Dear Editor, Care facilities for older adults, which house many older adults, experienced clusters of coronavirus disease 2019 (COVID-19).1, 2 Among these facilities, special elderly nursing homes (SENHs) for older adults and group living homes (GHs) for older people with dementia have high numbers of patients with dementia (PWD). These patients are more susceptible to infectious diseases3 and are at higher risk of developing severe infections.4 Therefore, it is crucial to control COVID-19 clusters in SENHs and GHs. Various efforts have been made in response to these clusters, such as the isolation of patients with COVID-19 and restrictions on outings and visits, but these have been reported to contribute to the worsening of neuropsychiatric symptoms, cognitive decline, and increased caregiver distress in the context of patients with dementia.5 Moreover, there are reports of difficulties faced by the staff of care facilities for older adults when implementing COVID-19 cluster control measures, particularly concerning dementia care.6 The most frequently reported challenges were the psychological burden on staff, followed by the physical burden on staff and a shortage of staff.6 Additionally, difficulties in coordinating with local governments and public health centers responsible for public health activities, such as the distribution of medical equipment and adjusting the acceptance of infected individuals, have also been reported.6 A previous study reported that the worsening of neuropsychiatric symptoms and increase in the severity of dementia during the COVID-19 pandemic were related to anxiety and stress among facilities' caregiver staff (FCS).7 However, previous studies have only examined psychological burden as a factor contributing to the adverse effects of symptoms in PWD and have not examined the physical burden of FCS or the characteristics of facilities. Hiroshima University and the Japan Geriatrics Society conducted an online, self-administered questionnaire survey of medical and care facilities between January and February 2023. The medical facilities included hospitals specializing in dementia treatment and care, mental disorders, and chronic diseases requiring long-term care. Among the 995 participating facilities, 323 (32.5%) had a PWD percentage of 75–100% and included SENHs (54.5%) and GHs (45.5%) where clusters occurred. Among these facilities, 257 (79.6%) reported negative effects on dementia symptoms. In terms of statical analysis, a binary logistic regression analysis was conducted. The presence of negative effects on dementia symptoms was the dependent variable, and cluster response difficulties and the number of items corresponding to the difficulties faced in cluster response (the number of applicable items), which were significantly different in the chi-square test, were the independent variables. Covariates included facility capacity, the number of new COVID-19 cases per 100 000 population in each prefecture, and a binary variable (SENHs and GHs). Logistic regression analysis was performed by entering the independent variables individually without and with covariates (Model 1 and Model 2, respectively). SPSS software version 29.0 (IBM, Armonk, NY, USA) was used for the analyses, and the significance level was set at 5%. The study revealed significant differences in the relationship between the presence of negative effects on dementia symptoms and difficulties faced in the cluster response. The significant items were difficulty in the alleviation of the infection and prolonged persistence of new infections (P = 0.044); staff being unable to work owing to fear of infection or infection itself, resulting in staff shortages (P = 0.042); significant physical burden on staff (e.g., overtime work) (P = 0.025); and the number of items corresponding to the difficulties faced in the cluster response (P = 0.004). The results of the logistic regression analysis are presented in Table 1. Among the significant items, the factor with the highest odds ratio was significant physical burden on staff (e.g., overtime work) (Model 1: odds ratio = 2.28, P = 0.028; Model 2: odds ratio = 2.26, P = 0.030). Binary variable (special elderly nursing homes and group living homes) This study revealed that the most influential factor in the worsening of dementia symptoms was the significant physical burden on staff. In previous studies, caregivers of PWD during the COVID-19 pandemic reported that the main burden was the perception of physical strain and loss of time.8 Caregivers spend most of their day feeding, bathing, and assisting with changing and managing residents' treatment,8 which can lead to physical fatigue. Additionally, PWD are vulnerable to changes in care and environment.6 Therefore, the physical burden on FCS during the cluster response may have affected the quality of medical and care support services, leading to the worsening of dementia symptoms. It is important to reduce the burden on FCS to prevent the worsening of dementia symptoms in PWD. We express our deepest gratitude to the participants of this study, the Japan Association of Medical and Care Facilities, Japan Association of Geriatric Health Services Facilities, Japanese Council of Senior Citizens Welfare Service, Japan Group-Home Association for People with Dementia, and Japanese Council of Daily Life Long-Term Care Service Facilities. We would also like to thank the COVID-19 Response Team of the Japan Geriatrics Society for their cooperation in this study. The authors declare no potential conflict of interest. JT: methodology, validation, writing–original draft, writing–review and editing. HJ: methodology, validation, writing–review and editing. YI: methodology, validation, writing–review and editing. SI: methodology, validation, writing–review and editing. This study was not supported by a specific grant. Because this study used an anonymous questionnaire survey method, it did not require approval by an ethics committee according to the ethical guidelines of the Ministry of Health, Labor and Welfare. The datasets analyzed in the present study are not publicly available. Informed consent for the secondary use of the data was not obtained from the participants.