Monday, March 19, 2018

Big Data and Machine Learning in Health Care | Clinical Decision Support | JAMA | JAMA Network

Nearly all aspects of modern life are in some way being changed by big data and machine learning. Netflix knows what movies people like to watch and Google knows what people want to know based on their search histories. Indeed, Google has recently begun to replace much of its existing non–machine learning technology with machine learning algorithms, and there is great optimism that these techniques can provide similar improvements across many sectors.
It is perhaps more useful to imagine an algorithm as existing along a continuum between fully human-guided vs fully machine-guided data analysis. To understand the degree to which a predictive or diagnostic algorithm can said to be an instance of machine learning requires understanding how much of its structure or parameters were predetermined by humans. The trade-off between the human specification of a predictive algorithm’s properties vs learning those properties from data is what is known as the machine learning spectrum. Returning to the Framingham study, to create the original risk score statisticians and clinical experts worked together to make many important decisions, such as which variables to include in the model, the relationship between the dependent and independent variables, and variable transformations and interactions. Since considerable human effort was used to define these properties, it would place low on the machine learning spectrum (#19 in the Figure and Supplement). Many evidence-based clinical practices are based on a statistical model of this sort, and so many clinical decisions in fact exist on the machine learning spectrum (middle left of Figure). On the extreme low end of the machine, learning spectrum would be heuristics and rules of thumb that do not directly involve the use of any rules or models explicitly derived from data (bottom left of Figure).
It is no surprise then that medicine is awash with claims of revolution from the application of machine learning to big healthcare data. Recent examples have demonstrated that big data and machine learning can create algorithms that perform on par with human physicians.1 Though machine learning and big data may seem mysterious at first, they are in fact deeply related to traditional statistical models that are recognizable to most clinicians. It is our hope that elucidating these connections will demystify these techniques and provide a set of reasonable expectations for the role of machine learning and big data in healthcare.

Machine learning was originally described as a program that learns to perform a task or make a decision automatically from data, rather than having the behavior explicitly programmed. However, this definition is very broad and could cover nearly any form of data-driven approach. For instance, consider the Framingham cardiovascular risk score, which assigns points to various factors and produces a number that predicts 10-year cardiovascular risk. Should this be considered an example of machine learning? The answer might obviously seem to be no. Closer inspection of the Framingham risk score reveals that the answer might not be as obvious as it first seems. The score was originally created2 by fitting a proportional hazards model to data from more than 5300 patients, and so the “rule” was in fact learned entirely from data. Designating a risk score as a machine learning algorithm might seem a strange notion, but this example reveals the uncertain nature of the original definition of machine learning.

There is no doubt that  'machine learning', artificial intelligence will gradually intrude upon our routines almost unnoticed, just as chatbots already have done so.

Big Data and Machine Learning in Health Care | Clinical Decision Support | JAMA | JAMA Network

Sunday, March 11, 2018

What Happens When Doctors Only Take Cash |

Coming very soon to your city..Direct pay...

When Art Villa found out, after one too many boating accidents, that he needed a total knee replacement, he began asking around to see how much it would cost. The hospital near his home in Helena, Mont., would charge $40,000 for the procedure, he says. But that didn't include the anesthesiologist's fee, physical therapy or a stay at a rehabilitation center afterward. A 2015 Blue Cross Blue Shield study found that one hospital in Dallas billed $16,772 for a knee replacement while another in the same area charged $61,585.
It was in the midst of this confounding research that Villa, who's 68, heard about the Surgery Center of Oklahoma, whose business model is different from that of most hospitals. There, the all-inclusive price for every operation is listed on the website. A rotator-cuff repair for the shoulder costs $8,260. A surgical procedure for carpal tunnel syndrome is $2,750. Setting and casting a basic broken leg: $1,925.

What Happens When Doctors Only Take Cash |

Friday, March 9, 2018

Uber launches healthcare platform | Healthcare Dive

"Every year an estimated 3.6 million Americans miss their appointments due to a lack of reliable transportation1. At Uber, we recognize that the path to health may not be easy, but we know the road to care can be."

“Within days of using Uber Health to arrange trips for our patients, we stopped using traditional transportation services. Coordination is seamless for us and rides are easy for patients.”
  • The ride-sharing company Uber is officially launching a new platform for providers to use the app to get patients to their facilities.
  • The dashboard, Uber Health, allows a healthcare worker to book a ride on demand or schedule a future ride for a patient. The passenger is alerted by text or phone call with trip details.
  • The product has been tested by about 100 hospitals and doctor's offices so far, according to The Atlantic.
  • The news may not come as a shock, but it does draw a line in the sand for competitors.
    Lindsay Elin, director of federal and community affairs at Uber, told Connected Health Conference in 2016 that the company was investing in a team to work solely on healthcare
    "I'd say there are two major things driving Uber's interest," Naveen Rao, health innovation analyst and founder of Patchwise Labs told Healthcare Dive.
    "First, they see how much Lyft has rolled up their sleeves in healthcare, so the official trigger was a simple reaction to their main competitor. Second, is they've been involved as an enabling partner in the non-emergency medical transport space for a couple of years now and seen the demand." 
    Companies like Veyo are targeting Medicaid and Medicare patients and Circulation is offering a platform connecting ride-share services like Uber and Lyft for NEMT.Ralph Decaro, Manhattan Centers for Women's Health
One study found when Uber enters a city, ambulance use decreases by at least 7%. However, using Uber as an ambulance can put a driver in an awkward position for suddenly taking on a job they weren't trained for.
Earlier this year, Uber and MedStar Health announced a partnership aimed to help alleviate patient no-shows for the Columbia, MD-based health system while presenting it as a forward-thinking, digital-savvy operation. With about a year's worth of experience under the partnership's belt, representatives from both Uber and MedStar shared their insights into the partnership and where such collaborations are going at the Connected Health Conference on Monday in the Washington, DC area.

Getting there for care

"Half our battle is getting the patient to the appointment," Pete Celano, director of consumer health initiatives at MedStar Institute for Innovation, said at a panel discussion on cities that promote health. Missed appointments are a costly problem for health plans and physician offices, with some estimates as high as $150 billion a year in the U.S., according to a Washington Business Journal report from earlier this year, a stat Lindsay Elin, director of federal and community affairs at Uber, also shared during the panel. An estimated 3.6 million people in the U.S. miss or delay medical care each year because of issues related to transportation.
For years, MedStar used taxis in effort to reduce no-shows. However, according to Celano, taxis came with three issues: They were expensive, clunky from an operations perspective and drivers historically wouldn't go the extra mile for patients.
MedStar was the first healthcare provider to partner with Uber but Elin said there has been increased interest from other healthcare providers. In fact, Uber is investing in the creation of a small team at the company that will do nothing but work with healthcare providers, Elin stated. Currently, Uber is doing over 5 million rides a day, she shared, adding 75% of the U.S. population lives in a county that has access to Uber. "We firmly believe with partnerships with healthcare providers, senior centers [and] transit agencies that we can do even more and reach more people," Elin said.
How it's working
While the partnership started in January beginning with UberX services, Uber created a a dashboard tool called UberCENTRAL with MedStar. The product allows healthcare providers to request and manage a ride on behalf of their patient who doesn't need to own a smartphone, an important fact to consider when dealing with low-income patients. MedStar moved to the UberCENTRAL platform two months ago.
Celano stated the system is already seeing returns on the partnership. With UberX, for example, the cost to the system is about 60% of the cost of a cab in DC, Northern Virginia or Maryland, according to Celano. In addition, using an UberX means having the ability to hail a car inside an hour which reduces the chance of an appointment going unfilled.
"For patients who can afford it, we say 'Please Uber if and as you want to,'" Celano said. "It could be less expensive to go to Georgetown University Hospital for example on an Uber from most places than to park there, if we even have parking spots available."
For patients with a medical and financial need (about 30% of the patients in the U.S. health system), MedStar can cover the Uber transportation fee. About half of indigent patients are able to get a ride to the facility, Celano shared. "People ask me how it's going and I say it's all about the power of going door-to-door," he said. The average roundtrip cost is $18 which can move an appointment that does not occur to one that does occur, whereby a given healthcare system can recognize new revenue or cost savings. These face-to-face visits can help result in better health outcomes for patients, Celano indicated.
We can expect more innovation cross-over from consumer oriented technology into the health space.
Uber launches healthcare platform | Healthcare Dive

Thursday, March 8, 2018

Measuring What Matters and Capturing the Patient Voice.

Incorporating the patient voice into health care delivery has proved challenging—particularly when there is little agreement about how to define and discuss the concept nor about how to measure its impact. NEJM Catalyst convened an in-person roundtable to address these issues in depth. The roundtable and an accompanying survey of our Insights Council point the way toward a framework for defining the patient voice and integrating it into care delivery.  Some have called this Patient-Centered Medicine.

At the same time  it assumes the physician ear is tuned to the same frequency. In the background, and at times many other voices are present. Some are wanted, most are extraneous, disruptive and irrelevant to physicians. After all physicians spend most of their time listening.

While communication is a good thing we must all master becoming better physicians, most of these interactions are distracting at the least. During out daily routine we manage to synthesize or eliminate these Voices

During a Roundtable sponsored by the NEJM Catalyst this article attempts to organize the cacaphony of voices.

Measuring What Matters and Capturing the Patient Voice

Measuring What Matters and Capturing the Patient Voice.pdf

Saturday, February 24, 2018

Anthem Calls On Eye Surgeons To Monitor Anesthesia During Cataract Surgery | California Healthline

Cost Containment overrides Safety
If you need cataract surgery, your eye surgeon may have to do double duty as your anesthetist under a new policy by health insurer Anthem. In a clinical guideline released this month, the company says it's not medically necessary to have an anesthesiologist or nurse anesthetist on hand to administer and monitor sedation in most cases.
Some ophthalmologists and anesthesiologists say the policy jeopardizes patient safety, and they are calling on Anthem to rescind it.
"The presence of anesthesia personnel is one of the key ingredients in the patient safety and effectiveness of cataract surgery today," says Dr. David Glasser, an ophthalmologist in Columbia, Md., who is secretary for federal affairs at the American Academy of Ophthalmology, a professional group for eye physicians and surgeons. "An ophthalmologist cannot administer conscious sedation and monitor the patient and do cataract surgery at the same time."
Anthem, which offers commercial insurance plans in 14 states, says anesthesia needs vary and so should coverage. According to a statement from the company:
"Anthem's Medical Policy and Technology Assessment Committee, a majority of whom are external physicians, reviewed the available evidence addressing the use of general anesthesia and monitored anesthesia care for cataract surgery. According to the literature reviewed, there is no one definitive approach regarding the use of anesthesia for cataract surgery and patient-specific needs should be taken into consideration as well as potential risk of harm to individuals who are sedated during surgical procedures."
Medicare, the health care program for people age 65 and older, covers cataract surgery, including anesthesia services.
A cataract, typically related to aging, is caused by clumps of protein that cloud the lens of the eye and can distort vision. During a cataract operation, the surgeon makes an incision in the surface of the eye with a laser or blade and then uses a tool to break up the clouded lens, pull it out and replace it with an artificial one.
Cataract surgery is common. More than half of Americans have either had a cataract or had cataract surgery by the time they reach age 80, according to the National Eye Institute.
Surgery is typically performed on an outpatient basis and takes less than an hour. Though drowsy while sedated, patients are generally conscious during the procedure and can hear what's said to them and speak if necessary.
Eye surgeons often have an anesthesiologist or a nurse anesthetist present to administer intravenous drugs to help keep the patient relaxed and ensure they don't move during the operation as well as monitor their vital signs and adjust medication as necessary.
Anthem's new policy states that this type of monitored anesthesia care is medically necessary only if the patient is under 18 years old, or is unable to cooperate or communicate because of dementia or other medical conditions, can't lie flat, has known problems with anesthesia, or if a complex surgery is anticipated.
But some ophthalmologists and anesthesiologists disagree.
"I wouldn't even consider doing a cataract surgery without an anesthesiologist or nurse anesthetist in the room," says Dr. David Aizuss, an eye surgeon who is president-elect of the California Medical Association. "If you're working inside the eye it's a very confined space, and if the patient gets agitated and starts moving around you have to get the equipment out of the eye very quickly."
Although Anthem posted the new policy online, providers are seeking clarification from the company about the timing of its implementation in their states, physicians said.
Until then, some practices are taking no chances. At the Freedom Vision Surgery Center in Encino, Calif., where Aizuss practices, Anthem patients who come in for cataract surgery are asked to pay $400 out-of-pocket upfront for anesthesia services.
Professional groups representing California eye physicians and anesthesiologists have written to Anthem requesting the policy be rescinded. In addition, the California Medical Association has lodged complaints with state regulators.
This isn't the first time Anthem has come under scrutiny for changes to its clinical guidelines that some have charged help the company's bottom line at patients' expense. Last year, the company said it would no longer pay for emergency department visits it later determined were not emergencies. Then in September it said it would no longer pay for imaging tests like MRIs in many cases if patients got them at hospital-owned centers rather than independent imaging centers.
Last week, Modern Healthcare reported that the company says it was modifying its ER rule so that certain types of visits would always be paid for, including those by patients who are directed to the emergency department by their provider or have recently had surgery.
Some safety experts say they were concerned about Anthem's new policy, even for routine cataract surgeries.
"If you're putting a knife in my eye, that's not routine for me," says Leah Binder, president and CEO of the Leapfrog Group, a nonprofit organization that advocates for improved safety and quality at hospitals. Noting that anesthesiologists and nurse anesthetists were pioneers in the patient safety movement, she says there are better ways for Anthem to save money than shutting them out of the operating room.
"How about identifying the surgeons who have the highest complication rates, and letting patients know about them?" she suggested.

Anthem Calls On Eye Surgeons To Monitor Anesthesia During Cataract Surgery | California Healthline

Tuesday, February 20, 2018

For Natives Americans, In Indian Health Service And Outside It : Shots - Health News : NPR

The life expectancy of Native Americans in some states is 20 years shorter than the national average.

There are many reasons why.
Among them, health programs for American Indians are chronically underfunded by Congress. And, about a quarter of Native Americans reported experiencing discrimination when going to a doctor or health clinic, according to findings of a poll by NPR, the Robert Wood Johnson Foundation and Harvard T.H. Chan School of Public Health.
Margaret Moss, a member of the Hidatsa tribe, has worked as a nurse for the Indian Health Service and in other systems. She now teaches nursing at the University of Buffalo.
She says she has seen racism toward Native Americans in health care facilities where she's worked, and as a mom trying to get proper care for her son.
Once, when she was on a health policy fellowship with a U.S. Senate committee, Moss' son had a broken arm improperly set at a non-IHS health facility in Washington, D.C.
She asked the physician about options to correct it, but he told her it was fine, she said. "Even when I, as an educated person using the right words, was saying what needed to happen, [he] didn't want to do anything for us, even though we had a [health insurance] card."
Moss then reluctantly pulled out a business card with the Senate logo, she recalled, and was instantly transformed in the doctor's eyes from "this American Indian woman with my obviously minority son" to someone he could not afford to dismiss.
"It wasn't until the person ... felt they could get in trouble for this ... then the person did something," said Moss. "I felt like it was racism. Not everybody has a card they can just whip out."
She says she feels discrimination is more overt, "in areas where American Indians are known about," like the Dakotas and parts of the American Southwest, but also exists in places without big tribal populations.
In the NPR poll, Native Americans who live in areas where they are in the majority reported experiencing prejudice at rates far higher than in areas where they constituted a minority.
In places where there are few American Indians, Moss says, "people don't expect to see American Indians; they think they are from days gone by, and so you are misidentified. And that's another form of discrimination."
Health care systems outside the Indian Health Service generally see very few Native American patients, because it's so hard for American Indians to access care in the private sector. A lot of that has to do with high poverty and uninsured rates among American Indians, who also often live in rural areas with few health care providers.
"The strikes against people trying to get care are huge: geographic, transportation, monetary," Moss says.
A persistent myth inside and outside Indian Country is that Native Americans get free health care from the federal government.
A persistent myth inside and outside Indian Country is that Native Americans get free health care from the federal government.
Anna Whiting Sorrell, a health care administrator for her tribes, says she is optimistic that the Affordable Care Act will make a big difference for Native Americans. It gives lower-income people access to affordable insurance coverage outside the IHS. Many Natives Americans who weren't eligible for Medicaid before the ACA now are, too.
Moss is more skeptical that the ACA will make a big difference, in part because of entrenched institutional discrimination toward Native Americans in healthcare.
"Until attitudes change," Moss says, "we're still going to be in a sad situation."

For Natives Americans, In Indian Health Service And Outside It : Shots - Health News : NPR

Saturday, February 10, 2018

Get creative in making space for new technologies in benefits plans

Rapid changes in Health, (digital health) in the U.S. are mirrored around the globe. Canada is making a uniform change and analysis of additions to their health system and adjusting benefits accordingly, using the hard test of time prior to funding.
It’s an exciting time for group benefits in Canada. That’s not a phrase you hear every day, but it’s one that’s coming up more and more often.
Terms like “digital health” and “disruption” are new words to the Canadian benefits vernacular. All of a sudden, it seems like a new digital health startup launches every day and incubators, accelerators and labs are popping up everywhere.
Over the past decade, health plans at all levels in Canada have become much better at managing costs. Employer health plans increasingly require proof of value before covering a new medical service or pharmaceutical product. The challenge, however, arises when the item under consideration is genuinely new and innovative. As difficult as it is to assess a new prescription drug, evaluating a service unlike anything currently offered can be nearly impossible. By definition, a service that has only existed for two years can’t have demonstrated a long-term return on investment. As a result, good ideas and new technologies face implementation delays.
But what does it all mean for group insurance? What will these new services, startups, innovation labs and disruptors really do to change how plan sponsors and employees manage their health and insurance?
Where to find help?
Pharmacogenetics: Personalized medicine is one of the great developments in health care. It has begun to arrive, and pharmacogenetics represent one of the first ways in which we see personalized medicine becoming available to group insurance plans.
Telemedicine: Access to prompt medical advice is one of the most commonly cited shortcomings of the Canadian health-care system. Due to an apparent gap in the definitions under the Canada Health Act, provincial plans generally don’t cover remote or virtual access to physicians. There are now multiple providers of the service on a per-visit or annual subscription basis. For employers, the potential time savings to their employees are significant. Is covering an on-demand video call with a doctor or nurse not more cost-effective than an employee taking time off work to wait at a clinic or in an emergency room?
Pharmacists: The idea of pharmacists doing more than simply dispensing pills by providing value-added services, such as coaching for patients with chronic disease, has been a solution waiting to happen. After years of discussion, the idea has yet to take hold. Why is that? Pharmacists will offer the services if employer benefits plans provide reimbursement. Employer benefits plans, in turn, are open to covering those services once there’s a proven track record of added value.
Gene therapy: The ability to alter someone’s genes to cure a disease (or eliminate the risk of developing one) will soon be a reality. How will benefits plans handle a genetic treatment for a disease, such as cancer, that someone is at risk for but doesn’t yet suffer from?
Health coaching and navigation:Unfortunately, most Canadians don’t get enough time with public health professionals either to stay healthy or get help, when they fall sick, in navigating the system efficiently. As a result, wellness, coaching and navigation alternatives are growing fast, often using technology to personalize and automate the services and thus drive down the cost and improve employee health at the same time.
Virtual care: In-person care has historically been the default way to see a health practitioner. But what if, instead of seeing a psychologist in person for $150 per hour, plan members could treat their mental-health issue by using a cognitive behavioural therapy app on their phone? Is that not worth paying the $60 annual subscription to access it? Despite the potential, most benefits plans haven’t developed a framework to assess the value of or reimburse virtual care.
Health-care spending accounts: New entrants into the group insurance market have sparked a growing debate recently about the role health-care spending accounts could and should play in benefits plans. Given the predictability of many kinds of expenses covered (such as paramedical services, dental recall exams and vision care), the concept of a health-care spending account as the primary source of coverage, with adequate real insurance for catastrophic expenses, is a valid structure to consider.
Medical marijuana: For a variety of reasons, information on both the therapeutic value and cost effectiveness of marijuana in a benefits plan is often still anecdotal. Would a plan that covers medical marijuana see an increase or a decrease in overall costs? For whom and under what circumstances could it be a benefit that outweighs the cost?
When looking at whether to add a new plan feature, employers don’t need to offer it at 100 per cent coverage or to all employees and under all circumstances. Being creative in merging new solutions with established ones will differentiate those plan sponsors that are willing to embrace change and use technology to engage employees in their health.Get creative in making space for new technologies in benefits plans | Benefits Canada

Friday, February 9, 2018

Algorithm identifies people with diabetes in Apple Watch heart data

Researchers have developed an algorithm that can identify people previously diagnosed with diabetes in Apple Watch heart data. The learning algorithm provides a strong model for identifying people with diabetes but has yet to prove itself against the tougher task of spotting undiagnosed patients without also racking up false positives.
Cardiogram, an Apple Watch app developer, ran the study (PDF) on 14,011 of its users in conjunction with researchers at UCSF.The data is displayed in the article for several states of heart activity After using data on some of the participants to train a deep neural network, called DeepHeart, the team tested the algorithm on results from the remaining cohort of subjects. The best version of the algorithm recorded a c-statistic of 0.85 in diabetes, making it a strong model.

San Francisco-based Cardiogram is able to identify people with diabetes from heart data because of earlier work that spotted a correlation between variability in cardiovascular activity and the condition.

It was not stated whether the . Apple Watch is HIPAA compliant. HIPAA compliance is a federal regulation which requires all personal identifiying information to be scrubbed from data sources.  The article did not expand on this issue.

RELATED: Apple eyes FDA approval filing for investigational heart device

If Apple files for approval of the app, it will mark a major advance in its long-running flirtation with the healthcare sector. The tech giant stepped up its interest in the space with the introduction of Watch in 2015 and rollout of its ResearchKit framework. Talks with the FDA and involvement in its software precertification pilot program followed. But Apple has yet to seek FDA clearance of a device. 

In another development:

AliveCor gets first FDA nod for an Apple Watch accessory

AliveCor has gained clearance to sell a medical device accessory for Apple Watch. The regulatory nod covers AliveCor’s KardiaBand, a device that clips onto Apple’s smartwatch and performs EKG readings.
Users of the $199 AliveCor device and accompanying $99-a-year service replace the wristband on their Apple Watch with KardiaBand. Machine learning algorithms, dubbed SmartRhythm, then sift through data gathered by Apple Watch’s sensors to establish a normal band of heart rate activity. If the wearer’s heart rate deviates from these historic norms, the app directs the user to take an EKG.
This is where the band itself comes in. The user places a finger on a sensor built into the strap. The band then performs an EKG, also known as an electrocardiogram, to assess whether the electrical impulses that modulate cardiac contractions are firing properly. The resulting 30-second waveform is shown on the Apple Watch screen, after which the user can share it with their doctor as a PDF.

Aspects of the technology are the same as the credit card-sized device, KardiaMobile, AliveCor already sells for use with smartphones. But the incorporation of the Apple Watch sensors and data into the process stands to change when and why users decide to take an EKG reading.

The big question now is whether this more objective approach to assessing when an EKG is needed will translate into improved outcomes for patients. AliveCor has clinical trial data showing its smartphone-based EKG outperformed routine care. In theory, the Apple Watch-based approach should improve on that product, for the reasons outlined by Topol, but that hypothesis is yet to be tested in the wild.
For the broader digital health sector, the important thing is AliveCor has gained clearance to start finding out how its device performs in the real world at all. AliveCor hustled through the regulatory process far faster in Europe than in the U.S. When AliveCor unveiled the device in March 2016, it talked up the prospect of “late spring” availability. That target came and went as AliveCor grappled with the FDA review process.

Vic Gundotra, the ex-Googler who runs AliveCor, told TechCrunch getting the device to market was “one of the hardest things I’ve ever done in my life.” Under the leadership of Scott Gottlieb, M.D., the FDA may provide an easier route to market for companies that try to follow in AliveCor’s wake. But having put in the hard yards, AliveCor has the field to itself for now.

The FDA process is lengthy and costly for device manufacturers, a barrier for smaller manufacturers. Hopefully this will encourage other ground breaking consumer medical devices to proceed.