Synergistic effects of family members as caregivers after acute hemorrhagic stroke in Jeju: a preliminary cohort study in Jeju, Korea

Abstract

Objective

All patients with hemorrhagic stroke aspire to recover to their pre-stroke condition. Their initial condition is often poor, but neurological status gradually improves in the absence of postictal complications. After neurosurgical treatment, patients are transferred to a rehabilitation department or an appropriate care facility for active rehabilitation. Subsequently, patients are monitored in the neurosurgical outpatient clinic at the appropriate time, together with their caregivers. During this process, attention was drawn to changes in the modified Rankin Score (mRS) of hemorrhagic stroke patients, and this study focuses on identifying the factors influencing mRS, with a particular emphasis on the role of the caregiver.

Methods

Data from hemorrhagic stroke patients treated between January 2020 and November 2021 were analyzed. The modified Rankin Scale (mRS) was assessed at 1 year after stroke onset. Patients whose mRS score improved by 1 point were classified into the “better” group. Patients who transitioned from an mRS score of 4–5 to a score of 0–3, or those with an improvement of 2 or more points in any mRS range, were classified into the “best” group. Various factors influencing these outcomes were investigated, with a particular focus on analyzing the results based on the family member providing caregiving.

Results

A total of 37 patients with Intracerebral Hemorrhage (ICH) participated in the study. The “better” outcomes were observed in 8 patients, while the “best” outcomes were seen in 16 patients. Most patients who showed improved outcomes received support from family members such as spouses, siblings, or daughters. Notably, when the caregiver role was taken by a daughter, non-independent patients were more likely to regain independence, with mRS scores improving from 4 to 3 or higher or achieving a remarkable improvement of 2 or more points in mRS, leading to the best outcomes.

Conclusions

Functional recovery is promoted not only by the patients’ own efforts but also through sustained support from their families. Early and intensive rehabilitation constitutes a cornerstone of favorable outcomes, with family involvement emerging as a decisive factor.

Introduction

Acute hemorrhagic stroke is associated with severe clinical outcomes and may result in permanent neurological deficits. Most patients with intracerebral hemorrhage (ICH) present to the emergency department with significant neurological impairments, and their symptoms often progress during the acute phase. Unlike ischemic stroke, hemorrhagic stroke lacks a clearly defined therapeutic time window, underscoring the importance of prompt medical evaluation and early supportive management. Rebleeding represents a major complication of hemorrhagic stroke and frequently occurs during hospitalization, with a well-recognized association with worse clinical outcomes. Although no definitive therapy exists to reliably prevent rebleeding, early and sustained rehabilitation remains a cornerstone of recovery and should be initiated as soon as clinically feasible.

Following the acute phase, patients with hemorrhagic stroke are typically transferred to a rehabilitation department and subsequently return to neurosurgical outpatient follow-up, often several months later. Accumulating clinical and neuroscientific evidence indicates that the first six months after stroke constitute a critical period for neurological recovery [1, 2]. This interval is commonly referred to as the “golden phase” of neuroplasticity, during which the brain undergoes reorganization and demonstrates an enhanced capacity for functional recovery through alternative neural pathways.

Nevertheless, more recent studies have demonstrated that meaningful neurological recovery may continue beyond six months after stroke onset. These findings suggest that intensive rehabilitation may also benefit patients in the chronic phase of stroke. For example, patients more than six months post-stroke have been shown to achieve significant improvements in gait and balance through structured and intensive physical therapy [3]. In addition, Harvey and colleagues reported that rehabilitation interventions positively affect not only motor function but also activities of daily living [4]. Together, these observations highlight the persistence of neuroplasticity beyond the early phase and underscore the relevance of evaluating functional recovery at one year after stroke, as well as the factors that may influence long-term outcomes.

In this study, functional status was assessed using the modified Rankin Scale (mRS). Although the mRS does not capture every dimension of neurological impairment, it is widely accepted as a global outcome measure that reflects functional ability, independence, and participation in daily and social activities [5–7].

The overarching goal of stroke rehabilitation is to maximize patient independence and thereby improve quality of life [8, 9]. From a functional perspective, prior studies have demonstrated comparable quality of life between patients with mRS scores of 2 and 3, and favorable outcomes are generally defined as an mRS score of 0–3 [9]. During outpatient follow-up, we observed that a subset of patients exhibited particularly favorable functional recovery as assessed by changes in mRS. This observation prompted further exploration of potential factors associated with such recovery, including the possible contribution of caregiver involvement. We therefore sought to explore whether caregiver-related factors might be associated with functional improvement, while recognizing that such relationships may be complex and multifactorial.

In parallel, growing evidence suggests that the clinical presentation and early course of acute spontaneous ICH vary according to hemorrhage topography. Lobar ICH has been reported to demonstrate a distinct clinical profile compared with deep subcortical ICH, with a higher prevalence of cortical symptoms such as headache, seizures, and visual disturbances [10]. Moreover, some studies have suggested that lobar ICH may be associated with a less favorable early in-hospital prognosis than deep subcortical hemorrhage, although this association remains incompletely defined and appears to be influenced by multiple clinical and etiologic factors [10]. From a pathophysiological standpoint, non-hypertensive mechanisms (including cerebral amyloid angiopathy, vascular malformations, sympathomimetic drug exposure, and coagulation disorders) predominate in lobar hemorrhages, whereas hypertensive small-vessel disease is more strongly associated with deep subcortical ICH. These distinctions underscore the importance of considering hemorrhage location when interpreting the clinical spectrum, early outcomes, and long-term functional recovery in patients with spontaneous ICH.

Materials and methods

Patient population

From January 2020 to November 2021, patients diagnosed with acute hemorrhagic stroke at Jeju National University Hospital were enrolled, and serial evaluations of the modified Rankin Scale (mRS) were conducted at predefined time points, including baseline functional status prior to stroke onset, at hospital discharge, and at 3, 6, and 12 months after discharge. The follow-up period was limited to 1 year, primarily due to loss to follow-up and patient relocation.

Inclusion criteria required that patients had received inpatient treatment exclusively at Jeju National University Hospital, had not relocated to other regions, and had maintained continuous outpatient follow-up. Only patients with spontaneous intracerebral hemorrhage (ICH) were included. Patients with subarachnoid hemorrhage (SAH) or isolated intraventricular hemorrhage (IVH) were excluded because their clinical courses were considered either too favorable or too severe to allow meaningful comparison of functional recovery.

The study protocol was approved by the Institutional Review Board (IRB) of Jeju National University Hospital, and the requirement for written informed consent was waived due to the retrospective nature of the study (IRB No. 2024–09-002).

Patients were classified into five outcome groups: Best, Better, Steady, Worse, and Worst. The Best group included patients whose mRS improved from 4 to 3, indicating a transition from a dependent to an independent status, as well as those who showed an improvement of two or more points within the mRS ranges of 0–3 or 4–5. The Better group consisted of patients who demonstrated a one-point improvement in mRS within the same ranges. The Steady group included patients whose mRS scores remained unchanged, whereas the Worse and Worst groups were defined as the inverse counterparts of the Better and Best groups, respectively.

In this study, mRS scores of 2 and 3 were categorized within the same functional group. Patients with an mRS score of 3, although requiring some assistance, are able to ambulate independently and are clinically distinct from those with an mRS score of 4, who are unable to walk without assistance. Previous studies have demonstrated that patients with mRS scores of 2 and 3 report comparable health-related quality of life despite differences in functional disability, suggesting that mRS 3 is more closely aligned with mRS 2 than with mRS 4 [9].

Statistical analysis

All statistical analyses were performed using IBM SPSS Statistics for Windows, version 27.0 (IBM Corp., Armonk, NY, USA). Categorical variables were compared across outcome groups using Pearson’s chi-square test or the Fisher–Freeman–Halton exact test, as appropriate. Continuous variables were analyzed using one-way analysis of variance (ANOVA). Variables with a p-value < 0.10 in univariate analyses, along with clinically relevant covariates, were included in multivariable models. Multivariable ordinal logistic regression was performed to identify independent predictors of functional outcomes, and multivariable linear regression was used to explore associations between clinical factors and outcome measures. All statistical tests were two-sided, and p-values < 0.05 were considered statistically significant.

General treatment and management of spontaneous intracerebral hemorrhage

Upon arrival at the emergency department of Jeju National University Hospital, all patients underwent evaluation of vital signs, airway, breathing, and circulation. Airway management and oxygen supplementation were provided as needed, and systolic blood pressure was controlled to below 140 mmHg. Level of consciousness and neurological status were assessed using the Glasgow Coma Scale (GCS), a standardized and widely used clinical scle for the assessment of impaired consciousness [11].

Non-contrast computed tomography (CT) was performed to determine the location and volume of hemorrhage, the presence of perihematomal edema, and midline shift. CT angiography was additionally performed in patients with suspected aneurysms, arteriovenous malformations, or tumor-related hemorrhage, particularly when the hemorrhage was located in an atypical region. Decisions regarding surgical intervention were based on hematoma volume, level of consciousness, and overall physiological status.

Factors influencing modified rankin score

Clinical variables analyzed included Sex, age, hematoma volume, hemorrhage location, presence of ventricular extension, surgical intervention, length of intensive care unit (ICU) stay, reoperation, transfer to the department of rehabilitation medicine, total duration of hospitalization, discharge disposition (home versus transfer to another hospital), and in-hospital complications such as pneumonia and urinary tract infection. Vascular and medical risk factors assessed included hypertension, smoking status, heavy alcohol consumption, dyslipidemia, use of anticoagulant or antiplatelet agents, and chronic kidney disease. Smoking status was analyzed separately due to its potential clinical relevance.

Results

Demographics and clinical presentation

Among the 42 patients with intracerebral hemorrhage (ICH), four were lost to follow-up and one declined participation after discharge. All patients received standard acute management and were discharged after completion of treatment. Neurological status was subsequently evaluated during outpatient follow-up, and a total of 37 patients were included in the final analysis. Functional outcomes were assessed by comparing modified Rankin Scale (mRS) scores at discharge with those obtained one year later. Based on longitudinal changes in mRS scores, patients were classified as follows: 16 in the “Best” group, 8 in the “Better” group, 9 in the “Steady” group, 1 in the “Worse” group, and 3 in the “Worst” group (Table 1).

Table 1.

Clinical and radiological characteristics according to functional outcome groups in patients with intracerebral hemorrhage

	Best (n = 16)	Better (n = 8)	Steady (n = 9)	Worse (n = 1)	Worst (n = 3)	P value
Sex						P = 0.2498
M	9	8	6	1	2
F	7	0	3	0	1
Age	67.4 ± 13.7	60.5 ± 12.7	60.1 ± 16.9	75	79.3 ± 7.4	P = 0.16
Hematoma volume	20 ± 19.0	8.9 ± 16.6	33 ± 14.0	3.2	5.13 ± 3.50	P = 0.044
Hematoma location
Basal ganglia	5	3	5	0	0
Thalamus	5	1	2	0	2
Lobe	4	3	2	1	0
Cerebellum	2	1	0	0	1
Extension to ventricle						P = 0.438
Yes	4	2	4	0	0
No	12	6	5	1	3
Smoking						P = 0.075
Yes	4	5	4	0	0
No	12	3	5	1	3
Risk factor HTN, Smoking, Heavy alcohol, Dysplipidemia, Anticoagulant, Antiplatelet, Chronic kidney disease, Multiple birth, Harsh working environment
Yes	33	16	15	2	6
No	2	0	1	0	1
Operation						P = 0.008
Yes	9	1	4	0	0
No	7	7	5	1	3
Length of stay in the ICU	5.1 ± 4.2	3.8 ± 7.0	11.8 ± 13.4	1	4.5 ± 5.5	P = 0.0224
Reoperation						P = 0.448
Yes	1	1	0	0	0
No	15	7	9	1	3
Transfer to the Rehabilitation meidicine						P = 0.112
Yes	12	2	4	0	2
No	4	6	5	1	1
Total length of stay	39.4 ± 21.5	23.1 ± 29.3	37.2 ± 21.9	7	32.3 ± 2.5	P = 0.355
Type of discharge						P = 0.012
Home	9	7	1	1	1
Transfer	7	1	8	0	2
Complication						P = 0.129
Pneumonia	1	0	2	0	0
UTI	1	0	0	0	0
Bleeding (SAH, ICH)	2	0	0	0	0

There were no statistically significant differences in age or sex among the groups. In contrast, mean hematoma volumes differed significantly across outcome groups (p = 0.044), measuring 20.0 ± 19.0 mL in the Best group, 8.9 ± 16.6 mL in the Better group, 33.0 ± 14.0 mL in the Steady group, 3.2 mL in the Worse group, and 5.1 ± 3.5 mL in the Worst group.

Hematoma location varied across groups. In the Best group, hematomas were located in the basal ganglia (n = 5), thalamus (n = 5), lobar regions (n = 4), and cerebellum (n = 2). The Better group included three basal ganglia, one thalamic, three lobar, and one cerebellar hemorrhage. In the Steady group, five hemorrhages were located in the basal ganglia, with two each in the thalamus and lobar regions. The single patient in the Worse group had a lobar hemorrhage, whereas the Worst group included two thalamic and one cerebellar hemorrhage.

When risk factors were assessed without subdivision, three patients (18%) in the Best group and one patient (11%) in the Steady group had no identifiable risk factors. All patients in the Better group and the single patient in the Worse group had at least one risk factor. In the Worst group, two patients (66%) had identifiable risk factors, while one (33%) did not. When smoking status was analyzed separately, differences among groups approached but did not reach statistical significance (p = 0.075). The proportion of non-current smokers was 75% in the Best group, 37.5% in the Better group, and 56% in the Steady group. The single Worse patient and all patients in the Worst group were non-smokers (Table 2.)

Table 2.

distribution of risk factors according to functional outcome groups

Risk factor
	Best (n = 16)	Better (n = 8)	Steady (n = 9)	Wore (n = 1)	Worst (n = 3)
HTN	12	4	5	0	2
Smoking	3	5	4	0	0
Heavy alcohol	5	3	3	1	1
Dyslipidemia	2	2	0	0	0
Anticoagulant	5	1	1	0	1
Antiplatelet	1	0	1	0	0
Chronic kidney disease	0	0	0	1	1
Multiple birth	4	0	1	0	1
Harsh working environment	1	1	0	0	0

Rates of surgical intervention differed significantly among groups (p = 0.008), whereas reoperation rates did not (p = 0.448). In the Best group, 9 of 16 patients (56%) underwent surgery, including one patient who required reoperation. In the Better group, one of eight patients underwent surgery and subsequently required reoperation. In the Steady group, four of nine patients underwent surgery without reoperation. Neither the Worse patient nor any patients in the Worst group underwent surgery; all presented with a Glasgow Coma Scale score of 15 on admission.

Discharge disposition also differed significantly among groups (p = 0.012). In the Best group, 56.3% were discharged home, whereas 43.8% were transferred to another hospital. In the Better group, 87.5% were discharged home. In contrast, only 11.1% of patients in the Steady group were discharged home. The single Worse patient was discharged home, while two-thirds of patients in the Worst group were transferred to another facility.

Although not statistically significant, caregiver type emerged as a noteworthy finding (Fig. 1, Table 3). Among the 16 patients in the Best group, 11 transitioned from dependency to independence, and five demonstrated an improvement of two or more mRS points. Among those who achieved independence, daughters were the most common caregivers, followed by spouses. In contrast, most patients with a ≥ 2-point improvement managed their care independently.

Fig. 1.

Distribution of primary caregivers according to functional recovery outcomes after intracerebral hemorrhage

Table 3.

Caregiver distribution in the best outcome group according to type of functional improvement

Best (n = 16)
Caregiver	Dependent → Independent (n = 11)	Upgrade 2 Score (n = 5)
Daugther	7	0
Spouse	4	0
Patient	0	4
Son	0	0
Daughter-in law	0	1
Brother	0	0

Overall, hematoma volume, surgical intervention, and discharge disposition were significantly associated with functional outcomes. In contrast, ventricular extension, smoking status, ICU length of stay, total hospitalization duration, reoperation, transfer to rehabilitation medicine, and in-hospital complications were not significantly associated with outcomes.

Patients were divided into five functional outcome groups (Best, Better, Steady, Worse, and Worst). The Best group had a larger mean hematoma volume (20.0 ± 19.0 mL) compared to the Worse and Worst groups (P = 0.044). Basal ganglia and thalamus were the most common hematoma locations. Surgical treatment was most frequent in the Best group (P = 0.008), and patients in this group were more often discharged home (P = 0.012). Other factors, including ventricular extension, smoking, ICU stay, and complications, showed no significant differences among groups.

The stacked bar chart demonstrates the proportional distribution of primary caregivers (daughter, spouse, patient self, son, daughter-in-law, and brother) across five functional outcome groups—Best, Better, Steady, Worse, and Worst. Daughters were notably predominant in the Best outcome group, followed by spouses, while other family members such as sons, daughters-in-law, and brothers were more frequently involved in the lower outcome categories. This trend suggests a potential association between caregiver type and the degree of functional improvement following hemorrhagic stroke.

In the Best group (n = 16), daughters were the main caregivers among patients who improved from dependent to independent status (n = 11), followed by spouses. Among those with a two-point mRS improvement (n = 5), most managed their own recovery. Overall, daughters played the key caregiving role in the best recovery cases.

Discussion

In this study, in-hospital factors including ICU length of stay, total duration of hospitalization, and common complications such as pneumonia, urinary tract infections, and rebleeding did not differ significantly between outcome groups. These findings suggest that, within this cohort, neither the intensity nor the duration of acute inpatient care, nor the occurrence of these complications, exerted a substantial influence on long-term functional outcomes. Although clinically important, these factors may not represent primary determinants of functional recovery in patients with intracerebral hemorrhage.

Similarly, contrary to findings reported in some previous studies, hemorrhage topography classified as lobar versus deep cerebral hemorrhage was not significantly associated with functional outcome in the present cohort. Although the distribution of lobar and deep hemorrhages varied across outcome groups, these differences did not reach statistical significance, suggesting that hemorrhage location alone did not independently determine long-term recovery in this population. This finding contrasts with prior reports indicating differences in early clinical profile and prognosis according to hemorrhage location, highlighting the potential influence of cohort characteristics, outcome definitions, and follow-up duration on observed associations.

Previous studies have consistently emphasized the prognostic significance of intraventricular hemorrhage (IVH) in patients with intracerebral hemorrhage, identifying ventricular extension as a strong determinant of poor early outcome and increased in-hospital mortality [12]. Several hospital-based registry studies have demonstrated that IVH is independently associated with worse neurological outcomes and higher short-term mortality rates [12]. However, in contrast to these earlier findings, our study did not identify a significant association between the presence of IVH and functional outcomes at one year. While IVH is widely regarded as a marker of disease severity in the acute phase, its impact on long-term functional recovery may be attenuated in selected patient populations or outweighed by other dominant factors such as hematoma volume, surgical intervention, discharge disposition, and post-discharge support.

By contrast, functional outcomes in this study were significantly influenced by hematoma volume, surgical intervention, and discharge disposition. Larger hematoma volumes were associated with worse outcomes, consistent with established evidence linking hematoma burden to neurological injury. Surgical intervention was correlated with improved prognosis, underscoring the importance of timely operative management in appropriately selected patients, whereas reoperation did not demonstrate a significant effect. Referral to rehabilitation medicine was not independently associated with long-term recovery; however, discharge to home rather than transfer to another facility was linked to better outcomes, suggesting that the post-discharge environment may play an important role in recovery trajectories.

Although this study included only patients with intracerebral hemorrhage and excluded those with subarachnoid hemorrhage, the findings may provide a conceptual framework for future investigations in other hemorrhagic stroke populations. Given the abrupt onset, high acute mortality, and frequent long-term cognitive impairments associated with subarachnoid hemorrhage, the burden on family caregivers may be comparable to or greater than that observed in intracerebral hemorrhage. Previous studies suggest that structured family involvement in subarachnoid hemorrhage care is associated with improved rehabilitation adherence and long-term outcomes [12], and emerging advances in treatment may further mitigate caregiver burden by reducing neurological disability and accelerating recovery [13].

An exploratory observation of this study was the distribution of caregiver type across functional outcome groups (Fig. 1, Table 3). Although daughters and spouses were more frequently observed among patients who demonstrated clinically meaningful recovery, this pattern did not reach statistical significance and should therefore be interpreted with caution. Accordingly, caregiver involvement is not described here as an independent prognostic determinant, but rather as a potential contributing factor that may support recovery through practical and psychosocial pathways.

Existing literature indicates that caregivers can facilitate post-stroke recovery by managing medications, coordinating medical appointments, monitoring key health indicators such as blood pressure and blood glucose, and supporting adherence to rehabilitation programs and home-based exercises [14–17]. Emotional support and motivation provided by caregivers may also promote psychological stability and sustained engagement in rehabilitation, both of which are important components of functional improvement [18]. In the present cohort, daughters and spouses were more commonly involved in patients who transitioned from dependency to greater independence. However, given the absence of statistical significance, this finding should be regarded as hypothesis-generating rather than confirmatory.

In the Korean cultural context, particularly within close-knit communities such as Jeju, caregiving by family members, especially daughters, often reflects traditional values of filial responsibility and family solidarity. Beyond physical assistance, caregivers may provide emotional reassurance and sustained encouragement, potentially supporting patients’ efforts toward greater independence. While these cultural factors may shape caregiving practices, their direct impact on functional outcomes cannot be conclusively determined from the present data and warrants further investigation.

Future research should explore the multifactorial determinants of functional recovery after intracerebral hemorrhage beyond traditional clinical and radiological factors. Prospective multicenter studies with larger cohorts are needed to clarify the independent effects of caregiver characteristics, social support structures, and rehabilitation intensity on long-term outcomes. In addition, age-specific analyses focusing on very elderly patients aged 85 years or older are particularly important, as this population is known to have distinct demographic characteristics, comorbidity profiles, and prognostic trajectories [19].

This study has several limitations. The relatively small sample size may have limited statistical power and external validity, and the cohort size precluded a dedicated analysis of very elderly patients aged 85 years or older. Age-related factors such as frailty, cognitive impairment, and reduced tolerance to rehabilitation, which may substantially influence functional recovery and independence, could not be adequately evaluated. Furthermore, among patients with favorable outcomes, clinically meaningful recovery was more accurately reflected by a transition from dependency to independence rather than by a simple two-point increase in the modified Rankin Scale score. Although patients cared for by daughters tended to demonstrate more favorable outcomes, this association did not reach statistical significance, underscoring the need for larger longitudinal studies to validate the potential influence of caregiver dynamics.

Conclusion

Although the dataset is limited in size, the findings suggest that functional outcomes after intracerebral hemorrhage are influenced by hematoma volume, surgical intervention, and discharge disposition. Caregiver involvement, particularly by spouses or daughters, was more frequently observed among patients with favorable recovery; however, this observation should be interpreted cautiously and viewed as hypothesis-generating rather than definitive. Further large-scale, prospective studies are required to clarify the role of family dynamics and caregiver characteristics in long-term functional recovery after hemorrhagic stroke.

Author’s contributions

All authors: conception and design; JKR, EHL: provision of study materials; JKR, EHL: collection and assembly of data; All authors: data analysis and interpretation; All authors: manuscript writing and editing; All authors: final approval of manuscript.

Funding

This work was supported by a 2025 research grant from Jeju National University Hospital.

Data availability

The datasets generated or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

This study was approved by the Institutional Review Board of Jeju National University Hospital (IRB No. 2024–09-002).

The requirement for informed consent was waived due to the retrospective design of the study.

The study was conducted in accordance with the Declaration of Helsinki.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.