Effect of Regular, Low-Dose, Extended-release Morphine on Chronic Breathlessness in Chronic Obstructive Pulmonary Disease: The BEAMS Randomized Clinical Trial

Key Points

Question  Does regular, low-dose, extended-release morphine improve the intensity of worst breathlessness in people with chronic obstructive pulmonary disease (COPD) and severe chronic breathlessness?

Findings  In this randomized clinical trial including 156 people with COPD and chronic breathlessness, treatment with 8 mg/d and 16 mg/d of oral extended-release morphine for 1 week resulted in a mean difference of −0.3 and −0.3, respectively, for change in the intensity of worst breathlessness compared with placebo; neither difference was statistically significant. The scores range from 0 to 10 (a score of 10 being the worst or most intense) on the numerical rating scale that was used.

Meaning  Extended-release morphine compared with placebo did not significantly reduce the intensity of worst breathlessness in people with COPD.

Importance  Chronic breathlessness is common in people with chronic obstructive pulmonary disease (COPD). Regular, low-dose, extended-release morphine may relieve breathlessness, but evidence about its efficacy and dosing is needed.

Objective  To determine the effect of different doses of extended-release morphine on worst breathlessness in people with COPD after 1 week of treatment.

Design, Setting, and Participants  Multicenter, double-blind, placebo-controlled randomized clinical trial including people with COPD and chronic breathlessness (defined as a modified Medical Research Council score of 3 to 4) conducted at 20 centers in Australia. People were enrolled between September 1, 2016, and November 20, 2019, and followed up through December 26, 2019.

Interventions  People were randomized 1:1:1 to 8 mg/d or 16 mg/d of oral extended-release morphine or placebo during week 1. At the start of weeks 2 and 3, people were randomized 1:1 to 8 mg/d of extended-release morphine, which was added to the prior week’s dose, or placebo.

Main Outcomes and Measures  The primary outcome was change in the intensity of worst breathlessness on a numerical rating scale (score range, 0 [none] to 10 [being worst or most intense]) using the mean score at baseline (from days −3 to −1) to the mean score after week 1 of treatment (from days 5 to 7) in the 8 mg/d and 16 mg/d of extended-release morphine groups vs the placebo group. Secondary outcomes included change in daily step count measured using an actigraphy device from baseline (day −1) to the mean step count from week 3 (from days 19 to 21).

Results  Among the 160 people randomized, 156 were included in the primary analyses (median age, 72 years [IQR, 67 to 78 years]; 48% were women) and 138 (88%) completed treatment at week 1 (48 in the 8 mg/d of morphine group, 43 in the 16 mg/d of morphine group, and 47 in the placebo group). The change in the intensity of worst breathlessness at week 1 was not significantly different between the 8 mg/d of morphine group and the placebo group (mean difference, −0.3 [95% CI, −0.9 to 0.4]) or between the 16 mg/d of morphine group and the placebo group (mean difference, −0.3 [95%, CI, −1.0 to 0.4]). At week 3, the secondary outcome of change in mean daily step count was not significantly different between the 8 mg/d of morphine group and the placebo group (mean difference, −1453 [95% CI, −3310 to 405]), between the 16 mg/d of morphine group and the placebo group (mean difference, −1312 [95% CI, −3220 to 596]), between the 24 mg/d of morphine group and the placebo group (mean difference, −692 [95% CI, −2553 to 1170]), or between the 32 mg/d of morphine group and the placebo group (mean difference, −1924 [95% CI, −47 699 to 921]).

Conclusions and Relevance  Among people with COPD and severe chronic breathlessness, daily low-dose, extended-release morphine did not significantly reduce the intensity of worst breathlessness after 1 week of treatment. These findings do not support the use of these doses of extended-release morphine to relieve breathlessness.

Trial Registration  ClinicalTrials.gov Identifier: NCT02720822

Chronic breathlessness, defined as breathlessness at rest or with low levels of exertion that persists despite optimal treatment for the underlying conditions,¹ affects many people with chronic obstructive pulmonary disease (COPD). It often restricts activities of daily living and results in decreased physical activity and muscular and cardiovascular deconditioning, which further worsens breathlessness.² Chronic breathlessness is associated with anxiety and depression, reduced health-related quality of life,³ and increased use of health care services and mortality.⁴ Chronic breathlessness is often assessed using the modified Medical Research Council (MRC) scale.⁵

Opioids may reduce symptoms in people with chronic breathlessness and severe disease.^6,7 Meta-analyses of small, short-term trials suggest beneficial effects of low-dose opioids in people with chronic breathlessness,⁸ including those with COPD.⁹ However, a 1-week randomized clinical trial (RCT) demonstrated no statistically significant improvement in breathlessness with 20 mg/d of oral extended-release morphine vs placebo.¹⁰ Another RCT of people with COPD and chronic breathlessness reported improved disease-specific health status after 4 weeks of treatment with low-dose, sustained-release oral morphine, and breathlessness improved in people with more severe breathlessness (modified MRC breathlessness scale score of 3 or 4).¹¹

The safety of morphine for persistent breathlessness is unclear. Population-based studies of people with COPD have reported increased hospitalizations and mortality with use of morphine,¹² although these effects were not documented in people with severe, oxygen-dependent COPD.¹³ Daily use of low-dose opioids (maximal dose of 20 mg/d of morphine) has not been associated with serious adverse events (hospitalization or death) in RCTs.^9-11,14

The Breathlessness, Exertion, And Morphine Sulfate (BEAMS) trial was designed to evaluate the effect of oral extended-release morphine (8 mg/d or 16 mg/d) vs placebo on the intensity of worst breathlessness after 1 week of treatment compared with baseline. The BEAMS trial also used blinded uptitration of morphine to evaluate change in daily step count (measured using an actigraphy device) from baseline to the end of week 3.

This trial was a multicenter, placebo-controlled, phase 3, double-blind, parallel-group, dose-increment RCT of regular, low-dose, oral extended-release morphine for people with COPD and chronic breathlessness. The trial protocol was approved by the Hunter New England human research ethics committee. Each participating center obtained approval from the research governance office to recruit people to this RCT. Enrolled people provided written informed consent. The trial was conducted and monitored in accordance with Good Clinical Practice.¹⁵ The trial protocol appears in Supplement 1 and has been published¹⁶ and the statistical analysis plan appears in Supplement 2. This RCT is in compliance with the Consolidated Standards of Reporting Trials statement.¹⁷ All adverse event reports and the conduct of the trial were reviewed by an independent data and safety monitoring committee.

Individuals were eligible for inclusion if they met the following criteria: (1) were aged 18 years or older; (2) had a physician diagnosis of COPD⁷ with postbronchodilator spirometry revealing a forced expiratory volume in the first second of expiration/forced vital capacity of less than 0.7; (3) had severe chronic breathlessness¹ defined as a modified MRC breathlessness scale score of 3 or 4 (corresponding to patient statements of “I stop for breath after walking about 100 yards or after a few minutes on the level,” “I am too breathless to leave the house,” or “I am breathless when dressing”)⁵ despite optimal treatment for underlying causes as confirmed by a respiratory physician; (4) had a score of 3 or greater for the intensity of worst breathlessness on a numerical rating scale (score range, 0-10) within the 24 hours before study enrollment; (5) had the ability to complete the assessments as determined by the study investigator; and (6) had been receiving stable COPD treatment during the previous week (as-needed inhaler medications were permitted; Figure 1).

Individuals were excluded if they (1) were currently taking opioids at any dose for breathlessness or had taken opioids during the previous week at an oral morphine equivalent dose of 8 mg/d or greater; (2) had previously experienced adverse reactions to morphine; or (3) had central hypoventilation syndrome, were pregnant, had liver or kidney failure, or had a gastrointestinal obstruction. A complete list of the inclusion and exclusion criteria appears in the trial protocol (Supplement 1).

The trial was coordinated by the Australian national Palliative Care Clinical Studies Collaborative and conducted at 20 oncology, palliative, or respiratory centers in Australia.¹⁶ People were also recruited through the Lung Foundation Australia and the Primary Health Networks. All people were referred by their treating clinician to the principal investigator at each participating center.¹⁶

Participants were randomized 1:1:1 to 8 mg/d or 16 mg/d of oral extended-release morphine or placebo during week 1. All randomizations were performed at baseline using computer-generated random sample tables and permuted block sizes of 6. Each person was provided with a wrist actigraphy unit (Fitbit) and charger at the baseline visit. People were instructed to wear the actigraphy unit continuously. At the start of weeks 2 and 3, people were randomized 1:1 to 8 mg/d of extended-release morphine, which was added to the prior week’s morphine dose, or placebo. Participants could then enter an optional blinded extension treatment period of 6 months (eFigure 1 in Supplement 3). The baseline step count was obtained using the wrist actigraphy unit prior to study enrollment (day −1).

The trial medications, morphine sulfate pentahydrate (Kapanol) and placebo, were provided by Mayne Pharma. They were indistinguishable in appearance and were provided by the local pharmacy in a blister pack each week. Participants were instructed to take 2 capsules of trial medication orally each morning.

Participants were also given 2 blinded tablets of laxative (docusate with sennosides for the morphine group) or placebo tablets (for those not receiving morphine) each morning. The blister packs were collected at the end of each week. Open-label laxative tablets were also available to all participants (≤2 tablets twice daily as needed).

Participants were instructed to record their level of breathlessness in a diary each evening using a validated numerical rating scale,¹⁸ which ranged from a score of 0 (none) to 10 (being the worst or most intense). All participants received a battery-operated, handheld fan and an information sheet with standard breathlessness self-management strategies (ie, how to use the handheld fan, breathing control techniques, positions to reduce breathlessness, and suggestions to keep active). The participants were instructed to continue taking all other medications and therapies in accordance with their clinicians’ recommendations throughout the study period.

The research staff, treating clinicians, and participants were blinded to the treatment allocation. Unblinding could occur only after collection of the last data point for the final participant or in an emergency situation after consultation with the principal investigator.

The primary outcome was change in the intensity of worst breathlessness on a numerical rating scale (score range, 0 [none] to 10 [being the worst or most intense]) using the mean score at baseline (from days −3 to −1) to the mean score after week 1 of treatment (from days 5 to 7) in the 8 mg/d and 16 mg/d of extended-release morphine groups vs the placebo group.

The secondary outcomes included (1) change in mean daily step count measured using an actigraphy device (Fitbit) at baseline (day −1) to the mean step count at week 3 (from days 19 to 21), (2) mean change in the intensity of breathlessness and overall breathlessness distress level during the previous 24 hours, (3) Global Impression of Change in health status score,¹⁹ (4) Hospital Anxiety and Depression Scale score,²⁰ (5) Australia-modified Karnofsky Performance Scale score,²¹ (6) health-related quality of life measured using scores from the COPD Assessment Test, (7) the Clinical Respiratory Questionnaire Dyspnea and Mastery scores, (8) overall well-being measured using a 0- to 100-mm visual analog scale for the 5-level version of the 5-dimension EuroQol,²² (9) symptom scores using the revised Edmonton Symptom Assessment System,²³ (10) blinded participant treatment preference, (11) end-tidal partial pressure of carbon dioxide level, (12) respiratory rate, and (13) oxygen saturation as measured by pulse oximetry.

Treatment-emergent adverse events were assessed during a mid-week telephone call and at the review at the end of each week¹⁶ using version 4 of the National Cancer Institute Common Terminology Criteria for Adverse Events.²⁴ A treatment-emergent adverse event was any event that occurred or worsened after baseline. Treatment-emergent adverse events with a grade of 3 or greater using the National Cancer Institute Common Terminology Criteria for Adverse Events, or those otherwise reported as serious adverse events as part of the trial’s pharmacovigilance reporting, were categorized as severe treatment-emergent adverse events. Outcomes were evaluated after 1 week for the primary outcome and after 3 weeks for the secondary outcomes.

We estimated that a sample of 135 people would provide 80% power to detect a minimal clinically important difference of 1.1 point for the primary outcome using a numerical rating scale with a score range from 0 (none) to 10 (being the worst or most intense)²⁵ and with an SD range of 2.0 to 2.5 each day during week 1. Sample size calculations were based on a recent RCT.¹⁰ The type I error rate was prespecified as 2.5% (2-sided α level of .025) for the primary analysis and comprised 2 comparisons: 8 mg of morphine vs placebo and 16 mg of morphine vs placebo. Using a mixed-linear regression model and variance-covariance matrix based on the mean responses from the recent RCT,¹⁰ 45 participants per group were required to remain in the trial after week 1.

The categorical variables are expressed as number (percentage) and the continuous variables are expressed as median (IQR). The analyses were conducted according to randomization group. For both the efficacy analyses and the safety analyses, the analysis set was all randomized participants who received at least 1 dose of study medication.

The treatment effects for the intensity of worst breathlessness and the step count outcomes were analyzed using random-effects mixed-linear regression models, with the change from baseline as the dependent variable; treatment group, day, and treatment group × day as interaction factors; and adjustment for the baseline values of age, sex, modified MRC breathlessness scale score, Australia-modified Karnofsky Performance Scale score, end-tidal partial pressure of carbon dioxide, oxygen saturation as measured by pulse oximetry, and Charlson Comorbidity Index. An unstructured variance-covariance structure was used for the primary outcome of change in the intensity of worst breathlessness.

Compound symmetry and Toeplitz variance-covariance structures were used for the secondary outcome of step count. For the other secondary outcomes, the treatment effects were analyzed using analysis of covariance with change from baseline to end of the week as the dependent variable, treatment group as a factor, and adjustment for the baseline values of age, sex, modified MRC breathlessness scale score, Australia-modified Karnofsky Performance Scale score, end-tidal partial pressure of carbon dioxide, oxygen saturation as measured by pulse oximetry, and Charlson Comorbidity Index. The primary analysis was conducted without and with imputation for missing outcome data using multiple imputations with 100 samples redrawn. The secondary analyses were conducted using observed data without imputation. After week 1, treatment with 8 mg/d and 16 mg/d of extended-release morphine was compared with placebo for the primary analysis of change in the intensity of worst breathlessness, and after week 3, each allocated morphine treatment group (8 mg/d, 16 mg/d, 24 mg/d, and 32 mg/d) was compared with placebo.

All between-group analyses were reported with 95% CIs. Statistical significance at the end of week 1 was defined as a 2-sided P value of <.025 for the primary outcome analysis of change in the intensity of worst breathlessness after treatment with morphine (8 mg/d or 16 mg/d) vs placebo. The secondary outcome of step count was assessed at the end of week 3 and used a 2-sided significance level of .0125 after treatment with morphine (8 mg/d, 16 mg/d, 24 mg/d, or 32 mg/d) vs placebo. A hierarchical testing procedure was used whereby if the primary outcome was not significant, the secondary outcome of step count would be tested but only reported as exploratory or hypothesis-generating. All other statistical tests were 2-sided at the 5% significance level. Statistical analyses were conducted using SAS version 9.4 (SAS Institute Inc).

People were enrolled between September 1, 2016, and November 20, 2019, and followed up through December 26, 2019, at 20 centers. Of the 160 people who were randomized, 55 were randomized to 8 mg/d of morphine, 53 to 16 mg/d of morphine, and 52 to placebo (Figure 1). During week 1, there were 2 people in 16 mg/d of morphine group and 2 people in the placebo group who did not take any study drug and these people were excluded from the primary analyses. Of the 156 people included in the primary analyses (median age, 72 years [IQR, 67-78 years]; 48% were women), 121 (78%) had a score of 3 on the modified MRC breathlessness scale and 35 (22%) had a score of 4.

The baseline characteristics were similar among the treatment groups (Table 1). At baseline, the mean step counts were similar among the groups: 2526 (SD, 2139) for the 8 mg/d of morphine group, 2214 (SD, 1801) for the 16 mg/d of morphine group, and 2242 (SD, 1708) for the placebo group (Table 2). A total of 21 people completed 3 weeks of treatment in the 8 mg/d of morphine group, 27 people in the 16 mg/d of morphine group, 22 people in the 24 mg/d of morphine group, 8 people in the 32 mg/d of morphine group, and 7 people in the placebo group (Figure 1).

After 1 week of treatment, the primary outcome of change in the intensity of worst breathlessness was not significantly different for the 8 mg/d of morphine group vs placebo (mean difference, −0.3 [95% CI, −0.9 to 0.4]) or for 16 mg/d of morphine group vs placebo (mean difference, −0.3 [95% CI, −1.0 to 0.4]) (Table 2 and Figure 2). There was no significant treatment effect when accounting for missing data using multiple imputation for the 8 mg/d of morphine group vs the placebo group (mean difference, −0.20 [95% CI, −0.86 to 0.46]) or for the 16 mg/d of morphine group vs the placebo group (mean difference, −0.30 [95% CI, −0.98 to 0.38]).

Overall in the primary outcome analysis including 156 people, data were missing for 10 (6%). Of the missing data for the 10 people, 1 was in the 8 mg/d of morphine group, 6 were in the 16 mg/d of morphine group, and 3 were in the placebo group. The primary outcome of change in the intensity of worst breathlessness was not significantly different among people with a modified MRC breathlessness scale score of 3 or 4 (eFigure 2 in Supplement 3) or when analyzed per protocol (mean difference, −0.20 [95% CI, −0.86 to 0.46] for the 8 mg/d of morphine group vs the placebo group; mean difference, −0.30 [95% CI, −0.98 to 0.38] for the 16 mg/d of morphine group vs the placebo group).

After 3 weeks of treatment, there were no significant differences for the secondary outcome of change in mean daily step count for the 8 mg/d of morphine group vs the placebo group (mean difference, −1453 [95% CI, −3310 to 405]), for the 16 mg/d of morphine group vs the placebo group (mean difference, −1312 [95% CI, −3220 to 596]), for the 24 mg/d of morphine group vs the placebo group (mean difference, −692 [95% CI, −2553 to 1170]), or for the 32 mg/d of morphine group vs the placebo group (mean difference, −1924 [95% CI, −47 699 to 921]) (Figure 2). None of the secondary outcomes were significantly different between the treatment groups after week 1 (Table 2 and eTables 2-3 in Supplement 3), including for change in mean daily step count (eFigure 3 in Supplement 3) and change in the mean intensity of breathlessness and overall breathlessness distress level (eFigure 4 in Supplement 3).

Based on capsule count at the end of the study periods, adherence to the allocated treatment was very high and similar among the groups. Mean adherence was 97% to 100% during week 1 (eTable 4 in Supplement 3), 95% to 100% during week 2 (eTable 5 in Supplement 3), and 93% to 100% during week 3 (eTable 6 in Supplement 3).

Treatment-emergent adverse events during week 1 occurred in 35 of 55 people (64%) in the 8 mg/d of morphine group, 40 of 51 people (78%) in the 16 mg/d of morphine group, and 24 of 50 people (48%) in the placebo group (eTables 7-8 in Supplement 3). The treatment-emergent adverse events were common morphine-related adverse events and included constipation, fatigue, dizziness, nausea, and vomiting. However, these treatment-emergent adverse events did not appear to be associated with study drug adherence (eTables 4-6 in Supplement 3). Treatment-emergent adverse events during weeks 2 and 3 appear in eTables 9-10 in Supplement 3.

During week 1, the number of people who discontinued treatment due to adverse events was 2 in the 8 mg/d of morphine group and 5 in the 16 mg/d of morphine group, and 0 in the placebo group (Figure 1). During week 2, study discontinuation due to adverse events occurred in 1 person in the 8 mg/d of morphine group, in 3 people in the 16 mg/d of morphine group, in 2 people in the 24 mg/d of morphine group, and in 1 person in the placebo group (Figure 1). During week 3, study discontinuation due to adverse events occurred in 1 person in the 8 mg/d of morphine group, in 1 person in 16 mg/d of morphine group, in 4 people in the 24 mg/d of morphine group, in 0 people in the 32 mg/d of morphine group, and in 0 people in the placebo group.

Throughout the trial, serious treatment-emergent adverse events (which included hospitalizations and deaths) occurred in 46 of 139 people (33%; 85 episodes) who received any dose of morphine compared with 2 of 17 people (12%; 2 episodes) who received only placebo (Table 3). Serious treatment-emergent adverse events included increased breathlessness, infections, morphine-related adverse events, and 2 events that met the criteria²⁴ for respiratory failure (both in the morphine group) of the National Cancer Institute Common Terminology Criteria for Adverse Events.

In this RCT of people with COPD and severe chronic breathlessness, daily treatment with extended-release morphine did not significantly improve the primary outcome of change in the intensity of worst breathlessness after 1 week of treatment. There was also no significant improvement in the secondary outcome of mean daily step count using an actigraphy device or in any of the other secondary outcomes, including symptoms, function, health-related quality of life, and physiological parameters. The lack of efficacy was consistent across severities of breathlessness (modified MRC breathlessness scale score of 3 or 4).

This RCT tested the efficacy and adverse event profile of several different uptitrated doses of extended-release morphine across clinically relevant doses (8 mg/d, 16 mg/d, 24 mg/d, and 32 mg/d) over 3 weeks in people with COPD and severe chronic breathlessness. To our knowledge, this study is the first to measure physical function using an actigraphy device concurrently with blinded uptitration of morphine in an adequately powered trial. These doses of extended-release morphine were not associated with a decreased total step count compared with placebo. The unchanged total step count does not support the hypothesis that the lack of efficacy on breathlessness was due to people increasing their physical activity (thereby masking a true symptom improvement).^10,26

The study findings may not be applicable to people with very advanced COPD and breathlessness who are in palliative care or near the end of life, at which time treatment with opioids may be useful to provide relief of severe dyspnea.^4,27 Further research is needed to determine if specific groups of people with COPD are more likely to experience a reduction in breathlessness with morphine, if some may benefit from higher doses of morphine, and to clarify the role of short-acting opioids for severe episodes of breathlessness.

Strengths of this study include its multicenter, double-blind, parallel-group randomized design and blinded uptitration during each stage of the trial vs placebo. Blinded administration of laxative (or placebo) improved blinding; participants were not allowed to take as-needed morphine during the trial, and clinically relevant efficacy and safety outcomes were assessed using validated instruments.²⁸

This study has several limitations. First, the number of participants in each group decreased over the 3-week period, and only 42% completed treatment at week 3. However, the trial was adequately powered for the primary outcome and recruited the target number of people.

Second, breathlessness was measured in daily life and not during standardized exercise testing. Third, extended-release morphine can be administered every 12 hours to 24 hours so it is possible that the 24-hour dosing interval used in this study may have provided a suboptimal dose.

Among people with COPD and severe chronic breathlessness, daily low-dose, extended-release morphine did not significantly reduce the intensity of worst breathlessness after 1 week of treatment. These findings do not support the use of these doses of extended-release morphine to relieve breathlessness.

Corresponding Author: Magnus Ekström, MD, PhD, Department of Medicine, Blekinge Hospital, Bättringsvägen 1, SE-37185, Kalskrona, Sweden (pmekstrom@gmail.com).

Correction: This article was corrected February 28, 2023, to fix the misspelling of a last name in the nonauthor collaborator supplement.

Accepted for Publication: October 12, 2022.

Author Contributions: Dr Currow had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Ekström, Ferreira, Johnson, Eckert, Fazekas, Clark, Agar, Currow.

Acquisition, analysis, or interpretation of data: Ekström, Ferreira, Chang, Louw, Johnson, Eckert, Fazekas, Agar, Currow.

Drafting of the manuscript: Ekström, Agar, Currow.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: Louw, Currow.

Obtained funding: Johnson, Currow.

Administrative, technical, or material support: Ekström, Ferreira, Eckert, Fazekas, Agar, Currow.

Supervision: Fazekas.

Conflict of Interest Disclosures: Dr Ekström reported receiving an unrestricted grant from the Swedish Research Council. Ms Louw reported being employed by McCloud Consulting Group and receiving consulting fees for her work on this study and receiving personal fees from the Palliative Care Clinical Studies Collaborative. Dr Johnson reported receiving consulting fees from Mayne Pharma for serving as a clinical advisor. Dr Eckert reported receiving a Collaborative Research Centre Consortium grant among his institution, the Australian government, and Oventus Medical; receiving grants from the National Health and Medical Research Council (NHMRC) of Australia, Takeda, Bayer, Invicta Medical, and Apnimed; being supported by a leadership fellowship sponsored by the NHMRC of Australia; serving as a consultant to Bayer, Apnimed, and Takeda; receiving personal fees from Eisai; and serving on advisory boards for Eisai, Invicta Medical, and Apnimed. Dr Agar reported receiving payment from AstraZeneca to facilitate a workshop on the multidisciplinary care of older people with lung cancer. Dr Currow reported receiving an unrestricted grant from Mundipharma; being an unpaid member of an advisory board for Helsinn Pharmaceuticals; and being a consultant for and receiving intellectual property payments from Mayne Pharma. No other disclosures were reported.

Funding/Support: This study was funded by grant APP1065571 from the National Health and Medical Research Council of Australia and sponsored by Flinders University, Adelaide, Australia.

Role of the Funder/Sponsor: The National Health and Medical Research Council of Australia and Flinders University had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Group Information: The members of the Australian National Palliative Care Clinical Studies Collaborative appear in Supplement 4.

Data Sharing Statement: See Supplement 5.

Additional Contributions: We thank all participants who made this research possible at a very difficult time of their life, and to the families and friends who supported them. Thanks also go to the staff at each site for their enduring efforts in working with participants to complete the study.