Background and importance Records of prescribed medication in the primary care setting have a high level of discrepancies regarding the medication that geriatric patients are really taking, when they are consulted at the hospital level.

Aim and objectives To review the rate of discrepancies between medication prescribed at the primary care level and medication that patients really need to take, in the geriatric population consulted with a multidisciplinary team in the hospital setting.

Material and methods An ambispective study was carried out by a multidisciplinary team (nurse, geriatrician, pharmacist) in all patients at a geriatric specialised outpatient office attended by a doctor in the hospital level setting, during the first fortnight of January 2020. An interview with every patient was carried out by a nurse, who was responsible for documenting in the medication records what every patient was really taking in that moment. Afterwards, a geriatrician made an evaluation of the clinical situation and adjusted treatment accordingly in every patient who attended. The pharmacist was then responsible for reconciliation of the medication. Demographic data, number of drugs prescribed, types of discrepancies and rate of acceptance by the physician were collected.

Results 34 patients, median age 85.5 years, were reviewed (8 men, 26 women), with a median of 10 (IQR 8.25–14) drugs prescribed. 61 discrepancies were detected with a mean of 1.8/patient (IQR 0–4). These discrepancies were: 47 (77.05%) inappropriate dosage, 8 (13.11%) drug omission, 2 (3.28%) drug duplication, 2 drug interaction and 2 commission. 46 discrepancies were reconciled with Turriano (tool for prescribing drugs in the primary care setting by general practitioners): 9 (19.56%) interventions adding drugs to the prescription, 15 (32.60%) referrals for drug discontinuing and 22 (47.82%) proposals to change the dosage. The remaining 15 were not accepted: 8 were unfunded drugs, 4 posology was conditioned to the clinical situation (intentional discrepancies), 2 due to ignorance of the prescription and 1 due to a computer problem. Only 4 patients (11.76%) did not present any discrepancy.

Conclusion and relevance The high percentage of patients with discrepancies in Turriano represents a significant safety problem for patients. In this study, a large number of discrepancies were found and corrected, leading to an improvement in quality of treatment and patient safety. These interventions are essential in elderly, multipathological and polymedicated patients.

Conflict of interest No conflict of interest