Abstract
The authors describe the current practices of occupational therapy practitioners for evaluating cognitive function in adults.
Cognitive impairment is common among adults in hospital settings and is a significant contributor to poor functional outcomes (Fogg et al., 2018; Linkens et al., 2020; Rochette et al., 2021). Cognitive impairment affects the ability to complete everyday activities from basic activities of daily living such as dressing, to more complex instrumental activities of daily living such as managing medications or personal finances. Occupational therapy practitioners play a primary role in identifying how cognitive impairments can limit a person’s ability to accomplish everyday functional skills in community environments (i.e., functional cognitive skills). Despite this key role, the methods used by occupational therapy practitioners to evaluate functional cognition across treatment settings are not well understood.
In general, occupational therapy practitioners use two standardized methods to evaluate the impact of cognitive status on everyday function. In this article, we use the terms neurocognitive and functional cognitive to describe these separate methods. A second important distinction is between screening and assessment. Screening is used to identify whether a person may have a problem, and assessment is used to identify the nature of the problem and indicate intervention avenues (Maxim et al., 2014). Neurocognitive screens—for example, the Montreal Cognitive Assessment (MoCA; Nasreddine et al., 2005) and the St. Louis University Mental Status Exam (SLUMS; Tariq et al., 2006)—rely on a bottom-up approach that measures specific cognitive skills (e.g., attention, memory, executive functions) to develop an overall score indicating the potential for cognitive impairment. Neurocognitive screening measures are widely used by occupational therapists to determine overall cognitive capacity, but their relationship to independent community functioning is limited (Marcotte et al., 2022; Sherman et al., 2022). Neurocognitive assessments are not typically performed by occupational therapy practitioners and are reserved for use by professionals licensed to administer them. In addition, occupational therapy practitioners who administer neurocognitive screens may follow up with functional cognitive screening or assessment (Jaywant et al., 2021).
In contrast to the bottom-up approach used in neurocognitive screening, functional cognitive screening (e.g., Allen Cognitive Level Screen [ACLS; Allen et al., 2007] and the Menu Task [Edwards et al., 2019]) and assessment (e.g., Weekly Calendar Planning Activity [WCPA; Toglia, 2015]) uses a top-down approach to evaluate the ability to integrate cognitive skills, performance skills, and performance patterns to accomplish everyday tasks (Giles et al., 2017). Occupational therapy practitioners may progress from using a functional cognitive screen to a more in-depth functional cognitive assessment or may go directly to assessment (e.g., the 10-item WCPA may be used with or without prior screening) (Jaywant et al., 2021; Marks et al., 2021). Although occupational therapy practitioners recognize the distinction between neurocognitive screening and assessment, such a clear distinction is not maintained in everyday practice. The distinction is blurred further between functional cognitive screening and assessment such that a specific functional cognitive measure may be used to both screen and assess (Jaywant et al., 2021). Functional activities require the integration of cognitive abilities in dynamic real-world environments. Functional cognitive measures provide information regarding a person’s capacity to plan, self-initiate, sequence a complex behavior, and terminate an activity, and it may also assess self-awareness of performance in real time: See Giles et al. (2025) for elaboration on the definition of functional cognition. Despite evidence that performance-based functional cognitive measures may more accurately represent real-world performance than other forms of measurement (Jekel et al., 2015), the use of standardized functional cognitive measures by occupational therapy practitioners remains relatively new.
Published research is limited on occupational therapists’ clinical practice patterns related to cognitive screening and assessment. Most studies have been conducted outside of North America (Goodchild et al., 2023; Manee et al., 2020; Stigen et al., 2018; Ward et al., 2024). Thus, the degree to which occupational therapy practitioners in the United States use neurocognitive and functional cognitive measures is largely unknown. Previous surveys indicate that many therapists rely on unstandardized evaluation approaches such as interviews, clinical judgment, or skilled observation of functional task performance (Friend et al., 2025; Goodchild et al., 2023; Manee et al., 2020). How prevalent this approach to evaluation is in the United States is unclear, now that the American Occupational Therapy Association (AOTA) has established the evaluation of functional cognition as a priority for the profession across areas of practice. The aim of this study is to describe current practice for the evaluation of cognitive function in adults across occupational therapy treatment settings. Furthermore, we aimed to determine whether unstandardized measures are used more frequently than standardized measures and whether neurocognitive measures are used more frequently than functional cognitive measures.
Method
Design
A cross-sectional survey of occupational therapy practitioners working in adult treatment settings examined current practice for evaluating cognition. The study was approved by the University of Missouri Institutional Review Board.
Respondents
Licensed occupational therapy practitioners with at least 6 mo of experience working in adult treatment settings were eligible to respond to the survey. The online survey was distributed through the AOTA CommunOT online discussion board, occupational therapy social media groups, professional networks of cognition researchers, word of mouth, and flyer distribution. The survey remained open from September 2022 to September 2023. Respondents provided informed consent electronically and received a $5 electronic gift card.
Survey Development
The study authors have prior experience in survey design. All authors reviewed previously published surveys examining cognitive assessment to ensure our questions were pertinent to previously reported issues in clinical practice (Friend et al., 2025; Goodchild et al., 2023; Manee et al., 2020). After the review of published surveys, the study authors developed a draft survey regarding ways in which practitioners evaluate cognition in clinical practice. Occupational therapy practitioners who are known to the authors and practicing in diverse settings piloted the survey and shared detailed feedback on the items and response options. The survey questions and instructions were refined, resulting in minor editorial, item wording, and grammatical changes, before implementation.
The respondents’ demographic characteristics were collected in the first part of the survey (e.g., years of practice, degree). The next section of the survey had respondents rate how often they used various methods to evaluate cognition, on a sliding scale ranging from 0% to 100% of the time. In the survey questions, we distinguished between neurocognitive screening and assessment; however, we did not distinguish between functional cognitive screening and assessment, because functional cognitive measures may be used as either screens or assessments. Respondents then indicated (in a yes-or-no format) whether they typically used specific measures from a comprehensive list; the frequency of typical use is reported. The various measures included in the survey were identified by a literature search. Respondents also rated the factors that influenced their choices, perceived barriers to use, and perceived workplace support for standardized screening and assessment using a Likert scale (on which 1 = not at all, 2 = to a minor degree, 3 = to a moderate degree, and 4 = to a substantial degree).
Data Collection
The survey was administered electronically through Research Electronic Data Capture (REDCap) software (Harris et al., 2019). Respondents provided their license number, which was cross-checked by research assistants with publicly available records to identify invalid survey responses (e.g., from survey bots). License numbers were used for screening purposes only and were deleted before the creation of the anonymized dataset.
Data Analysis
All survey responses were inspected for plausibility and responses deemed by author consensus (AB, GG, TM) to result from question ambiguity or misinterpretation were removed. We computed descriptive statistics for continuous variables and frequency distributions for categorical variables. To allow for statistical comparison using χ2 analysis, sliding-scale responses were categorized as never, 1% to 25% of the time, 26% to 50% of the time, 50% to 75% of the time, and 76% to 100% of the time. A χ2 test of independence was used to determine whether there were significant differences in the distributions relevant to the research aims. We used separate pairwise χ2 analyses to compare the distributions between the use of clinical judgment and the use of neurocognitive screening tools, distributions between the use of clinical judgment and the use of assessments of specific cognitive domains, distributions between the use of clinical judgment and the use of functional cognitive assessment, and distributions between the use of neurocognitive screening and the use of functional cognitive assessment. To calculate valid χ2 statistics, we combined the categories of Never and 1%–25% of the time into one category of 0% to 25%. For ease of presentation, Likert scale responses were combined into two categories: not at all–minor degree or moderate–substantial degree (Figure 1).
Factors, workplace supports, and barriers that affect the choice of standardized screening and assessment measures (N = 446).
In separate analyses, we categorized respondents’ endorsement of the use of a list of standardized measures as either neurocognitive or functional cognitive on the basis of literature review, information provided in test administration manuals, and consultation with topic experts. We categorized measures as neurocognitive if they measured one or more discrete cognitive functions such as attention, memory, or executive function, and we categorized measures as functional cognitive if they used a simulated basic or instrumental everyday activity to evaluate cognitive abilities at the performance level (for the categorization of measures, see Table A.1 in the Supplemental Material, available online with this article at https://research.aota.org/ajot). The top five typically used neurocognitive measures and functional cognitive measures were calculated. To further describe the relationship between typical standardized measurement use (neurocognitive and functional cognitive) of our respondents we computed a cross-tabulation between respondents who used and who did not use neurocognitive measures and those who used and who did not use functional cognitive measures (i.e., some respondents used both, and some used neither). On the basis of endorsing the typical use of one or more measures from the list provided, respondents were categorized as a user of neurocognitive measures, a user of functional cognitive measures, a user of neurocognitive and functional cognitive measures, or as not using either type of measure. All analyses were performed using IBM SPSS Statistics (Version 29).
Results
We collected and analyzed 446 complete survey responses, with 91.3% (n = 407) of the respondents being occupational therapists and 8.7% (n = 39) being occupational therapy assistants. Respondents work with a wide variety of client populations in their primary work settings. More than half of the respondents work with neurologic, orthopedic, postsurgical, and general medical conditions (see Table 1).
Respondent Characteristics (N = 446)
Note. EdD = doctor of education; OTD = doctor of occupational therapy; PhD = doctor of philosophy.
The distribution and percentages of the frequencies of use of different cognitive evaluation methods of the of 446 respondents are shown in Table 2. The χ2 analysis showed a statistically significant difference between the distribution of frequencies of using clinical judgment and frequency of using neurocognitive screening, χ2(9) = 52.39, p < .001. The percentage of clinicians who report using clinical judgment to evaluate cognition is higher than the percentage of clinicians who use neurocognitive screening. The χ2 analysis showed a statistically significant difference between the distribution of frequencies of using clinical judgment and frequency of using standardized assessments of specific neurocognitive domains, χ2(9) = 80.85, p < .001. Similarly, the percentage of clinicians who report using clinical judgment to evaluate cognition is higher than the percentage of clinicians who report using standardized assessments of specific neurocognitive domains. There was also a statistically significant difference between the distribution of frequency of using clinical judgment to evaluate cognition and the frequency of using standardized functional cognitive assessment, χ2(9) = 77.49, p < .001. Similarly, the percentage of clinicians who report using clinical judgment to evaluate cognition is higher than for those who report using functional cognitive assessments. There was a statistically significant difference between the distribution of frequency of using neurocognitive screening and frequency of using standardized functional cognitive assessment χ2(9) = 68.63, p < .001. Similarly, the percentage of clinicians who report using neurocognitive screening to evaluate cognition is higher for those who report using functional cognitive assessments.
Frequencies of Using Different Methods of Cognitive Evaluation (N = 446)
aItem included in χ2 analyses examining distribution of the frequencies of use of different cognitive evaluation methods.
Table 3 shows the five most used neurocognitive and functional cognitive measures. The most common neurocognitive measure was the MoCA (25.6%), followed by the SLUMS (25.3%) and the Clock Drawing Test (Rouleau et al., 1992; 21.3%). Respondents indicated that the ACLS (15.2%) was their most commonly used functional cognitive measure, followed by the Cognitive Performance Test (Burns, 2018; 6.5%) and the Assessment of Motor and Process Skills (Fisher & Jones, 2014; 6.1%). Therefore, occupational therapy practitioner respondents rely more on neurocognitive measures than on functional cognitive measures to evaluate their clients.
Top Five Neurocognitive and Top Five Functional Cognitive Measures Typically Used by the Respondents (N = 446)
On the basis of the individual measures that respondents reported typically using, we developed a cross-tabulation table and assigned each respondent into one of four groups on the basis of their typical use of standardized measures. The two largest groups were respondents who reported using neurocognitive measures alone (n = 189; 42.4%) and respondents who reported using both neurocognitive and functional cognitive measures (n = 197; 44.2%). A smaller number of occupational therapy practitioners reported the use of functional cognitive measures alone (n = 30; 6.7%; i.e., no use of neurocognitive measures) or reported no use of either type of measure (n = 30; 6.7%).
The three most common factors influencing the use of standardized screening and assessment measures were “time required to use the measure” (85.2%), “appropriateness for client needs” (79.6%), and “availability of the measure” (77.8%; see Figure 1). The most common barriers that limited the use of standardized measures were “cost of measures” (73.3%), “lack of resources” (69.5%), and “lack of knowledge of use and interpretation” (66.6%). The most reported workplace support was “ongoing peer or administrative support” (56.5%), and the least common workplace support was “training on how to perform the measure” (40.1%).
Discussion
The main goal of this study was to describe current practice for evaluating cognitive function in adults across occupational therapy treatment settings. In general, we found that standardized measures are not widely used. Although the survey included questions regarding a variety of unstandardized methods that clinicians use to evaluate cognition, we chose the use of clinical judgement as representative of unstandardized methods of evaluation, and we chose neurocognitive screening, assessment of specific neurocognitive domains, and functional cognitive assessment as representative of standardized methods of evaluation. It is important to note that unstandardized methods to evaluate cognition are used more frequently than standardized measures (neurocognitive and functional cognitive) and that respondents use neurocognitive screening far more often than functional cognitive assessments. The MoCA was the most reported neurocognitive measure, and the ACLS was the most reported functional cognitive measure.
Despite AOTA’s adoption of functional cognition as a focus for the profession, our results are consistent with previous reports of persistent use of unstandardized methods to evaluate cognition, as well as widespread reliance on neurocognitive screening rather than the use of functional cognitive measures across occupational therapy practice settings (Friend et al., 2025; Goodchild et al., 2023; Manee et al., 2020). AOTA clearly promotes practitioner use of evidence-based, standardized, and/or structured assessment tools and protocols during the screening, evaluation, and reevaluation process (AOTA, 2018). Nonetheless, we found that this is not the current standard of practice. Occupational therapy practitioners believe that unstandardized observations are an essential part of evaluating functional cognition. This approach is used because practitioners believe that this approach is less intimidating for clients and more readily accommodating of motor or language deficits (Ward et al., 2024). However, it has been widely demonstrated that unstandardized methods alone are inadequate to identify subtle but clinically important cognitive changes (Belchior et al., 2015; Edwards et al., 2006). Furthermore, unstandardized observations are not a reliable or validated method of measurement because unstandardized observations are fully dependent on the expertise of the therapist completing the observation. It has also been found that the types of tasks that are usually observed during simplified hospital routines do not evoke the cognitive skills that are essential for real-world functional performance (Morrison et al., 2015). Standardized methods of screening and assessment empirically anchor the information derived from unstandardized observations, and other sources of information (e.g., chart review, client interview, and caregiver reports) and together provide a comprehensive evaluation of the client’s capacity and needs for further treatment or environmental support.
Many factors may account for the apparent preference for use of neurocognitive screening measures, with the MoCA and the SLUMS both reported as being typically used by about a quarter of the respondents. These and many other validated neurocognitive measures have been available for a long time and are used by a variety of health care disciplines (Lezak et al., 2012). Occupational therapy practitioners may feel pressured to administer neurocognitive screens, because the tests and resulting scores are more familiar to the health care teams, despite evidence of limited predictive and ecological validity for complex task performance (Ward et al., 2024; Boone et al., 2025). Additionally, most neurocognitive screens are readily available (cited as the principle reason for choosing a cognitive assessment internationally; Manee et al., 2020), are free to use, and generally consist of paper-and-pencil administration materials. We found that 42.4% of our sample typically used neurocognitive measures but not functional cognitive measures. Although neurocognitive measures are an important source of clinically relevant information, they are only modestly associated with real-world functioning and are not recommended for use in isolation to predict everyday function (McAlister et al., 2016). We strongly encourage occupational therapy practitioners to adopt a combination of bottom-up and top-down approaches to obtain a complete understanding of cognitive capacity to perform complex occupational activities.
Functional cognitive measures may provide additional information, such as strategy use and self-awareness of performance, over information that was derived from neurocognitive measures (Marks et al., 2024). Functional cognitive measures that include performance-based testing in context evaluate a key domain of occupational therapy practice (Ward et al., 2024; Boone et al., 2025). Nonetheless, we found minimal use of any functional cognitive measure. Our findings are consistent with the results of prior surveys in similar adult practice settings (Friend et al., 2025; Goodchild et al., 2023). The most frequently reported functional cognitive measure, the ACLS, was only used 15.2% of the time. The ACLS is one of the first standardized screening measures of functional cognition developed to predict general functional abilities (Allen, 1985). No other functional cognitive measure listed in the survey was used more than 7% of the time. In comparison with neurocognitive screening, functional cognitive measures are seen as more challenging and time consuming to administer and have greater environmental barriers (e.g., the need for a kitchen) that limit their use in acute care and inpatient practice settings (Boone et al., 2025; Friend et al., 2025; Goodchild et al., 2023). Wider dissemination of functional cognitive screening measures that are highly portable may reduce this barrier. Additionally, occupational therapy managers could incentivize or provide practitioners with promotional ladders dependent on meeting evidence-based practice standards (Lin et al., 2010). It is important to note that AOTA could endorse a set of preferred functional cognitive measures, similar to the way the physical therapy profession has adopted preferred measures for physical functioning, for both clinical practice and educational curriculum (Academy of Neurologic Physical Therapy, n.d.). Currently, the occupational therapy profession has not developed a critical mass around teaching and learning of specific measures, as indicated by the low adoption into the discipline of any specific functional cognitive measure.
Although standardized measures offer substantial advantages over unstandardized measures, they are often more complex to administer and interpret. Lack of easy access and the time needed to administer and score the measures were seen by respondents as limiting factors when choosing a cognitive measure, echoing results from recent surveys (Friend et al., 2025; Goodchild et al., 2023; Manee et al., 2020). Time and availability are the most common reasons cited by the respondents for the low use of cognitive measures. Recent focus has been placed on developing functional cognitive screening measures that could be applied in a variety of settings with simple administration and scoring procedures (Al-Heizan et al., 2022; Arieli et al., 2022; Zartman et al., 2013). The survey identified key barriers to standardized measurement use, such as cost and lack of resources. Another obstacle reported by the respondents was the perceived complexity of the measures. The respondents reported their feeling of being unqualified to administer and interpret measures as a significant obstacle. One solution to this barrier may be employer- or practitioner-based training, which, in another survey, was associated with occupational therapy practitioners’ actual use of both neurocognitive and functional cognitive measures (Friend et al., 2025). Respondents indicated that money to cover the costs of measures and training on how to administer the measures were the least offered workplace supports.
Regarding the limitations of this study, the survey was distributed on general discussion boards and sent to e-mail lists that included therapists who work with adults with cognitive impairments, and there is a chance of selection bias because those who responded may be more interested and more aware of cognitive evaluation methods than nonresponders. Additionally, we were unable to formally generate a response rate because we posted the survey on publicly available discussion boards. Another limitation of this study is that we did not provide pictures or definitions of the measures we listed, and there is a chance that therapists did not recognize the name of a measure. Therefore, some measures may have been selected accidentally by respondents because of the name of the measure having the same name as common functional tasks. Although we attempted to include a comprehensive list of measures in the survey, several measures were not included in the list, and this may have affected the degree to which a specific test was endorsed; for example, only 14 (3.1%) of the respondents wrote in the Brief Interview of Mental Status (Saliba et al., 2012), which has been mandated for use across all post–acute care settings by the Centers for Medicare and Medicaid Services. Thirty-nine responses were occupational therapy assistants, and their role in cognitive evaluation varies by state. The Medi-Cog was listed as a single measure without differentiating its paper-and-pencil and practical variants. We categorized it as functional cognitive in our list of top measures because of its practical variant and believe that this is the most widely used form among occupational therapy practitioners, but we failed to distinguish between the two variants in our survey, marginally increasing the risk of inflating the use of functional cognitive measures in the findings.
Implications for Occupational Therapy Practice
The findings of this study have the following implications for occupational therapy practice: ▪ Comprehensive cognitive evaluation should include data from unstandardized observations, and from neurocognitive and functional cognitive measures to create a full occupational profile. ▪ Occupational therapy practitioners need increased education and workplace training and support to implement standardized methods to evaluate functional cognition.
Conclusion
The survey results indicate that occupational therapy practitioners use a variety of methods to evaluate cognitive function. Unstandardized observation was the most frequently used method to evaluate cognition, and neurocognitive measures were more commonly used than functional cognitive measures. Respondents identified factors influencing their choice of cognitive measures and barriers to their use. The overall low rates of support and resources available across treatment settings may suggest opportunities to increase evidence-based standardized measurement use.
Supplemental Material
Supplementary material for A National Survey of Occupational Therapy Practitioners’ Evaluation of Cognition in Adults
Supplementary material, sj-pdf-1-aot-10.5014_ajot.2025.051113.pdf for A National Survey of Occupational Therapy Practitioners’ Evaluation of Cognition in Adults by Anna E. Boone, Timothy S. Marks, Timothy J. Wolf, Gordon M. Giles and Dorothy Farrar-Edwards in The American Journal of Occupational Therapy
Footnotes
Acknowledgments
This research has been supported in full by an American Occupational Therapy Foundation Implementation Research Grant (AOTFIR22Boone).
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
