Mapping Effectiveness Studies of Occupational Therapy in Africa: A Scoping Review

Background Contextual relevance is an important consideration for evidence-based practice, especially in low- and middle-income countries where the nature of practice may differ from high-income countries. Resources and access to rehabilitation are constrained, and service-users face a range of intersecting challenges to activity and participation. Aim To evaluate the body of evidence for the effectiveness of occupational therapy in Africa, and to determine if systematic reviews with meta-analysis and/or meta-synthesis are feasible. Methods and Analysis. We conducted a systematic scoping review of published and grey literature by following PRISMA-ScR guidelines across 13 databases and through personal contact with occupational therapists across Africa. Covidence software was used to manage a blind review process by at least three reviewers per included article. The McMaster Quantitative Review Form, NHMRC levels of evidence, the Cochrane PROGRESS-Plus health equity criteria, and the TIDieR checklist informed data extraction using Microsoft Forms. Results The search yielded 4199 articles, of which 45 were included. Evidence in six fields of practice included paediatrics, mental health, physical rehabilitation, hand therapy, work practice, and community development, although the evidence was largely limited to South Africa (93% studies). Levels of evidence varied but included 13 RCTs. In all, 1957 participants were included, ranging in age from 25 days to 99 years, with a wide range of health conditions. Most studies reported a positive outcome for occupational therapy. Conclusion Findings suggest a moderate body of evidence to support occupational therapy in Africa. Systematic review with meta-analysis, assessment of risk of bias, and in-depth analysis of specific areas of practice are now required. All effectiveness studies in occupational therapy should include measures of occupational performance or participation, minimum reporting standard checklists should be used more consistently, and effect sizes should be consistently calculated and reported in effectiveness research.


Introduction
The overarching role of occupational therapy in Africa is to promote participation [1].Evidence-based practice (EBP) should be a foundation for enacting this role, since EBP is a key attribute of occupational therapy graduates [2].Working in an evidence-based way means making rehabilitation decisions based on the experience of the service user (what works well for me), the experience of the occupational therapist (what works well for my service users), and the research evidence (what has worked well for other people).
The nature of occupational therapy is also shaped by the context in which evidence-based services are provided.Nevertheless, systematic reviews are widely considered the gold standard for evidence-based practice and are the highest in the hierarchy of evidence.
Occupational therapy service delivery is already informed by a number of systematic reviews internationally, which have provided evidence of the effectiveness of occupational therapy across the lifespan.For example, Novak and Honan's [3] systematic review of paediatric occupational therapy for children with disabilities identified 75 systematic reviews and 54 randomised controlled trials that met their inclusion criteria.From these studies, 40 interventions were graded Green/Go, indicating sufficient evidence for implementation.Bennett et al.'s [4] systematic review and meta-analysis demonstrated moderate effectiveness of occupational therapy at home for people with dementia and their carers, across a range of outcomes.The economic value of occupational therapy in acute and subacute hospital settings was also demonstrated by Wales et al. [5], with five studies showing value for money.Meta-analysis of 16,718 patients in seven studies also showed that occupational therapy reduces readmission rates for adults hospitalised for surgical and general medical care, particularly if the intervention focused on the transition from the hospital to the home [6].Finally, for adults with depression, Christie et al. [7] found strong evidence for the effectiveness of return-to-work interventions on depressive symptoms.These selected systematic reviews demonstrate the breadth of occupational therapy and suggest its effectiveness throughout life for people with a range of health conditions.However, none of these reviews specifically considered how the evidence was shaped by the context of the research.
Although van Vuuren et al. [1] explored the role and scope of occupational therapy in Africa, we were unable to find any previous systematic or scoping reviews of the effectiveness of occupational therapy in Africa or other low-and middle-income country (LMIC) regions.This is problematic because effectiveness studies conducted in high-income countries (HICs) are not always applicable to evidencebased practice in LMICs, for a number of reasons.First, the nature of occupational therapy in LMICs may be different.For example, in South America, social occupational therapy has emerged as an area of practice that focuses on the needs of people who experience structural barriers to participation.Social occupational therapists work to address these barriers and their consequences [8].Occupational therapists from 11 African countries highlighted the challenges of working with very limited resources and the need for culturally appropriate occupational therapy and emphasized a focus on spirituality as a core component of service delivery [1].Occupational therapy also often needs to be population-based or include groups in contrast to working individuals [1].Second, occupational therapists in Africa face unique intersecting challenges.These include poverty and malnutrition, violence and political instability, stigma and discrimination against people with disabilities, and the need to understand cultural gender norms and their impact on participation [1].Third is access to rehabilitation services.In Africa, the number of occupational therapists per population is eight times lower than the level recommended by the World Federation of Occupational Therapists [9].This means that therapists have less time available per person than would typically be available in a HIC.Structural inequities in occupational therapy service provision in Africa are another potential source of difference between LMICs and HICs.Ned and colleagues found the occupational therapy workforce of 5180 registered with the Health Professions Council of South Africa in 2018 to be predominantly white (66%), under age 40 (67.7%),working in urban areas, and employed in the private sector (74.8%) [9].This means that people in rural areas, those living on low incomes, and those who speak a different language to their therapist may experience inequitable access to occupational therapy services.
These contextual challenges mean that it is important for occupational therapists in Africa and other LMICs to have access to research evidence that is relevant to their context.Using interventions that have been shown to be effective in similar contexts may enhance the impact of occupational therapy services.As an emerging profession in Africa, it is also important to raise the profile of the profession.Evidence-based practice increases the credibility of occupational therapy in Africa because therapists will be able to demonstrate their commitment to using scientifically valid and reliable evidence to inform their decision making.In a resource-constrained environment, evidence-based practice also helps to maximise the use of available resources.For example, research evidence may inform optimal treatment frequency and duration of treatment or the likelihood of further functional improvement after an injury.This helps to reduce waste or harm.A final benefit of having better access to research evidence from Africa is that therapists will be able to engage more easily in continuing professional development that is the most relevant to their practice.
If a systematic review has not been done, scoping reviews can assist researchers to map the available evidence and then identify and define specific research questions for subsequent systematic reviews [10].Scoping reviews can also promote evidence-based decision-making, by evaluating the available body of evidence.This means determining the level of confidence we can place in research findings, based on the quantity and quality of the available evidence.This includes the size of the body of evidence, the consistency of findings, and the type of research conducted within a hierarchy of evidence.
The aim of this study was to evaluate the body of evidence for the effectiveness of occupational therapy in Africa, to determine if systematic reviews with meta-analysis and/or meta-synthesis are feasible.These further reviews would need to evaluate the strength of the evidence, to enable occupational therapists and policymakers in Africa to make informed decisions about service delivery.The research question that guided our review was as follows: "What is the size, scope, and characteristics of the available research investigating the effectiveness of occupational therapy interventions in Africa?" Using the PICOS acronym, we included participants of any age or diagnosis, any intervention provided by an occupational therapist or under the direction of an occupational therapist, any comparison, any outcomes, and any study design included in the National Health and Medical Research Council's (NHMRC) Levels of Evidence [11].

Methods
We conducted a systematic scoping review following the procedures described in the JBI Manual for Evidence Synthesis [12].For reporting, we used the PRISMA extension for Scoping Reviews (PRISMA-ScR): Checklist and Explanation guidelines [13].A review protocol was completed as  [15] was used to identify important study characteristics including title, authors, year of publication, country where research was done, journal of publication, study design, research aim, sample size and type of sampling used, outcomes used and reporting of their psychometric properties, statistical tests used and identified level of significance, and whether effect size was reported or not.We categorised the field of occupational therapy inductively, according to our experience of the field.The NHMRC's Levels of Evidence [11] were used to classify the hierarchy of evidence available.Level I included a systematic review of randomised controlled trials (RCTs).Level II included RCTs.Level III included comparative studies such as pseudo-RCTs (Level III-1); studies with a concurrent control group (not randomised), cohort studies, interrupted time series (ITS) with a control group, and case control studies (Level III-2); and studies with a historical control, more than two single arm studies, or ITS without control (Level III-3).Level IV included case series, posttest studies, or pretest/posttest studies.
The generalizability of results to other populations was assessed using the Cochrane PROGRESS-Plus health equity criteria [16,17].This included the following 10 items: place of residence, any indication of race/ethnicity/culture/home language/religion, occupation, gender or sex, level of education, socioeconomic status, diagnosis, level of disability, HIV status, and marital status or living arrangements.For studies including children, occupation and level of education were considered reported if the authors reported the children went to school and their grade anywhere in the article.Living arrangements were considered reported if the study reported with whom the child lived.
The TIDieR checklist (Template for Intervention Description and Replication) [18] was used to extract data on the nature of the intervention and evaluate the quality of reporting.High-quality reporting enables replicability of an intervention.The 12-item checklist includes the name of the intervention, rationale, what (materials and procedures), person delivering the intervention, how, when, where and how much, how the intervention is tailored to individual needs, changes made during the trial, quality of planning, and fidelity of implementation.
To evaluate the consistency of the body of evidence, we first identified each outcome measure used, the measurement intervals used, and then the mean and standard deviation for the intervention and control groups at pretest, posttest, and follow-up.
The data extraction was completed using Microsoft Forms.Two authors piloted the use of Microsoft Forms with five articles, capturing data in duplicate, and then evaluating whether data had been consistently extracted.The included articles were divided between three authors based on available capacity.Data extraction was completed independently on allocated articles by each researcher.Queries or concerns were discussed in regular meetings throughout the time of data extraction.The final data set was checked for missing data and consistency of reporting by the first author.
As this was a scoping review, we did not complete a risk of bias assessment of the included studies.We did, however, analyse for publication bias and selective reporting within the studies.
3 Occupational Therapy International 2.5.Summary Measures.The size of the body of evidence was determined by counting the number of unique interventions evaluated; i.e., if a study was available as both a thesis and publication, it was only counted once.The type and quality of study design was determined by the frequency and percentage of publications at each of the NHMRC's levels of evidence.The quality of intervention reporting was determined by a two-point scoring system based on Plastow et al. [19] to indicate if each item on the TIDieR checklist was reported or not reported.The consistency of the findings and the quantity of evidence was determined by counting the number of studies that reported a positive result, no change, or a negative result.For generalisability, data on the characteristics of participants were summarised using frequencies and percentages.For age and sample size, we calculated the range and mean across studies where data was available.

Study Selection.
The study selection is presented in Figure 1.The initial database search yielded 4194 articles.An additional five articles were identified using other methods.When duplicates were removed, 2391 articles remained for screening.This suggests substantial overlap between the database findings and provides confidence that we identified most eligible studies.At title and abstract screening, we excluded 2111 studies.Thereafter, 280 studies were assessed for full-text eligibility.Of these, 230 were excluded, with reasons for exclusion indicated in Figure 1.When data extraction commenced, a further five studies were excluded.

Size and Scope of the Evidence.
A total of 45 studies evaluating the effectiveness of occupational therapy were identified.The earliest publication was published in 2000.The overwhelming majority (80%, 36/45) were published in the past 10 years (see Figure 2).Published articles appeared predominantly in the South African Journal of Occupational Therapy (n = 19, 42%), with two articles published in the Journal of Hand Therapy and Occupational Therapy International.Eleven other articles (24%) were each published in different journals.Eleven studies were unpublished master's theses (24%).
Six fields of occupational therapy were identified among effectiveness studies: paediatrics (n = 19), mental health (n = 10), physical rehabilitation (n = 9), hand therapy (n = 4), work practice (n = 2), and community development (n = 1).One study was included in both the paediatrics and hand therapy categories as it included splinting for children with cerebral palsy (n = 1).

Levels of Evidence.
A wide range of terminology was used to describe the study designs.We therefore classified each of the studies using terminology used in the NHMRC's levels of evidence [11] to enable comparison.Analysis using the NHMRC levels of evidence found no Level I study (systematic reviews of RCTs).At Level II, there were 13 randomised controlled trials, of which two were feasibility or pilot trials, and one used cluster randomisation.Level II evidence was only available in the fields of paediatrics (n = 7), mental health (n = 5), and hand therapy (n = 1).At Level III, there was 1 pseudorandomised controlled trial (Level III-1), 9 comparative studies with a concurrent control group (Level III-2), and 7 studies without concurrent controls (Level III-3).One-third of the studies (15/45) used a pretest, posttest, or case study design (Level IV).Most of these were physical rehabilitation (n = 5).Table 3 illustrates the distribution of studies within the various fields of occupational therapy, as well as how they are represented under the levels of evidence.

Reporting of Participant Characteristics
. A total of 1957 participants aged between 25 days and 99 years were included in the 45 studies.Using the PROGRESS-Plus criteria, participant characteristics were inconsistently reported, especially in journal articles.Of the 10 PROGRESS-Plus criteria, the authors reported a mean of 4.8 items (range 2-8, SD = 1 62).
Almost all of the research was conducted in South Africa (n = 42, 93%), with one study in Morocco, Uganda, and Namibia each.Most studies reported the location of the  5 Occupational Therapy International research within each country (40/45).Gender was similarly well-reported, with 40 of the 45 studies reporting the gender of participants.People receiving occupational therapy in Africa were almost evenly split between men (n = 722) and women (n = 771).There was, however, no report of any genders beyond the male/female dichotomy.Participants' health conditions were included in all 45 of the studies.However, the level of disability or functioning of participants was poorly reported, with only 12 studies reporting this equity characteristic.
The ten studies in the field of mental health included people with mood disorders [36][37][38][39][40][41], psychotic disorders [36,37,40], conduct disorders [42], and substance depen-dence [43].Two studies included participants without a mental health diagnosis.One addressed the needs of impoverished community-living people experiencing stress [44] and another teenagers at risk of alcohol use [45].González-Bernal et al.'s [46] community intervention for people with disabilities included those with unspecified mental disabilities and emotional difficulties in Morocco.
People with physical disabilities were also included in the Moroccan study.The remaining participants in the nine studies using physical rehabilitation included people with stroke [47][48][49][50][51], diabetic foot ulcerations [52], traumatic brain injury [53], and upper and lower limb injuries [49].In the field of work practice, one study included nursing auxiliaries working in a residential facility for people with intellectual disabilities [54] while another study included participants with traumatic brain injuries [55].
We identified five studies that evaluated the effectiveness of hand therapy, including children with cerebral palsy [56], people with mallet finger injury [57], flexor tendon injuries [58,59], and crush injuries [60].
Other demographic characteristics were less frequently reported.Only 29% (n = 13) of the articles reported either race, ethnicity, culture, religion, or home language of the participants.Of these, three studies only included people  Eight of ten studies in the field of mental health reported a positive effect on at least one outcome.These included the number of admissions, number of days in hospital, functional performance and participation (n = 5), and performance components (e.g., stress, mood, attention, concentration, and aggression).Function and participation were measured differently in all eight studies.One of the two studies in the field-ofwork practice demonstrated an improvement in cognition [55].In that study, seven of 10 participants returned to work.

Occupational Therapy International
All three studies investigating the effectiveness of hand therapy reported a positive outcome.Improvements were noted in the upper limb function in children with cerebral palsy [56], overall hand function and ability to work [60], and patient satisfaction [58].Similarly, all eight of the physical rehabilitation studies demonstrated at least some level of effectiveness.Positive outcomes included improved functional independence [49,50,53], occupational performance [48], reintegration and creative participation [52], self-care [49,51], mobility [49,53], and knowledge of caregivers of people with disability [62].

Discussion
This study intended to map the evidence base for occupational therapy in Africa and to make recommendations for future systematic reviews.The 45 studies we identified suggest that future systematic reviews of the effectiveness of occupational therapy in Africa are feasible.This is particularly important to ensure that research findings are translated into health policy and practice [65].The highest number of articles published was in the field of paediatrics, suggesting this is a good starting point.Six of these studies were RCTs, which suggests that a meta-analysis of outcomes may be feasible.The purpose of meta-analysis is to pool the results of a range of studies, to determine the overall effectiveness of an intervention.This requires a certain degree of similarity between the studies in their design, target population, outcome measures used, characteristics of the intervention, and methods of analysis [66].Although the calculation of heterogeneity is beyond the remit of this scoping review, at face value, there are significant differences in the paediatric studies presented in Table 4.In addition, only high-quality studies with a low risk of bias should be included in a meta-analysis.This suggests that researchers would need to carefully consider their approach to metaanalysis, under the guidance of a biostatistician.For mental health, physical rehabilitation, and hand therapy, most international systematic reviews focus on specific diagnostic categories.The body of evidence is not yet large enough for meta-analysis in any of these three fields of practice.However, an in-depth narrative review of practice in mental health and physical rehabilitation services would provide useful information for clinicians establishing new services or enhancing existing ones.
After establishing the size of a body of evidence, it is important to consider the consistency of the findings.The most consistent findings of effectiveness were in the fields of physical rehabilitation and hand therapy.The effectiveness of occupational therapy for people with mental health problems was also demonstrated relatively consistently.In all three of these areas, there was a focus on measuring occupational performance or participation, which is where we would expect to see the greatest gains from occupational therapy.This contrasts with the effectiveness of occupational therapy for children and young people [1].The evidence for this group of service users was much less consistent, with a third of studies demonstrating no effectiveness.However, this was also the group of studies that most frequently used only one outcome measure.In addition, many of the studies measured performance components, rather than occupational performance.Notably, none of the studies included a child-rated outcome measure like the Canadian Occupational Performance Measure [67] or the Child Occupational Self-Assessment [68].Occupational therapy for older people is an important area of practice internationally that is supported by a strong body of evidence [69][70][71].Our scoping review found that no studies focused exclusively on the needs of older people and no studies on the effectiveness of occupational therapy for people with dementia.Community development is a similarly important field of practice in Africa that is not being assessed for effectiveness.
The results of this scoping review also highlight a number of inequities in this body of evidence that limit the generalisability of the evidence base to healthcare systems across Africa.Studies to evaluate the effectiveness of occupational therapy were limited to four African countries.This is problematic since 15 African countries are now training occupational therapists for their local healthcare systems.These include Ghana, Kenya, Malawi, Mauritius, Nigeria, South Africa, Tanzania, Uganda, and Zimbabwe [72].More recently, Cameroon, Ethiopia, Madagascar, Morocco, Namibia, and Rwanda have commenced programs.Race, ethnicity, and culture are contested social constructs that socially group people together according to how they look or the traditions they practice [73].The failure to report race and ethnicity in occupational therapy research in Africa disregards the reality of inequity, injustice, and social stratification that is a consequence of the continent's colonial past and that has influenced population health.The low number of articles reporting race, ethnicity, culture, language, or religion also highlights the risk of inequity in occupational therapy service provision.Of the 13 articles that reported any of these constructs, almost a quarter (n = 3) only included white participants.Because these descriptors may hide health disparities, and limit the generalisability of findings, it is important to report these participant characteristics at this time [73].
The least reported health equity characteristic across the studies was HIV status.This is despite the fact that Africa is the continent with the largest number of HIV-positive people [74].The importance of this equity characteristic was highlighted by Philip et al.'s qualitative study of the lived experience of people with disabilities and HIV in accessing HIV services [75].That study found that HIV status, disability, stigma, poverty, and gender "collide in a hierarchy of identities to impede accessibility to HIV services" [75].It is expected that access to occupational therapy services and the effectiveness of those services will similarly be impeded by the intersection of a range of different health equity characteristics.For example, De Klerk et al. [76] found that male gender, being unemployed, and receiving state-subsidised healthcare because of low income were some of the factors predicting nonattendance at a regional rehabilitation centre in Cape Town, South Africa [76].This is why the limited reporting of socioeconomic status in the identified studies is also problematic.A number of the PROGRESS-Plus health equity characteristics point to the economic resources that may be available to a person that enable them to access 8 Occupational Therapy International      Occupational Therapy International healthcare.These include people's level of education, employment status, type of work, and level of income.None of these were adequately reported.

Conclusion
Evidence-based practice is a core attribute of occupational therapy graduates (Nicola Ann [2]).The findings from this review indicate emerging contextually relevant evidence to support the effectiveness of occupational therapy in Africa.A number of studies across the scope of occupational therapy practice have been conducted.More research is needed before making a judgement about the strength of this evidence.Future research should include a systematic review with meta-analysis of study findings, assessment of risk of bias, an in-depth analysis of assessments and interventions used in specific areas of practice, and an interrogation of the reasons why effectiveness is not always demonstrated.It is also recommended that all effectiveness studies in occupational therapy include a measure of occupational performance or participation, that minimum reporting standard checklists are consistently used to improve the quality of reporting, and that effect sizes are consistently calculated and reported in effectiveness research.

Table 1 :
Inclusion and exclusion criteria.

Table 2 :
Database search information.
* OR efficacy OR efficiency OR benefit OR influence OR outcome OR result OR improve * OR evidence OR significance OR implication OR value OR consequence * OR impact OR usefulness OR qualit * OR importance OR insight * OR barrier * OR challenge * OR problem

Table 3 :
[48][62][63]idence for occupational therapy in Africa.Of the 23 studies that reported employment status, 12 included children in education.Of the remaining 11, only five studies reported the type of work by adult participants.These included being ECD practitioners; skilled, semiskilled, and unskilled occupations; nursing auxiliaries; unqualified/casual or professional; sedentary, light, medium, or heavy work; and maintenance worker.Similarly, levels of education and socioeconomic status of participants were poorly reported.Only nine studies (20%) reported the level of education of adult participants.Of these, only three studies included participants with any tertiary qualifications, and three studies included participants with only primary school or no formal education.Similarly,only 13 of the studies (29%) reported the socioeconomic status of the participants.Of these, 12 used a range of terminology to describe participants living on low incomes.One study was conducted in a private clinic, suggesting a higher socioeconomic status of participants.The only two studies that reported HIV status were Ramugondo's feasibility RCT to improve play of children who were HIV positive[26]and Msengana et al.'s intervention to improve upper limb mobility and personal management in people with stroke[51].In addition, only four studies reported the marital status or living arrangements of participants.Occupational therapy research was not limited to healthcare settings.Although 15 interventions were delivered in hospitals, five in specialist clinics (e.g., hand clinic and diabetic foot clinic), and three in rehabilitation centres, nine were delivered in the school setting, and one was delivered partly in participants' workplace.byteachersandnurses[61][62][63].One study used telehealth as the mode of delivery[48].3.8.Dose.There was substantial variation in the duration and frequency of the interventions and the number of sessions offered by occupational therapists.
[47][48][49][50]ics of InterventionReporting.Studies reported between one and ten of the 12 TIDieR characteristics, with a mean and median of seven items reported per study.All of the studies provided a name for their intervention somewhere in the article.Interventions varied between multicomponent occupational therapy programs and specific treatment techniques.Most of the studies (n = 33) provided a rationale, goal, or theory informing the intervention.Less than half of the studies described how the interventions were tailored to the individual (n = 19).Only ten described any modifications made to the planned intervention, and only ten reported any plans to monitor intervention fidelity, although 14 studies provided at least some information on actual intervention fidelity.3.6.Location of Intervention.3.7.Mode of Intervention.The mode of delivery, reported in 37 studies, also varied.An equal number of interventions were delivered in groups (n = 15) and one-to-one sessions with the occupational therapist (n = 16).Two studies included both individual and group intervention components[36,49].In a small number of studies, occupational therapists provided training and then supervised the inter-vention delivered 3.9.Characteristics of Outcome Measures.Studies used a mean of 2.38 measures (range 1-9, SD = 1 64), although 18 studies only used one outcome measure.There was little consistency in outcomes used, even within the same field of practice.Less than half (21/45) of the studies included a measure of quality of life, participation, or occupational performance.The most commonly named outcomes at a participation level were the Barthel Index[47][48][49][50], level of creative

Table 4 :
Effectiveness studies from Africa by field of practice and level of evidence.