Trial-Based Economic Evaluations Conducted alongside Studies Embedded in Australian Healthcare Services and Programs: A Scoping Review

Abstract

Objectives

The objective of this study was to map the literature and provide an overview of economic evaluations (EEs) conducted alongside studies embedded in Australian healthcare services or programs by: (1) providing a broad overview of the volume, characteristics and types of EEs and (2) examining the subset of EEs impacting older Australians in greater depth including characteristics (type, costs, outcomes, perspectives adopted, time horizon) and quality.

Methods

We conducted a scoping review of full trial-based EEs of studies embedded in Australian healthcare services and programs, published in peer-reviewed journal publications. Medline, Informit, ProQuest Central, Web of Science, Econlit, PsychINFO, Cumulative Index to Nursing and Allied Health Literature (CINAHL), Cost-Effectiveness Analysis (CEA) Registry, and the International Health Technology Assessment (HTA) database were searched from inception to May 2025. Quality assessment was conducted using the Consensus Health Economic Criteria (CHEC-list).

Results

We identified 88 articles (86 studies) reporting 107 EEs. Cost-effectiveness analyses (CEA) were most common (n = 53, 50%). Most interventions were delivered in ambulatory care settings (n = 41, 47%) and included participants from metropolitan areas (n = 81, 92%).

A total of 28 studies (29 comparisons) were with older adults, reporting 38 EEs. Among these, CEAs (n = 16, 42%) and adoption of a health system perspective (n = 11, 29%) were most frequent. Most EEs included intervention costs (n = 25, 86%) and healthcare costs (n = 28, 97%). Most interventions evaluated in CEA and cost-utility analysis were found to be more costly and more effective (n = 16, 43%) or less costly and more effective (n = 15, 41%). The median CHEC-list was 16/19 (IQR 13–17).

Conclusions

Our review identified a diverse range of intervention types and evaluation methods, but a relative lack of cost–benefit analyses, despite their value for cross-sector comparisons. Most evaluations were conducted in metropolitan settings and few incorporated broader costs such as aged care, particularly regarding interventions targeting older adults. Future research should prioritize evaluations that include rural and remote populations and adopt broader perspectives, ensuring the full value of interventions is captured to better inform policy and decision-making.

Supplementary Information

The online version contains supplementary material available at 10.1007/s41669-026-00652-z.

Key Points for Decision Makers

There are a growing number of economic evaluations being conducted alongside studies embedded in Australian healthcare services and programs, with diversity in settings and intervention types.

Future economic evaluations in this area should prioritize evaluations conducted outside of metropolitan areas, and consider a broader range of costs (such as aged care costs, particularly in studies with older adults) to ensure the full value of interventions are being captured to better inform policy and decision-making.

Introduction

Embedded research integrates research activities into routine clinical practice, is tailored to local needs and resources, and is co-produced by researchers, local clinicians, and/or program managers [1]. Embedded research offers a pragmatic approach to generating contextually relevant evidence that can directly inform decision-making and strengthen health systems [2, 3]. A related but distinct concept is implementation research, which studies strategies to increase uptake of evidence-based interventions into practice [4] and may support the uptake of interventions evaluated in embedded studies [5]. By adopting elements of pragmatic trial design and leveraging existing healthcare resources and infrastructure, embedded research can improve research efficiency and relevance [6].

The value of healthcare interventions needs to be considered within the context of finite resources and health budgets [1]. Economic evaluations (EEs) support decision-making by comparing intervention costs and outcomes and are a crucial element of embedded research [1]. A 2017 review described characteristics and costs of embedded pragmatic clinical trials in healthcare and community settings across multiple countries in North America, Europe, Oceania, Africa, and Asia [7]. While the review explored trial costs, it did not focus on full trial-based EEs [7]. Additionally, as embedded research is an expanding area, an updated exploration of studies is warranted [7].

Understanding economic evidence within a particular context is critical, given the unique structure of each country’s healthcare system and population characteristics. Funding arrangements shape service delivery, perspectives adopted in EEs, and how findings are used in decision-making. Australia’s rapidly aging population is placing increasing pressure on health and aged care systems [8]. Aged care is projected to be one of the fastest growing areas of government spending in the next decade, with an estimated average annual growth of 5.7% [9]. Concurrently, there is a strong commitment to delivering high-quality care for older Australians, who are valuable members of their communities [10]. This was supported by the Aged Care Royal Commission, which also highlighted the need for a sustainable aged care system in the face of changing population needs [10]. Given the importance of cost-effectiveness in supporting program sustainability, to support healthy aging, we must understand the impacts and costs of health interventions targeted at older Australians. This ensures future resource allocation and policy development are informed by robust, context-specific economic evidence.

This scoping review aims to map the literature and provide an overview of EEs of studies embedded in healthcare services/programs. Scoping review methodology was selected for this review as the goal was to identify and map the available evidence and to describe the key characteristics of studies in this area [11, 12]. Findings of the scoping review would facilitate the identification of research gaps in the field, highlighting areas for future research with more targeted research questions [11, 12]. To inform about the conduct and impact of these EEs, we reviewed the broad literature in any clinical area. Given the priority of addressing the needs of an aging population, we undertook a more detailed exploration of evidence impacting older Australians. Our objectives were:

1. To provide a broad overview of the volume, study characteristics, and types of EEs.

2. To examine in detail the characteristics (type, costs, outcomes, perspectives, time horizon) and quality of EEs that impact older Australians.

Methods

This review was guided by the Joanna Briggs Institute (JBI) scoping review methodology [12] and reported according to the Preferred Reporting Items for Systematic reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR) statement [13]. The protocol was prospectively registered with Open Science Framework (10.17605/OSF.IO/EANWH).

Eligibility Criteria

Population

Studies investigated Australian participants receiving healthcare. For objective 1, no limitations were placed on age or health condition. For objective 2, study inclusion criteria specified individuals aged ≥ 65 years, or an average age of ≥ 65 years among included participants. This age threshold was selected as it aligns with the population grouping defined by the Australian government and also represents the general age requirement for access to aged care support services in Australia [8, 14].

Concept

Studies investigated intervention effect on patient health outcomes or implementation outcomes (e.g., reach, adoption, fidelity) and reported results of full trial-based EEs [e.g., cost-effectiveness analysis (CEA), cost-utility analysis (CUA), cost-benefit analysis (CBA), cost-minimization analysis (CMA)] conducted alongside studies embedded in Australian healthcare services or programs.

Context

Studies were conducted in Australia, or in multiple countries provided that Australian results were reported separately. Interventions focused on the treatment or management of a patient’s health/health condition. An intervention was defined as an act performed for, or with, a person to improve, maintain, promote, or modify health, functioning, or health conditions [15]. Interventions were embedded in the healthcare system—delivered within primary, secondary, or tertiary care settings and in hospital or community-based environments.

To be considered embedded, studies needed to meet all the following criteria:

• recruited existing patients within a health service,

• delivered (at least part of) an intervention within a health service,

• included at least one author with a clinical affiliation (affiliations with a local health district, hospital, or other health/clinical service were included, including those with conjoint academic and clinical affiliations), and

• investigated at least one effectiveness outcome using routinely collected data.

We recognize that the embeddedness of a study can sit on a continuum and may be defined by different criteria. The criteria we are using to define embedded studies have been selected to reflect some of the key aspects of embedded research. The recruitment of participants through a health service and utilizing existing clinical staff to deliver interventions supports the integration of research activities into clinical practice [16, 17]. As embedded research is targeted at generating contextually relevant evidence to improve patient outcomes, the inclusion of an author with a clinical affiliation serves as a proxy for a collaborative effort between health service staff and researchers in implementing interventions that are tailored to the specific health service context. Routinely collected data support the feasibility of outcome collection and inclusion of outcomes that are of particular relevance to patients and health services, optimizing the relevance of evidence generated on the impact of a particular intervention [18]. For transparency, studies excluded due to interventions being delivered external to a health service or not using routinely collected outcome data were reported in Appendix 1.

Types of Evidence Sources

Interventional studies including a comparator (control group/period) were eligible. Randomized and nonrandomized controlled trials (e.g., effectiveness-implementation trials, stepped-wedge designs), and quasi-experimental designs (e.g., before-and-after studies, interrupted time series) were included.

Peer-reviewed full-text journal articles were included. No restrictions were placed on study publication date. Reviews, comments, and editorials were excluded.

Search Strategy

The research team designed the search strategy with support from an experienced research librarian. An initial strategy was developed in Medline and adapted to search the other databases (Appendix 2). Search terms for study design were combined with terms related to: EEs, healthcare services and programs, and Australia. Databases included Medline, Informit, ProQuest Central, Web of Science, Econlit, PsycINFO, Cumulative Index to Nursing and Allied Health Literature (CINAHL), CEA Registry and International Health Technology Assessment (HTA) database (Appendix 2), searched from inception to May 2025.

Study/Source of Evidence selection

Study Screening, Data Extraction, and Quality Assessment

Title and abstract screening were conducted independently by two reviewers (BW, LP, SS, JC, JS, ZS, NC, JY, MP). Full-text screening was conducted independently by two reviewers (BW, LP, JC, SS, JY, MP). Quality assessment for included studies was conducted independently by two reviewers (BW and LP, JC, NC, JS, or ZS) using the Consensus Health Economic Criteria (CHEC-list) [19] (Appendix 3). A total of 19 items (encompassing EE design, parameters, costs, outcomes, and analysis) are rated yes, no, or not applicable. Scoring was conducted per intervention comparison. Disagreements were resolved through discussion and consensus, or by a third reviewer (MP) where agreement could not be reached.

A data charting tool was developed in Excel. This template was pilot tested by two reviewers (BW, LP) and refined as needed prior to data charting for the remaining studies. Data were extracted by one reviewer (BW, LP, JC) and checked by a second (BW or LP). Data on EE type, study characteristics, and location were extracted for objective 1. Additionally, data on key economic evaluation parameters, outcomes, cost categories, and quality were extracted for objective 2.

Interventions were classified on the basis of their primary content using intervention types in The Cost-Effectiveness Analysis Registry User Manual [20] (Appendix 4). Where multiple categories applied, we classified the primary intervention category via reviewer consensus. Study locations were classified using the Modified Monash Model [21]. Postcodes were derived from named study sites and clinical affiliations. Where this information was not available, information on study location was extracted as stated within available reports. Elements of the Reach, Effectiveness, Adoption, Implementation, and Maintenance (RE-AIM) framework [22] were searched in the papers reporting effectiveness results to scope the measurement or consideration of implementation measures. “Adherence” was used as a proxy term for consideration of implementation [22].

Deviations from Protocol

A search of the reference lists of systematic reviews was forgone due to the volume of studies included. The decision for all studies to be extracted by one reviewer and checked by a second was made pragmatically, and this approach is supported by the JBI guidelines [23].

Results

Our search yielded 20,410 studies, identifying 13,026 unique records (Fig. 1). Full-text screening was conducted for 776 studies, with 88 papers (86 unique studies) included in objective 1, and 28 studies included in objective 2. A complete list of included studies (and related publications) can be found in Appendix 5.

Fig. 1.

Flow diagram of studies included in the review

Objective 1 (All Clinical Areas)

Since the earliest included publication in 1994, there has been an upward trend in the volume of EE publications over time, with the greatest number of publications in 2018 and 2024 (n = 11 each) (Fig. 2). Among the 86 unique studies, 88 intervention comparisons were investigated. Across these comparisons, 107 EEs were included. CEAs were most frequently conducted (n = 53, 50%), followed by CUA (n = 35, 33%), CBA (n = 12, 11%), and CMA (n = 7, 7%). Among 19 comparisons that included multiple EE types, the majority conducted CEA and CUA (n = 13), followed by CEA and CBA (n = 3), and CUA and CBA (n = 3).

Fig. 2.

Number of published reports included in objective 1

Interventions were most commonly conducted in Victoria (n = 26, 30%), Queensland (n = 22, 25%), and New South Wales (n = 15, 17%). A total of 15 (17%) interventions were conducted in multiple states. On the basis of the Modified Monash Model [21] classification of location, the majority of interventions included participants in metropolitan areas (n = 81, 92%), followed by regional (n = 12, 14%), rural (n = 8, 9%), and remote (n = 2, 2%). Of these interventions, 13 were conducted in mixed locations. Further details on intervention locations can be found in Appendix 6.

Interventions were conducted across the ambulatory care (n = 41, 47%), inpatient (n = 29, 33%), and emergency department (n = 8, 9%) settings, with ten studies (11%) being conducted across multiple settings. The most common intervention category was care delivery (n = 38, 43%), followed by pharmaceutical (n = 10, 11%), medical procedure and health education or behavior (n = 9 each, 10%), and medical device (n = 8, 9%). Details on economic evaluation type by setting, intervention category, geography, and perspective can be found in Appendix 7. The population most commonly investigated was adults (18–64 years) and older adults (> 64 years) (n = 29 each, 33%). Detailed intervention characteristics can be found in Table 1. A minority of studies considered elements of implementation: reach (n = 1, 1%), adoption (n = 1, 1%), adherence (n = 21, 24%), and maintenance (n = 7, 8%).

Table 1.

Type of economic evaluation and intervention characteristics of studies included in objective 1

Study	Population group	Setting (delivery)	Intervention category (CEA Registry categories)	Who delivered?	Type of economic evaluation
Ariyaratne 2018 [55]	Older adults	Inpatient	Surgical	Surgeon	CEA
Asha 2014 [56]	Adults	ED	Care delivery	Nurses, doctors	CEA
Atey 2023 [57]	Older adults	ED	Care delivery	Pharmacist (and nurse)	CEA + CBA
Austin 2015 [58]	Pregnancy	Inpatient	Medical procedure	Doctors	CEA
Avent 2024 [59]	Not specified	Ambulatory care	Care delivery	Doctors	CEA
Bell 2007 [60]	Adults	Ambulatory care	Pharmaceutical	Doctors	CEA
Bramble 2021 [61]	Older adults	Inpatient	Care delivery	Nurse	CBA
Brown 2015 [62]	Children	Ambulatory care	Medical procedure	Nurse	CEA
Brusco 2015a [63]	Older adults	Inpatient	Care delivery	Nurse, physiotherapist, OT	CEA + CUA
Brusco 2023 [64]	Older adults	ED and inpatient	Care delivery	Nurse	CEA + CBA
Byrnes 2019 [65]	Older adults	Ambulatory care	Care delivery	Nurse	CUA + CBA
Carter 2022 [66]	Children	Ambulatory care	Care delivery	Allied Health Liaison Officer	CUA
Conway 2018 [67]	Older adults	Inpatient	Medical device	Nurse	CBA
Cook 1994 [68]	Adults	Inpatient and ambulatory	Medical procedure	Not specified	CUA
CRISTAL group 2024 [25]	Older adults	Inpatient and ambulatory	Pharmaceutical	Doctor, nurse	CEA + CUA
Crittenden 2022 [69]	Adults	Inpatient	Surgical	Surgeon	CUA
Cronin 2023 [70]	Adults	Ambulatory care	Health education or behavior	Not specified	CBA
Deeming 2024 [71]	Adults	Inpatient	Pharmaceutical	Surgical staff	CEA
Dinh 2016 [72]	Mixed	Inpatient	Care delivery	Trauma director, trauma clinical nurse consultant, trauma case manager	CEA
Doyle 2004 [73]	Infants	Inpatient	Care delivery	Not specified	CEA + CUA
East 2006 [74]	Pregnancy	Inpatient	Medical device	Midwife	CEA
Ekert 2001 [75]	Children	Ambulatory care	Pharmaceutical	Not specified	CUA
Fairhall 2015 [76]	Older adults	Ambulatory care	Care delivery	Physiotherapist, doctor	CEA + CUA
Fehily 2020 [77]	Adults	Ambulatory care	Health education or behavior	Occupational therapist	CEA
Foote 2007 [78]	Adults	Inpatient	Surgical	Surgeon	CUA
Frear 2021 [79]	Children	ED	Medical procedure	Burns clinician	CEA
Freeman 2021 [80]	Older adults	Ambulatory care	Care delivery	Doctor, pharmacist, nurse	CBA
Gabbe 2023 [81]	Adults	Inpatient	Care delivery	Allied health clinicians (physiotherapy, OT, social worker, speech pathology, nutrition, psychology, neuropsychology, orthotics and prosthetics, allied health assistants, and grade 4 team leader)	CUA + CBA
Gao 2023 [82]	Older adults	ED	Pharmaceutical	Neurologist, stroke research nurse, CT radiographers and two paramedics	CUA
GeeKee 2017 [83]	Children	ED and ambulatory care	Medical procedure	Nurse	CEA
Giles 2021 [84]	Pregnancy	Ambulatory care	Immunization	Not specified	CEA
Harris 2005 [85]	Adults	Ambulatory care	Pharmaceutical	Doctors	CUA
Heard 2017 [86]	Pregnancy	Inpatient	Medical procedure	Obstetrician	CEA + CUA
Ho 2021 [87]	Adults	Inpatient	Medical device	Radiologist	CEA
Hua 2022 [88]	Adults	Ambulatory care	Medical device	Doctors	CUA
Huang 2018 [89]	Infants	Inpatient	Medical device	Not specified	CEA
Hwang 2019 [90]	Older adults	Ambulatory care	Care delivery	Physiotherapists	CUA
Ibrahim 2019 [91]	Mixed	ED and ambulatory care	Pharmaceutical	Doctor	CEA + CUA
Inderjeeth 2022 [92]	Older adults	ED	Care delivery	Not specified	CEA + CUA
Jansons 2018 [26]	Older adults	Ambulatory care	Health education or behavior	Exercise physiologist	CUA
Johnston 1998 [93]	Mixed	Ambulatory care	Care delivery	Not specified	CEA
Kim 2022 [94]	Older adults	ED	Care delivery	Doctor	CUA
Laurence 2010 [95]	Pregnancy	Ambulatory care	Diagnostic	Not specified	CEA
Laurie 2023 [96]	Pregnancy	Ambulatory care	Care delivery	Midwife	CMA
Lewandowska 2021 [97]	Mixed	Ambulatory care	Care delivery	Doctors	CEA
Ling 2021 [98]	Older adults	Inpatient	Care delivery	Nurse	CEA
Lockwood 2025 [99]	Older adults	Inpatient and ambulatory	Health education or behavior	OT	CEA
Maher 2012 [100]	Adults	Inpatient	Surgical	Surgeon	CMA
Martin 2025 [101]	Older adults	Inpatient	Screening	Physiotherapist or OT (staff or student)	CMA
Maru 2018 [102]	Older adults	Ambulatory care	Care delivery	Nurse	CUA
Maru 2019 [103]	Older adults	Ambulatory care	Health education or behavior	Physiotherapist, exercise physiologist	CUA + CBA
McPhail 2022 [104]	Children	Ambulatory care	Medical procedure	OT	CUA
McPhail 2022 [104]	Children	Ambulatory care	Medical procedure	OT	CUA
Moodie 2006 [105]	Older adults	Inpatient	Care delivery	Not specified (MDT)	CEA
Moodie 2006 [105]	Older adults	Inpatient	Care delivery	Not specified (MDT)	CEA
Mortimer 2013 [106]	Adults	Ambulatory care	Care delivery	Doctors, nurses	CEA
Mortimer 2018 [107]	Adults	ED	Care delivery	Doctors, nurses	CEA
Moss 2007 [108]	Pregnancy	Ambulatory care	Care delivery	Dietitian, doctor	CEA
Mudiyanselage 2023b [27]	Older adults	Ambulatory care	Care delivery	Doctor, nurse	CUA
Neo 2024 [109]	Pregnancy	Ambulatory care	Care delivery	Midwife, GP, obstetric care providers, obstetric specialist, hospital staff	CMA
Pakneshan 2024 [110]	Adults	Ambulatory care	Screening	Endoscopists	CEA
Pasieczny 2011 [111]	Adults	Ambulatory care	Health education or behavior	Psychologists, social workers, psychiatrists, OTs, nurse	CBA
Pham 2023 [112]	Older adults	Inpatient	Medical device	Nurse, physiotherapist	CEA
Pouliopoulos 2024 [113]	Adults	Ambulatory care	Diagnostic	Cardiologist/MRI consultant	CUA
Russell 2024 [114]	Not specified	ED and ambulatory care	Care delivery	Doctor, nurse, Aboriginal health practitioner	CEA
Sekhon 2023 [115]	Pregnancy	Ambulatory care	Medical device	Not specified	CEA
Sharma 2018 [116]	Older adults	Inpatient	Health education or behavior	Dietitian	CEA + CUA
Sharma 2022 [117]	Infants	Ambulatory care	Screening	Not specified	CUA
Siskind 2013 [118]	Adults	Ambulatory care	Care delivery	Nurse, OT	CMA
Smith 2002 [119]	Adults	Ambulatory care	Pharmaceutical	Doctor, nurse	CMA
Smith 2022 [120]	Children	Ambulatory care	Care delivery	Physiotherapist	CEA
Snowdon 2021 [121]	Mixed	Ambulatory care	Care delivery	Not specified	CEA
Stark 2019 [122]	Infants	Ambulatory care	Diagnostic	Doctor	CEA + CUA
Taylor 2016 [123]	Adults	Inpatient	Pharmaceutical	Not specified	CEA + CUA
Taylor 2020 [124]	Older adults	Inpatient	Care delivery	Allied health MTD (OTs, physiotherapists, social workers, AHAs)	CEA
Tew 2023 [125]	Adults	Inpatient	Diagnostic	Radiologist	CEA + CUA
Treacy 2018 [126]	Older adults	Inpatient	Health education or behavior	Physiotherapists	CEA
Trentino 2021 [127]	Adults	Ambulatory care	Screening	Not specified	CEA
Tuffaha 2014 [128]	Adults	ED and inpatient	Care delivery	Doctors, nurses	CBA
Twigg 2013 [129]	Adults	ED and inpatient	Care delivery	Nurses	CEA
Walsh 2025 [130]	Adults	Ambulatory care	Care delivery	Podiatrist	CUA
Ward 2005 [131]	Older adults	Inpatient	Medical device	Surgeon	CEA
Whitty 2023 [132]	Pregnancy	Inpatient	Medical procedure	Obstetrician	CEA + CUA
Wigg 2018 [133]	Adults	Inpatient and ambulatory	Care delivery	Nurse, specialist, GP, dietician, alcohol counselors	CEA
Williams 2019 [134]	Adults	Ambulatory care	Health education or behavior	Health coaches, physiotherapist	CEA + CUA
Xu 2008 [135]	Children	Ambulatory care	Care delivery	ENT specialist	CMA
Yap 2018 [136]	Adults	ED	Pharmaceutical	Emergency physician	CEA + CBA
Young 2018 [137]	Older adults	Inpatient	Surgical	Surgeon	CUA

CBA cost-benefit analysis, CEA cost-effectiveness analysis, CMA cost-minimization analysis, CUA cost-utility analysis, ED emergency department, MDT multidisciplinary team

^aBrusco (2015) reported the results for a longer follow-up duration for the intervention investigated in Brusco (2014)

^bData for this study was also drawn from the publication by Mudiyanselage (2019)

Objective 2 (Older Adults)

Among 28 included studies, 29 intervention comparisons were included and 38 EEs were conducted. CEAs (n = 16, 42%) were most commonly conducted, followed by CUA (n = 14, 37%) and CBA (n = 7, 18%). Among the nine comparisons that included multiple EE types, the majority conducted CEA and CUA (n = 5).

Interventions were most commonly conducted in the inpatient setting (n = 14, 48%), followed by ambulatory care (n = 8, 28%). The most common intervention category was care delivery (n = 16, 55%), followed by health education and behavior (n = 5, 17%). Intervention characteristics are detailed in Table 2 and Appendix 8.

Table 2.

Intervention and outcome characteristics of studies included in objective 2 (older adults)

Study	Setting (delivery)	Intervention description	Control description	Effectiveness outcomes used in economic evaluation	Effectiveness outcome instrument
Ariyaratne 2018 [55]	Inpatient	Coronary artery bypass grafting (CABG) performed as an isolated procedure.	Percutaneous coronary intervention (PCI) using bare-metal stents or drug-eluting stents.	Major adverse cardiac and cerebrovascular event, all-cause mortality, life years Other cardiovascular event, non-cardiovascular event	Clinical records linked to same-hospital admissions and national death index data
Atey 2023 [57]	ED	Partnered pharmacist medication charting (PPMC): a pharmacist-documented best-possible medication history (BPMH), followed by a clinical conversation between a pharmacist and a medical officer to jointly develop a treatment plan and chart medications at the earliest possible time.	Usual care: medical-officer-led traditional medication charting without a pharmacist-obtained BPMH or clinical conversation.	Number of high/extreme risk medication error (cost-effectiveness) Relative stay index (RSI) (cost–benefit)	Digital medical records, Healthcare Software Clinical Suite
Bramble 2021 [61]	Inpatient	Employment of a specialist Parkinson’s nurse in a regional community.	Usual care: standard healthcare services for Parkinson’s disease without access to a specialist nurse.	LOS	Patient health record
Brusco 2015 [63]	Inpatient	Rehabilitation service on Saturday (1 h of physiotherapy and 1 h of occupational therapy, planned by the weekday therapist) + usual care	Usual care: rehabilitation from Monday to Friday and weekend care that included full nursing care and limited medical service.	QALY, functional independence	EQ-5D-3L and Functional Independence Measure
Brusco 2023 [64]	ED and inpatient	Sepsis Pathway: a whole of hospital pathway supporting nurse initiation and the use of early warning criteria in combination with severity indicators, focusing on the first 6 h.	Usual care: sepsis management practices in place prior to the implementation of the Sepsis Pathway.	In-hospital all-cause mortality	Hospital database
Byrnes 2019 [65]	Ambulatory care	SAFETY: an atrial fibrillation-specific nurse-led, home-based management strategy involving comprehensive health assessment, risk delineation, individualized care, and annual clinic visits.	Standard post-discharge management.	QALY	EQ-5D-3L
Conway 2018 [67]	Inpatient	Forced air warming for the duration of the procedure + usual care. Nursing staff monitored for signs of hyperthermia and thermal discomfort so that the temperature could be titrated.	Usual care: passive warming with heated cotton blankets throughout the peri-procedural period applied at the discretion of the nurse.	Thermal comfort	5-point thermal comfort scale
CRISTAL group 2024 [25]	Inpatient and ambulatory	Enoxaparin (40 mg/d) subcutaneously for 35 days after THA and for 14 days after TKA.	Aspirin (100 mg/day) orally for 35 days after THA and for 14 days after TKA.	QALY and venous thromboembolism avoided	EQ-5D-5L and online data collection form
Fairhall 2015 [76]	Ambulatory care	12-month interdisciplinary, multifactorial, individualized intervention targeting frailty. 10 physiotherapy visits and home program of lower limb balance and strength exercises (20–30 min, 3–5 times per week). Those meeting the weight loss frailty criterion underwent dietitian assessment and management. Medical management: medication review and chronic condition management.	Usual care: assessment and delivery of care needs, and medical and allied health management, if needed.	QALY, transition out of frailty	EQ-5D, Cardiovascular Health Study (CHS) frailty criteria
Freeman 2021 [80]	Ambulatory care	Comprehensive face-to-face medicine management consultation with an integrated practice pharmacist within 7 days of discharge, followed by a consultation with the GP and further pharmacist consultations as needed.	Usual care at the patient’s GP.	Incidence of hospital re-admissions and ED presentations	Hospital and medical center records
Gao 2023 [82]	ED	Tenecteplase (0.25 mg/kg [maximum 25 mg], administered as an intravenous bolus over 10 s).	Standard of care: alteplase (0.9 mg/kg [maximum 90mg] administered intravenously with 10% as a bolus over 1 min and 90% as an infusion over 1 h).	QALY	Utility-weighted mRS score
Hwang 2019 [90]	Ambulatory care	Home-based telerehabilitation program consisted of 12-week exercise and education sessions delivered twice-weekly. Exercise session was supervised by a physiotherapist with education facilitated by a physiotherapist and a nurse via videoconferencing for up to four participants.	Center-based group rehabilitation program based on current recommended guidelines encompassing education, aerobic, and strength training exercise. Each 60-min exercise session was led by a physiotherapist and nurse, while same-day education sessions (approximately 60 min), were conducted by a multidisciplinary team.	QALY	EQ-5D-3L
Inderjeeth 2022 [92]	ED	Fracture liaison service: patients with minimal trauma fractures were identified via ED information system, assessed for osteoporosis, and provided with a management plan.	Usual care.	QALY, recurrent fracture rate	EQ-5D and hospital records
Jansons 2018 [26]	Ambulatory care	12-month individualized home-based exercise program with health coaching, supervised by an exercise physiologist through five 30-min telephone calls over the first 10 weeks. Participants were encouraged to complete three 1-h exercise sessions per week.	12-month individualized gym-based exercise program and health coaching supervised by an exercise physiologist. Participants were encouraged to complete three 1-h exercise sessions per week.	QALY	EQ-5D-3L
Kim 2022 [94]	ED	Victorian Stroke Telemedicine program where ED clinicians can contact a stroke specialist who reviews brain images, provided a diagnosis, and recommended best-practice treatments.	Control period preceding the implementation of the program.	QALY	EQ-5D-3L
Ling 2021 [98]	Inpatient	No-CAUTI: a nurse-led multifaceted intervention with four components: catheter insertion criteria guidelines, insertion and maintenance care bundle, nurse-led catheter removal protocol, and clinician competency assessment framework.	Usual urinary catheterization practice.	Prevalence of indwelling urinary catheterizations (catheterization rate)	Clinical records
Lockwood 2025 [99]	Inpatient and ambulatory	Pre-discharge home assessment visit + usual-care hospital-based discharge planning.	Usual-care hospital-based discharge planning.	Functional independence	Functional Independence Measure
Martin 2025 [101]	Inpatient	Transdisciplinary Initial Neurological Screening Assessment (TINSA): assessing domains of social situation, home environment, cognition, mood, and function.	Usual discipline-specific allied health stroke assessment.	QoL	EQ-5D-3L
Maru 2018 [102]	Ambulatory care	Nurse-led Intervention for Less Chronic Heart Failure (NIL-CHF): a multidisciplinary (specialist nurses with close collaboration with community healthcare providers) home and clinic-based program designed to prevent progressive cardiac dysfunction.	Standard post-discharge care.	Life-years and QALY	Follow-up records and EQ-5D-3L
Maru 2019 [103]	Ambulatory care	Twice-weekly exercise sessions for 12 weeks followed by once-weekly sessions for 12 weeks, 50 min of graded moderate intensity interval and resistance training using traditional gym-based equipment, with balance exercises included as appropriate + HF-DMP.	12 weeks of weekly review with disease monitoring, review and progression of the home exercise program, and weekly structured self-management education classes.	QALY	AQoL-4D
Moodie 2006 [105]	Inpatient	Specialized stroke care unit with a multidisciplinary team.	General medical ward care.	Thorough adherence to process of care measures and at least one severe complication (one or more severe complication avoided)	Audit of patient medical records
Moodie 2006 [105]	Inpatient	Mobile service with a multidisciplinary team that reviewed stroke patients across different wards.	General medical ward care.	Thorough adherence to process of care measures and at least one severe complication (one or more severe complication avoided)	Audit of patient medical records
Mudiyanselage 2023 [27]	Ambulatory care	Personalized telehealth remote patient monitoring intervention where patients were provided with an Android tablet and peripheral equipment, to submit daily health data. Nurses monitored data 7 days a week, responded to alerts, and provided fortnightly health coaching via videoconferencing. Patients continued routine GP care.	Usual care: advise to manage their condition in conjunction with their community GP, outpatient clinic appointments and ED where appropriate.	QALY	AQoL-8D
Pham 2023 [112]	Inpatient	Ambient Intelligent Geriatric Management (AmbIGeM) System that combined wearable sensors with artificial intelligence to trigger alerts to hospital staff before a fall.	Usual care: routine best-practice falls prevention activities.	Experience of one or more injurious falls during study patients’ stay (injurious falls prevented/change in injurious falls rate)	Hospital records
Sharma 2018 [116]	Inpatient	Individualized nutritional plans provided by a dietitian and initiated within 24 h of admission, including oral nutrition supplements (ONS), snacks, food fortification, counseling, and monthly follow-up calls for 2 months post-discharge.	Usual care: patients underwent nutrition screening, but dietitian referrals depended on clinicians.	Unit improvement in nutrition status, QALY	PG-SGA and EQ-5D-5L
Taylor 2020 [124]	Inpatient	Comprehensive multidisciplinary allied health service provided on Saturdays in addition to the usual weekday service.	Usual staffing of allied health from Monday to Friday.	Functional independence	Functional Independence Measure
Treacy 2018 [126]	Inpatient	Six 1-h group-based progressive standing balance circuit classes for 2 weeks + usual therapy.	Usual therapy: assessment and treatment by a multidisciplinary team. Physiotherapy was mainly group-based, participants typically spent at least 2 h in the rehabilitation gym once or twice daily.	Mobility	Short physical performance battery
Ward 2005 [131]	Inpatient	Drug eluting stents (sirolimus-coated) for the treatment of restenosis.	Bare stents.	Target lesion revascularization rates	Follow-up telephone call and hospital reports
Young 2018 [137]	Inpatient	Self-expanding metallic stent placed through the obstructing lesion using a combined endoscopic and fluoroscopic approach. If stenting was unsuccessful, surgery was performed as deemed appropriate by the operating surgeon.	Surgical decompression based on the surgeon’s judgment and pathology encountered.	QALY	EQ-5D-3L

AQoL-4D Assessment of Quality of Life-4 Dimensions, AQoL-8D Assessment of Quality of Life-8 Dimensions, ED emergency department, EQ-5D-3L European Quality of Life 5 Dimensions 3 Level Version, GP general practitioner, LOS length of stay, PG-SGA Patient-Generated Subjective Global Assessment, QALY quality adjusted life years, RCT randomized controlled trial

Effectiveness study designs included randomized controlled trials (n = 16, 57%), cohort studies (n = 6, 21%), nonrandomized controlled trials (n = 3, 11%), and before-and-after studies (n = 4, 14%). The effectiveness outcomes and their measurement instruments can be found in Table 2.

In Australia, the healthcare system is jointly funded by federal and state governments, private health insurers, and individuals through out-of-pocket payments [24]. The costs covered by the government (such as health services and products under the Medicare Benefits Schedule and Pharmaceutical Benefits Scheme, hospital-related costs) would need to be considered within a health system perspective. The health service perspective, such as that of a hospital, would be limited to the costs incurred at the service level. Perspectives adopted in the EEs investigating older adults included those of the health system (n = 11, 38%), health service (n = 6, 21%), health system and out-of-pocket costs (n = 3, 10%), state health authority and societal (n = 2 each, 7%), and health system and community payer (n = 1, 3%) (Table 3). The perspective was not clearly reported in four (14%) evaluations. The time horizon of the included EEs ranged from 1 to 60 months, with 12 months being the most common (n = 11, 38%). Across the comparisons, a range of costs including implementation, intervention, healthcare, community and aged care, productivity, and out-of-pocket costs were included (Appendix 9). Incremental cost-effectiveness ratios (ICERs) from the included CEA and CUA comparisons (n = 37) fell most commonly in the northeast (more effective, more costly) (n = 16, 43%) and southeast (more effective, less costly) (n = 16, 43%) quadrants of the cost-effectiveness plane (Figs. 3 and 4). In the three cost-utility analyses where ICERs fell in the northeast quadrant, ICERs ranged from $10,666 to $491,572 per QALY gained [25–27], when adjusted to 2025 AUD [28]. When compared against a suggested Australian reference cost-effectiveness threshold of $40,253 (2025 AUD) [29], only one of these interventions would be considered cost-effective [27].

Table 3.

Characteristics of economic evaluations included in objective 2 (older adults)

Study	Type of economic evaluation	Stated perspective	Time horizon (months)
Ariyaratne 2018 [55]	CEA	Public hospital payer	32
Ling 2021 [98]	CEA	State public sector health authority	16
Moodie 2006 [105]	CEA	Broad health sector including costs and outcomes for the government and patients	7
Pham 2023 [112]	CEA	Healthcare system	60
Taylor 2020 [124]	CEA	Not reported	6
Treacy 2018 [126]	CEA	Hospital system	3
Ward 2005 [131]	CEA	Not reported	12
Lockwood 2025 [99]	CEA	Health service perspective	6
Hwang 2019 [90]	CUA	Healthcare provider	6
Jansons 2018 [26]	CUA	Societal	12
Kim 2022 [94]	CUA	Societal	12
Maru 2018 [102]	CUA	Healthcare system	50
Mudiyanselage 2023 [27]	CUA	Health service	12
Young 2018 [137]	CUA	Not reported	1
Gao 2023 [82]	CUA	Healthcare system perspective	3
Brusco 2015 [63]	CEA + CUA	Health system inclusive of private costs	12
Inderjeeth 2022 [92]	CEA + CUA	Health system	12
Fairhall 2015 [76]	CEA + CUA	Health and community care funder (including both health and aged care)	12
Sharma 2018 [116]	CEA + CUA	Healthcare sector (Medicare)	3
CRISTAL group 2024 [25]	CEA + CUA	Health system perspective	Not reported
Atey 2023 [57]	CEA + CBA	Health system’s perspective	12
Brusco 2023 [64]	CEA + CBA	Healthcare sector	12
Byrnes 2019 [65]	CUA + CBA	Public healthcare system	24
Maru 2019 [103]	CUA + CBA	State health department	12
Bramble 2021 [61]	CBA	Not reported	Not reported
Conway 2018 [67]	CBA	Organization responsible for deciding which resources to use for thermal comfort during procedures in their cardiac catheterization laboratory	Not reported
Freeman 2021 [80]	CBA	Healthcare system	12
Martin 2025 [101]	CMA	Health sector	3

CBA cost-benefit analysis, CEA cost-effectiveness analysis, CMA cost-minimization analysis, CUA cost-utility analysis

Fig. 3.

Incremental cost-effectiveness ratio (ICER) quadrants of economic evaluations included in objective 2 (older adults), by intervention category. a ICER quadrants of the included cost-utility analyses, b ICER quadrants of the included cost-effectiveness analyses. The numbers in each triangle represent the number of ICER comparisons under each intervention category. The value of the respective ICERs is not represented in this figure. If a point ICER was not reported, the cost-effectiveness plane was used to determine the quadrant that the majority of bootstrapped estimates fell into. Fairhall (2015) was excluded from a as no ICER or cost-effectiveness plane was presented for cost-utility analysis outcome

Fig. 4.

Incremental cost-effectiveness ratios for included cost-effectiveness analyses of interventions targeted at older adults. Negative values indicate ICERs in the southeast plane (interventions that were more effective and less costly). ICERs for $/death avoided, $/life-year gained, and $/non-cardiovascular event avoided in Ariyaratne (2018) indicated that the intervention was more costly and less effective than the comparison (northwest plane) and are therefore not included in this figure

Scores for quality assessment of the EEs investigating older adults using the CHEC-list ranged from 10 to 19 out of 19 points. The median score was 16 (IQR 13–17). Items 1 (study population description), 3 (research question), 4 (economic study design), and 12 (outcome valuation) were appropriately reported in all included EEs. Poorly reported items included: item 7, identification of all relevant costs; item 15 (19/29, 67%), related to the conduct of appropriate sensitivity analysis; and item 19 (12/29, 41%), related to the consideration of ethical aspects and impact of population characteristics on distributional implications. Quality assessment for all studies included in objective 2 can be found in Appendix 10.

Discussion

This review provides a first comprehensive mapping of EEs of studies embedded in Australian healthcare services. We identified various intervention types and evaluation methods across healthcare settings, with CEAs and CUAs being the most common approaches. Most studies were conducted in metropolitan settings. While many studies assessed intervention and healthcare costs, few incorporated broader categories such as aged care costs, particularly in evaluations targeting older adults.

The increasing number of publications in this area likely reflects growing recognition of the importance of EEs for supporting decision-making in healthcare [30, 31]. Our findings suggest that to date, CEA and CUA have been the dominant approaches, while CBAs formed a small proportion of the included studies. There is increasing interest in the use of CBAs by government bodies to inform allocative efficiency [32]. CBA can extend beyond health-sector outcomes, enabling cross-sector comparisons of interventions and capturing broader economic, social, environmental, and cultural impacts [32, 33]. Greater application of CBA may help to highlight the societal benefits of healthcare interventions.

The lack of published CBAs may, in part, reflect the challenges of conducting CBAs within healthcare. CBA requires the valuation of health benefits in monetary terms using revealed or stated preferences [34]. Where a market value for a health outcome is unavailable, stated preferences (e.g., willingness to pay [WTP]) are used to determine the value of those health benefits. While it is recommended that metrics such as statistical life year, quality-adjusted life-years, and disability-adjusted life-years are used as the basis for valuation, there is no definitive consensus on their monetary values [32]. The WTP for a certain health outcome may vary significantly across population groups, raising concerns about consistency and fairness [35]. This variation underscores the need to ensure that WTP estimates reflect the preferences of disadvantaged populations, as overlooking these groups may have important implications for health equity [35].

Quality assessment revealed that the majority of EEs did not adequately consider the equity implications of delivering interventions. In recent years, addressing inequity in health has been identified as a key objective of the Australian government’s strategy for optimizing population health [36]. When collecting data on health-related quality of life, including responses from groups experiencing health disparities can help to ensure representative reference value sets [35]. Distributional cost-effectiveness analysis can be used as a method to consider the equity implications of an intervention [37]. This involves modeling the distribution of health outcomes and costs on the basis of social characteristics relevant to an intervention and evaluating intervention impacts on the distribution of health. This provides information on equity in the distribution of costs and effects to be considered alongside cost-effectiveness [38]. Future EEs should consider factors impacting equity in older adults including social characteristics and affordability and access to healthcare [39, 40], and clearly report methods used to consider equity implications.

Few studies included participants in regional, rural, and remote settings. Australians living outside of metropolitan areas experience a higher rate of chronic health conditions, with increased burden of disease with increasing remoteness [41]. Yet, these groups often face additional barriers to accessing healthcare. The increased time and costs (including patient out-of-pocket costs) associated with service delivery and access, and limited access to services providing preventive and primary healthcare, may exacerbate health disparities [42]. Relatedly, in Australia the proportion of Aboriginal and Torres Strait Islander people increases with increasing remoteness [43]. This has the potential to increase health inequities among this group, contrary to the target outcomes set out in the Closing the Gap agreement, aimed at improving the health and well-being of Aboriginal and Torres Strait Islander people [44]. It is worth nothing that none of the included studies specifically targeted Aboriginal and Torres Strait Islander people. Future research should prioritize a greater understanding of the costs and benefits of interventions conducted in regional, rural, and remote Australia to inform resource allocation that supports the optimization of health outcomes for people living in these areas, while also holding broader economic benefits [42].

Few comparisons of interventions impacting older adults included aged care and broader cost categories. In the context of Australia’s aging population, use of aged care services is expected to steadily increase over the next two decades [45]. As the aging population will be a key driver of increased government spending [46], it is important to understand the impact that interventions targeted at older Australians may have on aged care costs to inform subsequent funding of cost-effective programs. Additionally, the consideration of broader cost categories, including those related to patient out-of-pocket expenditure, community services use, and productivity will help to ensure that the true value of interventions are being accounted for as Australia increasingly moves toward the delivery of value-based healthcare [47].

In comparison with previous reviews of EEs of interventions in older adults [48–51], a greater proportion of ICERs fell in the southeast quadrant (indicating an intervention is more effective and less costly) in our review. While considering the diverse range of interventions and settings, and the potential for non-publication of cost-effectiveness results where no significant differences in effectiveness are found, these findings indicate the potential for greater efficiency when embedding research in healthcare. This may be supported by the co-design element of embedded research. Health service staff’s deep understanding of their local clinical context may lead to the development of targeted, patient-centered intervention, which may lead to greater intervention outcomes [52]. However, we recognize that this finding may be impacted by publication bias in the non-publication of unfavorable cost-effectiveness results, as well as the majority of evaluations being conducted across a short time horizon of 12 months or less and a number of evaluations adopting an appropriate yet limited perspective of the health service.

Limitations

While we undertook a comprehensive search, we acknowledge that a hand-search of systematic reviews and grey literature search may have captured further relevant studies. In particular, government-led CBAs often are not published in peer-reviewed journals but may hold particular relevance in the Australian policy context. However, the decision to include journal publications only was made pragmatically, and we were able to identify a large number of studies. Our criteria for embedded studies were selected to reflect key aspects of definitions found in the literature. However, we recognize that embedded studies hold many important characteristics, and may be conducted in various ways [1]. For instance, it is possible that studies that did not use routinely collected outcomes while otherwise being integrated within service delivery were excluded. Most studies included clinical affiliations relating to the sites at which interventions were delivered. However, the eligibility of studies including authors with any clinical affiliation may have in reality captured interventions developed with varying levels of collaboration between health service staff and researchers. Studies were also excluded when we were unable to determine whether interventions were delivered as part of a healthcare service or program using the reported information (Appendix 1). Thus, while our included studies provide a representation of EEs of studies embedded within Australian healthcare, some studies may not have been captured. The RE-AIM framework is commonly used to evaluate implementation research across a range of clinical areas [53]. However, a range of implementation frameworks exist, with different terms denoting similar concepts [54]. While our scoping of implementation measures in this review was intended to be exploratory, additional terms may have identified further studies including implementation measures. The consideration of implementation measures in embedded research may be useful in future research. In defining studies that impacted older adults, we recognize that participants under 65 years may be represented in the included samples. This approach was taken with the intention of providing a broad overview of EEs impacting older Australians.

Conclusions

Our findings highlight the growing role of EE within embedded research in Australia and underscore opportunities to strengthen its contribution to health system decision-making. The findings may assist health services in considering whether to implement similar interventions within their patient population. Researchers may use these data to identify gaps in understanding the value for money of interventions of interest and guide future EEs of studies embedded in Australian healthcare services.

Supplementary Information

Below is the link to the electronic supplementary material.

Funding

BW is supported by an Australian Government Research Training Program (RTP) Scholarship.

Declarations

Conflict of Interest

Alison Pearce is an editorial board member of PharmacoEconomics - Open. She was not involved in the selection of peer reviewers for the manuscript nor any of the subsequent editorial decisions. All other authors have no competing interests to declare that are relevant to the content of this article.

Ethical Approval

Not applicable.

Consent to Participate

Not applicable.

Consent for Publication (from Patients/Participants)

Not applicable.

Availability of Data and Material

The data supporting the findings of this study are available within the article and/or its supplementary material.

Code Availability

Not applicable.

Author Contributions

Conceptualization: Marina B. Pinheiro, Catherine Sherrington, and Belinda Wang; investigation: Belinda Wang, Luiza R Pivotto, Stanley Saputra, Josielli Comachio, Joanne Scarfe, Zoe Szewczyk, Nathalia Costa, Justin Yu, and Marina B. Pinheiro; formal analysis: Belinda Wang; supervision: Marina B. Pinheiro and Catherine Sherrington; visualization: Luiza R. Pivotto and Belinda Wang; writing—original draft: Belinda Wang; writing—review and editing: Catherine Sherrington, Luiza R. Pivotto, Stanley Saputra, Josielli Comachio, Joanne Scarfe, Zoe Szewczyk, Nathalia Costa, Alison Pearce, Leanne Hassett, Andrew Milat, Christopher Williams, Wing S. Kwok, Justin Yu, Kirsten Howard, and Marina B. Pinheiro.