# Exploring the Impact of Safewards on Aggression and Coercion in Psychiatric Inpatient Care: Findings From a Swedish Longitudinal Quasi‐Experimental Trial

**Authors:** Jenny Karlsson, Anna Björkdahl, Nina Gårevik, Lars Kjellin, Veikko Pelto‐Piri, Naimi Johansson

PMC · DOI: 10.1111/inm.70228 · International Journal of Mental Health Nursing · 2026-02-02

## TL;DR

This study examined the Safewards model's impact on reducing aggression and coercion in psychiatric care in Sweden, finding limited effectiveness but some promising trends.

## Contribution

The study provides new insights into the Safewards model's implementation and outcomes in a Swedish psychiatric context.

## Key findings

- No statistically significant reductions in coercive measures or aggressive incidents were observed.
- A significant increase in normalisation of the Safewards model was found.
- A declining trend in mechanical restraint use was noted despite non-significant results.

## Abstract

Coercive measures in psychiatric inpatient care remain controversial and are often associated with negative experiences for both patients and staff. The Safewards model aims to reduce conflict and containment by fostering a safer and more therapeutic ward environment. However, evidence regarding its effectiveness is mixed. This study investigated the implementation and impact of Safewards in nine Swedish psychiatric inpatient wards, focusing on coercive measures, aggressive incidents, and the normalisation of the intervention. A quasi‐experimental, longitudinal design with comparison wards was used. Data were collected through administrative records on coercive measures, staff surveys for incident reports (SOAS‐R), and normalisation (S‐NoMAD). Mixed model regression analyses assessed changes over time in coercive measures. Wards implemented between two and five Safewards interventions. No statistically significant reductions were found in coercive measures or aggressive incidents. Although the effects on mechanical restraint were not statistically significant, the significant increase in normalisation and the declining trend in mechanical restraint suggest a potential shift. Partial implementation and contextual challenges likely restricted the model's full impact. The study was reported according to the TREND checklist.

## Full-text entities

- **Diseases:** Neurotic (MESH:D009497), fatigue (MESH:D005221), Schizophrenia (MESH:D012559), Aggression (MESH:D010554), Mental and behavioural disorders (MESH:D001523), self (MESH:D012652), somatoform disorders (MESH:D013001), Mood (affective) disorders (MESH:D019964), schizotypal (MESH:D012569), Intellectual disabilities (MESH:D008607), COVID-19 (MESH:D000086382), psychotic disorders (MESH:D011618), delusional disorders (MESH:D012563), trauma (MESH:D014947)
- **Chemicals:** psychoactive substances (-)
- **Species:** Homo sapiens (human, species) [taxon 9606]

## Full text

_Full body text omitted from this summary view._ Fetch the complete paper as Markdown: https://tomesphere.com/paper/PMC12865338/full.md

## Figures

3 figures with captions in the complete paper: https://tomesphere.com/paper/PMC12865338/full.md

## References

43 references — full list in the complete paper: https://tomesphere.com/paper/PMC12865338/full.md

---
Source: https://tomesphere.com/paper/PMC12865338