Free EMR Newsletter Want to receive the latest news on EMR, Meaningful Use, ARRA and Healthcare IT sent straight to your email? Join thousands of healthcare pros who subscribe to EMR and HIPAA for FREE!!

Has Amazon Brought Something New To Healthcare Data Analytics?

Posted on November 29, 2018 I Written By

Anne Zieger is a healthcare journalist who has written about the industry for 30 years. Her work has appeared in all of the leading healthcare industry publications, and she's served as editor in chief of several healthcare B2B sites.

Amazon’s announcement that it was getting into healthcare data analytics didn’t come as a major surprise. It was just a matter of time.

After all, the retail giant has been making noises about its health IT ambitions for a while now, and its super-sneaky 1492 team’s healthcare feints have become common knowledge.

Now, news has broken that its massive hosting division, Amazon Web Services, is offering its Comprehend Medical platform to the healthcare world. And at the risk of being a bit too flip, my reaction is “so?” I think we should all take a breath before we look at this in apocalyptic terms.

First, what does Amazon say we’re looking at here?

Like similar products targeting niches like travel booking and supply-chain management, the company reports, Comprehend Medical uses natural language processing and machine learning to pull together relevant information from unstructured text.

Amazon says Comprehend Medical can pull needed information from physician notes, patient health records and clinical trial reports, tapping into data on patient conditions and medication dosage, strength and frequency.

The e-retailer says that users can access the platform through a straightforward API call, accessing Amazon’s machine learning expertise without having to do their own development or train models of their own. Use cases it suggests include medical cohort analysis, clinical decision support and improving medical coding to tighten up revenue cycle management.

Comprehend Medical customers will be charged a fee each month based on the amount of text they process each month, either $0.01 per 100-character unit for the NERe API, which extracts entities, entity relationships, entity traits and PHI, or $0.0014 per unit if they use its PHId API, which only supports identifying PHI for data protection.

All good. All fine. Making machine learning capabilities available in a one-off hosting deal — with a vendor many providers already use — can’t be wrong.

Now, let’s look coldly at what Amazon can realistically deliver.

Make no mistake, I understand why people are excited about this announcement. As with Microsoft, Google, Apple and other top tech influencers, Amazon is potentially in the position to change the way things work in the health IT sector. It has all-star brainpower, the experience with diving into new industries and enough capital to buy a second planet for its headquarters. In other words, it could in theory change the healthcare world.

On the other hand, there’s a reason why even IBM’s Watson Health stumbled when it attempted to solve the data analytics puzzle for oncologist. Remember, we’re talking IBM here, the last bastion of corporate power. Also, bear in mind that other insanely well-capitalized, globally-recognized Silicon Valley firms are still biding their time when it comes to this stuff.

Finally, consider that many researchers think NLP is only just beginning to find its place in healthcare, and an uncertain one at that, and that machine learning models are still in their early stages, and you see where I’m headed.

Bottom line, if Google or Microsoft or Epic or Salesforce or Cerner haven’t been able to pull this off yet, I’m skeptical that Amazon has somehow pole-vaulted to the front of the line when it comes to NLP-based mining of medical text. My guess is that this product launch announcement is genuine, but was really issued more as a stake in the ground. Definitely something I would do if I worked there.

Software Marks Advances at the Connected Health Conference (Part 2 of 2)

Posted on October 31, 2018 I Written By

Andy Oram is an editor at O'Reilly Media, a highly respected book publisher and technology information provider. An employee of the company since 1992, Andy currently specializes in open source, software engineering, and health IT, but his editorial output has ranged from a legal guide covering intellectual property to a graphic novel about teenage hackers. His articles have appeared often on EMR & EHR and other blogs in the health IT space. Andy also writes often for O'Reilly's Radar site (http://oreilly.com/) and other publications on policy issues related to the Internet and on trends affecting technical innovation and its effects on society. Print publications where his work has appeared include The Economist, Communications of the ACM, Copyright World, the Journal of Information Technology & Politics, Vanguardia Dossier, and Internet Law and Business. Conferences where he has presented talks include O'Reilly's Open Source Convention, FISL (Brazil), FOSDEM, and DebConf.

The first part of this article focused on FDA precertification of apps and the state of interoperability. This part covers other interesting topics at the Connected Health conference.

Presentation at Connected Health Conference

Presentation at Connected Health Conference

Patient engagement

A wonderful view upon the value of collecting patient data was provided by Steve Van, a patient champion who has used intensive examination of vital signs and behavioral data to improve his diabetic condition. He said that the doctor understands the data and the patient knows how he feels, but without laying the data out, they tend to talk past each other. Explicit data on vital signs and behavior moves them from monologue to dialogue. George Savage, MD, co-founder and CMO of Proteus, described the value of data as “closing the loop”–in other words, providing immediate and accurate information back to the patient about the effects of his behavior.

I also gained an interesting perspective from Gregory Makoul, founder and CEO of PatientWisdom, a company that collects a different kind of data from patients over mobile devices. The goal of PatientWisdom is to focus questions and make sure they have an impact: the questionnaire asks patients to share “stories” about themselves, their health, and their care (e.g., goals and feelings) before a doctor visit. A one-screen summary is then provided to clinical staff via the EHR. The key to high adoption is that they don’t “drill” the patient over things such as medications taken, allergies, etc. They focus instead on distilling open-ended responses about what matters to patients as people, which patients like and providers also value.

Sam Margolis, VP of client strategy and growth at Cantina, saw several aspects of the user experience (UX) as the main hurdle for health IT companies. This focus was reasonable, given that Cantina combines strengths in design and development. Margolis said that companies find it hard to make their interfaces simple and to integrate into the environments where their products operate. He pointed out that health care involves complex environments with many considerations. He also said they should be thinking holistically and design a service, not just a product–a theme I have seen across modern business in general, where companies are striving to engage customers over long periods of time, not just sell isolated objects.

Phil Marshall, MD, co-founder and chief product officer of Conversa Health, described how they offer a chatbot to patients discharged from one partnering hospital, in pursuit of the universal goal by US hospitals to avoid penalties from Medicare for readmissions. The app asks the patient for information about her condition and applies the same standards the hospital uses when its staff evaluates discharged patients. Marshall said that the standards make the chatbot highly accurate, and is tuned regularly. It is also popular: 80 percent of the patients offered the app use it, and 97 percent of these say it is helpful. The chat is tailored to each patient. In addition to relieving the staff of a routine task, the hospital found that the app reduces variation among outcomes among physicians, because the chatbot will ask for information they might forget.

Jay V. Patel, Clinical Transformation Officer at Seniorlink, described a care management program that balances technology and the human touch to help caregivers of people with dementia. Called VOICE (Vital Outcomes Inspired by Caregiver Engagement) Dementia Care, the program connects a coach to family caregivers and their care teams through Vela, Seniorlink’s collaboration platform. The VOICE DC program reduced ER visits by 51 percent and hospitalizations by 18 percent in the six-month pilot. It was also good for caregivers, reducing their stress and increasing their confidence.

Despite the name, VOICE DC is text-based (with video content) rather than voice-based. An example of the advances in voice interfaces was provided at this conference by Boston Children’s Hospital. Elizabeth Kidder, manager of their digital health accelerator, reported using voice interfaces to let patients ask common questions, such as when to get vaccinations and whether an illness was bad enough to keep children home from school and day care. Another non-voice app they use is a game that identifies early whether a child has a risk of dyslexia. Starting treatment before the children are old enough to learn reading in school can greatly increase success.

Nathan Treloar, president of Orbita, reported that at a recent conference on voice interfaces, participants in a hackathon found nine use cases for them in health.

Pattie Maes of the MIT Media Lab–one of the most celebrated research institutions in digital innovation–envisions using devices to strengthen the very skills that our devices are now blamed for weakening, such as how to concentrate. Of course, she warned, there is a danger that users will become dependent on the device while using it for such skills.

Working at the top of one’s license

I heard that appealing phrase from Christine Goscila, a family nurse practitioner at Massachusetts General Hospital Revere. She was describing how an app makes it easier for nurses to collect data from remote patients and spend more time on patient care. This shift from routine tasks to high-level interactions is a major part of the promise of connected health.

I heard a similar goal from Gregory Pelton, MD, CMO of ICmed, one of the many companies providing an integrated messaging platform for patients, clinicians, and family caregivers. Pelton talks of handling problems at the lowest possible level. In particular, the doctor is relieved of entering data because other team members can do it. Furthermore, messages can prepare the patient for a visit, rendering him more informed and better able to make decisions.

Clinical trials get smarter

While most health IT and connected health practitioners focus on the doctor/patient interaction and health in the community, the biggest contribution connected health might make to cost-cutting may come from its use by pharmaceutical companies. As we watch the astounding rise in drug costs–caused by a range of factors I will cover in a later article, but only partly by deliberate overcharging–we could benefit from anything that makes research and clinical trials more efficient.

MITRE, a non-profit that began in the defense industry but recently has created a lot of open source tools and standards for health care, presented their Synthea platform, offering synthetic data for researchers. The idea behind synthetic data is that, when you handle a large data set, you don’t need to know that a particular patient has congestive heart failure, is in his sixties, and weighs 225 pounds. Even if the data is deidentified, giving information about each patient raises risks of reidentification. All you need to know is a collection of facts about diagnoses, age, weights, etc. that match a typical real patient population. If generated using rigorous statistical algorithms, fake data in large quantities can be perfectly usable for research purposes. Synthea includes data on health care costs as well as patients, and is used for FHIR connectathons, education, the free SMART Health IT Sandbox, and many other purposes.

Telemedicine

Payers are gradually adapting their reimbursements to telemedicine. The simplest change is just to pay for a video call as they would pay for an office visit, but this does not exploit the potential for connected health to create long-range, continuous interactions between doctor, patient, and other staff. But many current telemedicine services work outside the insurance system, simply charging patients for visits. This up-front payment obviously limits the ability of these services to reach most of the population.

The uncertainties, as well as the potential, of this evolving market are illustrated by the business model chosen by American Telephysicians, which goes so far as to recruit patients internationally, such as from Pakistan and Dubai, to create a telemedicine market for U.S. specialists. They will be starting services in some American communities soon, though. Taking advantage of the ubiquity of mobil devices, they extend virtual visits with online patient records and a marketplace for pharmaceuticals, labs, and radiology. Waqas Ahmed, MD, founder and CEO, says: “ATP is addressing global health care problems that include inaccessibility of primary, specialty, and high-quality healthcare services, lack of price transparency, substandard patient education, escalating costs and affordability, a lack of healthcare integration, and fragmentation along the continuum of care.”

The network is the treatment center

We were honored with a keynote from FCC chair Ajit Pai, who achieved notoriety recently in the contentious “net neutrality” debate and was highlighted in WIRED for his position. Pai is not the most famous FCC chair, however; that honor goes to Newton Minow, who as chair from 1961 to 1963 called television a “vast wasteland.” More recently, Michael Powell (who became chair in 2001, before the confounding term “net neutrality” was invented) garnered a lot of attention for changing Internet regulations. Newton Minow, by the way, is still on the scene. I heard him talk recently at a different conference, and Pai mentioned talking to Minow about Internet access.

Pai has made expansion of Internet access his key issue (it was mentioned in the WIRED article) and talked about the medical benefits of bringing fast, continuous access to rural areas. His talk fit well with the focus many companies at the Connected Health conference placed on telemedicine. But Pai did not vaunt competition or innovation as a solution to reaching rural areas. Instead, he seemed happy with the current oligopoly that characterizes Internet access in most areas, and promoted an increase in funding to get them to do more of what they’re now doing (slowly).

The next day, Nancy Green of Verizon offered a related suggestion that 5G wireless will make batteries in devices last longer. This is not intuitive, but I think can be justified by the decrease in the time it will take for devices to communicate with the cloud, decreasing in turn the drain on the batteries.

Devices that were just cool

One device I liked at Connected Health coll was the Eko stethoscope, which sends EKG data to a computer for display. Patients will soon be able to use Eko devices to view their own EKGs, along with interpretations that help non-specialists make sense of the results. Of course, the results are also sent to the patients’ doctors.

Another device is a smart pillbox by CUEMED that doubles as a voice-interactive health assistant, HEXIS. Many companies make smart pill boxes that keep track of whether you open them, and flash or speak up to remind you when it’s time to take the pills. (Non-compliance with prescription medications is rampant.) HEXIS is a more advanced innovation that incorporates Alexa-like voice interactivity with the user and can connect to other medical devices and wearables such as Apple Watch and blood pressure monitors. The device uses the data and vital signs to motivate the user, and provides suggestions for the user to feel better. Another nice feature is that if you’re going out, you can remove one day’s meds and take them with you, while the device continues to do its job of reminding and tracking.

I couldn’t get to every valuable session at the Connected Health conference, or cover every speaker I heard. However, the conference seems to be achieving its goals of bringing together innovators and of prodding the health care industry toward the effective use of technology.

Software Marks Advances at the Connected Health Conference (Part 1 of 2)

Posted on October 29, 2018 I Written By

Andy Oram is an editor at O'Reilly Media, a highly respected book publisher and technology information provider. An employee of the company since 1992, Andy currently specializes in open source, software engineering, and health IT, but his editorial output has ranged from a legal guide covering intellectual property to a graphic novel about teenage hackers. His articles have appeared often on EMR & EHR and other blogs in the health IT space. Andy also writes often for O'Reilly's Radar site (http://oreilly.com/) and other publications on policy issues related to the Internet and on trends affecting technical innovation and its effects on society. Print publications where his work has appeared include The Economist, Communications of the ACM, Copyright World, the Journal of Information Technology & Politics, Vanguardia Dossier, and Internet Law and Business. Conferences where he has presented talks include O'Reilly's Open Source Convention, FISL (Brazil), FOSDEM, and DebConf.

The precepts of connected health were laid out years ago, and merely get updated with nuances and technological advances at each year’s Connected Health conference. The ideal of connected health combines matching the insights of analytics with the real-life concerns of patients; monitoring people in everyday settings through devices that communicate back to clinicians and other caregivers; and using automation to free up doctors to better carry out human contact. Pilots and deployments are being carried out successfully in scattered places, while in others connected health languishes while waiting for the slow adoption of value-based payments.

Because I have written at length about the Connected Health conference in 2015, 2016, and 2017, I will focus this article on recent trends I ran into at this year’s conference. Key themes include precertification at the FDA, the state of interoperability (which is poor), and patient engagement.

Exhibition floor at Connected Health conference

Exhibition floor at Connected Health conference

Precertification: the status of streamlining approval for medical software

One of the ongoing challenges in the progress of patient involvement and connected health is the approval of software for diagnosis and treatment. Traditionally, the FDA regulated software and hardware together in all devices used in medicine, requiring rigorous demonstrations of safety and efficacy in a manner similar to drugs. This was reasonable until recently, because anything that the doctor gives to the patient needs to be carefully checked. Otherwise, insurers can waste a lot of money on treatments that don’t work, and patients can even be harmed.

But more and more software is offered on generic computers or mobile devices, not specialized medical equipment. And the techniques used to develop the software inherit the “move fast and break things” mentality notoriously popular in Silicon Valley. (The phrase was supposedly a Facebook company motto.) Software can be updated several times a day. Although A/B testing (an interesting parallel to randomized controlled trials) might be employed to see what is popular with users, quality control is done in completely different ways. Modern software tends to rely for safety and quality on unit tests (which make sure individual features work as expected), regression tests (which look for things that no longer work they way they should), continuous integration (which forces testing to run each time a change is submitted to the central repository), and a battery of other techniques that bear such names as static testing, dynamic testing, and fuzz testing. Security testing is yet another source of reliability, using techniques such as penetration testing that may be automated or manual. (Medical devices, which are notoriously insecure, might benefit from an updated development model.

The FDA has realized that reliable software can be developed within the Silicon Valley model, so long as rigor and integrity are respected. Thus, it has started a Pre-Cert Pilot Program that works with nine brave vendors to find guidelines the FDA can apply in the future to other software developers.

Representatives of four vendors reported at the Connected Health conference that the pilot is going quite well, with none of the contentious and adversarial atmosphere that characterizes the interactions between the FDA with most device manufacturers. Every step of the software process is available for discussion and checking, and the inquiries go quite deep. All participants are acutely aware of the risk–cited by critics of the program–that it will end up giving vendors too much leeway and leaving the public open to risks. The participants are committed to closing loopholes and making sure everyone can trust the resulting guidelines.

The critical importance of open source software became clear in the report of the single open source vendor who is participating in the pilot: Tidepool. Because it is open source, according to CEO Howard Look, Tidepool was willing to show its code as well as its software development practices to independent experts using multiple evaluation assessment methods, including a “peer appraisal” by fellow precert participants Verily and Pear Therapeutics. One other test appraisal (CMMI, using external auditors) was done by both Tidepool and Johnson & Johnson; no other participants did a test appraisal. Thus, if the FDA comes out with new guidelines that stimulate a tremendous development of new software for medical use, we can thank open source.

Making devices first-class players in health care

Several exhibitors at the conference were consulting firms who provide specific services to start-ups and other vendors trying to bring products to market. I asked a couple of these consultants what they saw as the major problems their clients face. Marcus Fontaine, president of Impresiv Health, said their biggest problem is the availability of data, particularly because of a lack of interoperable data exchange. I wanted to exclaim, “Still?”

Joseph Kvedar, MD, who chairs the Connected Health conference, spoke of a new mobile app developed by his organization, Partners Connected Health, to bring device data into their EHR. This greatly improves the collection of data and guarantees accuracy, because patients no longer have to manually enter vital signs or other information. In addition to serving Partners in improving patient care, the data can be used for research and public health. In developing this app, Partners depended heavily for interoperable data exchange on work by Validic, the most prominent company in the device interoperability space, and one that I have profiled and whose evolution I have followed.

Ideally, each device could communicate directly with the EHR. Why would Partners Connected Health invest heavily in creating a special app as an intermediary? Kvedar cited several reasons. First, each device currently offers its own app as a user interface, and users with multiple devices get confused and annoyed by the proliferation of apps. Second, many devices are not designed to communicate cleanly with EHRs. Finally, the way networks are set up, communicating would require a separate cellular connection and SIM card for each device, raising costs.

A similar effort is pursued by Indie Health, trying to solve the problem of data access by making it easy to create Bluetooth connections between devices and mobile phones using a variety of Bluetooth, IEEE, Continua, and other standards.

The CEO of Validic, Drew Schiller, spoke on another panel about maximizing the value of patient-generated data. He pointed out that Validic, as an intermediary for a huge number of devices and health care providers, possesses a correspondingly huge data set on how patients are using the devices, and in particular when they stop using the devices. I assume that Validic does not preserve the data generated by the devices, such as blood pressure or steps taken–at least, Schiller did not say they have that data, and it would be intrusive to collect it. However, the metadata they do collect can be very useful in designing interactions with patients. He also talked about the value of what he dubs “invisible health care,” where behavior change and other constructive uses of data can flow easily from the data.

Barry Reinhold, president and CTO of Lamprey Networks, was manning the Continua booth when I came by. Continua defines standard for devices used in the home, in nursing faciliies, and in other places outside the hospital. This effort should be open source, supported by fees by all affected stakeholders (hospitals, device manufacturers, etc.). But open source is spurned by the health care field, so Continua does the work as a private company. Reinhold told me that device manufacturers rarely contract with Continua, which I treat as a sign that device manufacturers value data silos as a business model. Instead, Continua contracts come from the institutions that desperately need access to the data, such as nursing facilities. Continua does the best it can to exploit existing standards, including the “continuing data” profile from FHIR.

Other speakers at the conference, including Andrew Hayek, CEO of OptumHealth, confirmed Reinhold’s observation that interoperability still lags among devices and EHRs. And Schiller of Validic admitted that in order to get data from some devices into a health system, the patient has to take a photo of the device’s screen. Validic not only developed an app to process the photo, but patented it–a somewhat odd indication that they consider it a major contribution to health care.

Tasha van Es and Claire Huber of Redox, a company focused on healthcare interoperability and data integration, said that they are eager to work with FHIR, and that it’s a major part of their platform, but they think it has to develop more before being ready for widespread use. This made me worry about recent calls by health IT specialists for the ONC, CMS, and FDA to make FHIR a requirement.

It was a pleasure to reconnect at the conference with goinvo, which creates open source health care software on a contract basis, but offers much of it under a free license.

A non-profit named Xcertia also works on standards in health care. Backed by the American Medical Association, American Heart Association, DHX Group, and HIMSS, they focus on security, privacy, and usability. Although they don’t take on certification, they design their written standards so that other organizations can offer certification, and a law considered in California would mandate the use of their standards. The guidelines have just been released for public comment.

The second section of this article covers patient engagement and other topics of interest that turned up at the conference.

Patient Billing And Collections Process Needs A Tune-Up

Posted on October 1, 2018 I Written By

Anne Zieger is a healthcare journalist who has written about the industry for 30 years. Her work has appeared in all of the leading healthcare industry publications, and she's served as editor in chief of several healthcare B2B sites.

A new study from a patient payments vendor suggests that many healthcare organizations haven’t optimized their patient billing and collections process, a vulnerability which has persisted despite their efforts to crack the problem.

The survey found that while the entire billing collections process was flawed, respondents said that collecting patient payments was the toughest problem, followed by the need to deploy better tools and technologies.

Another issue was the nature of their collections efforts. Sixty percent of responding organizations use collections agencies, an approach which can establish an adversarial relationship between patient and provider and perhaps drive consumers elsewhere.

Yet another concern was long delays in issuing bills to patients. The survey found that 65% of organizations average more than 60 days to collect patient payments, and 40% waited on payments for more than 90 days.

These results align other studies that look at patient payments, all of which echo the notion that the patient collection process is far from what it should be.

For example, a study by payment services vendor InstaMed found that more than 90% of consumers would like to know what the payment responsibility is prior to a provider visit. Worse, very few consumers even know what the deductible, co-insurance and out-of-pocket maximums are, making it more likely that the will be hit with a bill they can’t afford.

As with the Cedar study, InstaMed’s research found that providers are waiting a long time to collect patient payments, three-quarters of organizations waiting a month to close out patient balances.

Not only that, investments in revenue cycle management technology aren’t necessarily enough to kickstart patient payment volumes. A survey done last year by the Healthcare Financial Management Association and vendor Navigant found that while three-quarters of hospitals said that their RCM technology budget was increasing, they weren’t necessarily getting the ROI they’d hoped to see.

According to the survey, 77% of hospitals less than 100 beds and 78% of hospitals with 100 to 500 beds planned to increase their RCM spending. Their areas of investment included business intelligence analytics, EHR-enabled workflow or reporting, revenue integrity, coding and physician/clinician documentation options.

Still, process improvements seem to have had a bigger payoff. These hospitals are placing a lot of faith in revenue integrity programs, with 22% saying that revenue integrity was a top RCM focus area for this year. Those who would already put such a program in place said that it offered significant benefits, including increased net collections (68%), greater charge capture (61%) and reduced compliance risks (61%).

As I see it, the key takeaways here are that making sure patients know what to expect financially and putting programs in place to improve internal processes can have a big impact on patient payments. Still, with consumers financing a lot of their care these days, getting their dollars in the door should continue to be an issue. After all, you can’t get blood from a stone.

Applying AI Based Outlier Detection to Healthcare – Interview with Dr. Gidi Stein from MedAware

Posted on September 17, 2018 I Written By

John Lynn is the Founder of the HealthcareScene.com blog network which currently consists of 10 blogs containing over 8000 articles with John having written over 4000 of the articles himself. These EMR and Healthcare IT related articles have been viewed over 16 million times. John also manages Healthcare IT Central and Healthcare IT Today, the leading career Health IT job board and blog. John is co-founder of InfluentialNetworks.com and Physia.com. John is highly involved in social media, and in addition to his blogs can also be found on Twitter: @techguy and @ehrandhit and LinkedIn.

Most people who receive healthcare understand that healthcare is as much art as it is science. We don’t expect our doctors to be perfect or know everything because the human body is just too complex and there are so many factors that influence health. What’s hard for patients to understand is when obvious human errors occur. This is especially true when technology or multiple layers of humans should have caught the obvious.

This is exactly why I was excited to interview Dr. Gidi Stein, CEO and Co-founder of MedAware. As stated on their website, their goal is to eliminate prescription errors. In the interview below, you’ll learn more about what MedAware and Dr. Stein are doing to achieve this goal.

Tell us a little about yourself and MedAware.

Early in my career, I worked in the Israeli high-tech industry and served as CTO and Chief Architect of several algorithm-rich startups. However, after many years working in technology, I decided to return to school and study medicine. In 2002, I graduated from Tel Aviv University Medical School with a specialization in internal medicine, treating patients and teaching students and residents in one of Israel’s largest hospitals.

After working as a physician for several years, I heard a heartbreaking story, which ultimately served as my motivation and inspiration to found MedAware. A physician was treating a 9-year-old boy who suffered from Asthma. To treat the symptoms, the physician entered the electronic prescribing environment and selected Singulair from the drop-down menu, a standard treatment for asthma. However, unfortunately, he accidentally clicked Sintrom, an anticoagulant (blood thinner). Tragically, neither the physician, pharmacist nor parent caught this error, which resulted in the boys’ untimely death. This avoidable, medication-related complication and death was caused by a typo.

Having worked as a physician for many years, I had a difficult time understanding that with all the medical intervention and technological support we rely on, our healthcare system was not intelligent enough to prevent errors like this. This was a symptom of a greater challenge; how can we identify and prevent medication related complications before they occur? Given my combined background in technology and medicine, I knew that there must be a solution to eliminate these types of needless errors. I founded MedAware to transform patient safety and save lives.

Describe the problem with prescription-based medication errors that exists today.  What’s the cause of most of these errors?

Every year in the U.S. alone, there are 1.5 million preventable medication errors, which result in patient injury or death. In fact, medication errors are the third leading cause of death in the US, and errors related to incorrect prescription are a major part of these. Today’s prescription-related complications fall into two main categories: medication errors that occur at the point of order entry (like the example of the 9yr old boy) and errors that result from evolving adverse drug events (ADEs). Point of order entry errors are a consequence of medication reconciliation challenges, typos, incorrect dosage input and other clinical inconsistencies.

Evolving ADEs are, in fact, the bulk of the errors that occur – almost 2/3 of errors are those that happen after a medication was correctly prescribed. These are often the most catastrophic errors, as they are completely unforeseen, and don’t necessarily result from physician error. Rather, they occur when a patient’s health status has changed, and a previously safe medication becomes unsafe.

MedAware uses AI to detect outlier prescriptions.  It seems that everything is being labeled AI, so how does this work and how effective is it at detecting medication errors?

AI is best used to analyze large scale data to identify patterns and outliers to those patterns. The common theme in industries, such as aviation, cyber security and credit card fraud, is that they are rich with millions of transactions, 99.99% of which are okay. But, a small fraction of them are hazardous, and these dangers most often occur in new and unexpected ways. In these industries, AI is used to crunch millions of transactions, identify patterns, and most importantly, identify outliers to those patterns as potential hazards with high accuracy.

Medication safety is similar to these industries. Here too, millions of medications are prescribed and dispensed every day, and in 99.99% of cases, the right medication is prescribed and dispensed to the right patient. But, on rare occasions, an unexpected error or oversight may put patients at risk. MedAware analyzes millions of clinical records to identify errors and oversights as statistical outliers to the normal behavioral patterns of providers treating similar patients. Our data shows that this methodology, identifies errors and ADEs with high accuracy and clinical relevance and that most of the errors found by our system would not have been caught by any other existing system.

Are most of the errors you find obvious errors that a human could have detected but just missed or are you finding surprising errors as well?  Can you share some stories of what you’ve found?

The errors that we find are obvious errors; any physician would agree that they are indeed erroneous. These include: prescribing chemotherapy to healthy individuals, not stopping anticoagulation to a bleeding patient, birth control pills to a 70-year-old male and prescribing Viagra to a 2-year-old baby. All of these are obvious errors, so why didn’t the prescribers pick these up? The answer is simple: they are human, and humans err, especially when they are less experienced and over worked. Our software is able to mirror back to the providers the crowdsourced behavioral patterns of their peers and identify outliers to these patterns as errors.

You recently announced a partnership with Allscripts and their dbMotion interoperability solution.  How does that work and what’s the impact of this partnership?

Today’s healthcare systems have created a reality where patient health information can be scattered across multiple health systems, infrastructures and EHRs. The dbMotion health information exchange platform aggregates and harmonizes that scattered patient data, delivering the information clinicians need in a usable and actionable format at the point of care, within the provider’s native and familiar workflow. With dbMotion, all of the patient’s records are in one place. MedAware sits on top of the bdMotion interoperability platform as a layer of safety, accurately looking at the thousands of clinical inputs in the system and warning with even greater accuracy. MedAware catches various medication errors that would have been missed due to a decentralized patient health record. In addition to identifying prescription-based medication errors, MedAware can also notify physicians of patients who are at risk of opioid addiction.

This partnership will allow any institution using Allscripts’ dbMotion to easily implement MedAware’s system in a streamlined manner, with each installation being quick and effortless.

Once MedAware identifies a prescription error, how do you communicate that information back to the provider? Do you integrate your solution with the EHR vendor?

Yes, MedAware is integrated with EHR platforms. This is necessary for error detection and communication of the warning to the provider. There are two intervention scenarios: 1) Synchronous – when errors are caught at the point of order entry, a popup alert appears within the EHR user interface, without disturbing the provider’s workflow, and the provider can choose to accept or reject the alert. 2) Asynchronous – the errors/ADE is caught following a change in the patient’s clinical record (i.e. new lab result or vital sign), long after the prescription was entered. These alerts are displayed as a physician’s task, within the physician’s workflow and the EHR’s user interface.

What’s next for MedAware?  Where are you planning to take this technology?

The next steps for us are:

  1. Scale our current technology to grow to 20 million lives analyzed by 2020
  2. Create additional patient safety centered solutions to providers, such as opioid dependency risk assessment, gaps in care and trend projection analysis.
  3. Share our life-saving insights directly to those who need it most – consumers.

Does NLP Deserve To Be The New Hotness In Healthcare?

Posted on August 30, 2018 I Written By

Anne Zieger is a healthcare journalist who has written about the industry for 30 years. Her work has appeared in all of the leading healthcare industry publications, and she's served as editor in chief of several healthcare B2B sites.

Lately, I’ve been seeing a lot more talk about the benefits of using natural language processing technology in healthcare. In fact, when I Googled the topic, I turned up a number of articles on the subject published over the last several weeks. Clearly, something is afoot here.

What’s driving the happy talk? One case in point is a new report from health IT industry analyst firm Chilmark Research laying out 12 possible use cases for NLP in healthcare.

According to Chilmark, some of the most compelling options include speech recognition, clinical documentation improvement, data mining research, computer-assisted coding and automated registry reporting. Its researchers also seem to be fans of clinical trial matching, prior authorization, clinical decision support and risk adjustment and hierarchical condition categories, approaches it labels “emerging.”

From what I can see, the highest profile application of NLP in healthcare is using it to dig through unstructured data and text. For example, a recent article describes how Intermountain Healthcare has begun identifying heart failure patients by reading data from 25 different free text documents stored in the EHR. Clearly, exercises like these can have an immediate impact on patient health.

However, stories like the above are actually pretty unusual. Yes, healthcare organizations have been working to use NLP to mine text for some time, and it seems like a very logical way to filter out critical information. But is there a reason that NLP use even for this purpose isn’t as widespread as one might think? According to one critic, the answer is yes.

In a recent piece, Dale Sanders, president of technology at HealthCatalyst, goes after the use of comparative data, predictive analytics and NLP in healthcare, arguing that their benefits to healthcare organizations have been oversold.

Sanders, who says he came to healthcare with a deep understanding of NLP and predictive analytics, contends that NLP has had ”essentially no impact” on healthcare. ”We’ve made incremental progress, but there are fundamental gaps in our industry’s data ecosystem– missing pieces of the data puzzle– that inherently limit what we can achieve with NLP,” Sanders argues.

He doesn’t seem to see this changing in the near future either. Given how much money has already been sunk in the existing generation of EMRs, vendors have no incentive to improve their capacity for indexing information, Sanders says.

“In today’s EMRs, we have little more than expensive word processors,” he writes. “I keep hoping that the Googles, Facebooks and Amazons of the world will quietly build a new generation EMR.” He’s not the only one, though that’s a topic for another article.

I wish I could say that I side with researchers like Chilmark that see a bright near-term future for NLP in healthcare. After all, part of why I love doing what I do is exploring and getting excited about emerging technologies with high potential for improving healthcare, and I’d be happy to wave the NLP flag too.

Unfortunately, my guess is that Sanders is right about the obstacles that stand in the way of widespread NLP use in our industry. Until we have a more robust way of categorizing healthcare data and text, searching through it for value can only go so far. In other words, it may be a little too soon to pitch NLP’s benefits to providers.

Can Providers Survive If They Don’t Get Population Health Management Right?

Posted on August 27, 2018 I Written By

Anne Zieger is a healthcare journalist who has written about the industry for 30 years. Her work has appeared in all of the leading healthcare industry publications, and she's served as editor in chief of several healthcare B2B sites.

Most providers know that they won’t succeed with population health management unless they get some traction in a few important areas — and that if not, they could face disaster as their volume of value-based payment share grows. The thing is, getting PHM right is proving to be a mindboggling problem for many.

Let’s start with some numbers which give us at least one perspective on the situation.

According to a survey by Health Leaders Media, 87% of respondents said that improving their population health management chops was very important. Though the article summarizing the study doesn’t say this explicitly, we all know that they have to get smart about PHM if they want to have a prayer of prospering under value-based reimbursement.

However, it seems that the respondents aren’t making nearly as much PHM progress as they’d like. For example, just 38% of respondents told Health Leaders that they attributed 25% or more of their organization’s net revenue to risk-based pop health management activities, a share which has fallen two percent from last year’s results.

More than half (51%) said that their top barrier to successfully deploying or expanding pop health programs was up-front funding for care management, IT and infrastructure. They also said that engaging patients in their own care (45%) and getting meaningful data into providers’ hands (33%) weren’t proving to be easy tasks.

At this point it’s time for some discussion.

Obviously, providers grapple with competing priorities every time they try something new, but the internal conflicts are especially clear in this case.

On the one hand, it takes smart care management to make value-based contracts feasible. That could call for a time-consuming and expensive redesign of workflow and processes, patient education and outreach, hiring case managers and more.

Meanwhile, no PHM effort will blossom without the right IT support, and that could mean making some substantial investments, including custom-developed or third-party PHM software, integrating systems into a central data repository, sophisticated data analytics and a whole lot more.

Putting all of this in place is a huge challenge. Usually, providers lay the groundwork for a next-gen strategy in advance, then put infrastructure, people and processes into place over time. But that’s a little tough in this case. We’re talking about a huge problem here!

I get it that vendors began offering off-the-shelf PHM systems or add-on modules years ago, that one can hire consultants to change up workflow and that new staff should be on-board and trained by now. And obviously, no one can say that the advent of value-based care snuck up on them completely unannounced. (In fact, it’s gotten more attention than virtually any other healthcare issue I’ve tracked.) Shouldn’t that have done the trick?

Well, yes and no. Yes, in that in many cases, any decently-run organization will adapt if they see a trend coming at them years in advance. No, in that the shift to value-based payment is such a big shift that it could be decades before everyone can play effectively.

When you think about it, there are few things more disruptive to an organization than changing not just how much it’s paid but when and how along with what they have to do in return. Yes, I too am sick of hearing tech startups beat that term to death, but I think it applies in a fairly material sense this time around.

As readers will probably agree, health IT can certainly do something to ease the transition to value-based care. But HIT leaders won’t get the chance if their organization underestimates the scope of the overall problem.

An Interesting Overview Of Alphabet’s Healthcare Investments

Posted on June 27, 2018 I Written By

Anne Zieger is a healthcare journalist who has written about the industry for 30 years. Her work has appeared in all of the leading healthcare industry publications, and she's served as editor in chief of several healthcare B2B sites.

Recently I’ve begun reading a blog called The Medical Futurist which offers some very interesting fare. In addition to some intriguing speculation, it includes some research that I haven’t seen anywhere else. (It is written by a physician named Bertalan Mesko.)

In this case, Mesko has buried a shrewd and well-researched piece on Alphabet’s healthcare investments in an otherwise rambling article. (The rambling part is actually pretty interesting on its own, by the way.)

The piece offers a rather comprehensive update on Alphabet’s investments in and partnerships with healthcare-related companies, suggesting that no other contender in Silicon Valley is investing in this sector heavily as Alphabet’s GV (formerly Google Ventures). I don’t know if he’s right about this, but it’s probably true.

By Mesko’s count, GV has backed almost 60 health-related enterprises since the fund was first kicked off in 2009. These investments include direct-to-consumer genetic testing firm 23andme, health insurance company Oscar Health, telemedicine venture Doctor on Demand and Flatiron Health, which is building an oncology-focused data platform.

Mesko also points out that GV has had an admirable track record so far, with five of the companies it first backed going public in the last year. I’m not sure I agree that going public is per se a sign of success — a lot depends on how the IPO is received by Wall Street– but I see his logic.

In addition, he notes that Alphabet is stocking up on intellectual resources. The article cites research by Ernest & Young reporting that Alphabet filed 186 healthcare-related patents between 2013 and 2017.

Most of these patents are related to DeepMind, which Google acquired in 2014, and Verily Life Sciences (formerly Google Life Sciences). While these deals are interesting in and of themselves, on a broader level the patents demonstrate Alphabet’s interest in treating chronic illnesses like diabetes and the use of bioelectronics, he says.

Meanwhile, Verily continues to work on a genetic data-collecting initiative known as the Baseline Study. It plans to leverage this data, using some of the same algorithms behind Google’s search technology, to pinpoint what makes people healthy.

It’s a grand and somewhat intimidating picture.

Obviously, there’s a lot more to discuss here, and even Mesko’s in-depth piece barely scratches the surface of what can come out of Alphabet and Google’s health investments. Regardless, it’s worth keeping track of their activity in the sector even if you find it overwhelming. You may be working for one of those companies someday.

Healthcare AI Needs a Breadth and Depth of Data

Posted on May 17, 2018 I Written By

John Lynn is the Founder of the HealthcareScene.com blog network which currently consists of 10 blogs containing over 8000 articles with John having written over 4000 of the articles himself. These EMR and Healthcare IT related articles have been viewed over 16 million times. John also manages Healthcare IT Central and Healthcare IT Today, the leading career Health IT job board and blog. John is co-founder of InfluentialNetworks.com and Physia.com. John is highly involved in social media, and in addition to his blogs can also be found on Twitter: @techguy and @ehrandhit and LinkedIn.

Today I’m enjoying the New England HIMSS Spring Conference including an amazing keynote session by Dale Sanders from Health Catalyst. Next week I’ll be following up this blog post with some other insights that Dale shared at the New England HIMSS event, but today I just wanted to highlight one powerful concept that he shared:

Healthcare AI Needs a Breadth and Depth of Data

As part of this idea, Dale shared the following image to illustrate how much data is really needed for AI to effectively assess our health:

Dale pointed out that in healthcare today we really only have access to the data in the bottom right corner. That’s not enough data for AI to be able to properly assess someone’s health. Dale also suggested the following about EHR data:

Long story short, the EHR data is not going to be enough to truly assess someone’s health. As Google recently proved, a simple algorithm with more data is much more powerful than a sophisticated algorithm with less data. While we think we have a lot of data in healthcare, we really don’t have that much data. Dale Sanders made a great case for why we need more data if we want AI to be effective in healthcare.

What are you doing in your organization to collect data? What are you doing to get access to this data? Does collection of all of this data scare anyone? How far away are we from this data driven, AI future? Let us know your thoughts in the comments.

Google And Fitbit Partner On Wearables Data Options

Posted on May 7, 2018 I Written By

Anne Zieger is a healthcare journalist who has written about the industry for 30 years. Her work has appeared in all of the leading healthcare industry publications, and she's served as editor in chief of several healthcare B2B sites.

Fitbit and Google have announced plans to work together, in a deal intended to “transform the future of digital health and wearables.” While the notion of transforming digital health is hyperbole even for companies the size of Google and Fitbit, the pairing does have plenty of potential.

In a nutshell, Fitbit and Google expect to take on both consumer and enterprise health projects that integrate data from EMRs, wearables and other sources of patient information together. Given the players involved, it’s hard to doubt that at least something neat will emerge from their union.

Among the first things the pair plans to use Google’s new Cloud Healthcare API to connect Fitbit data with EMRs. Of course, readers will know that it’s one thing to say this and another to actually do it, but gross oversimplifications aside, the idea is worth pursuing.

Also, using services such as those offered by Twine Health– a recent Fitbit acquisition — the two companies will work to better manage chronic conditions such as diabetes and hypertension. Twine offers a connected health platform which leverages Fitbit data to offer customized health coaching.

Of course, as part of the deal Fitbit is moving to the Google Cloud Platform, which will supply the expected cloud services and engineering support.

The two say that moving to the Cloud Platform will offer Fitbit advanced security capabilities which will help speed up the growth of Fitbit Health Solutions business. They also expect to make inroads in population health analysis. For its part, Google also notes that it will bring its AI, machine learning capabilities and predictive analytics algorithms to the table.

It might be worth a small caution here. Google makes a point of saying it is “committed” to meeting HIPAA standards, and that most Google Cloud products do already. That “most” qualifier would make me a little bit nervous as a provider, but I know, why worry about these niceties when big deals are afoot. However, fair warning that when someone says general comments like this about meeting HIPAA standards, it probably means they already employ high security standards which are likely better than HIPAA. However, it also means that they probably don’t comply with HIPAA since HIPAA is about more than security and requires a contractual relationship between provider and business associate and the associated liability of being a business associate.

Anyway, to round out all of this good stuff, Fitbit and Google said they expect to “innovate and transform” the future of wearables, pairing Fitbit’s brand, community, data and high-profile devices with Google’s extreme data management and cloud capabilities.

You know folks, it’s not that I don’t think this is interesting. I wouldn’t be writing about if I didn’t. But I do think it’s worth pointing out how little this news announcement says, really.

Yes, I realize that when partnerships begin, they are by definition all big ideas and plans. But when giants like Google, much less Fitbit, have to fall back on words like innovate and transform (yawn!), the whole thing is still pretty speculative. Just sayin’.