Paperless but not effortless: EHR design limits in a digital lab ordering workflow

Introduction Previously, outpatients at Amsterdam UMC brought printed laboratory request forms to the phlebotomy desk. These forms listed the ordered tests via a grouper number and carried patient instructions. In October 2023, routine printing for patient-portal users was stopped to reduce paper use. We describe the implementation of this paperless workflow and its impact on clinicians, phlebotomy staff (hereafter referred to as “staff”) and patients.

Approach In the new workflow, patient identification and date logic became the main anchors, without guidance of the grouper number. Staff now select orders in Epic based on expected sampling date (today to +1 day) or, when marked “approximate” date, within +2 wks; additional visual cues were added. Governance relied on HFMEA/SAFER risk analysis and multidisciplinary consensus. Communication and training were provided through multiple channels for clinicians, staff and patients. Some tests were reconfigured so key instructions show digitally.

Evaluation After implementation, paper use in this process was reduced by about 70%, but several operational issues emerged. Order selection in Epic was error-prone: each test is a separate order, complicating correct selection. Fixed versus “approximate” dates were sometimes misunderstood, and changes in treatment schedules were not consistently mirrored in orders, causing missing or mistimed draws and requiring order re-entry. Staff incorrectly assumed that overtesting would increase patients’ out-of-pocket costs and therefore devised their own workarounds, which in turn increased workload. Conflicting expectations between staff and patients, combined with limited digital information for patients, led to discussions, uncertainty at the desk, delays, frustration and occasional aggression. These issues were more pronounced at one site than at the other.

Discussion Despite thorough preparation over more than a year, including governance, communication and training, the paperless workflow remained fragile, indicating that change management alone cannot overcome electronic health record design limits and restricted patient-visible fields. Better alignment of clinic schedules and lab orders requires stronger Epic support. Careful selection of both enthusiastic champions and critical contrarians from user groups helps surface issues and gauge buy-in, and should be combined with monitoring of process and safety metrics.