The optimal and effective organisation of primary care resources in the NHS requires information on performance and associated costs of the different types of doctors who are employed in the NHS. However, the lack of evidence on performance for particular types of GP doctors has led to concerns among policymakers and professional associations about the quality of provision of primary care services and patient safety. Very little research has been done about how and when locum doctors are used in primary care, or if their use can influence quality and safety.
NHS organisations need staffing flexibility and capacity that will allow them to effectively provide high quality healthcare which is safe for patients and minimises costs. The aim of our research is to help the NHS to understand how best to use locum doctors, and how to ensure that locum doctors can provide safe, high quality care.
Our research will compare the working patterns of different types of GP doctors (partners/salaried/locum) and examine what differences in quality and safety may exist. To do so, we will use data which is routinely collected during GP consultations and contains information on the type of doctor involved. We will be able to determine whether there exists variation in performance and safety between locums and other types of GP doctors, and whether variation in locum use is associated with worse outcomes for patients, such as having to return to see another doctor, being prescribed potentially hazardous medications or not being referred for further tests.
The aim of the study will be to collect and analyse quantitative data sets to compare the working patterns and performance of locums to those of other GP types in primary care. We will use the Clinical Practice Research Datalink (CPRD) to access anonymised general practice records on all registered patients, including diagnoses, tests, prescribed treatments and referrals. We will obtain data on each patient’s interaction with a health professional and the staff role field will be used to identify GP types (partner, salaried, or locum) involved in any recorded interaction. This will be our key parameter of interest and we will evaluate its association with patient outcomes. A cross sectional analysis will be conducted for each financial year for the period 2015-2021 to explore the associations between staff role and patient outcomes at the patient level. Moreover, our cohort study is eligible for linkage to Hospital Episodes Statistics (HES) data and this will allow us to obtain detailed information on any form of hospital attendance for all patients in the practices that participate in the linkage scheme.
We will focus on working patterns, service utilisation and patient outcomes. Our primary patient outcome will be patients revisiting general practice within 7 days or 15 days. Secondary outcomes will include: secondary care referrals, unplanned hospital admissions within 7 days or 15 days (HES linkage), A&E attendance within 7 days or 15 days (HES linkage), ambulatory care sensitive conditions (ACSC) hospitalisations within 7 days or 15 days (HES linkage), antibiotic, strong opioids and hypnotics prescriptions and validated prescribing safety indicators.
We will use multi-level mixed effects logistic regression models to quantify the association between exposure (staff role) and outcomes, controlling for all available covariates. Outcomes and key covariates will be appropriately described over time, at the patient, GP and general practice level.
• Patients revisiting the general practice within 7 days or 15 days
• Referrals to secondary care (from CPRD GOLD)
• Unplanned hospital admissions within 7 days or 15 days (HES linkage)
• A&E attendance within 7 days or 15 days (HES linkage)
• Ambulatory care sensitive conditions (ACSC) hospitalisations within 7 days or 15 days (HES linkage)
• Antibiotics prescriptions
• Strong Opioid prescriptions
• Hypnotics prescriptions
• Validated prescribing safety indicators (STOPP/START criteria, PINCER indicators)
Kieran Walshe - Chief Investigator - University of Manchester
Christos Grigoroglou - Corresponding Applicant - University of Manchester
Darren Ashcroft - Collaborator - University of Manchester
Evangelos Kontopantelis - Collaborator - University of Manchester
Gemma Stringer - Collaborator - University of Manchester
Jane Ferguson - Collaborator - University of Manchester
Thomas Allen - Collaborator - University of Manchester
2011 Rural-Urban Classification at LSOA level;HES Accident and Emergency;HES Admitted Patient Care;Patient Level Index of Multiple Deprivation;Practice Level Index of Multiple Deprivation