Q&A: Is practice software a safety risk?

Practice software has become a key tool in the work of the GP since it was widely introduced in the 1990s. But does it improve healthcare outcomes?

Australian Doctor talks with Associate Professor Farah Magrabi (pictured), leader of patient safety informatics at the Australian Institute of Health Innovation, Macquarie University, about its risks and benefits.

Related News:

Australian Doctor: Is there much evidence that practice software systems have improved patient safety? 

Professor Magrabi: Previous studies have looked at using computers for clinical purposes and evaluated their impact on care quality. But, no, there has not yet been a systematic evaluation of the overall patient safety impact of computerisation in Australian general practice.  

AD: What improvements would you like to see?

Professor Magrabi: There is an urgent need for user-friendly software that fits clinical workflow. The many systems we have today can be incredibly hard to use, increasing the risk of errors. We are able to pick up a smartphone and use it for the most basic tasks without any formal training. We need that for practice software.

Formal approaches to software and user interface design that are well developed in other sectors need to be applied in health. We also need to develop safety standards for clinical software. There is no common safety standard in Australia and no systematic operational oversight of software used in healthcare.

Related News:

AD: What issues has your research identified? 

Professor Magrabi: My research has focused on the patient safety risks associated with e-health. E-health is a major element of healthcare reform, but the work of my team has consistently shown new clinical IT systems have the potential to introduce new risks to patients when they are poorly designed, implemented or used. 

Our study examining IT-related patient safety risks is not yet published. But our 2015 analyses of incidents from England’s national program for IT describe the types of practice software issues that pose threats to patient safety.  

We found deficient user interfaces gave rise to errors — for instance, when a GP mistakenly selected the default dose of 100mg/mL for morphine instead of 10mg/mL. The patient who was already on morphine suffered from excessive side effects. A dose of 100mg/mL could have been lethal for a first-time user.

Related Opinions:

In another case, a GP mistakenly selected methadone concentrate 10mg/mL, the first option in an alphabetically arranged list, instead of 1mg/mL. The patient was admitted to an emergency room with respiratory arrest but made a full recovery. 

Problems with software configuration and migrating records from one package to another also generated errors. One patient received only half of their usual quantity of blood pressure medication because a repeat prescription didn’t transfer to a new software package when their records were migrated. The patient tried to stretch out the dose by taking the medication on alternate days and had a stroke, but fully recovered.

AD: How will practice software develop in the future? 

Professor Magrabi: In coming years, we will see sophisticated clinical decision capabilities appearing in practice software, where analytics will be performed in real-time at the point of care. This will support communities of learning among general practices to improve quality, safety and effectiveness of care.

Last month, Professor Magrabi won the Sax Institute’s Research Action Awards, which honours researchers whose work has changed the way healthcare is delivered.