Article Text
PDF
Abstract
Objective The aim of this study was to evaluate the cost-effectiveness of a home-based palliative care model relative to usual care in expediting discharge or enabling patients to remain at home.
Design Economic evaluation of a pilot randomised controlled trial with 28 days follow-up.
Methods Mean costs and effectiveness were calculated for the Palliative Care Extended Packages at Home (PEACH) and usual care arms including: days at home; place of death; PEACH intervention costs; specialist palliative care service use; acute hospital and palliative care unit inpatient stays; and outpatient visits.
Results PEACH mean intervention costs per patient ($3489) were largely offset by lower mean inpatient care costs ($2450) and in this arm, participants were at home for one additional day on average. Consequently, PEACH is cost-effective relative to usual care when the threshold value for one extra day at home exceeds $1068, or $2547 if only within-study days of hospital admission are costed. All estimates are high uncertainty.
Conclusions The results of this small pilot study point to the potential of PEACH as a cost-effective end-of-life care model relative to usual care. Findings support the feasibility of conducting a definitive, fully powered study with longer follow-up and comprehensive economic evaluation.
- Cancer
- Home care
- Service evaluation
- Supportive care
- Terminal care
Statistics from Altmetric.com
Introduction
Provided symptoms can be adequately controlled, up to 90% of people at the end of life prefer to be cared for and, often, to die at home.1 However, inability to rapidly mobilise nursing and social services can cause delays in receiving home-based care.2 Cost-effective models of care that overcome these barriers are required to support patient preferences for end-of-life care.
Economic evaluations of end-of-life care models informing healthcare resource allocation decisions are often incomplete, being largely limited to costing analyses.3 This paper presents a prospective cost-effective analysis of the Palliative Care Extended Packages at Home (PEACH) pilot randomised controlled trial (RCT) which assessed impacts on palliated patients’ time at home over 28-days follow-up. The aim was to evaluate the incremental resource use, cost and patient-related consequences of PEACH compared with usual care.
Methods
Context
Australian community-based end-of-life care is publicly funded.4 This study was undertaken in southwest Sydney, New South Wales where community care is provided by the specialist palliative care team, general practitioners and community nurses. Ethical approval was granted by the research ethics committee.
Trial design
The PEACH trial was a pilot, Phase II, parallel arm RCT involving 32 consented participants with predominantly advanced cancer (ACTRN12610001032044) (http://www.anzctr.org.au/trial_view.aspx?ID=335019). Patients were eligible if they had complex or unstable symptom management and high care needs (see online supplementary table S1). Study aims were to assess whether PEACH increases time at home and the feasibility, value and design of a larger definitive RCT. Participants were randomised to receive PEACH or usual care in a 3 : 1 ratio, increasing experience with PEACH and aiding recruitment and ethics approval. PEACH is an individualised care package determined by local protocols for community and inpatients. Services are rapidly mobilised, essential equipment is secured, allied health is coordinated and higher intensity nursing is provided (up to 24 h/day for up to 5 days) compared with usual care. Usual care encompassed conventional discharge planning with existing community services including specialist palliative care, access to an after-hours number, and equipment from loan pools.
Data
Patient-level data were collected prospectively, including: days at home; place of death; PEACH intervention costs (staff administration, travel and direct patient contact time, overheads and consumables); specialist palliative care service use; acute hospital and palliative care unit inpatient lengths of stay and outpatient visits. Resource use was costed according to the Australian Manual of Resource Items and their Associated Costs (Available at http://www.pbs.gov.au/html/industry/static/useful_resources/manualCosts) in 2010 Australian dollars. Inpatient stays were costed using case-mix weights for Australian Refined Diagnosis Related Groups (http://www.health.gov.au/internet/main/publishing.nsf/Content/5F8B6BE822DC75B3CA25748300164037/$File/NHCDC%20-%20Round%2011%20-%20Cost%20Weights.pdf) inpatient stays as recommended by the Australian Medical Services Advisory Committee guidelines (http://www.msac.gov.au/internet/msac/publishing.nsf/Content/D81BE529B98B3DB6CA2575AD0082FD1B/$File/Economics%20subcommittee%20guidelines%20June%202009.pdf). Specialist palliative care services and PEACH costs were estimated using hourly rates of local salaries (plus 30% on-costs), agency staff costs and equipment hire. PEACH administrative costs were included. Outpatient visits were costed using the National Hospital Cost Data Collection.
Analysis
Within-trial analysis, undertaken from a healthcare provider perspective, provided estimates of incremental costs, effects and cost-effectiveness. Incremental net monetary benefit (INMB) and cost-effectiveness acceptability curves (CEACs) were estimated at potential threshold values for one extra day at home (the primary outcome). INMB represents the monetary value of additional effects of care minus the additional costs of care.5 CEACs present the probability of positive INMB over plausible threshold values for the primary outcome.6 The threshold value reflects the decision-maker maximum acceptable monetary value for one unit gain of effect.5 The perspective of the analysis and 28 days follow-up period were determined by the available research resources. Based on previous evidence,3 community cost estimates were not expected to differ substantially between arms.
Bootstrapping on participants’ costs and effects across 10 000 replicates allowed robust assessment of uncertainty for costs, effects and cost-effectiveness.7 Ninety-five per cent CIs were calculated for the INMB and CEACs using these replicates.6 ,8 Sensitivity analyses considered: (1) only including an average daily cost for inpatient days within the study period instead of complete inpatient episodes and; (2) excluding programme costs given some overlap between PEACH and RCT administration costs. Costs and benefits were not discounted because the trial lasted less than a year. Statistical analyses were conducted using Microsoft Office Excel 2003.
Results
Complete data were available for 31/32 participants, with inpatient costs missing for one PEACH participant only. Baseline characteristics are summarised in table 1. The study population had a slightly younger profile than the national average palliative care population.9 Overall, 68% participants died during follow-up.
Baseline participant characteristics
Cost and effectiveness estimates
PEACH participants had 13.1 mean days at home, one more than usual care (see online supplementary table S2). The mean length of the PEACH intervention was 6.7 days (SD 7.2; median 5.0 days, range 0–23 days). PEACH costs varied widely with one high cost outlier attributable to a high quantity of additional nursing hours. Of participants who died, 56.3% in the PEACH arm died at home versus 80.0% in the usual care arm. The mean direct cost associated with PEACH was $3489 per participant and was largely offset by $2450 inpatient cost savings (see online supplementary table S2).
Incremental net monetary benefit
An INMB curve (figure 1) shows that mean INMB over 28 days for PEACH versus usual care becomes positive when the threshold value for one extra day at home exceeds $1068. At this point PEACH is the preferred service model because expected benefits exceed expected costs, although 95% CIs are wide.
Expected incremental net benefit curve for Palliative Care Extended Packages at Home versus usual care over 28 days with 95% CIs.
The INMB estimates are sensitive to the direction of treatment effect and PEACH programme costs (see online supplementary table S2). Costing only within-study days of hospital admission instead of complete inpatient episodes increases the threshold value above which the mean INMB becomes positive to $2547. Increasing the trial follow-up beyond 28 days in future studies would reduce this uncertainty, allowing for long-term treatment effects of PEACH which are likely to extend until death. Removal of the high cost outlier from the analysis reduced the threshold value above which the mean INMB becomes positive to $846. The CEAC (see online supplementary figure S1) indicates the probability of PEACH being cost-effective versus usual care for potential threshold values for one extra day at home
Discussion
The higher nursing costs of PEACH compared with usual care were largely offset by lower inpatient costs resulting in a net incremental cost of about $1000 per patient. At 28 days PEACH participants had, on average, one extra day at home although a greater proportion died as an inpatient. In practice, incremental costs of PEACH compared with usual care may be lower due to: overlapping PEACH and RCT administration costs; economies of scale; learning by doing effects; and treatment effects beyond 28 days.
This is the first prospective economic evaluation of a model of care supporting Australian patients with advanced cancer who want home-based rather than inpatient end-of-life care. Findings are consistent with other RCT evidence that suggests home-based palliative care can reduce hospitalisations and associated costs.3 ,10 An adequately powered trial with longer follow-up is required to definitively assess the cost-effectiveness of the home-based model of palliative care. Additionally, further research is needed to determine whether a threshold value for one extra day at home is meaningful to decision makers and to guide the choice of threshold value.
Limitations
A higher proportion of usual care (3/8, 37.5%) relative to PEACH participants (4/23, 17.4%) were recruited as inpatients, which may restrict days at home. This is not unexpected given the small randomised sample. There may have been important differences in baseline characteristics strongly associated with a home death between the intervention arms, for example, low functional status.11
Cost estimates did not include claims data for any additional costs of community care so the true cost of the models of care in each arm may be underestimated. However, these costs are not expected to differ by arm and are unlikely to influence total incremental costs.
Informal care costs were not included because a healthcare provider perspective was taken. Costs may be shifted from service providers to the family.12 Ideally, future evaluations would also capture such cost transfers so that a full societal perspective could be taken.
Generalisability of cost-effectiveness results will be limited to care provided by similar costing and funding models.13
There was a substantial proportion of missing data for other secondary outcomes hindering interpretation of these data. Robust analysis using imputation methods for missing data as recommended in the most recent CONSORT statement was unfeasible due to very small numbers.14 However, such analysis may be possible with data from a larger trial.
Despite these limitations, the PEACH pilot data have clarified the need for an adequately powered, Phase III multisite RCT stratified by recruitment site using 1 : 1 randomisation. The pilot suggests data should be collected beyond 28 days as only 68% of participants died during study follow-up and the impacts of PEACH are expected to extend until death. Secondary outcomes including symptom control, functional status, preference for place of care, patient and carer quality of life, and the ability of patients to finalise their affairs15 should also be covered in a definitive trial. Methods such as multiple outcome comparison on the cost-disutility plane could be used to more accurately estimate the true net benefit of PEACH through joint consideration of multiple outcomes under uncertainty (days at home; proportion of home deaths).16
Conclusion
Based on the results of this small pilot study most of the costs of intensive, home-based care packages are suggested to be offset by reduced inpatient stays, while days at home are increased. There is potential for such packages to be cost-effective but there was high uncertainty within the study. Findings suggest the need for a definitive trial with longer follow-up and support the feasibility of conducting an adequately powered study with a comprehensive economic evaluation.
Acknowledgments
The authors thank the patients and carers involved in PEACH, the staff from South West Sydney Local Health, previous reviewers for their insightful comments and Ms Debbie Marriott for her assistance with the manuscript.
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Files in this Data Supplement:
- Data supplement 1 - Online figure
- Data supplement 2 - Online table 1
- Data supplement 3 - Online table 2
Footnotes
-
Contributors NM: conception and design; analysis and interpretation of data; drafting of manuscript; critical revision of manuscript; statistical analysis; and administrative, technical and material support. MA: conception and design; obtaining funding; acquisition of data; analysis and interpretation of data; critical revision of manuscript; and technical support. JH: conception and design; acquisition of data; analysis and interpretation of data; critical revision of manuscript; JK: analysis and interpretation of data; critical revision of manuscript; and technical support. DC: conception and design; analysis and interpretation of data; critical revision of manuscript; administrative, technical and material support; SE: conception and design; obtaining funding; analysis and interpretation of data; critical revision of manuscript; technical support; and supervision.
-
Funding This study was funded by the Australian Government Department of Health and Ageing under the National Palliative Care Program, Palliative Care for People at Home. NM was also funded through the National Palliative Care Program and Flinders University.
-
Competing interests None.
-
Ethics approval Sydney South West Area Health Service Human Research Ethics Committees.
-
Provenance and peer review Not commissioned; externally peer reviewed.
-
Data sharing statement Quantities of resource use are available by request to the corresponding author. No other additional data are available from this study.