Clinical decision support / Technology

The Intuitive Potential of the Clinical Decision Support Tool

January 19, 2016

Digital tools, in their essence, provide a means of making tedious and arduous tasks easier and faster. The Clinical Decision Support (CDS) tool, a functionality integrated into many EHR (Electronic Health Records), is a digital, evidence-based content tool designed to support clinicians in their decision-making process at the point of care. Despite its conceptual acuity, clinician adoption rates have been low. However, that shouldn’t disqualify the CDS tool from being highly impactful in optimizing both the clinician and patient experience. In fact, with only a few, minor usability changes, CDS tools are incredibly useful in the clinical decision-making process, and even critical in delivering the highest quality patient care.

The primary adoption hurdle amongst busy clinicians has been regarding excessive alerts and triggers that are commonly prompted by CDS tools during a patient episode of care. This trigger fatigue would lead clinicians to immediately dismiss and override any and all triggers, particularly in the order entry section of the EHR1. Unfortunately, this established the perception that the CDS tool was time consuming, and ultimately led to low compliance and acceptance rates2.

Recent user experience and preference studies conducted at the Hofstra North Shore Department of Emergency Medicine, in Manhasset, New York identified several keys areas of improvement to the CDS tool during the treatment of Pulmonary Embolism (PE) presentations at the Emergency Department (ED).

Researchers used the think-aloud methodology on ED residents, having them verbalize their rational and behavior as they interacted with a mock EHR and CDS tool. Results from this exercise were then analyzed by coders who iterated the tool based on the resultant impact on the users’ workflow. In Phase II of the study, the residents encountered near-live clinical scenarios of mock patients presenting with varying PE risk categories (i.e. low, intermediate, and high). Their computer screens were captured as they interacted with the CDS tool throughout the duration of the patient encounter.

This study revealed that clinicians interacted more positively with the CDS tool when triggers and alerts were placed upstream of their clinical decision-making process, upon the chief complaint, as opposed to downstream during order entry, when their management plan was less likely to change. They also stated that they relied more on the CDS tool when treating intermediate PE cases, where diagnosis was unclear, rather than in more predictable PE scenarios. Coders made chances to the CDS tool based on study results and direct user feedback, mimicking “real-world” user behaviour and integrating it with highly variable ED workflow as best as possible3.

Continued user analysis and feedback will foster further development of EHR-integrated tools that support and cater to the unique needs and preferences of a given institution, department, or clinic. The potential of these quality initiatives will lead to more intuitive tools that actually benefit clinicians and accomplish the task of standardizing, and ultimately, optimizing their workflow, rather than hindering it. Like many digital health technologies, CDS tools are still in their infancy, and open up innumerable, exciting possibilities for the future of health care.

Notes

  1. Ash JS, Sittig DF, Campbell EM, Guappone KP, Dykstra RH. Some unintended consequences of clinical decision support systems. AMIA Annu Symp Proc 2007:26-30
  2. Drescher DS, Chandrika S, Weir ID, Weintraub JT, Berman L, Lee R, Van Buskirk Patricia D, et al. Effectiveness and acceptability of a computerized decision support system using modified Wells criteria for evaluation of suspected pulmonary embolism. Ann Emerg Med 2011 Jun;57(6):613-621
  3. Press A, McCullagh L, Khan S, Schachter A, Pardo S, McGinn T. Usability Testing of a Complex Clinical Decision Support Tool in The Emergency Department: Lessons Learned. JMIR Human Factors 2015;2(2):1-11
0 likes