Abstract
Background:
The palliative care population is prescribed a large number of drugs, increasing as patients deteriorate. The cumulative effects of these medications combined with underlying symptom burden can result in significant morbidity. There is an urgent need to describe possible symptomatic events that could be exacerbated by commonly prescribed drugs in palliative care and their impact.
Objectives:
To trial the feasibility and acceptability of determining baseline symptomatic event rates for community palliative care patients from which a composite measure of symptomatic events can be developed.
Design:
This prospective pilot study of patient-reported symptomatic events recruited a convenience cohort of 27 community palliative care patients in a metropolitan specialist palliative care service in Australia.
Results:
This study has demonstrated a high prevalence rate of symptomatic events (total crude event/participant day rate 0.87) in the study population.
Conclusion:
Data collection of patient-centered symptomatic events was acceptable and feasible to participants. This pilot supports a fully powered study.
Introduction
Symptomatic events that may be iatrogenic have not been well characterized in the palliative care population. 1 There is a risk of attributing these events solely to progression of the underlying life-limiting illness or associated comorbidities, whereas a component could be a direct result of medications. At times, this may be more than individual medications. For example, the cumulative anticholinergic effects of medications and their metabolites can be associated with significant morbidity. 2 Recent studies in palliative care populations (consultative, inpatient, and community) report patients are prescribed, on average, five medications per individual.3–6 Furthermore, there is a tendency for the number of medications prescribed to increase as death approaches. 3
There is a need to describe symptomatic events that have the potential to be, in part, related to commonly prescribed medications used in palliative care and the impact of these events. 7 Systems for identifying the rate of these events and monitoring their effects are required.
Results of a retrospective case note review of a palliative care inpatient population demonstrated that symptomatic events impacting peoples' mobility and independence as well as patient and caregiver confidence appear to be ubiquitous in the palliative care population. 7 This study was retrospective and limited to events recorded in the clinical notes of palliative care inpatients, therefore, potentially underestimating the rate. Further investigation is necessary to identify prospectively community palliative care patients' reported event rates. Although impossible to determine the extent to which symptomatic events are a direct result of medications, it is necessary to begin to understand the frequency and severity of such events across the whole palliative care population.
This study builds on previous work conducted in the inpatient setting. 7 It aims to assess the feasibility and acceptability of determining baseline symptomatic event rates for community palliative care patients. This study can help to inform the design of future studies to investigate whether rates and severity of symptomatic events can be reduced systematically.
Methods
Study setting
This study engaged a convenience cohort of community palliative care patients in a metropolitan specialist palliative care service in Australia. Ethics approval was granted from the hospital human research ethics committee.
Participants' eligibility
People living in the community who were newly referred to the specialist palliative care service, aged >18 years, clinically assessed to have a minimum prognosis of three or more months, and with sufficient English language skills (or caregiver) to complete the study diary, were eligible for this study if they consented to participate. Exclusion criteria included inability to give informed consent or complete the study diary, a Mini-mental State Exam (MMSE) <24/30, 8 or severely restricted function at baseline (Australia-modified Karnofsky Performance Status [AKPS] <50%). 9
Symptomatic events
This study prospectively evaluated a list of predefined symptomatic events using a daily self-assessment diary for eight weeks. The diary had a list of symptomatic events that had previously been used in the inpatient audit. These were chosen in the previous study because they may all be related to prescribing, especially for comorbid illnesses, or contribute to decline in function 10 : light headedness/dizziness; feeling muddled; difficulty remembering; difficulty concentrating; vivid dreams/hallucinations; nightmares; and nighttime confusion, blurred vision, falls, headaches, and easily irritated. Each event was reported as present or absent for each day the diary was used.
Data collection
Baseline demographic data were obtained from the medical record. A study nurse performed a face-to-face baseline assessment, followed by weekly phone assessments. A suite of data collection tools were used to evaluate feasibility of data collection and provide baseline normative date:
Baseline only (a) Charlson Comorbidity Index
11
(b) Folstein MMSE
8
Baseline and weekly (a) AKPS
9
(b) Resource utilization groups—activities of daily living (RUG-ADL) scale (c) The Australian National Subacute and Non-Acute Patient (AN-SNAP) phase
12
(d) Symptom assessment scale (SAS)
13
(e) McGill quality of life
14
(f) Life space assessment
15
(g) Medication review Weekly only (a) Clinical events checklist Daily (a) Self-assessment diary
Data analysis
A descriptive analysis was undertaken to establish a total crude rate of symptomatic events. A crude rate was calculated for each symptomatic event. The following group of symptoms was collapsed into one “event” to account for the possibility of symptom crossover: hallucinations, nightmares, nighttime confusion, blurred vision, and feeling muddled.
Results
Acceptability and feasibility
Participation and response rate
Thirty-three participants were enrolled in this study. Six patients were unable to complete the diary due to clinical deterioration and were excluded from the analysis.
Patient characteristics
For the 27 participants considered in this study, 16 (59%) were male, the median age was 69.3 years (interquartile range [IQR] 59.5–78.2), the median Comprehensive Complication Index (CCI) was 6.5 (IQR 3–9), and 25 (93%) had cancer as their life-limiting illness. At baseline, the median AKPS was 60 (IQR 60–70), total RUG-ADL scale was 4.6, and participants were prescribed a median of 9.0 (IQR 5.75–10.25) regular medications.
Symptomatic events
A composite measure for a crude event rate was created by examining the sum of all events during the study (1603 events) divided by the number of days reported for all symptomatic events (1848 participant days), giving a total crude event rate of 0.87 events per participant day). At baseline (on admission to the specialist palliative care service), only six patients reported none of the specific symptoms. In total, only 14% of days had none of these symptoms reported (Table 1).
Prospectively Collected Self-Reported Symptomatic Events in Community-Dwelling Hospice/Palliative Care Patients
Hallucinations, nightmares, nighttime confusion, blurred vision, muddled thinking.
Discussion
This pilot study has provided data to support the feasibility and acceptability of a prospective collection of patient-reported symptomatic events that may be related to prescribing for people late in life.
Drug interactions and medications assessed as “high risk” on Beers criteria have been linked with hospitalization and increased mortality in other populations such as the elderly and nursing home residents, but no association with anticholinergic medications specifically has been described in the palliative care setting.16,17 Apart from the traditional adverse effects of anticholinergic drugs such as dry mouth and constipation, cumulative anticholinergic load has been associated with delirium2,18–21; falls, reduced functional status, and impaired motor performance19,22,23; and poor cognitive outcomes (in particular, in those with prior cognitive impairment).10,15,24–28
This study has demonstrated a high prevalence rate of symptomatic events in people with an advanced life-limiting illness. Although it is impossible to attribute symptomatic events to one single factor, the importance of defining a cumulative baseline measure to monitor and evaluate this population for possible and preventable events is critical given that some of these events will be iatrogenic. Some of these events may be due to the life-limiting illness itself or comorbidities, but this should not be assumed without considering iatrogenic harms. Given the reduced physiological reserve of the community-dwelling palliative care population, an event such as a fall has the potential to impact irreversibly on the person's function and quality of life, and may even contribute to deterioration and premature death.
Limitations
The study has a number of limitations. The researchers predefined the events of interest. The data collection tool has not been validated. Thus, patients may have had other medication-related events not captured by the diaries. Furthermore, participants may have been unable to “isolate” specific symptoms, and there is a high probability of a crossover of several symptoms. As this study was a pilot, the numbers are too small to make generalizations based on our analysis. Nevertheless, this study has shown that patients themselves find these symptoms burdensome enough to document them. The study also highlights the ubiquitous nature of these symptomatic events in this population.
Given the lack of an international consensus on criteria for referral to palliative care services, the disease and functional status of palliative care populations varies widely. This is most obviously reflected in the time from referral to death. As such, the generalizability of these findings to other palliative care settings may be limited.
Further research
Our study provides the basis for future prospective studies that quantify the effect of reducing multiple medications to potentially reduce the cumulative effects of medications alone and in combination. Refinement of the data collection tool to take account of overlapping events would enable empirical investigation of interventions such as a structured model of reduction of multiple medications, to improve patient outcomes using patient-defined symptomatic event rates as an outcome measure.
Conclusion
The majority of patients are cared for in the community after it is recognized that they have a life-limiting illness for most of their remaining life. There is an urgent need to describe symptomatic events and their impact on these patients and their families in the palliative care setting. 7 This study has shown that it is feasible to collect data from patients living at home to explore event rates, especially with interventions designed to reduce these rates. The study points toward a significant symptom burden that warrants further investigation.
Footnotes
Acknowledgments
Thanks are due to all who participated in this study, at such a difficult time of life.
Funding Information
No funding was received for this study.
Author Disclosure Statement
All authors declare there are no conflicts of interests.