• adverse events
• incident reporting
• latent failures
• risk management

A key task in the enhancement of patient safety involves the ability to learn from error.¹ The cultural change needed to achieve this requires staff to report the errors and near misses they commit or see others commit, and to use these data appropriately to change policy and practice. In the UK the National Patient Safety Agency has been set up as a body for the collection of errors so that the lessons—written large at a national rather than a local level—can be appreciated more easily. However, this all depends upon errors being reported, and considerable research shows that this is very far from the case today.

The paper by Lawton and Parker in this issue of QSHC² is important in showing what types of errors are likely to be reported and by whom—which is useful if we are to bring about change where reporting is not taking place. It shows that nurses and, to a lesser extent, midwives are much more likely to report incidents than doctors; that reporting is more common where protocols are in place and not adhered to; and that reporting is also more likely to occur when patients are harmed by the error.

These results begin to show the ways in which errors are perceived by different groups. They show the importance of protocols, which govern nurses far more than they do doctors, and that near misses are likely to go unreported, as are errors which occur when staff have to improvise outside protocols. This means that the lack of formal recognition of these types of errors may therefore fail to provide the opportunity for the development of new guidelines in this less charted territory. The importance of using all types of error to bring about safer care needs emphasising to staff, but this can only be done in an atmosphere of trust.

We may be heartened by the finding that all staff are more likely to report errors that cause actual harm to patients. This may be because they see these areas as the most important to address. However, it is also true that reporting of such incidents is much more difficult to avoid than is the reporting of less serious errors or near misses. Ironically, it is probably easier to learn from incidents which cause only minimal or no harm to patients, and are therefore less emotionally charged, than from serious events which may be surrounded by guilt, anguish, and fear. Staff need to be encouraged to report incidents which lead to less serious outcomes, but this will only happen in a non-punitive atmosphere that allows innovation and learning to flourish.