bs_bs_banner
Geriatr Gerontol Int 2015; 15: 465–471
ORIGINAL ARTICLE: EPIDEMIOLOGY, CLINICAL PRACTICE AND HEALTH
Home death is associated with frequency of physician home medical care visits: A questionnaire survey on communications in home medical care settings Takuma Kimura,1,2 Teruhiko Imanaga,2 Makoto Matsuzaki3 and Tohru Akahoshi1 1
Department of General Medicine, Kitasato University School of Medicine, Sagamihara, Kanagawa, 2Department of General Medicine, National Hospital Organization, Higashisaitama Hospital, Hasuda, Saitama, and 3Center for Clinical Training and Education, Toho University School of Medicine, Tokyo, Japan
Aim: To investigate the factors related to communications in home medical care settings, and the association between such factors and a patient’s place of death. Methods: A questionnaire survey of 295 families of patients who had previously received home medical care was carried out in June and July 2011. The response rate was 83.8% (n = 227). Following the exclusion of families where the patient was still alive, or where the place of death was unknown, 143 questionnaires were available for analysis. Logistic regression was used to identify significant associations between possible factors related to communication and occurrence of home death. Results: Home death was observed in 66.4% (n = 95) of the families analyzed. Home death was significantly associated with the frequency of doctor home-visits per week (OR 2.835, 95% CI 1.436–5.597, P = 0.003). There was no statistically significant association between home death and any of the other variables included: malignant tumors as primary disease, independence in daily activity, duration of home medical care, duration of doctor’s visits, experience of doctor–patient communication without family, doctor–family communication without the patient or explanation from the doctor on the phone, existence of home-visit nursing services, existence of family’s anxieties and/or questions, age of primary caregiver(s) and sex of primary caregiver(s). Conclusion: The frequency of doctor home-visits was the only factor identified that was positively associated with the occurrence of home death in home medical care settings. Geriatr Gerontol Int 2015; 15: 465–471. Keywords: communication, doctor, family, home death, home medical care.
Introduction Japan has the fastest rate of population aging and the highest life expectancy at birth of all industrialized countries.1 Recently in Japan, there have been extensive discussions on the benefits of being cared for, and dying, at home.1–9 It has been shown that home death has a positive influence not only on the quality of life (QOL) of the patients themselves, but also their families
Accepted for publication 11 March 2014. Correspondence: Dr Takuma Kimura Master of Clinical Epidemiology, PhD, Department of General Medicine, Kitasato University School of Medicine, 1-15-1 Kitasato, Minami-ku, Sagamihara, Kanagawa 252-0374, Japan. Email: [email protected]
© 2014 Japan Geriatrics Society
when compared with deaths occurring at hospital.2 Hence, factors associated with dying at home have been a focus of research. Factors identified to date fall into four general categories: (i) patient-related factors, such as activities of daily living, diagnoses and the patient’s preferences for death at home; (ii) family-related factors, such as number of caregivers and the family’s preferences for patient’s death at home; (iii) doctor-related factors, such as control of physical symptoms (e.g. pain), and explanations about the dying process/death given to the patient and the family; and (iv) health system-related factors, such as the home-visit nursing system and referral support system.3–9 Doctors should have sufficient communication with the patient and their family, to allow them to be fully aware of both parties’ preferences regarding home doi: 10.1111/ggi.12297
| 465
T Kimura et al.
death. Effective communication is also required to successfully achieve “detailed explanations about the dying process/death to the patient and the family.”4 In home care settings, interprofessional work is important. In particular, home-visit nurses play an important role in many countries.10–12 In Japan, visiting doctors are considered to play an important role in the case of a family’s and/or patient’s decision-making of weighty issues, such as whether to choose home death, whether to ask the doctor to explain in detail about dying process/ death, and so on. However, to date there is no experimental evidence on the relationship between the occurrence of a patient’s home death and the factors related to the doctor’s communication, such as communication place/environment, how and what to convey, frequency of doctor visits and duration of a doctor visit, and whom to communicate with.13 Our first hypothesis is that, in home medical care settings, the number of times and the length of the doctor’s communication are associated with the occurrence of patients’ home deaths. Specifically, a higher visit frequency and a longer stay duration lead to a higher incidence of home deaths. Our second hypothesis is that, in home medical care settings, a doctor’s communication with the patient in the absence of the family, and vice versa, is associated with the occurrence of patients’ home deaths. This is because, in previous studies on outpatients and inpatients, it is suggested that in order for the attending doctor to gain full knowledge about patients’ and families’ preferences for home deaths, it is occasionally necessary for them to communicate with the patient in the absence of the family and vice versa.14–19 Further clarity is required on the contribution of the various strands of communication to improve home medical care and enhance home deaths. Therefore, we carried out the present study to investigate the possible factors related to doctor’s communication, such as frequency of doctor visits, duration of a doctor visit and those whom the doctor communicates with during home medical care and, subsequently, the association between such factors and place of death.
Methods Sampling and enrolment criteria An anonymous, self-completed, postal questionnaire survey was carried out on “communications in home medical care settings” in June and July 2011. Of the 323 families of patients who received home medical care from Higashisaitama Hospital physicians between May 2006 and April 2011, 295 families that met the following three criteria were selected as potential participants: (i) those who received home medical care more than three times, excluding a “deathbed” visit; (ii) those who were 466 |
present at least once during home medical care; and (iii) in cases where the patient had deceased, those where more than 50 days had elapsed since the patient’s death (out of respect for the bereaved family).
Ethical considerations The present study was carried out with the prior approval of the ethics committee of Higashisaitama Hospital Hasuda, Saitama, Japan. The researchers explained to the potential families the purpose and contents of the study, the protection of their personal information, and the anonymity of the obtained data. The questionnaires were then mailed only to families who had consented to being mailed on the phone or during a doctor visit, along with an explanatory document. Completing and returning the questionnaire by mail was interpreted as proof of consent to participate in the study. In cases where home medical care was still ongoing during the survey period, the contents of the survey were also explained to the patient, from whom consent was obtained.
Questionnaires and other data source The data sources of the present study were medical records and completed questionnaires. The questionnaires consisted of 53 questions, including some questions collected for a related study.20 Information on patient characteristics and place of death was collected from the medical records, whereas the information on patient background, family background and caregiving situation was extracted from the completed questionnaires. To study the characteristics of communication occurring during home medical care, information on communication between the patient and the doctor, and between the family and the doctor was also extracted from the completed questionnaires. The aforementioned questions and survey items were selected by the authors based on the results of previous studies on communication methods during home medical care.14–17,21
Patient characteristics For the present study, the following patient characteristics were examined: patient age at the start of home medical care, sex, primary disease, alive or dead (alive, died at home, died at hospital, unknown), duration of primary disease (in months), duration of home medical care (in days) and frequency of doctor visits per week. In addition, the level of independence in daily activities (eight levels) and the level of cognitive function (nine levels) assessed at the start of home medical care were described in accordance with guidance contained in the Japanese Long-term Care Insurance program, the attending physician’s report or the visiting nurse’s directions.1,22 © 2014 Japan Geriatrics Society
Communication and home death occurrence
Patient background The level of care required for the patient at the start of home medical care (i.e. requiring some assistance or less, care levels 1, 2, 3, 4 and 5) in accordance with the guidance contained in the Japanese Long-term Care Insurance program1,22 was recorded. The families were also asked if they had used home-visit nursing services and to estimate the mean duration (in minutes) of a typical doctor visit.
Communication between the patient and the doctor To assess the extent of communications occurring directly between the patient and the doctor, in the absence of the family during home medical care (doctor–patient communication without family; hereafter referred to as “DPC without family”) the primary caregivers were asked if they had witnessed such communication, and were requested to answer either “Yes, I have” or “No, I haven’t.”
Family background and caregiving situation For the present study, the following were examined: age and sex of the primary caregiver(s), the number of caregivers, and the existence of caregiving advisors with whom the family could consult. In addition, the existence of family’s anxieties and/or questions about patient’s symptoms was asked and described in five levels (never, almost nothing, a little, much and very much).
Communication situation between the family and the doctor To assess communications between the family and the doctor, the following two questions were asked: (i) if they had experience of communications exchanged just between the doctor and the family in the absence of the patient during home medical care (doctor–family communication without patient; hereafter referred to as “DFC without patient”); and (ii) if they had experience of receiving an explanation from the doctor on the phone. The primary caregivers were requested to answer either “Yes, I have” or “No, I haven’t.”
Data analysis Analysis model We constructed a multivariable logistic regression model to examine the relationship between place of death and the possible factors related to a doctor’s communication, such as frequency of a doctor’s visits, duration of a doctor’s visit and those whom the doctor communicates with during home medical care (the © 2014 Japan Geriatrics Society
principal subject of the present study) and other important explanatory factors previously identified in the literatures.3–10,13–21,23,24 The model included the following elements: 1 In order to test our first hypothesis, “frequency of doctor visits” and “mean duration of a doctor visit” were included in the model.13 2 In order to test our second hypothesis, “experiences of DPC without family” and “experiences of DFC without patient” were also included in the model.14–21 3 Also included in the model was “experience of receiving an explanation from the doctor on the phone,” which is reported in a previous study to be useful to complement communication during home medical care.23 4 “Malignant tumor as the primary disease,” “level of care required,” “use experience of home-visit nursing services” and “existence of family’s anxieties and/or questions about patient’s symptoms” were also included as factors previously identified in the literature associated with place of death.3–10,24 For this analysis, the primary disease was categorized as either “malignant tumor” or “other than malignant tumor.” When categorizing patients with multiple diagnoses, a diagnosis of a malignant tumor was given precedence over any other diagnosis. 5 As demographic characteristics, “age of primary caregiver(s)” and “sex of primary caregiver(s)” were also included in the model.
Statistical analysis Differences in place of death by patient and family characteristics were tested using the t-test for continuous variables, Wilcoxon rank–sum test for ordinal rating scale and the χ2-test for categorical variables. Factors associated with place of death (“home death” = 1 vs “death at hospital” = 0) were assessed using multivariate logistic regression analysis, and the adjusted odds ratios (OR) and their 95% confidence intervals (95% CI) were calculated. The explanatory factors included in the model were: “frequency of doctor’s visits”; “mean duration of a doctor’s visit”; “experience of DPC without family”; “experience of DFC without patient”; “experience of receiving an explanation from the doctor on the phone”; “malignant tumor as primary disease”; “level of care required”; “home-visit nursing services”; “existence of family’s anxieties and/or questions about the patient’s symptoms”; “age of primary caregiver(s)” and “sex of primary caregiver(s).” Analysis was carried out using the statistics package SPSS, version 19 for Windows (SPSS, Chicago, IL, USA). All P-values were two-sided, and P-values less than 0.05 were considered statistically significant. The Hosmer– Lemeshow test was used to assess the goodness-of-fit of the final model. | 467
T Kimura et al.
Results Of the potential 295 families who met the initial inclusion criteria, 17 did not consent to be contacted and seven were missing contact addresses. The questionnaire was mailed to the remaining 271 families, and 227 (83.8%) responded. Of those 227 families, 84 were excluded because the patient was still alive or the place of death (at home or hospital) was unknown, although the outcomes of all patients were known. As a result, data on 143 families remained in the final analysis of the present study. Table 1 shows the patient characteristics, and Table 2 shows the family characteristics. Home death occurred in 66.4% (n = 95) of the families analyzed.
Differences in patient characteristics Table 1 shows the patient characteristics by place of death. In patients with malignant tumors, home deaths were significantly more frequent than hospital deaths (P < 0.001). “Duration of home medical care” was significantly longer in patients who died in hospital than in
patients who had a home death (P < 0.001). Doctors made significantly more home visits per week to those who died at home than those who died in hospital (P < 0.001). No statistically significant difference in place of death was observed by “age,” “sex,” “primary disease duration,” “independence of daily activity,” “cognitive function,” “level of care required,” “duration of doctor’s visits,” “experience of DPC without family” and “use experience of home-visit nursing services.”
Differences in the family characteristics Table 2 shows the family characteristics by place of death. “Experience of DFC without patient” was significantly more frequent in home deaths than in hospital deaths (P < 0.014). No statistically significant differences were observed between home deaths and hospital deaths by “age of primary caregiver(s),” “sex of primary caregiver(s),” “number of caregivers,” “existence of caregiving advisors,” “experience of an explanation from doctor on the phone” and “existence of family’s anxieties and/or questions about patient’s symptoms.”
Table 1 Patient characteristics Home death (n = 95) Age, years (mean ± SD) Sex Male, n (%) Female, n (%) Disease classification Malignant tumor, n (%) Other than malignant tumor, n (%) Primary disease duration, months (mean ± SD) Duration of home medical care, days (mean ± SD) Frequency of doctor’s home-visits, per week (mean ± SD) Independence in daily activity (median) Cognitive function, median (1st quartile, 3rd quartile) Level of care required, median (1st quartile, 3rd quartile) Use experience of home-visit nursing services Yes, n (%) No, n (%) Duration of a doctor visit, min (mean ± SD) Experience of doctor–patient communication without family Yes, n (%) No, n (%)
Death at hospital (n = 48)
P-value
76 ± 14
77 ± 11
46 (48.4) 49 (51.6)
25 (52.1) 23 (47.9)
59 (63.1) 36 (37.9.) 40 ± 64.5 119 ± 224.2
14 (29.2) 34 (70.8.) 61 ± 64.5 356 ± 415.8
0.059†
Geriatr Gerontol Int 2015; 15: 465–471
ORIGINAL ARTICLE: EPIDEMIOLOGY, CLINICAL PRACTICE AND HEALTH
Home death is associated with frequency of physician home medical care visits: A questionnaire survey on communications in home medical care settings Takuma Kimura,1,2 Teruhiko Imanaga,2 Makoto Matsuzaki3 and Tohru Akahoshi1 1
Department of General Medicine, Kitasato University School of Medicine, Sagamihara, Kanagawa, 2Department of General Medicine, National Hospital Organization, Higashisaitama Hospital, Hasuda, Saitama, and 3Center for Clinical Training and Education, Toho University School of Medicine, Tokyo, Japan
Aim: To investigate the factors related to communications in home medical care settings, and the association between such factors and a patient’s place of death. Methods: A questionnaire survey of 295 families of patients who had previously received home medical care was carried out in June and July 2011. The response rate was 83.8% (n = 227). Following the exclusion of families where the patient was still alive, or where the place of death was unknown, 143 questionnaires were available for analysis. Logistic regression was used to identify significant associations between possible factors related to communication and occurrence of home death. Results: Home death was observed in 66.4% (n = 95) of the families analyzed. Home death was significantly associated with the frequency of doctor home-visits per week (OR 2.835, 95% CI 1.436–5.597, P = 0.003). There was no statistically significant association between home death and any of the other variables included: malignant tumors as primary disease, independence in daily activity, duration of home medical care, duration of doctor’s visits, experience of doctor–patient communication without family, doctor–family communication without the patient or explanation from the doctor on the phone, existence of home-visit nursing services, existence of family’s anxieties and/or questions, age of primary caregiver(s) and sex of primary caregiver(s). Conclusion: The frequency of doctor home-visits was the only factor identified that was positively associated with the occurrence of home death in home medical care settings. Geriatr Gerontol Int 2015; 15: 465–471. Keywords: communication, doctor, family, home death, home medical care.
Introduction Japan has the fastest rate of population aging and the highest life expectancy at birth of all industrialized countries.1 Recently in Japan, there have been extensive discussions on the benefits of being cared for, and dying, at home.1–9 It has been shown that home death has a positive influence not only on the quality of life (QOL) of the patients themselves, but also their families
Accepted for publication 11 March 2014. Correspondence: Dr Takuma Kimura Master of Clinical Epidemiology, PhD, Department of General Medicine, Kitasato University School of Medicine, 1-15-1 Kitasato, Minami-ku, Sagamihara, Kanagawa 252-0374, Japan. Email: [email protected]
© 2014 Japan Geriatrics Society
when compared with deaths occurring at hospital.2 Hence, factors associated with dying at home have been a focus of research. Factors identified to date fall into four general categories: (i) patient-related factors, such as activities of daily living, diagnoses and the patient’s preferences for death at home; (ii) family-related factors, such as number of caregivers and the family’s preferences for patient’s death at home; (iii) doctor-related factors, such as control of physical symptoms (e.g. pain), and explanations about the dying process/death given to the patient and the family; and (iv) health system-related factors, such as the home-visit nursing system and referral support system.3–9 Doctors should have sufficient communication with the patient and their family, to allow them to be fully aware of both parties’ preferences regarding home doi: 10.1111/ggi.12297
| 465
T Kimura et al.
death. Effective communication is also required to successfully achieve “detailed explanations about the dying process/death to the patient and the family.”4 In home care settings, interprofessional work is important. In particular, home-visit nurses play an important role in many countries.10–12 In Japan, visiting doctors are considered to play an important role in the case of a family’s and/or patient’s decision-making of weighty issues, such as whether to choose home death, whether to ask the doctor to explain in detail about dying process/ death, and so on. However, to date there is no experimental evidence on the relationship between the occurrence of a patient’s home death and the factors related to the doctor’s communication, such as communication place/environment, how and what to convey, frequency of doctor visits and duration of a doctor visit, and whom to communicate with.13 Our first hypothesis is that, in home medical care settings, the number of times and the length of the doctor’s communication are associated with the occurrence of patients’ home deaths. Specifically, a higher visit frequency and a longer stay duration lead to a higher incidence of home deaths. Our second hypothesis is that, in home medical care settings, a doctor’s communication with the patient in the absence of the family, and vice versa, is associated with the occurrence of patients’ home deaths. This is because, in previous studies on outpatients and inpatients, it is suggested that in order for the attending doctor to gain full knowledge about patients’ and families’ preferences for home deaths, it is occasionally necessary for them to communicate with the patient in the absence of the family and vice versa.14–19 Further clarity is required on the contribution of the various strands of communication to improve home medical care and enhance home deaths. Therefore, we carried out the present study to investigate the possible factors related to doctor’s communication, such as frequency of doctor visits, duration of a doctor visit and those whom the doctor communicates with during home medical care and, subsequently, the association between such factors and place of death.
Methods Sampling and enrolment criteria An anonymous, self-completed, postal questionnaire survey was carried out on “communications in home medical care settings” in June and July 2011. Of the 323 families of patients who received home medical care from Higashisaitama Hospital physicians between May 2006 and April 2011, 295 families that met the following three criteria were selected as potential participants: (i) those who received home medical care more than three times, excluding a “deathbed” visit; (ii) those who were 466 |
present at least once during home medical care; and (iii) in cases where the patient had deceased, those where more than 50 days had elapsed since the patient’s death (out of respect for the bereaved family).
Ethical considerations The present study was carried out with the prior approval of the ethics committee of Higashisaitama Hospital Hasuda, Saitama, Japan. The researchers explained to the potential families the purpose and contents of the study, the protection of their personal information, and the anonymity of the obtained data. The questionnaires were then mailed only to families who had consented to being mailed on the phone or during a doctor visit, along with an explanatory document. Completing and returning the questionnaire by mail was interpreted as proof of consent to participate in the study. In cases where home medical care was still ongoing during the survey period, the contents of the survey were also explained to the patient, from whom consent was obtained.
Questionnaires and other data source The data sources of the present study were medical records and completed questionnaires. The questionnaires consisted of 53 questions, including some questions collected for a related study.20 Information on patient characteristics and place of death was collected from the medical records, whereas the information on patient background, family background and caregiving situation was extracted from the completed questionnaires. To study the characteristics of communication occurring during home medical care, information on communication between the patient and the doctor, and between the family and the doctor was also extracted from the completed questionnaires. The aforementioned questions and survey items were selected by the authors based on the results of previous studies on communication methods during home medical care.14–17,21
Patient characteristics For the present study, the following patient characteristics were examined: patient age at the start of home medical care, sex, primary disease, alive or dead (alive, died at home, died at hospital, unknown), duration of primary disease (in months), duration of home medical care (in days) and frequency of doctor visits per week. In addition, the level of independence in daily activities (eight levels) and the level of cognitive function (nine levels) assessed at the start of home medical care were described in accordance with guidance contained in the Japanese Long-term Care Insurance program, the attending physician’s report or the visiting nurse’s directions.1,22 © 2014 Japan Geriatrics Society
Communication and home death occurrence
Patient background The level of care required for the patient at the start of home medical care (i.e. requiring some assistance or less, care levels 1, 2, 3, 4 and 5) in accordance with the guidance contained in the Japanese Long-term Care Insurance program1,22 was recorded. The families were also asked if they had used home-visit nursing services and to estimate the mean duration (in minutes) of a typical doctor visit.
Communication between the patient and the doctor To assess the extent of communications occurring directly between the patient and the doctor, in the absence of the family during home medical care (doctor–patient communication without family; hereafter referred to as “DPC without family”) the primary caregivers were asked if they had witnessed such communication, and were requested to answer either “Yes, I have” or “No, I haven’t.”
Family background and caregiving situation For the present study, the following were examined: age and sex of the primary caregiver(s), the number of caregivers, and the existence of caregiving advisors with whom the family could consult. In addition, the existence of family’s anxieties and/or questions about patient’s symptoms was asked and described in five levels (never, almost nothing, a little, much and very much).
Communication situation between the family and the doctor To assess communications between the family and the doctor, the following two questions were asked: (i) if they had experience of communications exchanged just between the doctor and the family in the absence of the patient during home medical care (doctor–family communication without patient; hereafter referred to as “DFC without patient”); and (ii) if they had experience of receiving an explanation from the doctor on the phone. The primary caregivers were requested to answer either “Yes, I have” or “No, I haven’t.”
Data analysis Analysis model We constructed a multivariable logistic regression model to examine the relationship between place of death and the possible factors related to a doctor’s communication, such as frequency of a doctor’s visits, duration of a doctor’s visit and those whom the doctor communicates with during home medical care (the © 2014 Japan Geriatrics Society
principal subject of the present study) and other important explanatory factors previously identified in the literatures.3–10,13–21,23,24 The model included the following elements: 1 In order to test our first hypothesis, “frequency of doctor visits” and “mean duration of a doctor visit” were included in the model.13 2 In order to test our second hypothesis, “experiences of DPC without family” and “experiences of DFC without patient” were also included in the model.14–21 3 Also included in the model was “experience of receiving an explanation from the doctor on the phone,” which is reported in a previous study to be useful to complement communication during home medical care.23 4 “Malignant tumor as the primary disease,” “level of care required,” “use experience of home-visit nursing services” and “existence of family’s anxieties and/or questions about patient’s symptoms” were also included as factors previously identified in the literature associated with place of death.3–10,24 For this analysis, the primary disease was categorized as either “malignant tumor” or “other than malignant tumor.” When categorizing patients with multiple diagnoses, a diagnosis of a malignant tumor was given precedence over any other diagnosis. 5 As demographic characteristics, “age of primary caregiver(s)” and “sex of primary caregiver(s)” were also included in the model.
Statistical analysis Differences in place of death by patient and family characteristics were tested using the t-test for continuous variables, Wilcoxon rank–sum test for ordinal rating scale and the χ2-test for categorical variables. Factors associated with place of death (“home death” = 1 vs “death at hospital” = 0) were assessed using multivariate logistic regression analysis, and the adjusted odds ratios (OR) and their 95% confidence intervals (95% CI) were calculated. The explanatory factors included in the model were: “frequency of doctor’s visits”; “mean duration of a doctor’s visit”; “experience of DPC without family”; “experience of DFC without patient”; “experience of receiving an explanation from the doctor on the phone”; “malignant tumor as primary disease”; “level of care required”; “home-visit nursing services”; “existence of family’s anxieties and/or questions about the patient’s symptoms”; “age of primary caregiver(s)” and “sex of primary caregiver(s).” Analysis was carried out using the statistics package SPSS, version 19 for Windows (SPSS, Chicago, IL, USA). All P-values were two-sided, and P-values less than 0.05 were considered statistically significant. The Hosmer– Lemeshow test was used to assess the goodness-of-fit of the final model. | 467
T Kimura et al.
Results Of the potential 295 families who met the initial inclusion criteria, 17 did not consent to be contacted and seven were missing contact addresses. The questionnaire was mailed to the remaining 271 families, and 227 (83.8%) responded. Of those 227 families, 84 were excluded because the patient was still alive or the place of death (at home or hospital) was unknown, although the outcomes of all patients were known. As a result, data on 143 families remained in the final analysis of the present study. Table 1 shows the patient characteristics, and Table 2 shows the family characteristics. Home death occurred in 66.4% (n = 95) of the families analyzed.
Differences in patient characteristics Table 1 shows the patient characteristics by place of death. In patients with malignant tumors, home deaths were significantly more frequent than hospital deaths (P < 0.001). “Duration of home medical care” was significantly longer in patients who died in hospital than in
patients who had a home death (P < 0.001). Doctors made significantly more home visits per week to those who died at home than those who died in hospital (P < 0.001). No statistically significant difference in place of death was observed by “age,” “sex,” “primary disease duration,” “independence of daily activity,” “cognitive function,” “level of care required,” “duration of doctor’s visits,” “experience of DPC without family” and “use experience of home-visit nursing services.”
Differences in the family characteristics Table 2 shows the family characteristics by place of death. “Experience of DFC without patient” was significantly more frequent in home deaths than in hospital deaths (P < 0.014). No statistically significant differences were observed between home deaths and hospital deaths by “age of primary caregiver(s),” “sex of primary caregiver(s),” “number of caregivers,” “existence of caregiving advisors,” “experience of an explanation from doctor on the phone” and “existence of family’s anxieties and/or questions about patient’s symptoms.”
Table 1 Patient characteristics Home death (n = 95) Age, years (mean ± SD) Sex Male, n (%) Female, n (%) Disease classification Malignant tumor, n (%) Other than malignant tumor, n (%) Primary disease duration, months (mean ± SD) Duration of home medical care, days (mean ± SD) Frequency of doctor’s home-visits, per week (mean ± SD) Independence in daily activity (median) Cognitive function, median (1st quartile, 3rd quartile) Level of care required, median (1st quartile, 3rd quartile) Use experience of home-visit nursing services Yes, n (%) No, n (%) Duration of a doctor visit, min (mean ± SD) Experience of doctor–patient communication without family Yes, n (%) No, n (%)
Death at hospital (n = 48)
P-value
76 ± 14
77 ± 11
46 (48.4) 49 (51.6)
25 (52.1) 23 (47.9)
59 (63.1) 36 (37.9.) 40 ± 64.5 119 ± 224.2
14 (29.2) 34 (70.8.) 61 ± 64.5 356 ± 415.8
0.059†
