Katie Hafner did a great job with this article as it shows how the really smart doctors, like Dr Dhaliwal, use tools like Isabel. I hope that Dr Dhaliwal can be a role model for the “most of us (who) don’t think we need help at diagnosis”.
The quality of the 200 plus comments has been outstanding and is a testament to the NYT and its readership. However, much of the discussion still revolves around the issue of whether the computer is or will be better than the doctor. I believe that the real issue here is not the ‘competition’ between doctor and computer but when the medical profession as a whole will embrace these tools to make them collectively smarter.
Over the 12 years that we have developed Isabel, we have never once thought about whether Isabel could replace the doctor – that was never our intention. We built it as tool to help doctors build a differential diagnosis in the very limited time they have. We wanted to reduce the burden of memory and buy them time to think which, as Dr Dhaliwal says, is their “most important procedure”
For the record, I don’t believe that computers will ever replace the doctor’s role of diagnosis. What I do believe is what has happened since the beginning of human history which is that humans design tools which make them better at doing what they were doing before. So doctors, using computers, will become much better and more reliable diagnosticians. As Don Berwick once said “Genius diagnosticians make great stories, but they don’t make great health care. The idea is to make accuracy reliable, not heroic.”.
I get the greatest pleasure from hearing, for example, how a nurse practitioner using Isabel has diagnosed two patients with cancer that her physician colleagues didn’t think of. The computer, in this case, Isabel has extended her capabilities. Imagine how this could help in places where there is a shortage of medical expertise? It’s wonderful to hear Dr Dhaliwal using Isabel to second check his conclusions even though most of the time he was right and didn’t need to. The important point was that he recognizes the importance of double checking his diagnosis; he is not too proud to check as he knows how easy it is to miss something.
It is estimated that 40,000 to 80,000 Americans die each year from, almost certainly preventable, misdiagnosis. Countless more will experience unnecessary discomfort, pain and worry about an unnecessarily delayed diagnosis. Much of this could be prevented if tools, like Isabel, were widely adopted into routine practice. We need to move the debate on from whether a computer will replace the doctor to asking why these tools aren’t in common use today. Dr Dhaliwal has shown us how easily they can be used, even as an app on the phone. The stakes are too high to worry about whether doctors will be replaced by computers. We will always need doctors and, as medicine gets ever more complex, we need them to be consistently smart - which they can be by using computers. Without them, they have little chance and will be letting us all down just when we need them most.