Cookie policy notice

By continuing to use this site you agree to our cookies policy below:
Since 26 May 2011, the law now states that cookies on websites can ony be used with your specific consent. Cookies allow us to ensure that you enjoy the best browsing experience.

This site is intended for health professionals only

At the heart of general practice since 1960

GPs only act on 2% of computer prescribing alerts, says study

GPs need to see computerised prescribing alerts earlier in their consultations in order to make any changes, conclude the authors of an NHS study that showed that only 2% resulted in any action from the GP.

Their research suggested that drug safety alerts may be ignored at least partly because they tend not to be flagged until the very of end of the prescribing process.

The results of the INTERACT-IT study come after a GMC-funded analysis published last year found errors in one in 20 GP prescriptions, and suggested that some of the errors were due to ‘alert fatigue’ during consultations.

The study - commissioned by the NHS Connecting for Health Evaluation Programme - looked at how eight GPs interacted with computerised decision support systems within the context of a consultation.

The University of Edinburgh researchers filmed the GPs over a total of 112 patient consultations, during which 132 prescriptions were issued. Five of the GPs were using EMIS software and the other three were using INPS Vision.

Out of a total of 117 alerts that occurred in the process of issuing these prescriptions, of which 55 were safety alerts, only three (2%) resulted in the GP taking action and none was ultimately altered.

Using the Roter Interaction Analysis System (RIAS), a validated tool for analysing routine medical dialogue, the team looked at the conversation between doctor and patient in each consultation, and found that over half of the discussion about the medical problem and how to treat it had already taken place before the computer was used to generate a prescription.

Furthermore, the team found that before an alert appeared, GPs had already gone through most stages of working up the prescription – from formulating the problem, to negotiating and explaining the options and then reaching agreement with the patient. In some cases GPs had even already given instructions on taking the medication and printed an information leaflet.

The authors concluded: ‘An alert in the final seconds of the task of generating the paper prescription is likely to be regarded as intrusive and unwelcome, and increases the probability of it being ignored,’ the researchers say.

They suggest that personalised information about potential treatments should be available earlier on, at the point when GPs are considering prescribing, for example by keeping core history information available once the prescribing dialogue box is opened.

Study leader Dr Hilary Pinnock commented: ‘Previous evidence has suggested that the effectiveness of these systems is limited as prescribers notoriously ignore/override the alerts.  

‘Our data, using systems in widespread use at the time of our recordings, showed that too many alerts occurred far too late in the consultation process to be of any use – and goes some way to explaining the problem.’   

Speaking to Pulse, Professor Tony Avery, professor of primary care at the University of Nottingham and a part-time GP in Chilwell, who led the recent GMC-commissioned PRACTICE study of prescribing safety, said the Edinburgh research was ‘very interesting’.

He said it was not clear whether GPs in the study had ignored any serious alerts, but that ‘alerting systems are still not designed as well as they could be, so the signal to noise ratio is not good’.

He added: ‘We also need to think also how we can adjust things ourselves. You only need to talk generally with the patient about their condition, you can then go to the computer and make your decision –  and only commit yourself and tell the patient once you have decided what you are going to give them.’

Professor Avery also noted that the types of software used in the study are very basic and that newer systems such as EMIS Web address at least some of the problems highlighted.

Dr Shaun O’Hanlon, clinical and development director at EMIS, told Pulse that some prescribing decisions do not follow a predictable algorithm that can be easily modelled ‘due to patient and clinician factors that are unrelated to the information held on the IT system’.

He said that EMIS’s most advanced system, EMIS Web, allows GPs to use more proactive prescribing tools – such as presenting medication options within the consultation ‘as the clinical scenario is elucidated within the system’.

However, Dr O’Hanlon noted that ‘while many users respond well to this level of support, others choose not to use these tools as they find them intrusive - indicating that, ultimately, safe prescribing relies on a combination of system support and individual clinical preferences’.

J Am Med Inform Assoc 2013, online 3 March

 

Pulse Live: 30 April - 1 May, Birmingham

Pulse Live

Dr Kartik Modha, a GP in London and founder of Tiko’s GP Group, will look at how technology and social media is changing GPs’ practice at Pulse Live, Pulse’s new two-day annual conference for GPs, practice managers and primary care managers, will cover the latest developments in telehealth.

Pulse Live offers practical advice on key clinical and practice business topics, as well as an opportunity to debate the future of the profession, and a top range of speakers includes NICE chair designate Professor David Haslam, GPC deputy chair Dr Richard Vautrey and the Rt Hon Stephen Dorrell MP, chair of the House of Commons health committee.

To find out more and book your place, please click here.

Readers' comments (1)

Have your say