Five spine surgeons discuss the next big healthcare technologies to make waves in the spine field.
Ask Spine Surgeons is a weekly series of questions posed to spine surgeons around the country about clinical, business and policy issues affecting spine care. We invite all spine surgeon and specialist responses.
Next week's question: What is the biggest obstacle facing spine providers in the modern landscape?
Please send responses to Alan Condon at acondon@beckershealthcare.com by 5 p.m. CDT Wednesday, Sept. 23.
Note: The following responses were lightly edited for style and clarity.
Question: What is the next big health IT innovation you expect to see?
Grant Shifflett, MD. DISC Sports & Spine Center (Newport Beach, Calif.): Artificial intelligence and augmented/virtual reality offer great promise but still need refinement for widespread adoption. Advances to make these tools more user-friendly and cost-effective will spur a new era of surgical advances and improvements in quality of care.
Vladimir Sinkov, MD. Sinkov Spine Center (Las Vegas): What I hope to see is greater and more streamlined interoperability of various health IT components such as physician office EMRs, hospital EMRs, state prescription drug monitoring program websites, e-prescribing software and web-based picture archiving and communication systems. For surgical patients, I need to log in to all of those systems, with separate logins and passwords (that change every 3 months) and remember how each system works so I can document the office visit, write postoperative orders, write hospital progress, prescribe postoperative pain medications and view X-ray and MRI images. The amount of time I spend navigating these systems that do not "talk" to each other could be better spent providing patient care.
Colin Haines, MD. Virginia Spine Institute (Reston): While it isn't new, I expect telemedicine to continue to grow by leaps and bounds. But I do think there is room for innovation to include blood pressure checks, blood draws and other tests performed at home. During the height of the pandemic, my practice rapidly implemented and began using telemedicine for the first time. Certainly, there are limitations due to lack of a physical exam, but I have been pleased with how effective our medical care has been with these virtual visits. Furthermore, many patients are electing to pursue telemedicine visits independent of COVID. I understand the plight of the busy individual; my biggest limitation to seeing my primary care physician or dentist is simply taking the requisite time off from work. As stretched thin as most of us are, if we can do more of our healthcare from the convenience of our computers, we may get better overall care due to more physician checkups.
Brian Gantwerker, MD. Craniospinal Center of Los Angeles: The next big innovation in IT will be the sharing of patient images and virtual visit technology. We will be seeing a time when there will be no state lines when it comes to patient care. Patients will soon be able to visit physicians anywhere in the country. I do see trouble though, as I think there will be a lack of the ability of future physicians to perform a decent physical exam.
Michael Gordon, MD. Hoag Orthopedic Institute (Irvine, Calif.): Machine learning has gained tremendous popularity in recent years, with some techniques such as search engines, voice recognition software and autonomous driving vehicles now part of our daily lives. Artificial intelligence shows great promise in promoting practice efficacy, personalizing patient management and improving research capacity. The most popular area of machine learning research is image interpretation, including the diagnosis of fractures, osteoarthritis, bone age and bone strength. Rather than replacing the radiologist, machine learning helps improve diagnostic accuracy and prevent errors and observer fatigue. Another major potential use is in predicting the clinical outcome of patients based on a clinical dataset, genomic information and medical images.
Risk assessment and outcome prediction have always been challenging in clinical medicine and are laborious, requiring review of multiple variables. Machine learning in the next decade will include helping with risk analysis. The dream is a machine pooling personal medical info and directing relevant information to a physician, decreasing clutter and reducing workload. The combination of a physician skilled in clinical evaluation and a machine assisting in amassing relevant information is key.
Problems with machines mimicking humans are legion. Everyone knows that speech to text is riddled with auto correct errors but only a human knows that I'm referring to multiple holes and not multiple jokes. Your self-driving car can calculate the shortest route but not necessarily the fastest or cheapest one, and doesn't know why you are going there. At present, a robot does not know how it has been incorrect: surgical robots may place screws or bone cuts in exactly the position calculated, but will not understand that they are misplaced, nor be able to self-correct at a later time.
Smart machines excel at correcting for human frailties — fatigue, inattention, distraction due to multiple inputs, bias and emotion. Smart humans excel at correcting for machine frailties — poor pattern-recognition capability without exhaustive databases, propagation of unintended errors, lack of self-correction in nonlinear human-based actions and a lack of "human intuition learning." Machine learning is limited by the high capital cost, time needed for its use (in preparation and intraoperatively), the variable reliability of machine learning technologies and the absence of long-term follow-up studies. Another unsolved issue arises in cases of misdiagnosis or complication: who is responsible, machine or human?
At present do not hope for a machine performing the entire surgery. Expect machines to help with streamlining workflow, improving efficiency and being an unblinking eye in the operating room. The machine can sew and drill at perfect spacing, but a human must tell it where to start.