Abstract
Abstract
Background:
Many studies have shown a lack of advantages to tube feeding for elderly with advanced dementia, but tube feeding is still considered standard care in Japan. The aim of this study is to investigate what nutrition method health care professionals want for themselves or their families, if they fall into a bedridden state due to irreversible impaired cognition in old age.
Methods:
In 2010 we surveyed 1321 Japanese health care professionals including 251 medical doctors and 1070 nurses. Their attitudes toward tube feeding were assessed by using an anonymous questionnaire, which included desired feeding methods for themselves or their families and propriety of card-based declaration of intent for end-of-life care.
Results:
Rates of accepting tube feeding for themselves and their families were 14.4% and 43.4%, respectively. In multivariate analyses, working at a municipal hospital and high frequency of taking care of tube-fed elderly patients were predictors of refusing tube feeding for themselves. Working at a municipal hospital and being a medical doctor were predictors of refusing tube feeding for their families. The rate of welcoming card-based declaration of intent for end-of-life care including feeding methods was 65.2%.
Conclusions:
Many doctors and nurses, especially with more frequent contact with tubefed patients, rejected tube feeding for themselves on their own deathbed, but did not always refuse this option for their families.
Introduction
Methods
Sample
Anonymous questionnaires were administered in 2010 to all of the full-time medical doctors and nurses at the five hospitals. They included 251 medical doctors and 1070 nurses. The hospitals were located in Saitama, Fukuoka, and Oita Prefectures in Japan, and included two city-run hospitals, two urban teaching facilities, and one university hospital, ranging in size from 190 to 978 beds. The work environment of the subjects varied from an intensive care unit for critically ill patients to a hospice for the terminally ill elderly patients. Respondents to the questionnaire remained anonymous and therefore the study did not require the approval of the ethics review boards at participating institutions.
The national health insurance system in Japan
The universal health care system in Japan, which was designed to provide all citizens with equal access to health care at a reasonable cost, was established in 1961. However, medical expenses have been increasing toward a major financial crisis with the rapid aging of the Japanese population. A new system of inclusive per-diem payments was introduced in 1990 as an option for almost all hospitals. These systems allow the elderly with advanced dementia to be placed on a feeding tube. All of the five hospitals in this study utilized the system of inclusive per-diem payments.
Structure of questionnaire
This was a simple, anonymous one-page pencil-and-paper questionnaire. Questions were presented as a number of options for each item, and respondents selected one answer from among the options. Question items requested as shown in Table 1. nutrition, central intravenous hyperalimentation, oral nutrition alone if possible, or delegation of the decision to their family) and their family (PEG, nasogastric tube feeding, peripheral intravenous nutrition, central intravenous hyperalimentation, or oral nutrition alone) with bedridden status due to irreversible dementia in old age; whether they had ever discussed with their family their wishes for medical treatment near the end of life; and the propriety of a card-based declaration of intent for near end-of-life care (welcome, not welcome, or indeterminate). In this questionnaire, “near the end of life” was defined as a condition of bedridden status due to irreversible dementia in old age, with reference to the National Institutes of Health State-of-the-Science Conference statement. 9
PEG, percutaneous endoscopic gastrostomy; QOL, quality of life.
Statistical analysis
Statistical analyses were performed using PASW Statistics version 18.0 software (IBM SPSS, Tokyo, Japan). Statistical significance was defined by a value of p<0.05 for all analyses. The χ2 test was applied for comparing categorical variables, unless one of the categories had fewer than 20 observations, in which case Fisher's exact test was applied. The χ2 test with linear trend test was conducted for testing trends. For the evaluation of predictors associated with professionals wanting tube feeding for themselves and their families, multiple logistic regression analysis was used to compare relevant backgrounds.
Results
Respondent characteristics
Questionnaires were completed by 203 doctors and 1013 nurses, with response rates of 80.9% and 94.7%, respectively. The 203 doctors consisted of internal medicine physicians [105 (52%)], neurologists [49 (24%)], psychiatrists [21 (10%)], pediatricians [13 (6%)], anesthetists [4 (2%)], residents [4 (2%)], and rehabilitation doctors [2 (1%)]. Workplaces of respondents were either university hospitals or municipal hospitals. Urban teaching facilities and city-run hospitals were merged into municipal hospitals, because these were relatively similar types of hospitals that mainly treated acute-phase patients. A significant difference in workplace was seen between doctors and nurses (p=0.015; Table 2). With the present respondents, more doctors were working at university hospitals than at municipal hospitals, while fewer nurses were working at university hospitals than at municipal hospitals. Although nurses were younger than doctors (p=0.049), no significant difference in length of work experience was seen between these two groups (p=0.472). No difference was seen in the number of tube-fed elderly patients cared for during the last three months between doctor and nurse groups (p=0.53).
A comparison of the background characteristics between the respondents in university hospital (n=465) and municipal hospital (n=751) revealed no significant difference in proportion of males [20% (95/465) versus 24% (177/751), p=0.201], age groups (p=0.102), length of service (p=0.352), and the frequency of taking care of tube-fed patients over the last three months (p=0.088). The only significant difference was observed in the proportion of doctors among medical professionals [doctor; 23% (108/465) versus 13% (95/751), p=0.015].
Attitudes of professionals toward nutrition for themselves and their families
Respondents selecting ‘oral nutrition alone’ for themselves comprised the largest group, followed by a group that selected ‘delegate to family’ (Table 3). The rate of respondents who wanted tube feeding using PEG or nasogastric tube for themselves under a situation of bedridden state with irreversible dementia in old age was 14.4%. The numbers of doctors who wanted tube feeding for themselves according to their specialities were as follows: 2 out of 2 rehabilitation doctors, 5 out of 6 neurosurgeons, 2 out of 4 residents, 4 out of 9 neurologists, 8 out of 39 surgeons except neurosurgeons, 2 out of 13 pediatricians and 3 out of 105 physicians. In multiple logistic regression analysis, working at a municipal hospital and more-frequent contact with tube-fed elderly patients were significant predictors of rejecting tube feeding for themselves, after adjusting for other variables (Table 4). Health care professionals who more frequently took care of tube-fed patients rejected tube feeding for themselves at a significantly higher rate with a significant difference (p=0.002; Figure 1). In contrast, 43.4% of respondents wanted tube feeding (35.0% in PEG, and 8.4% in nasal feeding) for their family members in the same situation (Table 3). The number of doctors who wanted tube feeding for their families according to their specialties revealed that rehabilitation doctors, anesthetists, residents, and neurosurgeons ranked first at 100% (2/2, 4/4, 4/4, 6/6, respectively) followed by neurologists at 89% (8/9), surgeons at 31% (14/45), pediatricians at 29% (5/13), psychiatrists at 29% (6/21), and physicians at 17% (18/105). In multivariate modelling, working at a municipal hospital and being a medical doctor were significant predictors of rejecting tube feeding for the subject's family (Table 5).
Rates of professionals accepting tube feeding for themselves according to the frequency of contact with tube-fed elderly patients with severe dementia.
PEG, percutaneous endoscopic gastrostomy; QOL, quality of life.
CI, confident interval; OR, odds ratio.
CI, confidence interval; OR, odds ratio.
Purpose of tube feeding for elderly patients
The views of professionals regarding the main purpose of tube feeding for elderly patients with bedridden status due to irreversible dementia are shown in Table 3. Responses of ‘life extension,’ ‘improving QOL,’ or ‘prevention of aspiration’ were commonly selected, while ‘social reasons' was least commonly selected, at approximately 10%. No significant difference in attitudes between medical doctors and nurses was identified in this section (p=0.121). A comparison of the rate of professionals who thought the medical benefits of tube feeding (e.g., life extension, improvement of QOL, prevention of aspiration, prevention of pressure ulcer, and weight gain) showed no significant difference between professionals who wanted tube feeding and those who did not [83% (145/175), versus 84% (872/1041), p=0.849].
Attitudes of professional toward end-of-life treatment
Close to half of the respondents had discussed with their families medical care at the end of life, and 65.2% of respondents welcomed the concept of a card-based declaration of intent for end-of-life care, including feeding methods (Table 6).
Discussion
This study demonstrated that only 14.4% of Japanese health care professionals wanted tube feeding for themselves under a hypothetical situation of bedridden status with irreversible dementia, and that working at a municipal hospital and higher frequency of taking care of tube-fed patients were significant predictors of rejecting tube feeding for themselves.
Of note was the finding that many professionals believe that tube feeding extends survival, whereas relatively few professionals wished to receive tube feeding for themselves. These results indicate that the lack of evidence for any survival advantage from tube feeding has not become widely known among professionals in Japan,1,3–5 and that they did not necessarily want artificial life-prolonging treatment near the end of life. One reason why working at a municipal hospital was a strong predictor associated with refusing tube feeding near the end of life may be that university hospitals treat a relatively large number of younger patients who rarely die a slow death. In addition, professionals who had more contact with tube-fed elderly patients tended to reject the notion of tube feeding for themselves. Such professionals probably do not regard tube feeding as an ideal nutrition method for near the end of life, due to greater familiarity with the real role of that method. However, the rate of wanting tube feeding for themselves was relatively high among professionals in contact with over 30 tube-fed patients during the past three months (Fig. 1). Such professionals may be getting accustomed to tube feeding as end-of-life treatment due to this environment.
In contrast, professionals' wishes for their families differed from those for themselves. Almost half of the professionals were receptive to the idea of tube feeding for members of their family. This result may show the presence of emotional influences, such as wanting to do something for their family. Aita and co-authors reported that factors such as legal barriers, emotional barriers, and cultural values that promote family-oriented end-of-life decision making influence Japanese physicians' decision to provide tube feeding. 8 The rates at which professionals wished to administer peripheral infusion and central intravenous infusion for their families were higher than for themselves. This result may also be affected by these decision making factors. In addition, all rehabilitation doctors, anesthetists, and residents among the respondents wanted tube feeding for their families, thus indicating that these specialists might tend to believe in the advantages of tube feeding even for end-of-life care.
In this study, 65.2% of professionals welcomed the idea of a card-based declaration of intent for end-of-life care. This was consistent with the result that one-quarter of all respondents wanted to delegate decisions for their own end-of-life feeding to their family. They may recognise that the right of self-determination at the end of life belongs not only to themselves, but also to their family. A simple explanation of this result with the aphorism ‘Do unto others as you would have them do unto you’ is therefore difficult.10,11
Several limitations must be considered when interpreting the results. First, the scenario outlined for this questionnaire involved a bedridden state with irreversible dementia in old age, differing from locked-in syndrome or brain death in a narrow sense. Second, this survey was targeted to professionals at only five hospitals in Japan. Therefore, a nationwide survey may be required.
In conclusion, this survey revealed that many medical doctors and nurses rejected tube feeding for themselves, and that working at a municipal hospital and a higher frequency of taking care of tube-fed patients were significant predictors for rejecting tube feeding for themselves. All health care professionals should inform patients and their families of the results of the present survey, in addition to the existing evidence of the disadvantages of tube feeding in elderly with advanced dementia. While it is difficult to resolve emotional issues, the formulation of appropriate guidelines for artificial nutrition near the end of life is therefore required. It is hoped that attitudes of the medical establishment and health insurance system will be adjusted based on the evidence.
Footnotes
Acknowledgments
We wish to thank Dr. Kenji Umeki, Dr. Hiroaki Oka, Dr. Atsuko Iwata, Dr. Ryo Shirai, Dr. Kenji Kishi, and Dr. Issei Tokimatsu (Internal Medicine 2, Oita University Faculty of Medicine), Dr. Noriyuki Ebi, Dr. Ou Yamaguchi, Dr. Yukihiro Sugimoto, and Dr. Kosuke Tsuruno (Respiratory Medicine, Aso Iizuka Hospital), Dr. Toshiharu Ueno (Internal Medicine, Aso Iizuka Hospital), Dr. Eiji Okabe, Dr. Taisuke Matsumoto, Dr. Tomohiko Iwashita, Dr. Tetsuo Tsubone, Dr. Bunroku Matsumoto, and Dr. Hajime Miyajima (Internal Medicine, Tenshindo Hetsugi Hospital), Dr. Yoshiaki Mori (San-ai Medical Center) for their advice and expertise.
Author Disclosure Statement
No competing financial interests exist.