Do Canadian Doctors Hate Their Computers?

Atul Gawande’s thought-provoking New Yorker article, entitled “Why Doctors Hate Their Computers” has caused quite the stir in the health tech community, while sparking some lively dialogue here at Think Research. Our President, Founder and practicing critical care physician Dr. Chris O’ Connor recently sat down with Will Falk, Executive-in-Residence, to discuss the applicability of the article to the Canadian healthcare setting.

Both Chris and Will are huge fans of Dr. Gawande’s ideas and writing, and in many ways, The Checklist Manifesto is the intellectual antecedent for the founding of Think Research.

WF: Hey Chris. Do doctors hate their computers?

CO: It’s an excellent question, and Atul Gawande’s article uncovers real issues, with serious consequences for both patients and providers around the world. While it may not be true for all doctors everywhere, doctors can often hate their computers.  

The first reason, as Gawande describes, is software complexity. With the advent of complex clinical information systems, simple tasks have became more difficult to carry out. Even something as simple as creating an order for a test can now take multiple mouse clicks to complete.

The second is that the work previously carried out by allied health professionals has now shifted onto the physician (what Gawande calls the “Revenge of the Ancillaries”). This both burdens physicians and disempowers the rest of the healthcare team, as healthcare workers are now rigidly stuck in their roles and are often unable to help each other. As a result, the team becomes frustrated and disconnected.

The third problem is documentation. Gawande’s article explores the notion of software creating clinical notes that become a  “massive monster of incomprehensibility” for today’s clinicians. Computerized systems are allowing clinicians to create ever-longer notes that are difficult to read or extract useful information from. As Gawande describes, these notes are “long, deficient and redundant”, utlimately lacking in meaning.

WF: In your opinion, do you see this situation as better or worse in Canada?

CO: In Canada, paradoxically, we may be a bit better off, just because our healthcare system has been slower to implement computerization than the US. Also, the Canadian system is less driven by billing requirements than the American system, where many of the big hospital information systems have to serve both clinical and billing needs (making them cumbersome). Yet despite this, the problems identified by Gawande are still affecting Canadian clinicians and absolutely need to be addressed.

WF: Chris, I often hear about how healthcare clinical notes can provide a rich source of data, but healthcare providers still don’t have proper, consistent clinical documentation. What are your thoughts on that statement?

CO: Gawande touches on some dimensions of the problem with usable clinical notes, but doesn’t, in my opinion, get to the core issue at hand, which is that clinical notes, at this point in time, are largely  free text narrative ‘blobs’. Working with these can be extremely difficult for clinicians, and as these ‘blobs’ grow, the burden of working with them only gets worse. In this case, more is not necessarily better.

WF: And paradoxically, EMRs’ copy and forward functionality makes it more enticing for clinicians to just copy and paste previous notes, making them even longer.

CO: Exactly! In his article, Gawande also mentions that handwriting forces paper notes to be brief and to the point which can make them a better source of information than very long dictated or copied EMR notes.

WF: I know you’ve been working on this problem for a long time – initially in your own hospital (Trillium Health Partners) and now at other hospitals in Ontario and Ireland. What needs to be done?

CO: We have to entirely rethink our approach to clinical documentation, moving from a free text paradigm to a structured, orderly knowledge-based approach. Clinical documentation  needs to accommodate today’s patients, who face multiple comorbidities and acute illnesses – and who are cared for by large interprofessional teams, across multiple phases of care.

In his article, Gawande outlines “one of the fastest-growing occupations” in healthcare:  “medical scribes”. This field involves “trained assistants who work alongside physicians to take computer-related tasks off their hands.” It’s a deeply ironic ‘solution’ to the problems generated by software complexity and challenges of free text documentation. Until the documentation paradigm shift happens, and until we collectively move away from free text documentation, even this solution can only partially address the problem.

At Think Research we see documentation as one of the most pervasive challenges facing global healthcare systems today. It’s an even bigger problem than ordering (or Order Set checklists) as documentation exists everywhere, across every clinical setting. Since clinical documentation is such a fundamental driver of quality care, rethinking our approach to it has the potential to radically transform healthcare systems around the world.

WF: So if the Checklist Manifesto was the start of one revolution, has Gawande started another revolution? Where do we go from here?

CO: I would say that Gawande’s latest article identifies very real problems and sets the foundation for addressing them (similar to The Checklist Manifesto). Having Gawande articulate these concerns will make it harder for people to dismiss these issues. Ultimately, it’s up to both software developers and clinicians to bring those two worlds together to drive real change. This mission is one we’re working towards here at Think Research, particularly with our Order Sets and Progress Notes solutions. The journey is far from over, but we’re making meaningful strides everyday.

This post is part of a series, sharing perspectives on key issues in the healthcare IT space. Stay tuned for the next post in the series!