At our knitting and crocheting group this past Friday, we got onto the subject of AI, or artificial intelligence. I had just posted an article to my newsletter (Just a Job) on LinkedIn and we began to talk about the impact of AI on the fiber arts community.
To be sure, I know a LOT LESS about AI in that context than I do my primary professional sphere (healthcare), but it's an interesting phenomenon to consider in any area.
I have been thinking about the impact of AI and the advanced technologies that accompany its emergence through the lens of knitting, crocheting and yarn. I'll visit that in a future blog post, but wanted to begin with sharing this article, on AI in the healthcare space. I hope you enjoy it!
Top 5 Risks to Health Professions (my AI predictions)
I have a bit of a "reputation" for saying the unpopular stuff out loud and often before others are publicly acknowledging the issue. In 1997 I earned the scorn of a Medical Transcription supervisor who told me, with hands on her hips, that "They ain't never gonna figure out how to transcribe all them doctors with their accents,..."
As it turned out, "never" was just over 5 years, and the solution was discovered outside of healthcare when the events of September 11th 2001 made it a priority for the United States government to be able to monitor hundreds of thousands of phone and other conversations by people for whom English was a second language.
After being a reasonable profession for many years, Medical Transcription is a poster child for what can happen to a profession when technology takes over and can be an important bellwether if we are willing to pay attention.Â
"Yeah, so what's new?"
I'm glad that you asked.
A brief and admittedly unscientific review of past historical advancements that have displaced workers show that most of history is littered with the lost jobs, incomes and family sustaining work done primarily by the working class. This goes at least as far back to the buggy whip makers that were put out of business by Henry Ford's "horseless carriage" and extends all the way to and beyond the Medical Transcriptionists mentioned above. What I see as changing, right before our eyes, is the target of the technological shift. In other words, education and job titles aren't going to be as much of a shield as they once were when this new wave of technology takes hold.
Before I go on, I want to be clear: these predictions are my own. They are based on what I have observed in the healthcare delivery system (which I've been a part of for more years than I care to admit), my understanding of the emerging technologies, and general trends as reported in the larger news cycle. Some of these are "old news" but you may be surprised by my take on some others.Â
NOTE: as the hints of voice recognition technology were emerging as a distinct reality (~2015), everyone that I spoke with insisted that even with technology, transcription would need experienced people and that technology would not be able to do what they do.Â
That has not been the case, and it will continue to not be the case as we look at other areas in the healthcare space - areas that may surprise you AND tempt you to also say "Technology can never replace what I/we do,..."Â
Here are my predictions for significant technology disruptions within the healthcare sector, specific to roles that we have previously assumed to be untouchable:
Radiologists (MD)
Pharmacists (PharmD)
Oncologists (MD)
Physical Therapists (DPT)
Nutrition and Food Science professionals (RDN)
Some unifying traits among these roles include that they require advanced degrees and credentials to practice, and involve complex decision-making. While there are studies that hem and haw about whether AI can replace radiologists, those comments sound a LOT like what we heard around voice recognition software limitations and what we are hearing today around AI and medical coding.Â
In one study published in March 2023, autonomous AI returned "...a 99.1 sensitivity rate for AI on abnormal radiographs in comparison to a 72.3 percent sensitivity..." for reports read by a radiologist.Â
The data coming out of reporting like this leads me to predict that Pharmacists are next in line for AI vulnerability. For YEARS the model in pharmacies has been that the Pharmacy Technicians do customer service, ring up the prescriptions, hand out the meds and ask if the customer has any questions, which they would refer to the pharmacist. The Pharmacists generally provide the deep expertise, as do radiologists in the imaging space. This deep expertise is accessible through technology - and allows for a much broader pool of knowledge than 1 person can possess.Â
Oncologists may be having whiplash, as IBM's Watson came out with some promising capabilities around cancer diagnostics in the mid-2010's and provided an early "AI scare" in the medical profession. This was quickly shot down as the capabilities seemed not to be as robust as initially believed. but no one should take that to the bank.Â
Those same early fits and starts were common in the journey of medical transcription, and while I am very aware that the task of a transcriptionist and that of an oncologist are very different, the lesson is not in their roles and tasks, but in the evolution of technology. In the same way that AI is performing exceptionally well in imaging, it will "learn" and evolve to be as effective in areas like oncology.
While Physical Therapists are on my list, they are likely a bit further down the road in terms of disruption than their radiology and pharmacy colleagues, but they are not out of the woods. Ever see the workout mirrors that are interactive? They're not fully vetted yet, but stay tuned because the technology is there, and the capabilities continue to expand.
Similar to the need for fewer pharmacists, I predict that eventually, rehabilitation departments will be primarily staffed with Physical Therapy Assistants (PTA's) and patients will be monitored by the PTA team in conjunction with the interactive mirrors. The medical histories will be available from the Electronic Health Records, (EHR) and the initial assessment of pain, range of motion, stiffness, etc. can be evaluated by the PTA with validation and/or guidance by the expertise in the interactive workout mirror.Â
Lastly for this list, I believe that professionals in Nutrition and Food Science are at risk from AI. Again, similar to the pharmacy and radiology vulnerabilities, the Registered Dietician (RDN) evaluates, reviews the patient history and current status, makes recommendations based on their deep knowledge and expertise around nutrition,... all things that can be done TODAY with the support of advancements such as large language models (LLMs).
As in the others areas, this corner of healthcare can incorporate more technician-level people and fewer higher-credentialed folks to achieve the same, or perhaps even better, results.
Now, before folks take to their keyboards to argue and refute, let me make a couple things clear. I am not happy about these things, and wish no one any ill. I also happen to think that an ounce of prevention is worth MORE than a pound of cure, so yelling at me (the messenger) may make you FEEL better,... but in the same way that arguing with me in the 1990's about voice recognition technology and the transcription profession didn't stop it from taking over, being mad at me isn't going to impact AI's forward march into these other areas.
My "BONUS" category for today is management and in the short term, I believe that this will impact mostly non-clinical managers (people like me, actually).
The COVID-era remote work options have been WONDERFUL for work/life balance, but they have a dark side that some of us have noticed from the beginning. If we can manage our teams remotely - without putting eyes on them as they do their work; why can't their productivity and accuracy be monitored by technology?
The truth is that it can. Even before COVID sent many of us scrambling to set up home offices, there were companies monitoring remote employees not only for accuracy and productivity, but for actual time in front of their computer screens and engagement in work specific to their jobs.
Scary, right?
We should all be a little concerned about what this means, and not because some giant government robot is coming for us. The force of nature that is coming for us - ALL of us - is called the bottom line. The job roles I have highlighted are not entry-level individuals. These are professionals who command large salaries and generous benefits packages. Most of these folks are well within the 6-digit salary and benefits range, and as we know - labor costs are THE MOST SIGNIFICANT expense in most organizations.
The potential value proposition for ALL of these roles being replaced by technology is that healthcare systems can not only eventually save money by reducing FTEs, but they can also eliminate a lot of the support services that go along with 'X' number of employees. Some quick examples: AI doesn't call off, leave for higher pay, have biases that get in the way of critical thinking (although I do recognize that there is the potential for bias being programmed in at development), or have difficulty "working/playing well with others".
I realize as I come to the end of this initial article that at least 2 other major threads can come from this unfolding reality: 1) the impact to Higher Education, who is ALREADY struggling, and 2) the impact to our larger society when it's not just entry level folks getting sidelined by technology.
Sadly, it has been the case that when working class folks get their walking papers due to technology, people will say "they should have gotten more education and THEN they'd have job security".
In a world where doctoral level healthcare professionals can be replaced by technology, that "common wisdom" about the protective nature of advanced education and credentials begins to take on a whole new flavor.
These are hard conversations, I know. BUT,... ignoring the issues around technology (AI, specifically), stomping your feet and insisting that I'm wrong about it, or finding a couple articles that take a contrary viewpoint aren't going to change things. It's better to face the truth and be able to think about what needs to be done BEFORE we are in panic mode. And that goes for individuals as well as institutions.
Buckle up - this is going to be quite a ride.
(( originally posted on my LinkedIn newsletter, "Just a Job" ))
For now at least, I don't see many parallels between the disruptions to the healthcare space and the individual knitter or crocheter's space, however; I do think that the larger crafting/fiber arts community is experiencing a number of interesting shifts right now.
Some of this has to do with the general financial status (unease?) across the world. Other impacts include the proliferation of online shopping, the availability of online (and often free) classes on all kinds of crafting and fiber arts, as well as the life realities that accompany increased costs for housing, transportation and food.
I described to one friend that it feels like we are looking at today's issues through the lens of what has traditionally worked (e.g. 20th century solutions), but that's not working in today's world (the increasingly high-tech 21st century).
I wonder if our great-grandparents felt this in the early 1900's,...?!
I'm guessing they did, and the world figured it out,.. but the world ended up engaging in 2 World Wars. Maybe we should embrace the fact that a NEW way of thinking and problem solving needs to emerge in this new century, and hopefully, we'll figure it out without the need for a global war.
This is a bit depressing, so I think I'll close for now and go back to my mesh bag in progress. I have finished one side, and am crocheting up the second side. I'm excited to get it finished and use it for quick trips to the store!
This subject "pivot" aligns perfectly with a quote I have penned for my upcoming class on Yarn and Self-Care at the Pittsburgh Creative Arts Festival in August (2024).
"Crocheting and Knitting: lifelines, made of yarn, and crafted with hook or needles." ~ Stitch 'n Dish
Until next time,... keep on stitchin' (and go grab some yarn!)
(C) 2024 Stitch 'n Dish
Comments