RecruitingACTRN12625000909426

Implementation and evaluation of a dashboard of predictive analytics and decision support to drive care quality and person-centred outcomes in aged care

Implementation and evaluation of a dashboard of predictive analytics and decision support to drive care quality and person-centred outcomes in aged care: a pragmatic cluster randomised controlled trial

Sponsor

Macquarie University

Enrollment

20 participants

Start Date

Nov 11, 2024

Study Type

Interventional

Conditions

Falls Wellbeing

Summary

We will implement an intervention to improve the quality of care for residents in aged care facilities. The intervention consists of an electronic dashboard on falls and quality of life. It is intended for use by aged care staff and predicts the risk of falls and poor wellbeing and presents information, action areas and clinical evidence-based recommendations that can be inputted by staff minimize resident risk of poor health outcomes. To evaluate the dashboard we will be conducting a cluster randomised controlled trail where we will randomise 20 facilities into intervention and control groups (i.e. 10 in each group). The intervention will be introduced across all intervention sites at the same time in early 2023. A 1-month intervention wash-in period will be allowed to allow the integration of the dashboard into routine practice. Since the intervention is an add-on to an existing system, 1 month will be sufficient to allow users to familiarise themselves with the dashboard. The impacts of the dashboard will then be compared between the intervention and the control sites after 12 months (excluding the wash-in period data). We will include two additional sites for the pilot testing. The primary outcome we will look at is rate of all falls (i.e., any falls regardless of whether an injury was involved, or hospitalisation was required). We hypothesise that the intervention will reduce the rate of falls in the intervention group in comparison to the facilities in the control group. The secondary outcomes include: injurious falls, falls requiring hospitalisation, client wellbeing, social service use (attendance at leisure and lifestyle activities), hospital service use, use of the Peninsula Health Falls Risk Assessment Tool and change in use of Falls-Risk Increasing Drug use.

Eligibility

Sex: Both males and femalesMin Age: 65 Yearss

Plain Language Summary

Simplified for easier understanding

Falls are one of the leading causes of injury and hospitalisation for older people living in aged care facilities, and predicting who is at risk can be difficult for busy staff. This study tests a digital decision-support dashboard that uses predictive analytics to identify residents at risk of falls and poor wellbeing, and provides care staff with evidence-based recommendations for what to do about it. Think of it as an intelligent assistant that helps aged care workers spot problems before they happen. The dashboard will be introduced across 10 aged care facilities (with another 10 serving as a comparison group that doesn't use the dashboard). After 12 months, researchers will compare falls rates, injuries, wellbeing scores, hospital admissions, and staff use of standardised assessment tools between the two groups. This study is run by Macquarie University. This study is aimed at residents of participating residential aged care facilities (aged 65 and over) owned by a specific partner provider. Individual residents' existing health records and data will be used — residents and their families may be contacted to participate in surveys about quality of life as part of the evaluation.

This summary was AI-generated to explain the trial in plain language. It is not medical advice. Always discuss eligibility with your doctor before enrolling in a clinical trial.

Interested in this trial?

Get notified about updates and connect with the research team.

Interventions

We will conduct a two-arm, parallel-group, non-blinded, pragmatic cRCT with baseline measurement. Randomisation will be stratified by the size and location of the facilities with a 1:1 allocation ratio. The unit of randomisation will be a cluster (i.e., residential aged care facilities). The intervention sites will receive the dashboard of predictive analytics and decision support while the control sites will remain on usual care (i.e., no dashboard). The 20 study sites (10 intervention and 10 control sites) will be randomly selected from a total of 24 facilities managed by Anglicare. The intervention will be introduced across all intervention sites at the same time in Nov 2024. A 1-month intervention wash-in period will be allowed to allow the integration of the dashboard into routine practice. Since the intervention is an add-on to an existing system, 1 month will be sufficient to allow users to familiarise themselves with the dashboard. The impacts of the dashboard will then be compared between the intervention and the control sites after 12 months (excluding the wash-in period data). We will include two additional site for the pilot testing. The intervention involves the implementation of a dashboard of predictive analytics and decision support to be used by aged care staff to improve the care of residential aged care clients in relation to falls related hospitalizations and quality of life. Staff and Macquarie University researchers will embed the dashboard within the care management systems within facilities and accessed via routine avenues within the facility (i.e. computer/tablet).. The dashboard utilizes data that has already been collected within Anglicare's electronic care management system to provide information on wellbeing scores as well as incorporating a dynamic falls risk predictive tool to inform residents daily falls risk. The dashboard will be used by staff in addition to other standard electronic and paper-based forms used to provide standard care. The dashboard will also include interventions or recommended actions to take to reduce resident falls risk based on clinical guidelines and the Peninsula Health Falls Risk Assessment Tool (PH-FRAT), commonly used in residential aged care. The content, design and functionality of the dashboard have been co-developed to ensure that it is suitable for implementation and use by staff. It is considered complementary to standard care. Access to four datasets was obtained: the resident profile (includes residents’ demographics and admission information); all medications administered to residents; the organization’s PH-FRATs (this aged care providers used the PH-FRAT) to obtain information related to falls risk assessments; and incident dataset (includes information related to all fall incidents and pressure injuries; The profile dataset included a free text field that reported the comorbidities along with other special needs of the patient at admission (ie, health status). From this field, comorbidities present at admission were identified using the R-programmed version of the “aged care health status algorithm” within PBI. The algorithm identifies the health conditions using free text fields from EHRs. All medications in the dataset were coded using the Anatomical Therapeutic Chemical codes. These datasets were then linked in the dashboard backend. The extracted datasets for the study period underwent an external analysis for this study using R programming language (version 4.3.3; R Core Team) and were also subsequently used for its intended purpose within PBI. A descriptive analysis of the admission-related information from the resident profile dataset is reported appropriately. Resident characteristics from the resident profile dataset, FRAT dataset, and daily medication administration data are recorded. MQ-Dash is a system used to present data already extracted from other Anglicare systems, there is no data input required to use MQ-Dash and therefore there is no time burden on staff. The dashboard practice points are taken from a rapid literature review conducted by the MQ team. Examples of practice points from MQ-Dash include: recommending a medication review with resident medical team, environmental recommendations such as clearing the floors of any potential trip hazards, extra resident supervision. To monitor use of MQ-Dash a study specific questionnaire is used at the 3-month, 6-month and 12-month check-in point for the trial. This has questions on frequency of MQ-Dash use. For MQ-Dash usage we also use PowerBI analytics, which reports which users have opened and used MQ-Dash.