Patient safety research in primary care: where are we now?

RECENT PROGRESS

This issue of the BJGP reflects recent progress in patient safety research in primary care with the publication of three articles addressing safety culture and teamwork in community care; harms following transfer of care responsibilities between primary care, secondary care, and other sector services; and the approaches to clinical reasoning that are associated with diagnostic error.^1–3 Traditionally, research in patient safety has focused on hospital-based, specialist care provision. The epidemiology of patient safety in these settings is established: around 1 in 10 patients experiences avoidable harm.⁴ Despite 90% of healthcare encounters occurring in the community setting in most developed nations, there has been an assumption that, due to the lower-risk nature of patient encounters in primary care, harms will be less significant.⁵ This claim is hard to disprove until robust population-level epidemiological studies are conducted to determine the frequency and burden of harms occurring in primary care.

A patient safety incident is any unintended or unexpected incident that could have harmed or did harm a patient during healthcare delivery. This can be the result of a wrong or inappropriate action (‘error of commission’) or failing to do the right thing (‘error of omission’).⁴ Current estimates suggest that 1 in 50 patient encounters in primary care will result in a patient safety incident and, of these, substantial patient harm occurs in 1 in 20.⁴ As over 340 million consultations are undertaken in general practice in the UK every year, this equates to substantial harm affecting, in the region of, 300 000 patients every year. When considering other areas of primary care such as dentistry, pharmacy, and nursing, the harm could be appreciably higher.

THE COMPLEXITY OF PRIMARY CARE

Strong primary care systems are advocated for better population health outcomes, economic, and patient preference reasons. Interventions to improve patient safety need to consider the diversity of settings, variety of patients, different clinical conditions, non-specific symptoms and undifferentiated presentations, and the vast array of healthcare professionals. Furthermore, primary care is changing, with care provision and responsibility being shifted from secondary to primary care at a time where concern has already been raised about excessive GP workload affecting patient safety.⁶ Additionally, GPs, who specialise in complex disease management, also help coordinate the social care and healthcare needs of their patients. This includes vulnerable patient groups such as those with multimorbidities and those at the extremes of age, who are at higher risk of patient safety incidents.⁵ Regardless of this, GPs are still the most trusted professionals in the eyes of the public. Even in the post-Shipman, Bristol Royal Infirmary, and Mid Staffs era the public still place enormous trust in their GPs and their ability to care for them. The expectation is that general practice is very safe.

CREATING URGENCY FOR CHANGE

The World Health Organization’s Safer Primary Care Expert Group first met in 2012 and carried out an international Delphi prioritisation exercise to identify cross-cutting priorities for patient safety research and development in primary care in low-, middle-, and high-income countries. Participants confirmed the need to recognise the importance of unsafe primary care, a willingness to share data, support for quality improvement, and practical proposals to bridge knowledge gaps; suggestions for action were made.⁷ So, what research and improvement initiatives have been undertaken in the UK to date?

To recognise the importance of unsafe primary care, the initial step is to understand the problem. Early work noted marked diversity in the reported frequency and nature of errors.⁸ De Wet and Bowie⁹ used the trigger review method to identify unsafe care in primary care and found it an effective tool to establish previously undetected harm. Further work on priority setting¹⁰ generated a list of ‘never events’ for general practice through practitioner and consensus-building methods, and there is now work underway to determine the incidence of missed diagnostic opportunities in English general practice.¹¹

Classification systems have been developed to analyse primary care patient safety incident reports.^12,13 Carson-Stevens and colleagues analysed over 13 000 primary care patient safety reports from the NHS National Reporting and Learning System, characterising the incidents that are being reported by healthcare professionals and the severity of harm outcomes.¹² The number of reports is the largest in the world and has never previously been systematically analysed. The volume of these data supports the identification of themes for priority setting and intervention, and the generation of hypotheses about the underlying causes of safety incidents. The work has already resulted in publications on childhood vaccination,¹⁴ with suggestions for safety improvement during hospital discharge included in this issue.²

Analysis of incident reports cannot reflect true epidemiology because it is subject to reporter bias, especially for clinical decision making. Avery and colleagues are conducting a retrospective case-notes review study in a random sample of general practices in England using triggers such as a hospital admission and a range of other factors to identify avoidable patient harm (T Avery, personal communication, 2015). This is important because it embraces the concept of errors of omission, as well as commission, as the role of the GP includes prevention as well as cure. It is only by understanding how and why patient safety incidents are caused in primary care, along with their contributory factors, that learning can be derived and systems set up to prevent such incidents reoccurring. For example, Wellesley and colleagues conducted a case-notes review in primary care of over 1000 patients living with HIV and found evidence of poor documentation of specialist contact and antiretroviral medication, with some contraindicated drug coprescribing.¹⁵ They also highlighted a higher rate of HIV diagnosis in general practice than the national average, and indicated conditions that people commonly present to primary care with that may aid earlier diagnosis. The work has led to recommendations to address these issues and endeavours to increase the testing and diagnosis of HIV in primary care.

WHAT CAN BE DONE NOW?

While awaiting the outcomes of the national studies discussed here, toolkits are already available for use in identifying areas where efforts can be made to improve patient safety for the individual GP, and at practice level.^16,17 The Royal College of General Practitioners has recently launched an online Patient Safety Toolkit,¹⁶ building on the Scottish Patient Safety Programme in Primary Care.¹⁷ Both can be used by primary healthcare professionals to identify (and tackle) safety problems in general practice, and to contribute to personal appraisal and revalidation. Another suggestion for action is the pharmacist-led IT intervention (PINCER) that was shown in a randomised controlled trial to reduce a range of medication errors.¹⁸

Due to the diversity of general practice contexts, local solutions with co-production of protocols and improvement interventions by multidisciplinary practices or locality-based teams will often be needed to improve patient safety. As a specialty we need to embrace the research and utilise the tools that we currently have available, so that we can uphold the public expectation that primary care is safe for patients.

Provenance

Commissioned; not externally peer reviewed.