	(30.3)			755	(27.2)			1,271	(32.5)
15–21	1,045	(15.6)			368	(13.3)			677	(17.3)
≥22	1,091	(16.3)			423	(15.2)			668	(17.1)
Pre-hospital home care users	806	(12.1)			257	(9.3)			549	(14.1)			<0.001
Pre-hospital facility users	665	(10.0)			198	(7.1)			467	(12.0)			<0.001
Discharge services
Discharge planning	1,263	(18.9)							1,263	(32.3)			<0.001
Coordination with community care	3,378	(50.6)							3,378	(86.5)			<0.001
Rehabilitation discharge instruction	1,374	(20.6)							1,374	(35.2)			<0.001
Readmission	1,872	(28.0)			377	(13.6)			1,495	(38.3)			<0.001
Confirmation of death by home doctor after discharge	199	(3.0)			37	(1.3)			162	(4.1)			<0.001
In-hospital death on readmission	38	(0.7)			18	(0.8)			20	(0.6)			0.301

[i] SD = standard deviation; IQR = interquartile range.

* Indicates Fisher’s exact test, Welch’s t test for comparison of expected values for ages, and Wilcoxon rank sum test for length of stay.

§ Includes the absence of data for 3 months after the end of the person-days censoring.

Table 2

Hazard ratio until readmission event occurs in the group with discharge services and each discharge service compared to the group without discharge services.

	UNIVARIABLE ANALYSIS LOG-RANK TEST				MULTIVARIABLE ANALYSIS COX PROPORTIONAL HAZARDS MODEL
	HR	95% CI		P VALUE	HR	95% CI		P VALUE
Without discharge services	1.00				1.00
With discharge services	2.94	2.63	3.29	<0.001	2.92	2.6	3.27	<0.001
Age					1.03	1.02	1.03	<0.001
Sex					1.19	1.08	1.3	<0.001
Complications*					1.00	0.91	1.10	0.98
Each discharge service
Discharge planning	2.15	1.95	2.37	<0.001	2.02	1.83	2.23	<0.001
Age					1.02	1.01	1.03	<0.001
Sex					1.17	1.06	1.28	<0.001
Complications					1.02	0.93	1.12	0.706
Coordination with community care	2.88	2.60	3.19	<0.001	2.83	2.55	3.14	<0.001
Age					1.02	1.02	1.03	<0.001
Sex					1.15	1.05	1.27	0.003
Complications					1.02	0.93	1.12	0.624
Rehabilitation discharge instruction	1.76	1.60	1.95	<0.001	1.77	1.60	1.95	<0.001
Age					1.03	1.02	1.03	<0.001
Sex					1.22	1.11	1.34	<0.001
Complications					1.03	0.94	1.13	0.51

[i] HR = hazard ratio; CI = confidence interval.

* Complications are adjusted to 4 or less and 5 or more.

Survival function curve for readmission events with or without discharge services.

Table 3

Subgroup analysis of hazard ratios to readmission event for each category in the group with discharge services and each discharge service compared to without discharge services.

	DISCHARGE SERVICES				EACH DISCHARGE SERVICE
	DISCHARGE SERVICES				DISCHARGE PLANNING				COORDINATION WITH COMMUNITY CARE				REHABILITATION DISCHARGE INSTRUCTION
	HR	95% CI		p-VALUE	HR	95% CI		p-VALUE	HR	95% CI		p-VALUE	HR	95% CI		p-VALUE
Without discharge support	1.00
Age, years
65–74 (n = 1,535)	3.54	2.69	4.67	<0.001	2.86	2.22	3.68	<0.001	3.39	2.65	4.35	<0.001	1.82	1.42	2.33	<0.001
75–84 (n = 2,874)	3.07	2.57	3.66	<0.001	2.16	1.85	2.52	<0.001	3.05	2.60	3.58	<0.001	1.74	1.49	2.02	<0.001
85– (n = 2,255)	2.48	2.08	2.96	<0.001	1.68	1.44	1.95	<0.001	2.36	2.01	2.77	<0.001	1.78	1.52	2.08	<0.001
Sex
Male (n = 2,900)	3.25	2.74	3.86	<0.001	2.05	1.76	2.38	<0.001	3.08	2.64	3.61	<0.001	1.93	1.66	2.24	<0.001
Female (n = 3,781)	2.66	2.29	3.10	<0.001	2	1.74	2.28	<0.001	2.64	2.3	3.03	<0.001	1.67	1.46	1.90	<0.001
Length of stay
–7 (n = 2,519)	3.03	2.53	3.64	<0.001	2.02	1.67	2.44	<0.001	3.08	2.59	3.66	<0.001	2.24	1.87	2.68	<0.001
7–14 (n = 2,026)	2.84	2.28	3.55	<0.001	2.23	1.86	2.67	<0.001	2.63	2.17	3.19	<0.001	1.64	1.37	1.97	<0.001
15–21 (n = 1,045)	2.28	1.74	2.98	<0.001	1.59	1.27	2	<0.001	2.38	1.87	3.03	<0.001	1.37	1.09	1.72	0.006
22– (n = 1,091)	3.31	2.50	4.38	<0.001	1.99	1.59	2.48	<0.001	3.03	2.36	3.88	<0.001	1.56	1.24	1.97	<0.001

[i] HR = hazard ratio; CI = confidence interval.

Hazard ratios were calculated using the Cox proportional hazards model.

Table 4

Subgroup analysis of interaction of each category (dummy variable) and the group with discharge services and each discharge service.

	DISCHARGE SERVICES				EACH DISCHARGE SERVICE
	DISCHARGE SERVICES				DISCHARGE PLANNING				COORDINATION WITH COMMUNITY CARE				REHABILITATION DISCHARGE INSTRUCTION
	HR	95% CI		p FOR INTERACTION	HR	95% CI		p FOR INTERACTION	HR	95% CI		p FOR INTERACTION	HR	95% CI		p FOR INTERACTION
Age, years
65–74 (n = 1,535)	1				1				1				1
75–84 (n = 2,874)	0.85	0.61	1.18	0.324	0.74	0.55	0.99	0.042	0.88	0.66	1.18	0.402	0.93	0.70	1.24	0.639
85– (n = 2,255)	0.68	0.49	0.93	0.018	0.56	0.42	0.75	<0.001	0.67	0.50	0.91	0.009	0.94	0.70	1.26	0.687
Sex
Male (n = 2,900)	1				1				1				1
Female (n = 3,781)	1.22	0.97	1.53	0.090	1.01	0.83	1.23	0.939	1.17	0.95	1.43	0.147	1.16	0.95	1.41	0.151
Length of stay
–7 (n = 2,519)	1				1				1				1
7–14 (n = 2,026)	0.92	0.69	1.22	0.559	1.04	0.81	1.34	0.778	0.84	0.65	1.08	0.177	0.72	0.56	0.92	0.009
15–21(n = 1,045)	0.73	0.53	1.01	0.058	0.73	0.55	0.98	0.037	0.76	0.56	1.02	0.066	0.58	0.44	0.78	<0.001
22– (n = 1,091)	1.08	0.77	1.51	0.653	0.92	0.69	1.23	0.593	0.98	0.72	1.32	0.871	0.69	0.51	0.92	0.011

[i] HR = hazard ratio; CI = confidence interval. p < 0.1 for interaction.

Hazard ratios were calculated using the Cox proportional hazards model.

Impact of Elderly Acute Care Discharge Services on Prevention of Rehospitalisation: A Retrospective Cohort Study Using National Health Data from Kita Ward, Tokyo

Figures & Tables

Figure 1

Table 1

Table 2

Figure 2

Table 3

Table 4

Paradigm

My account