Who Shall Receive Community-Based Occupational Therapy Services? A Logistic Regression Analysis

PURPOSE: Timely and appropriate referral for occupational therapy (OT) services is critical for optimizing clients’ functioning and successfully integrating them into the community. We aimed to establish a triage protocol by exploring correlates of long-term care referral for more effective and efficient community-based OT.

DESIGN: A correlational research design was applied. Data were gathered via a cross-sectional survey from eight urban and rural areas in Taiwan. Proportional sampling was conducted to reflect the composition of mentally and physically disabled adults in the surveyed cities and counties.

METHOD: Data were collected in face-to-face interviews using the Multidimensional Assessment Instrument (MDAI), a comprehensive long-term care needs assessment tool developed by the Ministry of Health and Welfare as a potential screening tool for Taiwanese long-term care insurance. Participants were 221 community-dwelling participants age 18 yr and older who reported disabilities in the previous national survey of long-term care in Taiwan (2011). Experienced occupational therapists reviewed the results of the MDAI (chart review) and made referral decisions. Thirty-four participants were also randomly selected and visited on-site by an experienced occupational therapist and a field interviewer. The OT referral decisions from these in-person visits were used to validate those from the chart reviews. Logistic regressions and generalized additive models were used to explore the predictors of the community OT service referral.

RESULTS: Of the 221 participants, 171 (77.4%) were recommended for OT intervention. They were more likely to be older, to be widowed or single, and to have less than a bachelor’s degree. They also had more multiple disorders; were more likely to have impaired vision, hearing, and cognitive function; and were more likely to be limited in basic activities of daily living (BADLs) and instrumental activities of daily living (IADLs). Those who were recommended for OT were more likely to have limited ambulation (37.8%), limited joint range of motion (ROM; 46.8%), poor balance function (∼31%), environmental barriers at home (22.5%), and a history of fall within 6 mo (32.0%). They had more severe behavioral and emotional problems, and their caregivers also had significantly higher caregiver strain.

Two predictive models of OT referral, one based on BADLs and one on IADLs as the major predictor, were proposed. Limitations in either IADLs or BADLs, dementia, psychiatric disorders, cognitive impairment, joint ROM limitations, fear of falling, behavioral or emotional problems, and expressive deficits (only in the BADL-based model) were significantly correlated with community-based OT referral. Both predictive models showed high area under the curve (AUC) values on receiver operating curve testing (AUC = .977 for BADLs, .972 for IADLs).

CONCLUSION AND IMPACT STATEMENT: Two predictive models of OT referral were proposed as triage protocols for community decision making. They could help in facilitating communication among health care professionals, such as case managers, to make appropriate decisions for OT referrals. On the policy level, it can be a reference model for third-party payers to validate the appropriateness of referrals. The models can also help occupational therapists focus the major factors of decision making in a shorter time frame. The methods and findings should be useful for developing referral protocols for other long-term care services.

References

Hobbs, J. A., Boysen, J. F., McGarry, K. A., Thompson, J. M., & Nordrum, J. T. (2010). Development of a unique triage system for acute care physical therapy and occupational therapy services: An administrative case report. Physical Therapy, 90, 1519–1529. https://doi.org/10.2522/ptj.20090166