Evaluation and improvement of the quality of the discharge prescription in cystic fibrosis children hospitalised for intravenous antibiotic therapy

Resumen

Background Cystic Fibrosis (CF) is a complex disease that requires multiple pharmaceutical treatment. Patients come to hospital to receive intravenous (IV) therapy as part of their optimisation therapy and part of their regular treatment is amended during their stay. At discharge time, the regular medication patients are on need to be mostly continued and there might be other therapies introduced that need assessment. The role of pharmacists caring for these patients has been documented, but no evaluation of revalidating discharge prescriptions and evaluation of potential interventions has occurred to date. Objective To evaluate and improve the quality of the discharge prescription in Cystic Fibrosis children hospitalised for intravenous antibiotic therapy. Methods The study took place at the Royal London Hospital for Children, Barts Health NHS Trust. Consent by the Clinical Effectiveness Unit was granted. The type of the study was ambispective observational with two years of retrospective phase and eight months prospective period. Inclusion criteria were paediatric CF patients admitted to respiratory ward to receive IV antibiotic therapy. Variables related to patient, to hospital stay and to drug therapy were defined. Discrepancies in drug therapy were assessed, and when the discrepancies were not justified then, these were considered medication errors. The medication errors were classified in 6 types: drug, dose, frequency, duration, pharmaceutical form and route of administration. A subtype of errors was also defined as omitted or committed error. After analysing the results obtained in the retrospective phase of the study, improvement strategies and safety actions were agreed and implemented to take place during the prospective phase. The quality of the prescription was evaluated through quality indicators defined. Results Tables 1 and 2 summarise the results obtained in the discharge errors found during the retrospective and prospective period with statistical analysis. Table 1. Type of errors found in discharge prescriptions Medication error Retrospective N=148 Prospective N=135 P Mean + SD (min, max) 1.48 + 1.38 (1,7) 4.35 + 5.27 (1,22) <0.001 Type of ME Drug N ; % (95%CI) 38 ; 25.68 (19.32-33.27) 5 ; 3.70 (1.59-8.38) <0.001 Dose N ; % (95%CI) 35 ; 23.65 (17.52-31.11) 4 ; 2.96 (1.16-7.37) <0.001 Frequency N ; % (95%CI) 3 ; 2.03 (0.69-5.79) 8 ; 5.93 (3.03-11.26) 0.091 Duration N ; % (95%CI) 25 ; 16.89 (11.71-23.75) 2 ; 1.48 (0.4-5.24) <0.001 Pharm form N ; % (95%CI) 44 ; 29.73 (22.95-37.53) 31 ; 22.96 (16.68-30.75) 0.198 Route of adm N ; % (95%CI) 3 ; 2.03 (0.69-5.79) 85 ; 62.96 (54.56-70.64) <0.001 Table 2. Subtype of errors found in discharge prescriptions Retrospective (N=148) Prospective (N=135) p Committed N ; % (95%CI) 78 ; 52.70 (44.69-60.58) 19 ; 14.07 (9.2-20.94) <0.001 Omitted N ; % (95%CI) 70 ; 47.30 (39.42-55.31) 116 ; 85.93 (79.06-90.8) <0.001 The global quality indicator in the retrospective phase had a value of 22 whilst in the prospective phase the value obtained was 41.82, with statistically significance differences (p=0.0095), although not achieving the defined value of quality achieved of 50. Discussion During the study there was a change of the computer system for discharge prescriptions. The evaluation was made by the same investigator in all cases, hence confounding factors were minimal. The new computer system had not been designed by clinical experts in paediatrics and prompt alerts to improve the prescribing system were not exploited, hence omission errors were the main cause of errors prospectively. The percentage of committed errors detected prospectively was less, indicating that the improvement strategies and safety actions introduced had an impact in the final results of the quality of the prescription. However, there was a high number of omitted errors found caused by technological aspects. If the computer system used in the prospective period had been the same as the one used in the retrospective phase, the likelihood of the percentage of errors would have been decreased as the new system lacked of a prompting alert for omitted information. Conclusions The deeper participation of the pharmacy team in the multidisciplinary team improve the optimisation of patient’s pharmacotherapy. The quality of the discharge prescription shows an improvement in the quality despite not achieving the standard value defined.