GP practices in England provide over 30 million appointments per month. Some are for urgent medical conditions that require urgent treatment where delays in care can lead to patient harm. Since 2020, almost all GP practices in England have provided online consultations where patients can request help over the internet. Online consultations are received by GP practices unprioritised which can worsen care delays. A potential solution is for the online consultation system to triage patients using artificial intelligence (AI). AI Triage is already used in NHS GP practices despite there being little evidence for its safety or whether this varies based on patient characteristics.

The aim is to evaluate the safety of AI Triage in GP practices and whether it varies according to patient clinical and sociodemographic characteristics including age, sex, socioeconomic deprivation, and ethnicity.

Reducing the provision of low-value interventions that expose patients to unnecessary risk of harm, both directly, and indirectly, is key to promoting patient safety and reducing healthcare costs. Over the last two decades, there has been significant attention from the policy and academic community on reducing the provision of low-value care that exposes patients to risk of harm. For example, the development of guidance (i.e. “do not do” recommendations from the National Institute for Health and Care Excellence), shared decision making (i.e. the Choosing Wisely initiative), and clinical audit and feedback systems (i.e. integrated within electronic health records). Despite this, many ineffective and harmful interventions continue to be used in routine clinical practice.

The aim of this PhD is to declutter primary care by reducing the provision of low-value interventions with limited clinical- and cost-effectiveness, and where the risk of harm exceeds likely benefit. This will be achieved by: (i) establishing a list of indicators to measure low-value care that expose patients to risk of harm in English general practice, (ii) analysing routinely collected administrative data to examine micro (i.e patient and clinician) and macro (i.e practice characteristics) factors that drive the provision of low-value care that expose patients to risk of harm, and (iii) identifying enablers and barriers to sustainable behaviour change in primary care required to reduce provision of low-value care that exposes patients to risk of harm

This PhD project will take a mixed-methods approach and the PhD candidate will develop skills and undertake training in evidence synthesis, consensus studies, quantitative analyses of healthcare administrative datasets, and qualitative research techniques. Collectively, this PhD project will develop actionable evidence that can inform policy and promote the delivery of safe, effective, and high-quality care delivery in English general practice.

Deadline for application: 28 May, 2024

Click here to find out more and apply