Use of the AusTOMs-OT to record outcomes in an occupational therapy homeless service

Abstract

Introduction

Occupational therapists provide services to homeless people internationally, yet relatively little research evidence underpins this practice. This study aimed to describe the process of selecting an outcome measure for use in a homeless occupational therapy service and demonstrate the changes experienced in the occupational performance of individuals participating in this service.

Method

The Glasgow homeless occupational therapy service staff reviewed eight outcome measures against six pre-determined criteria. The selected tool was then used to collect data pre and post an occupational therapy intervention with 58 participants. Data were analysed using descriptive statistics and Wilcoxon Signed Ranks Test to determine statistically significant change from admission to discharge in the different domains of the selected outcome measure.

Results

The AusTOMs-OT was implemented in daily practice and 70% of participants were discharged following achievement of therapy goals. Clients made clinically and statistically significant improvements over their admission as measured on AusTOMs-OT in areas such as Transfers, Self-Care and Domestic Life-Home, across the health domains of activity, participation and wellbeing.

Conclusion

AusTOMs-OT was successfully implemented and demonstrated that participants improved over their time with the service. This evidence can be used to support the effectiveness of occupational therapy services with people experiencing homelessness.

Introduction

Internationally, homelessness is recognised as a societal issue (Boland and Cunningham, 2019; Thomas et al., 2011). With diverse causes and presentations, homelessness significantly impacts on a person’s ability to participate in meaningful occupation. This study examined the process an occupational therapy homeless service used to select a suitable outcome measure for use with clients receiving occupational therapy interventions, in order to track change over time. Data were then used to describe the impact of this service on client’s occupational performance.

Literature review

The number of people experiencing homelessness is increasing around the world. This includes in Scotland, where approximately 30,000 people were recognised as homeless in 2019 (Scottish Government, 2019). However, with no internationally agreed upon definition, homelessness statistics are estimates only. In Scotland, people are defined as homeless if they have no accommodation or if their accommodation cannot be accessed due to threat of violence (Scottish Government, 2019). Causes of homelessness are varied and can range from substance abuse, mental illness, domestic violence or economic reasons (Schultz-Krohn and Tyminski, 2018). Other authors relate the cause of homelessness to a mixture of structural (housing supply, unemployment and welfare priorities) and individual factors (Boland and Cunningham, 2019). Because of this, individuals experiencing homelessness are not homogenous and the needs of this group are often diverse (Cunningham and Slade, 2019; Thomas et al., 2011). While mental health issues are frequently recognised as being associated with homelessness, mortality and comorbidity through physical health issues are similarly common (Marshall et al., 2020c; Lloyd and Bassett, 2012; Queen et al., 2017). Higher rates of morbidity are seen when this group is compared to the general population with a range of health conditions present (Lloyd and Bassett, 2012; Queen et al., 2017).

Occupational therapists, with the goal of enabling participation in meaningful occupation, are well placed to work in the complex field of homelessness. Theoretical foundations highlighting the interplay between the person, environment and occupation (Townsend and Polatajko, 2007), person centred practice and the concepts of occupational justice and occupational deprivation (Nilsson and Townsend, 2010; Townsend and Wilcock, 2004) can guide occupational therapists in their work with homeless individuals. A significant body of literature outlines the contribution occupational therapy is currently making in this growing area of occupational therapy practice.

Occupational therapists work in homelessness across different practice settings from emergency departments to acute inpatient stay through to community assertive outreach programs and dedicated homeless services (Grandisson et al., 2009; Lloyd et al., 2017; Lloyd and Bassett, 2012). Occupational therapy interventions in these settings range from linking homeless clients with services in their community, providing education regarding physical comorbidities, equipment prescription, treatment for physical injury, support for substance use through to case management for ongoing mental health issues and achieving housing or employment (Grandisson et al., 2009; Lloyd and Bassett, 2012; Lloyd et al., 2017). A scoping review of occupational therapy interventions with homeless individuals identified the following as being commonly used: occupational performance skills training, enrichment of occupational repertoire, employment and education-related interventions, provision of physical rehabilitation services, child and family occupational therapy services and community building (Roy et al., 2017).

To better understand the occupational experience of homeless individuals, and enhance occupational therapy practice, occupational therapy researchers have also explored the lived experience (Cunningham and Slade, 2019; Illman et al., 2013; Marshall et al., 2017, 2020c). Important themes of restrictions on time use, activity as a means of belonging and developing a sense of identity through engagement in activity were identified. Included in this exploration of homelessness is emerging evidence linking the pervasive nature of boredom commonly experience by homeless individuals with an impact on mental wellbeing, community integration and substance use in this population (Marshall et al., 2019a, 2019b).

Occupational therapists are also involved in the development of life skills and assisting individuals to transition from homelessness to housing (Boland, 2018; Gutman and Raphael-Greenfield, 2017; Marshall et al., 2020a). A systematic review determined that most of the research in this field describes either individual or group based life skill interventions aimed at sustaining accommodation, with training in how to apply for housing and then how to maintain tenancy (money management, cleaning, shopping and laundry) (Marshall et al., 2020a).

Occupational therapy practice with individuals experiencing homelessness is varied and growing. Demonstrating the effectiveness of these interventions is important for occupational therapists. Evidence-based practice depends on the routine collection of outcome measure data to evaluate interventions (Unsworth, 2017). Examples of outcome measures used to evaluate the effectiveness of homelessness services include the Ansell-Casey Life Skills Assessment (Boland, 2018; Marshall et al., 2020a; Thomas et al., 2011), Occupational Self-Assessment (Marshall et al., 2020a) and the Model of Human Occupation Screening Tool (Parmenter et al., 2013). These outcome measures have been used in effectiveness studies targeting a particular intervention. However, it is clear that occupational therapy interventions for homelessness are based across all domains of health, that is a person’s impairments, activity limitations, participation restrictions and their level of distress/wellbeing (World Health Organization, 2001). An outcome measure that can accommodate all health domains is required for routine collection of data as a part of daily occupational therapy practice for homelessness. At present, the use of this type of outcome measure has not been documented in the literature. Other important considerations when choosing an outcome measure include evidence of psychometric properties (Jerosch-Herold, 2005) and the clinical utility of the measure such as cost, time to administer, portability across different settings, training required and availability (Fawcett, 2013).

The aims of this study were to (i) describe the process of selecting an outcome measure for use in a homeless occupational therapy service that included information about the client’s physical status and occupational performance and (ii) to demonstrate the changes experienced in the occupational performance of individuals participating in this service using the selected outcome measure data, as part of a service evaluation.

Method

Setting

The National Health Service Greater Glasgow and Clyde is the largest Health Board in Scotland, serving a population of 1.14 million people. The Homeless Occupational Therapy Service (HOTS) is based in Glasgow at Hunter Street Homeless & Asylum service. This is a dedicated community service which focuses on working with homeless individuals with complex needs across the Greater Glasgow area. The HOTS provide interventions to individuals experiencing mental health issues, substance addictions and physical impairments who are living in homeless accommodation, typically emergency accommodation and temporary furnished flats. The HOTS accept individuals aged 16 and above who are in emergency, bed and breakfast, roofless (sleeping outside) or care of address accommodation (moving from one accommodation to another). Following assessment, the occupational therapy role is to provide appropriate interventions to ensure individuals are resettled in appropriate accommodation with adequate support.

Selection of an outcome measure

Although, like many other homeless services, the HOTS was successfully using the Model of Human Occupation Screening Tool (MOHOST) (Kramer et al., 2009; Parmenter et al. 2013) to record mental health outcomes, the team had identified an increasing number of clients being referred to the service with physical impairments. The MOHOST environment section was found to be adequate to record and monitor the functional impact (activity limitations) of these physical impairments, within the context of people being homeless. However, it does not address changes in impairment and participation also considered important with this client group. To facilitate clinical improvement and provide evidence of change in clients with physical impairments, an outcome measure that considered all health domains was needed. A critiquing process was used (Clarke et al., 2001) to identify a measure. A working group of occupational therapists from the HOTS team formed and met on several occasions to explore outcome measures. The working group compiled the following criteria for the requirements of an outcome measure:

• • Sensitive to measure small changes in function/status over time;

• • Able to demonstrate change in, or maintenance of function, specifically physical, abilities in relation to occupational performance;

• • Relatively easy and fast to administer;

• • Established psychometric properties, with ability to aggregate data;

• • Low purchase cost;

• • Suitable for use with clients aged 16 and above.

As it was beyond the purpose and resources available for any form of systematic or scoping review to be undertaken, the HOTS therapists conducted an informal and unstructured review of the literature from 2010 on PubMed and Cumulative Index to Nursing and Allied health Literature (CINAHL) and reviewed measurement texts such as those by Asher (2014), Hempill-Pearson and Urish (2019) and Bortnick (2017) using the key terms ‘homeless’, ‘outcome measure’ and ‘function’ or ‘occupational therapy’ to determine what measures were commonly used, or could be used, in the area of homelessness. Although the review may not have met strict systematic review guidelines, it did generate several outcome measures that were then evaluated against the set criteria.

Service evaluation

Population: Clients were included in the service evaluation if they were experiencing a physical impairment causing functional deficits, in addition to any mental health, addiction or cognitive issues. Clients were excluded if there was no physical impairment present.

Intervention: The HOTS team identified the supports clients’ need to maintain a tenancy and break the cycle of repeated homelessness. This could involve interventions which maximise the individual’s safety and independence through development of personal and domestic activities of daily living, equipment provision, promoting social skills and integration and improving coping mechanisms.

Outcome Measure: Staff met over several months to review potential outcome measures against the set criteria. The Australian Occupational Therapy Outcome Measure-Occupational Therapy (AusTOMs-OT) (Unsworth and Duncombe, 2014) was identified as the best tool to meet the criteria to evaluate service outcomes. A brief overview of the AusTOMs-OT is therefore provided here, and some further details described in the Results and Discussion. The AusTOMs-OT is based on the International Classification of Functioning, Disability and Health (ICF) (World Health Organization, 2001) and is designed for international use (downloaded from www.austoms.com). Occupational therapists establish collaborative goals for therapy with their clients and then choose from 12 scales representing core areas of function related to these goals. The 12 scales include (1) learning & applying knowledge, (2) functional walking & mobility, (3) upper limb use, (4) carrying out daily life tasks & routines, (5) transfers, (6) using transport, (7) self-care, (8) domestic life – home, (9) domestic life – managing resources, (10) interpersonal interactions and relationships, (11) work/employment/education and (12) community life/recreation/leisure/play. Each scale is rated across the four health domains of impairment, activity limitation, participation and distress/wellbeing on an ordinal scale (zero to five, with half points allowed, providing a total of 11 levels with higher scores indicate improvement).

Scores on the impairment and activity limitation domains are unique for each scale. However, as participation and distress/wellbeing are global constructs, these are scored once, across all scales at admission and again at discharge. Studies have demonstrated the validity (Abu-Awad et al., 2014; Chen and Eng, 2015; Unsworth, 2005; Unsworth et al., 2004) and reliability (Fristedt et al., 2013; Morris et al., 2005; Scott et al., 2006; Unsworth et al., 2018) of the AusTOMs-OT. In addition, Unsworth et al. (2015) investigated the minimal clinically important difference (MCID), indicating a clinically significant change, for each AusTOMs-OT scale as being a one-point shift.

Procedure

Once AusTOMs-OT was identified as the measure for the service evaluation, a working group designed a two-month pilot project to incorporate administration of the tool into existing caseloads. Initial results showed AusTOMs-OT could demonstrate change from initial assessment to discharge. The AusTOMs-OT was then implemented across the larger HOTS team by conducting in services on how to administer the measure which included cases studies and scoring practice as a group to improve consistency and reliability. For the service evaluation, the data were collected with individuals as part of the usual service of the HOTS team. At the time the data were collected, there was no plan to analyse the data or disseminate the results, therefore specific consent for use of the data was not sought. A Caldicott application was approved by the Data Collection Officer, Information Governance Department, National Health Service (NHS) Greater Glasgow & Clyde when a decision was later made to evaluate this routinely collected data, and ethical approval was sought and gained from Central Queensland University, #22823. The data analysed for this service evaluation were collected over a 13-month period between February 2019 and March 2020 from clients using the Homeless Occupational Therapy Service. Compliance with completing the AusTOMs-OT was monitored and supported using audits, AusTOMs-OT champions and availability of experienced staff to answer questions and provide clarification. The AusTOMs-OT was incorporated into documentation and discharge information to General Practitioners to demonstrate change in individuals receiving the occupational therapy service.

Data analysis

The HOTS team worked together to review possible outcome measures and develop a grid of features to identify an outcome measure to test in practice. To meet the second aim of evaluating the interventions used in the service using the identified outcome measure, data collected with clients were analysed using pre-post techniques. A power analysis was not conducted but a minimum sample size was determined using previous research as a guide (Unsworth et al., 2009). All data were analysed using IMB- SPSS (version 26) software using techniques described in Pallant (2016). Data were initially reviewed for missing data and normality. Descriptive statistics were performed for participants at admission and discharge. As a normal distribution of data was not expected, non-parametric testing was planned using a Wilcoxon Signed Ranks Test to determine statistically significant change from admission to discharge in the different domains of the selected outcome measure, the AusTOMs-OT. A p-value of .05 or less was considered statistically significant and an effect size was calculated representing the magnitude of difference between admission and discharge scores. For the Wilcoxon Signed Ranks Test, an absolute Z-score >1.96 was considered statistically significant at the .05 level. The larger the absolute Z-score, the smaller the probability the results were due to chance. Effect sizes were calculated and interpreted according to Cohen (1988) where an effect size of 0.20 or less represents a small change; 0.50 represents moderate change; and 0.80 represents a large change. The minimum clinically important difference (MCID) was used to determine clinical significance. For the AusTOMs-OT, the MCID has previously been established as a change of one point (Unsworth et al., 2015). The number and percentage of participants who improved, and who achieved or were above this clinically important difference was then calculated.

Results

Identification of an outcome measure for the service

Based on discussions among the occupational therapists at HOTS, and drawing on the summary data developed in Table 1, the Australian Occupational Therapy Outcome Measure-Occupational Therapy (AusTOMs-OT) (Unsworth, 2014) was identified as the best measure to implement across the service. The AusTOMs-OT was identified as a good fit for the client group due to the focus of scales on client function (including physical function) and its sensitivity to measure small changes in these areas (Unsworth, 2005; Unsworth et al., 2015).

Table 1.

Review of outcome measures against criteria developed for use in a homeless service, with particular focus on including client physical function.

Outcome measure	Sensitive	Function change	Ease	Psychometrics	Cost	Age
Ansell-Casey Life Skills Assessment (Marshall et al., 2020a; Boland, 2018; Thomas et al., 2011)	0	0	0	X	X	X
Australian Therapy Outcome Measures- Occupational Therapy (AusTOMs-OT) (Unsworth, 2014).	X	X	X	X	X	X
Canadian Occupational Performance Measure (Grajo et al., 2020; Marshall et al., 2020c)	X	X	X	0	X	X
Goal Attainment Scale (Gutman and Raphael-Greenfield, 2017; Chapleau et al., 2012)	X	X	0	0	X	X
Manchester Short Assessment of Quality of Life Scale (Boland, 2018; Gutman and Raphael-Greenfield, 2017)	X	0	X	X	X	X
Model of Human Occupation Screening Tool (Parmenter et al., 2013)	0	0	X	X	X	X
Occupational Self-assessment (Marshall et al., 2020a)	0	0	X	X	X	X
Therapy Outcome Measures (TOM) (Enderby et al., 2013)	X	0	X	0	X	X

Key to rating: X = feature present, 0 = feature absent. Key to columns: Sensitive: Sensitive to measure small changes in function/status over time, Function change: Able to demonstrate detailed change in, or maintenance of function, specifically physical abilities in relation to occupational performance, Ease: Relatively easy and fast to administer, Psychometrics: Established psychometric properties, with ability to aggregate data across individuals, Cost: Low purchase cost, Age: Suitable for use with clients aged 16 and above.

Participant demographic details

Data were collected from 58 participants (50% male) with an average age of 49 years (standard deviation (SD)−12.6). Fifteen (26%) of the participants reported having a carer present in their lives. This could have been a family member or a friend who assisted the participant in their day to day activities and whom the person relied on for assistance. All participants presented with physical (and in many cases mental) health conditions alongside their homelessness status, with 24 (42%) having both. Specific health conditions ranged from acute presentations (for example, sprains and fractures) to more chronic health conditions (for example, substance abuse, depression, anxiety, diabetes, chronic obstructive pulmonary disease, fibromyalgia and persistent pain) with chronic conditions being more prevalent. Comorbidities were present in 40 (70%) of the participants.

On average, participants were in contact with the homeless service for eight weeks (SD−10) receiving occupational therapy interventions to work on the goals reflected in the AusTOMs-OT scales. All participants were involved in an initial occupational performance review including the establishment of therapy goals and rating of AusTOMs-OT. This usually required 30–75 min dependent on the client and often incorporated assessments for equipment or housing needs. The next most common intervention was a domestic occupations assessment (usually 60–90 min) and then a variety of other occupational therapy interventions were used depending on the needs of the client (for example, personal care assessment, mental health interventions, energy conservation and support to carers). Forty-one (70%) participants were discharged from the homeless service following achievement of therapy goals, with the remaining participants ceasing their involvement due to a variety of reasons (for example, being transferred to another service, non-attendance or one participant died). While all AusTOMs-OT scales were available to occupational therapists and their clients, the scales most commonly reflected in collaborative goal setting and therefore used to measure change were Scale 5-Transfers (n = 32), Scale 7-Self Care (n = 22) and Scale 8-Domestic Life-Home (n = 19). Participants were also scored once in the Participation (n = 50) and Distress/Wellbeing (n = 50) domains across all three scales. Seven participants were rated on other AusTOMs-OT scales (Scale 1-Learning and Applying knowledge n = 3, Scale 2-Functional Walking and mobility n = 1, Scale 4- Carrying out daily life tasks and routines n = 1, Scale 11-Work, employment and education n = 1, Scale 12-Community life/recreation/leisure/play n = 1). However, there were insufficient data to be included in the analysis.

Service evaluation using AusTOMs-OT

Table 2 presents results from the Wilcoxon Signed Ranks Test examining participant change from admission to discharge from the service. Positive and negative ranks indicate those participants who had AusTOMs-OT scores that improved or deteriorated, respectively. Ties indicate those participants who did not change from admission to discharge. Large numbers of ties relative to positive/negative rankings can be seen for scores in the impairment domains, in addition to higher p values and small effect sizes. This is to be expected as occupational therapy interventions are more targeted towards the domains of activity, participation and wellbeing rather than impairment. Moderate effect size and statistically significant p values are demonstrated in all three scales for scores in the activity limitation domain. Scores in the participation and distress/wellbeing domains effect sizes are smaller but continue to be statistically significant.

Table 2.

Client score change from admission to discharge on AusTOMs-OT scales.

AusTOMs-OT domain	Admission	Discharge	Wilcoxon signed ranks test with effect size
Scale 5 – transfers: Impairment	n = 41	n = 40	−ve ranks = 1	Z = −2.218
	Median (IQR) = 4 (3,4)	Median (IQR) = 4 (3.5,4)	+ve ranks = 9	p = .027
			Ties = 30	r = 0.25
Scale 5 – transfers: Activity limitation	n = 42	n = 40	−ve ranks = 1	Z = −5.011
	Median (IQR) = 4 (3,4)	Median (IQR) = 5 (4,5)	−ve ranks = 32	p = .000
			Ties = 7	r = 0.55
Scale 7 – Self-care: Impairment	n = 23	n = 22	−ve ranks = 0	Z = −1.732
	Median (IQR) = 4 (3,4)	Median (IQR) = 4 (3,4)	+ve ranks = 3	p = .083
			Ties = 19	r = 0.26
Scale 7 – Self-care: Activity limitation	n = 24	n = 23	−ve ranks = 0	Z = −3.771
	Median (IQR) = 3 (2.5,4)	Median (IQR) = 4.5 (4,5)	+ve ranks = 18	p = .000
			Ties = 5	r = 0.55
Scale 8 – Domestic life- home: Impairment	n = 20	n = 18	−ve ranks = 0	Z = −1.342
	Median (IQR) = 4 (3.5,4)	Median (IQR) = 4 (3.5,4.5)	+ve ranks = 2	p = .180
			Ties = 16	r = 0.22
Scale 8 – Domestic life- home: Activity limitation	n = 20	n = 18	−ve ranks = 0	Z = −3.332
	Median (IQR) = 3 (3,4)	Median (IQR) = 5 (3,5)	+ve ranks = 14	p = .001
			Ties = 4	r = 0.54
Participation	n = 50	n = 48	−ve ranks = 1	Z = −3.473
	Median (IQR) = 3.5 (3,4)	Median (IQR) = 4 (3,4.5)	+ve ranks = 17	p = .001
			Ties = 30	r = 0.35
Distress/Wellbeing	n = 49	n = 46	−ve ranks = 0	Z = −4.161
	Median (IQR) = 3.5 (3,4)	Median (IQR) = 4 (4,5)	+ve ranks = 22	p = .000
			Ties = 24	r = 0.43

The clinical significance of the amount of change for participants as recorded on the AusTOMs-OT is represented in Table 3 indicating the percentage of clients who had more than (improved), equal to (same) or less than (deteriorated), the MCID of one point. Higher percentages of participants improved in their AusTOMs-OT scores by at least one point for the activity limitation and participation domains (between 86% and 91%) with distress/wellbeing percentages being slightly lower (77%).

Table 3.

Number and percentage of clients who achieved equal or more than the Minimum Clinically Important Difference (MCID) (when positive ranks attained on the Wilcoxon signed ranks test).

AusTOMs-OT scale and domain	Number of clients with positive ranks meeting the MCID (percentage)
Scale 5 – Transfers: Impairment	5/9 (55)
Scale 5 – Transfers: Activity limitation	29/32 (91)
Scale 7 – Self-care: Impairment	0
Scale 7 – Self-care: Activity limitation	16/18 (89)
Scale 8 – Domestic life-home: Impairment	1/2 (50)
Scale 8 – Domestic life-home: Activity limitation	12/14 (86)
Participation	15/17 (88)
Distress/Wellbeing	17/22 (77)

Discussion and implications

The AusTOMs-OT was introduced and was found to be a useful tool to record client change over time and to assist the HOTS to evaluate their service. The HOTS typically work with people who have complex needs, with functional difficulties resulting from both mental health conditions and physical impairments. These are often experienced alongside drug or alcohol misuse which can, in turn, contribute to chaotic lifestyles (Marshall et al., 2020c; Lloyd and Bassett, 2012; Queen et al., 2017). The HOTS was successfully using the MOHOST (Kramer et al., 2009) to record mental health outcomes; however, the team had no physical outcome measure with which to record and monitor the functional impact of these physical impairments for people experiencing homelessness. The results of this study indicate that individuals involved in the HOTS who receive a variety of occupational therapy interventions based on collaborative goal setting across different health domains do improve over time, and this may be related to participation in occupational therapy interventions. In particular, clinical and statistically important improvements were seen in the domains of activity limitation, participation and distress/wellbeing for Scale 5-Transfers, Scale 7-Self Care and Scale 8-Domestic Life-Home. Effect sizes were smaller in the Participation and Distress/Wellbeing domains and this may be indicative of the primary focus of interventions being the improvement of activity levels with participants. The multidimensional constructs of participation and wellbeing require multifaceted and often a team-based approach to improvement. However, the change in participant scores in these domains is promising.

An outcome measure sensitive to capturing change, the AusTOMs-OT has been introduced and is now used routinely in occupational therapy practice at the Glasgow Hunter Street Homeless & Asylum service. Client-led goal planning has improved through using the AusTOMs-OT as the selection of scales supported discussions between the occupational therapists and their clients and negotiation for goal setting. The routine collection of outcome measure data is also an important resource to enable senior therapists to quickly access information for funders and management as required to support and promote services or enable benchmarking against other homeless service providers. The complexity of health conditions experienced by individuals who are homeless necessitates a holistic approach. Use of an outcome measure that captures the change associated with occupational therapy homeless interventions like equipment provision through to more complex behaviour change interventions and training with a variety of length of treatment times, is essential. The use of outcome measures which are responsive to change is one of the most important aspects of conducting high quality research. This study has demonstrated that the AusTOMs-OT is such an outcome measure.

Research aimed at evaluating the effectiveness of interventions with homeless individuals is in the early stages of development (Marshall et al., 2020a). The majority of recent studies examine intervention programs aimed at the transition from homelessness to tenancy and focus on the development or re-learning of life skills associated with being housed. All describe evidence mostly gathered in the USA and make recommendations for further research to be completed in other countries (Marshall et al., 2020a; Munthe-Kaas et al., 2018), use measures that reflect occupation based goals (Thomas et al., 2011) and focus on community integration as an outcome with both housing and psychosocial interventions (Marshall et al., 2020b). Future investigations also need to focus on defining occupational therapy interventions in homeless services with more clarity. Evaluation of the impact of interventions, information regarding type (enablement skills used and occupational focus) and dosage (time spent and repetition), together with length of time engaged in the HOTS is important. This study provides a basis for addressing research gaps identified in the existing literature with insight into routine occupational therapy with homeless individuals in Scotland and the use of an outcome measure informed by an individual’s occupation-based goals across several domains including activity engagement, participation and wellbeing.

There are limitations to this study. It is not possible to infer that the improvements seen among individuals involved in the HOTS were achieved as a result of the interventions they received. In the future, a randomised control design, potentially with a wait list control group, is required to determine if participation in the occupational therapy interventions led to positive changes recorded. Only one homeless service was included in this research and generalisation is therefore limited. However, the experience of introducing an outcome measure is applicable to multiple occupational therapy services

Conclusion

This study has demonstrated the process of choosing an outcome measure to capture the change in individuals involved in a homeless occupational therapy service. Data collected on the AusTOMs-OT were able to be analysed to show that individuals do improve in their capacity for involvement in meaningful activity, levels of participation and the ability to cope with their current living environment. While research examining the effectiveness of homeless interventions is in development, an important precursor to this is the establishment of routine outcome measure use in clinical care. Research is now required to explore the effectiveness of specific occupational therapy interventions used in homeless services and the impact of these on not only housing outcomes for individuals, but participation in community life.

Key findings

• • The AusTOMs-OT can be incorporated as a routine outcome measure in a homeless service;

• • Data collected on two AusTOMs scales have reliably demonstrated change over time for clients who have physical as well as psychosocial problems, who received occupational therapy interventions.

What the study has added

The AusTOMs-OT has been identified as an outcome measure that can be used within the field of homelessness to help determine the value of occupational therapy interventions as part of a service evaluation.

ORCID iD

Carolyn A Unsworth https://orcid.org/0000-0001-6430-2823