Abstract
This study is the first to leverage the perspectives of clinical providers and administrators within the Veterans Health Administration to elicit explanations for disparities in access to inpatient psychiatric occupational therapy services.
Approximately 100,000 Veterans each year require inpatient psychiatric care because of severe symptoms that elevate suicide risk and limit their ability to engage in daily activities (U.S. Department of Veterans Affairs, 2013, 2023). Such psychiatric hospitalizations are an opportunity to intervene with the aim of promoting Veterans’ full recovery, stabilizing symptoms, and empowering them to engage in meaningful and health-promoting activities upon discharge (Substance Abuse and Mental Health Services Administration, 2012; U.S. Department of Veterans Affairs, 2023). Occupational therapy practitioners are uniquely equipped to enable such recovery-oriented services in the inpatient psychiatric setting by delivering skilled interventions aimed at promoting independent completion of activities of daily living (ADLs) and other occupations necessary for a successful transition after discharge (American Occupational Therapy Association, 2017; Synovec, 2015). Indeed, inpatient psychiatric occupational therapy services are associated with increased performance of ADLs and other life activities as well as decreased risk for 30-day readmission in the VHA (D’Amico et al., 2018; Kinney et al., 2024a; Lipskaya-Velikovsky et al., 2016, 2020). Therefore, beneficial occupational therapy services should be distributed equitably.
Our understanding of factors shaping the equitable distribution of occupational therapy services in the VHA inpatient psychiatric setting is limited. Equitable access occurs when patient need (e.g., ADL limitations) is the primary driver of care utilization rather than nonclinical factors, such as organizational characteristics (e.g., facility staffing; Anderson et al., 2014). Our recent study leveraged a national sample of Veterans to uncover evidence of such disparities in access to inpatient psychiatric occupational therapy in the VHA (Kinney et al., 2024b). Specifically, Veterans with ADL limitations were more likely to use inpatient psychiatric occupational therapy, indicating that Veterans in need tend to receive such services. However, occupational therapy utilization was also influenced by nonclinical factors, including characteristics of VHA facilities. Although such evidence sheds light on the potential presence of facility-level disparities in access to beneficial occupational therapy services, the mechanisms explaining such an inequitable distribution of these services are unclear.
This study is the final phase of an explanatory–sequential mixed-methods design (Creswell & Plano Clark, 2017). We conducted an in-depth qualitative exploration of plausible explanations for previously reported quantitative findings regarding facility-level disparities in access to inpatient psychiatric occupational therapy services in the VHA (Kinney et al., 2024b). Knowledge of such mechanisms, informed by the perspectives of VHA providers and administrators, is needed to identify specific and modifiable targets of strategies aimed at promoting equitable access to inpatient psychiatric occupational therapy services in the VHA.
Method
An explanatory–sequential mixed-methods design was used, whereby the qualitative component was used to explain observed findings from the quantitative component (Creswell & Plano Clark, 2017). The stated purpose of the previously reported quantitative component (Kinney et al., 2024b) was to first investigate whether a direct measure of occupational therapy need, ADL limitations, was associated with inpatient psychiatric occupational therapy utilization in the VHA. In addition, we examined whether the relationship between ADL limitations and occupational therapy utilization differed according to facility-level characteristics. In this article, we highlight select findings from the original quantitative study according to their statistical significance and theoretical relevance. Additional details can be found in Kinney et al. (2024b). All study procedures were approved by local regulatory institutions.
Quantitative Component
Participants and Procedures
Study procedures for the quantitative strand of this study have been detailed elsewhere (Kinney et al., 2024b). In brief, the analysis included 133,844 Veterans who received inpatient psychiatric care in VHA facilities between 2015 and 2020. Inclusion criteria were as follows: receipt of care at a VHA facility with occupational therapy services available, documented ADL performance, and survival of hospitalization.
Data Collected
Utilization of inpatient psychiatric occupational therapy (yes vs. no) was identified through documentation of occupational therapy–specific Current Procedural Terminology ® codes, workload codes, or provider designation. ADL performance was classified as independent or limited using a validated algorithm (Kinney et al., 2024b) applied to nursing assessments completed at admission. We obtained information about the following facility-level factors, representing characteristics of VHA facilities to which each Veteran had been admitted: facility complexity, psychiatric care quality, and psychiatric bed supply. Facility complexity (low, medium, or high) is a composite index calculated by the VHA that reflects factors such as the complexity of patients served (e.g., clinical risk) and resources associated with the facility (e.g., academic affiliation; Veterans Health Administration Office of Productivity, Efficiency, and Staffing, 2020). Psychiatric care quality (low, moderate, or high based on tertile values for the distribution) at the facility was measured using the Hospital Based Inpatient Psychiatric Services indicator, which reflects the proportion of Veterans (1) who were screened appropriately (e.g., with respect to risk to self and/or others) and (2) were discharged using appropriate justification for ≥2 antipsychotic drugs (Joint Commission, 2016). Psychiatric bed supply was defined as the number of operational inpatient psychiatric beds at the time of admission (low, moderate, or high based on tertile values for the distribution). Please see Kinney et al.’s (2024b) article for a descriptive analysis of study variables.
Data Analysis
We used modified Poisson regression to examine associations among ADL performance, facility characteristics, and occupational therapy utilization, while adjusting for clustering within facilities and other covariates. Interactions between ADL performance and facility characteristics were tested, with statistical significance set at α = .05.
Qualitative Component
Participants and Procedures
We used criterion sampling to select providers and administrators (hereafter referred to as clinical informants) within the VHA on the basis of their expertise in the delivery of inpatient psychiatric occupational therapy services (Palinkas et al., 2015). We complemented this strategy by posting study information to listservs and using snowball sampling. Providers were eligible if they delivered occupational therapy or related services (e.g., psychiatrists) in VHA inpatient psychiatric settings. Administrators were eligible if they were involved in administrative decision making regarding the delivery of occupational therapy in VHA inpatient psychiatric settings at any level of the VHA (e.g., facility; national). We invited 25 participants, 21 of whom agreed to participate. Five clinical informants were lost to follow-up, leaving 16 who completed semistructured interviews.
Data Collection
Clinical informants completed an online survey using Research Electronic Data Capture (REDCap; Harris et al., 2009). Surveys included sociodemographic (e.g., gender) and professional characteristics (e.g., discipline, experience).
Semistructured interviews lasted approximately 1 hr and were conducted using Microsoft Teams. Interviews were audio recorded and transcribed verbatim. Interview guide development was informed by study findings and conceptual frameworks explaining factors that shape health care access (e.g., Woodward et al., 2019). Interviews included questions designed to elicit perspectives on potential explanations for observed study findings from the quantitative strand of the study (see Table 1). Interviews were anchored to a description of methods used to conduct the quantitative strand of the study as well as narrative and visual summaries of study findings.
Exemplar Qualitative Interview Questions
Note. ADLs = activities of daily living; OT = occupational therapy.
Data Analysis
Descriptive–interpretive analysis was used to analyze interview data (Elliott & Timulak, 2021). An analytic memo summarizing salient concepts was completed after each interview. Analytic memo content was used to construct a preliminary codebook composed of meaning units, or text that reflected a salient idea that advanced the study purpose. This preliminary codebook was independently applied to two transcripts by three researchers (Adam R. Kinney, Morgan Nance, and Molly E. Penzenik). Meetings were then convened to refine the codebook before applying it to the remaining 14 interviews.
Next, a primary coder with experience delivering inpatient psychiatric occupational therapy (Kinney) assigned meaning units to the textual data. One of the two secondary coders (either Nance or Penzenik) independently reviewed each transcript to confirm the coding. Discrepancies were identified and resolved by reaching a negotiated consensus on each meaning unit (Bradley et al., 2007). Meaning units were then consolidated into broader themes, which in turn were merged with other themes on the basis of shared interpretation and the corresponding finding from the quantitative strand of the study. Qualitative procedures were consistent with practices intended to bolster the credibility and trustworthiness of study findings, including achieving a consensus on meaning units and themes (Chwalisz et al., 2008; Tong et al., 2007; Yardley, 2008). We also identified illustrative quotes to demonstrate consistency between textual data and assigned themes. Qualitative analysis was conducted using ATLAS.ti software (Version 23).
Synthesis of Quantitative and Qualitative Strands
We created a joint display table that grouped quantitative and qualitative data together, enabling comparisons of results (Creswell & Plano Clark, 2017). The first author (Kinney) drew meta-inferences, or interpretations drawn from the merged quantitative and qualitative strands, that revealed explanations for results observed in the quantitative strand of the study (Creswell & Plano Clark, 2017; Teddlie & Tashakkori, 2009). Meta-inferences were then reviewed by coauthors and a consensus was reached.
Results
The majority of clinical informants were female (81%), direct service providers (75%), and occupational therapy practitioners (75%). More than one-third reported at least 20 yr of experience in psychiatric care in the VHA (38%; Table 2).
Clinical Informants (n = 16)
Note. ref. = reference category.
Meta-inferences drawn from the synthesis of quantitative and qualitative results revealed plausible explanations for quantitative findings interpretations. Table 3 is a joint display table that summarizes quantitative results, qualitative findings, and meta-inferences.
Joint Display Table of Mixed-Methods Findings
Note. ADL = activities of daily living; IADLs = instrumental activities of daily living; OT = occupational therapy.
ADL Limitations Associated With Increased Likelihood of Occupational Therapy Utilization
Qualitative interviews elicited several potential explanations for the observed relationship between ADL limitations and inpatient psychiatric occupational therapy utilization. Mechanisms included the importance of ADLs as a target of occupational therapy intervention but also included aspects of care delivery, such as artificially narrow perceptions of the occupational therapy scope of practice. Specific themes included (1) influences health and well-being, (2) most evident occupational therapy target, and (3) occupational therapy scope simplification.
Facility-Level Factors Associated With Inpatient Psychiatric Occupational Therapy Utilization
Facility Complexity Influences the Relationship Between ADL Limitations and Occupational Therapy Utilization
Explanations for why facility complexity influences the relationship between clinical need (i.e., ADL limitations) and access to inpatient psychiatric occupational therapy primarily focused on resources, most notably staffing levels. Such resources enable the ability of high-complexity facilities to satisfy demand for a range of occupational therapy services and provide a strong foundation for advocacy efforts that increase awareness of the value of occupational therapy services in this setting. Specific themes included (1) high-complexity staffing and (2) advocating for occupational therapy value.
Psychiatric Care Quality Influences the Relationship Between ADL Limitations and Occupational Therapy Utilization
Qualitative interviews revealed several potential explanations for the role of care quality in shaping the relationship between clinical need (i.e., ADL limitations) and access to inpatient psychiatric occupational therapy. Explanations included aspects of facility resources (i.e., staffing) and culture, including variability in facilities’ commitment to evidence-based processes and interdisciplinary care. Themes included (1) high-quality staffing, (2) evidence-based processes, and (3) interdisciplinary care.
Psychiatric Bed Supply Influences the Relationship Between ADL Limitations and Occupational Therapy Utilization
The primary explanation for why bed supply may influence the relationship between clinical need and access to inpatient psychiatric occupational therapy concerned the presence of a favorable staff:patient ratio in smaller psychiatric units.
Discussion
In this explanatory–sequential mixed-methods study, we elicited the perspectives of VHA providers and administrators regarding plausible explanations for previously reported facility-level disparities in access to inpatient psychiatric occupational therapy services (Kinney et al., 2024b). Although the prior quantitative study advanced our understanding of the presence of varying levels of access to inpatient psychiatric occupational therapy across VHA facilities, before the present study little was known regarding potential mechanisms by which these disparities occurred. Our study revealed that facility resources and aspects of facility culture may explain the observed disparities in access, which represent modifiable targets that can inform the development of corresponding strategies aimed at achieving equitable access to inpatient psychiatric occupational therapy services in the VHA, in turn eliminating potential disparities in occupational performance in this population.
Facility resources—in particular, occupational therapy practitioner staffing levels—were identified as potential explanations for the observed influence of each of the three measured facility characteristics on access to inpatient psychiatric occupational therapy. This extends previous findings indicating that Veterans receiving care at facilities with fewer resources experience an unmet need for occupational therapy services in the VHA, suggesting that a lack of staffing is a modifiable factor capable of reducing disparities in access to these beneficial services (Kinney et al., 2023, 2024b). A multipronged approach should be considered to address such staffing shortages and promote equitable access to inpatient psychiatric occupational therapy. First, advocacy efforts should aim to increase hiring of occupational therapy practitioners in this setting. Administrators may perceive efforts aimed at increasing occupational therapy staffing as being cost prohibitive; however, recent evidence indicates that inpatient psychiatric occupational therapy utilization is associated with a lower risk of psychiatric readmissions and thus may improve outcomes while reducing excess costs associated with repeat admissions (Kinney et al., 2024a). Such evidence can be used to support advocacy efforts by highlighting the value of occupational therapy services in this setting. Second, in the short term, novel service delivery models can accommodate staffing shortages by leveraging resources from VHA facilities with more staff members to address unmet needs for occupational therapy in clinical settings with fewer resources. Telehealth or in-person models have been used, for example, to expand access to primary care and mental health services in the VHA, and similar approaches aimed at eliminating observed disparities in access to inpatient psychiatric occupational therapy should be considered (Leung et al., 2022).
Likewise, aspects of facility culture were identified as potential explanations for the observed relationship between inpatient psychiatric care quality and access to occupational therapy. Culture in a clinical setting has been defined as comprising the values, beliefs, and norms shared among staff, and it is considered an important factor that shapes health care delivery (Damschroder et al., 2022). In the current study, facilities with a shared commitment to evidence-based and interdisciplinary care were perceived as having a better understanding of the distinct value of occupational therapy, in turn leading to increased demand for a broader range of services, above and beyond treatment targeting ADL limitations. Prior work has highlighted the value of VHA facilities adopting recommended practices for inpatient psychiatric care, including the use of an interdisciplinary team. For example, a recent study revealed that receiving care at a VHA facility that used an interdisciplinary team was associated with reduced risk for psychiatric readmission (McGuire et al., 2022). However, our study is the first to uncover evidence that a shared commitment to evidence-based and interdisciplinary inpatient psychiatric care may increase access to occupational therapy in this setting. Strategies capable of promoting a shared commitment to inpatient psychiatric care that values the distinct role of occupational therapy should be developed and tested. For example, identifying and preparing champions in particular VHA facilities, or those who dedicate themselves to overcoming resistance to the adoption of care practices, may help forge a shared commitment to evidence-based and interdisciplinary inpatient psychiatric care, including increased access to beneficial occupational therapy services (Powell et al., 2015; Shea, 2021).
Perspectives regarding explanations for the relationship between ADL limitations and inpatient psychiatric occupational therapy utilization were mixed. Some participants highlighted that ADL limitations are critical indicators of safe community reentry and overall well-being and are therefore worthy indicators of clinical need for occupational therapy in this setting. Others, however, noted that the interdisciplinary team lacks an awareness of the breadth of occupational therapy’s scope of practice, leading to unmet need for other relevant targets, including engagement in personally meaningful activities. Efforts to educate members of the interdisciplinary team regarding the myriad ways in which occupational therapy can help address the needs of Veterans receiving inpatient psychiatric care is warranted. Furthermore, decision support systems that automatically trigger referrals on the basis of limitations in engagement in social and meaningful activities, in addition to ADL limitations, could expand access to beneficial services among Veterans with diverse occupational performance challenges. Indeed, such systems have been deployed in other settings to promote appropriate access to therapy services (Martinez et al., 2024).
Limitations
One limitation of the current study is its exclusive focus on the perspectives of providers and administrators. Although such perspectives yield important information capable of facilitating the development of strategies aimed at equitable access to beneficial occupational therapy services in the inpatient psychiatric setting, future research should gather Veterans’ perspectives regarding barriers to accessing such care. In addition, the majority of our participants were occupational therapy practitioners; future research should sample more members of the interdisciplinary care team, referring providers in particular, to substantiate our findings. Furthermore, we were unable to sample participants from VHA facilities with varying levels of resources, despite the importance of this factor in explaining access to inpatient psychiatric occupational therapy. Future studies should use purposive sampling strategies to advance our understanding of the relationship between facility resources and access to inpatient psychiatric occupational therapy. Last, our sample exclusively comprised VHA providers and administrators, and thus whether our findings are transferrable to other care systems is unclear. Future research should attempt to replicate findings in non-VHA systems.
Implications for Occupational Therapy Practice
Our findings have the following implications for occupational therapy practice: ▪ According to the perspectives of clinical informants, previously reported facility-level disparities in access to beneficial inpatient psychiatric occupational therapy services may be explained by aspects of facility resources and culture. ▪ Our findings lay the foundation for the development of strategies that modify practice and policy to promote equitable access to these beneficial services.
Conclusion
This mixed-methods study leveraged the perspectives of VHA providers and administrators to identify specific and modifiable factors explaining previously reported facility-level disparities in access to inpatient psychiatric occupational therapy. Observed disparities in access are potentially explained by facility resources and aspects of facility culture. Such factors may represent modifiable targets that can inform the development of strategies aimed at achieving equitable access to inpatient psychiatric occupational therapy in the VHA.
Footnotes
Acknowledgments
This work was supported by a Health Services Research grant to Adam R. Kinney from the American Occupational Therapy Foundation. This work was not supported by and does not necessarily represent the views of the U.S. Department of Veterans Affairs (VA) or the U.S. government. Lisa A. Brenner reports grants from the VA, U.S. Department of Defense (DoD), the National Institutes of Health (NIH), and the State of Colorado; editorial remuneration from Wolters Kluwer and the Rand Corporation; and royalties from the American Psychological Association and Oxford University Press. In addition, she consults with sports leagues via her university affiliation. Jeri E. Forster reports grants from the VA, DoD, NIH, and the State of Colorado.