Falls risk score removal does not impact inpatient falls: A stepped-wedge, cluster-randomised trial.

AIMS AND OBJECTIVES: To investigate the impact of removing a falls risk screening tool from an overall falls risk assessment programme on the rate of falls, injurious falls and completion of falls prevention activities by staff. BACKGROUND: Falls in older patients are common adverse events in hospital settings. Screening and assessing individual patients for risk of falls are a common, but controversial element of falls prevention strategies in hospitals. DESIGN: A stepped-wedge, cluster-randomised controlled trial using a disinvestment approach. METHODS: This trial was carried out according to the Consolidated Standards of Reporting Trials (CONSORT). All patients were admitted to 20 health service wards (9 units) over the 10-month study period. The control condition contained a falls risk screening tool element, a full falls risk factor assessment and intervention provision section. In the intervention condition, only the full falls risk factor assessment and intervention provision section was applied, and the falls risk screening tool element was removed. Fall rates were extracted from hospital level data, files were audited for tool completion, and nurses surveyed about tool use. RESULTS: There did not appear to be an impact on the falls rate per month when the risk screening tool component was removed (incidence rate ratio (IRR) = 0.84-favours intervention, 95%CI = 0.67 to 1.05, p = .14) nor on the falls rate with serious injury (IRR = 0.90, 95%CI = 0.26 to 3.09, p = .87). There was a thirty-six second reduction of time per patient reported by staff to complete paperwork (p < .001). There was no difference in the proportion of patients for whom the tool was completed, nor the number of falls prevention interventions identified for implementation. CONCLUSION: Removing the falls risk screening tool section did not negatively impact falls and reduced time spent completing falls prevention paperwork. RELEVANCE TO CLINICAL PRACTICE: Falls prevention is an important issue in health services. Removal of a screening risk tool is unlikely to impact falls. This has the potential to reduce nursing administration time that may be otherwise redirected to individual approaches to falls prevention.

Mobilisation alarm triggers, response times and utilisation before and after the introduction of policy for alarm reduction or elimination: A descriptive and comparative analysis.

BACKGROUND: Mobilisation alarms are a falls prevention strategy used in hospitals to alert staff when an at risk patient is attempting to mobilise. Mobilisation alarms have an estimated annual cost of $AUD58MIL in Australia. There is growing evidence from randomised controlled trials indicating mobilisation alarms are unlikely to prevent falls. AIM: The primary aim of this study was to describe the rate of mobilisation alarm false triggers and staff response time across different health services. The secondary aim was to compare pre to post mobilisation alarm utilisation following the introduction of policy to reduce or eliminate mobilisation alarms. METHODS: This descriptive and comparative study was conducted through Monash Partners Falls Alliance across six health services in Melbourne, Australia. This study described true and false alarm triggers and trigger response times across three health services and usual care mobilisation alarm utilisation across six health services; and then compared alarm utilisation across two health services following the introduction of policy to reduce (<2.5%) or eliminate (0.0%) mobilisation alarms in the acute and rehabilitation settings. RESULTS: The most frequent observation was a false alarm (n = 74, 52%), followed by a true alarm (n = 67, 47%) and no alarm (n = 3, 2%). Time to respond to the true and false alarms was an average of 37 seconds (SD 92) and this included 61 occasions of 0 seconds as a member of staff was present when the alarm triggered. If the 61 occasions of staff being present when the alarm triggered were removed, the average time to respond was 65 seconds (SD114). Usual care mobilisation alarm utilisation in acute was 7% (n = 171/2,338) and in rehabilitation was 11% (n = 286/2,623). Introducing policy for reduced and eliminated mobilisation alarm conditions was successful with a reduced utilisation rate of 1.8% (n = 11/609) and an eliminated utilisation rate of 0.0% (n = 0/521). CONCLUSION: Half of mobilisation alarm triggers are false and when alarms trigger without staff present, staff take about a minute to respond. While usual care has one in fourteen patients in acute and one in nine patients in rehabilitation using a mobilisation alarm, it is possible to introduce policy which will change practice to reduce or eliminate the use of mobilisation alarms, providing evidence of feasibility for future disinvestment effectiveness studies that it is feasible to disinvest in the alarms.