How accurate is the electronic health record? - a pilot study evaluating information accuracy in a primary care setting.

BACKGROUND

Electronic health records (EHR) are increasingly used for both administrative and clinical tasks with major implications for patient safety and quality of care. This study aims to determine a baseline EHR level of accuracy present on measurable information fields within an Australian general practice.

METHODS

Quantitative and descriptive pilot study of patients attending a private general practice. Patients who consented to participate in the study had their patient records reviewed to determine how many items were correct, incorrect or not recorded in each EHR information field. Statistical analysis was performed on the data collected.

RESULTS

A total of 33 patients gave consent to participate in this study. High levels of accuracy were found in the area of demographic details (94%). Moderately high levels of accuracy were reported for allergies (61%) but also a considerable percentage of non-recorded information was present (36%). Inaccuracies in medication lists were reported in 51% of records reviewed with 32.1% of all medications being inaccurately recorded. While over 91% of participants had a history summary with eight or less items present, omissions were reported for one in every five participants. There were no significant associations present between inaccurate data and frequency of practice visits or those with more than five past medical conditions listed in the EHR.

CONCLUSION

The study has confirmed that errors and inaccuracies exist in EHR in our Australian pilot study. The pilot study has also allowed us to complete a trial ensuring that a study of this type can be done safely and with correct methodology. As health informatics plays an increasingly important role in health care, studies of this type will better inform practitioners/ researchers in designing systems to ensure quality electronic patient information.