Health Care is Institutionally Resistant to Technology March 9, 2017Posted by Peter Varhol in Technology and Culture.
Tags: health, technology
add a comment
That is an overarching and controversial statement, and is probably not true under all circumstances. I will only touch on a few points, based on this article in WSJ (paywall) and my own recent experiences.
The WSJ article notes a pretty complete failure of the University of Texas MD Anderson Cancer Center to leverage IBM Watson AI technology to help diagnose and treat cancer.
Of course my own recent experiences include a referral to what is purportedly one of the leading cardio institutes in the country, which asked me to fill out forms using a Number 2 pencil. Like I did when I was in elementary school. When I went to the website, there were obvious misspellings and bad grammar, including in their bragging about being a leading institution.
My doctor objected to my objection. “They don’t do their own website!” My response: “And they can’t even be troubled to read it, either. If you can’t get the easy things right, it leaves a lot of doubt that you can get the hard things right.”
I see a couple of forces at work here. First, health care remains incredibly complex. Every patient is different, and has to be treated with individuality. (To be fair, that is not how many human practitioners treat their patients, but that is a tale for another day). This approach may not be amenable to current machine learning endeavors.
That being said, however, it is clear that health care practitioners and institutions are rooted in routine and learned practice, and passively or actively resist new approaches. In a sense, it is sad that otherwise highly intelligent and educated people are so steeply rooted in their routines that they cannot adapt to changes for the better.
But the institutions and bureaucracies themselves force this attitude on many. It’s simply less friction to do things the way you always have, as opposed to trying something new. And that, more than anything, is where health care needs to change.
The Conundrum of Health Care December 12, 2016Posted by Peter Varhol in Technology and Culture, Uncategorized.
Tags: adrian gill, death, health, nhs
add a comment
I write about health care because now, about 18 months ago, I was under a death sentence. I ultimately didn’t die (I will someday, but it won’t be today), but such a situation does tend to focus your thoughts.
Health care in the US is mostly private, based on employment. In other countries, notably the UK, it is largely public. I came across the story of a writer in Britain, Adrian Gill, who passed of cancer a few days ago, and offers praises for the National Health Service there.
There is a fatal flaw in health care that societies in general refuse to acknowledge. Health care is not an unlimited resource. There is not an infinity of doctors, nurses, and hospitals. Drugs cost money. Out of necessity, health care must be rationed. Here in the US, we largely ration through, as I said, employment. That isn’t particularly satisfactory, of course. But in places such as the UK and Canada, they ration through availability. That isn’t satisfactory, either.
But it’s a discussion that no society is willing to have, and that’s the real problem. I have a friend who says, crassly but no doubt reflecting what many people truly think, “I love nationalized health care, as long as I can afford to buy what I need on my own.”
No one wants to say, “You can get as much health care as you can afford.” Or “You can have health care as long as you are willing to wait a long time for it.”
And there is the rub. They are the realities. But we ignore them because of the friction built into our health care systems. We can dismiss rationing as being an unintended consequence of the broken systems.
I do have one objection to Gill’s description of the NHS. He says that you don’t get the humanity in private health care that you do in public. I respectfully disagree. People are people, whoever signs their paycheck. We have a human connection. He was wrong.