Nearly 7 million patients might have been subject to a general practice error in the past year. With new models of care proposed, Sukhmeet Panesar and colleagues urge GPs to have a greater awareness of patient safety

Illustration showing patients falling through nets to be caught in bed trolleys below

General practice has evolved slowly since the inception of the NHS in 1948, but a consensus is growing around the need for radical change to address issues of morale, recruitment and retention.

The traditional small scale of practices is perceived to be a particular problem. Yet, although new network models of practice are emerging, the implications for patient safety of changes in structure and governance have received scant attention.

‘Errors in primary care occur in about 1 to 2 per cent of all consultations’

Perhaps this is because little is known about patient safety in primary care. But at a time when GPs face ever greater complexity, more risk and an ageing population’s multimorbidity – all of which make errors more likely – the lack of focus on safety initiatives is worrying. 

It has been estimated that errors in primary care occur in about 1 to 2 per cent of all consultations, although some studies put this figure higher.

This is lower than the 10 per cent error rate in hospitals, reflecting the nature of risk in the two sectors.

There were an estimated 340 million GP consultations, in absolute terms, in 2013-14. So 6.8 million patients might have experienced an error over the last year.

The National Reporting and Learning System is the largest repository of patient safety incidents in England and Wales. Since 2003, 11 million incidents have been reported.

However, only 5,706 incidents were reported from general practice over the last year. It is likely that this reflects underreporting by GPs, and the reasons for this are complex.

Definitions of harm

There are two main problems with the promotion of patient safety in primary care. First, there is a lack of consensus about the patient safety terms in common use, which can cause problems for those attempting to engage primary care practitioners.

Definitions of patient safety terms

Patient safety termDefinition
HarmAny physical or psychological injury or damage to the health of a person, either temporary or permanent. Harm is usually classified as no harm, low harm, moderate harm, severe harm or death.
Near missAny patient safety incident that had the potential to cause harm but was prevented, resulting in “no harm”.
Adverse eventAny event involving unintended harm to a patient that resulted from medical care.
Preventable adverse eventAn event involving patient harm as a result of wrong or inappropriate action (“error of commission”) or failing to do the right thing (“error of omission”).
Adverse drug eventAny incident in which the use of medication results in harm to a patient, including adverse drug reactions – as in the case of known side effects that occur even when the medication is used as intended – as well as events in which the drug has been used erroneously.
Patient safety incidentAny unintended or unexpected incident that could have harmed or did harm the patient, including “near misses”. The term “patient safety incident” is preferred to “error”, as the latter has a more negative connotation.
Critical incident*A term first coined in the 1950s and made famous by a classic human factors study by Cooper of “anaesthetic mishaps”. They defined critical incidents as occurrences that are “significant or pivotal, in either a desirable or undesirable way”. This means that there was significant potential for harm – or actual harm – but also that the event has the potential to reveal important hazards in the organisation, and provide valuable opportunities to learn about individual and organisational factors that can be remedied to prevent future incidents.

Source: Cooper et al 1978 (reproduced with permission of Wolters Kluwer Health)

In general practice, it is less common for patients to come to serious harm than in acute settings (examples of the differing nature of risk in primary care settings are shown in the table, below).

Examples of patient safety incidents with corresponding levels of harm*

Category of harmDefinitionExample
No harm (impact prevented)Any unexpected or unintended incident which was noticed and halted or reversed before it was able to cause harm to a patient.A GP prescribes an inappropriate dose of a drug, which the local community pharmacist picks up when dispensing the prescription.
No harm  (impact not prevented)Any unexpected or unintended incident which did not lead to harm on this occasion.A patient is on medication that requires blood pressure monitoring. The hospital discharge letter does not mention this to the GP, which results in the patient not being followed up appropriately. However, it is noted when the patient visits the GP for a further prescription. The patient’s observations are then found to be normal.
LowAny unexpected or unintended incident which required extra observation or minor treatment and caused minimal harm, to one or more persons.A patient’s home visit is missed. The terminally ill patient required a pain assessment. This was picked up the following day, resulting in the patient continuing to be in pain until the medication was altered.
ModerateAny unexpected or unintended incident which resulted in further treatment, possible surgical intervention, cancelling of treatment, or transfer to another area and which caused short term harm, to one or more persons..Continuing treatment with warfarin without monitoring clotting levels for a length of time which results in an overdose and bleeding problems that require close monitoring and follow-up.
SevereAny unexpected or unintended incident that caused permanent or long term harm to one or more persons.A patient, who is a heavy smoker, visits the GP with a cough. The patient’s name is Mr Jones. He has a chest X-ray and the report suggests a suspicious lesion with the advice to refer for further investigations. The GP writes on the report that an urgent appointment is needed and the receptionist files the report in Mr Jones’s file. Mr Jones rings the surgery for his results and the receptionist looks in his file and says that no report has been received yet and that the practice will ring him if there is any news in the next week. Mr Jones does not hear so assumes everything is OK. Two months later Mr Jones visits the GP on a routine appointment and Mr Jones’s urgent request is found. Mr J Jones is referred and it is found that he has lung cancer.
DeathAny unexpected or unintended incident which caused the death of one or more persons.A patient suffering from chest pain is asked to wait for a free slot in the GP surgery. As he feels difficulty in getting his breath, he goes for a walk, collapses and dies in the GP surgery’s car park.

*Examples of incidents from Seven steps to patient safety for primary care.

Patchwork reporting

A second problem in promoting patient safety in primary care is that, in addition to the NRLS, there are several channels for reporting and learning from errors.

Given these disparate and disconnected reporting systems, it is hardly surprising that practices are confused and have failed to engage with the safety agenda in a systematic way.

‘It’s hardly surprising that practices have failed to engage with the safety agenda in a systematic way’

Change cannot be expected overnight – it has taken a decade for the patient safety movement to make significant gains in the acute setting.

However, we need to explore the benefits of a unified mechanism of reporting and learning from errors.

Big data has a role to play but a mechanism to capture errors from routine datasets without relying on self-reporting is needed. The Primary Care Trigger Tool, developed by the former NHS Institute for Innovation and Improvement, attempted to reduce this burden using a targeted case note review.

However, despite being considered the gold standard, case note reviews can be resource intensive and have never really been used in a “before and after” scenario to assess the efficacy of safety initiatives.

We need to develop a better understanding of safety in primary care and what this means to GPs and others. A renewed effort is needed to encourage patient safety experts, policy makers and GPs to enshrine patient safety within the newer service models for primary care.

‘Patient safety needs to be a priority in the creation of multispecialty community providers’

General practice needs to define the meaning of safety in line with its core principles of first contact, and comprehensiveness, continuous, coordinated and patient centred care.

Safety and quality improvement training needs to be provided in both undergraduate and postgraduate curricula.

GPs must be equipped with the knowledge and tools on how to design, test, implement and evaluate safety improvement initiatives, both at practice and population health levels.

New models

During this period of promoting various models of care, it is possible to imagine federations or networks having dedicated patient safety officers, risk managers and the infrastructure to promote reporting and learning from errors akin to those found in hospitals.

Patient safety needs to be a priority in the creation of multispecialty community providers, one of the new models of care suggested by the NHS Five Year Forward View.

Similarly, primary and acute care systems need hospitals to assess how their patient safety infrastructure can be translated into primary care.

‘Hospitals need to assess how their patient safety infrastructure can be translated into primary care’

The long term vision should answer two very important questions.

First, how can the newer models of care incorporate routine surveillance mechanisms to assess the burden of unsafe care, perhaps through the use of standardised safety indicators?

Second, how can an open culture of reporting and learning from errors be promoted locally within newer models of care and shared nationally?

Significant strides have been made in improving safety in hospitals and the time is now ripe for similar progress to be made in the primary care sector.

We should not wait for a patient’s primary care horror story to make the tabloids before we focus on delivering safer primary care.

Sukhmeet Panesar is honorary fellow at the Centre for Population Health Sciences, University of Edinburgh; Martyn Diaper is medical director (quality) of Southern Health Foundation Trust and head of patient safety (primary care) at NHS England; Donna Forsyth is senior head of patient safety (investigation) for NHS England; Andrew Carson-Stevens is clinical lecturer in healthcare improvement at the Cochrane Institute of Primary Care and Public Health; Sanjiv Ahluwalia is deputy head of GP School for London Deanery; Siân Rees is director of University of Oxford Health Experiences Institute; Martin Marshall is professor of healthcare improvement at UCL; and Charles Vincent is professor of psychology for the University of Oxford