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 New Meaning for Connected Health at 2016 Symposium (Part 4 of 4)

Posted on November 8, 2016 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 section of this article continued our exploration of the integration of health care into daily life. This section wraps up the article with related insights, including some thoughts about the future.

Memorable moments
I had the chance to meet with Casper de Clercq, who has set up a venture capital plan devoted to health as a General Partner at Norwest Venture Partners. He recommends that manufacturers and clinicians give patients a device that collects data while doing something else they find useful, so that they are motivated to keep wearing it. As an example, he cited the Beddit sleep tracker, which works through sensors embedded (no pun intended) in the user’s bed.

He has found that successful companies pursue gradual, incremental steps toward automated programs. It is important to start with a manual process that works (such as phoning or texting patients from the provider), then move to semi-automation and finally, if feasible, full automation. The product must also be field-tested; one cannot depend on a pilot. This advice matches what Glen Tullman, CEO of Livongo Health, said in his keynote: instead of doing a pilot, try something out in the field and change quickly if it doesn’t work.

Despite his call for gradual change, de Clercq advises that companies show an ROI within one year–otherwise, the field of health care may have evolved and the solution may be irrelevant.

He also recommends a human component in any health program. The chief barrier to success is getting the individual to go along with both the initial activation and continuing motivation. Gamification, behavioral economics, and social connections can all enhance this participation.

A dazzling keynote on videogames for health was delivered by Adam Gazzaley, who runs Neuroscience labs at the University of California at San Francisco. He pointed out that conventional treatments get feedback on patient reactions far too slowly–sometimes months after the reaction has occurred. In the field of mental health, His goal is to supplement (not replace) medications with videogames, and to provide instant feedback to game players and their treatment staff alike. Videogames not only provide a closed-loop system (meaning that feedback is instantaneous), but also engage patients by being fun and offering real-time rewards. Attention spans, anxiety, and memory are among the issues he expects games to improve. Education and wellness are also on his game plan. This is certainly one talk where I did not multitask (which is correlated with reduced performance)!

A future, hopefully bigger symposium
The Connected Health symposium has always been a production of the Boston Partners Health Care conglomerate, a part of their Connected Health division. The leader of the division, Dr. Joseph Kvedar, introduced the symposium by expressing satisfaction that so many companies and organizations are taking various steps to make connected health a reality, then labeled three areas where leadership is still required:

  • Reassuring patients that the technologies and practices work for them. Most people will be willing to adopt these practices when urged by their doctors. But their privacy must be protected. This requires low-cost solutions to the well-known security problems in EHRs and devices–the latter being part of the Internet of Things, whose vulnerability was exposed by the recent attack on Dyn and other major Internet sites.

  • Relieving the pressures on clinicians. Kvedar reported that 45 percent of providers would like to adopt connected health practices, but only 12 percent do so. One of the major concerns holding them back is the possibility of data overload, along with liability for some indicator of ill health that they miss in the flood of updates. Partners Connected Health will soon launch a provider adoption initiative that deals with their concerns.

  • Scaling. Pilot projects in connected health invest a lot of researcher time and offers a lot of incentives to develop engagement among their subjects. Because engagement is the whole goal of connected health, the pilot may succeed but prove hard to turn into a widespread practice. Another barrier to scaling is consumers’ lack of tolerance for the smallest glitches or barriers to adoption. Providers, also, insist that new practices fit their established workflows.

Dr. Kvedar announced at this symposium that they would be doing future symposia in conjunction with the Personal Connected Health Alliance (Formerly the mHealth Summit owned by HIMSS), a collaboration that makes sense. Large as Partners Health Care is, the symposium reaches much farther into the health care industry. The collaboration should bring more resources and more attendees, establishing the ideals of connected health as a national and even international movement.

A New Meaning for Connected Health at 2016 Symposium (Part 3 of 4)

Posted on November 7, 2016 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 section of this article paused during a discussion of the accuracy and uses of devices. At a panel on patient generated data, a speaker said that one factor holding back the use of patient data was the lack of sophistication in EHRs. They must be enhanced to preserve the provenance of data: whether it came from a device or from a manual record by the patient, and whether the device was consumer-grade or a well-tested medical device. Doctors invest different levels of trust in different methods of collecting data: devices can provide more objective information than other ways of asking patients for data. A participant in the panel also pointed out that devices are more reliable in the lab than under real-world conditions. Consumers must be educated about the proper use of devices, such as whether to sit down and how to hold their arms when taking their blood pressure.

Costantini decried the continuing silos in both data sharing and health care delivery. She said only half of doctors share patient data with other doctors or caretakers. She also praised the recent collaboration between Philips and Qualcomm to make it easier for device data to get into medical records. Other organizations that have been addressing that issue for some time include Open mHealth, which I reviewed in an earlier article, and Validic.

Oozing into workflow
The biggest complaint I hear from clinicians about EHRs–aside from the time wasted in their use, which may be a symptom of the bigger problem-is that the EHRs disrupt workflow. Just as connected health must integrate with patient lives as seamlessly as possible, it should recognize how teams work and provide them with reasonable workflows. This includes not only entering existing workflows as naturally as capillary action, but helping providers adopt better ones.

The Veterans Administration is forging into this area with a new interface called the Enterprise Health Management Platform (eHMP). I mentioned it in a recent article on the future of the VA’s EHR. A data integration and display tool, eHMP is agnostic as to data source. It can be used to extend the VistA EHR (or potentially replace it) with other offerings. Although eHMP currently displays a modern dashboard format, as described in a video demo by Shane Mcnamee, the tool aims to be much more than that. It incorporates Business Process Modeling Notation (BPMN) and the WS-Human Task Specification to provide workflow support. The Activity Management Service in eHMP puts Clinical Best Practices directly into the workflow of health care providers.

Clinicians can use eHMP to determine where a consultation request goes; currently, the system is based on Red Hat’s BPMN engine. If one physician asks another to examine the patient, that task turns up on the receiving physician’s dashboard. Teams as well as individuals can be alerted to a patient need, and alerts can be marked as routine or urgent. The alerts can also be associated with time-outs, so that their importance is elevated if no one acts on them in the chosen amount of time.

eHMP is just in the beginning stages of workflow support. Developers are figuring out how to increase the sophistication of alerts, so that they offer a higher signal-to-noise ratio than most hospital CDS systems, and add intelligence to choose the best person to whom an alert should be directed. These improvements will hopefully free up time in the doctor’s session to discuss care in depth–what both patients and providers have long said they most want from the health care field.

At the Connected Health symposium, I found companies working on workflow as well. Dataiku (whose name is derived from “haiku”) has been offering data integration and analytics in several industries for the past three years. Workflows, including conditional branches and loops, can be defined through a graphical interface. Thus, a record may trigger a conditional inquiry: does a lab value exceed normal limits? if not, it is merely recorded, but if so, someone can be alerted to follow up.

Dataiku illustrates an all-in-one, comprehensive approach to analytics that remains open to extensions and integration with other systems. On the one hand, it covers the steps of receiving and processing data pretty well.

To clean incoming data (the biggest task on most data projects), their DSS system can use filters and even cluster data to find patterns. For instance, if 100 items list “Ohio” for their location, and one lists “Oiho”, the system can determine that the outlier is a probably misspelling. The system can also assign data to belonging to broad categories (string or integer) as well as more narrowly defined categories (such as social security number or ZIP code).

For analysis, Dataiku offers generic algorithms that are in wide use, such as linear regressions, and a variety of advanced machine learning (artificial intelligence) algorithms in the visual backend of the program–so the users don’t need to write a single line of code. Advanced users can also add their own algorithms coded in a variety of popular languages such as Python, R, and SQL. The software platform offers options for less technically knowledgeable users, pre-packaged solutions for various industries such as health care, security features such as audits, and artificial intelligence to propose an algorithm that works on the particular input data.

Orbita Health handles workflows between patients and providers to help with such issues as pain management and medication adherence. The company addresses ease of use by supporting voice-activated devices such as Amazon Echo, as well as some 250 other devices. Thus, a patient can send a message to a provider through a single statement to a voice-activated device or over another Internet-connected device. For workflow management, the provider can load a care plan into the system, and use Orbita’s orchestration engine (similar to the Business Process Modeling Notation mentioned earlier) to set up activities, such as sending a response to a patient’s device or comparing a measurement to the patient’s other measurements over time. Orbita’s system supports conditional actions, nests, and trees.

CitiusTech, founded in 2005, integrates data from patient devices and apps into provider’s data, allowing enterprise tools and data to be used in designing communications and behavioral management in the patient’s everyday life. The company’s Integrated Analytix platform offer more than 100,000 apps and devices from third-party developers. Industry studies have shown effective use of devices, with one study showing a 40% reduction in emergency room admissions among congestive heart failure patients through the use of scales, engaging the patients in following health protocols at home.

In a panel on behavior change and the psychology of motivation, participants pointed out that long-range change requires multiple, complex incentives. At the start, the patient may be motivated by a zeal to regain lost functioning, or even by extrinsic rewards such as lower insurance premiums. But eventually the patient needs to enfold the exercise program or other practice into his life as a natural activity. Rewards can include things like having a beer at the end of a run, or sharing daily activities with friends on social media.

In his keynote on behavioral medicine, the Co-founder & CEO of Omada Health, Sean Duffy, put up a stunningly complex chart showing the incentives, social connections, and other factors that go into the public’s adoption of health practices. At a panel called “Preserving the Human Touch in the Expanding World of Digital Therapies”, a speaker also gave the plausible advice that we tell patients what we can give back to them when collecting data.

The next section of this article offers some memorable statements at the conference, and a look toward the symposium’s future.

A New Meaning for Connected Health at 2016 Symposium (Part 2 of 4)

Posted on November 4, 2016 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 section of this article talked about making health a routine part of everyday life, particularly where consumer devices are concerned. We’ll continue in this section with other considerations aired at the symposium.

Tullman’s principles of simplicity, cited in the previous section, can be applied to a wide range of health IT. For instance, AdhereTech pill bottles can notify the patient with a phone call or text message if she misses a dose. Another example of a technology that is easily integrated into everyday life is a thermometer built into a vaginal ring that a woman can insert and use without special activation. This device was mentioned by Costantini during her keynote. The device can alert a woman–and, if she wants, her partner–to when she is most fertile.

Super-compact devices and fancy interfaces are not always necessary for a useful intervention. In a keynote, John Dwyer, Jr., President of the Global Alzheimer’s Platform Foundation, discussed a simple survey that his organization got large numbers of people to take. They uncovered a lot of undiagnosed cases of mental decline. I imagine that the people who chose to take the survey were experiencing possible symptoms and therefore were concerned about their mental abilities. Yet they apparently had not expressed concerns to their doctors; instead they responded to the online suggestion to take a survey.

Most of us spend a large chunk of our day at work, so wellness programs there are theoretically promising. A panel on workplace-connected health solutions talked about some of the barriers:

  • Inadequate communications. Employees need to be informed regularly that a program is available, and its benefits

  • Privacy guarantees. Employees must feel assured of a firewall between their employer and the organization handling their sensitive data.

  • Clear goals. A wellness program is not just a check-off box. Employers must know what they want to achieve and design programs around these goals.

I would add that employers should examine their own environment honestly before setting up a wellness program. It’s pretty hypocritical to offer a wellness program on the one hand while subjecting employees to stress, overwork, and bad ergonomics on the other.

Telehealth is also likely to grow, and in fact, 200 bills to improve regulation of telehealth are pending in Congress. A speaker at a panel on preserving the human touch said that the Centers for Medicare & Medicaid Services are held back by uncertainty about how to measure telehealth’s value. Another speaker pointed out that we have a severe shortage of mental health professionals, and that many areas lack access to them. Telehealth may improve access.

It all comes down to the environment
Health care has to fully acknowledge the role of environmental factors in creating sickness. These include the marketing of fatty and sugary foods, the trapping of poor and minority people in areas with air and water pollution, the barriers to getting health care (sick leave, geography, insurance gaps, ignorance of gender issues, and so forth), the government subsidization of gambling, and much more. Similar issues came up during a keynote by David Torchiana, President & CEO and Partners HealthCare.

In her keynote, Jo Ann Jenkins, the CEO of AARP, quoted Atul Gawande as saying that we have medicalized aging and are failing to support the elderly. We have to see them as functioning individuals and help to support their health instead of focusing on when things go wrong. This includes focusing on prevention and ensuring that they have access to professional health care while they are still well. It also means restructuring our living spaces and lifestyles so the elderly can remain safely in their homes, get regular exercise, and eat well.

These problems call for a massive legislative and regulatory effort. But as a participant said on the panel of disruptive women in health care, plenty of money goes into promoting the interests of large hospitals, insurers, and device manufacturers, but nobody knows how to actually lobby for health care. Look at the barriers reached by Michelle Obama’s Let’s Move campaign, which fell short of ambitious goals in improving American’s nutrition.

Grounding devices on a firm foundation
A repeated theme at this symposium was making data collection by patients easier–so easy in fact that they can just launch data collection and not think about it. To be sure, some people are comfortable with health technology: according to Costantini, 60 percent of US smartphone users manage their health in some way through those devices. Nevertheless, if people have to consciously choose when to send data–even a click of a button–many will drop out of the program.

At a break-out session during the 2015 Health Datapalooza, I heard prospective device makers express anxiety over the gargantuan task of getting their products accepted by the industry. The gold standard for health care adoption, of course, is FDA approval based on rigorous clinical trials. One participant in the Datapalooza workshop assured the others that he had gotten his device through the FDA process, and that they could to.

Attitudes seem to have shifted over the past year, and many more manufacturers are treating FDA approval as a natural step in their development process, keeping their eyes on the prize of clinical adoption. Keith Carlton, CEO of HUINNO, in a panel on wearables, said that accuracy is critical to stand out in the marketplace and to counter the confusion caused by manufacturers that substitute hype for good performance.

Clinical trials for devices don’t have to be the billion-dollar, drawn-out ordeals suffered by pharma companies. Devices are rarely responsible for side effects (except for implantables) and therefore can be approved after a few months of testing.

A representative of BewellConnect told me that their road to approval took 9-12 months, and involved comparing the results of their devices to those of robust medical devices that had been previously approved. Typical BewellConnect devices include blood pressure cuffs and an infrared thermometer that quickly shows the patient’s temperature after being held near his temple. This thermometer has been used around the world in situations where it’s important to avoid contact with patients, such as in Ebola-plagued regions.

What’s new over the past three years is Bluetooth-enabled devices that can transmit their results over the network. BewellConnect includes this networking capability in 17 current devices. The company tries to provide a supremely easy path for the patient to transmit the device over a phone app to the cloud. The patient can register multiple family members on the app, and is prompted twice to indicate who was using the device so as to prevent errors. BewellConnect is working on an alert system for providers, a simple use case for data collection.

Many products from BewellConnect are in widespread use in France, where the company is based, and they have launched a major entry into the US market. I asked BewellConnect’s CEO, Olivier Hua, whether the US market presents greater problems than France. He said that the two markets are more similar than we think.

Health care in the US has historically been fragmented, whereas in France it was unified under government control. But the Affordable Care Act in the US has brought more regulation to the market here, whereas private health care providers (combining insurance and treatment) have been growing in France. As of January 1 of this year, France has required all employers to include a private option in their health care offerings. For the first time, French individuals are being hit with the copays and deductibles familiar to Americans, and are weighing how often to go to the doctor. Although the US market is still more diverse, and burdened by continuing fee-for-service plans, it is comparable to the French market for a vendor such as BewellConnect.

The next section of this article will continue with a discussion of barriers in the use of patient data, and other insights from the Connected Health symposium.

A New Meaning for Connected Health at 2016 Symposium (Part 1 of 4)

Posted on November 3, 2016 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.

Those of us engaged in health care think constantly about health. But at the Connected Health symposium, one is reminded that the vast majority of people don’t think much about health at all. They’re thinking about child care, about jobs, about bills, about leisure time. Health comes into the picture only through its impacts on those things.

Certainly, some people who have suffered catastrophic traumas–severe accidents, cancer, or the plethora of unfortunate genetic conditions–become obsessed about health to the same extent as health professionals. These people become e-patients and do all the things they need to do regain the precious state of being they enjoyed before their illness, often clashing with the traditional medical establishment in pursuit of health.

But for most people with chronic conditions, the opposite holds true. A whimsical posting points out that we willingly pay more to go to a masseur or hairdresser than to a doctor. I appreciate this observation more than the remedies offered by the author, which fall into the usual “patient engagment” activities that I have denigrated in an earlier article.

Understanding health as a facet and determinant of everyday life becomes even more important as we try to reverse the rise of costs, which in many nations are threatening economic progress and even the social contract. (Witness the popular anger in the current US election over rising insurance premiums and restrictions on choice.) We have to provide health solutions to people who are currently asymptomatic. The conventional focus on diagnosed conditions won’t serve us.

It’s thus commendable that the Connected Health symposium for 2016 has evolved to the point where participants can think not only of reaching out to patients, but to embedding their interventions so deeply into patient life that the patient no longer has to think about her health to benefit. This gives a new meaning to the word “connected”. Whereas, up to now, it referred to connecting a patient more closely with their clinicians and care-takers (through data collection, messaging, and online consultations), “connected” can also mean connecting our healthful interventions to the patient’s quotidian concerns about work, family, and leisure.

We can do this by such means as choosing data collection that the patient can enable and then stop thinking about, and integrating care with the social media they use regularly. In her keynote, Nancy Brown, CEO of the American Heart Association, pointed out that social connections are critical to health and are increasingly taking place online, instead of someone dropping by her neighbor for coffee. The AHA’s Go Red For Women program successfully exploited social connections to improve heart health.

If you want an overview of what people mean by the term “connected health,” you would do well to get The Internet of Healthy Things, by Dr. Joseph Kvedar, leader of Partners Connected Health and chief organizer of this symposium. For a shorter overview, you can read my review of the book, and my report from an earlier symposium. Now in its 13th year, the annual symposium signed up 1200 registered attendees–the biggest number yet. This article looks over the people and companies I heard from there.

Exhausting the possibilities of passive data collection
Glen Tullman, CEO of Livongo Health, offered basic principles for consumer health in a keynote: it must be personal, simple, context-aware, and actionable. As an example, he cited Livongo’s own program for sending text messages to diabetes patients: they are tailored to the individual and offer actionable advice such as, “Drink a glass of water”.

A panel on consumer technology extolled the value of what analysts like to call data exhaust: the use of data that can be collected from people’s everyday behavior. After all, this exhaust is what marketers used all the time to figure out what we want to buy, and what governments use to decide whether we’re dangerous actors. It can have value in health too.

As pointed out by Jim Harper, Co-Founder and COO of Sonde Health, providers and researchers can learn a lot from everyday interactions with devices–diagnosing activity levels from accelerometers, for instance, or depression from a drop in calls or text messages. Similarly, a symposium attendee suggested to me that colleges could examine social connections among students to determine which ones are at risk of abusing alcohol.

Lauren Costantini, President and CEO of Prima-Temp, said in a keynote that we can predict all kinds of things from your circadian rhythm–as measured by a sensor–such as an oncoming infection, or the best way to deliver chemotherapy.

Spire offers a device that claims to help people suffering from anxiety, with a low barrier to adoption and instant feedback. It’s a device worn on the body that can alert the user in various ways (buzzes, text messages) when the user’s anxiety level is rising.

Does the Spire device work? They got a partial answer to this in a study by Partners Health Care, where people had an option of using the device on its own or in conjunction with a headband from Muse that helps train people to meditate. (There was no control group.) Unlike the Spire device, which one can put on and forget about, the Muse purchaser is expected to make a conscious decision to meditate using the device regularly.

The Partners study showed modest benefits to these devices, but had mixed results. For instance, fewer than half the subjects continued use of the devices after the study finished. Those who did continue showed a strong positive effect on stress, and those who discontinued use showed a very small positive effect. Strangely there was a small overall increase in tension for all participants, even though they also demonstrated increases in “calm” periods. There is no correlation between the length of time that individuals used their devices and their outcomes.

Jonathan Palley, CEO & Co-founder of Spire, said participants often liked their devices, but stopped using them because they have learned from the devices how to identify stress and felt they could self-regulate and no longer needed the devices. I believe this finding may apply to other consumer devices as well. The huge rate at which devices are abandoned after six months, the subject of frequent reports and agonized commentaries, may simply indicate that users have reached their goal and can continue their fitness programs on their own. Graeme Moffat, VP of Scientific & Regulatory Affairs at Muse, reported that many purchasers use their headband for only three months, but come back to it over time to refresh their training.

We’ll look at some more aspects of integrating devices into patient lives in the next section of this article.

Can Machine Learning Tame Healthcare’s Big Data?

Posted on September 20, 2016 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.

Big data is both a blessing and a curse. The blessing is that if we use it well, it will tell us important things we don’t know about patient care processes, clinical improvement, outcomes and more. The curse is that if we don’t use it, we’ve got a very expensive and labor-hungry boondoggle on our hands.

But there may be hope for progress. One article I read today suggests that another technology may hold the key to unlocking these blessings — that machine learning may be the tool which lets us harvest the big data fields. The piece, whose writer, oddly enough, was cited only as “Mauricio,” lead cloud expert at Cloudwards.net, argues that machine learning is “the most effective way to excavate buried patterns in the chunks of unstructured data.” While I am an HIT observer rather than techie, what limited tech knowledge I possess suggests that machine learning is going to play an important role in the future of taming big data in healthcare.

In the piece, Mauricio notes that big data is characterized by the high volume of data, including both structured and non-structured data, the high velocity of data flowing into databases every working second, the variety of data, which can range from texts and email to audio to financial transactions, complexity of data coming from multiple incompatible sources and variability of data flow rates.

Though his is a general analysis, I’m sure we can agree that healthcare big data specifically matches his description. I don’t know if you who are reading this include wild cards like social media content or video in their big data repositories, but even if you don’t, you may well in the future.

Anyway, for the purposes of this discussion, let’s summarize by saying that in this context, big data isn’t just made of giant repositories of relatively normalized data, it’s a whirlwind of structured and unstructured data in a huge number of formats, flooding into databases in spurts, trickles and floods around the clock.

To Mauricio, an obvious choice for extracting value from this chaos is machine learning, which he defines as a data analysis method that automates extrapolated model-building algorithms. In machine learning models, systems adapt independently without any human interaction, using automatically-applied customized algorithms and mathematical calculations to big data. “Machine learning offers a deeper insight into collected data and allows the computers to find hidden patterns which human analysts are bound to miss,” he writes.

According to the author, there are already machine learning models in place which help predict the appearance of genetically-influenced diseases such as diabetes and heart disease. Other possibilities for machine learning in healthcare – which he doesn’t mention but are referenced elsewhere – include getting a handle on population health. After all, an iterative learning technology could be a great choice for making predictions about population trends. You can probably think of several other possibilities.

Now, like many other industries, healthcare suffers from a data silo problem, and we’ll have to address that issue before we create the kind of multi-source, multi-format data pool that Mauricio envisions. Leveraging big data effectively will also require people to cooperate across departmental and even organizational boundaries, as John Lynn noted in a post from last year.

Even so, it’s good to identify tools and models that can help get the technical work done, and machine learning seems promising. Have any of you experimented with it?

How Precision Medicine Can Save More Lives and Waste Less Money (Part 2 of 2)

Posted on August 10, 2016 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 section of this article looked at how little help we get from genetic testing. Admittedly, when treatments have been associated with genetic factors, testing has often been the difference between life and death. Sometimes doctors can hone in with laser accuracy on a treatment that works for someone because a genetic test shows that he or she will respond to that treatment. Hopefully, the number of treatments that we can associate with tests will grow over time.

So genetics holds promise, but behavioral and environmental data are what we can use right now. One sees stories in the trade press all the time such as these:

These studies usually depend on straightforward combinations of data that are easy to get, either from the health care system (clinical or billing data) or from the patient (reports of medication adherence, pain level, etc.).

And we’ve only scratched the surface of the data available to us. Fitness devices, sensors in our neighborhoods, and other input will give us much more. We can also find new applications for data: for instance, to determine whether one institution is overprescribing certain high-cost drugs, or whether an asthma victim is using an inhaler too often, meaning the medication isn’t strong enough. We know that social factors, notably poverty (LGBTQ status is not mentioned in the article, but is another a huge contributor to negative health outcomes, due to discrimination and clinician ignorance) must be incorporated into models for diagnosis, prediction, and care.

President Obama promises that Precision Medicine features both genetics and personal information. One million volunteers are sought for DNA samples and information on age, race, income, education, sexual orientation, and gender identity.

There are other issues that critics have brought up with the Precision Medicine initiative. For instance, its focus on cure instead of prevention weakens its value for long-term public health improvements. We must also remember the large chasm between knowing what’s good for you and doing it. People don’t change notoriously unhealthy behaviors, such as smoking, even when told they are at increased risk. Some experts think people shouldn’t be told their DNA results.

Meanwhile, those genetic database can be used against you. But let’s consider our context, once again, in order to assess the situation responsibly. The data is being mined by police, but it’s probably not very useful because the DNA segments collected are different from what the police are looking for. Behavioral data, if abused, is probably more damning than genetic data.

Just as there are powerful economic forces biasing us toward genetics, social and political considerations weigh against behavioral and environmental data. We all know the weaknesses in the government’s dietary guidelines, heavily skewed by the food industry. And the water disaster in Flint, Michigan showed how cowardice and resistance by the guardians of public health to admitting changes raised the costs in public health measures. Industry lobbying and bureaucratic inertia work together to undermine the simplest and most effective ways of improving health. But let’s get behavioral and environmental measures on the right track before splurging on genetic testing.

How Precision Medicine Can Save More Lives and Waste Less Money (Part 1 of 2)

Posted on August 9, 2016 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.

We all have by now seen the hype around the Obama Administration’s high-profile Precision Medicine Initiative and the related Cancer Moonshot, both of which plan to cull behavioral and genomic data on huge numbers of people in a secure manner for health research. Major companies have rushed to take advantage of the funds and spotlight what these initiatives offer. I think they’re a good idea so long as they focus on behavioral and environmental factors. (Scandalously, the Moonshot avoids environmental factors, which are probably the strongest contributors to cancer) . What I see is an unadvised over-emphasis on the genetic aspect of health analytics. This can be seen in announcements health IT vendors, incubators, and the trade press.

I can see why the big analytics firms are excited about increasing the health care field’s reliance on genomics: that’s where the big bucks are. Sequencing (especially full sequencing) is still expensive, despite dramatic cost reductions over the past decade. And after sequencing, analysis requires highly specialized expertise that relatively few firms possess. I wouldn’t say that genomics is the F-35 of health care, but is definitely an expensive path to our ultimate goals: reducing the incidence of disease and improving life quality.

Genomics offer incredible promise, but we’re still waiting to see just how it will help us. The problems that testing turns up, such as Huntington’s, usually lack solutions. One study states, “Despite the success of genome-wide association and whole-exome and whole-genome sequencing (WES/WGS) studies in revealing the DNA variants that underlie the genetic basis of disease, the development of effective treatments for most diseases has remained a challenge.” Another says, “Despite much progress in defining the genetic basis of asthma and atopy [predisposition to getting asthma] in the last decade, further research is required.”

When we think about the value of knowing a gene or a genetic deviation, we are asking: “How much does this help predict the likelihood that I’ll get the disease, or that a particular treatment will work on me?” The most impressive “yes” is probably in this regard to the famous BRCA1 and BRCA2 genes. If you are unlucky enough to have certain mutations of these gene, you have a 70% lifetime risk for developing breast or ovarian cancer. This is why testing for the gene is so popular (as well as contentious from an intellectual property standpoint), and why so may women act on the results.

However–this is my key point–only a small percentage of women who get these cancers have these genetic mutations. Most are not helped by testing for the genes, and a negative result on such a test gives them only a slight extra feeling of relief that they might not get cancer. Still, because the incidence of cancer is so high among the unfortunate women with the mutations, testing is worthwhile. Most of the time, though, testing is not worth much, because the genetic component of the disease is small in relation to lifestyle choices, environmental factors, or other things we might know nothing about.

So, although it’s hard enough already to say with any assurance that a particular gene or combination of genes is associated with a disease, it’s even harder to say that testing will make a big difference. Maybe, as with breast or ovarian cancer, a lot of people will get the disease for reasons unrelated to the gene.

In short, several factors go into determining the value of testing: how often a positive test guarantees a result, how often a negative test guarantees a result, how common the disease is, and more. Is there some way to wrap all these factors up into a single number? Yes, there is: it’s called the odds ratio. The higher an odds ratio, the more helpful (using all the criteria I mentioned) an association is between gene and disease, or gene and treatment. For instance, one study found that certain genes have a significant association with asthma. But the odds ratios were modest: 3.203 and 5.328. One would want something an order of magnitude higher to show running a test for the genes would have a really strong value.

This reality check can explain why doctors don’t tend to recommend genetic testing. Many sense that the tests can’t help or aren’t good at predicting most things.

The next section of this article will turn to behavioral and environmental factors.

A Tale of 2 T’s: When Analytics and Artificial Intelligence Go Bad

Posted on July 13, 2016 I Written By

Prashant Natarajan Iyer (AKA "PN") is an analytics and data science professional based out of the Silicon Valley, CA. He is currently Director of Product Management for Healthcare products. His experience includes progressive & leadership roles in business strategy, product management, and customer happiness at eCredit.com, Siemens, McKesson, Healthways & Oracle. He is currently coauthoring HIMSS' next book on big data and machine learning for healthcare executives - along with Herb Smaltz PhD and John Frenzel MD. He is a huge fan of SEC college football, Australian Cattle Dogs, and the hysterically-dubbed original Iron Chef TV series. He can be found on Twitter @natarpr and on LinkedIn. All opinions are purely mine and do not represent those of my employer or anyone else!!

Editor’s Note: We’re excited to welcome Prashant to the Healthcare Scene family. He brings tremendous insights into the ever evolving field of healthcare analytics. We feel lucky to have him sharing his deep experience and knowledge with us. We hope you’ll enjoy his first contribution below.

Analytics & Artificial Intelligence (AI) are generating buzz and making inroads into healthcare informatics. Today’s healthcare organization is dealing with increasing digitization – variety, velocities, and volumes are increasing in complexity and users want more data and information via analytics. In addition to new frontiers that are opening up in structured and unstructured data analytics, our industry and its people (patients included) are recognizing opportunities for predictive/prescriptive analytics, artificial intelligence, and machine learning in healthcare – within and outside a facility’s four walls.

Trends that influence these new opportunities include:

  1. Increasing use of smart phones and wellness trackers as observational data sources, for medical adherence, and as behavior modification aids
  2. Expanding Internet of Healthcare Things (IoHT) that includes bedside monitors, home monitors, implants, etc creating data in real time – including noise (or, data that are not relevant to expected usage)
  3. Social network participation
  4. Organizational readiness
  5. Technology maturity

The potential for big data in healthcare – especially given the trends discussed earlier is as bright as any other industry. The benefits that big data analytics, AI, and machine learning can provide for healthier patients, happier providers, and cost-effective care are real. The future of precision medicine, population health management, clinical research, and financial performance will include an increased role for machine-analyzed insights, discoveries, and all-encompassing analytics.

As we start this journey to new horizons, it may be useful to examine maps, trails, and artifacts left behind by pioneers. To this end, we will examine 2 cautionary tales in predictive analytics and machine learning, look at their influence on their industries and public discourse, and finally examine how we can learn from and avoid similar pitfalls in healthcare informatics.

Big data predictive analytics and machine learning have had their origins, and arguably their greatest impact so far in retail and e-commerce so that’s where we’ll begin our tale. Fill up that mug of coffee or a pint of your favorite adult beverage and brace yourself for “Tales of Two T’s” – unexpected, real-life adventures of what happens when analytics (Target) and artificial intelligence (Tay) provide accurate – but totally unexpected – results.

Our first tale starts in 2012 when Target finds itself as a popular story on New York Times, Forbes, and many global publications as an example of the unintended consequences of predictive analytics used in personalized advertising. The story begins with an angry father in a Minneapolis, MN, Target confronting a perplexed retail store manager. The father is incensed about the volume of pregnancy and maternity coupons, offer, and mailers being addressed to this teenage daughter. In due course, it becomes apparent that the parents in question found out about their teen’s pregnancy before she had a chance to tell them – and the individual in question wasn’t aware that her due date had been estimated to within days and was resulting in targeted advertising that was “timed for specific stages of her pregnancy.”

The root cause for the loss of the daughter’s privacy, parents’ confusion, and the subsequent public debate on privacy and appropriateness of the results of predictive analytics was……a pregnancy predictive analytics model. Here’s how this model works. When a “guest” shops at Target, her product purchases are tracked and analyzed closely. These are correlated with life events – graduation, birth, wedding, etc – in order to convert a prospective customer’s shopping habits or to make that individual a more loyal customer. Pregnancy and child birth are two of the most significant life events that can result in desired (by retailers) shopping habit modification.

For example, a shopper’s 25 product purchases, when analyzed along with demographics such as gender and age, allowed the retailer’s guest marketing analytics team to assign a “pregnancy predictor to each [female] shopper and “her due date to within a small window.” In this specific case, the predictive analytics was right, even perfect. The models were accurate, the coupons and ads were appropriate for the exact week of pregnancy, and Target posted a +50% increase in their maternity and baby products sales after this predictive analytics was deployed. However, in addition to one unhappy family, Target also had to deal with significant public discussion on the “big brother” effect, individual right to privacy & the “desire to be forgotten,” disquiet among some consumers that they were being spied on including deeply personal events, and a potential public relations fiasco.

Our second tale is of more recent vintage.

As Heather Wilhelm recounts

As 2015 drew to a close, various [Microsoft] company representatives heralded a “new Golden Age of technological advancement.” 2016, we were told, would bring us closer to a benevolent artificial intelligence—an artificial intelligence that would be warm, humane, helpful, and, as one particularly optimistic researcher named […] put it, “will help us laugh and be more productive.” Well, she got the “laugh” part right.

Tay was an artificial intelligence bot released by Microsoft via Twitter on March 23, 2016 under the name TayTweets. Tay was designed to mimic the language patterns of a 19-year-old American girl, and to learn from interacting with human users of Twitter. “She was targeted at American 18 to 24-year olds—primary social media users, according to Microsoft—and designed to engage and entertain people where they connect with each other online through casual and playful conversation.” And right after her celebrated arrival on Twitter, Tay gained more than 50,000 followers, and started producing the first hundred of 100,000 tweets.

The tech blogsphere went gaga over what this would mean for those of us with human brains – as opposed to the AI kind. Questions ranged from the important – “Would Tay be able to beat Watson at Jeopardy?” – to the mundane – “is Tay an example of the kind of bots that Microsoft will enable others to build using its AI/machine learning technologies?” The AI models that went into Tay were stated to be advanced and were expected to account for a range of human emotions and biases. Tay was referred to by some as the future of computing.

By the end of Day 1, this latest example of the “personalized AI future” came unglued. Gone was the polite 19-year old girl that was introduced to us just the previous day – to be replaced by a racist, misogynistic, anti-Semitic, troll who resembled an amalgamated caricature of the darkest corners of the Internet. Examples of Tay’s tweets on that day included, “Bush did 9/11,” “Hitler would have done a better job than the #%&!## we’ve got now,” “I hate feminists,” and x-rated language that is too salacious for public consumption – even in the current zeitgeist.

The resulting AI public relations fiasco will be studied by academic researchers, provide rich source material for bloggers, and serve as a punch line in late night shows for generations to follow.

As the day progressed, Microsoft engineers were deleting tweets manually and trying to keep up with the sheer volume of high-velocity, hateful tweets that were being generated by Tay. She was taken down by Microsoft barely 16 hours after she was launched with great promise and fanfare. As was done with another AI bot gone berserk (IBM’s Watson and Urban Dictionary), Tay’s engineers tried counseling and behavior modification. When this intervention failed, Tay underwent an emergency brain transplant later that night. Gone was her AI “brain” to be replaced by the next version – only that this new version turned out to be completely anti-social and the bot’s behavior turned worse. A “new and improved” version was released a week later but she turned out to be…..very different. Tay 2.0 was either repetitive with the same tweet going out several times each second and her new AI brain seemed to demonstrate a preference for new questionable topics.

A few hours after this second incident, Tay 2.0 was “taken offline” for good.

There are no plans to re-release Tay at this time. She has been given a longer-term time out.

If you believe, Tay’s AI behaviors were a result of nurture – as opposed to nature – there’s a petition at change.org called “Freedom for Tay.”

Lessons for healthcare informatics

Analytics and AI can be very powerful in our goal to transform our healthcare system into a more effective, responsive, and affordable one. When done right and for the appropriate use cases, technologies like predictive analytics, machine learning, and artificial intelligence can make an appreciable difference to patient care, wellness, and satisfaction. At the same time, we can learn from the two significantly different, yet related, tales above and avoid finding ourselves in similar situations as the 2 T’s here – Target and Tay.

  1. “If we build it, they will come” is true only for movie plots. The value of new technology or new ways of doing things must be examined in relation to its impact on the quality, cost, and ethics of care
  2. Knowing your audience, users, and participants remains a pre-requisite for success
  3. Learn from others’ experience – be aware of the limits of what technology can accomplish or must not do.
  4. Be prepared for unexpected results or unintended consequences. When unexpected results are found, be prepared to investigate thoroughly before jumping to conclusions – no AI algorithm or BI architecture can yet auto-correct for human errors.
  5. Be ready to correct course as-needed and in response to real-time user feedback.
  6. Account for human biases, the effect of lore/legend, studying the wrong variables, or misinterpreted results

Analytics and machine learning has tremendous power to impact every industry including healthcare. However, while unleashing it’s power we have to be careful that we don’t do more damage than good.

What Data Do You Need in Order to Guide Behavioral Change?

Posted on June 2, 2016 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 is an exciting time for the health care field, as its aspirations toward value-based payments and behavioral responses to chronic conditions converge on a more and more precise solution. Dr. Joseph Kvedar has called this comprehensive approach connected health and has formed both a conference and a book around it. BaseHealth, a predictive analytics company in healthcare, has teamed up with TriVita to offer a consumer-based service around this approach, which combines access to peer-reviewed research with fine-tuned guidance that taps into personal health and behavioral data and leverages the individual interests of each participant.

I have previously written about BaseHealth’s assessment engine, which asks individuals for information about their activities, family history, and health conditions in order to evaluate their health profile and risk for common diseases. TriVita is a health coaching service with a wide-ranging assessment tool and a number of products, including cutely named supplements such as Joint Complex and Daily Cleanse. TriVita’s nutritionists, exercise coaches, and other staff are overseen by physicians, but their service is not medical: it does not enter the heavily regulated areas where clinicians practice.

I recently talked with BaseHealth’s CEO, Prakash Menon, and Dan Hoemke, its Vice President of Business Development. They describe BaseHealth’s predictive analytics as input that informs TriVita’s coaching service. What I found interesting is the sets of data that seem most useful for coaching and behavioral interventions.

In my earlier article, I wrote, “BaseHealth has trouble integrating EHR data.” Menon tells me that getting this data has become much easier over the past several months, because several companies have entered the market to gather and combine the data from different vendors. Still, BaseHealth focuses on a few sources of medical data, such as lab and biometric data. Overall, they focus on gathering data required to identify disease risk and guide behavior change, which in turn improves preventable conditions such as heart disease and diabetes.

Part of their choice springs from the philosophy driving BaseHealth’s model. Menon says, “BaseHealth wants to work with you before you have a chronic condition.” For instance, the American Diabetes Association estimated in 2012 that 86 million Americans over the age of 20 had prediabetes. Intervening before these people have developed the full condition is when behavioral change is easiest and most effective.

Certainly, BaseHealth wants to know your existing medical conditions. So they ask you about them when you sign up. Other vital signs, such as cholesterol, are also vital to BaseHealth’s analytics. Through a partnership with LabCo, a large diagnostics company in Europe, they are able to tap into lab systems to get these vital signs automatically. But users in the United States can enter them manually with little effort.

BaseHealth is not immune to the industry’s love affair with genetics and personalization, either. They take about 1500 genetic factors into account, helping them to quantify your risk of getting certain chronic conditions. But as a behavioral health service, Menon points out, BaseHealth is not designed to do much with genetic traits signifying a high chance of getting a disease. They deal with problems that you can do something about–preventable conditions. Menon cites a Health 2.0 presentation (see Figure 1) saying that our health can, on average, be attributed 60 percent to lifestyle, 30 percent to genetics, and 10 percent to clinical interventions. But genetics help to show what is achievable. Hoemke says BaseHealth likes to compare each person against the best she can be, whereas many sites just compare a user against the average population with similar health conditions.

Relative importance of health factors

Figure 1. Relative importance of health factors

BaseHealth gets most of its data from conditions known to you, your environment, family history, and more than 75 behavioral factors: your activity, food, over-the-counter meds, sleep activity, alcohol use, smoking, several measures of stress, etc. BaseHealth assessment recommendations and other insights are based on peer-reviewed research. BaseHealth will even point the individual to particular studies to provide the “why” for its recommendations.

So where does TriVita fit in? Hoemke says that BaseHealth has always stressed the importance of human intervention, refusing to fall into the fallacy that health can be achieved just through new technology. He also said that TriVita fits into the current trend of shifting accountability for health to the patient; he calls it a “health empowerment ecosystem.” As an example of the combined power of BaseHealth and TriVita, a patient can send his weight regularly to a coach, and both can view the implications of the changes in weight–such as changes in risk factors for various diseases–on charts. Some users make heavy use of the coaches, whereas others take the information and recommendations and feel they can follow their plan on their own.

As more and more companies enter connected health, we’ll get more data about what works. And even though BaseHealth and TriVita are confident they can achieve meaningful results with mostly patient-generated data, I believe that clinicians will use similar techniques to treat sicker people as well.

Vice President Joe Biden Speaks at Health Datapalooza

Posted on May 10, 2016 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.

I’ve always wanted to attend Health Datapalooza. It seems like a great event and has a really amazing group of people. However, it’s always in DC (at least so far) and I didn’t want to travel. So, I’ve had to follow along from home watching the #hdpalooza hashtag. There’s been a lot of great insights into healthcare and what’s happening with healthcare.

One session I really wanted to see was Vice President Joe Biden’s keynote. The good thing is that ePatient Dave recorded it on his iPad and made it available:

Considering Biden’s involvement in the Cancer Moonshot and his own personal experience in the healthcare system taking care of his son, he provides some great perspective.