Whither palliative care?

Palliative care (pc) has evolved from a field of medicine devoted exclusively to end-of-life care into one that focuses on symptom management and quality of life throughout terminal illness. The patient population has also changed, with an increasing proportion of patients having non-cancer diagnoses and frailty. Advances in technology are providing new opportunities to identify health system issues and to improve patient care in this area. Here, we discuss some of the changes in pc over recent years and some of the newer approaches and ideas for delivering high-quality pc to patients and populations in the future.

From Early PC Referral to a Timely Palliative Approach

Recent pc research has featured randomized controlled trials of protocol-based early referral to a specialist pc team compared with usual care (that is, referral at the judgment of the most responsible physician). Those trials have shown that, compared with usual care, early pc referral can improve the patient’s quality of life, increase completion of advance care directives, reduce lengths of hospital stay, and decrease the economic burden, ultimately leading to current guidelines that recommend early integration of pc for all. However, not all studies found benefits from early intervention, and systematic reviews have identified important methodologic concerns, non-standardized or unclear interventions, and heterogeneous populations¹. Notably, there was no common definition of “early,” and usual care was often poorly described. Those concerns limit the generalizability of the studies to other contexts and populations.

But even if the validity and generalizability of the findings are accepted, early pc consultation faces a logistics obstacle. With almost 1% of the population dying annually, there will never be enough specialists and allied health team members to deliver this intervention to even a fraction of the individuals who might benefit, early or late. Instead, the challenge is to define a clear “palliative approach” that can be realistically scalable, to identify patients who would benefit from pc, and to determine the factors that are most predictive of that potential benefit. Shifting the focus from early consultation to timely intervention will incorporate pc at the time that the patient will benefit the most. Furthermore, increasing use of generalists, nurse practitioners, and clinical nurse specialists to provide pc will help to reduce the burden faced by specialist pc physicians and will allow for timely access to more expert pc when required. Future work at the population level will provide an improved understanding of the optimal time for specialist pc involvement and will help to identify the patient factors that can be managed successfully by general practitioners. That switch in focus will, in turn, maximize the finite resources for the pc population and lead to improved patient outcomes.

Rethinking Symptom Management

In recent years, the demographics of pc patients have shifted. Providers of pc are being asked to treat the growing number of cancer survivors, patients with nonmalignant diseases, and patients with histories of chronic pain, addiction, and mental health disorders. The traditional approach of pc providers, which features liberal symptom-targeted opioids and other medications, might not be appropriate for such patients. The new population requires a more tailored approach and consideration of nonpharmacologic and non-opioid alternatives for pain management over the longer term. Additionally, the understanding of how opioids are used by long-term pc patients and cancer survivors is limited and guidelines for de-prescribing remain unclear.

Amid the opioid crisis, the prescribing practices of pc physicians have been questioned. Early pc often leads to an earlier introduction of opioids and, ultimately, higher opioid doses before death². Although pc accounts for only a small fraction of total opioid prescriptions, the doses and schedules used can appear concerning to pharmacists and oversight bodies. Studies into the introduction and maintenance of opioid therapy and its contribution to opioid misuse are crucial to guide pc prescribing practices in the future.

The legalization of medical assistance in dying in Canada has renewed the focus on the problem of existential suffering, and the lack of effective pc interventions to treat it. Loss of autonomy and dignity, and fear of the unknown, significantly affect patient quality of life and are known to be major motivating factors in the pursuit of medical assistance in dying³. Currently, pc treatment for intractable suffering (both physical and psychological) is palliative sedation, in which a patient is given sedative medications to reduce consciousness, often for the remainder of their life. However, institutional guidelines and physician practices vary substantially with respect to indications for sedation, medications used, and the depth of sedation that should be targeted. Providers of pc would benefit from clearer guidelines, protocols, and oversight of this practice.

For many years, the role of psychedelics such as lsd, psilocybin, and ketamine in the management of existential distress has been considered. Although promising studies in the management of posttraumatic stress disorder and other mental health disorders have been published, the application of psychedelics in the pc population has been limited. Case reports suggest some benefit from psychedelics in appropriate pc patients, with limited drug interactions and fewer side effects than are seen with sedative medications⁴. The next logical step is to conduct large-scale multicentre studies to evaluate that approach in pc. As the practice of medical assistance in dying evolves and the understanding of existential suffering deepens, providers of pc will have to re-evaluate psychedelics and sedation.

Embracing Technology

Technological innovations have affected many aspects of medical care since the middle of the last century. Such innovations play an important role in pc, on both a micro and a macro level. Point-of-care ultrasonography (pocus) involves the use of ultrasound at the bedside for either diagnosis or procedural assistance. It has become more common in recent years and is the standard of care for providers in the intensive care unit and emergency department. In pc, pocus is not widely used, but might play a role in the diagnosis and drainage of symptomatic pleural and peritoneal fluid collections. Currently, providers of pc might be reluctant to drain such collections because of inexperience or fear of complications. Instead, they increase analgesia or transfer the patient offsite for guided drainage by a radiologist. Both approaches are suboptimal because of side effects, cost, and emotional burden on the patient. The use of pocus could allow providers of pc to more accurately diagnose and manage symptoms, while reducing costs and side effects. Future work in this area will have to focus on establishing both competence and comfort with pocus among providers of pc, given that so few currently have experience with this technology.

On a broader scale, an increasing number of studies are looking at population-level administrative data (“big data”) to learn about the use of pc in large groups of people over a large geographic region. In jurisdictions such as Canada, where administrative data are relatively rich and centrally collected, big-data approaches have facilitated a deep understanding of how many people at end of life are receiving pc, when they are receiving pc, and which factors are associated with the involvement of pc services⁵. Such studies can help to identify geographic and demographic areas of need, and possible targets for improving the delivery of pc on a population level. Big data can also be applied to monitor system-level performance through quality metrics such as the provision of pc at home or the use of acute care in the weeks before death.

Big data also creates an opportunity to overcome the “denominator challenge”⁶: the ability to prospectively identify the population that is nearing the end of life with a high burden of unmet palliative needs. Provider-dependent approaches such as the “surprise question” are cheap and easy to use, but can have low accuracy and reliability when applied at the population level⁷. Using administrative data to create automated algorithmic approaches to patient identification could improve the real-world effectiveness of virtually any palliative intervention and guide quality improvement efforts in pc.

Telemedicine has been available for decades and is another promising approach for improving the delivery of pc to populations, particularly when resources are more limited or population density is low. It can involve electronic communication between professionals (for example, teleconsultation) or with patients and family members (for example, mobile health). Studies of telemedicine in contexts other than pc have proved to be methodologically challenging, and the results of quality and cost-effectiveness analyses are inconsistent⁸. However, the challenge of delivering high-quality pc to populations spread over large areas is obvious, and telemedicine approaches will clearly be necessary. As with many implementation efforts, the challenge is not to determine whether a technology “works” in general, but how that technology can be adapted to support service provision in a specific area at a specific moment in time.

Ultimately, pc will continue to evolve. Reliance on technology and novel approaches to symptom management will be increasing. The use of big data has the potential to streamline the understanding of unmet needs in the pc population and to target future research and practice models accordingly. Understanding and embracing those changes will allow for an increasingly personalized, quality approach to the pc patient in future.