PS-060 Agreement between potential drug interactions identified by an electronic tool and clinical judgment: intercheck versus physicians

• Published in: European journal of hospital pharmacy. Science and practice Vol. 22; no. Suppl 1; p. A160
• Main Authors: Antoniazzi, S, Chiarelli, M, Venturini, F, Nobili, A, Pasina, L, Casiraghi, C, Damanti, S, Periti, G, Rossio, R, Mannucci, P
• Format: Journal Article
• Language: English
• Published: London BMJ Publishing Group LTD 01.03.2015
• Subjects: Drug interactions; Medical records; Physicians
• ISSN: 2047-9956; 2047-9964
• DOI: 10.1136/ejhpharm-2015-000639.385

BackgroundThe software INTERcheck provides physicians with the potential drug-drug interactions (pDDIs) associated with a patient’s treatment, classifying them based on clinical relevance from the literature, as A (minor), B (moderate), C (major), D (contraindicated).PurposeTo assess the concordance between the clinical relevance of pDDIs as classified by INTERcheck and the physician’s personal judgement.Material and methodsThis retrospective study, conducted in 4 wards between April–October 2014, identified pDDIs from medical records of elderly inpatients aged ≥65 years, on ≥5 drugs, using INTERcheck. Clinical relevance as classified by INTERcheck was then compared with physician’s judgement through a structured interview consisting of four questions, as follows: is the pDDI known? is it clinically relevant?, if yes, why?, would knowledge of the pDDI at the time of prescription have changed your prescribing approach?Concordance between INTERcheck and the physician’s judgement was defined as: classification of “clinically relevant = yes” by physician and class C or D by INTERcheck; classification of “clinically relevant = no” by physician and class A and B by INTERcheck.ResultsMedical records of 60 inpatients were analysed: 1,658 drugs were prescribed, 481 unduplicated pDDIs were detected by INTERcheck and subsequently evaluated by physicians. Of those, 229 (47.6%) were unknown and 235 (49%) were classified by them as clinically relevant: 158 (67%) for potential adverse effects, 58 (25%) for patient complexity/co-morbidity, 19 (8%) for other reasons.According to INTERcheck, pDDIs were classified as: 12 (2.5%) A; 300 (62.4%) B; 113 (23.5%) C and 56 (11.6%) D.Concordance between the physician’s judgment and INTERcheck was : 83% (10/12) for A, 63% (189/300) for B, 73% (83/113) for C, 70%(39/56) for D. The knowledge of the pDDI at the time of prescription would have resulted in a change of treatment in 28.5% (137/481) of cases.ConclusionA high concordance between INTERcheck and the physician’s judgement was found throughout all INTERcheck classes. The lowest concordance was retrieved in class B. This finding will be taken into account to improve this database according to physician needs.References and/or acknowledgementsNo conflict of interest.