This site is optimised for modern browsers. For the best experience, please use Google Chrome, Mozilla Firefox, or Microsoft Edge.

Case study: Reducing patient harm through medication errors

Using systems that are already in place

In the UK, just four classes of commonly prescribed drugs account for over half of preventable hospital admissions. Like other doctors, general practitioners (GPs) work from the first principle of do no harm, but mistakes get made and patients are frequently prescribed drugs that have serious consequences.

The annual hospital admission cost in England for adverse drug events was approximately £650 million in 2013.

Anthony J Avery, Professor of Primary Health Care and NIHR Senior Investigator, explains: “We found that prescribing errors were identified in five percent of prescription items, with 1 in 550 items containing a severe error. This equates to approximately two million serious prescribing errors in English general practices each year.”

He continues: “For example, patients with asthma are sometimes prescribed beta blockers, which can tighten their airways. Or treatments such as non-steroidal anti-inflammatory drugs (NSAID) that increase the risk of gastrointestinal bleeding are prescribed without giving protection against stomach ulcers and stomach bleeds. 

“Sometimes doctors need a prompt to check for conditions that suggest a drug should not be given or that the pros and cons should be very carefully weighed up.”  

Professor Avery has worked with Dr Sarah Rodgers of the University of Nottingham for nearly 25 years. They realised that the computer systems used in GP surgeries could be used to identify some of these patients to catch problems before they result in harm. 

Along with colleagues at the University of Manchester they developed safety indicators to identify patients at risk of medication error. Their idea was that a search could be done on the GP computer system to identify patients who are at risk and then have a pharmacist look in more detail to decide if changes need to be made. 

The pharmacist-led intervention

Together they developed a clinical trial to test a pharmacist-led IT-based intervention, PINCER, to reduce clinically important medication errors in primary care settings. The intervention focuses on medicines that are prescribed commonly for common conditions. 

The Department of Health funded a trial through its patient safety research portfolio that tested the intervention in 72 general practices. After searches were run in all practices to identify patients at risk, feedback was given in the form of computer generated summaries and brief educational materials to the control group, but in addition, the intervention group had a pharmacist work with practices to discuss cases and correct problems that were identified. 

Across the 72 practices there were 480,942 patients. At 6 months' follow-up, patients in the PINCER group were significantly less likely to have been prescribed a non-selective NSAID if they had a history of peptic ulcer without gastroprotection, a beta blocker if they had asthma, or an ACE inhibitor or loop diuretic without appropriate monitoring. 

Dr Rodgers explains: “One of the interesting things about the trial is that we were able to demonstrate that the intervention was effective compared to just running the searches, and was considered to be cost effective too. 

“To anyone who does not understand the pressures on general practice, it may not be obvious that you need to do more than just feedback the information. Indeed, since the trial, people have tried to suggest that just running the searches is enough, but we are able to show the value of the role of the pharmacist and push back on that.”

A comprehensive training package was developed for pharmacists, who worked for three days a week for up to 12 weeks in each of the GP surgeries. The focus was not solely on addressing issues that were identified but looking thoroughly at systems and processes, using tools and analysis to spot why patients slipped through the net, and changing processes to ensure errors were not repeated. Changing systems and processes in this way provides true long-term impact on patient safety. 

A successful national rollout

Following the trial, Dr Rodgers was awarded an NIHR primary research fellowship for three years to work on developing and updating the prescribing safety indicators. These indicators were then used in the scaling up of the work in the East Midlands, with funding from the Health Foundation and the East Midlands Academic Health Science Network (AHSN). 

Professor Avery says: “One of the brilliant things about NIHR funding is that it covered bridging time between the trial being published and us moving to the regional and then national rollouts. This was really important because it takes time to think things out properly. I think the tremendous achievement of this work is largely a result of our having time to work out the best way to scale it up and get it right.”

The Scaling Up PINCER project in the East Midlands achieved a 94% uptake of the intervention among general practices. Findings from the evaluation of the project showed a statistically significant reduction in hazardous prescribing, with the greatest differences demonstrated for prescribing safety indicators associated with gastrointestinal bleeding. 

In the Scaling Up PINCER project, practices fed back that some of the IT side was unintuitive and was off putting. They also wanted feedback on where they sat compared to other practices. Both are being addressed as the PINCER is rolled out across England. 

Working with PRIMIS at the University of Nottingham, the team have developed a more straightforward approach to run the searches, which are now embedded in the software used in GP surgeries. As part of the national rollout, information on numbers of patients at risk of potentially hazardous prescribing based on the 13 PINCER prescribing safety indicators contribute to a national comparative data service. This provides the ability to monitor changes in numbers of at-risk patients across localities and on a national basis.

Over the last two years, funding from the Health Foundation has enabled PRIMIS to work with Spring Impact, a non-profit global leader in social replication, on a model for the national scale and spread of PINCER. It has also enabled the secondment of Dr Sarah Rodgers from her post as Principal Research Fellow to work with PRIMIS for 18 months as PINCER National Programme Manager. As a result of this work, PINCER is currently being rolled out to all general practices in England via the AHSN Network. The team were regional winners in the NHS70 awards in 2018 for Excellence in Primary Care

In the 18 months to April 2020, 130 Clinical Commissioning Groups — almost 70% — have engaged in the PINCER rollout. Over 1,138 pharmacists have been trained to deliver the intervention through a combination of eLearning tools, online resources and face-to-face action learning set sessions.

In this time, 2,430 GP practices have uploaded baseline data to the national PINCER comparative analysis service. This shows that a minimum of 23.35 million patient records have been searched to identify instances of potentially hazardous prescribing using the 13 evidence-based prescribing safety indicators. In total, 187,534 at-risk patients have been identified in at least one prescribing safety indicator.

The NIHR Greater Manchester Patient Safety Translational Research Centre (GM PSTRC) contribution 

Researchers from the NIHR Greater Manchester Patient Safety Translational Research Centre (GM PSTRC) worked with colleagues from the University of Nottingham, to develop the PINCER safety indicators, and have played an ongoing pivotal role in testing and implementation. This includes developing new indicators.  

The GM PSTRC built on the principles of PINCER to develop SMASH, a web based application led by Dr Richard Williams and Professors Peek and Ashcroft. It’s a dashboard that updates in near real time and combines digital and behaviour/learning aspects of the safety PINCER indicators and is an automated system. The Medication Safety and Safety Informatics themes at the GM PSTRC have been working on the intervention together. 

SMASH has been tested and evaluated at 43 GP practices in Salford as part of a 12 month study. There was a marked reduction in the rates of hazardous prescribing which was sustained across the 12 month study. Dr Mark Jeffries has led a qualitative evaluation of the lived experience of using SMASH and by GPs pharmacists working in general practice. Now Health Innovation Manchester (part of the AHSN) are rolling it out across Greater Manchester to a population of 2.8 million.