Background Patient safety is a priority issue for health services. There has been an increase in patients included in clinical trials and these are increasingly complex and therefore establishing safe procedures is crucial.

Purpose To identify potential causes of failure in all procedures followed in the area of clinical trials by the pharmacy unit, in a tertiary hospital, as well as estimating their potential effects and proposing possible improvements.

Materials and methods A multidisciplinary team involved in clinical trial samples was formed in order to establish all the sub-processes, the potential ways of failure and their possible causes. The higher risk modalities of failure were identified by estimating the risk priority number (RPN). To do this, on the basis of Failure Mode and Effects Analysis (FMEA), the severity of the failure and the possibility of its occurrence and its detection were estimated (scoring from 1 to 10). Preventive actions were suggested for those modes of failure with an RPN of > 100 and the new RPN value was calculated.

Results Eight sub-processes were identified: initial visit, reception of samples, prescription, custody, preparation, dispensing, destruction/return of samples and final visit. In total, 36 modes of failure were evaluated, with 61 causes and effects, whose severity value varied from 1 to 9. 24 RPNs had values higher than 100. The modes of failure showing greater reduction of risk after implementation of the measure were:

• Initial visit: for ‘incorrect verbal information’ due to an uninformed clinical research associate, the RPN decreased from 160 to 36 if the pharmacy unit would request the information in advance and in writing.

• Sample preparation:

  • In cases of ‘erroneous sample preparation’ due to inexperienced staff, the RPN decreased from 189 to 48 after elaborating preparation sheets prior to the beginning of the trial, specific for the medicine and aimed at nursing staff.

  • For an ‘incorrect design in the chemotherapy programme’, the RPN decreased from 360 to 56 if the sponsor validated it prior to the first preparation.

Conclusions The FMEA methodology is a useful tool for improving quality in the area of clinical trials. Its application allows the prioritisation of risk-prevention actions in line with their occurrence, severity and possibility of detection.

No conflict of interest.