Today, clinical professionals have tools (like Ovid, ScienceDirect, UpToDate, Trip) that help answer complex questions and are connected to validated knowledge bases derived off of sources like PubMed. But how does a patient, with no access or expertise in the domain find and leverage this information? Medify tries to solve that.

The value proposition of Medify is not easy to describe. In fact, the ‘What is Medify‘ description on the site was banal enough to be dismissed, just like most other online social health startup marketing. They do a better (albeit prolix) job on the ‘How it works‘ page. Medify will appeal to the well-informed patients who are not afraid to sift through piles of academic articles burdened with medical jargon to understand and manage their own disease. Medify gives them a dashboard of existing literature – with it they can monitor things like which treatments are gaining traction in the provider community, which institutions are on the forefront of relevant research, etc. Affiliated web 2.0 functionality like faceted search, social sharing, tracking, annotating are bundled in to make it more personal.

Under the hood, it is smartly leveraging what public knowledge bases are already out there. The citation and abstract are free from PubMed. Interstitial phrases and terms in the content are further linked to sources like Wikipedia and MeSH. Brief outcomes or summaries are synthetically constructed from the article text.

Medify is not alone. There are other sites that try to help patients navigate the vast sea of research literature. PubMed’s parent NLM runs MedLinePlus,  UpToDate has a patient-oriented version, and niche startups like MyDailyApple, PatientsLikeMe are also tackling this to some extent.

In 2001 Brian Haynes, MD, PhD wrote an article describing the landscape of such ‘pre-appraised’ resources through a hierarchical structure that had four layers (called “4S” Model):

• Original ‘Studies’ (what PubMed provides) at the base
• ‘Syntheses’ (systematic reviews sources like The Chochrane Library) of evidence just above that
• ‘Synopses’ (like EBM, EBN Online) of studies and syntheses next up, and
• the most evolved evidence-based information ‘Decision Support Systems’ at the top.

He later expanded the model to 2 more layers (read about the “6S” paper here), but the basic argument remained same – Information seekers should begin looking at the highest level resource available for the problem that prompted their search. That is a good framework to understand why services like Medify are needed.

The skeptics would argue that offerings like Medify will do little more than empower hypochondriacs. But I believe that well-served health information only makes outcomes better. The lag time between published research being implemented in real-world medical practice can be in the order of decades. As consumers, we are entrusted to make choices about other important topics like money, and the market provides personal finance tools/services to help. Same can apply to healthcare, without diminishing the role of experts.

Isabel’s solution is based on sophisticated pattern-recognition technology from Autonomy, a huge pure software company based in UK. The powerful Autonomy engine is unleashed on a database built from medical textbooks and journal content, and claimed to have >100,000 documents covering >10,000 diseases. Think of it as a web crawler, that indexes authoritative medical text instead. The software’s strength lies in identification of the patterns that naturally occur in text, based on the usage and frequency of words or terms that correspond to specific medical concepts. Each diagnosis is assigned clinical weighting scores derived from expert opinions reflecting prevalence effects based on factors like age, gender, geography etc. The user interface incorporates Natural Language Processing to make the interaction more intuitive and easy for the clinician.

I’ve known about Isabel since 2007, and used it for a few months with very positive results. Was specially impressed with the software’s ability to handle free text, and the company’s openness to a peer-reviewed validation process in an effort to convince the medical community. It seems they have done a serious corporate make-over in 2010: renewed website, cleared pricing and changed management (Interesting fact- ex-CEO Joseph Britto now works for Autonomy).

Autonomy announced a clear healthcare-focused initiative this month, called Meaning-Based Healthcare platform (perhaps trying to capitalize the current buzz around Healthcare IT Meaningful Use regulation). Strangely, there is no clear cut mention of Isabel on the AutonomyHealth website. Regardless, with the vast amount of unstructured data being generated in healthcare, technologies like Autonomy have a bright future.

Rochester, NY based Logical Images is the company behind VisualDx. Understandably enough, it was founded by two dermatologists from University of Rochester in 1999. At its core lies a pretty big repository of clinical images (they claim > 60K) that are tagged with clinical metadata like symptoms, location, history etc. A user can browse through diagnosis domains or input findings to reach a relevant set of images.

There is a 15-day trial for non-enterprise users, and individual subscription pricing ranges from $49.99 (Pediatric) to $99.99 (Adult) for a year. Student discounts brings it down to a very reasonable level ($29.99 and $49.99 respectively) in my view. The company has leveraged their content to create an online consumer health resource called Skinsight. Besides the clever name, I really liked their slick ‘Skin Condition Finder‘ tool. Overall, it’s a smart approach for gaining market recognition and expanding services.

The concept of Diagnostic Decision Support (DDS) often comes with the implicit notion of text-based query, so I really like the ‘visual’ DDS approach. Dermatology, Pathology are obvious choices to implement such an approach and situations like ER, bioterrorism, infectious diseases are prime contexts since there is not enough time to engage a pure text-based knowledge base. Also, some conditions are rare enough that most practicing physicians and non-specialists haven’t actually seen one in real life. Other factors like dark skin, atypical presentation can be confounding. It’s handy to have a this as a medical education tool and as a care delivery aid in facilities where specialists are not available all the time.

I’m bit surprised why it’s not a widely subscribed resource in hospitals, public health entities, medical education and general EMR vendor products. Anyway, it’d be interesting to see this visual content embedded in other consumer tools (like iTriage, FreeMD) and enterprise offerings (Isabel Healthcare, SimulConsult, Curbside.MD etc.). Wonder which firms are competitors for VisualDx today…

At the core is a symptom-checker that lets users do a high-level categorization of what could be wrong with them, and how serious/not can it be. That functionality is nothing special, since there are multiple services that do that both online and mobile. Also included is a nationwide directory of care facilities (including urgent care, retail clinics, pharmacies and physician offices) and nurse advice phonelines (sorted by insurance provider). Your smartphone knows where you are, so the neat service is to be able to find directions to the nearest facility with few clicks. It may not sound like much; but when you are away from home, it’s pretty useful to be able to quickly figure out the nearest urgent care or retail clinic. iTriage is also trying to take this service to another level with the addition of ‘ER wait times‘ functionality in select cities. That way a user can further filter out the facilities based on wait times. Even if this is not really present/reliable in your area today, I think it’s a brilliant extension of a service like iTriage. Because its not a question of if this will be useful, but when. Once there is a critical mass of hospitals in an area submitting their ER-time feed to iTriage, the rest will find it hard not to do the same. There is a need for public to have a near real-time pulse on ER utilization, and services like iTriage provide a good platform for it.

The other interesting aspect is their partner service integration. Healthgrades provides the physician/hospital quality reports within iTriage, although users need to pay to get info beyond minimal basic report. Teladoc provides the 24×7 physician advise service (its phone consultations are for under 40$ mostly). They also partner with a claims adjudication organization (name?) that helps consumers negotiate thier medical bills. One can argue that a lot of the above functionality can be accomplished with a mobile browser or search app. But the value-add of a dedicated app is to provide fast, context-relevant info and be a steady companion as the user traverses the acute healthcare system.

I heard Paul Hudson, MD (co-founder) speak at the Mobile Health 2010 conference last month. He used an intriguing term to describe their technology: “symptom-to-provider technology”. Guess that is one way to think about a IT service that holds your hand from beginning to end of an acute care episode. He placed iTriage in the “multi-billion $ self-referral market”, and gave couple of interesting factoids about it’s current utilization:

• Around 89% iTraige members are commercially insured
• It’s use has gone beyond Emergency care in some cases- people have used the Healthgrades info to switch providers

Both of those underscore the fact that consumers need a trusted guide in handling how they interact with the acute care system. ERs, Urgent Care centers are the way most people experience care delivery most of the time, and having insurance doesn’t really help in making a good choice. With mobile phones getting smarter, the ‘self-referral’ market is ripe for innovative services.

FreeMD is the free online version of the same underlying triage application. It  uses video to conduct the interview, ask questions and then generates a custom web page that contains care instructions and suggested next steps. I tested it with a few hypothetical cases (from benign nose bleed to serious UTI) and it seemed to do fine for basic diagnosis. With vague complaints like diffuse abdominal pain and vomiting, it stayed roughly in the right categories at a high-level (appendicitis, pancreatitis, kidney stones, gallbladder disease, intestinal obstruction).

Is the underlying logic based on hierarchical structured programming or a more sophisticated expert system with forward and backward chaining algorithms? I don’t know. My interest in FreeMD spiked when I saw the 100K+ unique visitors/month statistic. They are consistently generating a lot of traffic, so there’s got to be fairly comprehensive content and at least some utility in the service. My personal impression is that as a patient-oriented triage tool, it does well. Of course, provider-oriented decision support is tougher and I don’t expect it to hold up like Mcyin or DXplain.

I’m also intrigued why DSHI systems chose to make their application available for free, when the revenue model seems to be based on licensing/co-branding with partners. I was half-expecting to see a feedback loop on FreeMD (like “Was this the right diagnosis? Let us know”) since one of the major reasons for open-sourcing anything is to leverage wisdom of crowds. But there isn’t anything like that, so maybe its all about gaining awareness and marketing the application.

Feb’10 Update: Connected with Dr. Steven Schueler after writing this post. He correctly identified that FreeMD is a triage system, so its a bit unfair to compare it with diagnostic decision support systems like Mycin/DXplain.

Dec’10 Update: FreeMD has a new look now. The user interface is much better and the interaction is much more easy (handy pain scale, descriptive pictures for example). Some interesting new functionality has been added too:

• Triage results now provide more information: FAQ, video explanations, and images
• Best options for care included in results, like ER, Urgent Care. Especially interesting are newer venues like eVisit and Retail Clinics. Hovering over each option gives some useful information like average charge expected at that venue. Very cool.
• The final ‘Triage Report’ can be printed or copied to clipboard.

These are all steps in the right direction. A more integrated future (like export capability to Google Health Record, or ER wait times from iTriage, or provider rankings from Vitals) would definitely establish FreeMD as a viable online triage destination.

At the heart of the model are a set of ordinary and differential equations that represent the physiological pathways relevant to diseases and their complications. The ‘variables’ in this model include signs, symptoms, patient behaviors (including adherence), provider behaviors, provider performance, encounters (e.g.  ER visits, office visits, admissions), protocols, guidelines, tests, treatments, etc. Basically, it tries to incorporate all aspects of diseases and healthcare system that are needed to analyze downstream clinical outcomes, utilization and costs.

A more loaded one-line description of Archimedes (taken from his original paper in 2002): “It’s an object-oriented, continuous-time, full simulation model for addressing a wide range of clinical, procedural, administrative, and financial decisions in health care at a high level of biological, clinical, and administrative detail.” Phew. I’ll confess that I don’t know what exactly is under the hood. But I know enough about the informatics field to believe that this approach is viable and very exciting.

This YouTube video explains how the model can be used to run virtual clinical trials. Kasier has already backed the findings of Archimedes to change their diabetes care delivery.  I think there are fantastic, unlimited opportunities for applying such a fundamental model to medicine- personalized health predictions, public health, health policy, cost-effectiveness and what not.  As a startup, they are doing fine. With an impressive list of partners/clients, and a $15.6M RJWF grant (2007), they have a good runway and momentum. They have all the right ingredients to be a change agent for next-generation Healthcare IT.

Jan 2011 Update: The FDA and Archimedes entered into a research agreement to understand the benefits of weight loss compared to the long-term risks of cardiovascular outcomes in patients treated with weight loss drugs.

Given the fact that Adverse Drug Effects (ADE) are one of the leading cause of death in the US, there is significant market opportunity in consumer education and support around it. Couple of nifty features that I like:

• Direct meds import from Healthvault
• Community Knowledge Base – an aggregation of information and experiences from people who are on multiple drugs. This feature is currently in private beta, but I think that it can be a great revenue opportunity once it gets some traction in terms of number of users. There are plenty of pharma companies who would pay good money for getting early (even though informal) insight into side-effects, efficacy, interactions, and usage patterns of their drugs.

Seems like they are planning to integrate with DestinationRx and Polka, which is a good idea since the traffic from those sites will already be primed for the services that PharmaSurveyor provides. The advisory board has some significant names, including Barney Pell, Matthew Holt and Mark Musen.

They do have some interesting marketing techniques like analyzing celebrity cocktails, not requiring registrations, etc. Business model seems to be only google ads for now, which is no surprise given their research-oriented background. The site is more a proof-of-concept for the underlying technology (although I’m not sure what it is exactly). It’ll make a lot of sense to integrate this service with commercial CIS offerings, and take it one step beyond just using RxNorm.

Aug’10 Update: PharmaSurveyor service changed its name to SurveyorHealth. Now they also offer mobile apps for iPhone and iPad.

In case you didn’t know, Google.org is the philanthropic arm of Google, and it was formed with the commitment of 1% of Google.com’s profits to address some of world’s most urgent problems (read the famous 2004 IPO letter by Larry and Sergey where they mention it). The site humbly admits that the Flutrends system is experimental. Nevertheless, it’s impressive that in some instances Flutrends was actually predicting flu before CDC.

Of course, not all people who search for flu have flu necessarily, but the power of this analysis comes from the coverage and promptness, not the granular accuracy. The basic idea of harnessing the collective thought (a.k.a. search needs) of the population to predict/monitor health events is fantastic. And this is just the beginning, IMHO. When a population is connected real-time and discussing what they think/want/need, abstracting that information can yield powerful insights- not just for prediction and monitoring, but for most aspects of healthcare (diagnosis, prognosis, news, followup etc).

The concept is applicable to domains outside of healthcare too. Take twitter for example. Twitter is another platform with mass adoption where people are having real-time conversations about what they are thinking/doing. Just look at what intelligent twitter mashups did for getting real-time snow report of the Feb’09 storm in UK or the Dec’08 Ice Storm in New Hampshire. There are health related examples too- the feb’09 salmonella-in-peanut-butter recall could be tracked promptly on a Twitter feed (btw, this slideshare presentation by PF Anderson at the University of Michigan explains ‘Twitter for Health’ in detail. Thanks to Christine Gorman for the link).

Healthmap extracts real-time information from more than 10,000 sites every hour and text mines them for disease and location patterns using bayesian filtering. The interface is clean and intuitive mashup with google maps. Links to the source of alert and a ‘heat index’ (composite score for each incidence based on things like recency of alerts, number of sources etc) are provided.

There are limitations too-  dependance on other sources, unstructured text, lack of integration between sources, not comprehensive, etc. But if you think of it as a free resource that supplements existing public health systems, its a great asset for general public and clinical professionals.

SimulConsult has an interesting logic behind it. Roughly speaking, its knowledge is derived from a ‘computational wiki’ that is restricted to physician users only. So its database is not only open for viewing, but users can submit modifications to the database. The system performs bayesian pattern matching and also considers temporal information like the age of onset and disappearance of each finding for each disease under consideration.

According to Dr.Segal, there are about 33,000 data points (disease findings) in the system and the future expansion plans are to include more of metabolism and genetic diseases. Interestingly, the wiki approach for knowledge gathering in healthcare is becoming more common (AskDrWiki, WikiDoc, wikiMD to note a few- more on them later).

Feb 20, 2009 Update: Got an email today announcing that those who contribute information to SimulConsult database will now get paid for their contributions. As they are adding sponsored links in the software, I guess it makes sense to share the wealth. More info here.