How do inter-organisational electronic health records affect hospital physician and pharmacist decisions? A scoping review.

Abstract Objective To provide an overview of the effects of inter-organisational electronic health records on inpatient diagnosis and treatment decisions by hospital physicians and pharmacists. Materials and Methods Five-stage scoping review, using distributed cognition and the information value chain as guiding conceptual models. Eligibility criteria: empirical studies addressing how shared health records were used in inpatient clinical decision-making, published 2008-18. Sources: Healthcare Databases Advanced Search, covering nine sources including PubMed. Charting methods: data extraction form completed by one author, with inter-rater reliability assessment at title and abstract review. Results Quantitative studies (n=14) often reported relatively low usage of shared records (6.8% to 37.1% of cases). Usage is associated with reduction in diagnostic testing and readmission and variable effects on admissions and overall costs. Qualitative studies (n=6) reported avoidance of duplicate diagnostics, changing clinical decisions, the value of historical laboratory results and optimising the timeliness of care. We found no explicit use of explanatory theoretical models, but there is implicit evidence of an information value chain. We found only one study specifically about pharmacists. Discussion Relatively low usage is due to clinical judgement whether (extra) data is needed, given current knowledge of the presenting condition and relative complexity. We suggest that extensive EHRs need recommender systems to highlight (sometimes unexpected) relevant content, in parallel with professional guidance on indications for consulting shared records. Conclusions Clinicians only consult shared health records when they must. Mixed effects on process outcomes are due to the hidden variables of patient complexity, clinician judgement and organisational context.