Systems thinking has revolutionised the way we think about improving safety in healthcare. Rather than considering errors as being due to individual mistakes, lessons from the aviation and other industries show that errors are often built into the design of healthcare systems and processes. Reason’s well-known ‘Swiss cheese model’ shows how major errors can arise from multiple small defects (the ‘holes’ in the cheese) in the existing system of care. These errors might be averted by understanding where these small defects might occur and by putting in further safeguards to prevent them.

‘Trigger tools’ can aid systematic detection, quantification and classification of errors. They are instruments developed to measure the rate of harm or changes in harm using a structured record review in a variety of healthcare settings. Another method is reviewing malpractice claims. In primary care, claims are most commonly due to delayed or missed diagnoses and prescribing errors.

A better understanding of when and where diagnostic errors might occur and why these might lead to delays in treatment can help to reduce these failures. For example, greater knowledge of common prescribing errors can allow prompts and reminders linked to prescribing safety indicators to be integrated into computer systems.

Serious errors, whether these might occur in a community or acute setting, are designated as ‘never events’, defined as ‘serious, largely preventable patient safety incidents that should not occur if the available preventative measures have been implemented by health care provider’. Examples of ‘never events’ include wrong-site surgery; serious medication errors, such as daily rather than weekly administration of methotrexate, or failure to refer a patient suspected of having cancer. Systematic measures to prevent ‘never events’ can help to provide healthcare more safely.