There is currently very limited reporting of patient safety incidents in the UK NHS. The National Reporting and Learning System (NRLS) has been shown to capture significantly fewer events than actually occur in practice. This problem is particularly severe in primary care, perhaps due to clinical staff being unaware of what exactly should be reported and because of limited time during consultations to report adverse events.
The NHS currently publishes a “Never Events” list for secondary care. Never events are wholly preventable serious incidents, and the list currently includes 14 such events including wrong-site surgery and in-hospital overdoses of certain drugs.
Although such a list does not currently exist for primary care, recent work by De Wet et al polled a quarter of Scottish GPs to determine what they believed constituted a “never event” in primary care, producing a list of 38 potential events.
This study will attempt to determine the incidence of a selection of potentially avoidable sentinel safety events in UK primary care by working with primary care consultation data with a view to developing standard algorithms that could be used within primary care data to flag these events for quality improvement purposes.
This study will attempt to capture the incidence of sentinel safety events in primary care (see appendix for proposed events)These events broadly fall into three categories: diagnostic errors, prescribing errors and administration errors (e.g. referrals not being mailed to patients).
The approach taken to capture these events within the data will vary depending on the type of event.
For example, when trying to capture the incidence of prescriptions made to those with allergies to the drug, the field “prescr” will be used to determine whether a patient has previously been recorded as having a prescribing exemption. This will then be compared with the “bnfcode” field to determine whether the patient has been prescribed the drug they have previously had problems with. Another example is following up planned referrals for cancer, which will utilise the HES data linkage to determine whether or not a referral has resulted in a subsequent hospital visit.
Paul Aylin - Chief Investigator - Imperial College London
Danny Furnivall - Corresponding Applicant - Imperial College London
Alex Bottle - Collaborator - Imperial College London
Azeem Majeed - Collaborator - Imperial College London
HES Admitted Patient Care;HES Outpatient;ONS Death Registration Data;Patient Level Index of Multiple Deprivation;Practice Level Index of Multiple Deprivation