ORCID

https://orcid.org/https://orcid.org/0000-0002-1384-657X

Date of Award

2026

Document Type

Dissertation

Degree Name

Psychology (Ph.D.)

Department

Psychology

First Advisor

Elizabeth Brondolo

Second Advisor

William Chaplin

Third Advisor

Melissa Peckins

Abstract

Research on the use of physical restraints has revealed high rates of heterogeneity between clinical settings and an array of potential risk factors at both the patient and provider level. This is most evident in general medical hospitals, where patient populations are more heterogenous than other settings and physicians often can rely only on observable characteristics of patients to assess safety risks. More recent work has shown that coherent subgroups of restrained patients can be derived from demographic and clinical indicators. Identification of coherent patient subgroups may aid in understanding both the heterogeneity of the restrained patient population and differences between providers in the decision to use restraints. The current study proceeded in three stages. First, we derived subgroups from a sample of all encounters to a general medical hospital over one year, replicating findings from previous studies. Second, we used these subgroups to populate vignettes which were presented to a sample of 23 physicians, asking them to rate each vignette both in terms of the likelihood that they would use restraints and the likelihood that they assumed others would use restraints. We found evidence for distinct decision-policies between physicians that varied in terms of the knowledge applied in the decision, the consistency of their decisions, and the accuracy of their decisions in determining if the patient described in the vignette had actually been restrained. The results of our study are discussed, along with implications for future research. Major limitations of the current study include a lack of data on nurses, who are vital stakeholders in the decision to restrain, and concerns with missingness of outcome data for 5/23 physicians.

Share

COinS