[Reliability of Primary Care computerised medication records]

To quantify and to evaluate the reliability of Primary Care (PC) computerised medication records of as an information source of patient chronic medications, and to identify associated factors with the presence of discrepancies.

A descriptive cross-sectional study.

General Referral Hospital in Murcia.

Patients admitted to the cardiology-chest diseases unit, during the months of February to April 2013, on home treatment, who agreed to participate in the study.

Evaluation of the reliability of Primary Care computerised medication records by analysing the concordance, by identifying discrepancies, between the active medication in these records and that recorded in pharmacist interview with the patient/caregiver. Identification of associated factors with the presence of discrepancies was analysed using a multivariate logistic regression.

The study included a total of 308 patients with a mean of 70.9 years (13.0 SD). The concordance of active ingredients was 83.7%, and this decreased to 34.7% when taking the dosage into account. Discrepancies were found in 97.1% of patients. The most frequent discrepancy was omission of frequency (35.6%), commission (drug added unjustifiably) (14.6%), and drug omission (12.7%). Age older than 65 years (1.98 [1.08 to 3.64]), multiple chronic diseases (1.89 [1.04 to 3.42]), and have a narcotic or psychotropic drug prescribed (2.22 [1.16 to 4.24]), were the factors associated with the presence of discrepancies.

Primary Care computerised medication records, although of undoubted interest, are not be reliable enough to be used as the sole source of information on patient chronic medications when admitted to hospital.