Background Medication reconciliation is becoming a priority as a safety strategy in care transitions.

Purpose To evaluate the incidence of mistakes in the pharmacotherapeutic profile of polymedicated patients on admission and discharge, and to classify the discordances detected in relation to home medications in order to prioritise possible hospital pharmacist interventions.

Material and methods Polymedicated patients were preselected from primary care pharmacy services through the information system software with the following criteria: patients with 60 or more prescriptions from October to December 2014. Those with a registered admission in the electronic clinical record during this period were finally selected. Pharmacotherapeutic profiles were compared: primary care prescription (home medications)/admission treatment and discharge treatment/home medications. Discordances were classified into three groups: (1) omission: home medication that was not prescribed on admission or discharge without justification, (2) initiation: drug that was not a home medication and was prescribed on admission or discharge without justification and (3) discrepancy: drug initiated during hospital admission with no prescription in primary care after discharge.

Total frequency of errors and by group on admission and at discharge were registered.

Results 18 patients, 24 admissions. 604 drugs prescribed: 161 (26.5%) were mistaken; 104 (17.2%) by omission, 31 (5.1%) by unjustified initiation and 26 (4.3%) by discrepancy.

At admission, 299 treatments were reviewed, 68 were mistaken (22.7%), 37 (12.3%) being by omission, 20 (6.7%) by unjustified initiation and 11 (3.7%) by discrepancy.

At discharge, 305 treatments were reviewed, 93 were mistaken (30.5%), 67 (21.4%) being by omission, 11 (3.6%) by unjustified initiation and 15 (4.9%) by discrepancy.

Conclusion The rate of mistakes observed on admission show the need for reconciliation in care transitions.

The highest incidence of mistakes was registered at discharge. These mistakes carried forward to primary care prescriptions, given that treatment at discharge is taken as the reference. Therefore, it is necessary to add a pharmaceutical validation at patient discharge.

It is also necessary to have a common pharmacotherapeutic record and for it to be appropriately used by prescribers of both care levels. This would avoid sources of error such as transcription of medication or patient questioning and could be used as a reliable information source.

It is essential that the hospital and primary care pharmacists have a more active role in the development of strategies to forestall these errors.

No conflict of interest.