Effects of Physiatrist and Physiotherapist-supervised Therapeutic Exercise on          Physical Function in Frail Older Patients with Multimorbidity

Koki Shiozaki; Makoto Asaeda; Tadaaki Hashimoto; Yasunori Umemoto; Hideyuki Ito; Tomoya Takahashi; Yukihide Nishimura; Fumihiro Tajima

doi:10.2490/prm.20230012

ABSTRACT

Objectives : To ease the burden on care facilities, there is a need to reduce the number of frail older people with multiple chronic and non-communicable diseases (multimorbidity) that require long-term care. We investigated the effects of therapeutic exercise in rehabilitation treatment for older individuals with frailty and multimorbidity.

Methods : We performed 4 weeks of inpatient rehabilitation treatment for frail older patients with multimorbidity. The therapeutic exercise was performed based on whole-body evaluations by qualified physiatrists and physiotherapists. Sixty-minute sessions were conducted twice a day and six times a week in accordance with the American College of Sports Medicine guidelines. Physical functions (grip strength, isometric lower muscle strength, 10-m walking test, and 6-min walking test) were measured at admission and discharge.

Results : Of the 33 patients, 8 were unable to complete physical function evaluations, and 1 patient had a stress fracture of the right calcaneus beyond the rehabilitation time. Twenty-four patients were included in the analysis. Compared to the admission period, significant improvements were observed for all evaluated parameters post-therapy (P < 0.05). However, no significant correlation was found between the number of chronic diseases and changes to physical function (P > 0.05).

Conclusions : Physical function of frail older individuals with multimorbidity improves through rehabilitation treatments supervised by physiatrists and physiotherapists. The number of chronic illnesses and the improvements in physical function were not related. Therefore, rehabilitation treatments for older patients may target both frailty and multimorbidity. Our results will aid in the rehabilitation treatment plans for this demographic.

INTRODUCTION

The aging rate in Japan is 28.4%, with about one-third of the total population being older adults. This is known as a “super-aging society.” Over time, the average life expectancy has increased such that Japan now has the highest percentage of older adults globally. More specifically, older adults aged 75 years and over account for 14.6% of the population.¹⁾ Consequently, the number of people requiring long-term care has increased, with 6.8 million people certified as requiring nursing care or support in 2020.²⁾ Over the past few years, the level of physical activity of older individuals living in the community has decreased because of the coronavirus pandemic.³⁾ Furthermore, the proportion of frail individuals is also increasing,⁴⁾ many of whom are at risk of future disability and the need of long-term care.⁵⁾ Therefore, the number of older individuals requiring long-term care may exceed current estimations and the capacity of the aged care system. As a result, there is a social need to reduce the number of frail older adults who need nursing care.

Frailty is not irreversible and can be changed to “pre-frail” or “robust” by appropriate interventions.⁶⁾ Typically, the recommended interventions are physical activities with resistance training components.⁷⁾ According to the guidelines of the American College of Sports Medicine (ACSM) for physical activity and resistance training in the geriatric population, therapeutic exercise with a load of 40% heart rate reserve (HRR) or higher for aerobic exercise and 40% of one-repetition maximum (1RM) or higher for resistance training is recommended.⁸⁾ In addition, past meta-analyses have shown greater improvement of physical functions in older frail patients through high-intensity training than by low-intensity training.⁹⁾ However, exercising has been reported as risky for older frail people, and the interventions for frail patients in long-term care facilities reportedly tend to increase the incidence of falls.¹⁰⁾ Therefore, there is a legitimate need to optimize the intensity of therapeutic exercise that can be effective while adequately assessing the risks in the rehabilitation treatment of frail older patients. Multimorbidity is a condition where the patient is diagnosed with more than one chronic and non-communicable disease.¹¹⁾ Aging is a risk factor for multimorbidity.¹²⁾ Among older people aged 80–85 years or older living in Japan, 13.2% had frailty syndrome, while 94.7% had multimorbidity.¹³⁾ Therefore, rehabilitation of older individuals with multimorbidity is common. However, in a meta-analysis of therapeutic exercise for people with multimorbidity, improving physical functions had no effect.¹⁴⁾ In addition, there are no reports on therapeutic exercise and rehabilitation treatment for older individuals with frailty and multimorbidity.

Rehabilitation under the guidance of a physiatrist and registered therapist, resembling the typical postoperative rehabilitation regimen, can improve the Functional Independence Measure in patients with cerebrovascular disease from the acute phase to the maintenance phase.¹⁵^,¹⁶⁾ Therefore, therapeutic exercise would be useful for patients with frailty and multimorbidity because it is effective in patients with cerebrovascular disease who are prone to multiple chronic diseases. Hence, in this study, we hypothesized that rehabilitation by a physiatrist and therapist improves the physical function of frail older people with multimorbidity. Therapeutic exercise in frail older patients with multimorbidity was performed under the supervision of a physiatrist and physiotherapist. We assessed the physical function of participants before and after rehabilitation treatment and evaluated the differences.

MATERIALS AND METHODS

Study Participants

This is a case-series study wherein the participants were outpatients at a regional general hospital. The patients were selected for inpatient rehabilitation by a physiatrist to address the subjective decline of activities of daily living (ADL) between September 2020 and March 2022. All patients had osteoarthritis (lower limbs, spine). The Japanese version of the Cardiovascular Health Study (J-CHS) criteria were used to determine participant frailty (Table 1). The J-CHS criteria include five items; if three or more criteria are matched, the patient condition is judged as “frailty,” whereas the matching of one or two criteria gives “pre-frail,” and no criteria matching gives “robust”.¹⁷⁾ The exclusion criteria were as follows: participants under 75 years of age, classified “pre-frail” or “robust” by J-CHS, and unable to gradually increase exercise load because of severe motor or mental dysfunction. Multimorbidity was defined as the “presence of multiple diseases in one individual” based on a previous study,¹⁸⁾ and the multiple diseases were the 20 diseases most included in the review by Pados-Torres et al.¹⁹⁾ All of the participants in this study had osteoarthritis and had at least one other disease, which was defined as multimorbidity. All patients were informed of the proposed regimens and associated risks of therapeutic exercise in accordance with the Declaration of Helsinki and the study was approved by the Ethics Committee of the institution to which the authors belong (Research Ethics Committee, Faculty of Wakayama Health Care Sciences, Takarazuka University of Medical and Health Care, Approval number: 221012-1). Oral and written informed consent was obtained for participation in the program and for use of the data.

Table 1. Japanese version of the Cardiovascular Health Study (J-CHS) criteria

Item	Description
Shrinking	Unintentionally lost 2 kg or more in past 6 months
Weakness	Grip strength <28 kg in men or 18 kg in women
Exhaustion	In the past 2 weeks, have you felt tired without reason?
Slowness	Gait speed <1.0 m/s
Low activity	Engages in moderate levels of physical exercise or sports aimed at health Engages in low levels of physical exercise aimed at health

Rehabilitation Program

On the first day of hospitalization, the physiatrist performed a rehabilitation diagnosis based on a whole-body evaluation (higher brain dysfunction, cranial nerves, musculoskeletal system, nutrition, and ADL). Therapeutic exercise was initiated on the same day. The therapeutic exercise was performed based on the whole-body evaluation by the physiatrist and in accordance with the ACSM guidelines.⁸⁾ Each session lasted 60 min, and two sessions were held each day, six times a week. The physiotherapist evaluated the weekly exercise intensity, which gradually increased based on the assessments. The therapeutic exercise consisted of aerobic, resistance, flexibility, and ADL training. Warm-ups were performed before each session (Table 2).

Table 2. Summary of frequency, intensity, time, and type of exercise therapy used in our study based on ACSM guidelines

	Aerobic exercise	Resistance training	Others
Frequency	2 sessions/day, 6 days/week	3–6 sessions/week	6 days/week
Intensity	Target range of heart rate was: 0.4 × HRR + HRrest to 0.6 × HRR + HRrest	40–60% 1RM
Time	20–30 min	15 repetitions/set, 2–4 sets/session
Type	Bicycle ergometer (Recumbent ergometer or arm-crank ergometer)	Medical fitness machine (Weltonics, Minato Medical): leg press, leg extension, hip abduction, and rowing	Flexibility exercise, and ADL training (gait, stair ascent/descent)

A bicycle ergometer was used for aerobic exercise. In the cases where the bicycle ergometer could not be ridden because of patient discomfort or limited range of motion of the limb joints, a recumbent ergometer or arm-crank ergometer was used. Karvonen methods were used to determine the aerobic exercise load, and a HRR of 40%–60% was targeted based on the resting heart rate (HRrest) and maximum heart rate (HRmax; 220−age).²⁰⁾ The target range of heart rate was 0.4 × HRR + HRrest to 0.6 × HRR + HRrest. If the target heart range could not be attained, the load was reduced while sustaining the cycling rate (40–50 rpm).

For resistance training, a medical fitness machine (Weltonics series, Minato Medical Science, Osaka, Japan) was used to perform leg press, leg extension, and hip abduction. This medical fitness machine has one-repetition maximum (RM) measurement mode and can change isometric exercise by mechanical fixation. Therefore, the value obtained by measuring the maximum pressure in that state was defined as 1RM in this study. One RM was measured before the session started and was performed with an exercise load of 40%–60% of 1RM. If the target RM could not be attained, the load amount was reduced while maintaining the number of repetitions.

Measurement of Physical Function

The muscle strength (grip strength and isometric muscle strength by medical fitness machine) and walking ability [10-m walking test, 6-min walking test (6MWT)] were measured at admission (baseline) and discharge (after therapy). The grip strength was measured in a standing position, and the higher values of two left and right measurements were recorded. The isometric muscle strength was performed in the following postures: 1) leg press: sitting with hip and knee flexion; 2) leg extension: sitting with the knee at 90-degree flexion; 3) hip abduction: sitting with hip at 10-degree abduction. The isometric muscle strength was obtained by dividing the higher value of the two measurements by the patient’s body weight.

The 10-m walking test measured patients’ walking speed over 10 m at both comfortable and maximum effort speeds. We permitted the use of walking aids such as canes or walkers. The same walking aid was used during the initial test at admission and at the final evaluation. The 6MWTs were conducted in accordance with the American Thoracic Society statement,²¹⁾ and the distance covered during 6 min of walking was measured (the use of walking aids was permitted).

Data Analysis

The differences in muscle strength and walking ability at baseline and after therapy were examined using paired t-tests. All the statistical analyses used SPSS ver. 27.0 (IBM, Tokyo, Japan), and the significance level was less than 5%. Using Pearson’s correlation coefficient, we investigated the relationship between the change (Δ) in muscle strength, walking ability, and the number of chronic illnesses. No statistical sample size calculations were conducted because this study is a case-series design. However, we analyzed the effect size (Cohen’s d) based on three categories, namely, 0.2 = small, 0.5 = moderate, and 0.8 = large, and power with 5% significance using SPSS.²²⁾ In addition, analysis of variance (ANOVA) was used to examine changes over time for the load (watts) and time (min) of aerobic exercise, and the load of resistance training (N), although repeated ANOVA was not used because some participants changed their exercise style during the exercise.

RESULTS

Between September 2020 and March 2022, 33 patients underwent inpatient rehabilitation. Of these, 8 were unable to complete the assessments for muscle strength and walking ability at baseline and after therapy because of functional deterioration, and 1 patient had a stress fracture of the right calcaneus beyond the rehabilitation time. A total of 24 patients were included in the analysis (Table 3). No adverse events (falls, fractures) requiring medical treatment during rehabilitation occurred, and there were no exacerbations of comorbidities. The average number of days the patients undertook therapeutic exercise was 26.5 ± 5.8 days, and the average exercise time was 48.9 ± 9.4 h. On completion of the therapy, all patients returned to the same home or care facility in which they resided before admission. Regarding the extent of long-term care required, 33% (8 patients) were not certified as requiring long-term care or support, and 29% (7 patients) needed long-term care. A total of 14 fall fractures occurred before hospitalization (Table 3). Heart failure was the most common contributor to multimorbidity (10 patients, 42%), followed by hypertension (8 patients, 33%) and cerebral infarction (7 patients, 29%); the average number of chronic diseases per patient was 4.3. Of the participants with knee osteoarthritis, 1 patient underwent total knee arthroplasty (after 3 years), and 4 patients were prescribed analgesics while hospitalized (2 used non-steroidal anti-inflammatory drugs, 1 used serotonin-norepinephrine reuptake inhibitors, and 1 used pregabalin). Medication was the same as prescribed at home, and no additional medication was administered during this hospitalization. In addition, no joint injections were administered during hospitalization for participants who had received joint injections at home. Patients with lumbar spinal canal stenosis also continued medication and did not receive an injection. Regarding the medication, 7 patients who had chronic heart failure used loop diuretics, and 3 patients used angiotensin II receptor blockers. None of the patients used beta-blockers or changed medications compared to the home prescription. None interrupted or discontinued the therapeutic exercise because of worsening heart failure, angina symptoms, or abnormal pulse.Increases in workload in resistance training and aerobic exercise (bicycle ergometer) are shown in Figs. 1 and 2. Some participants were unable to ride a bicycle ergometer because of lower back pain, knee joint pain, or limited range of motion in the knee joint. Participants who could not ride a bicycle performed stair-climb training (160–300 steps per session). In the trajectories excluding those without other continuous records, resistance training load and time of aerobic training (per session) were gradually extended. There were no significant differences for any exercise loads or times (P > 0.05, Figs. 1, 2). Almost all participants showed physical functional improvement (Fig. 3). While comparing the evaluations at admission and after therapy, significant improvements were observed for all the evaluated parameters (P < 0.05, Table 4). The effect size was 0.5 or higher, and power was 0.8 or higher for all the evaluated parameters except the comfortable gait speed.

Table 3. Summary of participant characteristics

	n=24
Age (years)	82.3±4.6
Sex (male/female)	8/16
Body mass index (kg/m²)	20.5±3.3
Musculoskeletal disease	Knee osteoarthritis: 12 Spondylosis: 12 Spine deformity (scoliosis, kyphosis): 6 Stenosis: 7
Chronic disease	Atrial fibrillation: 3 Angina pectoris: 6 Cerebral infarction: 9 Diabetes: 2 Chronic heart failure: 13 Hypertension: 14 Osteoporosis: 8 Parkinson’s disease: 2 Tachyarrhythmia: 1
Average number of chronic illnesses	4.3 ± 1.9 (min: 1, max: 7)
Care level	Not received: 8 Requiring support 1: 1 Requiring support 2: 4 Requiring long-term care 1: 7 Requiring long-term care 2: 2 Requiring long-term care 3: 2 Requiring long-term care 4: 0 Requiring long-term care 5: 0
History of fracture	Femoral neck: 3 Distal radial: 4 Vertebral compression fracture: 7
Rehabilitation period (days)	26.4 ± 5.8
Exercise time (h)	48.9 ± 9.4

Care level: requiring support: a person can carry out daily activities by themselves, but some support is required (levels 1 to 2); requiring long-term care: difficulty in performing basic actions in daily life on one's own and requiring nursing care (levels 1 to 5).

Fig. 1.

Trajectories of workload in resistance training (only cases with medical records).

Fig. 2.

Trajectories of workload (watts and minutes) on bicycle ergometer (only cases with medical records).

Fig. 3.

Individual performances in physical function and intra-individual differences between admission and after therapy. (A) Leg press, (B) leg extension, (C) hip abduction, (D) 10-m walking test (comfort), (E) 10-m walking test (maximum effort), (F) 6MWT.

Table 4. Comparison of evaluated progress parameters between admission and after therapy

Parameter		At admission	After therapy	Δ	P value	Cohen’s d	Power
Grip strength (kg)		16.9±4.5	18.5±4.4	2.2±1.1	<0.001	0.872	0.983
Isometric muscle strength (N/kg)	Leg press	10.0±3.1	13.8±4.1	5.0±3.0	<0.001	0.857	0.980
	Leg extension	5.1±1.6	6.5±1.7	1.9±1.2	<0.001	1.076	0.999
	Hip abduction	6.0±1.9	6.8±1.7	1.5±1.0	0.001	0.813	0.968
10-m walk test (s)	Comfort	17.9±9.9	14.0±5.6	5.0±8.1	0.032	0.465	0.588
	Maximum effort	15.4±7.9	11.8±4.9	4.0±6.0	0.008	0.598	0.800
6MWT (m)		234.6±80.8	285.9±97.2	56.4±45.1	<0.001	1.002	0.997

Data given as mean ± standard deviation.

When assessing the multimorbidity and physical function, no significant correlation was found between the number of chronic disease diagnoses and changes to muscle strength or walking ability (P > 0.05, Table 5). In addition, linearity was observed for all items (>0.05).

Table 5. Correlation between number of chronic illnesses and improvements in physical function

Factor	Pearson’s r	P value	Linearity
Δ Grip strength	0.026	0.906	0.818
Δ Leg press	−0.101	0.647	0.178
Δ Leg extension	−0.087	0.731	0.180
Δ Hip adduction	0.283	0.240	0.511
Δ 10-m walking test (comfort)	−0.153	0.496	0.599
Δ 10-m walking test (maximum effort)	−0.186	0.408	0.792
Δ 6MWT	−0.124	0.564	0.670

DISCUSSION

This study affirmed that the physical function of frail older patients with multimorbidity is improved by rehabilitation treatment supervised by physiatrists and physiotherapists. No relationship was found between the number of chronic illnesses and improvements in physical function. These results have not been previously reported.

Resistance band exercises for frail older patients can suppress frailty; however, there is only limited evidence that this therapy increases grip strength or knee extension muscle strength when compared with normal care.²³⁾ In the present study, the grip strength and isometric muscle strength improved post-rehabilitation treatment. A previous study showed that grip strength is associated with gradual muscle weakness (sarcopenia) but may not accurately reflect a functional recovery in frail older patients.²⁴⁾ Therefore, improvements in grip strength do not necessarily prevent frailty. Our results align with the previous results. However, we also found that the knee extension muscle strength increased significantly. In the above meta-analysis, an increased knee extension muscle strength was observed after high-load whole-body training (combined leg press, leg press, and Ped-deck with 50% to 75% of 1RM).²⁴⁾ Therefore, physical function can be improved even in frail 80-year-old patients through exercise regimens of appropriate intensities in an environment where sufficient medical management is available, such as an acute care hospital. Because the supervised program can improve balance ability and muscle strength for community-dwelling older people,²⁵⁾ frail older people can also benefit from supervised and medically managed therapeutic exercise to improve their physical function.

The hip abduction muscle strength can be used as a criterion for frailty,²⁶⁾ and can be used to predict walking speed in frail patients after discharge (brought about by increased muscle strength because of leg extension).²⁷⁾ Moderate exercise interventions for community-dwelling older people improve hip abduction muscle strength.²⁸^,²⁹⁾ However, in a previous study, 3 months of low-intensity exercise in frail older people did not improve hip abduction muscle strength.³⁰⁾ However, we consider that the exercise intensity used in the present study under the supervision of a physiatrist and physiotherapist allowed frail older patients with multimorbidity to increase their hip abduction muscle strength.

Walking ability was improved by managed and supervised exercise intervention for frail patients, similar to previous studies.²⁵⁾ The 6-min walking distance increased because of the therapeutic exercise, which combines endurance and resistance training in older people.³¹⁾ The minimal clinically significant improvement of the 6-min walking distance in frail patients in a previous study was 17.8 m.³²⁾ The results of our study showed a drastic effect (mean Δ=56.4±45.1 m). Therefore, as reported in previous studies, we found that the rehabilitation regimen recommended by the physiatrist and physiotherapist was effective for improving mobility.

Therapeutic exercise for multimorbidity improves physical functions.¹⁴⁾ The walking time in the 10-m walking test, walking distance for the 6MWT, and leg muscle strength were also significantly improved in our study. Furthermore, all the patients in this study were frail and were able to demonstrate that therapeutic exercise in frail older patients with multiple comorbidities is effective regardless of the number of chronic diseases. Frail patients are more likely to have multimorbidity (72%) than vice versa (16%).³³⁾ It has also been reported that with age, the likelihood of having multimorbidity becomes extremely high.¹²⁾ Therefore, therapeutic exercise for frail patients with multimorbidity must be recommended. Unsupervised home-based exercises improve adherence in frail patients but may be less effective than supervised exercises for improving physical function.³⁴⁾ Therefore, our results suggest that stronger recommendations will emerge for therapeutic exercise under medical management by physiatrists and physiotherapists because of the increase in the aging population.

The small number of cases in this study and its restriction to a single center are limitations of the study. In this study, we used daily rehabilitation treatment under the supervision of physiatrists and physiotherapists in an inpatient setting. However, it is not clear whether this approach is superior to outpatient or home-visit rehabilitation for older adults with regard to long-term care insurance. Furthermore, the intervention period, at 26.5 days, was very short. In addition, given limited information on the evaluation and definition of each chronic disease (e.g., the left ventricular ejection fraction by echocardiography, electrocardiogram, or lung function test), the effects of the severity of multimorbidities were unknown. The participants of this study were frail older adults with decreased ADL; however, the effects of the duration of the disease and previous activities were not objectively shown by indicators such as the Barthel index. Furthermore, because retrospective studies have not been able to investigate when the patient had frailty or multimorbidity, it is unclear whether improvement in physical function is caused by improvement of heart condition or the duration of the disease. Therapeutic exercise improves exercise tolerance and lower muscle strength in frail patients with chronic heart failure,³⁵⁾ and the duration of the disease affects the degree of improvement. Therefore, it is possible that more detailed grouping by stable duration can be used to examine more effective therapeutic exercises for frail older patients with multimorbidity.

CONCLUSION

Rehabilitation treatment, including therapeutic exercise, was performed for frail older patients with multimorbidity under the supervision of a physiatrist and physiotherapist. The physical function (muscle strength and walking ability) improved regardless of the number of chronic illnesses. In the future, we expect that rehabilitation treatment will be recommended for older patients with frailty and multimorbidity. Our results can be used as the basis for recommendations for safe and effective therapeutic exercise for frail older adults with multiple chronic diseases.

ACKNOWLEDGMENTS

We thank all the staff at Hashimoto Hospital involved in caring for the hospitalized patients.

CONFLICTS OF INTEREST

The authors have no conflicts of interest to declare.

REFERENCES

Corresponding author

Register with J-STAGE for free!