Association between performance of basic movements at admission and socioenvironmental factors and returning home following inpatient stroke rehabilitation

Abstract

[Purpose] The aim of this study was to investigate the association between the ability to perform basic movements at admission and returning home from convalescent rehabilitation in patients with subacute stroke after adjusting for socioenvironmental factors. [Participants and Methods] This retrospective cohort study used data from the medical records. The primary outcome was returning home, and the associated factors were basic movement ability and socioenvironmental factors. Basic movement ability was assessed using the revised version of the Ability for Basic Movement Scale. Socioenvironmental factors included pre-stroke cohabitation status, support at home, and marital status. [Results] Of the 480 participants included in the analysis, 380 had returned home. The revised version of the Ability for Basic Movement Scale scores were found to be significantly associated with returning home, even after adjusting for socioenvironmental factors. Among the various influencing socioenvironmental factors, only pre-stroke cohabitation status remained significant after adjustment. [Conclusion] Even after adjusting for socioenvironmental factors, basic movement abilities were found to be associated with returning home. Assessing the revised version of the Ability for Basic Movement Scale scores at admission and pre-stroke cohabitation status may be useful for reasonable hospitalization planning to return home after subacute stroke.

INTRODUCTION

Approximately 50% of stroke survivors are chronically disabled^{1, 2)}. Admission to long-term care is common after stroke, with up to 39% of stroke survivors being discharged to long-term care³⁾. In convalescent rehabilitation wards in Japan, even with intensive rehabilitation, 20–40% of stroke patients are unable to return home from the hospital^4,5,6). Therefore, it is necessary to know what types of patients are likely to be discharged to care facilities to ensure reasonable hospitalization planning to return home.

Factors associated with returning home reported in previous studies can be divided into physical functioning and socio-environmental factors^{7, 8)}. Physical function factors included body function, motor activity, and activities of daily living (ADL) independence^{6, 7)}. Motor activities, such as basic movements, form the basis of ADL independence, while similar discrimination has been reported as a factor associated with return home⁶⁾. Socio-environmental factors included pre-stroke cohabitation status, support at home, and marital status^8,9,10). While a greater understanding of physical and socio-environmental factors is needed, to the best of our knowledge, the basic movement ability and socio-environmental factors associated with returning home in individuals with stroke have not yet been examined using a multivariable analysis.

Therefore, the present study aimed to investigate whether the ability for basic movements at admission of subacute stroke patients is associated with returning home, even after adjusting for socio-environmental factors. Among these, we hypothesized that basic movement ability is more strongly related to returning home than environmental factors.

PARTICIPANTS AND METHODS

This study had a retrospective cohort design. The study protocol was approved by the Ethics Committees of Shinshu University and Toyama Prefectural Rehabilitation Hospital and Support Center for Children with Disabilities (Nos. 5557 and 86). The need for informed consent was waived due to the retrospective study design. However, we published an opt-out notice on our institution’s website and bulletin boards, allowing potential participants to refuse participation. This study involved the analysis of medical records of patients admitted to the rehabilitation facility between April 2021 and March 2023. The study was conducted in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement. The STROBE checklist is provided as a supplementary file.

The inclusion criteria for the participants were: age >18 years, a diagnosis of cerebral infarction or intracerebral hemorrhage, and living at home prior to stroke onset. The exclusion criteria were: transfer to an acute care facility, death during hospitalization, and missing data on outcomes and associated factors.

The primary outcome was returning home, defined as the place of residence before stroke onset. The associated factors included: the ability to perform basic movement upon admission to the convalescent rehabilitation ward and socio-environmental factors. Basic movement ability was evaluated using the revised version of the Ability for Basic Movement Scale (ABMS II). The ABMS II comprises five items: turning over, sitting up, remaining sitting, standing up, and remaining standing, assigning a score of 1 to 6 for a maximum of 30 points, with higher scores indicating better functionality¹¹⁾. Socio-environmental factors included pre-stroke cohabitation status, support at home, and marital status, which were confirmed to be significantly associated with returning home in a recent meta-analysis⁸⁾. Pre-stroke cohabitation status was defined as living with others (spouse, family, or friends), or living alone⁸⁾. We defined support at home as a first-degree relative living in the same prefecture as the patient or not⁸⁾. Marital status was defined as being married (excluding bereavement and divorce) or not⁸⁾.

The participants were divided into two groups, the home and facility groups, based on whether they returned home from the hospital. Demographic information and medical information at admission were compared between the two groups using the χ² test for categorical variables, and the Mann–Whitney U test for continuous variables. We used univariable logistic regression analysis to confirm the association between the possibility of returning home and four variables: ABMS II score at admission, pre-stroke cohabitation status, support at home, and marital status. Furthermore, we used multivariable logistic regression analysis to verify whether the ABMS II was associated with returning home, even after including socio-environmental factors in the analysis. To ensure the robustness of the results, we performed sensitivity analyses for four subgroups including (1) participants with high pre-stroke ADL (modified Rankin scale [mRS] score ≤2), (2) those with low pre-stroke ADL (mRS score >2), (3) those without previous stroke, and (4) those with previous stroke. We used the R package “pmsampsize” to calculate the sample size¹²⁾. We set the number of factors to be four (ABMS II score, pre-stroke cohabitation status, support at home, and marital status)^{6, 8, 11)}, the incidence rate of the outcome to 0.2^{4, 6)}, the C-statistic to 0.827⁶⁾, and the minimum required sample size to 246.

RESULTS

In total, 569 individuals with stroke and 89 were excluded from the study. The reasons for exclusion were as follows: 44 patients had subarachnoid hemorrhage, 41 were transferred to an acute care hospital or died, 2 were residents of the facility before stroke onset, and 2 had missing data. Finally, 480 participants were included in the analysis, of whom 380 (79.2%) were able to return home. Table 1 lists the characteristics of the participants. Compared with the facility group, the home care group had the following characteristics: younger age, higher BMI, independence in activities of daily living before onset, more frequent first onset, shorter time from onset to admission to a rehabilitation hospital, higher ABMS II and Functional Independence Measure (FIM) scores at admission and discharge, and longer hospital stay.

Table 1. Comparisons of participants’ demographic and medical characteristics.

Characteristics	Total (n=480)	Home group (n=380)	Facilities group (n=100)	p-value
Age, years	70.0 ± 13.0	68.6 ± 13.1	75.3 ± 11.0	***
Sex, female	181 (37.8)	141 (37.2)	40 (40.0)
BMI, kg/m2	22.5 ± 4.0	23.0 ± 4.0	20.7 ± 3.1	***
Pre-stroke cohabitation status, living with others	385 (80.2)	320 (84.2)	65 (65.0)	***
Support at home, available	438 (91.3)	357 (93.9)	81 (81.0)	***
Marital status, married	307 (64.0)	260 (68.4)	47 (47.0)	***
Pre-stroke ADL, mRS ≤2	447 (93.1)	359 (94.5)	88 (88.0)	*
Stroke type, ICH	155 (32.3)	125 (32.9)	30 (30.0)
Previous stroke, presence	100 (20.8)	69 (18.2)	31 (31.0)	**
Admission from stroke onset, days	33.6 ± 24.5	31.2 ± 21.7	43.0 ± 31.4	***
Total score of the ABMSII at admission, points	24.4 ± 6.8	26.1 ± 5.6	17.9 ± 6.9	***
Motor FIM at admission, points	50.5 ± 25.2	56.8 ± 23.2	26.7 ± 17.2	***
Cognitive FIM at admission, points	23.2 ± 8.1	25.1 ± 7.2	15.8 ± 7.0	***
Total FIM at admission, points	73.7 ± 31.5	81.9 ± 28.4	42.5 ± 21.2	***
Total score of the ABMSII at discharge, points	26.8 ± 5.4	28.3 ± 3.6	21.2 ± 7.1	***
Motor FIM at discharge, points	67.7 ± 24.2	75.5 ± 18.0	37.8 ± 21.6	***
Cognitive FIM at discharge, points	26.4 ± 8.1	28.7 ± 6.6	17.6 ± 7.2	***
Total FIM at discharge points	94.1 ± 30.7	104.2 ± 22.9	55.5 ± 26.0	***
Length of hospital stay, days	78.5 ± 37.7	71.2 ± 34.5	106.4 ± 36.2	***

*p<0.05, **p<0.01, ***p<0.001.

Values are presented as mean ± standard deviation or number (%). P-values indicate significant differences between the groups.

BMI: body mass index; mRS: modified Rankin Scale; ICH: intracerebral hemorrhage; ABMSII: revised version of the Ability for Basic Movement Scale; FIM: functional independence measure.

Table 2 presents the results of the univariable and multivariable logistic regression analyses. Univariable logistic regression analysis of the association between the four variables and returning home revealed a statistically significant difference. The odds ratio (OR) and 95% confidence interval (95% CI) were 1.19 (1.15–1.23) for ABMS II, 2.87 (1.74–4.70) for pre-stroke cohabitation status, 3.64 (1.88–7.00) for available support at home, and 2.44 (1.56–3.84) for marital status.

Table 2. Logistic regression analysis to determine the association with returning home.

Predictors	Univariate analysis	Multivariate analysis
	Odds ratios (95% CI)	Odds ratios (95% CI)	Regression coefficients
Intercept			−4.514
ABMSII	1.19 (1.15–1.23)***	1.21 (1.16–1.26)***	0.190
Pre-stroke cohabitation status	2.87 (1.74–4.70)***	2.45 (1.07–5.64)*	0.896
Support at home	3.64 (1.88–7.00)***	1.91 (0.75–4.84)	0.648
Marital status	2.44 (1.56–3.84)***	1.88 (0.96–3.69)	0.632

*p<0.05, ***p<0.001.

ABMSII: revised version of the Ability for Basic Movement Scale.

Furthermore, we performed multivariable logistic regression analysis to examine whether ABMS II was associated with returning home after adjusting for the three socio-environmental factors, confirming a significant association with an OR of 1.21 (1.16–1.26). Of the three socio-environmental factors, only pre-stroke cohabitation status was significantly associated with returning home, with an OR of 2.45 (1.07–5.64). The results of the sensitivity analyses were similar for all participants (Supplementary Tables 1–4). Across the two subgroup analyses, ABMS II was significantly associated with returning home, after adjusting for three socioenvironmental factors.

DISCUSSION

This is the first study to confirm the association between returning home using multivariable analysis assessing the ability for basic movement on admission and socio-environmental factors. The present study suggests that ABMS II scores are associated with returning home, even after adjusting for three socio-environmental factors: pre-stroke cohabitation status, support at home, and marital status in patients with subacute stroke. The main findings remained consistent across the four subgroup sensitivity analyses, supporting the robustness of the results of multivariable analysis.

Of the participants in this study, 100 (20.8%) were discharged from the facility, which is similar to the rates of 20–40% reported in previous studies of 20–40%^4,5,6). The ABMS II score at admission was 26.1 ± 5.6 points for the home group and 17.9 ± 6.9 points for the facility group, similar to the scores of 23.75 ± 5.83 points and 16.27 ± 5.13 points reported by Yang et al⁶⁾. In the home group, younger age, BMI, and total FIM score were higher, while admission from stroke onset and length of stay were shorter, which is also consistent with prior reports^4,5,6). Therefore, we believe that the participants in this study did not deviate from those representative of convalescent rehabilitation wards in Japan.

A previous study reported that ABMS II scores were associated with returning home, with an odds ratio of 1.241 (1.132–1.360)⁶⁾. The odds ratio obtained in this study was 1.19 (1.15–1.23), showing consistent association with a previous study. In addition, ABMS II was associated with an odds ratio of 1.21 (1.16–1.26), even after adjusting for socio-environmental factors, supporting the importance of the ability for basic movement as an association with returning home. The ABMS II score at admission associated with returning home reflects basic movement ability and may be associated with the FIM score at discharge. In this study, the ABMS II score at discharge, motor FIM score, cognitive FIM score, and total FIM score were significantly higher in the home group. Motor FIM scores of 61 or lower and 79 or lower indicate moderate and severe disability, respectively¹³⁾. The average motor and total FIM scores at discharge in the facility group were 37.8 and 55.5, respectively, indicating that many patients required assistance in their ADLs. Previous studies have reported that basic movement ability on admission predicts ADL independence at discharge^{6, 14)}. The ABMS II further includes an assessment of trunk control, such as sitting and standing¹¹⁾, while trunk function is one of the most important factors in performing ADL independently after stroke rehabilitation^{15, 16)}. Higher FIM motor scores are more strongly associated with sitting balance than with other FIM sub-items¹⁶⁾. These findings support the association between the higher ABMS II scores at admission and home discharge observed in the present study.

The main purposes of convalescent rehabilitation wards are not only to facilitate home discharge but also to avoid prolonged bed rest, reduce the risk of complications, and improve quality of life. Therefore, even for patients with poor basic movement ability upon admission, convalescent rehabilitation hospitals should provide intensive and comprehensive rehabilitation services that are focused on maximizing functional recovery, preventing complications during the recovery phase, and training patients and caregivers in adaptation strategies.

A meta-analysis reported that pre-stroke cohabitation status was associated with return home, with 2.60 (1.84–3.68), available support at home with OR 11.48 (95% CI 6.52–20.21), and marital status with 2.05 (1.80–2.33)⁸⁾. The odds ratios obtained in this study were 2.87 (1.74–4.70), 3.64 (1.88–7.00), and 2.44 (1.56–3.84) respectively. The large difference in the odds ratio for support at home may be due to differences in the severity of participants’ conditions and their social backgrounds. Pereira et al. investigated the discharge destinations in patients with severe stroke, and only one individual returned home in the absence of support at home⁹⁾. Because the present study included stroke survivors of all severities, the impact of home support may have been small. Moreover, Japan’s insurance system is unique; individuals certified as requiring long-term care can utilize facilities and in-home and community-based services¹⁷⁾. Consequently, using these services without the support of others may have led to returning home, which may help to explain the low odds ratio. It is considered that these socio-environmental factors overlap and are interrelated, while the results of the multivariable analysis showed that only pre-stroke cohabitation status had a statistically significant association. Furthermore, the results of the sensitivity analysis suggested that the socio-environmental factors that are important for returning home may differ depending on the degree of independence in ADL before stroke and the history of stroke. Therefore, further research is needed to investigate the socio-environmental factors associated with returning home.

This study has several limitations. First, this was a single-center study conducted in Japan. If all patients receive complete nursing care, then all patients should be able to return home. However, because this service situation differs depending on the medical system and country, to obtain more generalized findings regarding the association between basic movement ability and home discharge, it is necessary to verify this in multiple countries. Secondly, this was a retrospective study. Although some data were missing, only a few potential factors reported in previous studies were collected. Environmental factors associated with returning home include socioeconomic status, housing, and relational situation^{8, 18, 19)}. Furthermore, the association between cognitive function, bowel/bladder function, and return home has been clarified in recent years^{5, 6)}. These variables could potentially confound the association between basic movement abilities and returning home. Future cohort studies should therefore include these factors.

In conclusion, even after adjusting for socio-environmental factors, basic movement ability was found to be associated with returning home, as was the pre-stroke cohabitation status.

Conflicts of interest

None.