'Doing Safety' in General Practice
Dr Maureen Baker, Chair RCGP
When I graduated from Medical School in 1981 (I've been around a long time!), no-one had ever heard the term 'patient safety'. Likewise, the term was never used through my vocational training for general practice. In fact, it wasn't until I read the 1995 RCGP Occasional Paper on Significant Event Audit by Mike Pringle (now our President) and Colin Bradley that I came across any sort of structured approach to patient safety and even then it was not expressed as 'patient safety'. Although it should be said that the notion of 'First do no harm' has been around since the time of Hippocrates.
It wasn't until the publication of 'To Err Is Human' by the US Institute of Medicine in 1999 that the potential of applying modern safety science in medicine and healthcare was properly grasped and the patient safety movement, as such, was born. Since then, there has been a widespread understanding that patient safety is an important dimension in the teaching and practice of medicine. However, the vast majority of studies, reports and interventions in patient safety are focussed on the acute sector, leading some to contest that primary care - and general practice - doesn't 'do' patient safety.
To what extent is that allegation justified, or is it the case that because our discipline approaches patient safety differently, the work we do in this area is not recognised as such by colleagues elsewhere, even those who are safety specialists?
As an example, consider the difference between incident reporting through the National Reporting and Learning System (NRLS) and Significant Event Audit (SEA). General Practice, and GPs, are often castigated for not reporting into the NRLS, with less than 4% of reports coming from our sector. On the other hand, GPs have been using SEA for 20 years now - a technique that pre-dates the patient safety movement, as explained above. Structured SEA involves an individual in a practice deciding that something that has happened (either something that went well or something that went badly) is a significant event for them and them discussing this event with other team members within a structured SEA meeting. From the meeting, any agreed actions are noted and formal follow-up (as in conventional audit) agreed. This can be a powerful method for learning at the practice level, but generally this learning is not disseminated more widely.
The NRLS, set up by the now defunct National Patient Safety Agency (NPSA), was designed to capture anonymous incident reports (normally from nursing staff) that are routinely locally reported in the hospital setting. The idea, based on experience in other safety critical industries, is to instil a culture of reporting incidents so that lessons can be learned at a national level and then fed back to frontline staff. This concept is perfectly sound, although almost from the beginning it has been difficult to distil meaningful data from the vast number of incident reports to the NRLS (now over 300,000 incident reports a quarter). From the many millions of incident reports, a small number of Safer Practice Notices have been formulated, firstly by NPSA and currently by NHS England.
By comparison, every practising GP is required to submit written reports of two SEAs for annual appraisal and revalidation. These, often detailed, reports are then logged in the doctor's e-portfolio and then shared with the doctor's appraiser - and no-one else.
In looking at the NRLS and SEA, we can see that there are two different methods of learning from incidents that are operating in the acute sector and in general practice. Both have their drawbacks. NRLS have accumulated vast swathes of data, from which it is relatively difficult to identify avoidable harm, difficult to provide reporter feedback, and from which relatively few national alerts have been issued. SEA can be powerful - and almost immediate - learning technique, but learning is seldom shared wider than the specific practice in which the SEA takes place. It could be argued that in the NRLS we have an effective national reporting system that is not a powerful learning methodology. And in SEA we have an effective local learning system that is not utilised for national reporting or learning.
Nonetheless, SEA fulfils the professional requirement to reflect and learn from safety incidents and goes further in that action to avoid subsequent similar incidents is then taken. However the onus is on general practice to adopt reporting into the NRLS despite this being based on standard practice in the acute sector and using systems predicated on that experience. I suggest it would be far more effective to recognise the good practice that takes place within the context of general practice and find a way to access, anonymise and share the circa 70,000 written SEA reports generated annually by GPs in England.
Using this specific example, I submit that general practice does indeed 'do safety', but does so in way that is different from the acute sector and therefore poorly understood by many in the patient safety movement.