Evaluating initiatives to reduce seclusion and restraint

J Healthc Qual. 2007 Jul-Aug;29(4):46-55. doi: 10.1111/j.1945-1474.2007.tb00205.x.

Abstract

The use of institutional measures of control such as seclusion and restraint within psychiatric hospitals is common and arguably countertherapeutic; however, little is known about how best to reduce the use of these measures. The development and implementation of new institutional strategies to reduce the use of seclusion and restraint are important. Although traditional performance improvement (PI) project methodology might seem well-suited to helping managers and administrators identify effective hospital-wide interventions to decrease seclusion and restraint rates, the Logic of the standard PI model precludes managers from making valid inferences about which interventions actually cause change. This article presents a model (derivative of the multiple baseline time-series design with randomization) for testing individual elements of a Large-scale PI project to reduce the use of seclusion and restraint in a behavioral healthcare organization. The proposed model is flexible, accommodates overlapping organizational initiatives, and simultaneously allows for meaningful inferences to be made about the active components of the interventions. The ability to make meaningful inferences is important because, if the initiatives to reduce seclusion and restraint rates work, other healthcare organizations would benefit from knowing Key Words which specific interventions actually Lead engagement model to change and which interventions have multiple baseline design Little impact on secLusion and restraint performance improvement rates. Early experiences with this model psychiatry from a hospital manager's perspective sanctuary trauma are discussed, along with the costs and benefits of using it.

MeSH terms

  • Evaluation Studies as Topic*
  • Hospitals, Psychiatric*
  • Humans
  • Models, Theoretical
  • Patient Isolation / statistics & numerical data*
  • Restraint, Physical / statistics & numerical data*
  • United States