Physical function, depressive symptoms, and quality of life with post-acute stroke care

Lai C.L.
Tsai M.M.
Luo J.Y.
Liao W.C.
Hsu P.S.
Chen H.Y.

Post-acute care for stroke—a retrospective cohort study in Taiwan.

,

Weng S.-C.
Hsu C.-Y.
Shen C.-C.
Huang J.-A.
Chen P.-L.
Lin S.-Y.

Combined functional assessment for predicting clinical outcomes in stroke patients after post-acute care: a retrospective multi-center cohort in central Taiwan.

,

Mahoney and Barthel, 1965

Wu W.J.
Chang I.C.
Lin C.T.
Wang S.C.
Ku Y.R.

The effects of enhanced physical therapy on post-acute stroke patients.

Google Scholar

). However, studies have rarely investigated the recovery trends of physical function, psychological conditions, and QOL before, during, and after the PAC rehabilitation program.

In Taiwan, nurses work with survivors of stroke, but there is very little evidence of their involvement in rehabilitation after stroke. Most stroke survivors with stable neurological conditions were discharged to another facility or home. The hospital-based PAC program is an updated clinical practice that differs from previous rehabilitation for stroke survivors in Taiwan. Thus, the study aimed to investigate the trajectories of physical function, depressive symptoms, QOL, and factors associated with QOL in patients who received a hospital-based PAC program over 12 weeks following stroke. Using this information, we can address the nurses’ role in nursing practice and empower them to fulfil their roles in stroke rehabilitation while improving the outcomes for stroke survivors (

Clarke, 2014

Clarke D.J.

Nursing practice in stroke rehabilitation: systematic review and meta-ethnography.

,

Meng et al., 2020

Meng X.
Chen X.
Liu Z.
Zhou L.

Nursing practice in stroke rehabilitation: perspectives from multi-disciplinary healthcare professionals.

).

2. Participants, ethics, and methods

2.1 Study design, setting, and participants

In this quasi-experimental study, patients in the neurology ward of a regional teaching hospital in central Taiwan were screened for eligibility to participate in the PAC–CVD program between July 2017 and January 2019. The inclusion criteria for patients were: (1) onset of acute ischaemic or haemorrhagic stroke confirmed by imaging or diagnosed by a physician within the past 8–30 days, (2) at least 20 years of age, (3) clear consciousness and ability to communicate, and (4) an mRS score of 2–4 points. The exclusion criteria for patients were: (1) diagnosis of cognitive decline or mental health conditions by a physician before the stroke, (2) aphasia (including motor aphasia) or illiteracy combined with severe hearing loss and inability to use verbal or nonverbal communication, (3) diagnosis of depression by a physician before the stroke, or (4) a post-acute stroke stage with prior rehabilitation treatments unrelated to stroke. The patients who opted for PAC and agreed to participate in the follow-up study constituted the PAC group, whereas those who did not wish to participate in the PAC program but agreed to participate in the follow-up study constituted the regular care group.

The sample size was computed with a prior power analysis using the G-power version 3.1.9.2 software (Heinrich-Heine-Universität Düsseldorf, Düsseldorf, Germany). With a power of 80%, alpha of 0.05, an effect size of 0.25, and four repeats, the required sample size was calculated as 48. Of the 62 eligible patients, one participated in the first investigation alone and withdrew. Overall, 61 patients were enroled, with 41 in the PAC group and 20 in the regular care group. The ages ranged from 20 to 91 years, with an average age of 67.59 ± 15.64 years, and the majority were males (36, 59.02%), married (51, 83.61%), had an elementary school education (21, 34.43%), and were unemployed before stroke onset (36, 59.02%) (Table 1).

Table 1Comparison of the homogeneity of patients in the PAC and regular care groups (n = 61).

Variables	All patients		PAC group (n = 41)		Regular care group (n = 20)		t/x²
Variables	n (%)	Mean (SD)	n (%)	Mean (SD)	n (%)	Mean (SD)
Age (years)		67.59 (15.64)		65.00 (15.07)		72.90 (15.80)	−1.89	0.06
NIHSS score at admission		8.00 (4.74)		8.51 (4.94)		6.95 (4.25)	1.21	0.23
Sex							1.48	0.22
Male	36 (59.02)		22 (53.66)		14 (70.00)
Female	25 (40.98)		19 (46.34)		6 (30.00)
Marital status								0.67 ^a Fisher’s accurate test.
Married	51 (83.61)		33 (80.49)		18 (90.00)
Single/divorced/widow	10 (16.39)		8 (19.51)		2 (10.10)
Education level								0.73 ^a Fisher’s accurate test.
Illiterate	21 (34.43)		14 (34.15)		7 (35.00)
Elementary school	21 (34.43)		11 (26.83)		10 (50.00)
Junior high school	6 (9.84)		5 (12.20)		1 (5.00)
High school or higher	13 (21.58)		9 (21.95)		2 (10.00)
Work status before stroke							3.14	0.08
Yes	25 (40.98)		20 (48.78)		5 (25.00)
No	36 (59.02)		21 (51.22)		15 (75.00)
Type of stroke								0.29 ^a Fisher’s accurate test.
Ischaemic	57 (93.44)		37 (90.24)		20 (100.00)
Haemorrhagic	4 (6.66)		4 (9.76)		0 (0.00)
Location of stroke								0.03 ^a Fisher’s accurate test. * p < 0.05.
Left brain	23 (37.70)		19 (46.34)		4 (20.00)
Right brain	33 (54.10)		17 (41.46)		16 (80.00)
Diencephalon/pontine	5 (8.20)		5 (8.20)		0 (0.00)
Number of strokes							0.24	0.62
First time	48 (78.69)		33 (80.49)		15 (75.00)
Second time (inclusive)	13 (21.31)		8 (19.51)		5 (15.00)
CCI (0–5)		1.34 (1.34)		1.15 (1.20)		1.75 (1.55)	−1.68	0.10
BI score at admission (0–85)		35.16 (22.02)		35.12 (22.09)		35.25 (22.45)	−0.02	0.98
mRS score at admission (2–4)		3.84 (0.42)		3.88 (0.33)		3.75 (0.55)	0.96	0.35
Length of stay in PAC (days)				39.27 (16.02)

Note. t = t test; x² = chi-square; PAC = post-acute care; CCI = Charlson comorbidity index; BI = Barthel Index; mRS = modified Rankin Scale; NIHSS = stroke severity (National Institutes of Health Stroke Scale; SD = standard deviation.

* p < 0.05.

a Fisher’s accurate test.

Open table in a new tab

2.2 Hospital-based post-acute care program vs. regular care

2.2.1 Hospital-based post-acute care program

In this study, patients in the PAC group with a stable medical condition and physical functional status between mild and moderately severe dysfunction (mRS score, 2–4) and within 8–30 days after the onset of acute stroke were provided 3–6 weeks of integrated rehabilitation, including physical and occupational therapies and language training. The NHIA covers costs of this hospital-based PAC program for up to 12 weeks, and a well-trained care team comprising a rehabilitation physician, physical therapist, occupational therapist, speech-language pathologist, and case manager is assigned to continuously evaluate the patient’s progress, extending the period, if necessary, for no more than 12 weeks. The PAC program was scheduled for twice a day on weekdays, depending on the patient’s ability; each treatment lasted 90–120 min. Each patient received physiotherapy (PT) or occupational therapy (OT) according to their disability status; speech-language therapy was arranged once a week according to the needs of the patients.

2.2.2 Regular care

Patients in the regular care group were those who did not wish to participate in the PAC program but agreed to participate in the follow-up. These patients received PT or OT once a day during hospitalisation or were discharged to home or another facility to receive rehabilitation program or further medical management.

2.3 Measurements

2.3.1 Demographics and disease information

Demographics data and medical information, such as sex, age, marital status, education level and work status, the number of strokes, stroke type, stroke location, stroke severity (National Institutes of Health Stroke Scale [NIHSS]), and comorbidities (Charlson comorbidity index [CCI]) were collected from reviewing their medical records.

Self-administered questionnaires were used to collect data on physical function, depressive symptoms, and QOL. For illiterate participants, a family member or caregiver provided the responses on the patient’s behalf while the nurse filled out the questionnaire. For Geriatric Depression Scale (GDS-15) and Stroke Impact Scale Version 3.0 (SIS V3.0), the nurse read the questions and filled out the answers on behalf of the patient.

2.3.2 Physical function

The BI, developed by

Mahoney F.I.
Barthel D.W.

Functional evaluation: the Barthel Index.

, measures the progress of inpatient rehabilitation, using different activities of daily living (ADL) related to self-care and mobility. The scale comprises 0–100 points, ranging from total dependency (0–20 points) to independency (100 points) in performing ADL. In this study, the overall Cronbach’s α value of the BI was 0.90.

The IADL, developed by

Lawton and Brody, 1969

Lawton M.P.
Brody E.M.

Assessment of older people: self-maintaining and instrumental activities of daily living.

, comprises a questionnaire to assess independent living skills and everyday functions in older adults. The total score ranges from 0 to 24 points for adult females and 0–15 points for adult males, the higher the score, the better the independence.

Kadar et al., 2018

Kadar M.
Ibrahim S.
Razaob N.A.
Chai S.C.
Harun D.

Validity and reliability of a Malay version of the Lawton Instrumental Activities of Daily Living Scale among the Malay speaking elderly in Malaysia.

showed that Cronbach’s α of the internal consistency of IADL was 0.838 and that the test–retest and inter-rater reliabilities were 0.950 and 0.957, respectively.

The mRS, originally developed in 1957 and modified in the 1980s, is used to assess the degree of disability in stroke survivors (

Uyttenboogaart et al., 2005

Uyttenboogaart M.
Stewart R.E.
Vroomen P.C.
De Keyser J.
Luijckx G.J.

Optimizing cutoff scores for the Barthel Index and the modified Rankin Scale for defining outcome in acute stroke trials.

). The scale comprises 0–6 points, running from no symptoms (0 points) to dead (6 points). The test–retest reliability κ_w is 0.81–0.95, the internal consistency κ is 0.56–0.78, and the construct validity is related to the type and size of stroke and neural defect (

Banks and Marotta, 2007

Banks J.L.
Marotta C.A.

Outcomes validity and reliability of the modified Rankin Scale: implications for stroke clinical trials: a literature review and synthesis.

); if structured interviews are used, it can reach 81% consistency, and the unweighted κ is 0.74 (

Wilson et al., 2005

Wilson J.T.
Hareendran A.
Hendry A.
Potter J.
Bone I.
Muir K.W.

Reliability of the modified Rankin Scale across multiple raters: benefits of a structured interview.

).

2.3.3 Depression symptoms

The original GDS-30, developed by

Yesavage et al., 1982

Yesavage J.A.
Brink T.L.
Rose T.L.
Lum O.
Huang V.
Adey M.
et al.

Development and validation of a geriatric depression screening scale: a preliminary report.

, was condensed to GDS-15 in 1986 (

Sheikh and Yesavage, 1986

Sheikh J.I.
Yesavage J.A.

Geriatric Depression Scale (GDS): recent evidence and development of a shorter version.

). The latter comprises 15 yes/no-type questions to measure the depressive symptoms among older patients related to three levels of content: emotion, cognition, and behaviour. A total score of 0–4 points indicates a normal state of mind, 5 points or more indicates clinical depression and a more comprehensive assessment with the doctor should be done, and greater than or equal to 10 indicates major depression. A study found that GDS-15 had a sensitivity of 84% and specificity of 89%, according to the Diagnostic and Statistical Manual (4th edition) diagnostic criteria (

Baillon et al., 2014

Baillon S.
Dennis M.
Lo N.
Lindesay J.

Screening for depression in Parkinson’s disease: the performance of two screening questions.

).

2.3.4 Quality of life

The SIS V3.0, developed by

Duncan et al., 1999

Duncan P.W.
Lai S.M.
van Culin V.
Huang L.
Clausen D.
Wallace D.

Development of a comprehensive assessment toolbox for stroke.

, measures disability and the QOL of patients with stroke, and the SIS V3.0 comprises a self-reporting questionnaire with 8 domains (domains 1–8) and 1 self-assessment question (domain 9). The total score of each domain ranges from 0 to 100 points; a higher score indicates better QOL.

Liang and Lin, 2016

Liang P.J.
Lin P.S.

Reliability and validity of the Stroke Impact Scale 3.0. In Taiwanese stroke patients.

demonstrated a good internal consistency (Cronbach’s α = 0.89–0.96), except for the emotional aspect (Cronbach’s α = 0.70) of the psychometric properties of the SIS V3.0 in assessing patients with stroke in Taiwan. In this study, the overall Cronbach’s α value of the SIS V3.0 was 0.93.

2.4 Procedure

This study was approved by the institutional review board (IRB No: 201700582B0) before data collection. The principal investigator or coinvestigator verified the medical records of potential subjects who met the inclusion criteria and explained to them the purpose and follow-up procedures of this study. Then, data on demographics and disease were collected. The subjects were followed up in the 1st, 4th, 8th, and 12th weeks post stroke. They completed five questionnaires, namely the BI, IADL, mRS, GDS-15, and SIS V3.0, at each visit.

2.5 Data analysis

The basic information of stroke survivors was analysed using descriptive statistics. In inferential statistical analysis, independent sample t-tests, chi-square tests, and Pearson’s correlation coefficient were used to test for differences in the basic attributes between the PAC and regular care groups and the interrelationship among outcome variables. Generalized estimating equations (GEEs) were used to analyse the changing trends and differences in physical function, depression symptoms, and QOL of these patients at week one, four, eight and twelve after stroke. Each GEE model included a main effect of the PAC and regular care groups, a main effect of time, and a two-way interaction effect of group by times, which indicates group differences concerning the changes from time 1 (week 1) to time 4 (week 12). Finally, multiple linear regression analysis was used to explore the predictors that affected the QOL of patients in the post-acute stage of stroke. Data analysis was performed using SPSS 22 (IBM SPSS, Armonk, NY: IBM Corp)(

IBM Corp, 2013

IBM Corp

IBM SPSS statistics for windows, version 22.0.

Google Scholar

).

3. Results

3.1 Demographics and disease information of stroke survivors

Table 1 summarises the details of the basic information of 61 patients. In terms of stroke type, 57 patients (93.44%) had an ischaemic stroke; the majority were first-time infarctions (48, 78.69%) located in the right brain (33, 54.10%), with an average NIHSS score of 8.00 ± 4.74 at admission and an average CCI of 1.34 ± 1.34.

3.2 Changing trends in physical function, depression symptoms, and quality of life of patients

Parameters of the regular care group at admission or in the 1st week after stroke were used as baseline values to evaluate the changing trend of patients’ physical function (BI, mRS, and IADL scores), depression symptoms (GDS-15 score), and QOL (SIS V3.0 score) in the 1st, 4th, 8th, and 12th week after stroke (Table 2).

Table 2Changing trends in physical function, depressive symptoms, and QOL of patients with stroke (n = 61).

Subject	PAC group (n = 41)					Regular care group (n = 20)
Subject	T0	T1	T2	T3	T4	T0	T1	T2	T3	T4
Physical function
BI	35.12	42.07 ^** p < 0.01.	49.63 ^*** p < 0.001.	65.73 ^*** p < 0.001.	73.78 ^*** p < 0.001.	35.25	37.50	46.70	48.50	50.25
mRS	3.88	3.76 ^** p < 0.01.	3.59 ^*** p < 0.001.	2.93 ^*** p < 0.001.	2.51 ^*** p < 0.001.	3.75	3.65	3.55	3.50	3.55
IADL	–	3.46 ^** p < 0.01.	4.20 ^*** p < 0.001.	5.80 ^*** p < 0.001.	7.54 ^*** p < 0.001.	–	3.25	3.55	3.95	3.90
GDS-15		7.76 ^** p < 0.01.	6.46 ^*** p < 0.001.	4.66 ^*** p < 0.001.	3.41 ^*** p < 0.001.		9.30	8.15	8.80	8.55
SIS V3.0 domain 1–8		367.73 ^** p < 0.01.	447.38 ^*** p < 0.001.	539.69 ^*** p < 0.001.	577.68 ^*** p < 0.001.		374.32	406.57	411.15	412.16
SIS V3.0 domain 9		18.78 ^** p < 0.01.	32.20 ^*** p < 0.001.	55.37	63.41 ^*** p < 0.001.		19.00	26.00	28.50	30.50

Note. T0 = admission; T1 = 1st week after stroke; T2 = 4th week after stroke; T3 = 8th week after stroke; T4 = 12th week after stroke. PAC = post-acute care; BI = Barthel Index; GDS = Geriatric Depression Scale; IADL = Instrumental Activities of Daily Living; mRS = modified Rankin Scale; SIS V3.0 = Stroke Impact Scale Version 3.0.

** p < 0.01.

*** p < 0.001.

Open table in a new tab

3.2.1 Physical function

The BI scores of the PAC group in the 1st, 4th, 8th, and 12th weeks after stroke were 6.95 (p = 0.008), 14.51 (p < 0.001), 30.61 (p < 0.001), and 38.66 (p < 0.001) points higher, respectively, than those at the time of admission (35.12). In contrast, the BI scores of the regular care group in the 1st, 4th, 8th, and 12th weeks after stroke displayed a gradual upward trend over time; the difference was not significant (p > 0.05). The mRS scores of the PAC group in the 1st, 4th, 8th, and 12th weeks after stroke reduced by 0.12 (p = 0.048), 0.29 (p = 0.005), 0.95 (p < 0.001), and 1.37 (p < 0.001) points, respectively. The mRS scores of the regular care group in the 1st, 4th, 8th, and 12th weeks after stroke gradually decreased over time, there was no significant difference (p > 0.05). The IADL scores of the PAC group in the 4th, 8th, and 12th weeks after stroke were 0.73 (p = 0.006), 2.34 (p < 0.001), and 4.07 (p < 0.001) points higher, respectively, than that in the 1st week. In contrast, the IADL scores of the regular care group in the 4th, 8th, and 12th weeks after stroke displayed a gradual upward trend over time, there was no significant difference (p > 0.05).

3.2.2 Depression symptoms

For those who scored 5 or more points in the PAC group, case numbers gradually dropped as follows: 31 (75.6%), 29 (70.7%), 17 (41.5%), and 15 (36.6%) patients after the 1st, 4th, 8th, and 12th week after stroke, respectively. The mean GDS-15 scores of the PAC group were 7.76, 6.46, 4.66, and 3.41 after the 1st, 4th, 8th, and 12th week after stroke and were reduced by 1.29 points (p < 0.001), 3.10 points (p < 0.001), and 4.34 points (p < 0.001), respectively, compared with those in the 1st week. In contrast, a total of 5 points or more in the regular care group was scored by 16 (80.0%), 16 (80.0%), 17 (85.0%), and 16 (80.0%) patients after the 1st, 4th, 8th, and 12th week after stroke, respectively. The mean GDS scores of the regular care group were 9.30, 8.15, 8.80, and 8.55 in the 1st, 4th, 8th, and 12th week after stroke, respectively, and they plateaued over time, there was no significant difference (p > 0.05).

3.2.3 Quality of life

The SIS V3.0 scores of domains 1–8 of the PAC group in the 4th, 8th, and 12th week after stroke were 79.65 (p < 0.001), 171.96 (p < 0.001), and 209.94 (p < 0.001) points higher, respectively, than those in the 1st week. Similarly, their scores for domain 9 in the 4th, 8th, and 12th week after stroke were 13.42 (p < 0.001), 36.59 (p < 0.001), and 44.63 (p < 0.001) points higher, respectively, than those in the 1st week. The SIS V3.0 scores of domains 1–8 and domain 9 of the regular care group in the 4th, 8th, and 12th week after stroke displayed a gradual upward trend over time, there was no significant difference (p > 0.05).

3.2.4 Differences in physical function, depression symptoms, and quality of life of patients

Table 3 summarises the BI, mRS, IADL, GDS-15, and self-evaluated recovery (domain 9 of SIS V3.0) scores of the two patient groups. These data show that these parameters changed significantly, starting from the 8th week after stroke (p < 0.05). The QOL of the two patient groups differed significantly, starting from the 4th week after stroke (p < 0.01). Therefore, compared with stroke survivors receiving RC, those receiving PAC showed a significant improvement in physical function, depressive symptoms, and QOL from the 8th week after stroke.

Table 3Analysis of the effectiveness of PAC for improving the physical function, depressive symptoms, and QOL of stroke survivors.

Predictive variables	B	SE	Wald χ²	p-value
Group × timepoint (BI)
PACG × T1	4.70	3.43	1.88	0.170
PACG × T2	3.06	5.45	0.32	0.574
PACG × T3	17.36	6.84	6.45	0.011
PACG × T4	23.66	7.28	10.55	0.001
Group × timepoint (mRS)
PACG × T1	−0.02	0.09	0.06	0.810
PACG × T2	−0.09	0.24	0.15	0.703
PACG × T3	−0.70	0.25	7.65	0.006
PACG × T4	−1.17	0.26	19.40	<0.001
Group × timepoint (IADL)
PACG × T2	0.43	0.32	1.83	0.176
PACG × T3	1.64	0.78	4.47	0.035
PACG × T4	3.42	0.87	15.59	<0.001
Group × timepoint (GDS)
PACG × T2	−0.14	0.64	0.05	0.825
PACG × T3	−2.60	0.86	9.22	0.002
PACG × T4	−3.59	0.97	13.57	<0.001
Group × timepoint (SIS 1–8)
PACG × T2	47.41	14.44	10.78	0.001
PACG × T3	135.13	25.60	27.87	<0.001
PACG × T4	172.11	29.32	34.47	<0.001
Group × timepoint (SIS 9)
PACG × T2	6.42	4.69	1.87	0.171
PACG × T3	27.09	6.13	19.50	<0.001
PACG × T4	33.13	6.07	29.79	<0.001

Notes. The timepoint of the reference group for BI and mRS was the time of admission. The timepoint of the other groups was the 1st week after stroke. BI = Barthel Index; RG = regular care group; PAC = post-acute care; PACG = PAC group; GDS = Geriatric Depression Scale; IADL = Instrumental Activities of Daily Living; mRS = modified Rankin Scale; SE = standard error; SIS V3.0 = Stroke Impact Scale Version 3.0.

Open table in a new tab

3.3 Associated factors affecting the quality of life of patients with stroke

After the variables of demographic characteristics were converted to categorical variables, physical function (measured by BI and IADL) and depressive symptoms (measured by GDS-15) were regressed against the QOL at the four timepoints. The regression analyses revealed that the potential factors associated with QOL in the 4th, 8th, and 12th week after stroke (p < 0.05) included age, NIHSS score, CCI, marital status, education level, work status, physical function, and depressive symptoms.

The QOL in the 4th week reached statistical significance with F_{7, 53} = 39.48, p < 0.001, and the explanatory power of the regression model reached 0.839 (adjusted R² = 0.818). After adjusting for other covariates, higher BI score (B = 1.71, p < 0.001), higher IADL score (B = 14.06, p < 0.001), and lower GDS-15 score (B = −11.94, p < 0.001) were associated with better QOL in the 4th week in stroke survivors. The QOL in the 8th week reached statistical significance with F_{9, 51} = 45.05 (p < 0.001), and the explanatory power of the regression model reached 0.888 (adjusted R² = 0.869). After adjusting for other covariates, higher CCI score (B = 12.71, p = 0.042), higher BI score (B = 1.37, p = 0.026), higher IADL score (B = 7.36, p = 0.008), and lower GDS-15 score (B = −15.49, p < 0.001) were associated with better QOL in the 8th week post stroke. The QOL in the 12th week reached statistical significance with F_{9, 51} = 107.37, p < 0.001, and the explanatory power of the regression model reached 0.950 (adjusted R² = 0.941). After adjusting for other covariates, higher BI score (B = 2.06, p < 0.001), lower mRS score (B = −26.53, p = 0.018), higher IADL score (B = 6.80, p < 0.001), and lower GDS-15 score (B = −12.22, p < 0.001) were associated with better QOL in the 12th week post stroke (Table 4).

Table 4Predictors of QOL in the 4th_, 8th, and 12th weeks in stroke survivors.

Independent variables	T2				T3				T4
	B	SE	t value	p-value	B	SE	t value	p-value	B	SE	t value	p-value
Intercept	283.46	76.18	3.72	<0.001	467.62	101.18	4.62	<0.001	426.54	67.44	6.33	<0.001
Education level: junior high school or above ^a Education level: Below elementary school (reference).	−0.16	21.87	−0.01	0.994	10.47	24.86	0.42	0.675	5.97	18.37	0.33	0.746
Age	0.83	0.65	1.28	0.205	0.94	0.89	1.05	0.297	0.64	0.62	1.02	0.312
NIHSS score	−2.17	1.59	−1.36	0.179
Marital status: No ^b Marital status: yes (reference).					28.89	22.29	1.30	0.201	9.73	16.05	0.61	0.547
Work status: Yes ^c Work status: no (reference).					−3.83	18.82	−0.20	0.839	3.68	13.58	0.27	0.787
CCI					12.71	6.10	2.08	0.042	5.75	4.13	1.39	0.169
BI	1.71	0.45	3.77	<0.001	1.37	0.60	2.30	0.026	2.06	0.40	5.17	<0.001
MRS	15.75	9.39	1.68	0.099	−27.05	15.76	−1.72	0.092	−26.53	10.89	−2.44	0.018
IADL	14.06	3.02	4.65	<0.001	7.36	2.65	2.78	0.008	6.80	1.70	4.01	<0.001
GDS-15	−11.94	1.92	−6.21	<0.001	−15.49	2.38	−6.50	<0.001	−12.22	1.55	−7.87	<0.001
R²	0.839				0.888				0.950
Adjusted R²	0.818				0.869				0.941
F	39.48				45.05				107.37
p-value	<0.001				<0.001				<0.001

T2 = 4th week after stroke; T3 = 8th week after stroke; T4 = 12th week after stroke.

BI = Barthel Index; GDS = Geriatric Depression Scale; IADL = Instrumental Activities of Daily Living; mRS = modified Rankin Scale; SE = standard error; SIS V3.0 = Stroke Impact Scale Version 3.0.

a Education level: Below elementary school (reference).

b Marital status: yes (reference).

c Work status: no (reference).

Open table in a new tab

4. Discussion

4.1 Demographics and disease information of stroke survivors

In this study, the mean age of the 61 patients was 67.59 (±15.64) years, similar to that in studies by

Chien S.H.
Sung P.Y.
Liao W.L.
Tsai S.W.

A functional recovery profile for patients with stroke following post-acute rehabilitation care in Taiwan.

(66.01±15.36 years),

Huang H.C.
Tsai J.Y.
Liu T.C.
Sheng W.Y.
Lin T.C.
Lin C.W.
et al.

Functional recovery of stroke patients with postacute care: a retrospective study in a northern medical center.

(two groups—68.22±14.52 years vs. 68.51±11.22 years), and

Weng S.-C.
Hsu C.-Y.
Shen C.-C.
Huang J.-A.
Chen P.-L.
Lin S.-Y.

Combined functional assessment for predicting clinical outcomes in stroke patients after post-acute care: a retrospective multi-center cohort in central Taiwan.

(67.20 ± 14.70 years). In contrast, the average age of participants in the study by

Lai C.L.
Tsai M.M.
Luo J.Y.
Liao W.C.
Hsu P.S.
Chen H.Y.

Post-acute care for stroke—a retrospective cohort study in Taiwan.

was younger at 63.01 (±13.24) years. The average length of the PAC program in the current study was 39.27 (±16.02) days, shorter than those in other studies: 43.57 (±26.28) days and 58.15 (±16.93) days (

Chien S.H.
Sung P.Y.
Liao W.L.
Tsai S.W.

A functional recovery profile for patients with stroke following post-acute rehabilitation care in Taiwan.

,

Lai C.L.
Tsai M.M.
Luo J.Y.
Liao W.C.
Hsu P.S.
Chen H.Y.

Post-acute care for stroke—a retrospective cohort study in Taiwan.

).

4.2 Changing trends in physical function, depression symptoms, and quality of life

4.2.1 Physical function

This study revealed that the BI, mRS, and IADL scores of patients receiving PAC improved gradually over time, displaying significant differences at the 8th and 12th week after stroke, consistent with the results of

Wu W.J.
Chang I.C.
Lin C.T.
Wang S.C.
Ku Y.R.

The effects of enhanced physical therapy on post-acute stroke patients.

Google Scholar

,

Huang H.C.
Tsai J.Y.
Liu T.C.
Sheng W.Y.
Lin T.C.
Lin C.W.
et al.

Functional recovery of stroke patients with postacute care: a retrospective study in a northern medical center.

,

Lai C.L.
Tsai M.M.
Luo J.Y.
Liao W.C.
Hsu P.S.
Chen H.Y.

Post-acute care for stroke—a retrospective cohort study in Taiwan.

,

Chien S.H.
Sung P.Y.
Liao W.L.
Tsai S.W.

A functional recovery profile for patients with stroke following post-acute rehabilitation care in Taiwan.

, and

Weng S.-C.
Hsu C.-Y.
Shen C.-C.
Huang J.-A.
Chen P.-L.
Lin S.-Y.

Combined functional assessment for predicting clinical outcomes in stroke patients after post-acute care: a retrospective multi-center cohort in central Taiwan.

. This study also provided evidence that the PAC program enhances a stroke survivor’s functional recovery and may increase independence in their ADL (

Hsieh et al., 2018

Hsieh C.Y.
Tsao W.C.
Lin R.T.
Chao A.C.

Three years of the nationwide post-acute stroke care program in Taiwan.

,

Huang H.C.
Tsai J.Y.
Liu T.C.
Sheng W.Y.
Lin T.C.
Lin C.W.
et al.

Functional recovery of stroke patients with postacute care: a retrospective study in a northern medical center.

,

Peng et al., 2017

Peng L.N.
Lu W.H.
Liang C.K.
Chou M.Y.
Chung C.P.
Tsai S.L.
et al.

Functional outcomes, subsequent healthcare utilization, and mortality of stroke postacute care patients in Taiwan: a nationwide propensity score-matched study.

,

Weng S.-C.
Hsu C.-Y.
Shen C.-C.
Huang J.-A.
Chen P.-L.
Lin S.-Y.

Combined functional assessment for predicting clinical outcomes in stroke patients after post-acute care: a retrospective multi-center cohort in central Taiwan.

). Most previous studies measured two timepoints—before initiation and after completion of PAC—to assess the changes in the BI, mRS, and IADL scores and, thus, could not identify the critical changing timepoint (

Chien S.H.
Sung P.Y.
Liao W.L.
Tsai S.W.

A functional recovery profile for patients with stroke following post-acute rehabilitation care in Taiwan.

,

Hsien et al., 2015

Hsien H.H.
Liou W.C.
Yang Y.H.

Post-acute care improves daily activity functions of stroke patients: an effectiveness analysis.

;

Hsieh et al., 2018

Hsieh C.Y.
Tsao W.C.
Lin R.T.
Chao A.C.

Three years of the nationwide post-acute stroke care program in Taiwan.

;

Huang H.C.
Tsai J.Y.
Liu T.C.
Sheng W.Y.
Lin T.C.
Lin C.W.
et al.

Functional recovery of stroke patients with postacute care: a retrospective study in a northern medical center.

,

Lai C.L.
Tsai M.M.
Luo J.Y.
Liao W.C.
Hsu P.S.
Chen H.Y.

Post-acute care for stroke—a retrospective cohort study in Taiwan.

;

Peng et al., 2017

Peng L.N.
Lu W.H.
Liang C.K.
Chou M.Y.
Chung C.P.
Tsai S.L.
et al.

Functional outcomes, subsequent healthcare utilization, and mortality of stroke postacute care patients in Taiwan: a nationwide propensity score-matched study.

;

Weng S.-C.
Hsu C.-Y.
Shen C.-C.
Huang J.-A.
Chen P.-L.
Lin S.-Y.

Combined functional assessment for predicting clinical outcomes in stroke patients after post-acute care: a retrospective multi-center cohort in central Taiwan.

;

Wu W.J.
Chang I.C.
Lin C.T.
Wang S.C.
Ku Y.R.

The effects of enhanced physical therapy on post-acute stroke patients.

Google Scholar

). The study tracked these changes at four timepoints—1st, 4th, 8th, and 12th week after stroke after adjusting covariates to identify the changing trends in physical function recovery among stroke survivors.

4.2.2 Depressive symptoms

Although 36.6% of stroke survivors receiving PAC in this study reported depressive symptoms, the proportion decreased gradually but significantly over time. The result was similar to that of

Wang et al., 2018

Wang S.
Wang C.X.
Zhang N.
Xiang Y.T.
Yang Y.
Shi Y.Z.
et al.

The association between post-stroke depression, aphasia, and physical independence in stroke patients at 3-month follow-up.

and

Dong et al., 2021

Dong L.
Williams L.S.
Brown D.L.
Case E.
Morgenstern L.B.
Lisabeth L.D.

Prevalence and course of depression during the first year after mild to moderate stroke.

, who found that 34.7% and 35.3% of patients exhibited depressive symptoms three months after stroke. Moreover, the result was consistent with the American Heart Association/American Stroke Association’s statement that PSD is associated with poorer functional outcomes after stroke (

Towfighi et al., 2017

Towfighi A.
Ovbiagele B.
El Husseini N.
Hackett M.L.
Jorge R.E.
Kissela B.M.
et al.

Poststroke depression: a scientific statement for healthcare professionals from the American Heart Association/American Stroke Association.

). The physical function in the PAC group improves and becomes more independent after receiving PAC, which in turn may potentially positively affect their psychological condition and may potentially reduce depressive symptoms. In addition, the study enabled us to simultaneously identify changing trends in depressive symptoms as well as the functional recovery after stroke.

4.2.3 Quality of life

The QOL of stroke survivors receiving PAC and their stroke recovery increased significantly, but gradually at the 4th, 8th, and 12th week after stroke. These results were similar to those of

Lai C.L.
Tsai M.M.
Luo J.Y.
Liao W.C.
Hsu P.S.
Chen H.Y.

Post-acute care for stroke—a retrospective cohort study in Taiwan.

, who showed significant improvement in EQ-5D (mobility, self-care, usual activities, and anxiety/depression) among 168 patients after stroke with an average of 43.57 days in the PAC program. However, the results differed from those of

Schindel et al., 2021

Schindel D.
Schneider A.
Grittner U.
Jöbges M.
Schenk L.

Quality of life after stroke rehabilitation discharge: a 12-month longitudinal study.

, who followed patients after discharge from the rehabilitation program and found that the majority had neither maintained nor regained their initial QOL at three, six and twelve months after rehabilitation discharge.

4.3 Associated factors affecting the quality of life of stroke survivors

This study showed that physical function and depressive symptoms are significant factors associated with QOL in the 4th, 8th, and 12th weeks after stroke. The results suggest that patients with stroke who can perform ADL independently and have fewer depressive symptoms have a better QOL at three months after stroke compared with those who are dependent on others for ADL and exhibit depressive symptoms. The results were similar to those of

Hsien et al., 2015

Hsien H.H.
Liou W.C.
Yang Y.H.

Post-acute care improves daily activity functions of stroke patients: an effectiveness analysis.

,

Lai C.L.
Tsai M.M.
Luo J.Y.
Liao W.C.
Hsu P.S.
Chen H.Y.

Post-acute care for stroke—a retrospective cohort study in Taiwan.

,

Weng S.-C.
Hsu C.-Y.
Shen C.-C.
Huang J.-A.
Chen P.-L.
Lin S.-Y.

Combined functional assessment for predicting clinical outcomes in stroke patients after post-acute care: a retrospective multi-center cohort in central Taiwan.

, and

Heikinheimo and Chimbayo, 2015

Wu W.J.
Chang I.C.
Lin C.T.
Wang S.C.
Ku Y.R.

The effects of enhanced physical therapy on post-acute stroke patients.

Google Scholar

, demonstrating that the tendency toward depression and dependency on others for daily activities affected the QOL of stroke survivors.

Heikinheimo T.
Chimbayo D.

Quality of life after first-ever stroke: an interview-based study from Blantyre, Malawi.

and