Prévention des chutes dans les hôpitaux : une revue systématique de la mise en œuvre, des composantes, de l'observance et de l'efficacité.

Résumé (en anglais)

OBJECTIVES: To systematically document the implementation, components, comparators, adherence, and effectiveness of published fall prevention approaches in U.S. acute care hospitals.

DESIGN: Systematic review. Studies were identified through existing reviews, searching five electronic databases, screening reference lists, and contacting topic experts for studies published through August 2011.

SETTING: U.S. acute care hospitals.

PARTICIPANTS: Studies reporting in-hospital falls for intervention groups and concurrent (e.g., controlled trials) or historic comparators (e.g., before-after studies).

INTERVENTION: Fall prevention interventions.

MEASUREMENTS: Incidence rate ratios (IRR, ratio of fall rate postintervention or treatment group to the fall rate preintervention or control group) and ratings of study details.

RESULTS: Fifty-nine studies met inclusion criteria. Implementation strategies were sparsely documented (17% not at all) and included staff education, establishing committees, seeking leadership support, and occasionally continuous quality improvement techniques. Most interventions (81%) included multiple components (e.g., risk assessments (often not validated), visual risk alerts, patient education, care rounds, bed-exit alarms, and postfall evaluations). Fifty-four percent did not report on fall prevention measures applied in the comparison group, and 39% neither reported fidelity data nor described adherence strategies such as regular audits and feedback to ensure completion of care processes. Only 45% of concurrent and 15% of historic control studies reported sufficient data to compare fall rates. The pooled postintervention incidence rate ratio (IRR) was 0.77 (95% confidence interval = 0.52-1.12, P = .17; eight studies; I(2) : 94%). Meta-regressions showed no systematic association between implementation intensity, intervention complexity, comparator information, or adherence levels and IRR.

CONCLUSION: Promising approaches exist, but better reporting of outcomes, implementation, adherence, intervention components, and comparison group information is necessary to establish evidence on how hospitals can successfully prevent falls.