What are the factors that cause emergency home visit in home medical care in Japan?

Abstract Background In the home medical care setting, the factors causing emergency home visits (EHV) remain unclear. This study aimed to determine those factors and examine their relationship with EHV requests. Methods This is a single‐center retrospective observational study from data obtained from a home medical care clinic. We assessed the association between frequency of EHV and age, gender, level of care‐needed, cancer, and medical device in use with using Poisson regression analysis. Results A total of 608 EHV in 214 patients were analyzed. Common chief complaints were fever, death, and dyspnea. As factors that affect frequency of EHV because of fever, higher care‐needed level (RR: 3.35; 95% CI: 1.95‐5.74, P < .001), urinary catheter use (RR: 1.94; 95% CI: 1.22‐3.08, P = .005), and central venous port use (RR: 2.39; 95% CI: 1.44‐3.96, P = .001) showed significant correlation. Regarding EHV because of dyspnea, lung tumor (RR: 2.71; 95% CI: 1.26‐5.84, P = .011) and home oxygen use (RR: 3.96; 95% CI: 2.05‐7.68, P < .001) showed significant correlation. Regarding EHV because of all chief complaints, higher care‐needed level (RR: 1.59; 95% CI: 1.12‐2.26, P = .009), urinary catheter use (RR: 1.78; 95% CI: 1.13‐2.93, P = .014), and central venous port use (RR: 1.75; 95% CI: 1.04‐2.93, P = .034) showed positive correlation. Conclusion The factors associated with frequency of EHV because of fever or all chief complaints were urinary catheter use, central venous port use, and higher care‐needed level. As for dyspnea, they were lung cancer and home oxygen use. Our study suggests that the burdens on medical staffs, patients, and their families can be reduced through recognizing these risk factors.


| INTRODUC TI ON
While end-of-life care is attracting attention nowadays, it is reported that more than half of Japanese people desire to spend their time at home during their end of life. 1 The Ministry of Health, Labour, and Welfare promotes a home medical care system for older adults to supply medical and nursing care in places where they are used to living. 2 However, most people die in the hospital, and according to recent data, only about 10% die at home overall in Japan. 3 To achieve end-of-life care at home, there is an urgent need to improve the home medical care system.
On the other hand, more than 70% of the physicians in home medical care support clinics feel burdened by the 24 hour on-call system. 4 Furthermore, sudden changes in a patient's condition which require an emergency home visit (EHV) become a burden for patients and their family. 5 Moreover, it can become both a psychological burden and an obstacle to continue home medical care for home visiting nurses who have to request a physician's EHV. 6 For enhancement of home medical care, it is essential to take measures to reduce physical and psychological burdens for patients and their family, physicians, and home visiting nurses. For that reason, we need to know about the details relating to EHV.
In a previous study, there is cross-sectional work on contents of EHV, 7 and it shows that the common reasons for EHV are fever, death, dyspnea, and cough in that order of prevalence. However, there is no direct study that investigates the factors that cause EHV so far and data are lacking.
This study aims to clarify the factors that cause EHV at home medical care support clinics by analyzing the reason for EHV and background factors observable in our clinic.

| Study design and participants
This descriptive cross-sectional study was conducted by analyzing Toyama Machinaka Clinic's medical records related to EHV retrospectively. Toyama Machinaka Clinic is directly managed by Toyama city and is an enhanced home medical care support clinic with 3 physicians and 4 nurses. Toyama City has a population of approximately 417 000, 8 and this clinic covers patients in all area of Toyama City.
Most patients' demographics are complex, because of medical, social, and geographical reasons, which makes it difficult to be seen in local clinics.
We enrolled 214 patients over the age of 65 out of a total of 251 patients who had medical records within the 2 year duration from January 1, 2018, to December 31, 2019. In this study, patients aged 65 years or older were included because the population of patients under the age of 64 was small in our clinic, and therefore, this agegroup became nonrepresentative. We set the observation period as 2 years because the medical system of our clinic, which is in the fourth year since establishment, had been set up to some extent.
We determined that a sufficient number of samples could be obtained after confirming that there is no significant change or bias in the number and presenting problems of our patients in the first and second years.
This study was conducted in accordance with "Ethical Guidelines We defined EHV as an examination which was performed by a visiting physician, requested by a patient, family member, caregiver, or the insurance medical institution, and contact was made mostly by phone when a physician acknowledges the need for an emergency visit to the patient's home. 9

| Statistical analysis
We analyzed a total of 608 EHV in 214 patients.
First, we conducted a descriptive analysis about chief complaints requiring EHV, time period, and background of patients. We defined chief complaints in this study as the reason why EHV was requested.
We did not include the complaints that the patient stated or added when examined by the physician.
Second, we performed a Poisson regression analysis to determine the factors associated with the frequency of EHV. Variables include age, gender, level of care-needed, presence or absence of cancer, and medical device in use (urinary catheter, home oxygen, and central venous port).
The level of statistical significance was set at P < .05. All statistical analyses were conducted using IBM SPSS Statistics version 26.

| RE SULTS
The number of participants was 214. Mean age ± standard deviation was 81.9 ± 8.4 years. The number of male and female patients was 96 and 118, respectively.
The total number of EHV was 608 (during normal working hours: 384; after-hours: 224). Three common chief complaints that triggered the request for EHV were fever, death, and dyspnea. Among these, after-hour visits were 39 for fever (17.4%), 56 for death (25.0%), and 12 for dyspnea (5.4%). We found that death was the most common complaint triggering the request of EHV in after-hour visits. Table 1 shows the results of the cross-tabulation of each variable. This table shows the mean and standard deviation of the number of EHV for each chief complaint (column) for each subgroup of independent variables (row). Regarding the level of care-needed, the care need criteria classified by the Japanese Long-Term Care Insurance system was used to define the care need of individuals aged 65 or older into 7 levels: support-needed levels 1 and 2, and care-needed levels 1 to 5. 10 These are categorized into two groups, support-needed 1 and 2 and care-needed up to level 2 and care-needed levels 3-5 according to prior literature. 11 The presence or absence of cancer was divided into three groups: no cancer, lung tumor (primary and metastatic lung cancer), and other cancers, because it was expected that a lung lesion in all cancers would act as a confounding factor, especially in EHV with the chief complaint of dyspnea.
Spearman's rank correlation was used for "Age" category, a oneway ANOVA was then conducted for "Cancer" category, and a t test was used for other variables. For EHV with the chief complaint of fever, significance was shown in the higher level of care-needed, cancer, urinary catheter use, and central venous port use. For EHV with the chief complaint of dyspnea, it showed significance for lung cancer and home oxygen use. The evaluation for EHV for all chief complaints found significance in the higher level of care-needed, urinary catheter use, home oxygen use, and central venous port use.    16,17 It is crucial to be aware of risks, explaining them to patient and family in advance, clean operation and education, careful examination to recognizes risks, and early detection of abnormal findings. 18,19 Higher level of care-needed is also associated with fever events, as pointed out in previous studies. 11,20 Yokobayashi et al 20 state the reason why level of care-needed for fever events includes increased risk of aspiration because of less strength to cough, and increased susceptibility to infections from decreased muscle strength and poor nutrition status.

| D ISCUSS I ON
Second, dyspnea is a frequent chief complaint after fever, and an association to be noted was shown between EHV because of dyspnea and lung tumor/home oxygen use. Patients with lung tumors and home oxygen users should be considered a high-risk group.
Patients with lung cancer presenting with dyspnea are reported to have significantly shorter survival than patients with other cancers. 21 Thus, it is definitely important to discuss with patients with lung cancer and their family about the possible occurrence of dyspnea in advance. By and large, it is known that high rates of emergency room visits in terminally ill patients and physician house calls are low-quality indicators of end-of-life care. 22 believe it is important to recognize that cancer is an associated factor for EHV in home medical care as well. The occasions that patients and their family feel the necessity to request EHV can be reduced by appropriate symptom management, 26 increasing home care nursing, 27 palliative care with team coordination, 28 and education and support for family caregivers. 29,30 Fever and dyspnea are also valued symptoms in previous studies, 7,11,20 and the results in this study correspond with those of previous studies.  31 Based on the current situation above, we assessed the problem of patients who needed EHV from the perspective of patient care.
In conclusion, common chief complaints that triggered the request of EHV were fever, death, and dyspnea. For factors that affect the occurrence of EHV because of fever, urinary catheter use, central venous port, and a higher level of care-needed showed significant positive correlation. For EHV because of dyspnea, lung tumor and home oxygen use showed significant correlations. Regarding EHV because of all chief complaints, the higher care-needed level, urinary catheter use, and central venous port use showed positive correlations. Our study suggests that we can reduce the physical and psychology burden on medical staffs, patients, and their families through recognizing these risk factors for EHV.

ACK N OWLED G EM ENTS
We gratefully acknowledge Asami Usui, Yukiko Kanamori, Kana Kusano, Rieko Satake, Rumi Nakano, and Emi Higuchi for their cooperation. We also appreciate Dr R. Eugene Bailey, Dr Ryuichi Ohta, and Dr Makoto Kaneko for their helpful advice.

CO N FLI C T O F I NTE R E S T
Authors declare no conflict of interests in this article.