Beyond the ‘Go-Live’: why context matters in EHR implementations

Background

Electronic Health Records (EHR) systems are essential to the daily work of healthcare providers. However, the implementation or ‘go-live’ of a new EHR remains a costly and challenging endeavour.¹,³ Two recent editors’ choice papers in BMJ Health & Care Informatics examined how newly implemented EHR systems affect usability, workflow and clinical outcomes. In a cross-sectional survey of 1424 health professionals, Lohmann-Lafrenz et al⁴ compared the perceived usability of a newly implemented EHR across various occupational groups (eg, medical doctors and therapists) and clinical contexts (eg, oncology and laboratory medicine) as well as examined the relationships between EHR usability and provider burnout, insomnia and turnover intention. In this context, usability refers to ‘the extent to which technology can be used to achieve specific goals with effectiveness, efficiency and satisfaction’.⁴ Williams et al⁵ conducted clinical observation and analysed the actual use patterns of the Barcode Medication Administration (BCMA) system, a service available in their newly implemented EHR. While the scope and methods of these two papers^{4 5} differ, their synergistic findings offer a unified and convincing lesson that EHR implementations are challenging because we often do not adequately account for the variations in the clinical contexts where the implementations occur.

EHR implementation: context matters

After 6 months of implementing a new EHR system, Lohmann-Lafrenz et al⁴ found extremely low perceived EHR usability among healthcare professionals (median system usability scale score at 25 out of 100). However, contextual variations in usability were identified. For example, laboratory technician-rated usability is 1.67 times higher than that of medical doctors. Providers in radiology reported lower usability than those in internal medicine. Williams et al⁵ took a different approach in examining their EHR implementation by analysing the actual BCMA system use. From analysing 613 868 medication administration records over a 16-month period, they also found significant variations between hospital wards in barcode scanning compliance rates, overtime compliance trends and scanned barcode types (patient vs medication barcodes). For example, most wards showed declines in barcode scanning compliance, but one of them showed an improved trend.⁵ These findings show that a one-size-fits-all implementation strategy cannot produce the same results across diverse clinical environments because contextual factors (eg, clinical workflow) vary and matter. It is important to assess the contextual variations during the EHR implementation.

Appraise the sociotechnical system

The sociotechnical system approach, such as the Systems Engineering Initiative for Patient Safety (SEIPS) model, that analyses the interactions between technology (eg, EHR) and other contextual factors in a sociotechnical system (eg, workflow and user characteristics) may help identify the key issues contributing to these variations.⁶,⁸ Williams et al⁵ provided a good example. They analysed how workflow-related factors, such as medication administration rounds times and technology/tool-related factors, such as medication formulation (eg, tablets vs infusion), affect barcode scanning compliance. Through clinical observations, Williams et al⁵ further identified other system issues that negatively affected the scanning compliance, including the providers’ disbelief in the BCMA efficacy, uncertainty over the recommended workflow and a lack of confidence in raising concerns about inadequate technology. These misalignments between users, technology, organisation and other elements in a sociotechnical system led to observed workarounds of BCMA use⁵ and negative user outcomes—burnout, insomnia and turnover intention—identified by Lohmann-Lafrenz et al⁴ The knowledge gained through sociotechnical system appraisal during the EHR implementation is profound, offers critical implications and can inform improvement and future direction.⁴,⁶⁹

A call for sustained leadership, not just a go-live

Finally, the leadership support to establish a culture of safety and sustain a quality improvement effort is crucial because it empowers health professionals to identify the issues that arise long after the initial go-live and improves the chance for more successful long-term EHR adoption.^{9 10} The findings from Lohmann-Lafrenz et al⁴ and Williams et al⁵ demand a shift in focus from a one-time technical implementation to a sustained organisational commitment to continuous improvement. This requires leaders who understand contextual variations, show compassion toward EHR users and foster a culture of safety where staff can raise concerns without fear. Equally important is a commitment to acting on those concerns and supporting continuous improvements.^{9 10} Williams et al⁵ reported that the key reason why the one ward showed an improved trend (while others showed the opposite) was because the senior nurses in that ward openly discussed with the team about the issues that they encountered during BCMA implementation and worked together to resolve them. The profound knowledge we gain from system appraisals is only valuable if it is supported by leadership and used to inform a continuous, iterative improvement process—one that treats the human and contextual factors as being just as critical as the technology itself. Ultimately, technology alone cannot ensure safe and effective EHR use—people, processes and leadership matter just as much.