Factors Associated With Physical Restraint Use in Older Adults With Dementia in Geriatric Care Facilities During the COVID‐19 Pandemic in Japan

ABSTRACT

Objectives

We aimed to identify factors associated with the use of physical restraints in geriatric care facilities housing older adults with dementia during the COVID‐19 pandemic in Japan.

Methods

A cross‐sectional anonymous online survey was conducted from January to February 2023 in special nursing homes (SNHs) for older adults with moderate to severe dementia and group homes (GHs) for those with mild to moderate dementia. The survey gathered information on facility characteristics, the presence of residents with dementia infected during COVID‐19 clusters, and the use of physical restraints. Additional items addressed challenges encountered during a cluster outbreak, difficulties in managing residents with dementia infected with COVID‐19, measures implemented for these residents, and considerations when residents were unable to isolate in their rooms owing to wandering. Multiple logistic regression analysis was used to identify factors associated with physical restraint use.

Results

Data from 286 SNHs and 151 GHs were analyzed, all of which housed residents with dementia infected during COVID‐19 clusters. Among them, 49 SNHs (17.1%) and 14 GHs (9.3%) reported using physical restraints. In SNHs, multiple logistic regression analysis identified wandering, worsening dementia symptoms, and ongoing infections as significant factors associated with restraint use. In GHs, difficulty communicating the situation to other residents and families was notably associated with restraint implementation.

Conclusions

The use of physical restraints in facilities caring for older adults with moderate to severe dementia may be associated with worsening dementia symptoms during the COVID‐19 pandemic in Japan. These findings underscore the need for improved staff training and the development of care strategies aimed at minimizing restraint use during infectious disease outbreaks.

Key Points

• Physical restraints were more frequently used in special nursing homes than in group homes for older adults with dementia during COVID‐19 cluster outbreaks in Japan.

• In special nursing homes, restraint use was associated with ongoing new infections, wandering behaviors, and worsening dementia symptoms.

• In group homes, difficulties explaining pandemic‐related restrictions to other residents and families were significantly linked to restraint use.

• Our findings highlight the need for targeted strategies to manage behavioral symptoms such as wandering while minimizing restraint use during future pandemics.

1. Introduction

The coronavirus disease 2019 (COVID‐19) spread globally, resulting in unfavorable outcomes for many individuals. The mortality rate among particularly vulnerable older adults was high, due in part to the rapid transmission of COVID‐19 in geriatric care facilities and hospitals, where many older adults resided [1, 2]. To curb the spread of COVID‐19, many such facilities implemented social distancing measures and restricted family visits and outings [3]. However, these interventions contributed to declines in cognitive function and mental health, accompanied by new behavioral and psychological symptoms of dementia [4].

The use of physical restraints on older adults with dementia may have increased owing to challenges in maintaining high‐quality care during the pandemic. Shortages of personal protective equipment and an overwhelming number of inpatients placed immense pressure on medical staff and healthcare workers [5, 6, 7], potentially leading to increased use of restraints for older patients. Older patients subjected to physical restraints may experience adverse outcomes, including restraint‐related injuries (e.g., asphyxiation or chest compression), immobility‐related complications (e.g., pulmonary embolism or aspiration), and functional decline [8, 9, 10].

Several studies have reported the use of physical restraints on older adults with dementia during the pandemic. A survey conducted in July 2020 among hospitals in Japan that admitted patients with COVID‐19 found an increase in the use of restraints for older patients with dementia [11]. Internationally, an Italian national survey of nursing homes reported that 62.1% of facilities—where approximately 26% of residents were patients with dementia—implemented physical restraints after the onset of the pandemic [12]. In Singapore, a survey conducted during both pre‐pandemic and pandemic periods found that nearly half of caregivers reported using restraints when caring for older adults with dementia at home [13]. However, no studies have investigated the use of restraints in geriatric care facilities in Japan. Furthermore, factors associated with restraint use in these settings remain unclear.

In this study, we aimed to identify factors associated with the use of physical restraints in geriatric care facilities accommodating older adults with dementia during the COVID‐19 pandemic. We hypothesized that difficulty in behavioral control among older patients with dementia was associated with restraint use. The findings from this study will inform strategies for managing older adults with dementia during future outbreaks of contagious infectious diseases such as COVID‐19.

2. Materials and Methods

2.1. Study Design and Ethical Consideration

A cross‐sectional design was employed in this study. An anonymous online survey was administered to medical and nursing care facilities between January and February 2023, during Japan's eighth wave of COVID‐19. The survey was conducted by Hiroshima University in collaboration with the COVID‐19 response team of the Japan Geriatrics Society. As the survey was anonymous and did not collect personal information or facility names, ethics committee approval was not required according to the ethical guidelines of the Ministry of Health, Labor and Welfare [14]. The study adhered to the principles of the Declaration of Helsinki (2013 version).

2.2. Participants

Participants were provided with a link to an online questionnaire developed using Google Forms (Google Plex, Mountain View, CA, USA). Before completing the survey, they were informed of the study's purpose, procedures, and privacy policy. The survey addressed the situation of residents with dementia in medical and nursing care facilities from the onset of the COVID‐19 pandemic to the present. One representative per facility was asked to respond. Participants were advised that completing the survey indicated their consent to participate. They then completed the anonymous questionnaire.

2.3. Measurements

The study focused on facilities that serve a large number of older adults with dementia. In Japan, these include special nursing homes (SNHs) [15] and group homes (GHs) for older adults with dementia [16]. Participants were asked about the type of facility, its capacity, the proportion of older adults with dementia or suspected dementia, and whether any residents with dementia had been infected during a COVID‐19 cluster outbreak. Information was collected on the use of physical restraints, including types used. Additional data were obtained regarding challenges encountered during a cluster outbreak; difficulties in managing residents with dementia infected with COVID‐19; measures implemented for these residents; and considerations when residents were unable to remain in their rooms owing to behavioral and psychological symptoms, such as wandering. These items were assessed through single‐ and multiple‐answer questionnaire items. Detailed content is available in Supporting Information: Table S1.

2.4. Statistical Analysis

Categorical variables were analyzed using the χ2‐test or Fisher's exact test. To identify factors significantly associated with the use of physical restraints, a multiple logistic regression analysis with the forced entry method was performed. Independent variables were those that significantly differed between groups with and without restraint use. Statistical analyses were conducted using IBM SPSS Statistics 28. Statistical significance was set at p < 0.05.

3. Results

A total of 725 responses were collected, comprising 389 SNHs and 336 GHs. Data from facilities were included in the analysis if they had any residents with dementia among individuals infected with COVID‐19 during a cluster outbreak: 286 SNHs (74.1%) and 151 GHs (44.9%).

Among SNHs and GHs, 49 (17.1%) and 14 (9.3%) facilities, respectively, reported the use of physical restraints. The specific types of physical restraints implemented are listed in Supporting Information S1: Table S2. SNHs that employed physical restraints tended to have significantly larger capacities (Figure 1).

FIGURE 1.

Facility capacity by type and use of physical restraints.

No significant difference was observed in the proportion of older adults with dementia or suspected dementia between SNHs that did and did not use physical restraints (p = 0.529). The distribution of facilities by dementia prevalence categories was similar across both groups: 0%–25%, 1 facility (2.0%) versus 3 facilities (1.3%); 25%–50%, 1 facility (2.0%) versus 18 facilities (7.6%); 50%–75%, 14 facilities (28.6%) versus 68 facilities (28.7%); and 75%–100%, 33 facilities (67.3%) versus 148 facilities (62.4%).

In both SNHs and GHs using physical restraints, the response option “considered with the use of physical restraints” was significantly more common when managing residents with dementia infected with COVID‐19 who were unable to remain isolated in their rooms owing to behavioral and psychological symptoms (Figure 2).

FIGURE 2.

Considerations when residents were unable to remain in their rooms due to behavioral and psychological symptoms, such as wandering.

Compared to facilities not using physical restraints, SNHs that used restraints reported significantly higher rates of difficulties such as “Zoning was difficult due to the facility's structural layout” and “The infection did not subside quickly, and new infections continued to occur for an extended period” during cluster outbreaks. In GHs using physical restraints, “Difficulty explaining the situation to other residents and their families” was identified as a significant issue (Figure 3).

FIGURE 3.

Challenges encountered during a cluster outbreak.

SNHs employing physical restraints also reported significantly higher proportions of “Difficulty isolating residents due to wandering” and “Worsening of dementia symptoms and behavioral and psychological symptoms” when managing residents with dementia infected with COVID‐19 (Figure 4). No significant differences were observed between facilities with and without the use of physical restraints in the measures implemented during such cases (Figure 5).

FIGURE 4.

Difficulties in managing residents with dementia infected with COVID‐19.

FIGURE 5.

Measures implemented when managing dementia residents infected with COVID‐19.

The results of multiple logistic regression analysis using the forced entry method, with use of physical restraints as the dependent variable, are presented in Table 1. In SNHs, significant independent variables included “Considered with the use of physical restraints,” “The infection did not subside quickly, and new infections continued to occur for an extended period,” “Difficulty isolating residents due to wandering,” and “Worsening of dementia symptoms and behavioral and psychological symptoms.” In GHs, “Considered with the use of physical restraints” and “Difficulty explaining the situation to other residents and their families” were identified.

TABLE 1.

Results of multiple logistic regression analysis using the forced entry method, with the use of physical restraints as the dependent variable.

	Special nursing home (n = 286) a				Group home for older adults with dementia (n = 151) b
		95% CI				95% CI
Independent variables	OR	Lower limit	Upper limit	p	OR	Lower limit	Upper limit	p
Challenges encountered during a cluster outbreak: Zoning was difficult due to the structure of the facility	1.160	0.474	2.842	0.745	—	—	—	—
Challenges encountered during a cluster outbreak: The infection did not subside quickly, and new infections continued to occur for an extended period	2.603	1.051	6.444	0.039	—	—	—	—
Challenges encountered during a cluster outbreak: Difficulty explaining the situation to other residents and their families	—	—	—	—	5.959	1.234	28.782	0.026
Difficulties in managing residents with dementia infected with COVID‐19: Difficulty isolating residents due to wandering	5.862	1.749	19.649	0.004	—	—	—	—
Difficulties in managing residents with dementia infected with COVID‐19: Worsening of dementia symptoms and behavioral and psychological symptoms	3.326	1.118	9.891	0.031	—	—	—	—
Considerations when residents were unable to remain in their rooms due to behavioral and psychological symptoms: Considered with the use of physical restraints	31.115	12.395	78.110	<0.001	46.546	10.223	211.918	<0.001

Abbreviations: CI, confidence interval; OR, odds ratio.

^a Adjusted for facility capacity and the percentage of older adults with dementia or suspected dementia.

^b Adjusted for facility capacity.

Supporting Information S1: Tables S3–S5 present a comparison of variables associated with and without the use of physical restraints when responding to residents with dementia infected with COVID‐19 who were unable to remain in their rooms due to behavioral and psychological symptoms. No significant differences were found regarding difficulties during cluster outbreaks. However, SNHs using physical restraints reported a significantly higher proportion of “Worsening of dementia symptoms and behavioral and psychological symptoms” as a difficulty in managing residents. Additionally, “Signs and posters were placed in prominent areas, such as on room doors, to remind residents to exercise caution” was significantly more common in SNHs using physical restraints as part of their management measures.

4. Discussion

We examined factors associated with the use of physical restraints in geriatric care facilities housing older adults with dementia during the COVID‐19 pandemic. In SNHs, significant factors were “Difficulty isolating residents due to wandering (wandering),” “Worsening of dementia symptoms and behavioral and psychological symptoms (worsening dementia symptoms),” and “The infection did not subside quickly, and new infections continued to occur for an extended period (ongoing new infection)” contrasting with “Difficulty explaining the situation to other residents and their families” in GHs.

Japan has long sought to reduce physical restraint use among older adults. In 2001, the Ministry of Health, Labor and Welfare introduced the “Guideline Toward Zero Physical Restraints” for nursing care and healthcare sectors [17]. In 2005, reforms aligned with the Long‐Term Care Insurance Act emphasized “maintaining dignity” for older adults [17]. In 2021, operational standards were revised to mandate committee formation, guideline development, and staff training sessions to reduce restraint use [17]. Despite these efforts, physical restraints may have been difficult to avoid during the COVID‐19 pandemic, especially for older adults with dementia. These individuals are particularly susceptible to COVID‐19, may forget or not comprehend instructions on social distancing and mask use, and may require oxygen support due to conditions such as happy hypoxemia [18, 19]. A recent study in Japan reported that older adults with dementia infected with COVID‐19 and requiring nursing care were significantly more likely to be physically restrained in hospitals [11]. Internationally, two studies reported physical restraint use in 46.9% of caregivers of older adults with dementia at home in Singapore [13] and in 62.1% of nursing homes in Italy, where 26% of older adults with dementia were restrained [12]. Unlike these prior studies, to our knowledge, this study is the first in Japan to investigate the use of physical restraints in care facilities specifically housing older adults with dementia, in relation to infection prevention measures.

Worsening dementia symptoms were frequently observed among residents with COVID‐19 [20]. Our study identified three major factors associated with physical restraint use in SNHs, which typically house older adults with moderate to severe dementia [15]: ongoing new infections, wandering, and worsening dementia symptoms. Although the rate of new infections did not differ between facilities that did or did not use restraints, the rate of worsening dementia symptoms was significantly higher in facilities that implemented restraints. While the incidence of wandering was not significantly different between groups, worsening symptoms could have increased wandering behavior, indirectly contributing to restraint use [21]. Future research should investigate this relationship. Additionally, our findings indicate that staff burden remained high regardless of restraint use. Considering the persistent pandemic and elevated staff burden, physical restraints may have been used to help control the spread of COVID‐19.

By contrast, GHs showed no significant association between restraint use and worsening dementia symptoms. Only 14 GHs (9.3%) out of 151 used physical restraints, compared to 49 (17.1%) out of 286 SNHs. GHs primarily accommodate older adults with mild to moderate dementia [16]. Symptom progression in these individuals may be subtle, involving cognitive, emotional, and functional changes—such as memory decline, mild behavioral symptoms, and early functional impairment [22]. Furthermore, residents in GHs typically lack access to regular specialist consultations or full‐time physicians [23]. Consequently, dementia progression may go unnoticed, making it difficult for staff to identify cognitive or behavioral decline. Nevertheless, a lack of detection does not imply that symptoms did not worsen in restrained individuals with mild dementia. This interpretation aligns with previous studies reporting that physical restraints may exacerbate dementia symptoms [24, 25].

In GHs, “Difficulty explaining the situation to other residents and their families” was significantly associated with restraint use. The GH philosophy promotes a “free and homely life,” encouraging frequent family involvement [16]. However, pandemic‐related visitation restrictions limited communication about the need for physical restraints, complicating efforts to obtain understanding and consent from residents and families.

These findings have important implications, particularly regarding wandering in facilities housing older adults with moderate to severe dementia during COVID‐19 outbreaks. To reduce wandering, experts have recommended involving residents in meaningful activities, regularly screening for delirium and emotional changes related to isolation, and using environmental tools such as door alarms and tagged bracelets [26, 27, 28]. Reducing media exposure to pandemic news was also advised [26, 29]. Globally, care settings adopted measures such as placing orientation signs in residents' rooms [30]. In cases where wandering posed substantial risks, limited and careful use of restraints has been considered [31]. To minimize restraint use, we recommend that caregivers and medical staff undergo targeted training in managing wandering behaviors during pandemics, equipping them with strategies to address these challenges.

This study had several limitations. First, as a cross‐sectional study, it could not establish causal relationships between the identified factors and restraint use. Second, generalizability is limited as the findings reflect only SNHs and GHs in Japan and may not apply to other countries. Although the use of physical restraints was examined during the COVID‐19 pandemic, our data may not fully capture the patterns of restraint use throughout the pandemic. One reason is that the survey period (January–February 2023, during Japan's eighth COVID‐19 wave) did not coincide with the peak of COVID‐19 cases in Japan, which occurred in July to September 2022 during the seventh wave, according to the MHLW [32]. Restraint use during the period may also have been influenced by healthcare worker shortages and government restrictions. Furthermore, the challenges of isolating older adults with dementia during the period may have increased the likelihood of physical restraint use; possibly, the findings also do not reflect patterns of restraint use during the non‐COVID period. Third, a specific respondent type was not designated for this survey to enhance its feasibility. Because staffing requirements differ by facility type—for example, physicians are required in SNHs but not in GHs—the composition of respondents likely varied across facilities [33]. The results, therefore, may differ depending on whether the respondent was a doctor, nurse, or administrative staff member. However, because facility staff were generally familiar with the MHLW's guidelines for the Elimination and Prevention of Physical Restraint for Workers in Long‐Term Care Facilities and Services [17], the impact of this variation on the study results is likely minimal. Fourth, the use of questionnaires limited the precision of data collection regarding the application and duration of restraints, extent of wandering, and progression of dementia symptoms. Furthermore, the institutional‐level factors examined in this study were largely subjective, whereas restraint use is often more directly associated with patient‐ or caregiver‐level factors [34]. More objective indicators, such as doctor–patient ratio, number of patients with dementia with COVID‐19, proportion of residents restrained, or use of chemical restraints, were not available in this dataset. Therefore, although these findings provide new insights into institutional‐level influences, our results should be interpreted cautiously and complemented by future research that integrates both subjective and objective measures across patient, caregiver, and institutional domains.

In conclusion, the use of physical restraints during the COVID‐19 pandemic in geriatric care facilities housing older adults with moderate to severe dementia in Japan may be linked to worsening dementia symptoms. Strategies aimed at managing wandering behaviors are essential to reducing restraint use.

Author Contributions

Study design and data acquisition: H.J., S.I., Y.I., J.T., K.H., and M.A. Data analysis and interpretation: H.J. Manuscript preparation: H.J. Critical revision of the manuscript and approval of the final version: All authors.

Funding

The authors have nothing to report.

Ethics Statement

As the survey was anonymous and did not collect personal information or facility names, ethics committee approval was not required according to the ethical guidelines of the Ministry of Health, Labor and Welfare [14]. The study adhered to the principles of the Declaration of Helsinki (2013 version).

Consent

Participants were advised that completing the survey indicated their consent to participate.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Table S1: Survey items on dementia care and COVID‐19 responses in medical and nursing Care Facilities.

Supporting Information S1

Jung, Hungu , Akishita Masahiro, Iwamoto Yuji, Tanabe Junpei, Hirohama Kenta, and Ishii Shinya. 2025. “Factors Associated With Physical Restraint Use in Older Adults With Dementia in Geriatric Care Facilities During the COVID‐19 Pandemic in Japan,” International Journal of Geriatric Psychiatry: e70174. 10.1002/gps.70174.

Data Availability Statement

The authors have nothing to report.