To Force Feed the Patient With Dementia or Not to Feed: Preferences, Evidence Base, and Regulation

Author(s): 

Speakers: Michael D. Cantor, MD, JD, Jeanie Kayser-Jones, RN, PhD, FAAN, and Thomas E. Finucane, MD

Michael D. Cantor, MD, JD, Special Assistant to the Director, National Center for Ethics, Department of Veterans Affairs, Washington, DC, discussed when and if food and water can be withheld from a nursing home resident. There are three important considerations when making this determination: autonomy, surrogate decision-making, and the best-interest approach.

Dr. Cantor used a case discussion to illustrate a person who fits the common profile of someone who resides in a nursing home. Mr. Smith is an 89-year-old resident with advanced Alzheimer’s disease (AD) who is dependent for all of his activities of daily living (ADLs) and requires assistance with eating and drinking. His wife of 60 years is his health care proxy and surrogate. She has recently asked staff members if they would stop feeding Mr. Smith as “he would not want to live like this.”

The first question to be answered, according to Dr. Cantor, is regarding a patient’s autonomy and the authority of his family members to make this decision. The basic definition of autonomy is the concept that people have the right to make their own decisions. The idea of autonomy stresses that each person knows what is best for him or her as an individual, and under American medical practice and the law, it is recognized that patients have the right to choose what health care they do and do not accept, even when this decision can impact their length of life.

The above-mentioned case brings into question whether someone who has a terminal illness—in this instance,  advanced AD—can decide to stop eating and drinking, especially since this patient clearly lacks decisional capacity and does not have insight into the risks, benefits, or alternatives to his decision. Does this patient lose his autonomy to make this decision, or does someone else make that decision on his behalf?

The choice to eat and drink, unlike other decisions that health care proxies are usually asked to make, is not really a medical decision, according to Dr. Cantor. These actions fall into basic activities of living, aspects that are needed to continue to live, but are not necessarily technical or medically- directed. Therefore, the speaker asked, can people decide to have this nonmedical care withheld? Law and medical practice dictate that people have absolute power to make decisions about what medical treatments they choose or reject. In addition, it must be determined how much importance should be placed on a person’s previous wishes regarding eating and drinking. “Can we be sure that the person who now has dementia would have been able to predict  what he really would want in this condition?” questioned Dr. Cantor.

Generally speaking, surrogates are the best determinants for assessing what the patient would want; for example, in the case study, this person would be Mrs. Smith. However, the speaker stressed that the question remains as to whether some decisions are so fundamental to the care provided that others should not be allowed to make them. Mrs. Smith may not truly be capable of ascertaining whether Mr. Smith is being overburdened by eating and drinking and if he would really want to be “put out of his misery.”

When it becomes impossible to determine what someone would have wanted based on his or her prior decision, the “best-interest analysis” is employed by looking at this particular patient in the context of other patients who have received similar care or experienced similar circumstances and determining what the best outcome was for them. Dr. Cantor noted that,  generally speaking, it is difficult to argue that it is not in someone’s best interest to eat and drink, although there are exceptions to this likelihood. For instance, when the burden of eating and drinking is overwhelming, such as when the patient cannot swallow, chokes, or aspirates the food, it may not be in the patient’s best interests to be subjected to actions that are potentially life threatening in light of infection risk.

Dr. Cantor concluded by stating that in most instances, he does not believe patients should have food and water withheld, and that it is not an action a surrogate can decide to waive. Autonomy remains an important part of patient care decisions, but in the case of the above-noted patient, his autonomy is clearly limited. Offering basic support for human beings is a critical part of the health care profession, stressed Dr. Cantor, and this obligation cannot be fulfilled if food and drink are withheld.

**sub**Factors Contributing to Malnutrition, Dehydration, and Weight Loss in Nursing Homes: Empirical Data **endsub**
Dr. Jeanie Kayser-Jones, RN, PhD, FAAN, Professor, Department of Physiological Nursing and Medical Anthropology Program, University of California San Francisco School of Nursing, discussed an unpublished anthropological study in which she participated entitled “The Behavioral Context of Eating and Nutritional Support.” The study sought to investigate eating problems and the use of feeding tubes in nursing homes. Three strategies were utilized to obtain data. Study participants were observed, families, nursing staff, residents, and physicians were interviewed, and using event analysis, 100 residents who were not eating well were followed up two to three times a week at meal time for a period of about six months. Each resident had a general examination and a bedside dysphagia screening; a dietary analysis was conducted on 40 of these 100 residents.

Nutritional care was influenced by many factors, according to Dr. Kayser-Jones. There was both a lack of attention to individual food preference and to the provision of ethnic food choices. Although 33% of study participants belonged to a minority group, only western food was offered at meals. For example, the speaker recalled a Chinese resident who did not like western food, and whose weight subsequently fell from 102 pounds to 71 pounds during a six-week period that ended with his death.

Swallowing disorders also affected nutritional intake. Of the 45 residents who had a swallowing disorder, only 10 were referred for evaluation to a speech pathologist. This type of disorder requires that someone be fed slowly, but due to inadequate staffing, many residents were fed quickly and not given enough time to swallow, leading to coughing, choking, and in  some cases, aspiration pneumonia. Dysphagia was complicated by the fact that many residents had poor oral health—51% had no teeth, just a few teeth, or no dentures. Sixteen percent had dental caries, 15% reported pain, and 7% had oral lesions. As residents without teeth are usually placed on a pureed diet, the speaker noted that this blending could often result in very unappetizing food.

Inadequate staffing and lack of supervision was the most significant factor that influenced how people ate. Because it takes time to move residents from their beds to the dining room, many would eat in an improper position in bed, and due to minimal staffing, residents would be forced to eat quickly. If the meals were being consumed too slowly, some of the staff mixed all the pureed food together in a glass of milk and forced them to drink this rapidly. “Not only was this unpalatable, it was unsafe. Many coughed and choked when fed so quickly,” stated Dr. Kayser-Jones. Some residents received little or no food at all, and even if the trays were taken into the room, no one was there to feed them.

A dietary analysis has also been conducted to investigate the factors that influenced fluid intake among nursing home residents who were not eating well (Kayser-Jones J, Schell ES, Porter C, et al. J Am Geriatr Soc 1999). Researchers measured the amount of fluid residents drank. Only one out of 40 residents drank an adequate amount, some residents drank nothing at all with their meals or between meals, and some people went for 20 to 30 hours without anything to drink.

Dr. Kayser-Jones and other investigators are currently conducting a study into factors that influence the care of terminally ill residents. Again, they are finding that residents often are not fed and do not receive adequate liquids. The patients would often complain of thirst and ask for something to drink, with containers of water sometimes out of reach. The speaker relayed the story of an 82-year-old woman who came to the facility every day for seven years to feed her husband, because she would “wonder about what happens when I am not here.”

When patients with cognitive impairment do not eat well and begin to lose weight, families and professionals must decide whether to insert a feeding tube. A study by Finucane and colleagues (JAMA 1999) reviewed the literature on feeding tubes from 1966 through March 1999, and found no data to suggest that tube feedings improve clinical outcomes. They concluded that tube feedings should be discouraged for patients with severe dementia.

Dr. Kayser-Jones noted that while there has been considerable discussion on the use of feeding tubes, very little attention has been focused on the preventable factors that contribute to malnutrition, dehydration, and weight loss. “The lack of attention to problems such as inadequate staffing and lack of professional care in nursing homes leads me to ask if we have a tendency to ignore the obvious day-to-day ethical issues and focus instead on more complex, philosophical questions,” the speaker stated. She stressed that as professionals, there exists an ethical responsibility to intervene when the health and well-being of nursing home residents are at risk.

She concluded her presentation by stating that for people with AD and other dementias who have high levels of dependence, physicians and nurses must ensure that the basic needs, such as those for food and water, are met, and that these individuals are treated with dignity and respect during the final days of their lives.

**sub**Evidence Base for the Use of Feeding Tubes in Patients with Dementia **endsub**
Thomas E. Finucane, MD, Professor of Medicine, Johns Hopkins University School of Medicine, Johns Hopkins Bayview Medical Center, Baltimore, MD, discussed the data pertaining to the use of feeding tubes. “The clinical goals that we are shooting for are to prevent aspiration pneumonia and prevent the sequelae of malnutrition,” he stated. “The sequelae of malnutrition might reasonably be death from starvation, bad pressure sore outcomes, infections of other kinds, poor function and palliation, hunger, and thirst in a patient who cannot report his or her symptoms.”

Aspiration pneumonia is often a common reason to insert a feeding tube. However, Dr. Finucane stressed, there are basically two conceptual definitions of aspiration pneumonia. First, there is the misdirection of contaminated oral secretions into the lung, causing a mixed anaerobic and gram-negative infection in dependent segments of the lungs. Secondly, there is Mendelson’s syndrome, which is gastropulmonary regurgitation. The question to consider, the speaker stated, is whether the placement of a feeding tube will prevent either one of these events. He asserted that no published studies have presented data that have found any reduction in the risk of pneumonia by putting in a feeding tube, even in patients with a recurrent aspiration pneumonia.“There is no reason to imagine that we are helping this patient avoid pneumonia,” he stated. “We are not going to keep oral secretions out and we are not going to keep regurgitated material out.”

Dr. Finucane then examined the data pertaining to the use of a feeding tube to prevent malnutrition and thus improve survival rates. Multiple studies conducted on more than 80,000 patients have all shown no improvement in mortality rates for those fed by tube. “The overwhelming weight of evidence,” the speaker stated, “suggests that there is no benefit as far as survival.”

There are no randomized controlled clinical trials that have shown that patients with dementia who receive feeding tubes do better in terms of preventing pressure sores or healing those that are already present. No data exists showing that the types of infections found in the nursing home setting (such as cellulitis, urinary tract infection, influenza, and other viral infections) occur less commonly in patients with feeding tubes. On the contrary, stated the speaker, there are many documented cases of infectious complications of nose, middle ear, throat, pleura, lung, gut, peritoneum, abdominal wall, etc, where the cause of the infection was the feeding tube. When examining function, again, no randomized controlled trials provide any evidence that a patient who is permanently recumbent with dementia and is given a feeding tube will get stronger and become more able to provide care for him or herself.

Finally, the speaker examined feeding tubes in relation to palliation. A study conducted by  McCann and colleagues (JAMA 1994) examined patients who were dying of cancer and were asked on a daily basis whether they were hungry or thirsty, and then sought to find out what it would take to palliate them. The investigators found that in this population, careful oral hygiene and the use of a small amount of ice chips was almost 100% palliative.
 
In conclusion, the speaker noted that there are many common decisions in geriatric medicine, such as whether to use a ventilator or to conduct surgery to relieve a bowel obstruction, where two outcomes occur. If the decision is made to conduct the procedure, it may allow the patient to live longer but in more pain or emotional discomfort. If the decision is made to simply rely on palliative methods, comfort and dignity may be increased, but ultimately lead to a greater chance of death in the short term. However, Dr. Finucane opined, the decision about tube feeding is not one of those decisions. In these instances, even if all motivated and comprehensive efforts have been made to improve a patient’s eating habits, there is no compelling data that have shown insertion of a tube to be beneficial. In the face of negative evidence, the speaker stated that it is important to focus on what can and should be changed in nursing homes to improve the quality of care.