About the Project
This project is one of a number that are in competition for funding from the ‘GW4 BioMed MRC Doctoral Training Partnership’ which is offering up to 17 studentships for entry in October 2021.
+++ Note: Full funding may not be available for all applicants. Please see the Funding Notes below for more information +++
The DTP brings together the Universities of Bath, Bristol, Cardiff and Exeter to develop the next generation of biomedical researchers. Students will have access to the combined research strengths, training expertise and resources of the four research-intensive universities. More information may be found here: https://www.gw4biomed.ac.uk/.
SUPERVISORY TEAM:
Dr Matthew Jones (lead), University of Bath, Department of Pharmacy & Pharmacology https://researchportal.bath.ac.uk/en/persons/matthew-jones
Dr Anita McGrogan, University of Bath, Department of Pharmacy & Pharmacology
Dr Hannah Family, University of Bristol, Medical School
Dr Hélène De Ribaupierre, Cardiff University, School of Computer Science and Informatics
THE PROJECT:
There are 237 million medication errors in England per year. Around 28% cause harm to patients and cost the NHS more than £98 million. The international cost is $42 billion. This project will use artificial intelligence (AI) to identify possible causes of medication errors reported in the NHS. The student will gain skills in AI and statistics for large data sets through the project and training at all three universities, and knowledge of medicines safety via an NHS placement. This project will suit students from a wide variety of backgrounds, e.g. computer science, mathematics or a health profession.
There are many causes of errors and all need to be addressed to increase safety. One recognised cause in many settings is poorly written guidelines for health professionals leading to difficulty finding relevant, unambiguous information. Our research has identified tools that improve guidance and increase safety. We have published this recently in the Journal of Patient Safety and BMJ Quality and Safety, two of the leading patient safety journals (see references).
This project will enable us to target our tools at high-risk areas but identifying 1) the frequency of medication errors caused by poor quality guidance (by NHS sector and staff group); 2) the types of guideline frequently implicated; 3) the types and severity of the associated errors.
A systematic review and meta-analysis of professional guidelines as a cause of medication errors will address these knowledge gaps using international literature with sub-group analysis of NHS data. There is no equivalent study available in this under-researched area.
Next, the National Reporting and Learning System (NRLS), a database containing thousands of reports of NHS medication errors will be analysed. This is a rich data source, but it is not coded to identify errors caused by guidance, so analysis of free text is required. Given the scale of the NRLS, automation is needed. Natural language processing (NLP, a subfield of artificial intelligence) has been used to analyse incident reports in safety critical industries but it is not widely used in medicines safety. Because of the intrinsic nature of free text, research in these areas is still growing and presents different and interesting challenges. This project will use NLP and machine learning (ML) to identify relevant NRLS reports. Several algorithms will be trained and evaluated to identify medication errors caused by poor quality information. The best approach will be implemented to analyse the complete NRLS dataset and address the knowledge gaps outlined above. The transfer of this approach to different fields in medicines safety will be investigated.
If successful, at least three peer reviewed publications will results from this project (meta-analysis, algorithm development and NRLS analysis) and we will also work closely with the NHS to explain our findings and how they can be used to increase patient safety.
APPLICATIONS:
Applicants must have obtained, or be about to obtain, a First or Upper Second Class UK Honours degree, or the equivalent qualifications gained outside the UK, in an area appropriate to the skills requirements of the project.
IMPORTANT: In order to apply for this project, you should apply using the DTP’s online application form: https://cardiff.onlinesurveys.ac.uk/gw4-biomed-mrc-doctoral-training-partnership-student-appl-2
You do NOT need to apply to the University of Bath at this stage – only those applicants who are successful in obtaining an offer of funding form the DTP will be required to submit an application to study at Bath.
More information on the application process may be found here:
https://www.gw4biomed.ac.uk/doctoral-students/
APPLICATIONS CLOSE AT 17:00 ON 23 NOVEMBER 2020.
Funding Notes
Studentships cover tuition fees at the ‘Home’ level, research/training costs and a stipend (£15,285 p.a., 2020/21 rate) for 3.5 years.
The main categories of candidates normally eligible for 'Home' fees are:
UK nationals*
Irish nationals living in the UK/Ireland
Applicants with settled or pre-settled* status in the UK under the EU Settlement Scheme
Applicants with indefinite leave to enter/remain in the UK
* must have lived in the UK/EEA/Switzerland continuously since September 2018.
Those not meeting the nationality and residency requirements to be treated as a ‘Home’ student may apply for a limited number of full studentships for international students.
References
Jones MD, Franklin BD, Watson MC, Raynor DK. User testing to improve retrieval and comprehension of information in guidelines to improve medicines safety. Journal of Patient Safety. 2020; published ahead of print. https://doi.org/10.1097/PTS.0000000000000723
Jones MD, McGrogan A, Raynor DKT, Watson MW, Franklin BD. User-testing guidelines to improve the safety of intravenous medicines administration: a randomised in-situ simulation study. BMJ Quality and Safety. 2020; published online first. https://doi.org/10.1136/bmjqs-2020-010884