Skip to content
View commentaries on this research

This is a plain English summary of an original research article

A review of research from around the world suggests that patient safety incidents in primary care are relatively frequent but most do not result in significant harm to patients. While patient safety in hospitals has been well researched, less is known in primary care.

The World Health Organization funded this review. It found that diagnostic and prescribing incidents were most commonly associated with harm to patients. Given the breadth of the review, differences in health systems, variety of study types and incomplete information, it is not possible to identify completely accurate rates of incidents and harm. Estimates based on this review are that around 2 or 3 patients in every hundred have a safety incident, of which around 4% experience severe harm. Most evidence came from reviewing patient notes and medication records, but safety incident reports were also reviewed. One limitation of this is that we know that staff do not report every incident. Also, this review excluded incidents due to inaction, such as not receiving necessary treatment. This well conducted review provides the best evidence so far.

In the UK at present, 2015 NICE recommendations on safer prescribing include encouraging health practitioners and patients to report incidents and associated harms, for example using the National Reporting and Learning System. A large NIHR project in progress in England and Wales will cast more light on what can be done to improve patient safety.

Why was this study needed?

Primary care is generally the first point of contact people have with the health system, such as in general practices. Each day over a million NHS patients are treated by general practitioners in the UK and 90% of doctor-patient interactions take place in primary care. Despite this, very little is known about the possible risk to patients and the impact of those risks on patient health, compared with efforts to understand the number of occurrences, causes and consequences of patient safety in hospitals. The authors suggest that this may be in part because primary care medical records may not be easily accessible.

In order to look at preventing patient safety errors in primary care, it is important to know how often patient safety incidents occur, what type of incidents these are and what impact they have on patients in terms of severity.

What did this study do?

This systematic review included nine reviews and 100 observational studies on patient safety in primary care, which were assessed for quality. Included studies contained data about numbers or types of safety incidents or the severity or harm associated with safety incidents. Most of the studies measured safety incidents by reviewing records or prescriptions, with some using incident reporting systems, surveys or interviews.

Variation between the studies was too high to pool the results in meta-analyses, and most studies were scored moderate for quality. Seven studies were high quality with a low risk of bias. Thirty-six per cent of studies were from the USA or Canada, and only 20% from the UK, so study findings came from different health systems and might not apply to the UK.

What did it find?

  • Fifty-nine studies estimated the occurrence of patient safety incidents:
    • Twelve studies looking at ‘any type of patient safety incident’ reported between less than one and 24 incidents per 100 patient consultations.
    • These studies looked at patient records or prescribing data and found a median of around two to three incidents per 100 consultations/patient records reviewed.
  • The three most common types of safety incidents were administrative and communication incidents, diagnostic incidents and prescribing and medication management incidents. Ranges in estimates varied widely across the studies, due to different definitions, measurement approaches and quality of the studies, so it is not possible to estimate the precise frequency of these types of safety incidents.
  • Estimates of harm from safety incidents also varied widely across the studies. Seven studies that reviewed records estimated a median of 4% of incidents as being associated with severe harm (range less than 1% to 44% [see Definitions]).
  • Diagnostic and medication-related (prescribing) incidents were most commonly associated with harm to patients, with between 8% and 11% of medication incidents reported to result in harm of any severity.

What does current guidance say on this issue?

NICE provides recommendations about the safe and effective use of medicines in its 2015 guidance on medicines optimisation (NG5). This includes ensuring use of local and national reporting systems such as the National Reporting and Learning System, reporting all incidents consistently and in a timely manner, using tools such as patient safety alerts and considering assessing the training and education provided to health practitioners to identify and report medicines related patient safety incidents. The guidance also recommends considering a screening tool (e.g. the STOPP/START tool in older people) to identify potential medicines-related patient safety in particularly at risk groups.

What are the implications?

The global breadth of this WHO-funded review and the research challenges involved may mean that its main value is in raising awareness of patient safety in primary care, rather than providing accurate incident figures or making specific recommendations for UK practice. An ongoing NIHR study due for publication is looking at patient safety reported by healthcare staff in primary care across England and Wales to identify priority areas to minimise risk for future patients. The NIHR also funds the Greater Manchester Primary Care Patient Safety Translational Research Centre that promotes and generates patient safety evidence in primary care.

This review found that diagnostic and prescribing incidents were most commonly associated with harm to patients, which suggests that attention should focus on preventing these types of incidents. 2015 NICE guidance on medicines optimisation lists recommendations relating to prescribing, including encouraging health practitioners and patients to report all safety incidents (for example, in the National Reporting and Learning System).


Citation and Funding

Panesar SS, deSilva D, Carson-Stevens A, et al. How safe is primary care? A systematic review. BMJ Qual Saf. 2015. [Epub ahead of print].

This study was funded by the World Health Organization.



National Reporting and Learning System [internet]. NHS; undated.

NHS England. Patient safety in general practice [internet]. Leeds: NHS England; undated.

NICE. Medicines optimisation: the safe and effective use of medicines to enable the best possible outcomes. NG5. London: National Institute for Health and Care Excellence; 2015.

NIHR. HS&DR - 12/64/118: Characterising the nature of primary care patient safety incident reports in England and Wales: mixed methods study [internet]. 2015.

NIHR Greater Manchester Primary Care Patient Safety Translational Research Centre [internet]. Manchester; 2016.

O'Mahony D, O'Sullivan D, Byrne S, et al. STOPP/START criteria for potentially inappropriate prescribing in older people: version 2. Age and Ageing. 2014:44(2):213-18.

Produced by the University of Southampton and Bazian on behalf of NIHR through the NIHR Dissemination Centre

  • Share via:
  • Print article

In this review, primary care was defined as ‘first port of call’ generalist care, delivered outside hospital inpatient settings including the specialities of family practice, general practice, general internal medicine, general paediatrics, obstetricians and gynaecologists.

Harm was defined as impairing the structure or function of the body or mind including pain, nausea, psychological distress, disability or death.

No harm was defined as any patient safety incident that had the potential to cause harm but was prevented, resulting in no harm or that ran to completion but no harm occurred.

Low harm was defined as requiring extra observation or minor treatment and caused minimal harm.

Moderate harm was defined as resulting in a moderate increase in treatment and caused significant but not permanent harm (e.g. hospitalisation).

Severe harm was defined as resulting in permanent harm such as disability, death, or long-lasting physical or mental consequences.

Back to top