Although there was still some way to go to know how best to protect patients in primary care, safety research in other settings suggested it should begin with promoting a "safety culture".
Dr Wallis said the safety culture concept referred to the shared attitudes, beliefs, values and assumptions underlying how people perceived safety and how they acted on safety issues.
It was a systems or team approach, rather than the traditional approach with its emphasis on individual error.
Errors in primary care included delayed diagnosis, missed diagnoses, adverse reactions to drugs, administering the wrong drugs, losing results or not communicating them to the patient, and lost referral letters.
"Increasing awareness of patient safety in the developed world has mainly focused on the care provided in hospitals. However, there is research showing that patients are harmed in the primary care setting as well."
In her PhD research Dr Wallis received funding from the World Health Organisation World Alliance for Patient Safety to use one of the tools assessing safety culture, the Manchester Patient Safety Framework, in 12 Dunedin general practices.
The tool was not designed to be judgemental, but allowed practices to look at their attitudes and values, evaluate them and see where improvements could be made to reduce harm.
In her first visits to practices, Dr Wallis found some staff did not see patient safety as a separate issue from matters of occupational health and safety.
On her second visits, six months later, she found some had made changes .
These included having regular meetings where safety was on the agenda, improved error-reporting systems or the way they followed up results or recalled patients.
Public hospitals have to report serious events annually and Dr Wallis can see a time when there could be a national system for reporting general practice errors, but it should not rely on counting the mistakes yearly and then comparing, to see if fewer were occurring.
Asking practices to report over a month with the emphasis on evaluating and learning from mistakes, both their own and others', would be preferable.
"If you expect practices to report all the time, they would spend all their time reporting."
Dr Wallis said she could see the Manchester framework becoming more widely used in practices, particularly as it had the backing of the Royal College of General Practice and was suited to both small and large practices.
Dr Wallis will deliver an address on her research today at the New Zealand Bioethics Conference, being held at Salmond Hall, Dunedin.
Elements of a safety culture
• A reporting culture which allows easy reporting of errors and near-misses.
• A just culture where staff are encouraged and rewarded for reporting errors.
• A flexible culture with minimal hierarchy.
• A learning culture where there is a willingness to learn from mistakes and make changes.
- Prof James Reason, University of Manchester
