Unveiling the Impact of Occupational Therapy on Acute Care Outcomes: A Machine Learning Approach

Area of Science:

• Healthcare research
• Rehabilitation medicine
• Health services research

Background:

• Occupational therapy (OT) is crucial for patient recovery and functional independence.
• Understanding the optimal delivery of OT services is essential for improving patient outcomes.
• Limited resources necessitate efficient allocation of OT interventions.

Purpose of the Study:

• To investigate the relationship between occupational therapy (OT) service delivery models and patient discharge outcomes.
• To identify how the quantity and frequency of OT sessions influence readmission rates and functional status post-discharge.

Main Methods:

• Retrospective cohort study analyzing data from 36,300 patients at Cleveland Clinic (2017-2021).
• Decision tree algorithms were used to assess the impact of total OT visits, total OT minutes, and OT visit frequency on outcomes.
• Included patients with orthopedic conditions, hospitalized for less than 30 days, receiving at least one OT session.

Main Results:

• Prediction models demonstrated good to excellent performance (accuracy 0.69-0.91).
• Variations in the number, duration, and frequency of OT sessions were associated with 30-day readmission rates.
• Minimal detectable improvements in daily activities were observed for specific patient subgroups with orthopedic conditions.

Conclusions:

• The effective utilization of limited occupational therapy (OT) resources is key.
• Prioritizing certain patient subgroups for more intensive or frequent OT may enhance outcomes.
• Optimizing OT service delivery can improve patient discharge and functional recovery.