Improving the quality of drug error reporting

Background Drug errors are a common and persistent problem in health care and are also associated with serious adverse events. Reporting has become the cornerstone of learning from errors, but is not without its imperfections. Aim The aim of this study is to improve reporting and learning from drug errors through investigating the contributory factors in drug errors and quality of reporting in an acute hospital. Methods A retrospective, random sample of 991 drug error reports from 1999 to 2003 were subjected to quantitative and qualitative analysis. This was followed by 40 qualitative interviews with a volunteer, multi‐disciplinary sample of health professionals. The combined analysis has been used to develop a knowledge base for improved drug error reporting. Results The quality of reports varied considerably, and 27% of reports lacked any contributory factors. Documentary analysis revealed a focus on individuals, sometimes culminating in blame without obvious justification. Doctors submitted few reports, and there were notable differences in reporting according to clinical location. Communication difficulties commonly featured in causation, and high workload and interruptions were predominant contributory factors in the interview data. Interviewees viewed causation as multifactorial, including cognitive and psychosocial factors. Organizational orientation to error was predominantly perceived by interviewees as individual rather than systems‐based. Staff felt obliged to report but rarely received feedback. Implications and conclusion Drug errors are multifactorial in causation. Current reporting schemes lack a theoretical basis, and are unlikely to capture the information required to ensure learning about causation. Health professionals have reporting fatigue and some remain concerned that reporting promotes individual blame rather than an examination of systems factors. Reporting can be strengthened by human error theory, redesigned to capture a range of contributory factors, facilitate learning and foster supportive actions. It can also be feasible in routine practice. Such an approach should be examined through multi‐centred evaluation.