Regulating Insurance Overhead-Is it really a good thing?

Posted on December 15, 2011 by

We here at PrescribableApps try to stay as up to date on new developments in our country’s evolving healthcare system as possible. However, it can be pretty difficult to stay current on everything while getting a startup off the ground. Developments pertaining to new payment models are of course particularly interesting since we see ourselves fitting into any models that reward for quality of patient experience and improved outcomes (like BCBS MA’s Alternative Quality Contract program, PDF). Curtailing health expense growth to match the consumer price index (inflation), is a brilliant solution to reduce the absurd double digit percentage increases in premiums we have been seeing over the last decade. There are many reason these models are a good thing, among them the fact that our solution will have limited distribution and a short life if there is no financial incentive for health systems to reduce their version of ‘billable hours.’

I’ve been reading a number of articles lately about the recently finalized HHS rules (PDF) regarding the Medical Loss Ratio (MLR) provision in the Healthcare Reform Bill (HCR). To see what I’m talking about, see Kaiser Health News’ recap, a Forbes oped, Bloomberg or the WSJ Health Blog, or a particularly good review on the Health Affairs blog. For those of you just looking for a brief summary of the MLR provision, here is what the Kaiser Family Foundation published on the matter in their more comprehensive HCR summary (PDF):

  • Require health plans to report the proportion of premium dollars spent on clinical services, quality, and other costs and provide rebates to consumers for the amount of the premium spent on clinical services and quality that is less than 85% for plans in the large group market and 80% for plans in the individual and small group markets. (Requirement to report medical loss ratio effective plan year 2010; requirement to provide rebates effective January 1, 2011)
  • Establish a process for reviewing increases in health plan premiums and require plans to justify increases. Require states to report on trends in premium increases and recommend whether certain plan should be excluded from the Exchange based on unjustified premium increases. Provide grants to states to support efforts to review and approve premium increases. (Effective beginning plan year 2010)

Let me begin by saying I have not read the full MLR statement, and am not a legal expert, so I may have missed something in the actual provisions. With that out of the way, let’s take a good, 5th grade mathematical look at the provision. While I love the idea of receiving a tax-free refund from my insurance provider, I am pretty sure that percentage limits are not the best way to go about regulating costs. Sure, the provision states that any increase in premiums needs to be justified, but how exactly does this really go about saving the consumer money? It seems like there is a very simple loophole here: 20% of $5 BN is less than 20% of $6 BN. If insurance companies find they need to increase their bottom line and their administrative budget, what’s really there to stop them from

Average medical loss ratio is already ~80%. Does this really change anything?

Average medical loss ratio is already ~80%. Does this really change anything?

just paying more for medical procedures, justifying an increase in the premiums charged to patients? This most certainly looks like a ‘bomb’ ready to go off. Under this regulation, solutions like the one we provide would still be covered as a medical expense and would factor into the 80%, but why would they want to pay if they decide they do need to increase costs? (Image to right excerpted from 2009 Main Street Alliance report PDF).

If anyone has anything to add to this, please do. It would definitely help me sleep better at night!

Posted in Health, mHealth | Tagged , , , , , , , , | Leave a comment

Impressions from my first mHealth Summit

Posted on December 13, 2011 by

Inevitably, interest in the mHealth field has continued to grow, with over 3000 individuals attending this year’s mHealth Summit at the Gaylord Resort in Maryland. This provided a great resource for networking with some of the more inspired and knowledgeable minds in the industry, but also led to a lot of noise from both big players and small companies that haven’t thought through some of the business basics. As one individual put it, “There was a bit of whining about getting the government to force large corporations to form strategic partnerships with smaller organizations.” A hands-on government approach to establishing partnerships is something none of us truly want. What they are already doing should be enough, which is giving SBIR/STTR grant preference to companies that have a plan for obtaining strategic partners once they have validated key hypotheses. That, and HCR.

While there is some truth to the fact that small startups in the space flounder without help from the larger players, there were plenty of examples of small companies plugging away and slowly building some momentum by creative means. The value propositions need to be clear, and as Michael Skok informed a group of startup CEOs last night, the need for any new solution must be urgent since its adoption will require the reallocation of already-budgeted funds. As a startup ourselves, it seems the best way to find success is by starting with some bottoms-up adoption (consumers, small practices) and begin conversations in parallel with the enterprise organizations to identify what the true pain points are. Enterprises move notoriously slow, so start the conversations early and bring them evidence from the small pilots when you have it. If this doesn’t get them moving faster, you may need to rethink your strategy. This will require taking a hard look at your hypotheses and assumptions.

Hypotheses. This word brings up another point: the lean startup. This is a movement that has been gaining momentum in the last couple of years with evangelists such as Steve Blank, Eric Ries and Ash Maurya, to name a few. It is past time for more health IT startups to be applying lean methods to product development. If Intel of all people can get on stage and say they are developing new solutions using lean methods, then the smaller startups should be able to do the same. Maybe this is why they are having such a hard time finding strategics–they have a solution looking for a problem and have not done the proper customer development and discovery to understand where the real pain is.

Hopefully the accelerator model that has been taking off for this sector (Blueprint Health, Startup Health, Rock Health and Health Box), will help startups show clinical efficacy and start driving the big guys (enterprise employers, payers and providers) to adopt and invest in new solutions. Hype is good, as Health 2.0 has proven over the last few years, but over-hype could be a bad thing though if there is excessive private investment in companies that tank. The announcement that HealthTap closed an $11.5 MM Series A during the event after a $2.35 MM seed round earlier this year is one such example of a potentially over inflated company. Don’t get me wrong, the idea is a good one and the investors behind it bring a good deal of quality experience, but it is still unclear that this has revenue potential to sustain the expectations of such a large raise.

Finally, I attended the morning discussion group about mental health. We were all impressed by the size of the room we were given to discuss this pervasive set of illnesses, but there were only about 15 of us in attendance. It’s a good start, but anyone with a grasp of the impact these illnesses have on our society should be astonished there weren’t more people at the discussion. It is a bit of a stark reflection on how we are only just beginning to fully appreciate that impact. And one thing that came up a few times was the lack of solutions for behavioral disorders on display at the event. Most are merely pushing for behavioral change without addressing what is the root of many chronic illnesses, which are poor lifestyle choices often due to a comorbid psychiatric disorder. If all goes as we expect (or at least close), our solution will address these issues and help millions.

For other good readings on the event, also check out Chilmark Research, Healthcare Scene, and Venture Valkyrie. And here, a TED talk that everyone from the conference should watch. Have a great day everyone!

Posted in Health, mHealth | Tagged , , , , , , , | 1 Comment

Reflections on Health 2.0 San Francisco 2011

Posted on October 2, 2011 by

Just finished reading a particularly good recap of the Health 2.0 conference written by Chilmark Research’s managing director John Moore (@john_chilmark) and have been inspired to write my own piece. In the recap he goes into the strengths and weaknesses of this year’s conference relative to the past four he’s attended. As this was PrescribableApps first time attending, I will be focusing more on a first timer’s experience. Main takeaway from the conference: don’t go looking to learn much during the 3 minute demo sessions, the meat is in between sessions and outside the conference halls when you get to talk with those innovating and hear about their experiences. Definitely recommend the Chilmark article for a more in depth breakdown on companies.

Overall I thought it was a very interesting and informative event because of the people it brought together. We all have a pretty strong grasp of how messed up the healthcare industry is right now, and are all trying to improve it in some way. My first exposure was during volunteer training before the pre-conference sessions, where I was able to meet some of my peers, and that went really well. Most of us were volunteering because our companies are too young to justify the expense, but some volunteers were students or just curious about the industry. We were there to learn and make new connections.

After training I attended the Doctors 2.0 preconference session. The one piece of wisdom that really got driven home for us innovators was this: while doctors care about their patients, they won’t adopt a new and improved technology if it doesn’t simplify current processes or make them significantly more money for the inconvenience of using the technology. For demos, I did like what Avado and HealthLoop are trying to do by advancing care beyond the clinic. Met with both David Chase (@chasedave) from Avado and Jordan Shlain from HealthLoop after the preconference and was impressed by their respective grasps of the industry and how they intend to change care delivery by extending the reach of doctors. May have to recruit David as an advisor when we get to the point where we need a good sales strategy. Guy knows what he’s talking about.

The next couple days were the heart of the conference and we saw a lot of demos from new and established companies alike. Most demos were so-so, with many still missing the mark IMHO. Reportedly UI is improving, but still saw a number that made me cringe. Also, too many Facebook/Zynga wannabes. Gamification is understandably all the rage, but there are just too many ‘me too’ companies developing rewards-based wellness platforms for the worried well. Yawn.

One of my favorite talks was on Day 1, by Mark Smith from the California Healthcare Foundation. It was one of the few talks that had any kind of industry self-reflection and critique. The most memorable moment was when he made a comparison between the underpants gnomes from South Park and us young startups in the healthcare space. This was one of the only negative, reality-check comments made about the industry during the whole event.

Collect underpants, ???, Profit!

What's your Phase 2?

Too many young companies have an idea that they expect to make plenty of money, but skip the important phase 2. The allusion was referenced many times with good reason–too many of the companies have no clear revenue model and there are plenty of bodies piling up in the industry because of this. Apparently new companies are catching onto this fact and reacting accordingly, but I still saw a number of them that had no clear plan. And this is the most worrying thing about the hype machine that is Health 2.0: these companies are getting investors. Not sure what’s going on behind closed doors, but investors are dumping money into questionable companies. Some will succeed, many will not, and the backlash from the hype could hurt the industry more than it helped in the first place. Not looking like this will happen anytime soon, thankfully, but something to keep in mind if you’re an investor (or startup): don’t buy (build) the hype, be sure you are buying (building) a real vision!

Alexandra Drane (@adrane, blog) introduced the Unmentionables talk with a great presentation about how much emotional and mental wellbeing impact quality of life both at home and at work. She was an engaging speaker and need to see more discussion of these topics nationally. Mental wellbeing is now discussed in an offhand way, but unfortunately employers and our society as a whole do not take enough action. We really do need more discussion of our ‘unmentionables’ and I hope she keeps being asked to give these talks. Just one eye opening slide below (all the others were unfortunately too out of focus):

Health risks of 'unmentionable stressors'

Health problems increase significantly when just a few common life stressors occur together.

Posted in Health, mHealth | Tagged , , , , , | Leave a comment

Antidepressants are everywhere these days!

Posted on August 16, 2011 by

The journal Health Affairs recently published a study about the widespread prescribing of antidepressants without a psychiatric diagnosis that has been gaining quite a large amount of media attention (WSJ, WebMD, LA Times, CBS, PsychCentral). While some of the reporting is fairly on point such as the LA Times piece, other commentary, such as that found on the CBS page, is a little less useful and more inflammatory. Furthermore, some of the articles seem to have their numbers jumbled. I think it is about time we weighed in on the matter after taking a week and a half to reflect on what this all means in the bigger picture–specifically that by and large, mental health care is still not being delivered properly to those that need it. This is despite massive leaps in the scientific and biological understanding of disease over the past few decades.

The Study
Dr. Ramin Mojtabai, an associate professor from Johns Hopkins, and  Dr. Mark Olfson, a professor from Columbia, wrote the Health Affairs study that looked at prescription outcomes from 233,144 office visits to non-psychiatrists participating in the 1996-2007 National Ambulatory Medical Care Surveys. By crunching the numbers, a few startling conclusions emerged:

  • Non-psychiatrists currently write ~80% of all antidepressant prescriptions
  • Over 11 years, approximately 6.21% of all visits to non-psychiatrists resulted in a prescription for an antidepressant (see chart for change over time)
  • Primary care physicians were roughly 2.6x more likely than other non-psychiatric professionals to prescribe antidepressants (9.3% v 3.6%)
  • Of all antidepressant prescriptions written outside a psychiatrist’s office, by 2007 only 27.3% of visits resulted in a psychiatric diagnosis, down from 40.5% in 1996, despite increased prevalence of antidepressant prescriptions during that same time period.

Changes in antidepressant prescription behavior among non-psychiatrists 1996-2007, from Health Affairs 8/2011

While the data set is by no means complete—there are serious limitations to the information the surveys capture—this analysis highlights the trend of non-psychiatrists using antidepressants in situations where it is unclear if they will have an effect. The specific levels of antidepressant prescriptions is not so worrisome—an estimated 25% of our population suffers from a bout of clinical depression annually. It is the fact that so many of the prescriptions written come without a psychiatric diagnosis for follow up.

What this all means
Unlike many vocal pundits out there, I am a firm believer that if administered properly, antidepressants and other psychiatric medications can have a hugely positive impact on the lives of those living with behavioral illnesses. There is plenty of evidence showing that depression and other behavioral illnesses have a biological basis, so medications developed to treat the underlying pathology should be effective. As we are still learning much about this underlying pathology, medications are often not enough on their own. Furthermore, it can often take a long time for a proper regimen to be established, but I have witnessed how finding the right mix of medication and behavior modification can dramatically improve an individual’s quality of life.

However, antidepressants are now the 3rd largest class of drugs by sales numbers, and this study is showing us that of ALL antidepressant prescriptions written, only 42% are being given with a concurrent psychiatric diagnosis. There are a few explanations for this over abundance of prescriptions without diagnosis, none of which are particularly comforting.

  1. Antidepressants are increasingly being used for off-label indications. A 2006 report showed how many psychiatric medications (not just antidepressants) are being prescribed for off-label use. ‘Off-label’ is when a doctor prescribes a medication for an indication not explicitly listed on the FDA approved indications list. There is little evidence for most off-label uses, and unfortunately this could explain a significant portion of the undiagnosed visits, especially if people are just going in for general anxiety and stress-related depressive symptoms, which most antidepressant medications aren’t particularly good for.
  2. Patients are asking for their doctors to leave a specific diagnosis out of their medical record for employment and insurance reasons, as suggested by an interview we conducted with Dr. Stuart Gitlow, an addictions specialist and occasional Harvard lecturer. Dr. Gitlow finds that many of his patients seeking treatment want to keep it private for fear of discovery by an employer. He specifically lists government positions as being difficult for a patient to hold if they have a prior psychiatric diagnosis.
  3. Patients are exclusively seeing PCPs and other non-psychiatrists for financial reasons. We have conducted a number of market research interviews, and many of the patients we spoke to not seeing a mental health professional seek medical services from their PCP, specifically for psychiatric medications.
  4. Stigma is still a huge problem in our society. When I read some of these reports, which are for the most part pretty unbiased, there are still indications of stigma and a lack of understanding around the actual psychiatric illnesses. And as someone with a number of friends and family that live with such illness, it can be very difficult to remember that it isn’t just ‘all in their heads’ (although literally, it kind of is). This social stigma may be pushing patients to stay out of a psychiatrist’s office, despite a need for more professional services.
  5. Poor charting and interoperability across different practices. WebMD’s coverage of the article is probably one of the best I’ve read. They speak to a number of psychiatrists, and one of the problems they address, which was not noted elsewhere, is that this is also possibly just the result of bad charting information. It is entirely possible that a large number of diagnoses were made elsewhere and just don’t happen to be in the chart with the PCP/non-psychiatrist.
  6. Misdiagnosis of symptoms. As the Wall Street Journal points out here, there are many illnesses that can actually present as a psychiatric illness (article includes a list of such illnesses). While this is one possible explanation, the premise of the article is that individuals are being misdiagnosed with psychiatric illnesses when it might be something else, which would NOT explain the under diagnosing of depression, etc.

That’s a lot of potential explanations, and I could have kept going! Regardless of the reasons, if 80% of antidepressants are being prescribed by non-psychiatrists, then it is likely that a good number of those individuals are not receiving proper follow up care.

It is safe to say this is certainly a problem that needs to be addressed. PCPs are not trained to be mental health professionals and cannot be expected to administer the full level of care necessary to treat behavioral illnesses. There is already a national shortage of proper care providers for mental health, and these numbers indicate the problem is only getting worse. We hope that our platform will provide the non-psychiatric professionals with better tools to follow up with their patients and make more informed decisions about treatment. As this is clearly becoming more and more something in their domain, it is important that they be better equipped to deal with at least some of the responsibilities traditionally reserved for specialists. We could also aid in the process of passing care over to a mental health professional by better monitoring how a patient is responding to their new medication. More on that as we get closer to having a true product ready for launch.

In conclusion, these numbers are all based on visits before the recent recession. I am very curious to see what a more updated list would look like after all the recent turmoil. I know I went through what could have probably been classed as clinical depression when I couldn’t find work for a few months, so can only imagine what it must be like for those with very specific skills in struggling industries. Clearly, we as a society are failing if we can’t ensure individuals receive proper care, which is the one thing this article makes very clear.

Then there is the connection between socioeconomic inequality and higher societal incidence of mental illness. Could the fact that our country is currently the most economically unequal it has ever been also have a bearing on these numbers? That is an issue for an entirely different post. This one is plenty long already. Thanks for reading!

Posted in Health | Tagged , , , , , , , , , | Leave a comment

Why I am Loving Unbounce

Posted on August 9, 2011 by

I have to make something very clear at the beginning of this post: I am not a developer. I am interested in learning how to write code as I think it would be a very practical skill, especially given the company I am starting right now, but I have no background other than a life long appreciation for what finished products look like and what they can do. I can see how new developments in technology can be applied in everyday life, but until a year ago I was a molecular biologist by trade and training. With this in mind, this is my relatively uneducated opinion based on a number of frustrating experiences over the weekend.

WYSIWYG

As a relative noob in this arena, I decided that it would be best to use a ‘What you see is what you get’ (WYSIWYG) tool to help our main developer with some of the front end, freeing him to focus on the backend. This would also allow us to get something out there a bit sooner and start working on getting some interest in our platform so that when it is ready, there will be people ready to test it. From my understanding, this was a relatively easy thing to do.

I am a bit of a perfectionist at times, and as such, was grossly misinformed.

I spent a majority of the weekend researching and testing free and free demo versions of paid WYSIWYG builders. This was incredibly time consuming and, as it turns out, fairly fruitless. There are a lot of good tools out there, but a number of them are pretty limiting from what I could tell, or just not intuitive enough and therefore appeared limiting. I needed something simple, customizable and quick.

‘Why not just hire someone to do the website for you?’ you might be asking. That is definitely in the cards, but we are trying to stay true to lean startup methods as much as possible and bootstrapping right now, so the money would be better spent on other matters if we can do most of it on our own. Also, I feel it is my responsibility to help on the development end where possible. We will eventually be hosting the site on our own server and therefore we also don’t need much of the hosting options that are offered with the sites I tried out.

4,660,000? That is just way too many!

For research, I Googled ‘WYSIWYG editor’ which apparently is way too vague. Eventually I found a few good review articles that tested many editors. I tested Freeway 5.5, Webs, Weebly, Moonfruit, looked into Drupal editors, iWeb and one or two others that I am blanking on right now. I liked how Flux looked, but couldn’t find a demo version. I was going to test BlueGriffon, but by that time I was just tired of looking around.

Weebly and Webs were a bit frustrating as they did not allow me to edit certain elements of the page to my liking. The templates were decent in all the above, but only Moonfruit had one I really liked and allowed as much control over everything as I was hoping. Too bad it is entirely Flash and there are tons of complaints about translation into HTML for mobile browsers. I tested the site I had been developing on my phone and was NOT pleased with how it looked.

Once again, these are all good editors, and I think I will purchase RapidWeaver when the time comes for us to do a bit more fine tuning and cleaning of the HTML code. Weebly, Moonfruit and Webs all provide hosting and paid versions include credits for Facebook Ads and/or Google Adwords. They really are a good deal, and I have to say Moonfruit was my overall favorite, but the Flash thing was a deal breaker.

Unbounce is a service for rapid A/B testing. For those of you unsure of what this means, it is quickly developing and testing page variants to help decide best marketing strategy based on visitor conversions. The Obama campaign used this strategy (doubt they used Unbounce) to optimize their campaign contribution page. I had already signed up for a free account for our basic lead generation page, which can be found here: PrescribableApps. This was entirely based on one of their templates, and I think it looks pretty good. Not exactly the professional feel we were going for, but a solid start for lead generation.

If you click the ‘free account’ link above, you will see that the monthly rates for Unbounce are a bit higher than Moonfruit or some of the other sites listed previously. Not entirely sure why this is, but I am guessing it has something to do with the ability to create UNLIMITED subdomains and A/B split tests. As an early company trying to find leads to test a closed alpha, this is a great feature made particularly easy by the great marketing templates they provide users and the marketing tools provided by their integration partners. I personally feel this is a great hosting tool for any budding enterprise, but it would be quite nice and reasonable to expect some Google Adwords or Facebook Ads credits like other sites since this is a marketing tool and all.

However, the focus on marketing is a little limiting if you want to try to quickly build out something resembling a full website. The templates are not at all configured for a standard homepage, and therefore require that just about everything be done from scratch. Since we were already using the service for our homepage, I decided to play around a bit with the editor and see what could be accomplished. As you might be able to tell, I was very pleased. If only I had done this sooner, I would have saved myself a lot of time and frustration (although I did learn a lot from the process which is valuable).

This is (obviously) still a work in progress, but I am working on changing our homepage to look a little something like this. This website was entirely created using Inkscape (an Open Source alternative to Illustrator) and the editor tools in Unbounce. Granted, if I were a designer with HTML/JS/CSS skills, this wouldn’t be very impressive, but for a few hours work I think I did a pretty good job.

Workarounds

One of the significant limitations I realized is that there is no real JS or CSS editing possible. I tried a few times and found it very difficult. One way around that is to use the custom HTML boxes to include snippets of JS, as I did here with the Twitter box. The way around not having a standard CSS and master template for all pages is through the ‘Move to Another Page’ option. This allows the user to shuffle an entire variant to another page that is already set up and make the new variant the champion if they so desire (read: create new page first!). All the user needs to do is then remove unwanted elements and add new ones specific to the new page.

'Move to another page'-easy as pie

Another issue I encountered with Unbounce was a complication with ‘link color.’ The master theme will set all links on the page to appear as the same color and because of the colors I used, this just wasn’t going to be possible. Instead of starting over, I decided to make ‘buttons’ out of the links in the top right and then any links on the rest of the page were then the same color. Buttons are fully customizable and can be made transparent, so they look like normal links until closer inspection. This also made it possible for me to make the nav bar a little different than just a row of buttons. The standard templates have gradients in them, but they don’t come standard as an option in the blank template.

Finally, Unbounce supports the use of different forms for your conversion. There can only be one ‘form’ per page, but the options are comprehensive enough. I would include a screenshot of the form selection menu, but all my pages already have forms and I’m a little weary at this point. You’ll just have to see for yourself.

I’m not entirely sure if this is an entirely legitimate use of the Unbounce platform, but I looked through the TOS and it is certainly not forbidden. Granted this is also something that will only be hosted on their servers for a couple months while we build out our own site. And they claim to love entrepreneurs, so I don’t think they’ll mind too much.

Please let me know if you have any questions. I will answer what I can.

Posted in Development | Tagged , , , , , , , , | Leave a comment

WellDoc is Back in the News and That’s a Good Thing for mHealth

Posted on August 5, 2011 by

Over the last couple days I have been thinking about a post regarding the recent release of WellDoc’s 1 year clinical trial data. In case you missed it and don’t want to follow the jump, the study showed that patients using their DiabetesManager application for a year were able to maintain a decrease in A1C levels of 1.9%, whereas individuals receiving treatment as usual (TAU) only showed a decrease of 0.7%. Every 1% decrease in A1C levels reduces the risks of complication by 37%, so a 1.9% reduction is quite a feat. This is huge news for them, and it’s clear from their media push that they are pretty excited about it. The company has done a good job staying in the headlines without the study results, but the reason this specific study has been garnering such media attention is because of its broader implications for the field of mobile health (mHealth).

Diabetes

While I am no diabetes expert, I do have a general understanding of the disease and know how to use PubMed thanks to my former life as a laboratory (bench) scientist. As many of you may know, diabetes is quite an epidemic in our country—good ol’ USA—with 8.3% (25.8 million) of the population living with the illness, contributing to over 230,000 deaths a year. If left untreated or improperly cared for, the illness can lead to heart disease, high blood pressure, blindness, kidney disease, neuropathy, and amputation. Not only are there severe emotional costs to patients, friends and family, but medical expenses for treating diagnosed patients and known productivity loss are costing the American economy close to $200 billion annually. Clearly something needs to change, which is what WellDoc hopes to accomplish.

Will this prove to be the future of improved diabetes care? Emoticons? Hard results to argue.

WellDoc DiabetesManager is an attempt to improve the provision of care while also helping patients to understand their own illness. WellDoc has managed to create a system that can coach patients individually and provide doctors with clinical decision support based on patient reporting of glucose levels and dietary intake. Based on their two clinical trials, this approach is working, as A1C levels have been decreased significantly relative to TAU. Dr. Jon Shematek from BCBS: “Improved care coordination is a significant factor in gaining better patient outcomes…Ultimately, by making lasting improvements in care coordination and quality we can better control costs for the health industry as a whole.”

Care Coordination

This is the key phrase here: Care Coordination. In the case of WellDoc, coordination of care is achieved by increasing the feedback to health providers through improved patient self-tracking. There is a significant push these days to interest patients in their own care, and if the government initiatives to drive ACO creation are successful, this will become more of a priority as there will be financial incentives in keeping patients healthy outside of the care center. I am a member of a group here in the Boston area called the Collaborative Care Technology Working Group. The lead faculty member is Dr. Allan Goroll, head of primary care at Harvard Medical School, and it is a personal mission of his to increase the prevalence of proactive health management through technology that improves collaboration between doctors and patients. This is not always easy, as we are learning, but there are strides being made in the right direction, but we are all very excited about what a further push towards capitated care models could do to drive innovation in health. The WellDoc study is validation that what we believe is actually possible.

On the other hand, there are some individuals already very keen on tracking their vitals that don’t need a doctor to assign them the task. This is not sustainable long term as health conscious individuals tend not to be the ones driving the majority of healthcare spending, but it is a move in the right direction. I had the pleasure of attending the Boston Quantified Self Meetup yesterday evening and saw some examples of these consumer facing applications. Mike Nagle had assembled a great panel of speakers including Joe Kvedar from the Partners Center for Connected Health (PCCH), Paul Wicks from PatientsLikeMe, Jacqueline Thong from Ubiqi Health, and Rick Lee from Healthrageous. Each of these companies is taking a different approach to having individuals track their own vitals. Healthrageous uses gaming mechanics for wellness. Ubiqi uses mobile devices to monitor migraines. PatientsLikeMe gives patients a forum to discuss their experiences, and PCCH is opening the door for clinicians to provide just-in-time care. However, PCCH was the only example of a company that explicitly engages doctors and patients in a coordinated care situation. While fairly straightforward in their approaches, each company has had very different experiences with their users, which struck me as important. Keeping a closer eye on what works instead of just plowing ahead and blaming the patients as some companies have done will be the key to identifying a way to build better patient engagement in health.

Dr. Kvedar made an interesting analogy to the speed monitors on the side of the road that alert a driver if they are over the speed limit. Health tracking is a bellwether to better understand what contributes to health or disease. How much of the WellDoc effect is simply better knowledge and awareness? There is certainly something to be said for a more informed care provider. The initial 3-month clinical trial that WellDoc conducted saw 84% of health care providers in the trial arm titrate their patients’ medication. Only 23% of doctors in the control group made any such changes. This indicates a dramatic increase in clinician understanding of a patient’s illness. On the other hand, how much were these decisions influenced by patients having a better understanding of their own disease?

This is something that will hopefully be investigated. However, the point is that true care coordination is a many-faceted goal requiring engagement from a number of participants on the side of the care provider as well as a true desire to improve one’s health on the patient end. A lot of this can be achieved by better education, and as the other John Moore—the one from Chilmark Research—stated, patients are ready to take a more active role in their health. They just need to be prompted by their caregiver and informed of the benefits achievable through using a system to track their health (usability is a large deterrent, but relatively easy to overcome with some quality thinking on the part of designers). People don’t enjoy being sick!

A few final thoughts, mostly about trial design…

We at PrescribableApps are working to develop a platform that in theory has many similar components to WellDoc. We will be using care providers as our distribution channels. We will have patients monitoring symptoms and behaviors relevant to their illness. We will hopefully (fingers crossed!) be reimbursed by payers. With such similarities, we are thrilled that WellDoc has been able to reproduce their initial results and show that after an entire year and with a larger patient population, the outcomes still hold. Our solution will initially be dealing with behavioral health, however, so our outcomes won’t be quite as easy to quantify as A1C counts are with a disease like diabetes.

The actual study is yet to be released in its full form, but is scheduled for the September issue of Diabetes Care. I will likely look again at the entirety of the trial at that point as I am still an investigator at heart and like to know all about trial design before accepting what I am being told. I am curious to see if there is any quantifiable measure or even mention of the additional amount of time each care provider spent using the application and interacting with patients. I could see the increase of communication alone having quite an impact on health outcomes, even without the coaching.

I will like to see the attrition rates and use regularity among participants. A year is a long trial, but I imagine there were some added incentives around continued use by both doctors and patients that may not be present in the wider market. Is there any indication from their dropout rates that this could be an issue?

I would like to see what the trend charts look like regarding A1C levels. The 3-month proof of concept trial showed a decrease of approximately 2% and 0.68% (test group and TAU, respectively), which are almost identical numbers. Is there a long-term benefit to using the WellDoc DiabetesManager? Or are 3 months all that is necessary? My money would be on improved maintenance of care, but it is entirely possible that the initial 3-month period allows the doctor to get the medication right and that is the only change to care we are seeing reflected in the results. This is given a little more credibility since the TAU groups were essentially the same in both trials.

Finally, what are the liability issues around this type of medical utility, and how is WellDoc addressing them? And do they think filing for FDA approval was absolutely necessary with the recently released guidelines? Since they already have FDA approval, when will they be buying/partnering with someone like Agamatrix or a Bluetooth enabled glucometer to cut out the manual entry step?

Check back regularly and I’ll eventually get into a full breakdown of the trial sometime in September. Either way, this is huge validation for the improvement of care through better doctor-patient interaction and communication. I am thrilled to be working on a project with similar goals, even if the execution is a bit different.

Thanks for stopping by!

Posted in mHealth | Tagged , , , , | Leave a comment