Objectives: To compare two methods of data collection of patient-practitioner encounter data in general practice: from medical records consultation notes and from data recorded on encounter forms.
Method: Data were collected from two sources: (i) Medical Records Study: a study of the efficacy of an intervention designed to improve the quality of medical records provided details of 3107 patient encounters with 163 general practitioners (GPs) which had been photocopied from medical records; and (ii) Australian Morbidity and Treatment Survey (AMTS): from a national sample of 495 GPs and over 100,000 patient encounters, data from 47 GPs in the same geographical area as those in the Medical Records Study provided encounter forms for 10,392 patient encounters including details about patient demographics, reason for encounter, management, treatment, tests and investigations, admissions, referrals and planned follow-up. The International Classification of Primary Care (ICPC) was used to code reasons for encounter and problems managed. Drugs were classified according to an in-house classification by generic name and broad drug group.
Results: Patient details and all items of clinical information were recorded less frequently or were more often illegible in medical records than on encounter forms. There was a higher rate of management of problems classified as general or non-specific in the medical records. A lower prescribing rate for drugs acting on the cardiovascular system was recorded in the medical records, but higher rates were found for antibiotics, drugs acting on the immune system and miscellaneous drugs. Coding of all data was more reliable both between and within coders using the data from the encounter forms compared to the medical records.
Conclusion: General practice data obtained from encounter forms are more comprehensive and are coded more reliably than those drawn from medical records.