Identifying Homogeneous Outcome Groups in Adult Rehabilitation Using Cluster Analysis

Adults receiving occupational therapy in inpatient rehabilitation are a heterogeneous population with differing needs, outcomes, and rehabilitation processes. Outcome studies based on what works for all clients may obscure the needs of population subgroups who benefit unequally from rehabilitation services.

To identify subgroups on the basis of client satisfaction and progress in functional self-care among a diverse rehabilitation population and to understand subgroup differences in occupational therapy and rehabilitation processes and client discharge status.

Using an existing dataset, we used K-means cluster analysis of demographics, status at admission, and the outcomes of satisfaction and change in self-care to identify five homogeneous outcome groups. Occupational therapy and rehabilitation processes and discharge status were compared across subgroups.

Inpatient rehabilitation hospital.

The dataset included 1,099 inpatients age 18 yr and older who received occupational therapy over a 27-mo period.

Admission measures included the Inpatient Rehabilitation Facility-Patient Assessment Instrument and self-care items of the FIM™. The Satisfaction with Continuum of Care-Revised was administered after discharge.

Five subgroups showed statistically different patterns of medical complications, functional self-care, rates of progress, satisfaction with intervention, and course of treatment. The profile of each group suggests differing therapeutic needs. Although all groups made significant gains in functional self-care, two groups continued to need physical assistance at discharge.

and Relevance: Cluster analysis proved useful in segmenting a typical heterogeneous rehabilitation population into more homogeneous subgroups to enhance understanding of clinical needs and to potentially increase the potency of outcomes research.

This research identified subgroups within a typical population of rehabilitation clients receiving occupational therapy and identified their unique needs and outcomes using cluster analysis techniques.