Archive for the ‘Healthcare informatics’ category

Google Plays Doctor

April 17, 2008

And Microsoft also wants your health records as well. The New York Times reports on the NEJM article warning about the entrants of mega-players GOOG and MSFT as purveyors of your private healthcare information. These are not altruistic enterprises – they have to turn a profit on this somehow. So it does make one wonder about their product strategy – will Google flash consumer health ads at you while you review your meds and shots? Will Microsoft create a new Health Passport ID to qualify your access to your medical records on their servers? So, who will the early adopters be?

There are so many questions –  Have you actually tried to locate and consolidate your medical records? Unless you’re a veteran and on VistA, have you noticed they are paper? So, will Google scan them for you as well?  What happens to your records if you leave the country or die? What if laws change and you don’t know about it? Can your doctor get to your online records, and will they have to have a separate ID for all their patients?  Why can’t people just put all their records on a flash drive, that gets updated at the doctor’s office, and then keep it physically with them – and a backup on the laptop? What’s the real value-add of the big players here?

The NEJM authors consider the trend toward personal health records a positive development for personal responsibility for healthcare:

Despite their warnings, Dr. Mandl and Dr. Kohane are enthusiastic about the potential benefits of Web-based personal health records, including a patient population of better-informed, more personally responsible health consumers.

“In very short order, a few large companies could hold larger patient databases than any clinical research center anywhere,” Dr. Mandl said in an interview.

But the authors see a need for safeguards, suggesting a mixture of federal regulation — perhaps extending HIPAA to online patient record hosts — contract relationships, certification standards and consumer education programs.

Today you’ll see almost 200K Google hits on “personal health record.” The growth in this “space” in one year has been truly amazing. Remember that Microsoft bought Medstory a year ago, (and there’s been no news from them since). However, if the US had a national healthcare system (or even statewide systems), a feature of patient and cost management would be your access to personal versions of electronic health records, such as those available on the aforementioned VistA system. Your tax dollars already built VistA and it is a public domain application. Veterans already have personal eHealth records through My Health eVet (such a bad brand name, but you get the idea). In this area, government and veterans are at the leading edge, and not consumers.

Information services providers might consider the “records aftermarket,” rather than records access –  helping consumers make sense of their health records. The medical records themselves are used as intra-practice documentation, and are not easily readable by patients or family members. Perhaps this is the direction the big players want to go, but there’s a lot of gray area in the reading and interpretation of medical tests, procedure, physician orders and notes, etc.  Doing this well is an editorial process – currently services such as Medstory do a good job getting you to qualified information, but the level of difficulty of interpretation and sensemaking remains daunting.


Multisensory Medical Informatics

February 7, 2008

Wow – The Wii earns my respect as a serious haptic interface. A University of Arizona team has shown improvements in fine hand motor skill developed from exercises in continuous practice of the Wii for simulated laproscopy.

I’ve interacted with the virtual gall bladder removal and cauterization simulation at Riverside Methodist hospital’s Virtual Care Unit. It tallied a game score just like a Wii game – but the Wii interface may have leveled the playing field by making it possible to learn and tune fine-grained motor skills in the context of purposeful (and cheap) simulations.

The virtual OR lab at the National Center for Collaboration in Medical Modeling and Simulation has been developing alternatives to learning hands-on procedures, primarily based on practice of motor routines in roughly simulated situations. Mark Scerbo, human factors psychologist (and a Cincinnati grad), explains:

“It’s like doing very sophisticated surgery with chopsticks in your hands,” Scerbo said. “It takes a lot of training to look at a two-dimensional display and understand what your instruments are doing. There’s a real need to train doctors, and not on patients.”

Surgeon Leonard Weireiter said: “It turns out you don’t need the high-fidelity haptics. It’s the repetitive practice of the motion that counts.”

Consider the similarity to sports psychology research that shows significant performance improvements from visualization exercises and mental practice. The brain-body system entrains toward the optimal physical movements, timing, and interaction with devices over practice, even when roughly simulated.

Impact on Medical Practice?

Healthcare informatics and e-Learning are rapidly evolving, from several directions –  clinical decisionmaking, patient eMR and personal healthcare records, consumer health information, drug information, resident education, specialist informatics, nursing education, genomics, institutional workflow, finance and insurance integration, and collaborative diagnostics. (I’d link all these to examples, but this was a handful just to type – if I get a reply, I’l do the links!)

Healthcare services and institutions represent a massive information ecology and infoconomy. A significant activity for design research involves understanding these resources and content sources as living, growing players in an ecosystem that cannot be designed, but rather interfaced, linked, connected, and metadatabased. We need ways to visualize the resources, ontologies/taxonomies and information objects available in the overall emerging system – a picture of the stable niches, emerging services, publishers and providers, institutions and their drivers, the relationships among these, and the size/impact of each service in the overall scheme of things. (Let me know if you find one!)

“Mass consumerization of health information”

August 18, 2007

What does that really mean? Here’s the reference, from this week’s NY Times: Google and Microsoft Look to Change Health Care (And since they charge for content once its a few days old, I’ve posted a few paras fair-use style below).

I see the possible head-on collision of supply-side healthcare information services with the recognition of individual health needs. The fact is that people also use Google to search for sites that might contain valid information or perspectives on a disease condition. That does not ipso facto lead to people storing their healthcare records with Google’s servers.

In the US insurance system, I would not even want my search history of healthcare related issues to be available. Trust Google with personal information requiring that level of privacy? Would you trust your permanent record with Google?

The article suggests some powerful directions for systems and services designers to consider: Healthcare must become more collaborative. Patients with chronic disease conditions interact with multiple healthcare professionals, and ask different questions and express different needs at different times. With the U of T Laboratory for Collaborative Diagnostics we are exploring diagnostics tools and collaborative informatics for collaborative practice. We are also creating new tools for engaging the individual as a central participant in their healthcare intervention. Not just patient-centered medicine, but individual-centered collaborative healthcare.

What will Google and Microsoft do? Well, Microsoft has already purchased Medstory – We should start exchaning our experiences with using it for real clinical issues. (I’m not sure it helps any more that a straight Google search for the trials I’ve made with it.) So, what else do we think Google or Microsoft will show up with? Are these strategic acquisitions?

Here’s some of the article, from Steven Lohr – who covers this beat for the Times:

In politics, every serious candidate for the White House has a health care plan. So too in business, where the two leading candidates for Web supremacy, Google and Microsoft, are working up their plans to improve the nation’s health care.

By combining better Internet search tools, the vast resources of the Web and online personal health records, both companies are betting they can enable people to make smarter choices about their health habits and medical care.

“What’s behind this is the mass consumerization of health information,” said Dr. David J. Brailer, the former health information technology coordinator in the Bush administration, who now heads a firm that invests in health ventures.

It is too soon to know whether either Google or Microsoft will make real headway. Health care, experts note, is a field where policy, regulation and entrenched interests tend to slow the pace of change, and technology companies have a history of losing patience.

And for most people, typing an ailment into a Web search engine is very different from entrusting a corporate titan with personal information about their health.

Google and Microsoft recognize the obstacles, and they concede that changing health care will take time. But the companies see the potential in attracting a large audience for health-related advertising and services. And both companies bring formidable advantages to the consumer market for such technology.

Microsoft’s software animates more than 90 percent of all personal computers, while Google is the default starting point for most health searches. And people are increasingly turning to their computers and the Web for health information and advice. A Harris poll, published last month, found that 52 percent of adults sometimes or frequently go to the Web for health information, up from 29 percent in 2001.

If the efforts of the two big companies gain momentum over time, that promises to accelerate a shift in power to consumers in health care, just as Internet technology has done in other industries.

Today, about 20 percent of the nation’s patient population have computerized records — rather than paper ones — and the Bush administration has pushed the health care industry to speed up the switch to electronic formats. But these records still tend to be controlled by doctors, hospitals or insurers. A patient moves to another state, for example, but the record usually stays.

The Google and Microsoft initiatives would give much more control to individuals, a trend many health experts see as inevitable. “Patients will ultimately be the stewards of their own information,” said John D. Halamka, a doctor and the chief information officer of the Harvard Medical School.

Already the Web is allowing people to take a more activist approach to health. According to the Harris survey, 58 percent of people who look online for health information discussed what they found with their doctors in the last year.

It is common these days, Dr. Halamka said, for a patient to come in carrying a pile of Web page printouts. “The doctor is becoming a knowledge navigator,” he said. “In the future, health care will be a much more collaborative process between patients and doctors.”

Microsoft and Google are hoping this will lead people to seek more control over their own health records, using tools the companies will provide. Neither company will discuss their plans in detail. But Microsoft’s consumer-oriented effort is scheduled to be announced this fall, while Google’s has been delayed and will probably not be introduced until next year, according to people who have been briefed on the companies’ plans.

A prototype of Google Health, which the company has shown to health professionals and advisers, makes the consumer focus clear. The welcome page reads, “At Google, we feel patients should be in charge of their health information, and they should be able to grant their health care providers, family members, or whomever they choose, access to this information. Google Health was developed to meet this need.”

A presentation of screen images from the prototype — which two people who received it showed to a reporter — then has 17 other Web pages including a “health profile” for medications, conditions and allergies; a personalized “health guide” for suggested treatments, drug interactions and diet and exercise regimens; pages for receiving reminder messages to get prescription refills or visit a doctor; and directories of nearby doctors.

Google executives would not comment on the prototype, other than to say the company plans to experiment and see what people want. “We’ll make mistakes and it will be a long-range march,” said Adam Bosworth, a vice president of engineering and leader of the health team. “But it’s also true that some of what we’re doing is expensive, and for Google it’s not.”

At Microsoft, the long-term goal is similarly ambitious. “It will take grand scale to solve these problems like the data storage, software and networking needed to handle vast amounts of personal health and medical information,” said Steve Shihadeh, general manager of Microsoft’s health solutions group. “So there are not many companies that can do this.”

This year, Microsoft bought a start-up, Medstory, whose search software is tailored for health information, and last year bought a company that makes software for retrieving and displaying patient information in hospitals. Microsoft software is already used in hospitals, clinical laboratories and doctors’ offices, and, Mr. Shihadeh noted, the three most popular health record systems in doctors’ offices are built with Microsoft software and programming tools.

Microsoft will not disclose its product plans, but according to people working with the company the consumer effort will include online offerings as well as software to find, retrieve and store personal health information on personal computers, cellphones and other kinds of digital devices — perhaps even a wristwatch with wireless Internet links some day.

Mr. Shihadeh declined to discuss specifics, but said, “We’re building a broad consumer health platform, and we view this challenge as far bigger than a personal health record, which is just scratching the surface.”

Yet personal health records promise to be a thorny challenge for practical and privacy reasons. To be most useful, a consumer-controlled record would include medical and treatment records from doctors, hospitals, insurers and laboratories. Under federal law, people can request and receive their personal health data within 90 days. But the process is complicated, and the replies typically come on paper, as photocopies or faxes.

The efficient way would be for that data to be sent over the Internet into a person’s digital health record. But that would require partnerships and trust between health care providers and insurers and the digital record-keepers.

Privacy concerns are another big obstacle, as both companies acknowledge. Most likely, they say, trust will build slowly, and the online records will include as much or as little personal information as users are comfortable divulging.

A person might start, for example, by typing in age, gender and a condition, like diabetes, as a way to find more personalized health information. If a person creates a personal health record and later has second thoughts, a simple mouse click should erase it. The promise, the companies say, will be complete consumer control.

There are plenty of competitors these days in online health records and information from start-ups like Revolution Health, headed by AOL’s founder, Stephen M. Case, and thriving profit-makers led by WebMD.

Potential rivals are not underestimating the two technology giants. But the smaller companies have the advantage of being focused entirely on health, and some have been around for years. WebMD, for example, traces its lineage to Healtheon, a fallen star of the dot-com era, founded by the Netscape billionaire Jim Clark.

Google and Microsoft are great companies, said Wayne T. Gattinella, WebMD’s chief executive, but “that doesn’t mean they will be expert in a specific area like health.”

Specialized health search engines — notably Healthline — are gaining ground and adding partners. AOL recently began using Healthline for searches on its health pages, even though Google is a close partner.

Still, 58 percent of people seeking health information online begin with a general search engine, according to a recent Jupiter Research report, and Google dominates the field. “Google is the entry point for most health search, and that is a huge advantage,” said Monique Levy, a Jupiter analyst.

Indeed, it is the market reach and deep pockets that Google and Microsoft can bring to consumer health information that intrigues medical experts, and has lured recruits. Dr. Roni Zeiger, a graduate of Stanford’s School of Medicine, a medical informatics researcher and a former primary care doctor, joined Google last year. The 36-year-old, who still sees patients some evenings and weekends at a nearby clinic, said, “At Google, I can use my expertise and knowledge to potentially help millions of people each day.”

2Collaboration with Elsevier

August 15, 2007

Beta launch of 2collab – Elsevier’s new social bookmarking and networking tool has been released in Beta. I’ve registered and started tagging some articles – I’m finding it very simple to get in and working with it. Try it out yourself and see –

On June 26, the beta version of 2collab was launched to the Scopus and ScienceDirect Development Partners. 2collab is a new collaborative research tool that enables researchers to share bookmarks, references or any linked materials with their peers and colleagues. Users can share, collaborate and discuss resources either in private groups or openly with the wider scientific community.

A common scenario involves a researcher writing an article with co-authors around the globe. Using 2collab he/she can store and share information resources such as research articles centrally so colleagues can access them. Bookmarks can be tagged to allow new ways of searching and accessing information. In addition, researchers can comment, rate and evaluate these resources in their groups. This makes collaboration more efficient and helps researchers share, connect and explore. All without the need for long and complex email strings!

“2collab beta is just the starting point,” says Michiel van der Heyden, Senior Product Manager, ScienceDirect. “We plan to create a platform that allows researchers across the globe not just to collaborate on evaluating information but also to help them build new networks, share expertise, and discover new information resources. And we gain from having an opportunity to observe and learn from their behavior.”

While I realize Elsevier’s ScienceDirect and Scopus are the flagship services here, 2Collab would seem to be a great fit with Scirus. Since 2Collab tags open resources that others can locate from your tags, Scirus’ indexing across open science resources would be a good for the early discovery stages of lit research.  Also, Scirus recommends terms to you drawn from the search results. I know tagging is supposed to be user-specified, but the idea is to use meaningful tags recognized by other users – Scirus has a great index already from its (linked) suggested terms.  How about an icon to display Scirus terms accessible to 2Collab?

Procedures Consult – Immersive experience in rich content

July 16, 2007

Nothing but medical procedures. No tagging, no RSS feeds, no social networking. (Not yet anyway.) The innovation was in discovering exactly what was required for the intensive education of doctors in residency programs, delivering these capabilities and minimizing all other distractions. Here the user experience is in the transparent immersion in the details of anatomy, positioning, approach and entry points, techniques, and simulations of all the internal parts you never could see even with live patients.
Elsevier released ProceduresConsult this month, to institutional subscribers. This e-learning service provides animated and acted video and text details for common internal medicine procedures, so you can learn to do arthrocentesis at home. (Hopefully, only if you are in a residency program). Medical residents in internal medicine (and soon Orthopedics and Anethseia will be able to locate detailed interactive procedures education with extraordinary video produced by Harvard’s Dr. Todd Thomsen.


Both 320 (on-page) and 640 (video player) videos are provided for every procedure. Text is shown in two full-scrollable pages, Quick Review and Full Details. Navigation was based on how residents typically work with one procedure at a time – while you can search, browse and jump around, the nav model focuses attention within the procedure and limits the opportunity to link away from the page.

We started on this service from nearly scratch 10 months ago, and leveraged an existing platform (based on Elsevier’s Nursing Skills product) to deliver the testing and tracking capabilities rapidly. The interaction design was totally drawn from user research, from multiple interactions with physicians and residency program staff. The content design was generated in collaborative design with the product manager (the indefatigable Rolla Couchman), Dr. Thomsen, and the product and editorial team. We conducted multiple onsite user research sessions, evaluating the site structure and navigation, visual and interaction design, content layout, video interaction, and administrative tools in a series of iterations and increasingly-defined prototypes. Jez Alder of Elsevier produced the visual design, over multiple iterations. Look for navigation and content enhancements soon.