Palliative and End-of-Life Care: Vital Aspects of Holistic Diabetes Care of Older People With Diabetes

The three terms, palliative, terminal, and end-of-life care, are defined in Table 1, along with some suggestions for when to implement palliative and end-of-life care and some implementation strategies that may be used.

Historically, palliative care was associated with hospice and cancer care at the end of life and did not include curative care. Consequently, many clinicians equate palliative care with terminal care (17) and regard death as a treatment failure. Significantly, though, palliative care can be commenced at any time and can be used with a patient’s “usual” care (18,19). Generally, the earlier palliative care is implemented, the greater the benefit (18–20).

Many diabetes clinicians are not familiar with palliative and end-of-life care or comfortable discussing these issues with people with diabetes and their families. The National Coalition for Hospice and Palliative Care has noted that palliative care is the responsibility of all clinicians—not the sole preserve of palliative care specialists (20). The coalition’s guidelines recommend that service providers and clinicians integrate palliative care into the care plans of people with serious illnesses, including proactively discussing end-of-life care (20). However, clinicians often wait until a patient or patient’s family raises the issue. Likewise, the patient and family often wait until the doctor initiates such conversations. Consequently, nobody raises the issue, and conversations about palliative and end-of-life care do not occur (21).

Optimal palliative care requires many conversations with individuals about their values and care goals as their health status changes. Thus, communication and advance care planning is a continuing process and a key aspect of palliative care. However, many clinicians find it difficult to discuss these issues because of inadequate skills and training, time constraints, fear of conflict (17), and the challenge of prognostication. Clinicians’ intuitive prognostication can be inaccurate, and prediction tools are not particularly helpful (21).

Palliative care can be delivered in a person’s home, in a hospice, or in the hospital (Table 2). Many people prefer to die at home, but most do not actually do so (18,22).

Two guidelines and one article focus on palliative and end-of-life diabetes care (5,13,23), although the topic is also included within several other general guidelines. Most of the guidance in this area is based on low-level evidence and expert consensus; however, it makes clinical sense, is consistent with palliative care guidelines, and focuses on personalised care. Table 1 defines end-of-life care and outlines care strategies for people with diabetes requiring terminal and end-of-life care.

As previously noted, the care focus must gradually shift from tight blood glucose control to comfort and symptom control for patients with either type 1 or type 2 diabetes (5,8,12,13,23). Preventing hypoglycemia and hyperglycemia and their associated uncomfortable symptoms is essential. Both states are associated with increased risks of death, although hypoglycemia might be a marker, but not a cause, of poor outcomes, at least for people hospitalized after myocardial infarction (24). It might also be important to reduce glucose variability, especially during disease exacerbations (25).

Glucose-lowering medication doses may need to be reduced or different formulations used, while also simplifying the regimen as much as possible. Glucose-lowering medications are used to prevent hyperglycemia and its symptoms and reduce the risk of diabetic ketoacidosis or hyperglycemic hyperosmolar state. That is, they can be used with a palliative intent for symptom control in older people with type 1 or type 2 diabetes rather than to achieve tight blood glucose control. Preventing diabetes complications is largely irrelevant in older people approaching the end of life.

A generally accepted safe blood glucose range is 6–11 mmol/L (108–198 mg/dL) (13,23), although some Australian guidelines recommend 6–15 mmol/L (108–270 mg/dL) (4,5). Sliding-scale insulin (i.e., hyperglycemia correction doses) should not be used alone but may be a useful add-on to a basal-bolus insulin regimen. Sliding-scale insulin is more likely to cause glucose variability and hypoglycemia that can lead to a hospital admission (5,13,25) and is a predictor of death in critical illness (25). Transitioning to a basal-bolus insulin regimen can help avoid hypoglycemia and reduce the care burden on the patient, family, and care staff (5,13).

Usual prescribing guidelines and algorithms should be used to manage all classes of medications and individual drugs within each class. However, the following cautions apply with regard to glucose-lowering medicines (5,13,23).

• Metformin is generally safe but may be contraindicated in people with gastrointestinal cancers and other gastrointestinal problems.

• Sulfonylureas increase the risk of hypoglycemia, especially in people with low oral intake and/or liver disease. This risk is further increased if insulin is used with a sulfonylurea.

• Pioglitazone may be contraindicated in people with or at risk for bladder cancer.

• Dipeptidyl peptidase‐4 (DPP-4) inhibitors have low hypoglycemia risk alone but can increase the risk of hypoglycemia if used with sulfonylureas. Some DDP-4 inhibitors should be used with caution or stopped in people with renal disease.

• Glucagon-like peptide 1 receptor agonists have low or no hypoglycemia risk but should be stopped in patients who develop pancreatitis or abdominal pain.

• Sodium–glucose cotransporter 2 inhibitors have a low hypoglycemia risk. They should be stopped when patients develop an intercurrent illness or require surgery because of the risk of dehydration and associated hyperosmolar or ketoacidotic states. Ketone testing should be started if it is not already used.

Blood glucose monitoring can be a useful guide to the underlying cause of symptoms such as nausea, headache, and urinary frequency in people with type 1 and type 2 diabetes (26). Our research shows that people in the last stages of life value blood glucose monitoring because it enables the identification and treatment of the causes of some symptoms (27). It is also useful when corticosteroids are used (5,13,23). Monitoring frequency should be tailored to the action profile of the glucose-lowering medications and the individual’s food intake and activity levels. Food intake can be erratic for a range of reasons, including dementia, dysphagia, changed taste sensation resulting from aging or chemotherapy, and dissatisfaction with the food served in hospitals, aged care homes, or long-term care facilities.

The terms and processes used for advance care planning vary around the world to accommodate legislation, language, and religious and cultural imperatives. However, the purpose remains constant: to ensure that the individual has a good death. Diabetes clinicians are ideally placed to help older people with diabetes clarify their values, preferences, and care goals and to document them in an advance care plan because diabetes care requires regular follow-up appointments and screenings for complications.

Advance care planning involves people with diabetes discussing their preferred future care with clinicians and family members while they can make autonomous decisions (informed consent). Documenting the person’s values—the things that give meaning and purpose to the person’s life—is extremely important. Values are a useful guide later on to the type of care a person would choose to achieve quality of life.

People value having some control over the way they die, being free from pain, and having time to complete important business and to say goodbye to family and friends (5,22,27). For example, achieving personal goals such as attending a grandchild’s wedding may be more important than living a longer life (28). Thus, care should support people in achieving their goals, when possible. Significantly, people who document their values and goals and instructions about the care they want or do not want are more likely to receive care consistent with their preferences. Discrepancies between the values of patients and clinicians are well documented (29). For example, clinicians underestimate the burden of symptom severity compared with patient ratings, and many tools that measure such outcomes do not adequately capture the patients’ perspective (28,29).

Advance care plans are guides to and decision aids for determining an individual’s values and care goals. They can help clinicians and patient surrogates or caregivers make difficult treatment decisions when patients cannot decide for themselves, provided they meet legal requirements (e.g., they are signed and witnessed) (21,30). As previously noted, clinicians often miss opportunities to initiate conversations to help people make such plans (31). Indeed, some such conversations occur during rapid response or medical emergency interventions (32), which is clearly not an appropriate time to make decisions about emotion-laden issues.

Some clinicians believe that people do not want to discuss dying and advance care planning, and in some cases, this is true. Generally, however, people are more open to such discussions when they have disabilities and illnesses (33). Our research shows that people with diabetes want to discuss their end-of-life care in their last months of life, but they are often reluctant to initiate discussions because they worry about upsetting their clinicians and because clinicians do not ask them (27,34).

Most patients, caregivers, and clinicians want some idea of prognosis when a patient’s condition deteriorates, and not considering prognosis could result in lower-quality care (35). Prognostication, like deciding benefits and harms, is challenging, however. As Hippocrates said, “It appears to be a most excellent thing for the physician to cultivate prognosis” (16).

Palliative and diabetes prognostic tools that can help clinicians “cultivate prognosis” include the Supportive and Palliative Care Indicators Tool (36) and The Gold Standards Framework Prognostic Indicator (37). These tools can help clinicians identify people who would benefit from palliative care. Clinicians can then provide palliative care themselves or refer patients to specialist palliative care services (5), as well as help the identified patients develop or update their advance care plans, if necessary.

People with diabetes and their families need to make many complex decisions about their diabetes self-care on a daily basis and when considering palliative and end-of-life care. We interviewed older people with type 1 or type 2 diabetes receiving palliative care, their family carers, and palliative care clinicians when we developed the Guidelines for Managing Diabetes at the End of Life (38). Our findings suggest that people with diabetes and their families consider blood glucose testing to be a useful guide to identifying hypoglycemia and hyperglycemia and enables early recognition and treatment of their associated unpleasant symptoms (27). Blood glucose monitoring is also a familiar routine for many people with type 1 diabetes during a time of change.

We also found that most people wanted to continue taking their glucose-lowering medications until the terminal stage to manage hyperglycemia, which can cause confusion, frequent urination, and tiredness and compound other symptoms such as delirium, all of which cause discomfort and reduce quality of life. They also wanted to avoid hypoglycemia because of its associated loss of control and symptom burden (27,34).

Patients’ family members, mostly spouses, were concerned about having to undertake diabetes tasks such as blood glucose monitoring and insulin administration when their relative could no longer manage (27). They said they had not received education about how to do these tasks and worried they could accidently give the wrong insulin dose and hasten death. Underlying this concern was the risk of unintentionally causing death and being accused of murdering their relative.

Documenting an advance care plan is time-consuming and distressing for individuals, their family members, and clinicians. However, advance care planning specialists, when available, can help reduce the stress. Helping people create a life narrative enables them to find meaning and purpose, develop a sense of coherence, and maintain their identity (personhood), self-esteem, and coping capacity as they prepare for their end of life (39).

It is important to understand individuals’ journey with diabetes and changing care needs as they transition through older age and toward the end of life. Helping people with diabetes cope with a life-limiting illness is a privilege. It is also an ethical responsibility, given the current focus on autonomy, patient engagement, and personalized care. Generally, a more flexible approach to food choices, blood glucose monitoring, and the blood glucose target range is preferable to achieving tight glycemic control. The specifics of care must be decided with individuals and their families. Prognostication is challenging and contributes to moral distress. Thus, clinicians and people with diabetes need to know how to or get help to manage uncertainty and cope with such distress.

Documented values, care goals, and preferences can help reduce uncertainty and aid decision-making when older people cannot decide for themselves and can be used in conjunction with guidelines such as the Gold Standards Framework (37). Such plans do not need to be completed all at once; starting early allows people time to think through the issues involved. Once a plan is documented, it should be reviewed regularly because peoples’ goals of care can change, although their values tend to remain relatively constant. Significantly, living as long as possible is not always the most important goal. For many people, avoiding pain and suffering and having a good death are more important (28).

The author thanks her colleague, Professor Peter, for allowing her to adapt the information in Table 2 from their shared work. The author thanks Georgie Lee for her help with the reference list and manuscript submission. The author further acknowledges the people with diabetes at the end-of-life stage who have shared their experiences with her team.

No potential conflicts of interest relevant to this article were reported.

T.L.D. is the sole author and guarantor of this work and takes responsibility for the integrity of the information presented.