https://orcid.org/0000-0002-1421-2966
Figures
Abstract
Background
Over 65s are frequent attenders to the Emergency Department (ED) and more than half are admitted for overnight stays. Early assessment and intervention by a dedicated ED-based Health and Social Care Professionals (HSCP) team reduces ED length of stay and the risk of hospital admissions among older adults while improving patient health-related quality-of-life and satisfaction with care. This study aims to evaluate whether augmenting the treatment as usual for older adults admitted to ED is cost-effective.
Methods and findings
Cost-effectiveness analysis (CEA), conducted alongside the OPTI-MEND randomised controlled trial of 353 patients aged ≥65 with lower urgency complaints compared the effectiveness of early assessment and intervention by a dedicated HSCP team in the ED to treatment as usual (TAU). An economic analysis estimated the average cost per older adults randomised to the HSCP team, and compared to TAU, how contact with HSCP team changed health care use, and associated total costs, and estimated the effect of HSCP on Quality-Adjusted Life Years (QALYs). Within the OPTI-MEND trial, the average cost of a contact with the HSCP team during ED attendance is estimated to be €801 per patient. Compared to TAU, the incremental QALY of intervention is 0.053 (95% CI: 0.023 to 0.0826, p<0.0001). Accounting for cost savings because of contact with HSCP team, the average incremental saving in the total cost, compared to TAU, is -€6,128 (95% CI: -€9,217 to -€3,038, p<0.0001). Given the incremental health gains and significant cost savings, bootstrapped cost CEA suggests that dedicated HSCP care dominates over TAU for low urgency older adults attending the ED.
Conclusions
A dedicated HSCP team in the ED significantly improves overall health for lower acuity older adults and, by reducing inpatient length of stay, results in staggering cost savings. This economic evaluation conducted on the OPTI-MEND trial provides convincing evidence that HSCP should be adopted as part of treatment as usual in Irish EDs.
Trial registration
ClinicalTrials.gov, NCT03739515; registered on 12th November 2018. https://classic.clinicaltrials.gov/ct2/show/NCT03739515.
Citation: Trépel D, Ruiz-Adame M, Cassarino M, Ahern E, Devlin C, Robinson K, et al. (2024) The cost effectiveness of early assessment and intervention by a dedicated health and social care professional team for older adults in the emergency department compared to treatment-as-usual: Economic evaluation of the OPTI-MEND trial. PLoS ONE 19(6): e0298162. https://doi.org/10.1371/journal.pone.0298162
Editor: Arkers Kwan Ching Wong, The Hong Kong Polytechnic University, HONG KONG
Received: May 13, 2023; Accepted: January 10, 2024; Published: June 25, 2024
Copyright: © 2024 Trépel et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: Yes - all data are fully available without restriction. All data files are available within the paper or https://osf.io/5wpjm/.
Funding: Funded: Authors' Initials: KR, GM, RG. Grant Number: RCQPS-2017-2 Full Name: Health Research Board of Ireland URL: https://www.hrb.ie The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Abbreviations: ADL, Activities of daily living; CEA, Cost-effectiveness analysis; CEAC, Cost-effectiveness acceptability curve; CEP, Cost-effectiveness plane; CI, Confidence Interval; CONSORT, Consolidated standards of reporting trials; ED, Emergency department; GP, General practitioner; HCU, Healthcare use; HSCP, Health and Social Care Professional; HSE, Health Service Executive; ICER, Incremental cost-effectiveness ratio; LoS, Length of Stay; MTS, Manchester triage system; QALY, Quality-adjusted life year; TAU, Treatment as usual
Introduction
Older adults are frequent users of emergency departments (EDs) [1,2] and, under treatment as usual (TAU), more than half of ED attendances result in inpatient admissions with a median length of stay of nine nights (interquartile range 5 to 24 nights) [3]. Hospital admissions are associated with increased ED wait times [4], shortages in hospital beds and complex admission pathways result in delayed length of stay (LoS) in the ED. The OPTI-MEND trial tested the effect of adding a dedicated team of Health and Social Care Professionals (HSCP) to the ED and concluded that early assessment and intervention for low urgency older people can facilitate shorter stays in the ED, reduced risk of hospital admissions and improve satisfaction with overall care. This health economic study extends on the clinical trial and, by using OPTI-MEND trial data, conducts a cost effectiveness analysis to inform policy on whether HSCPs may represent value for money.
Appraisal of the best available evidence suggests that interventions centred around care coordination in the ED may increase clinical effectiveness for older adults [5]. Care Coordination Teams in the ED reduce rates of hospital admission [6], which motivated the hypothesis that, allocating a dedicated HSCP team who conducts early assessment and intervention, would result in better clinical and economic outcomes for lower acuity older adults. Specifically, HSCP teams are demonstrated to result in favourable discharge outcomes to home and enhanced continuity of care in the community [7,8] which, on aggregate, should improve health-related quality-of-life and reduce demand for limited healthcare resources.
The OPTI-MEND trial was conducted to determine whether ED-based interdisciplinary HSCP teams are effective to reduce LoS in the ED and incidence of hospital admission among older adults [3]. The HSCP team allocated into the ED included one additional senior physiotherapist, one senior occupational therapist, and one senior medical social worker to provide early assessment and intervention to lower acuity older adults. The team conducted interdisciplinary assessments of functional and mobility status, cognition, and psychosocial needs and subsequent interventions were tailored to individual older adults’ needs (including, but not limited to, patient and family education on the outcome of the HSCP assessment and ED discharge plan, prescription of mobility aids and enabling Activities of Daily Living (ADL) equipment, provision of home exercise programmes, education of self-management strategies and onward referral to alternative care pathways). The primary clinical study found that, compared to usual ED care, HSCP teams were clinically effective in reducing ED length of stay (6.4 versus 12.1 median hours, p < 0.001), and incidence of hospital admissions (19.3% versus 55.9%, p < 0.001); this motivated the further hypothesis that dedicated HSCP teams for older adults should also be a cost effective service, and potentially cost saving, within the Irish health system.
To determine whether dedicated HSCP teams for older adults in the ED represent value to the Irish health system, the national guidance recommends that incremental health gains from intervention over TAU be expressed as Quality-Adjusted Life Years (QALYs), and all costs relevant to a health and social care budget should be considered [9]. To be considered cost effective, the incremental cost effectiveness ratio of HSCP plus usual ED care, compared to usual ED care alone, would need to demonstrate producing health gain for less than the cost-effectiveness threshold, which is currently set at €45,000 per QALY in Ireland.
Economic evaluations of ED models of care have been shown, through systematic searches, to be largely absent in the evidence base [10]. This paper reports cost-effectiveness analysis, conducted alongside the OPTI-MEND trial, with a view to informing Irish decision-makers on whether routinely allocating HSCP teams, that provide early assessment and intervention for low urgency older people, represents value for money, compared to TAU.
Methods
All methods regarding the conduct of this economic evaluation conducted alongside the OPTI-MEND trial were described in a Health Economic Analysis Plan [11].
Trial design
A single-centre, parallel group, randomised controlled trial was conducted in the ED of a regional university teaching hospital in the Mid-West of the Republic of Ireland. The trial was registered on ClinicalTrials.gov (NCT03739515) and a protocol detailing the clinical- and cost-effectiveness analyses were published in advance [12]. The study received ethics approval from the Health Service Executive (HSE) Mid-Western Regional Hospital Research Ethics Committee on 20th September 2018 (Ref: 103/18).
Participant inclusion followed specific eligibility criteria described in Table 1. Recruitment of participants took place between December 2018 and May 2019. After giving written consent to take part, each participant underwent a baseline assessment and were then randomly allocated to the intervention or control group. A total of 353 older people aged ≥65 years were randomised to either receive HSCP plus TAU (n = 176), or TAU (n = 177). Full details on patient inclusion criteria are described elsewhere.
The research nurse (CD) who conducted the evaluations was blind to group allocation. Only the research nurse and the HSCP team had access to information that could identify the participants during data collection. Once all data were collected, data were anonymised and the final dataset that was used for analysis contained no identifiable information.
Intervention and control groups
To create the HSCP team, three full-time senior healthcare professionals were employed and allocated to work in the ED: one physiotherapist, one occupational therapist and one medical social worker, all at senior level. Participants who were eligible to be seen by the HSCP team were identified either through the ED triage system, or via consultation with the Emergency Medicine staff. The control group received treatment as usual (TAU) chosen because it represented routine ED care that patients would ordinarily receive on attendance at the ED and allowed incremental cost effectiveness analysis.
Cost required to provide HSCP team
To inform the cost of the intervention within the base-case cost-effectiveness analysis, the OPTI-MEND trial budget for allocating the HSCPs, over TAU, were utilised. The average cost per participant was calculated as the total budget divided by the number of trial recipients.
Resource use and associated costs
In addition to considering the cost of providing a HSCP team in the ED, and determining how this compares to TAU, resource use data were collected from all participants by a trained research nurse blind to group allocation. Resource use data was gathered from the hospital database of service use following discharge and included number of visits (if any) to the General Practitioner (GP), public health nurse, home help/home support, private consultations, outpatient department visits, and allied health service use.
The economic evaluation sought to examine wider resource use, and the associated cost, both immediately following the ED index visit, and in the interval between the two successive trial follow-up timepoints (30-days and 6-months). Immediately following presentation to the ED, time from ED registration to discharge was measured in hours and, where participants were admitted as inpatients, their length of inpatient stay was captured. Participants were also followed up at 30 days after their index visit to ED and the number of unscheduled ED re-visit and, where applicable, length inpatient stay, were captured. The final follow-up was conducted approximately 6 months after the index visit and individuals resource use after the date of their 30-day follow up was captured and reported unscheduled ED re-visit, inpatient length of stay, outpatient contact and community contacts (specifically with either general practitioner, nurse, physiotherapist, occupational therapist, dietician, or podiatrist).
For all items of resource use captured, a related Irish Unit Cost was identified (see Table 3) and, for each participant, the quantity of each resource item was multiplied by the related Irish Unit cost. Summing all individual costs, a total cost per participant was calculated and, by adding the incremental cost related to their intervention group (HSCP + TAU or TAU), a group average total cost, relevant from the perspective of the wider healthcare system, was calculated. Formally, the Total Cost equation is:
For each arm of the study, the average use of each resource, the associated average cost per item and the total average cost are summarised.
Health outcomes for economic evaluation
Participants’ responses to the EQ-5D-5L questionnaire were used to estimate health states utilities using the Irish value set [14]. Using an area-under-the-curve approach, the estimated health state utility at each timepoint and the specific dates of data collection, Quality-Adjusted Life Years (QALYs) were estimated across all timepoints.
Cost-effectiveness analysis
Unadjusted Incremental Cost Effectiveness Ratio (ICER) for each intervention group are calculated compared to treatment as usual (TAU) using the following formula:
where
is the average total costs and
is the average effect, expressed in terms of Quality-Adjusted Life Years.
To account for the joint distributions of cost and QALYs, the differences between groups were estimated using Zellner (1962) Seemingly Unrelated Regression Equation (SUR) [15]; SUR is selected as it is considered more efficient over unrelated ordinary least suare regression and reports correlation between costs and effects [16,17]. Non-parametric bootstrapping (10,000 replication) was conducted on random samples of the observed data and the results of the bootstrap are presented as a scatter plot on the cost effectiveness plane. Furthermore, joint distribution of costs and outcomes were illustrated using 50%, 75% and 95% confidence ellipses surrounding the ICER, indicating on the CE plane, the probability space within which we are confident the true ICER is found.
Guidelines for the Economic Evaluation of Health Technologies in Ireland require that probability analysis present “the probability of an ICER is being below €20,000 and €45,000 per QALY, respectively” [9]. Where relevant, the probability of HSCP being cost-effective at these willingness-to-pay thresholds were calculated and how the probability of HSCP being cost effective increases as willingness-to-pay increases, a cost effectiveness acceptability curve (CEAC) is generated. No discounting was applied as the study was less than 12 months in duration.
Patient and public involvement
The design and implementation of the trial was informed by extensive consultation with key ED stakeholders, including ED patients and caregivers, as well as hospital and ED medical, nursing and HSCP staff [18].
Results
Inspection of baseline characteristics indicated that trial randomisation produced well balanced groups across most characteristics and that the trial was sufficiently powered for cost effectiveness analysis.
The Consolidated standards of reporting trials (CONSORT) diagram was adapted to report key variables and explain the sample available for complete case cost-effectiveness analysis (see Fig 1). Overall, OPTI-MEND randomised 353 participants to either HSCP + TAU (n = 177) or TAU (n = 176) and all participants completed the intervention. For complete case cost effectiveness analysis, missing responses to EQ-5D-5L and/or dates for this data or missing health care use data (HCU) resulted in omission from the final analysis. Attrition at either the 30-day or 6-month follow-up was documented as either lost to follow up or participants discontinued in the study. For participants who died during the study, where the date of death was obtained, they were reported as “deaths” between the timepoint and were included in the complete case analysis (i.e., their health utility and health care use being zero on the date of death onwards).
In contrast the clinical effectiveness analysis [3], this diagram specifically explains data available for use in complete case cost effectiveness analysis.
With reference to trial budgets, the HSCP team were employed for a period of six months as part of the OPTI-MEND trial study at a cost of €118,792.89 for the duration of the trial. A budget of €7,500 was allocated to cover cost of aids and appliances for participants during the intervention and a dedicated assessment room in the ED for €14,600 for the six-month period. This total required budget was €140,892.89.
To calculate individuals cost related to their health care use, Table 2 provides the unit costs use to convert resource usage into costs.
Table 3 reports, by treatment group, the average health care use (left) and associated cost (right) for all resource use items obtained in the trial.
Table 4 reports the average and 95% Confidence Intervals for each timepoint, unadjusted utilities and costs by timepoint, and across the whole study, QALY and total costs and between group difference of QALYs and Total Cost. The difference in QALYs and Total Cost were subject to bootstrapping (10,000 replication) to provide unbiased 95% confidence intervals and finds the unadjusted between-group difference in QALYs is 0.053 (bootstrapped 95% CI: 0.019 to 0.086) and in total cost -€6,128 (bootstrapped 95% CI: -€9,180 to -€3,075).
To model the joint distribution of costs and QALYs for incremental cost effectiveness analysis of HSCP, and to control for baseline utility, seemingly unrelated regression was performed on the n = 322 complete cases available across all time points. Regression of the joint distribution find that total costs and QALYs were significantly correlated, and treatment group explained a large proportion of the variance in QALY (R2 = 0. 2442) and a smaller proportion of the variance in total cost (R2 = .0448). Correlation between Total Cost and QALYs was -0.2803 and negative correlation indicates individuals with worse outcomes have higher costs. Accounting for correlation in the joint distribution, the dedicated HSCP intervention reduced total cost to healthcare by €6,128 (95% CI: -€9,217 to -€3,038, p<0.001) and resulted in an incremental QALY of 0.0529 (95% CI: 0.0231 to 0.0826) (see Table 5).
Fig 2 illustrates results from the analysis of the uncertainty of the joint distribution of total cost and QALY. As the majority of bootstrapped replicates call in the bottom right quadrant, this indicates that HSCP has a 99.85% certainty that HSCP + TAU dominates (i.e. is more effective and saves money) over TAU alone suggesting a high probability the intervention should replace the usual arrangement of care.
Impact of patient and public involvement
Qualitative insights gathered during the consultation identify enablers and challenges associated with the introduction of the HSCP team and the trial data collection [15].
Discussion
The OPTI-MEND trial was performed on the hypothesis that, early assessment and intervention for low urgency older people can facilitate shorter stays in the ED, reduces risk of hospital admissions and improve satisfaction with overall care. Analysis of clinical effectiveness demonstrated such HSCP teams significantly reduce ED LoS, and incidence of hospital admissions [3] and, building upon these initial findings, this formal cost-effectiveness analysis now confirms the magnitude of potential cost savings the Irish health system, as well as significant improvement in health.
The economic evaluation conducted alongside the OPTI-MEND trial firstly estimates that the average cost of a contact with the HSCP team during ED admission is €801. In line with HIQA guidance from the Economic Evaluation of Health Technologies in Ireland require that probability analysis present “the probability of an ICER is being below €20,000 and €45,000 per QALY, respectively” [9]. Because of contact with HSCP team, there is an average incremental saving in the total cost, compared to TAU, of -€6,128 per patient, largely driven by averting inpatient admission and stay. As effectiveness analysis show an average benefit of 0.053 additional per QALYs and given the treatment results in overall cost saving, there is certainty from OPTI-MEND data that, HSCP teams are cost effective and may in fact ‘dominate’ usual care (i.e. that is would be efficient use of resource to replace the current arrangement of care in the subpopulation).
The OPTI-MEND trial has shown that a dedicated ED-based HSCP team, as compared to TAU, has positive clinical outcomes that allow a higher use of services for more populations (e.g. by reducing inpatient length of stay, lower rates of hospital admission) and a high reduction in cost per patient. From this analysis, we can reliably conclude that HSCP represents value to the Irish health system and should be adopted as part of treatment as usual in Irish EDs. While these dedicated HSCP teams are currently in situ across the majority of ED in Ireland, further work is ongoing to establish core standards of care across these teams.
The study was carried out on data collected in one setting within the Irish context and the results may not be generalisable to other contexts where healthcare infrastructure, processes and costs may be different.
Supporting information
S1 File. Supplementary materials (Appendix 1 and 2).
https://doi.org/10.1371/journal.pone.0298162.s001
(DOCX)
References
- 1. van Tiel S, Rood PP, Bertoli-Avella AM, Erasmus V, Haagsma J, van Beeck E, et al. Systematic review of frequent users of emergency departments in non-US hospitals: state of the art. European journal of emergency medicine. 2015;22(5):306–15. pmid:25647038
- 2. Soril LJ, Leggett LE, Lorenzetti DL, Noseworthy TW, Clement FM. Characteristics of frequent users of the emergency department in the general adult population: a systematic review of international healthcare systems. Health policy. 2016;120(5):452–61. pmid:26947060
- 3. Cassarino M, Robinson K, Trépel D, O’Shaughnessy Í, Smalle E, White S, et al. Impact of assessment and intervention by a health and social care professional team in the emergency department on the quality, safety, and clinical effectiveness of care for older adults: A randomised controlled trial. PLoS Medicine. 2021;18(7):e1003711. pmid:34319971
- 4. Driesen BE, Van Riet BH, Verkerk L, Bonjer HJ, Merten H, Nanayakkara PW. Long length of stay at the emergency department is mostly caused by organisational factors outside the influence of the emergency department: A root cause analysis. PloS one. 2018;13(9):e0202751. pmid:30216348
- 5. Hughes JM, Freiermuth CE, Shepherd‐Banigan M, Ragsdale L, Eucker SA, Goldstein K, et al. Emergency department interventions for older adults: a systematic review. Journal of the American Geriatrics Society. 2019;67(7):1516–25. pmid:30875098
- 6. Cassarino M, Robinson K, Quinn R, Naddy B, O’Regan A, Ryan D, et al. Impact of early assessment and intervention by teams involving health and social care professionals in the emergency department: A systematic review. PLoS One. 2019;14(7):e0220709. pmid:31365575
- 7. Waldron N, Dey I, Nagree Y, Xiao J, Flicker L. A multi-faceted intervention to implement guideline care and improve quality of care for older people who present to the emergency department with falls. BMC geriatrics. 2011;11(1):1–8.
- 8. Moss JE, Houghton LM, Flower CL, Moss DL, Nielsen DA, Taylor DM. A multidisciplinary Care Coordination Team improves emergency department discharge planning practice. Medical Journal of Australia. 2002;177(8):427–39. pmid:12381253
- 9. Information IH, Authority Q. Guidelines for the Economic Evaluation of Health Technologies in Ireland 2010: Health Information and Quality Authority; 2010.
- 10. Wylie K, Crilly J, Toloo G, FitzGerald G, Burke J, Williams G, et al. Emergency department models of care in the context of care quality and cost: A systematic review. Emergency Medicine Australasia. 2015;27(2):95–101.
- 11. Corcoran C, Ruiz-Adame M, Trépel D. Early assessment and intervention by a dedicated health and social care professional team in the emergency department in older adults compared to treatment-as-usual: Health Economic Analysis Plan for within-trial cost effectiveness analysis. HRB Open Research. 2023;6(22):22.
- 12. Cassarino M, Robinson K, O’Shaughnessy Í, Smalle E, White S, Devlin C, et al. A randomised controlled trial exploring the impact of a dedicated health and social care professionals team in the emergency department on the quality, safety, clinical and cost-effectiveness of care for older adults: a study protocol. Trials. 2019;20:1–7.
- 13. Parenti N, Reggiani MLB, Iannone P, Percudani D, Dowding D. A systematic review on the validity and reliability of an emergency department triage scale, the Manchester Triage System. International journal of nursing studies. 2014;51(7):1062–9. pmid:24613653
- 14. Hobbins A, Barry L, Kelleher D, Shah K, Devlin N, Goni JMR, et al. Utility values for health states in Ireland: a value set for the EQ-5D-5L. Pharmacoeconomics. 2018;36:1345–53. pmid:30051267
- 15. Zellner A, Huang DS. Further Properties of Efficient Estimators for Seemingly Unrelated Regression Equations. International Economic Review. 1962;3(3):300–13.
- 16. Willan AR, Briggs AH, Hoch JS. Regression methods for covariate adjustment and subgroup analysis for non‐censored cost‐effectiveness data. Health economics. 2004;13(5):461–75. pmid:15127426
- 17. Mutubuki EN, El Alili M, Bosmans JE, Oosterhuis T, Snoek F J., Ostelo RW, et al. The statistical approach in trial-based economic evaluations matters: get your statistics together! BMC health services research. 2021;21(1):475. pmid:34011337
- 18. Cassarino M, Quinn R, Boland F, Ward ME, McNamara R, O’Connor M, et al. Stakeholders’ perspectives on models of care in the emergency department and the introduction of health and social care professional teams: A qualitative analysis using World Cafés and interviews. Health Expectations. 2020;23(5):1065–73.
- 19. Gillespie P, Hobbins A, O’Toole L, Connolly D, Boland F, Smith SM. Cost-effectiveness of an occupational therapy-led self-management support programme for multimorbidity in primary care. Family Practice. 2022;39(5):826–33. pmid:35137039
- 20. Smith S, Jiang J, Normand C, O’Neill C. Unit costs for non-acute care in Ireland 2016–2019. HRB open research. 2021;4. pmid:35317302