Article Text
Abstract
Introduction Patients with chronic conditions account for a third of emergency hospital admissions, many of which are not life-threatening and result in unnecessary hospital stays. Focussed primary and community care could prevent one-in-five emergency admissions, improving health outcomes and saving money. Predictive risk models which use routine data to identify at risk populations for targeted care aim to reduce emergency admissions but we do not know how GPs react to their introduction.
The Prism tool predicts likelihood of patient's emergency admission to hospital and divides the entire practice population into risk categories low to high. Swansea University is leading the PRISMATIC trial, evaluating the introduction of Prism into 32 Welsh GP practices.
Prism, and other similar tools, effectively predict admissions but questions remain over their adoption. We asked potential users of Prism how they expect its introduction to affect patient care and emergency admissions.
Methods 43 GPs, practice managers and nurses from 32 general practices participated in one of four focus groups or an interview before receiving Prism. We recorded the groups and interviews, with participants' consent, and carried out thematic analysis of transcripts.
Results We found that respondents supported the principle of proactively identifying at risk patients and were willing to trial Prism. It appealed to their desire to improve care through targeting patients who could become emergencies. Perceived challenges included: integrating Prism with existing practice; limited time and community staff to follow up identified patients.
Summary Risk predictions tools like Prism effectively predict hospital emergency admissions and are increasingly being introduced. Without sufficient evidence about how practitioners perceive and use these tools, their introduction is unlikely to make a difference to patients—nor have the desired effect on resources. PRISMATIC will provide evidence about how context and expectations can affect implementation.
- publication