2011 Annual Conference

Welsh, Rod

The purpose of this study was to identify patient and clinical factors most strongly associated with consultation for physical therapy (PT) services among older patients hospitalized for acute illness.  

233 patients 65 years and older admitted to an Acute Care for Elders (ACE) unit were studied; 102 received a PT consult sometime during their stay.

All patients wore a Step Activity Monitor from admission to discharge.  Logistic Regression was used to examine the association between the dependent physical therapy consult variable (yes vs. no) and the independent clinical and demographic variables. To further examine the association between length of stay and likelihood of receiving a PT consult, the estimated probabilities of PT consultation were calculated from an unadjusted logistic regression model and plotted over the entire distribution of lengths of stay.

Patients who received a PT consult were more likely to be older (p=.002), have mobility impairments prior to admission (p=.001), have a neurologic or orthopedic diagnosis (p<.001), and have longer lengths of stay (p<.001). In the fully adjusted multivariate model, age (OR=1.07; 95% CI=1.02–1.11), length of stay (OR=1.27; 95% CI=1.10–1.46) use of a cane or walker prior to admission (OR=2.12; 95% CI, 1.04–4.33), and an orthopedic (OR=19.38; 95% CI=2.14–175.3) or neurologic diagnosis (OR=3.98; 95% CI=1.01–15.78) remained significantly associated with PT consultation. The longer the length of stay the greater the odds of receiving a PT consult. At length of stay=6 days, the odds of receiving a PT consult were approximately 50:50; at 11 days, the probability increased to over 90%.

Physicians consult PT for the most vulnerable subset of older patients during acute hospitalization but consults may be occurring late during the patient’s stay. Further research is needed into the timing of PT consultations during hospitalization.