The Necessity, Effectiveness, and Inhibiting Factors of Physical Therapy in the Psychiatric Ward of a General Hospital

ABSTRACT

Objectives:

Physical complications in psychiatric patients and the role of physical therapy (PT) are gaining attention. However, research remains limited. This study examined the necessity, effectiveness, and inhibiting factors of PT in a psychiatric ward.

Methods:

This retrospective observational study evaluated inpatients undergoing rehabilitation in the psychiatric ward of a general hospital between April 2017 and March 2022. The data collected included characteristics, psychiatric diagnosis, physical complications, time to rehabilitation initiation, training implementation rate, training time, Barthel Index (BI) at admission and discharge, length of hospital stay, and residence before and after hospitalization. Univariate analysis was performed to compare patients who received PT with those who did not receive PT. Among patients who received PT, factors inhibiting independence in ADLs (BI ≥85) and the BI gain were analyzed using binomial logistic regression and multiple regression analysis.

Results:

Of all patients, 772 received PT, and they were significantly older, more likely to be institutionalized, and had higher rates of organic mental disorders, musculoskeletal disorders, and internal conditions compared to the 317 patients who did not receive PT. PT resulted in a significant BI gain, with 596 individuals achieving independence. Independence and BI gain were significantly associated with pre-hospitalization factors, including age, psychiatric diagnosis, residence, and BI, whereas rehabilitation-related indicators showed no significant association.

Conclusions:

Many patients with advanced age, organic mental disorders, physical complications, and low BI achieved functional independence following PT. However, reduced activity levels before hospitalization hindered recovery. Preventive efforts should target physical decline in both community and hospital settings.

INTRODUCTION

In recent years, the number of individuals with mental disorders in Japan has steadily increased, reaching approximately 6,148,000 as of 2020.¹⁾ Alongside this trend, the number of patients with psychiatric disorders comorbid with physical complications is also growing. It is estimated that 68% of patients with mental health conditions also experience physical comorbidities.²⁾ A significant rise in psychiatric outpatient numbers has been observed, particularly among those under the age of 65 years with mood and neurotic disorders.¹⁾ Consequently, mental illnesses are often identified incidentally during treatment for physical conditions.

Although the number of psychiatric inpatients is declining, approximately 288,000 individuals, primarily with schizophrenia, remain hospitalized, with approximately 64% aged over 65 years.¹⁾ The prevalence of physical complications tends to increase with age. Moreover, prolonged hospitalization can negatively impact physical functioning and activities of daily living (ADLs).³⁾ The average length of stay in psychiatric wards is decreasing. However, it was reported to be 277 days in fiscal 2020, which is still a long period.¹⁾

The side effects of psychotropic medications are the most important factor to consider as a cause of physical complications. Schizophrenia, in particular, is associated with a substantially increased risk of fracture⁴⁾ and osteoporosis-related fracture, risks that are significantly exacerbated by the use of psychotropic drugs.⁵⁾ Adverse effects of psychotropic drugs include parkinsonism, neuroleptic malignant syndrome, rhabdomyolysis, hyponatremia, QT prolongation, dysphagia, and various metabolic abnormalities. These side effects collectively contribute to an elevated risk of osteoporosis, falls, and ultimately fractures.

Various reports have examined the relationship between internal disorders and mental illness. The risk ratios for heart failure, ventricular arrhythmia, and cardiovascular death are 1.6, 2.3, and 2.2, respectively.⁶⁾ These risks are associated with a reduced life expectancy of 15–20 years, lifestyle factors such as smoking and lack of exercise, and metabolic abnormalities induced by psychotropic drugs.⁷⁾ Additionally, several antipsychotic drugs have been linked to an increased risk of pneumonia, with certain drug combinations showing a synergistic effect that further exacerbates this risk.^8,9) Antipsychotics are particularly associated with side effects such as parkinsonian symptoms, acute dystonic reactions, tardive dyskinesia, drooling, xerostomia, and sedation. These effects can lead to dysphagia, aspiration pneumonia, and choking.¹⁰⁾

Additional contributors to physical complications include high-risk behaviors, such as overdosing and water intoxication, and organic mental disorders that encompass both physical and psychiatric symptoms. Behavioral abnormalities may contribute to malnutrition caused by eating disorders, alcohol-induced liver dysfunction, and pancreatitis. Patients with schizophrenia and physical comorbidities often have low treatment adherence and limited access to physical healthcare. These challenges may stem from barriers to healthcare access, health-related beliefs, and communication difficulties.¹¹⁾ Therefore, individuals with schizophrenia face increased risk of premature aging, weight gain, impaired daily functioning, and metabolic and cardiovascular diseases.¹²⁾

Traditionally, psychiatric occupational therapy (OT) has been the primary rehabilitative approach for patients with mental illness. Recently, with the rising incidence of physical complications, the role of physical therapy (PT) in conjunction with OT is gaining attention. However, few physical therapists are currently working in psychiatric settings, and research specifically focused on PT in mental health remains limited, both in Japan and internationally.¹³⁾ This study investigated the current state of PT in a psychiatric ward of a general hospital to examine its necessity, effectiveness, and inhibiting factors.

MATERIALS AND METHODS

Study Design and Participants

This retrospective observational study was approved by the Ethics Committee of Japanese Red Cross Ashikaga Hospital (approval number 2025–5). Informed consent was obtained using the opt-out method via our hospital website, and individuals who did not opt out were included in the study. All patient information was de-identified. This study was conducted in accordance with the principles of the Declaration of Helsinki.

Patients admitted to the Psychiatry Department of Japanese Red Cross Ashikaga Hospital between April 2017 and March 2022, who were referred to the Rehabilitation Department for mental disorders or physical complications and prescribed individualized rehabilitation, were included. Japanese Red Cross Ashikaga Hospital is an acute care hospital with 540 beds, including 40 beds in the psychiatric ward, and treats psychiatric patients with physical complications. The average length of stay is 14.2 days. The hospital has 60 rehabilitation staff members, including 1 physical therapist and 2 occupational therapists dedicated to the psychiatric ward. Medical records of these inpatients were reviewed retrospectively. Raw data were generated at the Japanese Red Cross Ashikaga Hospital, and all data supporting the findings of this study are fully available without restriction.

To identify psychiatric patients who may benefit from PT, patient characteristics, psychiatric diagnoses, and physical complications were compared between patients who received PT and those who did not receive PT. Moreover, rehabilitation interventions and outcomes were examined in each patient to assess the effectiveness of PT. The following exclusion criteria were applied: 1) patients who received only group psychiatric OT, 2) patients discharged because of death, and 3) patients whose PT time was less than 10% of their OT time (many of these patients were prescribed PT, but only underwent evaluation and were determined to have little need for training, making it difficult to determine their appropriate group classification).

In addition, for patients who received PT, factors inhibiting independence were analyzed by comparing the three groups: the independent group [Barthel Index (BI) ≥85 at discharge], the partially assisted group (60 < BI < 85), and the assisted group (BI ≤60). The BI assesses the degree of independence in ADLs on a scale of 0 to 100.¹⁴⁾ A BI of 85 or higher indicates full to modified independence, whereas a BI of 60 or lower indicates partial to full assistance.¹⁵⁾ Patients with BIs above 60 or below 85 were a mix of independent patients and patients requiring assistance.

Characteristics, Psychiatric Diagnosis, and Physical Complications

Age, sex, pre-hospitalization residence, the Global Assessment of Functioning (GAF) score, psychiatric diagnosis, and physical complications were retrospectively reviewed from medical records. The GAF scale is a numerical scale used by mental health clinicians to subjectively rate an individual’s psychological, social, and occupational functioning. Scores range from 100 (extremely high functioning) to 1 (severely impaired).¹⁶⁾ Psychiatric disorders were classified according to the tenth edition of the International Classification of Diseases.¹⁷⁾ The following classifications were used: F0, organic, including symptomatic, mental disorders; F1, mental and behavioral disorders caused by psychoactive substance use; F2, schizophrenia, schizotypal and delusional disorders; F3, mood (affective) disorders; F4, neurotic, stress-related, and somatoform disorders; F5, behavioral syndromes associated with physiological disturbances and physical factors; F6, disorders of adult personality and behavior; F7, mental retardation; F8, disorders of psychological development; and F9, behavioral and emotional disorders with onset usually occurring in childhood and adolescence.

Physical complications were classified based on the disease domains defined by the Japanese Society of Rehabilitation Medicine: Domain 1, cerebrovascular disorders, head trauma, and other brain diseases; Domain 2, musculoskeletal disorders and trauma; Domain 3, traumatic spinal cord injury; Domain 4, neuromuscular disorders; Domain 5, amputation; Domain 6, pediatric disorders; Domain 7, rheumatic disorders; Domain 8, internal disorders; and Domain 9, other disorders. When a decline in ADLs was observed despite the absence of any physical complications listed above, the cause was attributed to disuse syndrome.

Rehabilitation and Outcomes

OT was provided to all inpatients, either in group settings or individually, depending on the severity of their psychiatric symptoms. PT was prescribed when physical function and ADLs had declined or were expected to decline because of prolonged bed rest or physical complications. PT needs were assessed by physical and occupational therapists dedicated to the psychiatric wards, along with psychiatrists and rehabilitation physicians, during admission, at regular weekly conferences, and whenever significant changes in medical condition occurred. Speech-language therapy was provided only for patients with dysphagia and speech dysfunction.

Physical restraints were performed only when absolutely necessary to protect the patient’s life based on concerns about self-harm or harm to others accompanying mental symptoms. To address the risk of falls and other accidents associated with physical complications, we took measures such as frequent monitoring, bed exit sensors, low beds, and wearing headgear, while refraining from physical restraint. The continued use of these measures was reviewed periodically and whenever there were changes in the patient’s condition. These reviews were performed by a multidisciplinary team including psychiatrists, nurses, physical therapists, and occupational therapists.

Psychiatric OT included manual work, creative activities, relaxation exercises, cognitive function training, and social skills training. These activities involved the minimum necessary physical activity. However, their main purpose was to improve mental function. Physical therapists focused on preventing complications through respiratory physiotherapy, muscle strengthening training, early mobilization, and exercise therapy such as gait training and aerobic exercises. They also educated psychiatric ward staff on patient assistance, participated in discharge planning conferences, and conducted pre-discharge home visit assessments. The weekly multidisciplinary conference was attended by doctors, nurses, physical therapists, occupational therapists, speech therapists, social workers, nutritionists, and pharmacists. During the 30-min conference, each profession reported on the changes in each patient over the course of a week and the effectiveness of their approach. They also discussed and shared issues and goals among the various professions. To evaluate the implementation and effectiveness of rehabilitation, the following indicators were assessed: days from hospitalization to the start of rehabilitation, treatment days of PT and OT, training implementation rate relative to length of hospital stay of PT and OT, training time per treatment day of PT and OT, BIs at admission and discharge, BI gain calculated by subtracting the BI at admission from the BI at discharge, length of hospital stay, and post-discharge residence.

Statistical Analysis

Patient characteristics, psychiatric diagnoses, physical complications, rehabilitation details, and outcomes were compared between patients who received PT and those who did not receive PT, as well as among the independent, partially assisted, and assisted groups. Age, GAF score, days from hospitalization to the start of rehabilitation, training implementation rate, training time per treatment day, BIs at admission and discharge, BI gain, and length of hospital stay were compared using the Mann–Whitney U test and the Kruskal–Wallis test. Sex, psychiatric diagnoses, physical complications, pre-hospitalization residence, and post-discharge residence were compared using Fisher’s exact test. Multiple comparisons involving psychiatric diagnoses and physical complications, as well as intergroup comparisons between the independent, partially assisted, and assisted groups, were adjusted using the Bonferroni correction.

Factors inhibiting independence in ADLs and BI gain after PT were analyzed using binomial logistic regression and multiple regression analysis. The objective variables were the need for assistance with ADLs at discharge and the BI gain. Independent variables included age, sex, GAF score, psychiatric diagnosis, physical complications, pre-hospitalization residence, days from hospitalization to the start of rehabilitation, training implementation rate, training time per treatment day, BI at admission, and length of hospital stay. Statistical significance was set at P < 0.05. For psychiatric diagnoses and physical complications, the significance level was set at P<0.005 (0.05/10) using the Bonferroni method. All statistical analyses were performed using SPSS version 30.0 (IBM, Armonk, NY, USA).

RESULTS

Patient Characteristics in PT and Non-PT Groups

Figure 1 illustrates the participant selection process. Individualized rehabilitation was conducted in 1114 of the 3094 cases. After excluding 8 patients because of death, a total of 1106 inpatients were included in this study. Among them, 772 patients received sufficient PT, whereas 317 did not receive PT. Patient characteristics, psychiatric conditions, and physical complications in the PT and non-PT groups are shown in Table 1. PT was more commonly administered to older and pre-institutionalized patients with physical comorbidities, particularly musculoskeletal and internal disorders, as well as those with organic mental disorders (P<0.001). In contrast, schizophrenia was significantly more prevalent among patients who did not receive PT (P<0.001).

Fig. 1.

 Flowchart of the participant selection process. Patients were excluded if they had no request for rehabilitation, received only group psychiatric OT, or were discharged because of death. A total of 1106 patients were included in the study, comprising 772 who received PT and 317 who did not receive PT. Among the patients who received PT, three subgroups were further analyzed: 596 patients in the independent group (BI ≥85 at discharge), 50 patients in the partially assisted group (60< BI <85), and 126 patients in the assisted group (BI ≤60).

Table 1.  Patient characteristics, psychiatric disease, and physical complications for patients who received PT and those who did not receive PT.

Characteristic	Received PT (n=772)	No PT (n=317)	P
Age, years	60 [45–74]	47 [32–64]	<0.001***
Female sex	422 (54.7)	201 (63.4)	<0.01**
GAF score	30 [30–30]	30 [30–30]	<0.001***
Psychiatric disease			<0.001***
F0	185 (24.0)	25 (7.9)	<0.001a
F1	39 (5.1)	10 (3.2)	0.20
F2	218 (28.2)	133 (42.0)	<0.001a
F3	220 (28.5)	111 (35.0)	0.04
F4	22 (2.8)	11 (3.5)	0.56
F5	56 (7.3)	12 (3.8)	0.04
F6	1 (0.1)	4 (1.3)	0.03
F7	23 (3.0)	6 (1.9)	0.41
F8	7 (0.9)	5 (1.6)	0.35
F9	1 (0.1)	0 (0)	1.00
Physical complications			<0.001***
Cerebrovascular disorders	178 (23.1)	68 (21.5)	0.58
Musculoskeletal disorders	78 (10.1)	10 (3.2)	<0.001a
Spinal cord injury	5 (0.6)	0 (0)	0.33
Neuromuscular disorders	8 (1.0)	0 (0)	0.11
Amputation	0 (0)	0 (0)	1.00
Pediatric disorders	1 (0.1)	0 (0)	1.00
Rheumatic disorders	4 (0.5)	1 (0.3)	1.00
Internal disorders	76 (9.8)	9 (2.8)	<0.001a
Other	171 (22.2)	48 (15.1)	0.01
Disuse syndrome or no complication	251 (32.5)	181 (57.1)	<0.001a
Pre-hospitalization residence: home	623 (80.7)	285 (89.9)	<0.001***

Data are shown as median [interquartile range] or number (percentage).

^**P<0.01; ^***P<0.001; Mann–Whitney U test or Fisher's exact test.

^a P<0.01/10; multiple comparisons were corrected using the the Bonferroni method.

Rehabilitation and Outcomes in PT and Non-PT Groups

Table 2 presents rehabilitation details and outcomes for patients who received PT and those who did not receive PT. Both groups showed significant improvement in ADLs at discharge compared to admission (P<0.001). Patients in the PT group began rehabilitation significantly earlier (P<0.001) and had longer hospital stays (P<0.001) than the non-PT group. Although patients in the PT group had a significantly lower OT training implementation rate, the combined implementation rate and training time for PT and OT were significantly higher (P<0.001). Patients with significantly lower BIs at admission underwent PT, and their BI gain was significantly higher (P<0.001). However, their BIs at discharge remained significantly lower than those of the non-PT group (P<0.001). Among the patients who received PT, 596 (76.5%) achieved modified independence, whereas 126 (16.3%) still required assistance. Subsequently, we compared the 596 patients in the independent group with the 50 patients in the partially assisted group and the 126 patients in the assisted group (Fig. 1).

Table 2.  Rehabilitation and outcomes for patients who received PT and those who did not receive PT.

Variable	Received PT (n=772)	No PT (n=317)	P
Time from hospitalization to start of rehabilitation, days	0 [0–1]	1 [0–2]	<0.001***
PT treatment, days	8 [4–16]	―	―
PT training implementation rate, %	33.3 [20.0–48.7]	―	―
PT time per treatment day, min	22.2 [20–27.6]	―	―
OT treatment, days	11 [5.5–18]	8 [4–14]	―
OT training implementation rate, %	37.5 [25.9–52.2]	45.5 [33.3–59.4]	<0.001***
OT time per treatment day, min	23.7 [20–29.2]	24 [20–30]	0.52
BI at admission	65 [25–85]	85 [85–100]	<0.001***
BI at discharge	100 [85–100]	100 [100–100]	<0.001***
BI gain	20 [10–45]	15 [0–15]	<0.001***
Length of hospital stay, days	31 [17–51]	20 [10–33]	<0.001***
Post-discharge residence			<0.001***
Home	583 (74.8)	283 (89.3)
Nursing home	189 (24.3)	34 (10.7)

Data are shown as median [interquartile range] or number (percentage).

^***P<0.001; Mann–Whitney U test or Fisher's exact test.

Patient Characteristics in Independent Group, Partially Assisted Group, and Assisted Group

Patient characteristics, psychiatric conditions, and physical complications for the independent, partially assisted, and assisted groups are shown in Table 3. Participants in the assisted group were significantly older (P<0.001/3), had a higher rate of living at home prior to hospitalization (P<0.001/3), and exhibited more physical complications, especially spinal cord injuries and internal disorders, than the independent group (P<0.005/3). Mood disorders and behavioral syndromes were significantly more common in the independent group, whereas organic mental disorders were more prevalent in the assisted group (P<0.001/3).

Table 3.  Patient characteristics, psychiatric disease, and physical complications for the independent group (BI ≥85), the partially assisted group (60< BI <85), and the assisted group (BI ≤60).

Characteristic	Independent (n=596)	Partially assisted (n=50)	Assisted (n=126)	P
Age, years	56 [44–72]	63.5 [44.5–75]	70.5 [58.5–79]	<0.001***
Female sex	336 (56.4)	24 (48.0)	62 (49.2)	0.21
GAF score	30 [30–30]	30 [30–30]	30 [23.8–30]	0.18
Psychiatric disease				<0.001***
F0	105 (17.6)	22 (44.0)	58 (46.0)	<0.001b
F1	33 (5.5)	1 (2.0)	5 (4.0)	0.62
F2	167 (28.0)	16 (32.0)	35 (27.8)	0.46
F3	198 (33.2)	3 (6.0)	19 (15.1)	<0.001b
F4	19 (3.2)	0 (0)	3 (2.4)	0.57
F5	55 (9.2)	1 (2.0)	0 (0)	<0.001b
F6	1 (0.2)	0 (0)	0 (0)	1.00
F7	12 (2.0)	6 (12.0)	5 (4.0)	0.002a
F8	5 (0.8)	1 (2.0)	1 (0.8)	0.44
F9	1 (0.2)	0 (0)	0 (0)	1.00
Physical complications				<0.001***
Cerebrovascular disorders	122 (20.5)	18 (36.0)	38 (30.2)	0.006
Musculoskeletal disorders	56 (9.4)	7 (14.0)	15 (11.9)	0.40
Spinal cord injury	0 (0)	1 (2.0)	4 (3.2)	<0.001b
Neuromuscular disorders	6 (1.0)	0	2 (1.6)	0.79
Amputation	0 (0)	0	0 (0)	1.00
Pediatric disorders	1 (0.2)	0	0 (0)	1.00
Rheumatic disorders	4 (0.7)	0	0 (0)	1.00
Internal disorders	46 (7.7)	2 (4.0)	29 (22.2)	<0.001b
Other	140 (23.5)	11 (22.0)	20 (15.9)	0.50
Disuse syndrome or no complication	221 (37.1)	11 (22.0)	18 (14.3)	<0.001b
Pre-hospitalization residence: home	531 (89.1)	38 (76.0)	61 (48.4)	<0.001***

Data are shown as median [interquartile range] or number (percentage).

^***P<0.001; Kruskal–Wallis test or Fisher's exact test.

^a P<0.05/10; ^b P<0.01/10; multiple comparisons were corrected using the Bonferroni method.

Rehabilitation and Outcomes in Independent Group, Partially Assisted Group, and Assisted Group

Table 4 presents rehabilitation details and outcomes for the independent, partially assisted, and assisted groups. All the groups showed significant improvement in ADLs at discharge (P<0.001), and the partially assisted group showed the greatest BI gain. The assisted group began rehabilitation earlier (P<0.01/3), had a higher training implementation rate of PT (P<0.001/3), had more PT time per treatment day (P<0.001/3), and had longer hospitalizations (P<0.05/3) than the independent group. However, despite the earlier and more intensive rehabilitation, the assisted group had significantly lower ADL scores and a higher rate of institutionalization at discharge than the independent group (P<0.001/3). Table 5 presents the results of the binomial logistic regression analysis, identifying factors that inhibited ADL independence after PT. Table 6 presents the results of the multiple regression analysis, identifying factors that inhibited the BI gain after PT. ADL independence and BI gain were both significantly associated with pre-hospitalization factors such as age, psychiatric conditions, pre-hospitalization residence, and BIs at admission. In contrast, rehabilitation-related variables were not significantly associated with ADL outcomes.

Table 4.  Rehabilitation and outcomes for the independent group (BI ≥85), the partially assisted group (60< BI <85), and the assisted group (BI ≤60).

Variable	Independent (n=596)	Partially assisted (n=50)	Assisted (n=126)	P
Time from hospitalization to start of rehabilitation, days	0 [0–1]	0 [0–0]	0 [0–0]	<0.001***
PT treatment, days	7 [4–14]	11 [4–20]	15 [8–24]	―
PT training implementation rate, %	30.0 [17.9–45.5]	37.1 [19.8–54.3]	42.2 [30.3–53.3]	<0.001***
PT time per treatment day, min	21.3 [20–26.7]	25.5 [20–30.7]	24.2 [20.9–27.7]	<0.001***
OT treatment, days	10 [5–17]	15.5 [6–24.5]	15 [7–24]	―
OT training implementation rate, %	36.4 [25.0–52.0]	41.6 [31.4–52.0]	42.2 [30.1–53.0]	0.08
OT time per treatment day, min	23.6 [20–28.9]	24.9 [20–32.4]	23.5 [20–29.6]	0.28
BI at admission	80 [50–85]	40 [16.3–60]	5 [0–20]	<0.001***
BI at discharge	100 [100–100]	75 [70–80]	40 [15–50]	<0.001***
BI gain	20 [10–45]	37.5 [15–55]	20 [0–35]	<0.01**
Length of hospital stay, days	30 [17–47]	37 [27–67.75]	35 [17.3–69]	<0.01**
Post-discharge residence				<0.001***
Home	521 (87.4)	19 (38.0)	43 (34.1)
Nursing home	75 (12.6)	31 (62.0)	83 (65.9)

Data are shown as median [interquartile range] or number (percentage).

^**P<0.01; ^***P<0.001; Kruskal–Wallis test or Fisher's exact test.

Table 5.  Binomial logistic regression analysis for inhibiting factors of ADL independence after PT.

Variable	Odds ratio	95% Confidence interval	P
Age	0.98	0.96–0.99	0.01*
Sex	0.99	0.58–1.71	0.98
GAF score	1.03	0.98–1.08	0.24
Psychiatric disease	1.29	1.08–1.55	<0.01**
Physical complications	1.01	0.94–1.09	0.75
Pre-hospitalization residence	0.19	0.11–0.35	<0.001***
Time from hospitalization to rehabilitation	1.01	0.96–1.06	0.74
PT training implementation rate	0.82	0.11–5.88	0.84
PT time per treatment day	0.44	0.15–1.29	0.14
OT training implementation rate	1.34	0.26–7.02	0.73
OT time per treatment day	1.16	0.50–2.66	0.73
BI at admission	1.05	1.04–1.06	<0.001***
Length of hospital stay	1.00	0.99–1.01	0.74

^*P<0.05; ^**P<0.01; ^***P<0.001.

Table 6.  Multiple regression analysis of factors inhibiting BI gain following PT.

Variable	B	95% Confidence interval	P
Age	−0.10	−0.18 to −0.01	0.02*
Sex	−0.40	−3.27 to 2.47	0.79
GAF score	0.08	−0.16 to 0.31	0.52
Psychiatric disease	1.15	0.25 to 2.04	0.01*
Physical complications	0.03	−0.36 to 0.42	0.88
Pre-hospitalization residence	−14.51	−18.21 to −10.80	<0.001***
Time from hospitalization to rehabilitation	0.14	−0.22 to 0.51	0.43
PT training implementation rate	−0.82	−10.10 to 8.45	0.86
PT time per treatment day	−0.64	−5.44 to 4.15	0.79
OT training implementation rate	−3.65	−12.67 to 5.37	0.43
OT time per treatment day	1.62	−2.48 to 5.72	0.44
BI at admission	−0.64	−0.69 to −0.60	<0.001***
Length of hospital stay	−0.02	−0.06 to 0.03	0.46

^*P<0.05; ^***P<0.001.

B, Partial regression coefficient.

DISCUSSION

To the best of our knowledge, this is the first study to examine both the inhibitory factors and the effects of PT on physical complications in a psychiatric ward of a general hospital. Based on expert consensus, the necessity of PT was recognized in older, pre-institutionalized patients with musculoskeletal, internal, and organic mental disorders. Regarding the effectiveness of PT, approximately three-quarters of these patients achieved independence. However, 16.3% of patients still required assistance. Pre-hospitalization factors such as advanced age, psychiatric illness, prior institutionalization, and low BI were identified as significant barriers to achieving independence.

There was significant need of PT for patients with musculoskeletal and internal disorders. A previous study has shown that PT improves gait reacquisition, discharge-to-home rates, and gait prognosis in patients with femoral neck fractures and mental illness.¹³⁾ Therefore, PT should be appropriately administered to patients with musculoskeletal disorders, regardless of the presence of psychiatric comorbidities. Furthermore, as mentioned above, mental illness is a significant risk factor for internal disorders. The effectiveness of PT for musculoskeletal and internal disorders is widely known. This study suggests that PT can also contribute to improving ADLs and leading to independence in patients with mental disorders.

In this study, organic mental disorders, including dementia and sequelae of brain injury, were found to be significantly associated with the necessity for PT and to be significant factors inhibiting independence. Research has shown that dementia can hinder reacquisition of gait,¹¹⁾ and that the motor domain of the Functional Independence Measure (FIM) is significantly lower in patients with neurological conditions and psychiatric comorbidities in convalescent wards.¹⁸⁾ Ono et al.¹⁹⁾ reported that in men, ADL scores predicted discharge destination, whereas in women, scores on the Revised Hasegawa Dementia Scale were predictive of discharge destination. Declines in both ADLs and cognitive function often lead to institutional discharge, which further exacerbates physical complications and hinders improvement in functional independence.

In contrast, patients with schizophrenia were significantly less likely to be prescribed PT. According to a World Health Organization cohort study, 54% of patients with schizophrenia achieved remission, experiencing only residual symptoms without functional impairment. Additionally, 77% of middle-aged and older patients with schizophrenia exhibited no adverse symptoms.²⁰⁾ Psychosocial functioning also tends to improve with age, with diminished psychotic symptoms, reduced psychiatric relapses requiring hospitalization, and better self-management.^21,22) Patients admitted with schizophrenia were relatively young and had few physical comorbidities or age-related changes. Therefore, only OT may have been provided instead of PT in these patients.

Other mental disorders that showed significantly high effectiveness of PT included F3 and F5 diagnoses. Almost all patients with F5 disorders were included in the independent group rather than the assisted group. Anorexia nervosa, a typical F5 disorder, has a rising incidence among individuals aged younger than 15 years.²³⁾ Consequently, although disuse syndrome attributed to bed rest and ADL decline caused by muscle atrophy were observed, the relatively lower prevalence of physical comorbidities and aging effects may have contributed to improved ADL outcomes. Patients with F3 disorders are similar to those with F5 disorders, and these patients are significantly more likely to belong to the independent group, with their numbers increasing in the non-older population.³⁾ However, mortality rates for anorexia nervosa and bulimia nervosa are higher than for other psychiatric disorders, with anorexia nervosa having the highest mortality rate among patients hospitalized for psychiatric conditions.²³⁾ Given that deceased patients were excluded from the analysis, this study may not have included some patients with poor improvement in ADLs.

This research demonstrates that aging and institutionalization prior to hospitalization are significantly associated with reduced independence in ADLs. Mental illness impairs mobility, social interaction, daily functioning, and social participation.²⁴⁾ A previous study found that the prevalence of multimorbidity increases substantially with age and is present in most individuals aged 65 years and older. Furthermore, the presence of a mental health disorder becomes more likely as the number of physical morbidities increases.²⁵⁾ Another study on older inpatients with psychiatric disorders reported a significant decline in BI with advancing age.²⁶⁾ Institutionalized patients are often even more socially isolated. Loneliness is associated with impaired social cognition and is a known risk factor for broad-based morbidity across the adult lifespan.²⁷⁾ Aging and social isolation contribute to physical complications, ADL decline, and reduced rehabilitation effectiveness in patients with mental illness.

This study also indicated that ADL independence is significantly influenced by pre-hospitalization status. Therefore, patients should ideally be able to live in the community without experiencing a decline in their ADLs compared to their pre-hospitalization levels. The Ministry of Health, Labour and Welfare of Japan is promoting the establishment of a comprehensive community care system that also addresses mental health needs. Given that patients with both physical and mental disabilities require coordinated care, collaboration among various healthcare professionals is essential. These efforts are expected to improve pre-hospitalization status and outcomes of rehabilitation treatment.

Rehabilitation-related indicators were not significantly associated with ADL independence. In this research, patients who received PT had significantly lower BIs at discharge, despite starting rehabilitation significantly early. This finding may be explained by the fact that rehabilitation was initiated early across the board, with a median interval of 1 day, including patients who did not receive PT. The Japanese Society of Intensive Care Medicine defines early rehabilitation as a set of measures that assist in maintaining, improving, and restoring functions, initiated within 48 h of disease onset, surgery, or acute exacerbation.²⁸⁾ In this study, even patients who did not receive PT underwent early psychiatric OT. This may have contributed to improvements in ADL. Similarly, the lack of association between the timing of rehabilitation initiation and ADL independence may be attributable to the uniformly early rehabilitation implementation, with a median of initiation of 0 days in the independent, partially assisted, and assisted groups.

This study found that patients who received PT had significantly more training time and longer hospital stays than patients who did not receive PT; however, their BIs at discharge remained significantly lower. Even in the multivariate analysis, neither training implementation rate, training time, nor length of hospital stay was found to be associated with ADL independence. A previous study found that patients with motor impairment and mental disorder comorbidities trained in the convalescent wards had FIM scores and home discharge rates comparable to those of patients without mental disorders, although their training time and length of stay were significantly longer.¹⁸⁾ In the present study, the median PT time per treatment day was 22.2 min for patients who received PT and 21.3 min for the assisted group, and the median length of hospital stay was 31 days and 35 days, respectively, both shorter than typical stays in a convalescent ward. Given that greater training time and longer hospital stays could potentially lead to more improvement in ADLs and independence, more intensive rehabilitation programs in psychiatric wards, or the acceptance of patients with psychiatric conditions into convalescent wards, should be considered. To achieve this, it is desirable for rehabilitation staff to understand mental illness and for psychiatric staff to understand rehabilitation. However, this is hindered by the fact that many rehabilitation hospitals and psychiatric hospitals are single-specialty hospitals, and that the medical fees for treating both conditions are not adequately established.

This study has some limitations. First, as a single-center retrospective analysis, the generalizability of the findings may be limited. This study was conducted in the psychiatric ward of a general hospital, and the clinical setting differed from that of a psychiatric hospital in terms of the availability of doctors and physical therapists who could treat physical complications, as well as the calculation of rehabilitation costs. Multicenter prospective studies are necessary to validate these results. Second, the need for PT was determined by expert consensus, without the use of standardized criteria. Therefore, the potential impact of selection bias, such as PT being provided more frequently to elderly patients or patients with lower BIs during hospitalization, cannot be discounted. Rehabilitation programs were also individualized and tailored to each patient based on specialist judgment. Although a research setting requires uniform standards for rehabilitation, ethical considerations make it challenging to ignore individual patient needs. Third, mental disorders and physical complications were categorized broadly, and individual conditions were not examined in detail. Further studies are required to specifically investigate musculoskeletal disorders, internal disorders, organic mental disorders, and schizophrenia, conditions for which significant differences were identified. Fourth, this study did not examine the impact of treatment strategies, including pharmacotherapy. The effects and side effects of psychotropic drugs are important factors influencing physical complications and ADLs and warrant further investigation.

CONCLUSION

This study demonstrated the necessity of PT in psychiatric wards, particularly for patients with advanced age, organic mental disorders, motor disorders, internal disorders, and low ADL scores. The effectiveness of PT was recognized in the majority of these patients, who achieved independence. However, in some patients, pre-hospitalization indicators such as age, type of mental illness, institutionalization, and baseline ADL status inhibited independence, and this study did not provide direct statistical evidence of the effectiveness of PT in these patients. In psychiatric settings, it is effective to consider the need for PT early in hospitalization and to implement it in conjunction with OT. However, the system for patients who require more frequent training and longer training hours remains an issue. Furthermore, a comprehensive approach is needed in the community to prevent a decline in ADLs and isolation. Future studies are expected to more directly clarify the effectiveness of PT in this important clinical setting.