Multiple news sources have recently pointed out massive layoffs in the technology industry due to artificial intelligence (AI) improvements that lead to job redundancy. During an interview with Jimmy Fallon, Bill Gates expressed a vision where in the future AI will replace health care providers and other service professionals. A sentiment that is shared throughout the tech industry.
In health care, the first wave of AI-related workforce adjustment has occurred in two systems: Kaiser Permanente and Revere Health. However, this is also reflected in the business side of medicine by layoffs at Oracle EHR and health insurance companies. Most of the restructuring has occurred in the following roles: medical coding and billing, patient scheduling and documentation, claims denial prevention, and low-level tech-support roles. No clinical jobs have been replaced or will be permanently replaced in the near future by artificial intelligence.
My prediction is that the only jobs AI will replace are the jobs that were non-essential to the patient-physician relationship, as already evidenced in this first wave of layoffs. These roles have been artificially inserted into the health care field due to changes in the business landscape but none of them are essential to the service being rendered to the patient. Something that is confirmed indirectly by the rise in popularity of direct care practices.
Notwithstanding the above, rest assured that clinical providers will be fired during this AI hype cycle, later to be re-hired to reconstruct the workflows that AI helped “transform.” It has been documented that AI proof of concepts rarely make it to production citing factors related to insufficient organizational readiness and IT infrastructure. This problem will only become exponential when data is meant to be extracted from multiple interfaces that don’t talk to each other. A few hype cycles ago, cloud computing and the Internet of Things made promises that systems would be able to talk to each other and providers would have access to unified patient data, however today health care remains as fragmented as ever.
Real physician voices, twice a week
Free, and one click to unsubscribe.
Following restructuring efforts a phenomenon called “boomerang hiring” has gained traction. Boomerang hiring is the practice in which former employees return to the same employer. An event that previously came with disgrace, it is now being shrugged off as par for the course for business for many employers.
Where many entrepreneurs and top brass are misguided is by not acknowledging the complexity of our work as clinicians. Health care is messy and you can’t skip steps. Biologic systems are not as predictable as synthetic structures (i.e., logistics, IT, or production lines). Modern medicine with its evolving foundation in evidence-based medicine is based upon statistical inferences imposed upon an ideal “average human.” Try to extrapolate this statistical average into statistical predictive inference by an LLM and you are building a supposition on top of a supposition, a proverbial house of cards. Echoing the above, the clinical validity of the data in which these models are used has been questioned in multiple research studies.
AI transformation of health care is here, but it is decades away from having a meaningful positive impact. For the time being what we can expect is increased work expectations (to justify the institutional investment in these unproven tools), a round of provider layoffs (to push those remaining into being more efficient) and restructuring efforts.
An open question surfaces: How will we providers safeguard the practice of medicine from AI hubris?
Felix Reyes is a pulmonary and critical care physician.
