100 Ways of Using Data to Make Lives Better

Can Data Be Used to Make Sure Medications Are Used Safely?

Published on: 8th June 2017

Download case study

Learning Health Systems
Case Study 23

Project Lead: Prof Bruce Guthrie, University of Dundee

A team of researchers from Scotland and Australia used data to find a way of recognising patients that are at high-risk of reacting seriously to the medicines they have been prescribed.

The Challenge

Every year almost 1 billion drug prescriptions are made in the UK and more than 200,000 people are admitted to hospital as an emergency because they suffer a sudden, unexpected and adverse (bad) reaction to the medicines they have been prescribed. In some cases it is impossible to predict if and why this happens. In others however, there might be an opportunity to look at medical records held in primary care (such as GP surgeries) to identify patients who are ‘at risk’.

By looking at this data, healthcare professionals including pharmacists, doctors and nurses would be able to contact these ‘at risk’ patients to review the drugs they are taking. This would improve safety and reduce the risk of people being admitted to hospital because they have reacted badly to their medication.

Researchers in Scotland and Australia have worked together to find a solution to this and to identify a set of rules that can recognise patients with high-risk prescriptions, hopefully preventing some people from being admitted to hospital.

The Research

The research team identified what is known about the safety and negative experiences associated with using prescription drugs. They found 176 common uses of medicines that were thought to be high priority for initiatives that would improve quality and safety in general practice. The team also conducted a study funded by the Scottish Government to find out if flagging these high risk medicines to healthcare professionals could improve safety and reduce risk. To make sure improvements were carried out, there were interventions such as educational meetings, structured
reviews, financial incentives and smart-apps that helped identify patients in GPs electronic records.

The Results

By using data and apps, the researchers found that GPs stopped 40% of the drugs that were putting patients at risk of harm (reduced from 3.7% to 2.2% in almost 30,000 people). The benefits of the study continued for almost a year after the interventions had stopped. As a result, the rate of serious adverse reactions that patients had to their medications, like bleeding in the gut and heart failure, were significantly reduced.

The Impact

NHS Scotland has implemented a number of the researchers high-risk prescribing indicators in National Therapeutic Indicators and in NHS prescribing data analysis tools. Several Health Boards have used these indicators in improvement projects, with NHS Forth Valley, for example, stopping high-risk medicines in 1,200 patients as a result. The team is now working across Scotland to enhance the safety of prescriptions in 370 local practices with the aim of improving the safety of prescribing drugs to the large number of elderly people who take multiple medications every day. It is hoped that this work will lead to improvements in the safety and effectiveness of medicines on a large scale for patients across the population.


Marwick, CA, Guthrie, B, Pringle, JEC, Evans, JMM, Nathwani, D, Donnan, PT & Davey, PG 2014, ‘A multifaceted intervention to improve sepsis management in general hospital wards with evaluation using segmented regression of interrupted time series‘ BMJ Quality & Safety, vol 23, no. 12, e2. DOI: 10.1136/bmjqs-2013-002176

For more information about prescribed medicines visit

Enquiries to Cherry Martin, Communications Manager, The Farr Institute of Health Informatics Research, cherry.martin@ed.ac.uk

Back to top+