What happened in health care technology this week, and why it’s important.
Study finds AI can predict cybersickness in VR users
Artificial Intelligence (AI) could be used to help combat nausea, headaches, and dizziness among virtual reality (VR) users and make the technology more accessible, according to a new study with Charles Darwin University (CDU). The study by CDU and Torrens University researchers explored using AI techniques and machine learning algorithms to predict cybersickness for potential users of VR technology with headsets.
Why it’s important – In the era of technology and especially after the COVID-19 pandemic, the need for remote access and virtual education has grown significantly. Cybersickness is one of the factors that hinder users from using VR devices because of the significant discomfort due to the VR experience.
Infographic of the week – We can expect a Great Reconsideration, or churn, among healthcare providers who fast-adopted and pivoted to telehealth and remote patient monitoring, based on new research of health execs from Panda Health and Sage Growth Partners. The team explored 24 digital health applications, with 11 shown here in The Churn Index table from the report. Note that telemedicine and e-visits, and remote monitoring, are expected to exhibit high “churn” as healthcare providers re-consider their early investments into these two areas and may seek to replace, strip out, or otherwise change vendor partners for these tools. Other areas, like patients’ ability to self-schedule and care coordination, are expected to essentially stay in place.
Doctors Are Using Chatbots in an Unexpected Way
Gina Kolata reports that despite the drawbacks of turning to artificial intelligence in medicine, some physicians find that ChatGPT improves their ability to communicate empathetically with patients in her New York Times article. In one survey, 85 percent of patients reported that a doctor’s compassion was more important than waiting time or cost. In another survey, nearly three-quarters of respondents said they had gone to doctors who were not compassionate. And a study of doctors’ conversations with the families of dying patients found that many were not empathetic.
Why it’s important – Skeptics like Dr. Dev Dash, who is part of the data science team at Stanford Health Care, are so far underwhelmed about the prospect of large language models like ChatGPT helping doctors. In tests performed by Dr. Dash and his colleagues, they received replies that occasionally were wrong but, he said, more often were not helpful or were inconsistent. If a doctor is using a chatbot to help communicate with a patient, errors could make a difficult situation worse. Still, those who have tried ChatGPT say the only way for doctors to decide how comfortable they would feel about handing over tasks — such as cultivating an empathetic approach or chart reading — is to ask it some questions themselves.
How Penn Medicine’s Telemedicine Program Saved It $113 Per Patient Visit
MedCity News reporter Katie Adams reports that Penn Medicine recently published a study in the American Journal of Managed Care on the economics of providing telemedicine — it showed that when the health system began offering virtual urgent care services to its employees, the visits ended up being 23% less expensive to conduct than in-person appointments. The study analyzed how much it costs to provide visits via Penn Medicine OnDemand — a 24/7 copayment-free telemedicine program that Penn Medicine established for its employees in the summer of 2017.
Why it’s important – Despite the growing availability of telemedicine programs provided by employers, many health systems have been hesitant to embrace this approach. One reason for this is that past research has shown that the cost of care is not reduced when employers offer telemedicine via third-party platforms, the study said. The study authors argued that health systems are uniquely positioned to establish homegrown, money-saving telemedicine programs — they can use the providers they already have, make in-system referrals, and coordinate better follow-up care.
YouTube videos of the week – Last week’s headlines were all about the entry of Apple into the consumer AR space with their VisionPro. But what about health? I shared some thoughts in my blog post about WWDC 2023. But this video from Dr. Atul Gupta, Chief Medical Officer of Philips Image Guided Therapy, is a terrific look at the potential for the technology in healthcare.
And here’s another great video summary from Dr. Rafael Grossmann from the NextMed Conference.
NYU Langone Health LLM can predict hospital readmissions
Researchers at New York University’s Langone Health academic medical center developed a large language model, now deployed at three of its hospitals, that predicts a patient’s risk of 30-day readmission and other clinical outcomes. Andrea Fox from Healthcare IT News reports that coinciding with its study published in Nature this week, the release of the NYUTron model’s code base in GitHub enables other healthcare organizations to train their own LLM and provide doctors with insights that could help them identify which patients may require intervention to reduce readmissions. The model has been used to evaluate 50,000 patients discharged from NYU’s healthcare system. NYUTron shares its predictions of readmission risk with physicians by email. NYU partnered with NVIDIA to develop and run its LLM on several of its artificial intelligence platforms – its stack, library, and software.
Why it’s important – Research published in the Journal of Multidisciplinary Healthcare last year said that nearly 15% of all hospital patients are readmitted to the hospital within 30 days after initial discharge. Readmission rates are affected by countless variables – including five common treatments at emergency departments – that not only impact a patient’s overall care but they can also divert beds and resources from patients that may have more intensive healthcare needs. “Not all hospitals have the resources to train a large language model from scratch in-house, but they can adopt a pre-trained model like NYUTron and then fine-tune it with a small sample of local data using GPUs in the cloud,” said Dr. Eric Oermann, assistant professor of neurosurgery, radiology, and data science at NYU Langone Health, said in a blog post on NVIDIA’s website.
Recommended video – There have been multiple articles on the use of large language models in healthcare recently. One of the annual conferences I follow is the World Medical Innovation Forum, held in Boston. They just posted a video of the session on Generative AI, ChatGPT, and the Rapidly Changing Role of AI in Healthcare. They discussed the opportunities for these cutting-edge technologies in the short- and long-term while keeping a close eye on potential biases in their development and their impact on equity in care. They also highlighted the role of AMCs (Academic Medical Centers) in preparing for LLM integration, the ins and outs of insurance approvals and appeals, and the regulatory landscape for LLMs in clinical practice, research, and operations. Plus, they reviewed the unique assets for building and validating LLM datasets and approaches to industry collaboration and brainstormed ways to ensure safe and liability-free LLM implementation.
Uber Health Launches Grocery, Over-the-counter Delivery Service for Patients
MedCity News Maria Plescia reports that Uber Health is jumping on the food-as-medicine train. The company is adding grocery and over-the-counter delivery services to its platform, it announced Tuesday. San Francisco-based Uber Health is the healthcare arm of the transportation company Uber. Its solution offers non-emergency medical transportation and prescription delivery. Providers, payers, and case managers can use the same web-based platform they use for these services for grocery and over-the-counter delivery. Other companies that offer healthy food delivery services include FarmboxRx, Instacart, and Kroger. Several payers have also launched food-as-medicine programs, including Kaiser Permanente and Highmark Health.
Why it’s important – There is a growing understanding that supporting healthy food and nutrition can reduce costly healthcare incidents down the road. A recent study found that if payers provided medically-tailored meals to patients who need them, 1.6 million hospitalizations could be avoided annually, and payers could save $13.6 billion. Uber Health will be paid for these services through insurers’ benefit plans. Most (especially Medicaid and Medicare plans) cover transportation and delivery for prescriptions, groceries, and over-the-counter items. Still, some providers, particularly value-based providers, will pay for the benefits out of pocket. Patients won’t have to pay for the service.
Most People With Heart Disease Don’t Use Health Trackers
Christina Frank for Yale University writes in Futurity online that fewer than 1 in 4 people with or at risk for heart disease use wearable health trackers. Only half of those who wear them do so consistently. The population-based, nationally representative study, published in JAMA Network Open, included 9,303 participants from the Health Information National Trends Survey (HINTS). Eighteen percent of people with established CVD and 26% at risk for CVD reported using wearable devices compared with 29% of the general population.
Why it’s important – Older age, lower educational attainment, and lower household income were associated with significantly lower odds of wearable device use, likely due to these populations having less access to these devices or discomfort with using the technology. The people who benefit the most tend to use them the least. Part of it is awareness. People may not recognize the potential value of these devices to the health side, especially since the evidence for their value to health is still evolving. And the other piece is their cost and the fact that people who have health disorders might actually have additional costs to take care of, so these things may seem frivolous to them.
Singapore smart vending machines offer teleconsultations for pharma-only meds
Adan Ang reports that Pharmacy-only medicines can now be accessed via new smart vending machines at Changi General Hospital in his article in Healthcare IT News. Developed by pharmacists, the Pharmacy Smart Interface System (PharmaSis) was authorized by the Health Sciences Authority in April to dispense pharmacy-only medicines with guidance from a CGH pharmacist via teleconsultation. A user first seeks a teleconsultation with a CGH pharmacist, who will assess them before recommending appropriate health advice and medications. They will be guided on selecting medicines, which include those for common minor ailments such as cough, colds, diarrhea, constipation, and fever. PharmaSis also dispenses a range of OTC medicines, as well as hearing aid batteries and other audiology products.
Why it’s important – There is increased interest in developing these types of solutions globally. Two years ago, a similar vending concept was introduced by Fullerton Health, powered by US-based SmartRx’s proprietary drug dispensing hardware and software, to increase access to healthcare for thousands of migrant workers in Singapore amid the pandemic-induced lockdown. In Malaysia, health tech startup Qmed Asia also supports employees’ health by developing and launching the Qmed GO workplace telehealth kiosk. It features medical IoT devices that can measure up to 16 vital parameters, which are then transmitted to and accessed via a patient app and employer dashboard. Docty in India has also developed its own telehealth kiosks, which are being set up across rural and urban poor communities in the country to improve healthcare access.
Here’s what concerns me about the photo of the implementation above. Based on the image, how is patient privacy ensured? If someone stands behind the user, they can hear the whole conversation between the patient and the pharmacist.
A cyberattack is partly to blame for St. Margaret’s Health closing all operations
Two rural Catholic hospitals in Illinois are closing all operations as of Friday, June 16, due in part to a 2021 cyberattack. Healthcare Finance Executive Editor Susan Morse reports that St. Margaret’s Health in Peru and Spring Valley will be closing all current operations as of Friday, June 16, according to a notice on the St. Margaret’s Health website. A ransomware attack in 2021 hit SMP Health, which includes the hospitals of St. Margaret’s Health. The attack halted the hospital’s ability to submit claims to insurers, Medicare or Medicaid for months, sending it into a financial spiral.
Why it’s important – Last year, a report from the American Hospital Association highlighted the variety of causes that resulted in 136 rural hospital closures from 2010 to 2021 and a record 19 closures in 2020 alone. Cyberattacks both threaten patient safety and are costly to hospitals. The most expensive healthcare cyberattack cost an average of $4.4 million in the past 12 months, according to a Ponemon Institute study, including $1.1 million of productivity loss.
Cala Health Launches Wearable Device to Treat Hand Tremors
Katie Adams from MedCity News reports that Cala Health recently launched its kIQ system — which it calls the first noninvasive wearable device cleared by the FDA to temporarily relieve tremors in people with essential tremor and Parkinson’s disease. The kIQ system is an on-demand, at-home therapy used to manage action hand tremors — those which occur when the patient’s hand is undergoing a voluntary movement, such as typing, picking up a glass of water, or writing a note.
The wrist-worn device senses each patient’s unique tremor signature and individualizes stimulation. It then delivers transcutaneous afferent patterned stimulation (TAPS) therapy, which means it applies specific electrical stimulation patterns through the skin to modulate neural activity and promote therapeutic effects. Using TAPS, the device noninvasively counteracts the patient’s tremor through the nerves on the wrist. After the patient completes an at-home therapy session, they can view their body’s responses via the MyCala.com patient portal.
Why it’s important – More than 7 million Americans struggle with essential tremor (ET), but only one in 10 ET patients is satisfied with their current treatment. The klQ system is prescribed by movement disorder specialists and neurologists for patients with ET or Parkinson’s. Researchers say more studies are needed to determine whether wearable devices can create long-lasting tremor relief from a single session — beyond reducing tremors for an hour or two at a time. They also suggest that further research is needed to solidify wearables as a standalone treatment for essential tremor patients.