Differentiating Therapeutic Responses That Reduce Restrictive Practice Use and Situational Aggression in an Acute Mental Health Unit.

AIM: An analysis of mental health nursing de-escalation logs for 249 days from a regional adult inpatient unit in New South Wales, Australia, was completed to identify groups of cooccurring nursing therapeutic responses to aggression and examine their associations with reductions in restrictive practices and situational aggression.

DESIGN

A single-centre retrospective study was undertaken.

METHOD: Hierarchical clustering of nursing interventions established groups of cooccurring nursing responses. Poisson mixed-effect models were then used to determine the associations of the intervention clusters with restrictive practices.

RESULTS

Two intervention clusters emerged: Cluster 1 involved verbal de-escalation with active listening and rapport building, whereas Cluster 2 included additional limit setting and problem-solving, distraction, sensory modulation, environmental change and individual staff time. Cluster 1 was linked with a reduction in seclusion use by 83% [IRR = 0.17, 95% CI (0.07, 0.41), p < 0.001], physical restraint by 79% [IRR = 0.21, 95% CI (0.11, 0.40), p < 0.001] and average judged situational aggression by 1.56 [95% CI (0.86, 2.25), p < 0.001]. Cluster 2 was related to statistically insignificant increases in the three studied outcomes.

CONCLUSIONS

The intervention clusters prove the value of supplementary tools in surfacing nurses' therapeutic potential. The differences in restrictive practice use between intervention clusters signal the structure and progression of forming therapeutic relationships in aid of de-escalation and the possibility of assessing de-escalation components robustly.

RELEVANCE TO CLINICAL PRACTICE: Acknowledging and supporting nurses' therapeutic work support the development of recovery-oriented care and a positive professional identity for nurses.

REPORTING METHOD: This study followed the applicable STROBE guidelines.

PATIENT OR PUBLIC INVOLVEMENT: Due to the study's retrospective nature, there was no service user or public involvement.