Identifying ‘avoidable harm’ in family practice: a RAND/UCLA Appropriateness Method consensus study

Andrew Carson-stevens, Stephen Campbell, Brian G. Bell, Alison Cooper, Sarah Armstrong, Darren Ashcroft, Matthew Boyd, Huw Prosser Evans, Rajnikant Mehta, Christina Sheehan, Aziz Sheikh, Anthony Avery

Research output: Contribution to journalArticlepeer-review


Health care-related harm is an internationally recognized threat to public health. The United Kingdom’s national health services demonstrate that upwards of 90% of health care encounters can be delivered in ambulatory settings. Other countries are transitioning to more family practice-based health care systems, and efforts to understand avoidable harm in these settings is needed.

We developed 100 scenarios reflecting a range of diseases and informed by the World Health Organization definition of ‘significant harm’. Scenarios included different types of patient safety incidents occurring by commission and omission, demonstrated variation in timeliness of intervention, and conditions where evidence-based guidelines are available or absent. We conducted a two-round RAND / UCLA Appropriateness Method consensus study with a panel of family practitioners in England to define “avoidable harm” within family practice. Panelists rated their perceptions of avoidability for each scenario. We ran a k-means cluster analysis of avoidability ratings.

Panelists reached consensus for 95 out of 100 scenarios. The panel agreed avoidable harm occurs when a patient safety incident could have been probably, or totally, avoided by the timely intervention of a health care professional in family practice (e.g. investigations, treatment) and / or an administrative process (e.g. referrals, alerts in electronic health records, procedures for following up results) in accordance with accepted evidence-based practice and clinical governance. Fifty-four scenarios were deemed avoidable, whilst 31 scenarios were rated unavoidable and reflected outcomes deemed inevitable regardless of family practice intervention. Scenarios with low avoidability ratings (1 s or 2 s) were not represented by the categories that were used to generate scenarios, whereas scenarios with high avoidability ratings (7 s 8 s or 9 s) were represented by these a priori categories.

The findings from this RAND/UCLA Appropriateness Method study define the characteristics and conditions that can be used to standardize measurement of outcomes for primary care patient safety.

We have developed a definition of avoidable harm that has potential for researchers and practitioners to apply across primary care settings, and bolster international efforts to design interventions to target avoidable patient safety incidents that cause the most significant harm to patients.
Original languageEnglish
JournalBMC Family Practice
Issue number1
Early online date4 Oct 2019
Publication statusPublished - 2019


Dive into the research topics of 'Identifying ‘avoidable harm’ in family practice: a RAND/UCLA Appropriateness Method consensus study'. Together they form a unique fingerprint.

Cite this