Abstract
The findings of this systematic review provide preliminary evidence for the economic effectiveness of occupational therapy in acute and subacute care.
Occupational therapists in acute and subacute health care services work to facilitate independence in daily activities and improve quality of life (QoL) by addressing clients’ ability or adapting their environment (American Occupational Therapy Association [AOTA], 2019). Research supports the clinical effectiveness of occupational therapy in these settings to improve functional outcomes (e.g., Clemson et al., 2016; Lockwood et al., 2019; Nilsen et al., 2014; Uruma et al., 2019), and evidence for occupational therapy’s economic value is emerging (e.g., Sampson et al., 2014). However, the achievement of superior client outcomes (such as enhanced functional outcomes) often requires additional resources (such as greater amounts of practitioner time or increased length of hospital stay). It is uncertain whether the additional costs associated with occupational therapy services are offset by future savings or additional benefits.
Two recent studies with participants after hip replacement in acute and subacute care determined the benefit to participants’ function when occupational therapy services were included. First, a randomized controlled trial (RCT) found that people who received a predischarge occupational therapy home visit after hip replacement had fewer falls and lower hospital readmissions than those who did not receive such a visit (Lockwood et al., 2019). Clients who received a predischarge home visit also had higher functional ability as assessed by the FIM® (Lockwood et al., 2019 ; Uniform Data System for Medical Rehabilitation, 2009; FIM® is a trademark of the Uniform Data System for Medical Rehabilitation, a division of UB Foundation Activities, Inc.).
Second, a retrospective observational audit also found that people in the acute rehabilitation phase after hip replacement had increased functional ability when occupational therapy services were provided (Uruma et al., 2019). However, both studies identified increased resource use, in health professional time or in bed days, so the cost of improved outcomes may not be offset by the additional benefits or cost savings in terms of lower hospital readmission rates or decreased use of social services after discharge. Without information on effectiveness, the economic value of occupational therapy after hip replacement in acute and subacute care is uncertain.
Economic evaluations provide a means to compare the cost and effectiveness of health care services (Drummond et al., 2015). Even though occupational therapy is an integral component of health care, no synthesis of cost-effectiveness research has been performed. Previous systematic reviews have evaluated the cost effectiveness of occupational therapy with older adults (Nagayama et al., 2016) and with clients with cognitive or age-related functional decline (Rahja et al., 2018).
A review of the quality of published economic evaluations in occupational therapy found that although the number of economic evaluations had increased, there were fewer evaluations than in other disciplines and the quality was suboptimal, with only 3 of the 9 studies deemed high quality (Green & Lambert, 2017). This review recommended that future research incorporate economic evaluations of higher methodological quality to produce more robust results regarding the value of occupational therapy services (Green & Lambert, 2017).
An investigation of not only the clinical benefit of occupational therapy but also its value for money is needed. Therefore, the objective of this systematic review was to identify, describe, critique, and synthesize published economic evaluations of occupational therapy services provided to adults in acute and subacute care.
Method
Search Strategy
A comprehensive search strategy was designed in MEDLINE (see supplemental data, available online with this article at https://research.aota.org/ajot) and then translated into the following databases: CINAHL, CENTRAL, EconLit, Embase, National Health Services Economic Evaluation Database, PsycINFO, ProQuest (Health and Medicine and Social Science subsets only), and OTseeker. Gray literature searches included the websites of occupational therapy associations in the United States, Canada, Australia, United Kingdom, and New Zealand; government research bodies; and clinical trial registries. Manual searches of reference lists from included studies were also conducted. Searches were completed up to March 2021. The search was conducted by a health services librarian in conjunction with Kylie Wales.
For this study, the primary outcome examined was the value for money of occupational therapy services in acute and subacute care. Studies were included if they had an adult population (≥80% participants ages ≥18 yr) and assessments or interventions were administered by an occupational therapist or a multidisciplinary team that included an occupational therapist in acute or subacute care. We included interventions (e.g., discharge planning in the acute care setting) focused on activity and participation that often include occupational therapists even if they did not specifically mention occupational therapists. When the intervention had multiple components, the occupational therapy component was required to make up 50% of the program. In this review, occupational therapy was defined as assessment or intervention that targets the activity or participation domains of the International Classification of Functioning, Disability and Health (World Health Organization [WHO], 2002). Activity was defined as “execution of a task by an individual” and participation as “involvement in a life situation” (WHO, 2002, p. 2).
To be included, studies had to have at least one alternative intervention or comparator (such as standard care) for comparison. Randomized and nonrandomized studies (e.g., multiple-baseline cohort studies and pre–post studies) with embedded economic analyses (comparing cost and outcomes between interventions) and modeled economic analyses were included. Systematic reviews, study protocols, conference proceedings, editorials, and commentary papers were excluded, as were efficacy or effectiveness analyses with no analysis of costs, burden-of-disease studies, and cost-of-illness studies. Studies were also excluded if they were not published in English. There was no restriction on the country in which the study was conducted. We recognize that there are challenges in comparing economic evaluations across countries because variation in health care systems (including structure and payment systems) can influence the overall effectiveness and cost of a program; the intent of our systematic review is to synthesize the available information to identify the structures and environments that are conducive to cost-effective interventions (Mazzei et al., 2020).
Two independent reviewers screened studies by title and abstract (Kylie Wales and Miia Rahja), with disagreements resolved by a third reviewer (Natasha A. Lannin or Danielle Lang); when they could not reach a consensus, the study proceeded to full-text review. The full texts were assessed by two independent authors (Wales or Rahja), and disagreements were resolved by a third reviewer (Lannin or Lang).
Data Extraction and Appraisal
Data were extracted using a standardized form adapted from the Joanna Briggs Institute Data Extraction Form for Economic Evaluations (Joanna Briggs Institute, 2014) and entered into a Microsoft Excel spreadsheet. Information included study design, setting (acute or subacute), geographical location of the study, participant characteristics, description of intervention and comparator, source of effectiveness data, measurement and valuation of costs and outcomes, time horizon, analytical approach and perspective, economic results, sensitivity analysis, and study authors’ summary of findings. The primary reviewer (Wales) extracted data independently, and a second reviewer (Kate Laver) checked the extracted data; any discrepancies were resolved through discussion or consultation with a third reviewer (Lang).
Methodological quality of economic evaluations was assessed using the Evers Consensus on Health Economic Criteria (CHEC) list (Evers et al., 2005).
Cost Conversion
To allow for comparisons across studies, all costs were converted to 2020 U.S. dollars (USD) using the Campbell and Cochrane Economics Methods Group–Evidence for Policy and Practice Information and Coordinating Centre (2019) Cost Converter (Version 1.6). When an intervention was less expensive and more effective than its comparator, it is referred to as dominant (Drummond et al., 2015).
Results
The search resulted in 17,863 citations. After removal of duplicates, 13,176 articles were screened. A total of 190 articles were assessed in full text, and 10 articles were included in the systematic review (Figure 1).

Flow of articles through the systematic review process.
Overview of Included Studies
Most of the included studies (n = 8) described an economic evaluation embedded in an RCT (Haines et al., 2013; Mortimer et al., 2019; Nagayama et al., 2017; Rodgers et al., 2003; Sampson et al., 2014; Sigurdsson et al., 2008; Wales et al., 2018, Yoshida et al., 2019), and for most studies (n = 8) the comparator was described as usual care (Brandis, 1998; Haines et al., 2013; Ho et al., 2019; Mortimer et al., 2019; Nagayama et al., 2017; Rodgers et al., 2003; Sigurdsson et al., 2008, Yoshida et al., 2019). Five studies used cost-effectiveness analyses (Haines et al., 2013; Ho et al., 2019; Mortimer et al., 2019; Sigurdsson et al., 2008; Wales et al., 2018), 4 used cost–utility analyses (Nagayama et al., 2017; Sampson et al., 2014; Wales et al., 2018, Yoshida et al., 2019), 1 used a cost–benefit analysis (Brandis, 1998), and 1 was a cost-minimization analysis (Rodgers et al., 2003; Wales et al. [2018] conducted two analyses). Table A.1 in the Supplemental Appendix provides information on the studies included in the review as well as data on the clinical effectiveness of the interventions.
Study Population
Four studies included a population of stroke survivors undergoing rehabilitation, but each focused on different aspects of occupational therapy services (Ho et al., 2019; Nagayama et al., 2017; Rodgers et al., 2003; Sampson et al., 2014). The remaining interventions were (1) falls prevention activities for older adults admitted to acute or subacute wards (Haines et al., 2013), (2) people awaiting hip replacement receiving additional pre- and postoperative occupational therapy and physiotherapy (Sigurdsson et al., 2008), (3) people admitted to the hospital poststroke or classified as not coping who received a predischarge home visit (Brandis, 1998), (4) older adults admitted to the hospital who received pre- and postdischarge home visits from occupational therapists (Wales et al., 2018), (5) activities of daily living (ADL) retraining after traumatic brain injury (TBI; Mortimer et al., 2019), and (6) a mixed population of people with cerebral, spinal, or muscular disease undergoing inpatient rehabilitation (Yoshida et al., 2019).
Cost Data
Six studies either stated or were assessed as having evaluated costs from a health services perspective (Brandis, 1998; Haines et al., 2013; Ho et al., 2019; Mortimer et al., 2019; Sampson et al., 2014, Yoshida et al., 2019), 3 from a societal perspective (Nagayama et al., 2017; Sigurdsson et al., 2008; Wales et al., 2018), and 1 from a government perspective (Rodgers et al., 2003). The types of costs included in the studies varied depending on perspective and type of intervention. Nine studies included the cost of occupational therapist services, but cost inclusions varied across studies (Brandis, 1998; Haines et al., 2013; Ho et al., 2019; Mortimer et al., 2019; Nagayama et al., 2017; Rodgers et al., 2003; Sampson et al., 2014; Sigurdsson et al., 2008; Wales et al., 2018). Nine studies also included costs related to hospitalization (Brandis, 1998; Haines et al., 2013; Ho et al., 2019; Mortimer et al., 2019; Nagayama et al., 2017; Rodgers et al., 2003; Sigurdsson et al., 2008; Wales et al., 2018, Yoshida et al., 2019); 4 included travel costs (Brandis, 1998; Sampson et al., 2014; Sigurdsson et al., 2008; Wales et al., 2018); and 2 included the costs of social support services (Rodgers et al., 2003; Wales et al., 2018). Only 1 study (Sigurdsson et al., 2008) included the costs associated with productivity losses. Further information on costs included for each study can be found in Supplemental Table A.1.
Outcomes
Economic outcomes also varied across the studies; 4 studies included quality-adjusted life years (QALYs; Nagayama et al., 2017; Sampson et al., 2014; Wales et al., 2018, Yoshida et al., 2019), a commonly used approach for estimating survival and QoL benefits in economic evaluations (Drummond et al., 2015). QALYs are derived from health-related QoL measures such as the EuroQoL 5D (EQ-5D; EuroQoL Research Foundation, 2021); they are a measure of health outcomes that incorporates changes in survival and in QoL (0 = death, 1 = full health; Drummond et al., 2015). Nagayama et al. (2017) and Wales et al. (2018) derived QALYs using the six-dimensional health state short form; Sampson et al. (2014) used the three-level version of the EQ-5D; and Yoshida et al. (2019) used the five-level version of the EQ-5D. Haines et al. (2013) measured the incremental number of people without falls as the outcome of their cost-effectiveness analysis. The outcome assessed by Brandis (1998) was bed days avoided, which was valued in monetary terms on the basis of cost per bed day. The remaining 4 studies (Ho et al., 2019; Mortimer et al., 2019; Rodgers et al., 2003; Sigurdsson et al., 2008), as well as Wales et al. (2018), measured outcomes using functional assessment tools.
Risk of Bias
Supplemental Appendix Table A.2 describes the methodological quality of included studies as assessed with the Evers CHEC-list. No study fulfilled all criteria for methodological quality for economic analyses, with scores ranging from 9 (Ho et al., 2019) to 18 (Mortimer et al., 2019; Nagayama et al., 2017) out of a possible 19. The study with the lowest methodological quality was a cost-effectiveness analysis that did not describe relevant methods for the analysis, including clearly stating the perspective adopted for the economic analysis (Ho et al., 2019). One consistent shortcoming of the included studies was the short time horizon selected for reviewing economic outcomes. The time horizons often did not include the long-term outcomes, and they may have been too short to capture all costs and benefits of occupational therapy services after the hospital admission, such as less use of health care services after discharge or of formal or informal caregiving.
Levels of Evidence
Levels of evidence were assessed as per the American Journal of Occupational Therapy’s guidelines for systematic reviews (AOTA, 2020), which adopt the Oxford Levels of Evidence (OCEBM Levels of Evidence Working Group, 2009).
Narrative Synthesis of Economic Analysis
Because of the heterogeneity of the results, we were unable to draw conclusions as to the economic value of occupational therapy services. Supplemental Table A.1 describes the cost-effectiveness outcomes for each study (including costs transformed to a standard year and currency [2020 USD]).
Occupational Therapy in Acute and Subacute Care
Acute Care
One Level 3b study (Brandis, 1998) was exclusively based in acute care. This study sought to evaluate the value of discharge planning and services with an occupational therapy home visit. Although it included no detail on the number of occupational therapy sessions provided, the total cost of the first year of the discharge home visit program was USD18,286, with a cost per client of USD147. Brandis (1998) estimated that 1,092 bed days were saved (including 56 readmissions) and estimated that the reduction in bed days and readmissions saved approximately USD519,780. Clinical effectiveness was measured in client and clinician satisfaction with the program, which was reported to be positive. Details missing from this study included costing data (including equipment and home modifications costs) and the occupational therapy recommendations provided to inpatients.
Acute and Subacute Care
The remaining 9 studies had significant heterogeneity and were grouped as acute and subacute care for older people.
Focus on Older Adults
Two Level 1b studies provided services across acute and subacute care to older people. Haines et al. (2013) provided falls education via multimedia and health professional follow-up, and Wales et al. (2018) investigated discharge planning (with pre- and postdischarge home visits).
The outcome for the economic evaluation conducted by Haines et al. (2013) was the number of falls (including number of people without falls, number of injurious falls, and number of people with one or more injurious falls). The results of this study (which used data from an RCT) showed that the program’s cost-effectiveness compared with usual care was USD451 per number of people with falls prevented and USD252 per fall prevented, respectively. The study demonstrated that cognitively intact clients had a lower rate of falls and were less likely to experience a fall. This study was limited to a health service perspective, and costs related to community health follow-up, equipment, or modifications were not included.
Wales et al. (2018) found no differences in QALYs between the intervention and comparator intervention in a discharge planning program. The data were explored in terms of incremental cost per additional client with functional improvement (USD52,718) and incremental cost per additional client with improvement in functional ability (approximately USD46,000). Given that the decision makers’ willingness to outlay such costs in terms of functional improvement was unknown and no difference in QALYs was found, Wales et al. were unable to deem the intervention cost effective.
Focus on People Poststroke
Four studies focused on occupational therapy services for people poststroke. In 3 of these studies, therapy was provided at least 5 times a week: Ho et al. (2019; Level 3b) provided 40 min of conventional therapy and 20 min of virtual reality therapy every day for 1 wk, Nagayama et al. (2017; Level 2b) provided therapy focused on activity acquisition for at least 40 min per session, and Rodgers et al. (2003; Level 1b) delivered interdisciplinary upper limb therapy.
In the Nagayama et al. (2017) study, the occupational therapist and client used the Aid for Decision making in Occupation Choice (ADOC) to identify suitable meaningful activities that could be integrated into therapy. Most of the therapy provided to clients in stroke rehabilitation using the ADOC adopted occupation-focused interventions, which were expected to enhance participants’ QoL. The average length of stay was 129.3 days for participants using ADOC and 115.5 days for clients receiving usual care. Nagayama et al. used a cost–utility analysis to evaluate the ADOC intervention but did not present the incremental cost per QALY gained for the ADOC intervention versus usual care. However, the intervention was less expensive and more effective than usual care (i.e., it was dominant). The average cost per QALY gained was JPY3.544 million (USD 40,078; intervention) versus JPY3.913 million (USD 43,036; comparator). The probability that the intervention was cost-effective was estimated to be 65.3%, using a willingness-to-pay threshold of JPY5 million per QALY gained (USD54,992; Nagayama et al., 2017).
Ho et al. (2019) compared the differences between conventional treatment and conventional treatment plus 20 min of virtual reality rehabilitation each day for a week. The addition of virtual rehabilitation reduced medical costs and resulted in improved outcomes (including symptom severity and functional outcomes) and so was considered a dominant intervention (USD3,582 vs. USD4,800, respectively). Rodgers et al. (2003) compared interdisciplinary (occupational therapy and physiotherapy) upper limb intervention with a stroke care service. The upper limb intervention was based on neurodevelopmental treatment principles as well as engagement in meaningful activities. They found no difference in outcomes between interdisciplinary upper limb services and stroke care service, and a cost-minimization analysis found that the intervention was costly as the stroke care service (USD11,811 and USD11,514, respectively).
One study provided a predischarge home visit only to participants in subacute rehabilitation (Sampson et al., 2014). A cost–utility analysis found that the cost per QALY gained of home visits versus usual care was USD36,810. The authors reported that the likelihood of the intervention being cost-effective was 47% when a willingness to pay was set at GBP20,000–GBP30,000 per QALY gained (USD33,483–USD50,225).
Subacute Care
One Level 2b study (Sigurdsson et al., 2008) investigated the effect of preoperative and postoperative occupational therapy and physiotherapy for clients after hip replacement. Clients received preoperative training and information on required postoperative exercises. A home visit was conducted at discharge if required, and the client was then visited a median of 4 times to monitor adherence to the rehabilitation program. The study used a cost-effectiveness analysis approach; the Oxford Hip Score was used to evaluate pain and functional impairment before and after the intervention. Costs in this study were driven by rehabilitation stay rather than operation costs. An incremental cost-effectiveness ratio was not presented, but the intervention was found to be less expensive and more effective than usual care (i.e., it was dominant).
Another study (Mortimer et al., 2019), also Level 1b evidence, was based in an inpatient rehabilitation center where clients with TBI received structured ADL training during posttraumatic amnesia (PTA). The treatment was shown to be dominant over the comparator condition, demonstrating statistically significant improvements in functional independence and cost savings with varying sensitivity analysis (USD42,863 vs. USD49,278, respectively).
The Yoshida et al. (2019; Level 1b) study was based in inpatient rehabilitation; clients with cerebral, spinal, or musculoskeletal disease were provided rehabilitation services that adjusted the challenge and skill balance to ensure activities met the clients’ needs. The study found a statistically significant difference in QoL (measured with the EQ-5D-5L) for the intervention group versus the comparator group. Cost per QALY gained of the adjusting challenge–skill balance versus the comparator (usual care) was USD5,733.80.
Cost Effectiveness
Figure 2 depicts the cost-effectiveness plane for the included studies. Interventions that were found to be dominant, that is, whose costs were lower and whose outcomes were better than the comparator, were structured ADL retraining during PTA for people with TBI (Mortimer et al., 2019), occupation-focused interventions for people poststroke (Nagayama et al., 2017), and occupational therapy and physiotherapy interventions pre– and post–hip replacement (Sigurdsson et al., 2008).

Cost-effectiveness plane of included studies.
The other 2 studies identified as dominant (Brandis, 1998 ; Ho et al., 2019) should be interpreted with caution because they were assessed as being of lower quality, and their economic analyses did not explore uncertainty. The 4 studies with higher benefits and higher costs had significant uncertainty, making interpretations difficult—small sample size in Sampson et al. (2014), outliers in Haines et al. (2013), no exploration of uncertainty in Yoshida et al. (2019), and limited information on likelihood of willingness to pay for improvement in functional outcome in Wales et al. (2018). One study, Rodgers et al. (2003), was identified as not providing value for money, with no difference found in clinical outcomes and a higher cost associated with the intervention therapy.
Discussion
We identified 10 studies that included economic evaluations of occupational therapy services and found significant heterogeneity in both cost-effectiveness methodology and occupational therapy intervention. This heterogeneity made it difficult to draw definitive conclusions about the cost-effectiveness of occupational therapy services in acute and subacute care, and thus further research is required. However, five interventions included in these studies were found to be more effective and cheaper than usual care: delivery of structured ADL retraining during PTA for people with TBI; occupation-based interventions poststroke; virtual rehabilitation poststroke; acute discharge planning; adjusting the challenge–skill balance in rehabilitation for people with musculoskeletal, spinal, or cerebral disease; and combined occupational and physiotherapy pre- and postoperative interventions for people undergoing hip replacement (Brandis, 1998; Ho et al., 2019; Mortimer et al., 2019; Nagayama et al., 2017; Sigurdsson et al., 2008).
In the absence of information on willingness to pay for a unit of health gain, determining value for money when an intervention is more costly and more effective than its comparator is difficult. Another challenge for the profession lies in how one assesses value for money when different outcomes are measured. For example, should the benefit of falls prevention programs be reported as cost per fall prevented, cost per client avoiding a fall, cost per improvement on a functional scale, or QALY? Currently, little professional discussion regarding these issues has occurred in occupational therapy.
By evaluating the methodological quality of existing studies, this review highlights common methodological limitations in the current health economic evaluations of occupational therapy in acute and subacute settings. The most common limitation was that relevant costs for each alternative intervention were not always considered. Occupational therapy is a client-centered approach that can have wide-ranging effects on a person’s life, and it is no surprise that identifying relevant costs was difficult. Nonetheless, if the clinical benefit of occupational therapy interventions often relates to return to meaningful activity, studies of cost-effectiveness should capture costs from across the breadth of a person’s activity and participation.
Strengths and Limitations
Although the included studies had methodological limitations, the findings of this systematic review suggest that future research has a good foundation to build on. Most studies incorporated appropriate outcome measures to detect the effects of occupational therapy intervention. Strengths of this review include the comprehensive search strategy, which was developed in collaboration with an experienced health services librarian and was tested in several databases to ensure relevance. However, as with all systematic reviews, it is plausible that literature may have been missed.
A limitation of this review is that we were able to identify only a small number of eligible studies, which limited our ability to draw conclusions. This limitation is similar to those of other systematic reviews of economic evaluations in occupational therapy (e.g., Nagayama et al., 2016; Rahja et al., 2018). Another limitation is that we did not include studies published in languages other than English. Potentially relevant articles were thus not included; however, because of the lack of resources for translation, including these articles was not feasible.
Implications for Occupational Therapy Practice and Research
The findings of this systematic review of economic evaluations has the following implications for occupational therapy practice and research: Occupational therapy intervention for adults in acute and subacute care can be cost-effective, although further research is required to substantiate these results. Occupational therapists should continue to provide services in acute and subacute care and carefully document their outcomes and costs so that an evidence base can be built. The paucity of studies presenting economic analyses of occupational therapy in acute and subacute care highlights the need for future trials of clinical benefit to include an economic evaluation in their study design. Without such studies, conclusions regarding the economic value of occupational therapy cannot be made, despite its published clinical benefit. In addition, modeled economic analyses that capture costs and outcomes beyond the short follow-up of research trials are needed to ensure that all costs and benefits associated with occupational therapy services are included in economic evaluations.
Conclusion
Preliminary evidence provides support for the economic value of occupational therapy services in TBI rehabilitation, discharge planning, subacute stroke rehabilitation, and rehabilitation before and after hip replacement. Without further research to substantiate these findings, occupational therapy is at risk of being devalued in acute and subacute care. Research is urgently needed that assesses the economic value of occupational therapy, and similar economic perspectives and outcomes must be used so comparisons can be made between studies.
Supplemental Materials
Supplementary material for Economic Effects of Occupational Therapy Services for Adults in Acute and Subacute Care Settings: A Systematic Review
Supplementary material, sj-docx-1-aot-10.5014_ajot.2022.049078.docx for Economic Effects of Occupational Therapy Services for Adults in Acute and Subacute Care Settings: A Systematic Review by Kylie Wales, Danielle Lang, Miia Rahja, Lisa Somerville, Kate Laver and Natasha A. Lannin in The American Journal of Occupational Therapy
Supplementary material for Economic Effects of Occupational Therapy Services for Adults in Acute and Subacute Care Settings: A Systematic Review
Supplementary material, sj-docx-2-aot-10.5014_ajot.2022.049078.docx for Economic Effects of Occupational Therapy Services for Adults in Acute and Subacute Care Settings: A Systematic Review by Kylie Wales, Danielle Lang, Miia Rahja, Lisa Somerville, Kate Laver and Natasha A. Lannin in The American Journal of Occupational Therapy
Footnotes
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
