AI vs burnout
Veterinary hospitals all over the country are understaffed, overworked, and burned out. How is a chatbot going to save us?
ipopba / stock.adobe.com
I use OpenAI’s ChatGPT-4 every day. I use the chatbot to search the internet, as a copy editor for my writing, and to help me remember the name of a song when I can remember only a clip of the lyrics. It’s delightfully pluripotent, and the analysis, image generator, and internet browser are all combined right now. It’s wildly useful in writing professional communications, organization, ideation, and administrative tasks. And the new analysis feature computes math problems using a Python code for calculation.
As a daily user of something that augments almost everything I do on a computer, I forget that two-thirds of Americans haven’t tried it yet.1 It’s the naivete of the privileged to assume that everyone’s experience is similar to your own, and that’s the mistake I made. I happily go chasing the next interesting bit of knowledge and neglect to share the variety of ways that this technology has helped my practice.
But, recently, a conversation with a colleague reminded me of the value of this technology and the ways in which it can help alleviate burnout in the veterinary profession. Galyna Danylenko, a communications expert and the cofounder of Full Slice Agency, suggested I revisit the burnout issue through the lens of artificial intelligence (AI). She even graciously sent me some helpful resources on the topic for my edification and use, and reminded me that her work with Digitail, a practice information management software (PIMS) provider leading the way on AI integration in the veterinary industry, and folks like me, are not the norm. Many people haven’t tried AI, though they are often wary of its use and impact on our work. The software I employ in my practice is Digitail, but you don’t need that specific PIMS to use AI.
In the United States, we are often understaffed and overworked, and even the strongest among us have felt the effects of a relentless workload combined with a lack of support. Although burnout is certainly among the most prominent challenges we encounter, it’s also the one that I believe we will see the most improvement in by implementing AI into our workflows and processes.
Burnout
We know the mental or physical collapse caused by overwork or stress as burnout. It’s hardly unique to veterinary medicine, but it’s of particular concern in an industry with an ongoing workforce shortage2 and a profession-leading suicide rate.3
There are a lot of reasons for these things, and some of them are up for debate. However, my purpose here is to discuss the ways that AI can be used to improve productivity, reduce workload, and allow veterinarians to focus on the most rewarding and valuable parts of their job.
A Galaxy Vets white paper on burnout cites a 1999 framework put forth by Leiter and Maslach.4 The burnout triggers are defined as reward (financial, social, or intrinsic), values, workload, community, fairness, and control.
These are broad concepts with complex dynamics that vary from hospital to hospital. How is an AI large language model (LLM) going to solve any of these issues? I’m so glad you asked.
Workload
The easiest (and my favorite) burnout trigger to address with AI is workload. What automation has done best, dating back to waterwheels for energy and the flying shuttle for weaving, is to relieve humans of mundane tasks.
My favorite means of doing so in veterinary medicine is the use of LLMs to enhance and augment recordkeeping. It’s now possible for an ambient audio recording of the exchange between the veterinarian and the client during a physical examination to be automatically turned into a transcript of the conversation. No scribe required, just software. The transcription can then be run through an LLM to produce an appropriate SOAP (subjective, objective, assessment, and plan) note and client summary for the doctor to review, approve, and then add to the patient’s file.
Don’t think this is a big enough deal? At the University of Colorado Boulder, taking the computer out of the exam room and supporting doctors with human medical assistants led to a reduction in physician burnout from 53% to 13%.5 They just had somebody else keep the electronic health record (EHR) and they reduced the burnout rate by 75%. If you don’t think that natural language processing of LLMs will have the same effect, I think all you need to do is try it. The technology is persuasive.
How much time do doctors spend on medical records? Reports in human medicine range from about 41% of total doctor time6 to 4.5 hours per day7 to increased time on legacy EHR systems8 to the unanimous conclusion that time spent on medical records is excessive.9 Want to solve the productivity crisis or the workforce shortage? Giving veterinarians 41% of their day to do something besides recordkeeping would go a long way to do it.
At my hospital and others, we are actively employing AI-powered audio recording and transcription for medical recordkeeping. This isn’t a future possibility or some pipe dream; it’s an active functionality successfully implemented in hospitals. It’s dramatically decreased the veterinarian’s workload for recordkeeping.
AI also helps with client and professional communications. Although there are theoretical arguments for maintaining human empathy10 in health care, the empirical reality is that we are more empathetic when aided by AI11 and less empathetic than the AI alone.12 A recent study in the New England Journal of Medicine showed that we even get beat as diagnosticians.13 I’m a fan of these use cases and certainly employ ChatGPT in my professional communications, but the current impact and potential impact of dramatically reducing the time spent on medical recordkeeping could all but single-handedly solve the problem of workload as a source of burnout in veterinary medicine.
Community
My experience in veterinary medicine has often been deficient in a sense of community. Throughout much of my professional journey, I’ve encountered a notable absence of robust mentorship or guidance. The environments in which I worked were seldom conducive to support or collaboration, and interactions with colleagues were frequently less than amicable. I hold some of the responsibility for that, but in an industry where my story is far from unique, it seems that I might not bear it alone. Additionally, I experienced a recurring trend among my superiors: a predominant focus on enhancing my productivity as a veterinarian, coupled with a paradoxical expectation of bare minimum compensation. Despite my profound affection for my profession, some of my experiences have tempered my love of my industry.
In light of these experiences, it is not coincidental that I chose to establish a solo practice. My staffing approach has been deliberately atypical, favoring individuals with backgrounds outside the traditional veterinary sphere.
The advent of AI solutions presents an intoxicating opportunity. Though it may seem unconventional, leveraging these technological tools to augment and analyze communication offers the potential to cultivate a stronger sense of community within veterinary practices, as well as across the industry at large. Such impartial, third-party intervention can serve as a catalyst for positive development in our professional environments. We’ve already reduced the workflow, and now we can enhance the connection between colleagues and coworkers. They’re freed up and less stressed, which should make community building and connection easier. So how do we use AI to do it?
AI can allow employees to focus on creative and collaborative work, and it can ease the burdens of communication. It’s great at reworking tone. It can also facilitate learning and knowledge sharing, generate organizational assistance, create team-building activities, and even augment brainstorming.
I think the easiest way to see how AI helps community is to let it make your written messages nicer. Maybe it doesn’t apply to everyone, but I grew up in a suburb of Philadelphia known for its “extraordinarily direct” style of communication. Running messages through ChatGPT to evaluate tone has helped me soften the edges a little bit. It’s more anecdote than use case, but it’s helped me communicate in a gentler, more effective way.
Control
I love having the independence and autonomy of a practice owner. I don’t answer to boards or investors or anyone else about how to run my practice. If I want to block off a couple of hours and write an article, I can do just that.
How do we use AI to do that? ChatGPT can allow veterinarians and technicians to perform more productively and communicate at a higher level, which inherently lends itself to permitting more autonomy in workers at every level. Ethan Mollick, PhD, MBA, associate professor of management at Wharton School, University of Pennsylvania, pointed out in a blog post that you can now add intelligence of any sort to an organization by including AI.14 And his research backs that up.15
With the help of AI, you can have greater trust in employees at every level with less need for micromanagement and more autonomy. If people have more autonomy, they have more control. If they have more control, they have less burnout.16 It’s that simple.
Reward
Fairness and values are less concrete concepts, and I’m not sure there’s a defined use case for AI. I can invent some, but I’d be reaching. But reward? There are 2 ways that AI can augment reward in veterinary medicine.
The first way is simple: It can allow you to do more things, be more productive, and make more money. Your income reward increases in a traditional, causative, linear fashion. Do more, make more.
The second is a little less substantive but no less important. It can give you time. It can help you get records done so you can make it to your child’s soccer game, your book club, or your Pilates class. It can reduce the emotional lift required for communications, leaving more of you left for family, friends, or your spouse. By reducing the drudgery of the dreadful mundane, AI can give you more presence and focus to pay attention to the things to which you want to pay attention. It can do the easy stuff and give you the opportunity to be more of your best self. Time and presence are enticing rewards of the sort that not even money can buy.
Conclusion
How do we learn to use AI in these ways? Try it. Test it. Experiment with it. Learn its strengths and limitations.
The software is pluripotent and pliable. It’s rapidly evolving and adding new features all the time. There are few things in your day-to-day routine that it can’t enhance, augment, or improve. If your records are done in a fraction of the time, if your communications are done faster and better, if you have more control over your life, and you can have more of what you want, be it time, money, or attention, aren’t you better off for it?
The integration of AI into veterinary medicine is not just a step toward individual productivity and hospital efficiency; it is a leap toward redefining the industry’s future by dramatically relieving some of the burnout triggers. By embracing AI, veterinary practices can transform their work environments into spaces where burnout is not an inevitability but an exception. This technology paves the way for a new era in veterinary medicine, one in which veterinary professionals are empowered to provide the highest quality of care without sacrificing their mental health or work-life balance.
As we continue to explore and refine the use of AI, we are not only addressing the immediate challenges of workload and communication but also setting a precedent for innovative, humane, and sustainable veterinary practice. The journey with AI is just beginning, though it is decidedly underway, and its potential to fundamentally change our profession is boundless. This is a seismic shift. It is not merely a fundamental change of an organization but a change on a tectonic scale, where the ground we knew has shifted beneath our feet. By embracing this change, we are committing to a future where veterinary medicine is as rewarding for those who practice it as it is beneficial for the animals for whom we care.
William Tancredi, DVM, is a veterinarian and the owner of Old Ridge Veterinary Hospital in Chadds Ford, Pennsylvania. He is an experienced small animal clinician and entrepreneur with an interest in generative artificial intelligence (AI). He has lectured on AI and its use in veterinary medicine to the Independent Veterinary Practitioners Association, the Veterinary Hospital Managers Association, and private veterinary organizations; he has also published a number of articles in various publications, including dvm360.
References
1. Kochhar R. Which U.S. workers are more exposed to AI on their jobs? Pew Research Center. July 26, 2023. Accessed December 13, 2023.
https://www.pewresearch.org/social-trends/2023/07/26/which-u-s-workers-are-more-exposed-to-ai-on-their-jobs/
2. Straight talk about veterinary workforce issues. American Veterinary Medical Association. July 14, 2023. Accessed December 13, 2023.
https://www.avma.org/news/straight-talk-about-veterinary-workforce-issues
3. Morgan K. The acute suicide crisis among veterinarians: ‘You’re always going to be failing somebody.’ BBC. October 10, 2023. Accessed December 13, 2023.
https://www.bbc.com/worklife/article/20231010-the-acute-suicide-crisis-among-veterinarians-youre-always-going-to-be-failing-somebody
4. Leiter MP, Maslach C. Six areas of worklife: a model of the organizational context of burnout. J Health Hum Serv Adm. 1999;21(4):472-489.
5. Wright AA, Katz IT. Beyond burnout—redesigning care to restore meaning and sanity for physicians. N Eng J Med. 2018;378(4):309-311. doi:10.1056/NEJMp1716845
6. Finnegan J. For each patient visit, physicians spend about 16 minutes on EHRs, study finds. Fierce Healthcare. January 14, 2020. Accessed December 13, 2023.
https://www.fiercehealthcare.com/practices/for-each-patient-visit-physicians-spend-about-16-minutes-ehrs-study-finds
7. Payerchin R. Physicians spend 4.5 hours a day on electronic health records. Medical Economics. April 21, 2022. Accessed December 13, 2023.
https://www.medicaleconomics.com/view/physicians-spend-4-5-hours-a-day-on-electronic-health-records
8. Joukes E, Abu-Hanna A, Cornet R, de Keizer NF. Time spent on dedicated patient care and documentation tasks before and after the introduction of a structured and standardized electronic health record. Appl Clin Inform. 2018;9(1):46-53. doi:10.1055/s-0037-1615747
9. Lorkowski J, Grzegorowska O. Assessment of time spent by doctors on medical records: initial analysis and preliminary study. Education of Economists and Managers. 2020;58(4):47-64. doi:10.33119/EEIM.2020.58.4
10. Montemayor C, Halpern J, Fairweather A. In principle obstacles for empathic AI: why we can’t replace human empathy in healthcare. AI Soc. 2022;37(4):1353-1359. doi:10.1007/s00146-021-01230-z
11. Sharma A, Lin IW, Miner AS, Atkins DC, Althoff T. Human-AI collaboration enables more empathic conversations in text-based peer-to-peer mental health support. Nat Mach Intell. 2023;5:46-57. doi:10.1038/s42256-022-00593-2
12. Ayers JW, Poliak A, Dredze M, et al. Comparing physician and artificial intelligence chatbot responses to patient questions posted to a public social media forum. JAMA Intern Med. 2023;183(6):589-596. doi:10.1001/jamainternmed.2023.1838
13. McDuff D, Schaekermann M, Tu T, et al. Towards accurate differential diagnosis with large language models. arXiv. Preprint posted online November 30, 2023. doi:10.48550/arXiv.2312.00164
14. Mollick E. Reshaping the tree: rebuilding organizations for AI. One Useful Thing. November 27, 2023. Accessed December 13, 2023.
https://www.oneusefulthing.org/p/reshaping-the-tree-rebuilding-organizations
15. Dell’Acqua F, McFowland E, Mollick ER, et al. Navigating the jagged technological frontier: field experimental evidence of the effects of AI on knowledge worker productivity and quality. Harvard Business School Technology & Operations Mgt. Unit Working Paper No. 24-013. September 27, 2023. doi:10.2139/ssrn.4573321
16. Peart N. Making work less stressful and more engaging for your employees. Harvard Business Review. November 5, 2019. Accessed December 13, 2023.
https://hbr.org/2019/11/making-work-less-stressful-and-more-engaging-for-your-employees
