(32% response rate) showed that the feedback report had an appropriate format and length (median scores 3/5), although only 33% of the stewards had disseminated the results to colleagues. Conclusion and relevance Awareness of electrolyte disorders increased among physicians, but the direct impact of our feedback remains unclear. Other QIs showed little room for improvement and need re-evaluation. Overall results suggested a persistent need for training on intravenous fluids, especially on surgery wards, and feedback should include tailored communication with staff.

satisfied the following criteria: for undocumented infection, discontinuation of probabilistic antibiotic therapy at 72 hours of apyrexia; for documented infection, continuation of documented antibiotic therapy, according to the recommendations of the local antibiotic guidelines. Results Ninety infectious episodes were studied. The study population comprised 49 men (54%) and 41 women (46%). Average age was 56 years. Cefepime or piperacillin/tazobactam were systematically introduced as probabilistic therapy. If the infection was undocumented (n=61/90), the duration of probabilistic antibiotic therapy conformed in 41% of cases (n=25/61). For clinical documentation (n=6/90), the conformity rate was 67% (n=4/ 6). For microbiological documentation (n=23/90), compliance rate was 74% (n=17/23). Conclusion and relevance For most undocumented infections, probabilistic antibiotic therapy was prescribed for too long. This may be explained by the fragility of haematology patients and the fear of being confronted with recurrence of infection. For documented infections, conformity was very satisfying, as haematologists have extensive knowledge of infectiology. In order to harmonise prescription duration and continue to prevent the emergence of bacterial resistance, a guide for correct use of antibiotics and a second prospective study should be considered.

Background and importance Cancer patients are a vulnerable population for SAR-CoV-2 infection. Aim and objectives The aim of our study was to describe the epidemiology and clinical course of patients with cancer infected with SARS-Cov-2, attending hospital. Material and methods A retrospective observational study was conducted in cancer patients attending a tertiary hospital for SARS-CoV-2 infection during the period 3 January 2020 to 31 May 2020. Demographic and clinical variables were analysed: comorbidities, tumour diagnosis, tumour stage and whether they had received anticancer treatment in the last month (active treatment). The clinical course was evaluated by hospital admission, pneumonia, oxygen therapy requirements, the development of acute respiratory distress syndrome (ARDS), admission to ICU, mortality rate and mortality rate <30 days from admission. Quantitative variables were expressed as means (SD). The association between dichotomous variables or proportions was compared using Fisher's exact test and between quantitative variables using the Mann-Whitney U test. Results 112 patients were included, 59.8% (67) were men, mean age 67±13.4 years. 94.6% (106) were Caucasian (4.4% (5) Latino). 61.6% (69) were non-smokers, 25% (28) exsmokers and 13.4% (15) current smokers; 11.6% (13) had obesity. The most frequent comorbidities were: 57.1% (64) arterial hypertension, 34.8% (38) cardiovascular disease, 32.1% (36) diabetes mellitus and 21.4% (24) COPD. The most frequent cancer diagnosis were: 18.8% (21) breast cancer, 17.9% (20) lung cancer, 16.1% (18) colorectal cancer and 12.5% (14) prostate cancer. Tumour stage: 55.4% (62) metastatic disease, 25% (28) localised disease and 19.6% (22) locally advanced disease. 60.7% (68) of patients received active cancer treatment (42.7% chemotherapy, 32.3% hormonal treatment, 16.2% targeted therapy, 7% immunotherapy and 2.9% radiotherapy). At admission, 85.7% (96) of patients

BackgroundParenteral nutrition has been classified as a high-alert medication. In recent years, quality organisations such as the Joint Commission require hospitals to conduct proactive risk assessments of high-risk processes.PurposeTo describe the utilisation of Failure Modes, Effects and Criticality Analysis (FMECA) as a tool to evaluate the impact of the improvements implemented in the adult parenteral nutrition process.Material and methodsAs part of the departmental risk management strategy, a multidisciplinary team (two hospital pharmacists, two nurses, a technician and a safety specialist) were recruited for the analysis of the process. The team listed all the failure modes and the possible causes and effects. For each failure mode, the team assigned a score for likelihood of occurrence (1–10), severity (1–10) and likelihood of detection (1–10). Finally, the Risk Priority Number (RPN) was calculated by multiplying the three scores.ResultsThe process in the year 2008 included: manual prescription, manual transcription to the compounding software, validation, preparation and check of the medication tray, compounding in the laminar airflow hood and visual inspection of the parenteral nutrition and the used products. In the year 2016 the process included: a computerised physician order entry (CPOE) software, an automated transcription interface from CPOE to the compounding software and a built-in gravimetric end product quality control.For the process in the year 2008, a total of 32 failure modes were listed and an overall RPN of 3518 points was calculated. Manual prescription (1,188), manual transcription of the fax-transmitted prescription (665) and compounding (542) reached the highest RPN. Fifteen high-risk failure modes (RPN >100 points) were listed. After the implementation of the improvements, in the year 2016 only three high-risk failure modes were found. The total number of failure modes decreased to 31 and an overall RPN of 1540 points was calculated. The highest RPN were found in the medication tray preparation (504) and compounding (394) subprocesses. The most noticeable improvements were obtained with the implementation of CPOE (111) and the transcription interface (17).ConclusionFMECA was considered a valuable tool for the detection of areas for improvement and helped monitoring the effectiveness of the improvements after their implementation.No conflict of interest

interstitial pneumonia and 1 anxiety attack that forced a change in treatment) and 5 with the use of tofacitinib (1 herpes zoster, 1 dry lip, 1 tinnitus, 1 oedema and 1 dyslipidaemia). Of the 21 patients with baricitinib, 4 changed treatment due to ineffectiveness (2 to tofacitinib and 2 to biologics), and of the 11 treated with tofacitinib 2 switched to biologics and 1 suspended treatment due to cardiovascular risk. Conclusion and relevanceIn our clinical experience, baricitinib and tofacitinib are shown to be effective in the treatment of RA, with a good safety profile.