Feel the Burn, Heal the Burn: Job Crafting and Burnout Among Occupational Therapy Professionals

Burnout is “a state of physical, emotional, and mental exhaustion that results from long-term involvement in work situations that are emotionally demanding” (Schaufeli & Greenglass, 2001, p. 501), often leading to negative health- and job-related outcomes. Common outcomes of burnout include depression, anxiety, sleep disturbances, headaches, respiratory infections and gastrointestinal distress, poor job performance, absenteeism, and turnover (Bakker et al., 2023). Burnout among health care providers has also been negatively associated with patient satisfaction, safety, and mental health (Park, 2021).

Health care professionals are vulnerable to burnout because of unique stressors such as exposure to human suffering, pressures from patients and family members, and demanding working conditions (National Institute for Occupational Safety and Health [NIOSH], 2024). Occupational therapy professionals are at high risk for job stress and burnout because of unmanageable workloads, waitlists, lack of autonomy and respect, lack of professional identity, pressure to discharge, handling of urgent versus chronic care cases, and managers’ unrealistic demands (Edwards & Dirette, 2010; González-Gutiérrez et al., 2004; Gupta et al., 2012). Previous studies have consistently reported moderate to high burnout levels among occupational therapy professionals in the United States and globally (Abaoğlu et al., 2021; Juy et al., 2022; Shin et al., 2022). Juy et al.’s (2022) study in Spain reported burnout prevalence ranging from 15% to almost 25%. In Turkey, Abaoğlu et al. (2021) identified 26% of occupational therapy professionals experienced burnout, with another 38% at risk for burnout.

Burnout is often the result of exposure to work-related stressors. Thus, burnout reduction strategies must consider organizational and psychological aspects of work experiences (Park, 2021). Work setting is one organizational aspect that warrants attention because of mixed findings in previous research. According to Escudero-Escudero et al. (2020), previous studies have indicated that burnout is least prevalent among occupational therapists working in the areas of physical rehabilitation and intellectual disability, whereas the highest burnout levels are seen among those working with older adults and in child care. In addition, occupational therapists in government-owned facilities and general hospitals experienced significantly higher levels of burnout than those in other settings (Brollier et al., 1986). A meta-analysis of occupational therapists’ burnout found that those working in large hospitals reported higher strain levels compared with those in small hospitals (Park, 2021). However, these studies by Brollier et al. (1986) and Park (2021) only investigated hospital settings. In contrast, a study conducted in Portugal (Reis et al., 2018) found no significant differences in burnout across practice areas (e.g., pediatrics, hospital acute care, outpatient hand therapy), although these authors noted challenges categorizing occupational therapists’ specific practice areas.

Research has also considered the relationship between burnout and individual differences among occupational therapists, revealing that older age, more education, and longer work experience are associated with lower levels of burnout among occupational therapists (Park, 2021). Conversely, having 10 or fewer years of experience in occupational therapy is associated with higher burnout (Poulsen et al., 2014). In one of few studies investigating resources related to burnout, Bar-Nizan et al. (2024) examined evidence-based practice in occupational therapy and self-efficacy related to burnout. They found that self-efficacy mediated the relationship between evidence-based practice and burnout.

Research examining specific occupational therapy in relation to burnout is limited to comparing supervisors versus staff (Park, 2021). Although this dichotomy is important, job title offers more insight related to predictors of burnout. For example, an occupational therapy supervisor may have different responsibilities and stressors than a rehabilitation director despite both being supervisory positions. More research is needed to investigate specific job characteristics across occupational therapy job titles and settings to identify additional ways to reduce burnout.

This study investigated how burnout varies across occupational therapy work settings and job titles; explored specific job characteristics associated with burnout, including the dearth of research regarding occupational therapists’ job resources; and evaluated job crafting as a possible strategy to reduce burnout and improve work experiences. Specifically, we aimed to address the following research questions among occupational therapy professionals: (1) How do burnout and job crafting vary by work settings and job titles? (2) Which job demands predict burnout? and (3) Which job resources predict use of job crafting? We built upon previous studies that investigated protective psychosocial factors for burnout (Juy et al., 2022), expanding the scope to include job resources as buffers against burnout and job crafting as a preventive measure. In addition, in response to Park’s (2021) meta-analysis on occupational therapists’ burnout, which underscored the ambiguity in research regarding job challenges, we examined both hindering and motivating job challenges, thus introducing a nuanced perspective that may contribute to greater clarity.

Most burnout studies use the Maslach Burnout Inventory (MBI; Maslach et al., 1996), assessing burnout in terms of emotional exhaustion, depersonalization, and reduced personal accomplishment. However, the MBI does not investigate the domains in which burnout occurs. In comparison, the Copenhagen Burnout Inventory (CBI; Kristensen et al., 2005) measures personal burnout (i.e., general exhaustion), work-related burnout (i.e., exhaustion attributed to work), and client-related burnout. Using the CBI to examine domain-specific burnout experiences among professionals in client-facing roles can help identify the context in which burnout occurs, thus clarifying targets for intervention.

The job demands–resources model (Bakker & Demerouti, 2017) is a dominant framework for understanding burnout. According to the model, job factors (i.e., physical, psychological, social, and organizational components) can be categorized as either job demands, which require sustained effort, or job resources, which can help achieve work goals, reduce job demands, or stimulate personal growth, learning, and development (Bakker & Demerouti, 2017).

Drawing from our review of prevalent stressors and protective factors within occupational therapy (e.g., Juy et al., 2022; Park, 2021), we conceptualized job demands as workload, work hours, emotional labor, uncertainty about or competing demands within one’s job, misunderstanding by others about the occupational therapy profession, and negative relationships with one’s supervisor. Job-related resources included meaningful work, job involvement, professional identity, autonomy, and perceived organizational, supervisor, and coworker support.

Although previous research has studied the relationship between occupational therapists’ job demands and burnout, job resources have received less attention. Importantly, job resources can activate motivational processes and may reduce burnout (Bakker & Demerouti, 2017 ; Bakker et al., 2023).

One approach for reducing burnout is job crafting, in which an employee alters aspects of their jobs by modifying demands or acquiring resources (Tims et al., 2012). Job crafting is akin to providing the just-right challenge to patients and clients through occupational therapy intervention. For example, occupational therapists overwhelmed by high patient volume and insufficient time to prepare patient treatments may craft their job by working with their supervisor to adjust their schedule or caseload or developing strategies to stretch treatment themes (e.g., repurposing the same treatment activity for several patients). Job crafting involves increasing structural and social resources, increasing challenging job demands (i.e., motivational job characteristics), and decreasing job demands that increase work stress (Tims et al., 2012). Ideally, job resources should match or exceed what is necessary for one’s job.

This study contributes to occupational therapy research by investigating how job crafting varies among occupational therapists, and relates to burnout. Identifying job resources strongly related to job crafting could help occupational therapists identify and acquire beneficial resources.

After receiving institutional review board approval, we conducted an online cross-sectional survey using Qualtrics between June 29 and December 12, 2022 and analyzed data from a heterogeneous sample of 400 occupational therapy professionals from 45 states across the United States. Participants were recruited via convenience and snowball sampling and needed to be actively working in the field of occupational therapy to participate. The survey was distributed via email and social media (e.g., occupational therapy Facebook groups) to our professional network, lists of alumni from occupational therapy training programs (Nova Southeastern University and Colorado State University), and professional associations (Occupational Therapy Association of California). A total of 652 people responded. Participants who completed less than 65% of the survey or failed two of the three attention checks were removed, leaving a final sample of 400. Missing data were less than 5%. This final sample demonstrated a sizable amount of missing data (up to 37.3% on some items), largely because of the number of self-employed participants who did not answer questions that did not apply to them. Thus, missing data were deemed a nonissue for most analyses; otherwise pairwise deletion was used. The response rate cannot be calculated because the number of occupational therapy professionals who received the survey is unknown. We collected demographic information (e.g., race, gender) and job-related variables (e.g., work setting, position tenure). Most participants were women (65.8%) and Caucasian or White (73.3%). Common practice settings included pediatrics (13%), long-term care or skilled nursing facility (13%), home health (12.8%), and hospital acute inpatient rehab (10.5%). Most participants were in the early stages of their current position with a job tenure of 0 to 5 yr (73%). See Table A.1 in the Supplemental Material for participant demographics (available online with this article at https://research.aota.org/ajot).

Participants completed measures of job demands and job resources outlined below. We conducted factor analyses to confirm the unidimensionality of each scale and consistency with previous research (results are available from the authors). See Table A.2 in the Supplemental Material for example items and reliability coefficients.

We measured job satisfaction using a single item, “Overall I am satisfied with my job.” This item has shown similar reliability and validity as a multiple-item scale (Fisher et al., 2016). This item was rated on a scale from 1 (strongly disagree) to 5 (strongly agree).

We measured burnout using the CBI (Kristensen et al., 2005), which consists of three dimensions: personal burnout, work-related burnout, and client-related burnout. All items were rated on a scale from 1 (never/almost never) to 5 (always).

To measure job crafting, we used the Job Crafting Scale (Tims et al., 2012). The scale consists of four dimensions: increasing structural job resources, increasing social job resources, increasing challenging job demands, and decreasing hindering job demands. One item (“I try to make my work more challenging by examining the underlying relationships between aspects of my job”) was dropped from the increasing challenging job demands dimension because of poor factor loading. All items were rated on a scale from 1 (strongly disagree) to 5 (strongly agree).

We used the Emotional Labor Scale (Brotheridge & Lee, 2003) to examine both surface acting (i.e., displaying emotions that are not truly felt) and deep acting (i.e., authentically aligning one’s emotions with the desired emotional display). One item (“I try to actually experience the emotions that I must show”) was dropped from the deep acting dimension because of poor factor loading. All items were rated on a scale from 1 (never/almost never) to 5 (always).

We assessed workload using the Quantitative Workload Inventory (Spector & Jex, 1998). All items were rated on a scale from 1 (less than once a month or never) to 5 (several times per day).

We used the Role Stressor Scale (Bowling et al., 2017) to measure role ambiguity (i.e., uncertainty about expectations or responsibilities) and role conflict (i.e., experience of incompatible demands or expectations stemming from multiple roles). One item (“My supervisors often tell me to do two different things that can't both be done”) was dropped from the role ambiguity dimension because of poor factor loading. All items were rated on a scale from 1 (strongly disagree) to 5 (strongly agree).

We measured job involvement (i.e., the degree of psychological involvement one experiences with their job, aspects of the job, or work in general) using a subset of the Job Involvement Scale (Reeve & Smith, 2001). All items were rated on a scale from 1 (strongly disagree) to 5 (strongly agree).

To measure professional identity, we adapted items from the Professional Identity Scale in Counseling (Woo, 2013) and generated some of our own. Psychometric assessment demonstrated that these items fit two dimensions: strain–misidentification by others and engagement–association with the field. The strain–misidentification dimension identified stressors associated with professional identity that are related to others misunderstanding the field of occupational therapy. The engagement–association dimension identified ways in which occupational therapy professionals identify with their field and others in their field. All items were rated on a scale from 1 (strongly disagree) to 5 (strongly agree).

We used the Work Design Questionnaire (Morgeson & Humphrey, 2006) to measure three dimensions of autonomy: work scheduling autonomy, decision-making autonomy, and work methods autonomy. One item (“The job allows me to plan how I do my work”) was dropped from the work scheduling autonomy dimension because of poor factor loading. All items were rated on a scale from 1 (never) to 5 (always).

We measured meaningful work using the Work and Meaning Inventory (Steger et al., 2012). The scale consists of three dimensions: positive meaning (i.e., the extent to which work has personal significance and meaning), meaning making through work (i.e., ways in which people’s meaning in life benefits from meaningful work), and greater good motivations (i.e., ways in which work is meaningful based on its impact on others). All items were rated on a scale from 1 (strongly disagree) to 5 (strongly agree).

To measure social support, we used the Social Support Scale (Eisenberger et al., 1986), which assesses social support from one’s organization, supervisor, peers, and subordinates. All items were rated on a scale from 1 (strongly disagree) to 5 (strongly agree).

All analyses were conducted using IBM SPSS Statistics (Version 29.0.0.0) and R Studio (Version 4.4.1). We checked all variables for normality. A few scale items displayed kurtosis; however, when items were averaged to create scale scores, all scale scores were normally distributed. We used a variety of multivariate statistical analyses, including analyses of variance (ANOVAs) to examine group differences and relative weights analysis (RWA) to assess variables that predicted burnout and job crafting. RWA is an alternative to multiple regression analysis that calculates the relative importance of predictor variables to the outcome variable relative to the overall regression coefficient. RWA addresses multicollinearity by transforming the independent variables into a new set of predictor variables that are orthogonal to each other but still related to the original predictor variables. These transformed variables are then regressed onto the dependent variable, and standardized β coefficients are obtained as raw relative weights that sum to the overall R². The raw relative weights are then rescaled to provide a predictor variable percentage of the total variance explained in the dependent variable. The absolute value of the rescaled weights will always equal 100. Thus, rescaled weights can be interpreted similarly to a percentage explaining the amount that a single variable contributes to the overall regression effect size (Johnson & LeBreton, 2004; Tonidandel & LeBreton, 2015).

Table 1 presents descriptive statistics and ANOVA results evaluating mean differences in burnout across job titles. Personal, work-related, and client-related burnout varied by job title and setting (p < .05). Notably, occupational therapy professionals working in the community reported lower levels of work-related burnout than those in hospital acute inpatient rehab, long-term care or skilled nursing facilities, and pediatrics. Occupational therapy professionals in pediatrics reported higher work-related burnout than those in mental health and home health settings.

Nearly all job demands were correlated with personal, work-related, and client-related burnout (see Table 2). Results from RWA identified the strongest predictors of both personal and work-related burnout: workload (21% and 19%, respectively), professional identity strains and misidentification by others (11% and 8%, respectively), and role conflict (7% and 7%, respectively). The three strongest predictors of client-related burnout were role ambiguity (7%), surface acting (7%), and negative relationships with supervisors (8%; see Table 3).

Using an ANOVA, we found that ability to decrease hindering job demands differed by job title: Supervisors and rehabilitation liaisons reported greater opportunities to decrease hindering job demands than general occupational therapists. No differences across job titles were found for increasing structural job resources. Job crafting differed by setting for increasing structural job resources, which was lower in academia than almost all other settings. Increasing social job resources and increasing challenging job demands did not vary significantly across work settings.

Job crafting was positively related to nearly all job resources and negatively related to personal, work-related, and client-related burnout (see Table 2). Based on RWA, the three strongest predictors of increasing structural job resources were professional identity engagement–association with the field (18%), meaningful work (14%), and autonomy (3%; see Table 4). The three strongest predictors of decreasing hindering job demands were perceived organizational support (8%), job involvement (5%), and perceived supervisor support (2%). The three strongest predictors of increasing social job resources were perceived organizational support (5%), perceived supervisor support (5%), and meaningful work (4%). The top three predictors of increasing challenging job demands were job involvement (11%), meaningful work (3%), and professional identity engagement–association with the field (2%).

This study, based on the job demands–resources model (Bakker & Demerouti, 2017) and job crafting (Wrzesniewski et al., 2010), investigated how burnout varies across occupational therapy jobs and practice settings, and examined job demands and resources to identify work conditions most predictive of burnout and job crafting. Consistent with previous research, our findings highlight excessive workload as a strong risk factor for burnout and supervisor support as an important resource to prevent or mitigate burnout (Gupta et al., 2012; Shin et al., 2022). We extend prior studies of burnout among occupational therapists by investigating a different set of work-related variables as possible predictors of burnout, including job titles, work settings, and job crafting. Furthermore, we considered multiple burnout domains, including work-related, client-related, and personal burnout.

Burnout varies across job titles and practice areas. Community-based occupational therapists reported lower work-related burnout than those in hospital acute inpatient rehab, long-term care or skilled nursing facilities, and pediatrics. We speculate that community occupational therapists may experience less bureaucracy and/or a lower caseload than occupational therapists in clinical settings (Brollier et al., 1986). Community occupational therapists are more likely to work for nonprofit organizations, which may allow for more job crafting than other settings. Nonprofit employment may enhance personal and professional connection to a mission, increasing meaningfulness and protecting against burnout (Correia & Almeida, 2020).

Research has shown that high workload, inadequate supervisor support, professional identity strains, role ambiguity, and role conflict strongly correlate with burnout (e.g., Edwards & Dirette, 2010; Gupta et al., 2012; Shin et al., 2022). Role ambiguity and surface acting are most predictive of client-related burnout, possibly influenced by occupational therapy curriculum and training that encourages occupational therapists to engage in therapeutic use of self, a term used to refer to an occupational therapist’s conscious efforts to optimize the therapeutic relationship (Taylor et al., 2011). By training occupational therapists to adapt to patient needs and learning preferences, occupational therapists may foster empathy despite potential professional doubt or frustration. Our results explained 28% of the variance in client-related burnout (R² = 0.28), suggesting the need for future research on additional job demands that may lead to client-related burnout.

In addition to assessing burnout, we investigated job crafting among occupational therapists to identify strategies to reduce burnout. Occupational therapists can craft their jobs by redefining their tasks, relationships, or perceptions to better align with their motives, strengths, and passions (Wrzesniewski et al., 2010). This may be accomplished by seeking challenges (e.g., expanding the role to include a passion project) or increasing resources (e.g., cultivating personal relationships with coworkers for social support). To reframe perceptions, an occupational therapist might compartmentalize aspects of the job, viewing parts of the job as necessary evils (e.g., paperwork) and other parts as meaningful (e.g., positive client interactions).

Occupational therapy supervisors reported greater opportunities to decrease hindering job demands than did general occupational therapists, which may reflect greater autonomy and clinical freedom. The opportunity to increase structural job resources was significantly lower in academia than most other settings; academic occupational therapists may have sufficient resources and less need to increase structural job resources. Alternatively, academic bureaucracy may prevent academic occupational therapists from acquiring additional structural job resources. Future research should consider the supports and barriers that accompany job crafting opportunities.

By exploring job resources associated with job crafting, we identified professional identity engagement– association with the field and meaningful work as key predictors for increasing structural job resources and challenging job demands. Job crafting can increase meaningful work (Wrzesniewski et al., 2010), which can promote well-being and protect against burnout (Correia & Almeida, 2020). Perceptions of meaningful work can stem from the nature of work tasks, the ability to cultivate positive relationships, and the notion that occupational therapy is a calling (Steger et al., 2012). Understanding that job crafting is a path to meaningful work may help occupational therapists strategically align tasks, relationships, and perceptions with their motives, strengths, and passions, bolstering beliefs that their work matters (Wrzesniewski et al., 2010).

Autonomy emerged as a strong predictor of increasing structural resources, aligning with the idea that employees enhance their jobs by granting themselves more autonomy and that autonomy in and of itself can be considered a structural resource (Tims et al., 2012). Occupational therapists who already experience higher levels of autonomy may have more opportunities to increase structural resources. Perceived supervisor support and organizational support emerged as ways to increase social job resources and decrease hindering demands. Future research might consider another way to measure job crafting (Slemp & Vella-Brodrick, 2013) considering task, cognitive, and relational job crafting.

Personal, work-related, and client-related burnout showed strong negative correlations with most job crafting aspects, highlighting the importance of primary, secondary, and tertiary interventions (Schonfeld & Chang, 2017). Within the context of burnout, a primary intervention attempts to prevent burnout before it occurs, such as by engaging in job design to eliminate or reduce job demands. Secondary interventions strive to buffer the impact of job demands on burnout, often making use of job resources such as social support and autonomy. Tertiary interventions aim to treat burnout postoccurrence. Given job crafting’s negative association with burnout, job crafting may best serve as a primary intervention to prevent burnout in the first place. Some research demonstrates how job crafting can reduce fatigue during work hours (Shi et al., 2021), suggesting that this may also be a pliable intervention for occupational therapists. With increasing attention directed toward research and training on job crafting interventions for occupational therapists to use with their clients (Sundar & Brucker, 2021), there is an increasing chance that occupational therapists may adopt job crafting techniques for their own benefit. Future research should consider whether employees with higher levels of burnout are less likely to engage in job crafting (Bakker et al., 2023).

One limitation of the study was the exclusive focus on the United States, potentially limiting the generalizability of findings to other cultural and geographic contexts. In addition, our sample may not represent all specialties or work settings within the occupational therapy profession, also affecting generalizability. Last, this cross-sectional study captured a snapshot of work-related experiences at a specific moment, lacking the ability to infer causal associations among job demands, job crafting, and burnout and to track changes over time. Longitudinal research examining job demands, resources, job crafting, and burnout among occupational therapists is needed to determine how burnout and job crafting vary across occupational therapy jobs and settings over time.

We recommend three strategies to help reduce burnout:

• ▪ Employers should train occupational therapists on how to craft their jobs and facilitate the process.

• ▪ Employers and occupational therapists should reduce job demands (e.g., workload, role ambiguity) and increase job resources (e.g., social support).

• ▪ Occupational therapists should foster meaningful work by aligning daily tasks with core values. Finding significance in preferred tasks might reignite their connection to work, a form of cognitive job crafting (Slemp & Vella-Brodrick, 2013). If these tasks are infrequent or lack depth, collaborating with a supervisor to enhance access to meaningful work opportunities may help.

Burnout is a critical issue for occupational therapists (Abaoğlu et al., 2021; Bar-Nizan et al., 2024; Juy et al., 2022; Shin et al., 2022). Some occupational therapy subgroups reported higher levels of burnout, as well as reported job resources that predict job crafting. Using these findings, employers and occupational therapists can target areas where burnout is more likely and build resources to increase ability to job craft.

This work was supported by the NIOSH Mountain and Plains Education and Research Center (T42 OH009229).