Predictive risk stratification model: a progressive cluster-randomised trial in chronic conditions management...

Area of Science:

• Health Services Research
• Predictive Analytics in Healthcare
• General Practice Innovation

Background:

• An aging population increases demand on health and social care services.
• New approaches are needed to shift care from hospital to community and general practice settings.
• A predictive risk stratification tool (Prism) has been developed to estimate emergency hospital admission risk in general practice.

Purpose of the Study:

• To evaluate the effectiveness, cost-effectiveness, and implementation of the Prism tool in general practice.
• To assess Prism's impact on patient outcomes, care processes, and practitioner/policymaker perceptions.
• To identify barriers and facilitators to Prism implementation in primary care settings.

Main Methods:

• A mixed-methods progressive cluster-randomised trial comparing usual care with Prism implementation.
• Data collection includes routine data, postal questionnaires (at baseline, 6, 18 months), focus groups, and interviews.
• Economic evaluation includes cost-effectiveness and cost-consequence analyses; data analysis uses generalized linear models and survival analysis.

Main Results:

• Technical performance will be assessed by comparing predicted vs. actual emergency admissions.
• Patient outcomes, costs, satisfaction, and quality of life will be compared between groups.
• Qualitative data will explore perceptions and adoption of Prism by practitioners and policymakers.

Conclusions:

• The study will provide crucial information on the costs and effects of the Prism tool.
• Findings will illuminate practical use, implementation challenges, and perceived value in managing chronic conditions.
• Results will inform the integration of predictive tools into primary care for improved patient management.