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!!

A Learning EHR for a Learning Healthcare System

Posted on January 24, 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.

Can the health care system survive the adoption of electronic health records? When the HITECH act mandated the installation of EHRs in 2009, we all hoped they would propel hospitals and clinics into a 21st-century consciousness. Instead, EHRs threaten to destroy those who have adopted them: the doctors whose work environment they degrade and the hospitals that they are pushing into bankruptcy. But the revolution in artificial intelligence that’s injecting new insights into many industries could also create radically different EHRs.

Here I define AI as software that, instead of dictating what a computer system should do, undergoes a process of experimentation and observation that creates a model to control the system, hopefully with far greater sophistication, personalization, and adaptability. Breakthroughs achieved in AI over the past decade now enable things that seemed impossible a bit earlier, such as voice interfaces that can both recognize and produce speech.

AI has famously been used by IBM Watson to make treatment recommendations. Analyses of big data (which may or may not qualify as AI) have saved hospitals large sums of money and even–finally, what we’ve been waiting for!–make patients healthier. But I’m talking in this article about a particular focus: the potential for changing the much-derided EHR. As many observers have pointed out, current EHRs are mostly billion-dollar file cabinets in electronic form. That epithet doesn’t even characterize them well enough–imagine instead a file cabinet that repeatedly screamed at you to check what you’re doing as you thumb through the papers.

How can AI create a new electronic health record? Major vendors have announced virtual assistants (See also John’s recent interview with MEDITECH which mentions their interest in virtual assistants) to make their interfaces more intuitive and responsive, so there is hope that they’re watching other industries and learning from machine learning. I don’t know what the vendors basing these assistants on, but in this article I’ll describe how some vanilla AI techniques could be applied to the EHR.

How a Learning EHR Would Work

An AI-based health record would start with the usual dashboard-like interface. Each record consists of hundreds of discrete pieces of data, such as age, latest blood pressure reading, a diagnosis of chronic heart failure, and even ZIP code and family status–important public health indicators. Each field of data would be called a feature in traditional AI. The goal is to find which combination of features–and their values, such as 75 for age–most accurately predict what a clinician does with the EHR. With each click or character typed, the AI model looks at all the features, discards the bulk of them that are not useful, and uses the rest to present the doctor with fields and information likely to be of value.

The EHR will probably learn that the forms pulled up by a doctor for a heart patient differ from those pulled up for a cancer patient. One case might focus on behavior, another on surgery and medication. Clinicians certainly behave differently in the hospital from how they behave in their home offices, or even how they behave in another hospital across town with different patient demographics. A learning EHR will discover and adapt to these differences, while also capitalizing on the commonalities in the doctor’s behavior across all settings, as well as how other doctors in the practice behave.

Clinicians like to say that every patient is different: well, with AI tracking behavior, the interface can adapt to every patient.

AI can also make use of messy and incomplete data, the well-known weaknesses of health care. But it’s crucial, to maximize predictive accuracy, for the AI system to have access to as many fields as possible. Privacy rules, however, dictate that certain fields be masked and others made fuzzy (for instance, specifying age as a range from 70 to 80 instead of precisely 75). Although AI can still make use of such data, it might be possible to provide more precise values through data sharing agreements strictly stipulating that the data be used only to improve the EHR–not for competitive strategizing, marketing, or other frowned-on exploitation.

A learning EHR would also be integrated with other innovations that increase available data and reduce labor–for instance, devices worn by patients to collect vital signs and exercise habits. This could free up doctors do less time collecting statistics and more time treating the patient.

Potential Impacts of AI-Based Records

What we hope for is interfaces that give the doctor just what she needs, when she needs it. A helpful interface includes autocompletion for data she enters (one feature of a mobile solution called Modernizing Medicine, which I profiled in an earlier article), clear and consistent displays, and prompts that are useful instead of distracting.

Abrupt and arbitrary changes to interfaces can be disorienting and create errors. So perhaps the EHR will keep the same basic interface but use cues such as changes in color or highlighted borders to suggest to the doctor what she should pay attention to. Or it could occasionally display a dialog box asking the clinician whether she would like the EHR to upgrade and streamline its interface based on its knowledge of her behavior. This intervention might be welcome because a learning EHR should be able to drastically reduce the number of alerts that interrupt the doctors’ work.

Doctors’ burdens should be reduced in other ways too. Current blind and dumb EHRs require doctors to enter the same information over and over, and even to resort to the dangerous practice of copy and paste. Naturally, observers who write about this problem take the burden off of the inflexible and poorly designed computer systems, and blame the doctors instead. But doing repetitive work for humans is the original purpose of computers, and what they’re best at doing. Better design will make dual entries (and inconsistent records) a thing of the past.

Liability

Current computer vendors disclaim responsibility for errors, leaving it up the busy doctor to verify that the system carried out the doctor’s intentions accurately. Unfortunately, it will be a long time (if ever) before AI-driven systems are accurate enough to give vendors the confidence to take on risk. However, AI systems have an advantage over conventional ones by assigning a confidence level to each decision they make. Therefore, they could show the doctor how much the system trusts itself, and a high degree of doubt could let the doctor know she should take a closer look.

One of the popular terms that have sprung up over the past decade to describe health care reform is the “learning healthcare system.” A learning system requires learning on every level and at every stage. Because nobody likes the designs of current EHRs, they should be happy to try a new EHR with a design based directly on their behavior.

Key Articles in Health IT from 2017 (Part 2 of 2)

Posted on January 4, 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 set a general context for health IT in 2017 and started through the year with a review of interesting articles and studies. We’ll finish the review here.

A thoughtful article suggests a positive approach toward health care quality. The author stresses the value of organic change, although using data for accountability has value too.

An article extolling digital payments actually said more about the out-of-control complexity of the US reimbursement system. It may or not be coincidental that her article appeared one day after the CommonWell Health Alliance announced an API whose main purpose seems to be to facilitate payment and other data exchanges related to law and regulation.

A survey by KLAS asked health care providers what they want in connected apps. Most apps currently just display data from a health record.

A controlled study revived the concept of Health Information Exchanges as stand-alone institutions, examining the effects of emergency departments using one HIE in New York State.

In contrast to many leaders in the new Administration, Dr. Donald Rucker received positive comments upon acceding to the position of National Coordinator. More alarm was raised about the appointment of Scott Gottlieb as head of the FDA, but a later assessment gave him high marks for his first few months.

Before Dr. Gottlieb got there, the FDA was already loosening up. The 21st Century Cures Act instructed it to keep its hands off many health-related digital technologies. After kneecapping consumer access to genetic testing and then allowing it back into the ring in 2015, the FDA advanced consumer genetics another step this year with approval for 23andMe tests about risks for seven diseases. A close look at another DNA site’s privacy policy, meanwhile, warns that their use of data exploits loopholes in the laws and could end up hurting consumers. Another critique of the Genetic Information Nondiscrimination Act has been written by Dr. Deborah Peel of Patient Privacy Rights.

Little noticed was a bill authorizing the FDA to be more flexible in its regulation of digital apps. Shortly after, the FDA announced its principles for approving digital apps, stressing good software development practices over clinical trials.

No improvement has been seen in the regard clinicians have for electronic records. Subjective reports condemned the notorious number of clicks required. A study showed they spend as much time on computer work as they do seeing patients. Another study found the ratio to be even worse. Shoving the job onto scribes may introduce inaccuracies.

The time spent might actually pay off if the resulting data could generate new treatments, increase personalized care, and lower costs. But the analytics that are critical to these advances have stumbled in health care institutions, in large part because of the perennial barrier of interoperability. But analytics are showing scattered successes, being used to:

Deloitte published a guide to implementing health care analytics. And finally, a clarion signal that analytics in health care has arrived: WIRED covers it.

A government cybersecurity report warns that health technology will likely soon contribute to the stream of breaches in health care.

Dr. Joseph Kvedar identified fruitful areas for applying digital technology to clinical research.

The Government Accountability Office, terror of many US bureaucracies, cam out with a report criticizing the sloppiness of quality measures at the VA.

A report by leaders of the SMART platform listed barriers to interoperability and the use of analytics to change health care.

To improve the lower outcomes seen by marginalized communities, the NIH is recruiting people from those populations to trust the government with their health data. A policy analyst calls on digital health companies to diversify their staff as well. Google’s parent company, Alphabet, is also getting into the act.

Specific technologies

Digital apps are part of most modern health efforts, of course. A few articles focused on the apps themselves. One study found that digital apps can improve depression. Another found that an app can improve ADHD.

Lots of intriguing devices are being developed:

Remote monitoring and telehealth have also been in the news.

Natural language processing and voice interfaces are becoming a critical part of spreading health care:

Facial recognition is another potentially useful technology. It can replace passwords or devices to enable quick access to medical records.

Virtual reality and augmented reality seem to have some limited applications to health care. They are useful foremost in education, but also for pain management, physical therapy, and relaxation.

A number of articles hold out the tantalizing promise that interoperability headaches can be cured through blockchain, the newest hot application of cryptography. But one analysis warned that blockchain will be difficult and expensive to adopt.

3D printing can be used to produce models for training purposes as well as surgical tools and implants customized to the patient.

A number of other interesting companies in digital health can be found in a Fortune article.

We’ll end the year with a news item similar to one that began the article: serious good news about the ability of Accountable Care Organizations (ACOs) to save money. I would also like to mention three major articles of my own:

I hope this review of the year’s articles and studies in health IT has helped you recall key advances or challenges, and perhaps flagged some valuable topics for you to follow. 2018 will continue to be a year of adjustment to new reimbursement realities touched off by the tax bill, so health IT may once again languish somewhat.

Key Articles in Health IT from 2017 (Part 1 of 2)

Posted on January 2, 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.

This article provides a retrospective of 2017 in Health It–but a retrospective from an unusual perspective. I will highlight interesting articles I’ve read from the year as pointers to trends we should follow up on in the upcoming years.

Indubitably, 2017 is a unique year due to political events that threw the field of health care into wild uncertainty and speculation, exemplified most recently by the attempts to censor the use of precise and accurate language at the Centers for Disease Control (an act of political interference that could not be disguised even by those who tried to explain it away). Threats to replace the Affordable Care Act (another banned phrase) drove many institutions, which had formerly focused on improving communications or implementing risk sharing health care costs, to fall back into a lower level of Maslow’s hierarchy of needs, obsessing over whether insurance payments would cease and patients would stop coming. News about health IT was also drowned out by more general health topics such as drug pricing, the opiate crisis, and revenue pressures that close hospitals.

Key issues

But let’s start our retrospective on an upbeat note. A brief study summary from January 4 reported lower costs for some surgeries when hospitals participated in a modest bundled payment program sponsored by CMS. This suggests that fee-for-value could be required more widely by payers, even in the absence of sophisticated analytics and care coordination. Because only a small percentage of clinicians choose bold risk-sharing reimbursement models, this news is important.

Next, a note on security. Maybe we should reprioritize clinicians’ defenses against the electronic record breaches we’ve been hearing so much about. An analysis found that the most common reason for an unauthorized release of data was an attack by an insiders (43 percent). This contrasts with 26.8 percent from outside intruders. (The article doesn’t say how many records were compromised by each breach, though–if they had, the importance of outside intruders might have skyrocketed.) In any case, watch your audit logs and don’t trust your employees.

In a bracing and rare moment of candor, President Obama and Vice President Biden (remember them?) sharply criticized current EHRs for lack of interoperability. Other articles during the year showed that the political leaders were on target, as interoperability–an odd health care term for what other industries call “data exchange”–continues to be just as elusive as ever. Only 30% of hospitals were able to exchange data (although the situation has probably improved since the 2015 data used in the study). Advances in interoperability were called “theoretical” and the problem was placed into larger issues of poor communication. The Harvard Business Review weighed in too, chiding doctors for spending so much money on systems that don’t communicate.

The controversy sharpened as fraud charges were brought against a major EHR vendor for gaming the certification for Meaningful Use. A couple months later, strangely, the ONC weakened its certification process and announced it would rely more on the vendors to police themselves.

A long article provided some historical background on the reasons for incompatibility among EHRS.

Patients, as always, are left out of the loop: an ONC report finds improvements but many remaining barriers to attempts by patients to obtain the medical records that are theirs by law. And should the manufacturers of medical devices share the data they collect with patients? One would think it an elementary right of patients, but guidance released this year by the FDA was remarkably timid, pointing out the benefits of sharing but leaving it as merely a recommendation and offering big loopholes.

The continued failure to exchange data–which frustrates all attempts to improve treatments and cut costs–has led to the question: do EHR vendors and clinicians deliberately introduce technical measures for “information blocking”? Many leading health IT experts say no. But a study found that explicit information blocking measures are real.

Failures in interoperability and patient engagement were cited in another paper.

And we can’t leave interoperability without acknowledging the hope provided by FHIR. A paper on the use of FHIR with the older Direct-based interoperability protocols was released.

We’ll make our way through the rest of year and look at some specific technologies in the next part of the article.

Evolving Message Systems Learn To Filter And Route Alerts For Health Care Providers

Posted on December 11, 2017 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.

Because health care is a collaborative endeavor, patients can suffer if caretakers don’t get timely notifications. At the same time, the caretakers suffer when they are overloaded with alerts. Threading one’s way through this minefield (“Communications are complicated,” Vocera CMIO, Dr. Benjamin Kanter told me) was the theme of November’s Healthcare Messaging Conference and Exhibition at the Harvard Medical School. Like HIMSS, the major conference in health IT, something of a disconnect existed here between the conference and the exhibition. The speakers in the sessions implicitly criticized what the vendors were offering, information overload being the basic accusation.

Conference speakers told story after story of well-meaning installations of messaging systems that almost literally assaulted the staff with dozens of messages an hour. Kenny Schiff of CareSight reported seeing boxes full of expensive devices stuffed into closets in many hospitals. Dr. Trey Dobson reported research suggesting that 85% of standard hospital alarms require no intervention at all. He speculated that messaging has similar wasteful effects. In his facility, the Southwestern Vermont Medical Center at Dartmouth, they determined which lab results need to be delivered to the physician immediately and which could wait. They greatly reduced the number of messages sent about labs, which in turn decreased delivery time for important messages from an average of 50 minutes to only 7 minutes. These stories show both the benefits and drawbacks of current messaging systems.

State of the science
We all remember the first generations of pagers. Modern messaging systems, as represented by the vendors at the Healthcare Messaging Exhibition, offer a much sleeker experience, including:

  • Knowledge about who is responsible for a patient. No longer should messages be delivered to the nurse who left his shift an hour ago. The technical mechanism for tracking the role played by each clinician is group membership, familiar from the world of security. All clinicians who share a responsibility–such as working on a particular ward or caring for a particular patient–are assigned to a group. The status of each clinician is updated as he or she logs into the system, so that the message is delivered to the doctor or nurse currently on duty. A clinician dealing with one urgent situation should also not be interrupted by messages about another situation.

  • Full tracking of a message throughout its lifetime. The system records not only when a message was sent, but whether and when it was read. A message that goes ignored after a certain period of time can be escalated to the next level, and be sent to more and more people until someone addresses it.

  • Flexibility in delivery medium: mobile device, pager, computer, WiFi link, cellular network.

  • Sophisticated auditing. If a hospital needs to prove that a message was read (or that it was never read), the logs have to support that. This is important for both quality control and responses to legal or regulatory actions.

  • Integration with electronic health record systems, which allows systems to include information about the patient in messages.

  • HIPAA compliance. This essentially requires just garden-variety modern encryption, but it’s disturbing to learn how many physicians are breaking the law and risking their patients’ confidentiality by resorting casually to non-compliant messaging services instead of the ones offered at this exhibtion, which are designed specifically for health care use.

  • Cloud services. Instead of keeping information on devices, which can lead to it becoming lost or unavailable, it is stored on the vendor’s servers. This allows more flexible delivery options.

Although some of these advances generate more informative and useful messages, none of them reduce the number of messages. In fact, they encourage a vast expansion of the number of messsages sent. But some companies do offer enhancements over the common traits just cited.

  • Vocera has been connecting health care staff for many years. The company formed the subject of my first article on health IT in 2003, and of course its technology has evolved tremendously since then. Their services extend beyond the hospital to the primary care physician, skilled nursing facilities, and patients themselves. Dr. Kanter told me that they conceive of their service not simply as messaging, but as a form of clinical decision support. Their acquisition of Extension Healthcare in 2016 allowed them to add a new dimension of intelligence to the generation of messages. For instance, the patient’s health record can be consulted to determine the degree of risk presented by an event such as getting out of bed: if the patient has a low risk of falling, only the patient’s nurse may be alerted. Location information can also be incorporated into the logic, so that for instance a nurse who is already in the patient’s room will not receive an alert for that patient. Vocera has a rules engine and works with hospitals to develop customized rules.

  • HipLink has a particularly broad range of both input and delivery devices. In addition to all the common devices used by clinicians, HipLink can convert text to voice to call a plain telephone with a message. CEO Pamela LaPine told me it also accepts input not only from medical sensors, but from sensors embedded in fire alarms, doors, and other common props of medical environments.

  • OnPage helps coordinate secure communications through the use of schedules, individual and group messaging, and message tracking. For instance, the end of an operation may generate a message to the nursing staff to prepare for the arrival of a post-op patient. A message to the cleaning staff might be generated in order to prepare a room. All the necessary messages are presented to a dispatcher on a console.

  • 1Call, which provides a suite of innovative and integrated scheduling and communication applications, includes prompts to call center staff, a service they call Intuitive Call Flow Navigation. For a given situation, the service can help the staff give the information needed at the right point in each call. The same logic applies to the automated processes carried out with 1Call’s integration engine and automated notification software, which can also consolidate messaging based on rules, be customized to each organization’s needs, and improve efficiency throughout the organization.

Michael Detjen, Chief Strategy Officer of Mobile Heartbeat, laid out the pressures on messaging companies to evolve and become more like other cutting-edge high-tech companies. As messaging become universal through a health care institution, workflows come to depend on it, and thus, patient lives depend on it too. Taking the system down for an upgrade–or even worse, having it fail–is not acceptable, even at 2:00 in the morning. Both delivery and successful logging must be guaranteed, both for quality purposes and for compliance. To achieve this kind of reliability, developers must adopt the advanced development techniques popular among the most savvy software companies, such as DevOps and continuous testing and integration.

Looking toward the future
In his presentation, Schiff described some of the physical and logistical requirements for messaging devices. Clinicians should be able to switch devices quickly in case one is lost. They should be able simply to run their ID card through a reader, pick up a new device, and have it recognize them along with their message history (which means storing the messages securely in the cloud). Login requirements should be minimized, and one-hand operation should be possible. Schiff also looks forware to biometric identification of users.

Shahid Shah pointed out that the burden current messaging places on caregivers amounts to a form of uncompensated care. If messages are sent just to reassure patients, doctors and nurses will treat them as annoyances to be avoided. However, if the messages improve productivity, staff will accept them. And if they improve patient outcomes, so much the better–as long as fee-for-value reimbursements allow the health care provider to profit from improved outcomes.

To introduce the intelligence that would make messaging beneficial, Shah suggests more workflow analysis and the automation of common responses. A number of questions regarding patients could be answered automatically by bots, leaving only the more difficult ones for human clinicians.

The message regarding messaging was fairly consistent at the Healthcare Messaging Conference. Messaging has only begun to reap the benefits it can provide, and requires more analytics, more workflow analysis, and more integration with health care sites to become a boon to health care staff. The topic was a rather narrow one for a two-day conference, perhaps the reason it did not attract a large audience in its first iteration. But perhaps the conference will help drive messaging to new levels of sophistication, and become true life-savers while reducing burdens on clinicians.

Healthcare messaging and communication is also one of the focuses of our conference Health IT Expo happening May 30-June 1, 2018 in New Orleans. If you’re in charge of your hospital messaging systems, join us in New Orleans for an in depth look at best practices, hacks, and strategies for hospital messaging and communication.

This article is also available in a Portuguese translation by homeyou.

Measuring the Vital Signs of Health Care Progress at the Connected Health Conference (Part 3 of 3)

Posted on November 17, 2017 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 previous segment of this article covered one of the crucial themes in health care today: simplifying technology’s interactions with individuals over health care. This segment finishes my coverage of this year’s Connected Health Conference with two more themes: improved data sharing and blockchains.

Keynote at Connected Health Conference

Keynote at Connected Health Conference

Improved data sharing
The third trend I’m pursuing is interoperability. If data collection is the oxygen that fuels connected health, data sharing is the trachea that brings it where it’s needed. Without interoperability, clinicians cannot aid patients in their homes, analysts cannot derive insights that inform treatments, and transitions to assisted living facilities or other environments will lead to poor care.

But the health care field is notoriously bad at data sharing. The usual explanation is that doctors want to make it hard for competitors to win away their patients. If that’s true, fee-for-value reimbursements will make them even more possessive. After all, under fee-for-value, clinicians are held accountable for patient outcomes over a long period of time. They won’t want to lose control of the patient. I first heard of this danger at a 2012 conference (described in the section titled “Low-hanging fruit signals a new path for cost savings”).

So the trade press routinely and ponderously reports that once again, years have gone by without much progress in data sharing. The US government recognizes that support for interoperability is unsatisfactory, and has recently changed the ONC certification program to focus on it.

Carla Kriwet, CEO of Connected Care and Health Informatics at Philips, was asked in her keynote Fireside Chat to rate the interoperability of health data on a scale from 0 to 10, and chose a measly 3. She declared that “we don’t believe in closed systems at all” and told me in an interview that Philips is committed to creating integrated solutions that work with any and all products. Although Philips devices are legendary in many domains, Kriwet wants customers to pay for outcomes, not devices.

For instance, Philips recently acquired the Wellcentive platform that allows better care in hospitals by adopting population health approaches that look at whole patient populations to find what works. The platform works with a wide range of input sources and is meant to understand patient populations, navigate care and activate patients. Philips also creates dashboards with output driven by artificial intelligence–the Philips IntelliVue Guardian solution with Early Warning Scoring (EWS)–that leverages predictive analytics to present critical information about patient deterioration to nurses and physicians. This lets them intervene quickly before an adverse event occurs, without the need for logging in repeatedly. (This is an example of another trend I cover in this article, the search for simpler interfaces.)

Kriwet also told me that Philips has incorporated the principles of agile programming throughout the company. Sprints of a few weeks develop their products, and “the boundary comes down” between R&D and the sales team.

I also met with Jon Michaeli, EVP of Strategic Partnerships with Medisafe, a company that I covered two years ago. Medisafe is one of a slew of companies that encourage medication adherence. Always intensely based on taking in data and engaging patients in a personalized way, Medisafe has upped the sophistication of their solution, partly by integrating with other technologies. One recent example is its Safety Net, provided by artificial intelligence platform Neura. For instance, if you normally cart your cell phone around with you, but it’s lying quiet from 10:00 PM until 6:00 AM, Safety Net may determine your reason for missing your bedtime dose at 11:00 PM was that you had already fallen asleep. If Safety Net sees recurring patterns of behavior, it will adjust reminder time automatically.

Medisafe also gives users the option of recording the medication adherence through sensors rather than responding to reminders. They can communicate over Bluetooth to a pill bottle cap (“iCap”) that replaces the standard medicine cap and lets the service know when you have opened the bottle. The iCap fits the vast majority of medicine bottles dispensed by U.S. pharmacies and costs only $20 ($40 for a pack of 2), so you can buy several and use them for as long as you’re taking your medicine.

On another level, Mivatek provides some of the low-level scaffolding to connected health by furnishing data from devices to systems developed by the company’s clients. Suppose, for instance, that a company is developing a system that responds to patients who fall. Mivatek can help them take input from a button on the patient’s phone, from a camera, from a fall detector, or anything else to which Mivatek can connect. The user can add a device to his system simply by taking a picture of the bar code with his phone.

Jorge Perdomo, Senior Vice President Corporate Strategy & Development at Mivatek, told me that these devices work with virtually all of the available protocols on the market that have been developed to promote interoperability. In supporting WiFi, Mivatek loads an agent into its system to provide an additional level of security. This prevents device hacking and creates an easy-to-install experience with no setup requirements.

Blockchains
Most famous as a key technological innovation supporting BitCoin, blockchains have a broad application as data stores that record transactions securely. They can be used in health care for granting permissions to data and other contractual matters. The enticement offered by this technology is that no central institution controls or stores the blockchain. One can distribute the responsibility for storage and avoid ceding control to one institution.

Blockchains do, however, suffer from inherent scaling problems by design: they grow linearly as people add transactions, the additions must be done synchronously, and the whole chain must be stored in its entirety. But for a limited set of participants and relatively rate updates (for instance, recording just the granting of permissions to data and not each chunk of data exchanged), the technology holds great promise.

Although I see a limited role for blockchains, the conference gave considerable bandwidth to the concept. In a keynote that was devoted to blockchains, Dr. Samir Damani described how one of his companies, MintHealth, planned to use them to give individuals control over health data that is currently held by clinicians or researchers–and withheld from the individuals themselves.

I have previously covered the importance patient health records, and the open source project spotlighted by that article, HIE of One, now intends to use blockchain in a manner similar to MintHealth. In both projects, the patient owns his own data. MintHealth adds the innovation of offering rewards for patients who share their data with researchers, all delivered through the blockchain. The reward system is quite intriguing, because it would create for the first time a real market for highly valuable patient data, and thus lead to more research use along with fair compensation for the patients. MintHealth’s reward system also fits the connected health vision of promoting healthy behavior on a daily basis, to reduce chronic illness and health care costs.

Conclusion
Although progress toward connected health comes in fits and starts, the Connected Health Conference is still a bright spot in health care each year. For the first time this year, Partners’ Center for Connected Health partnered with another organization, the Personal Connected Health Alliance, and the combination seems to be a positive one. Certain changes were noticeable: for instance, all the breakout sessions were panels, and the keynotes were punctuated by annoying ads. An interesting focus this year was wellness in aging, the topic of the final panel. One surprising difference was the absence of the patient advocates from the Society for Participatory Medicine whom I’m used to meeting each year at this conference, perhaps because they held their own conference the day before.

The Center for Connected Health’s Joseph Kvedar still ran the program team, and the themes were familiar from previous years. This conference has become my touchstone for understanding health IT, and it will continue to be the place to go to track the progress of health care reform from a technological standpoint.

Measuring the Vital Signs of Health Care Progress at the Connected Health Conference (Part 2 of 3)

Posted on November 15, 2017 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 segment of this article introduced the themes of the Connected Health Conference and talked about the importance of validating what new technologies do using trials or studies like traditional medical advances. This segment continues my investigation into another major theme in health care: advanced interfaces.

Speaker from Validic at Connected Health Conference

Speaker from Validic at Connected Health Conference

Advanced interfaces
The compulsory picture of health care we’re accustomed to seeing, whenever we view hospital propaganda or marketing from health care companies, shows a patient in an awkward gown seated on an uncomfortable examination table. A doctor faces him or her full on–not a computer screen in site–exuding concern, wisdom, friendliness, and professionalism.

More and more, however, health sites are replacing this canonical photograph with one of a mobile phone screen speckled with indicators of our vital signs or thumbnail shot of our caregivers. The promise being conveyed is no longer care from a trusted clinician in the office, but instant access to all our information through a medium familiar to almost everyone everywhere–the personal mobile device.

But even touchscreen access to the world of the cloud is beginning to seem fusty. Typing in everything you eat with your thumbs, or even answering daily surveys about your mental state, gets old fast. As Dr. Yechiel Engelhard of TEVA said in his keynote, patients don’t want to put a lot of time into managing their illnesses, nor do doctors want to change their workflows. So I’m fascinated with connected health solutions that take the friction out of data collection and transmission.

One clear trend is the move to voice–or rather, I should say back to voice, because it is the original form of human communication for precise data. The popularity of Amazon Echo, along with Siri and similar interfaces, shows that this technology will hit a fever pitch soon. One research firm found that voice-triggered devices more than doubled in popularity between 2015 and 2016, and that more than half of Americans would like such a device in the home.

I recently covered a health care challenge using Amazon Alexa that demonstrates how the technology can power connected health solutions. Most of the finalists in the challenge were doing the things that the Connected Health Conference talks about incessantly: easy and frequent interactions with patients, analytics to uncover health problems, integration with health care providers, personalization, and so on.

Orbita is another company capitalizing on voice interfaces to deliver a range of connected health solutions, from simple medication reminders to complete care management applications for diabetes. I talked to CEO Bill Rogers, who explained that they provide a platform for integrating with AI engines provided by other services to carry out communication with individuals through whatever technology they have available. Thus, Orbita can talk through Echo, send SMS messages, interact with a fitness device or smart scale, or even deliver a reminder over a plain telephone interface.

One client of Orbita uses it platform to run a voice bot that talks to patients during their discharge process. The bot provides post-discharge care instructions and answers patients’ questions about things like pain management and surgery wound care. The results show that patients are more willing to ask questions of the bot than of a discharge nurse, perhaps because they’re not afraid of wasting someone’s time. Rogers also said services are improving their affective interfaces, which respond to the emotional tone of the patient.

Another trick to avoid complex interfaces is to gather as much data as possible from the patient’s behavior (with her consent, of course) to eliminate totally the need for her to manually enter data, or even press a button. Devices are getting closer to this kind of context-awareness. Following are some of the advances I enjoyed seeing at the Connected Health Conference.

  • PulseOn puts more health data collection into a wrist device than I’ve ever seen. Among the usual applications to fitness, they claim to detect atrial fibrillation and sleep apnea by shining a light on the user’s skin and measuring changes in reflections caused by variations in blood flow.
  • A finger-sized device called Gocap, from Common Sensing, measures insulin use and reports it over wireless connections to clinical care-takers. The device is placed over the needle end of an insulin pen, determines how much was injected by measuring the amount of fluid dispensed after a dose, and transmits care activity to clinicians through a companion app on the user’s smartphone. Thus, without having to enter any information by hand, people with diabetes can keep the clinicians up to date on their treatment.
  • One of the cleverest devices I saw was a comprehensive examination tool from Tyto Care. A small kit can carry the elements of a home health care exam, all focused on a cute little sphere that fits easily in the palm. Jeff Cutler, Chief Revenue Officer, showed me a simple check on the heart, ear, and throat that anyone can perform. You can do it with a doctor on the other end of a video connection, or save the data and send it to a doctor for later evaluation.

    Tyto Care has a home version that is currently being used and distributed by partners such as Heath Systems, providers, payers and employers, but will ultimately be available for sale to consumers for $299. They also offer a professional and remote clinic version that’s tailor-made for a school or assisted living facility.

A new Digital Therapeutics Alliance was announced just before the conference, hoping to promote more effective medical devices and allow solutions to scale up through such things as improving standards and regulations. Among other things, the alliance will encourage clinical trials, which I have already highlighted as critical.

Big advances were also announced by Validic, which I covered last year. Formerly a connectivity solution that unraveled the varying quasi-standard or non-standard protocols of different devices in order to take their data into electronic health records, Validic has created a new streaming API that allows much faster data transfers, at a much higher volume. On top of this platform they have built a notification service called Inform, which takes them from a networking solution to a part of the clinicians’ workflow.

Considerable new infrastructure is required to provide such services. For instance, like many medication adherence services, Validic can recognize when time has gone by without a patient reporting that’s he’s taken his pill. This level of monitoring requires storing large amounts of longitudinal data–and in fact, Validic is storing all transactions carried out over its platform. The value of such a large data set for discovering future health care solutions through analytics can make data scientists salivate.

The next segment of this article wraps up coverage of the conference with two more themes.

Measuring the Vital Signs of Health Care Progress at the Connected Health Conference (Part 1 of 3)

Posted on November 13, 2017 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.

Attendees at each Connected Health Conference know by now the architecture of health reform promoted there. The term “connected health” has been associated with a sophisticated amalgam of detailed wellness plans, modern sensors, continuous data collection in the field, patient control over data, frequent alerts and reminders, and analytics to create a learning health care system. The mix remains the same each year, so I go each time to seek out progress toward the collective goal. This year, I’ve been researching what’s happening in these areas:

  • Validation through clinical trials
  • Advanced interfaces to make user interaction easier
  • Improved data sharing (interoperability)
  • Blockchains

Panel at Connected Health Conference

Panel at Connected Health Conference

There were a few other trends of interest, which I’ll mention briefly here. Virtual reality (VR) and augmented reality (AR) turned up at some exhibitor booths and were the topic of a panel. Some of these technologies run on generic digital devices–such as the obsession-inducing Pokémon GO game–while others require special goggles such as the Oculus Rift (the first VR technology to show a promise for widespread adoption, and now acquired by Facebook) or Microsoft’s HoloLens. VR shuts out the user’s surroundings and presents her with a 360-degree fantasy world, whereas AR imposes information or images on the surroundings. Both VR and AR are useful for teaching, such as showing an organ in 3D organ in front of a medical student on a HoloLens, and rotating it or splitting it apart to show details.

I haven’t yet mentioned the popular buzzword “telehealth,” because it’s subsumed under the larger goal of connected health. I do use the term “artificial intelligence,” certainly a phrase that has gotten thrown around too much, and whose meaning is subject of much dissension. Everybody wants to claim the use of artificial intelligence, just as a few years ago everybody talked about “the cloud.” At the conference, a panel of three experts took up the topic and gave three different definitions of the term. Rather than try to identify the exact algorithms used by each product in this article and parse out whether they constitute “real” artificial intelligence, I go ahead and use the term as my interviewees use it.

Exhibition hall at Connected Health Conference

Exhibition hall at Connected Health Conference

Let’s look now at my main research topics.

Validation through clinical trials
Health apps and consumer devices can be marketed like vitamin pills, on vague impressions that they’re virtuous and that doing something is better than doing nothing. But if you want to hook into the movement for wellness–connected health–you need to prove your value to the whole ecosystem of clinicians and caretakers. The consumer market just doesn’t work for serious health care solutions. Expecting an individual to pay for a service or product would limit you to those who can afford it out-of-pocket, and who are concerned enough about wellness to drag out their wallets.

So a successful business model involves broaching the gates of Mordor and persuading insurers or clinicians to recommend your solution. And these institutions won’t budge until you have trials or studies showing that you actually make a difference–and that you won’t hurt anybody.

A few savvy app and device developers build in such studies early in their existence. For instance, last year I covered a typical connected health solution called Twine Health, detailing their successful diabetes and hypertension trials. Twine Health combines the key elements that one finds all over the Connected Health Conference: a care plan, patient tracking, data analysis, and regular check-ins. Their business model is to work with employer-owned health plans, and to expand to clinicians as they gradually migrate to fee-for-value reimbursement.

I sense that awareness is growing among app and device developers that the way to open doors in health care is to test their solutions rigorously and objectively. But I haven’t found many who do so yet.

In the next segment of this article continues my exploration of the key themes I identified at the start of this article.

Alexa Can Truly Give Patients a Voice in Their Health Care (Part 3 of 3)

Posted on October 20, 2017 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.

Earlier parts of this article set the stage for understanding what the Alexa Diabetes Challenge is trying to achieve and how some finalists interpreted the mandate. We examine three more finalists in this final section.

DiaBetty from the University of Illinois-Chicago

DiaBetty focuses on a single, important aspect of diabetes: the effect of depression on the course of the disease. This project, developed by the Department of Psychiatry at the University of Illinois-Chicago, does many of the things that other finalists in this article do–accepting data from EHRs, dialoguing with the individual, presenting educational materials on nutrition and medication, etc.–but with the emphasis on inquiring about mood and handling the impact that depression-like symptoms can have on behavior that affects Type 2 diabetes.

Olu Ajilore, Associate Professor and co-director of the CoNECt lab, told me that his department benefited greatly from close collaboration with bioengineering and computer science colleagues who, before DiaBetty, worked on another project that linked computing with clinical needs. Although they used some built-in capabilities of the Alexa, they may move to Lex or another AI platform and build a stand-alone device. Their next step is to develop reliable clinical trials, checking the effect of DiaBetty on health outcomes such as medication compliance, visits, and blood sugar levels, as well as cost reductions.

T2D2 from Columbia University

Just as DiaBetty explores the impact of mood on diabetes, T2D2 (which stands for “Taming Type 2 Diabetes, Together”) focuses on nutrition. Far more than sugar intake is involved in the health of people with diabetes. Elliot Mitchell, a PhD student who led the T2D2 team under Assistant Professor Lena Mamykina in the Department of Biomedical Informatics, told me that the balance of macronutrients (carbohydrates, fat, and protein) is important.

T2D2 is currently a prototype, developed as a combination of Alexa Skill and a chatbot based on Lex. The Alexa Skills Kit handle voice interactions. Both the Skill and the chatbot communicate with a back end that handles accounts and logic. Although related Columbia University technology in diabetes self-management is used, both the NLP and the voice interface were developed specifically for the Alexa Diabetes Challenge. The T2D2 team included people from the disciplines of computer interaction, data science, nursing, and behavioral nutrition.

The user invokes Alexa to tell it blood sugar values and the contents of meals. T2D2, in response, offers recipe recommendations and other advice. Like many of the finalists in this article, it looks back at meals over time, sees how combinations of nutrients matched changes in blood sugar, and personalizes its food recommendations.

For each patient, before it gets to know that patient’s diet, T2D2 can make food recommendations based on what is popular in their ZIP code. It can change these as it watches the patient’s choices and records comments to recommendations (for instance, “I don’t like that food”).

Data is also anonymized and aggregated for both recommendations and future research.

The care team and family caregivers are also involved, although less intensely than some other finalists do. The patient can offer caregivers a one-page report listing a plot of blood sugar by time and day for the previous two weeks, along with goals and progress made, and questions. The patient can also connect her account and share key medical information with family and friends, a feature called the Supportive Network.

The team’s next phase is run studies to evaluable some of assumptions they made when developing T2D2, and improve it for eventual release into the field.

Sugarpod from Wellpepper

I’ll finish this article with the winner of the challenge, already covered by an earlier article. Since the publication of the article, according to the founder and CEO of Wellpepper, Anne Weiler, the company has integrated some of Sugarpod functions into a bathroom scale. When a person stands on the scale, it takes an image of their feet and uploads it to sites that both the individual and their doctor can view. A machine learning image classifier can check the photo for problems such as diabetic foot ulcers. The scale interface can also ask the patient for quick information such as whether they took their medication and what their blood sugar is. Extended conversations are avoided, under the assumption that people don’t want to have them in the bathroom. The company designed its experiences to be integrated throughout the person’s day: stepping on the scale and answering a few questions in the morning, interacting with the care plan on a mobile device at work, and checking notifications and messages with an Echo device in the evening.

Any machine that takes pictures can arouse worry when installed in a bathroom. While taking the challenge and talking to people with diabetes, Wellpepper learned to add a light that goes on when the camera is taking a picture.

This kind of responsiveness to patient representatives in the field will determine the success of each of the finalists in this challenge. They all strive for behavioral change through connected health, and this strategy is completely reliant on engagement, trust, and collaboration by the person with a chronic illness.

The potential of engagement through voice is just beginning to be tapped. There is evidence, for instance, that serious illnesses can be diagnosed by analyzing voice patterns. As we come up on the annual Connected Health Conference this month, I will be interested to see how many participating developers share the common themes that turned up during the Alexa Diabetes Challenge.

Alexa Can Truly Give Patients a Voice in Their Health Care (Part 2 of 3)

Posted on October 19, 2017 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 introduced the problems of computer interfaces in health care and mentioned some current uses for natural language processing (NLP) for apps aimed at clinicians. I also summarized the common goals, problems, and solutions I found among the five finalists in the Alexa Diabetes Challenge. This part of the article shows the particular twist given by each finalist.

My GluCoach from HCL America in Partnership With Ayogo

There are two levels from which to view My GluCoach. On one level, it’s an interactive tool exemplifying one of the goals I listed earlier–intense engagement with patients over daily behavior–as well as the theme of comprehensivenesss. The interactions that My GluCoach offers were divided into three types by Abhishek Shankar, a Vice President at HCL Technologies America:

  • Teacher: the service can answer questions about diabetes and pull up stored educational materials

  • Coach: the service can track behavior by interacting with devices and prompt the patient to eat differently or go out for exercise. In addition to asking questions, a patient can set up Alexa to deliver alarms at particular times, a feature My GluCoach uses to deliver advice.

  • Assistant: provide conveniences to the patient, such as ordering a cab to take her to an appointment.

On a higher level, My GluCoach fits into broader services offered to health care institutions by HCL Technologies as part of a population health program. In creating the service HCL partnered with Ayogo, which develops a mobile platform for patient engagement and tracking. HCL has also designed the service as a general health care platform that can be expanded over the next six to twelve months to cover medical conditions besides diabetes.

Another theme I discussed earlier, interactions with outside data and the use of machine learning, are key to my GluCoach. For its demo at the challenge, My GluCoach took data about exercise from a Fitbit. It can potentially work with any device that shares information, and HCL plans to integrate the service with common EHRs. As My GluCoach gets to know the individual who uses it over months and years, it can tailor its responses more and more intelligently to the learning style and personality of the patient.

Patterns of eating, medical compliance, and other data are not the only input to machine learning. Shankar pointed out that different patients require different types of interventions. Some simply want to be given concrete advice and told what to do. Others want to be presented with information and then make their own decisions. My GluCoach will hopefully adapt to whatever style works best for the particular individual. This affective response–together with a general tone of humor and friendliness–will win the trust of the individual.

PIA from Ejenta

PIA, which stands for “personal intelligent agent,” manages care plans, delivering information to the affected patients as well as their care teams and concerned relatives. It collects medical data and draws conclusions that allow it to generate alerts if something seems wrong. Patients can also ask PIA how they are doing, and the agent will respond with personalized feedback and advice based on what the agent has learned about them and their care plan.

I talked to Rachna Dhamija, who worked on a team that developed PIA as the founder and CEO of Ejenta. (The name Ejenta is a version of the word “agent” that entered the Bengali language as slang.) She said that the AI technology had been licensed from NASA, which had developed it to monitor astronauts’ health and other aspects of flights. Ejenta helped turn it into a care coordination tool with interfaces for the web and mobile devices at a major HMO to treat patients with chronic heart failure and high-risk pregnancies. Ejenta expanded their platform to include an Alexa interface for the diabetes challenge.

As a care management tool, PIA records targets such as glucose levels, goals, medication plans, nutrition plans, and action parameters such as how often to take measurements using the devices. Each caregiver, along the patient, has his or her own agent, and caregivers can monitor multiple patients. The patient has very granular control over sharing, telling PIA which kind of data can be sent to each caretaker. Access rights must be set on the web or a mobile device, because allowing Alexa to be used for that purpose might let someone trick the system into thinking he was the patient.

Besides Alexa, PIA takes data from devices (scales, blood glucose monitors, blood pressure monitors, etc.) and from EHRs in a HIPAA-compliant method. Because the service cannot wake up Alexa, it currently delivers notifications, alerts, and reminders by sending a secure message to the provider’s agent. The provider can then contact the patient by email or mobile phone. The team plans to integrate PIA with an Alexa notifications feature in the future, so that PIA can proactively communicate with the patient via Alexa.

PIA goes beyond the standard rules for alerts, allowing alerts and reminders to be customized based on what it learns about the patient. PIA uses machine learning to discover what is normal activity (such as weight fluctuations) for each patient and to make predictions based on the data, which can be shared with the care team.

The final section of this article covers DiaBetty, T2D2, and Sugarpod, the remaining finalists.

Alexa Can Truly Give Patients a Voice in Their Health Care (Part 1 of 3)

Posted on October 16, 2017 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 leading pharmaceutical and medical company Merck, together with Amazon Web Services, has recently been exploring the potential health impacts of voice interfaces and natural language processing (NLP) through an Alexa Diabetes Challenge. I recently talked to the five finalists in this challenge. This article explores the potential of new interfaces to transform the handling of chronic disease, and what the challenge reveals about currently available technology.

Alexa, of course, is the ground-breaking system that brings everyday voice interaction with computers into the home. Most of its uses are trivial (you can ask about today’s weather or change channels on your TV), but one must not underestimate the immense power of combining artificial intelligence with speech, one of the most basic and essential human activities. The potential of this interface for disabled or disoriented people is particularly intriguing.

The diabetes challenge is a nice focal point for exploring the more serious contribution made by voice interfaces and NLP. Because of the alarming global spread of this illness, the challenge also presents immediate opportunities that I hope the participants succeed in productizing and releasing into the field. Using the challenge’s published criteria, the judges today announced Sugarpod from Wellpepper as the winner.

This article will list some common themes among the five finalists, look at the background about current EHR interfaces and NLP, and say a bit about the unique achievement of each finalist.

Common themes

Overlapping visions of goals, problems, and solutions appeared among the finalists I interviewed for the diabetes challenge:

  • A voice interface allows more frequent and easier interactions with at-risk individuals who have chronic conditions, potentially achieving the behavioral health goal of helping a person make the right health decisions on a daily or even hourly basis.

  • Contestants seek to integrate many levels of patient intervention into their tools: responding to questions, collecting vital signs and behavioral data, issuing alerts, providing recommendations, delivering educational background material, and so on.

  • Services in this challenge go far beyond interactions between Alexa and the individual. The systems commonly anonymize and aggregate data in order to perform analytics that they hope will improve the service and provide valuable public health information to health care providers. They also facilitate communication of crucial health data between the individual and her care team.

  • Given the use of data and AI, customization is a big part of the tools. They are expected to determine the unique characteristics of each patient’s disease and behavior, and adapt their advice to the individual.

  • In addition to Alexa’s built-in language recognition capabilities, Amazon provides the Lex service for sophisticated text processing. Some contestants used Lex, while others drew on other research they had done building their own natural language processing engines.

  • Alexa never initiates a dialog, merely responding when the user wakes it up. The device can present a visual or audio notification when new material is present, but it still depends on the user to request the content. Thus, contestants are using other channels to deliver reminders and alerts such as messaging on the individual’s cell phone or alerting a provider.

  • Alexa is not HIPAA-compliant, but may achieve compliance in the future. This would help health services turn their voice interfaces into viable products and enter the mainstream.

Some background on interfaces and NLP

The poor state of current computing interfaces in the medical field is no secret–in fact, it is one of the loudest and most insistent complaints by doctors, such as on sites like KevinMD. You can visit Healthcare IT News or JAMA regularly and read the damning indictments.

Several factors can be blamed for this situation, including unsophisticated electronic health records (EHRs) and arbitrary reporting requirements by Centers for Medicare & Medicaid Services (CMS). Natural language processing may provide one of the technical solutions to these problems. The NLP services by Nuance are already famous. An encouraging study finds substantial time savings through using NLP to enter doctor’s insights. And on the other end–where doctors are searching the notes they previously entered for information–a service called Butter.ai uses NLP for intelligent searches. Unsurprisingly, the American Health Information Management Association (AHIMA) looks forward to the contributions of NLP.

Some app developers are now exploring voice interfaces and NLP on the patient side. I covered two such companies, including the one that ultimately won the Alexa Diabetes Challenge, in another article. In general, developers using these interfaces hope to eliminate the fuss and abstraction in health apps that frustrate many consumers, thereby reaching new populations and interacting with them more frequently, with deeper relationships.

The next two parts of this article turn to each of the five finalists, to show the use they are making of Alexa.