Related Experiment Video
Updated: Mar 26, 2026
07:35
Author Spotlight: Rehabilitation of Stroke Patients With a Digital Occupational Training System
Published on: December 29, 2023
2.3K
Developing a Referral Protocol for Community-Based Occupational Therapy Services in Taiwan: A Logistic Regression
Hui-Fen Mao1, Ling-Hui Chang2,3, Athena Yi-Jung Tsai4
1School of Occupational Therapy, College of Medicine, National Taiwan University, Taipei, Taiwan.
Plos One
|February 11, 2016
Summary
Identifying key client characteristics is crucial for timely occupational therapy (OT) referrals in community-based long-term care. This study developed predictive models to improve rehabilitation service access for individuals with disabilities.
Area of Science:
- Rehabilitation Medicine
- Occupational Therapy
- Health Services Research
Background:
- Limited resources necessitate efficient referral systems for long-term care.
- Optimizing client function and community integration requires timely rehabilitation services.
- Understanding predictors for community-based occupational therapy (OT) referral is essential.
Purpose of the Study:
- To identify client characteristics predicting referral for community-based OT services in Taiwan.
- To develop and validate predictive models for OT referrals based on expert opinion and client assessments.
- To establish an evidence-based algorithm for calculating referral probability.
Main Methods:
- Face-to-face interviews using the Multidimensional Assessment Instrument (MDAI) with 221 community-dwelling adults with disabilities.
- Expert judgment from two experienced occupational therapists served as the referral standard.
- Logistic regressions and Generalized Additive Models were employed to develop predictive models based on basic and instrumental activities of daily living (BADLs/IADLs).
Main Results:
- Dementia, psychiatric disorders, cognitive impairment, joint limitations, fear of falling, behavioral issues, and expressive deficits were significant predictors.
- Two predictive models (BADL-based and IADL-based) demonstrated high accuracy (AUC = 0.977 and 0.972).
- A referral algorithm was developed to calculate the probability of needing community OT services.
Conclusions:
- The study successfully identified key factors influencing OT referral decisions in Taiwan.
- The developed predictive models and referral algorithm can enhance the efficiency and appropriateness of OT service allocation.
- These findings support the development of standardized referral protocols for community-based long-term care services.
