Help me with industry sign banking new york medical history
mmm my own oh yeah yeah that works little louder actually is it too loud we might paint at me we might be dealing do old all right but that's okay hopefully this is loud enough how's that it'll do I'll be off in a minute anyway hey Gary nice to you so that clock correct we get started 1130 welcome everybody my name is Aaron Cohen i'm professor and department medical informatics and clinical epidemiology i know many of you but not all of you so welcome this is our regular Thursday conference some Thursday conferences we designate as an IDL conference my mic is suckling a little bit which the IDL is the informatics discovery lab which is sort of an outreach collaboration initiative within the department and we bring in illuminaries from various industry positions to educate us on what's going on in the real world of informatics so today we have one of those luminaries dr. richard gibson who is a well-known colleague and friend of the department before i get into the bio i'm just going to say a couple of things i guess this is some talk is available for CME credit and anybody who needs to get CME credit please email lynn schwabe and she will give you the details but they can't take some kind of online test and she'll do that also we are streaming and recording so if you have any questions or comments during a Q&A period please press the little button on the black box in front of you so the light goes on so it gets recorded in strength okay anything else logistical I need to say that i forgot i don't think so okay so dr. richard gibson is our guest today dr. richard gibson md PhD is currently the executive director of the health record banking alliance and he's going to be talking about that today which is a nonprofit organization founded in 2006 to promote consumer centered consumer controlled comprehensive lifetime medical records dr. Gibson comes to us to from health record banking alliance previously to that he was in a position at Gartner Inc where he was the research director and previously he's held roses chief intelligence healthcare intelligence officer at providence health and services chief information officer at legacy health and chief medical officer at providence health systems for the oregon region he's an affiliate assistant so in the department of medical informatics and clinical epidemiology at ohsu his educational background includes a bachelor's science from Stanford and MD from case western reserve a PhD in medical informatics from the University of Utah fellowship in Intermountain Healthcare in salt lake city and an MBA from the Wharton School he's also that wasn't enough a retired family physician an emergency physician and I had to read all this because there's no way I could remember all that stuff so today doctor gives know people talk to us about health record banks and how complete data can lead to a better patient care and I am very interested to see what he says up to these days with that okay thank you thank you thank you for joining us so I was scheduled on december nine thursday to give you a talk on population health management really good talk but I've got snowed out then and then lapsed on I'm thinking oh maybe a groundhog day all over again for this talk but we're good and we'll get you out in time to avoid snowmageddon here in Portland so ah no no disclosures to make to the audience and in 1984 I finish my family medicine residency in Spokane and I owed the National Health Service a couple of years because they paid for a couple years of my medical school and so I move to forks Washington and when I came out to Fort switch about five hours north of here I was starting a new private family practice there and one of the doctors was retiring and so I met him on Main Street where his office was and he says on when I have a patient referred to you I said oh yeah and it's a young woman who's pregnant so happy to see her could I get a records and so he walks over to his death and brings back a five by eight card and this is just a facsimile so this is not real patient data and on the front it had her birthday her own birthday date and her birth weight and then a set of visits all on one line and then he flipped it over and again more visits through her adolescent years and positive pregnancy test referred to dr. Getz 1984 once you love to go back to those days I think it's ironic that I spent most of my career dealing with electronic health records and now I would not say that electronic records are in the beloved state that such a card was for some doctors back in nineteen eighty-four but I do think that there's hope to get back to that easier state with the electronic record is not thought thing in front of you in the exam room but supports your care and I think that health record banks are part of that so what we'll talk about today is the general setting for why we're using data and healthcare all known to you we'll talk about I'll review current status of health information exchanges lightly just to provide context for health record banks will go into health record banks will talk about a proposed business model we'll talk about blockchain just to slide on blockchain because they get to have so often and then we'll wrap up so that the challenge we have with health records is that if you're in the inpatient setting like an ohsu the patient has a complete record and not a challenge journey things everything that's happened to them at a weight issue and of course would care everywhere you can see everything that's happened in an epic facility across the country again at ohsu outpatient you have the same issue but for most of the care in America especially in the ambulatory environment doctors are progressing was only a part of the records that the patients have accumulated over their lifetime and so the problem is that we get under treatment over treatment and treatment with errors so Asher script survey showed that fifty-five percent of care was missing a complete history and that forty-nine percent of the time patients said that the doctor didn't even know what prescriptions they were on now that wouldn't apply to ohsu but we have a problem bigger than the challenge at only tissue so the case we're making that we need comprehensive care at the time of care in order to give good care to do quality to do policymaking and to understand what we're doing for our patients that means that a care has to be electronic to begin with so doctor is already on electronic health records so that's good and the challenges now we have scattered information over each electronic health record of consternation about how to get that information from the scattered hie spots to the place where the patient happens to be right now and oh by the way we also have dental records to do and we have home care and other kinds of interactions that people have with providers that need to be brought in as well we already have digital health care we have common standards and we need to find a place to pull it all together one of the challenges no mystery to you all is that we need care we need electronic health records for care at the point of care where they need to be immediate and need to have a rapid response time and we also need retrospective analysis care again for policymaking for quality analysis for population health management for understanding our outcomes and what works best now that requires that you get all the records together but it doesn't need immediate response time and you can vary even can be a delay and we'll talk about that why are we where we are well of course again no mr. to you all there's been a medical explosion in terms of data and specialties which means any given problem requires you to see more than one provider so that you have scattered records no individual brighter tend to have all their records unless you've gotten all your care at a given facility so there's an opportunity to manage records better we know that we need to engage patients and population health that we're going to get the best outcomes they need be motivated take care of themselves and be given the tools to do that and smartphones will play into this so 564 million dollars of the federal money spent starting 2010 to build state health information exchanges and HHS weighed in in 2013 said more will be needed and many of the hie s have already failed but some are working and we'll review some of those so the challenge is to help information exchanges is that there's an issue of trust in fact a lot of the time is spent with the attorneys building a distributed rights agreement to examine each other's data and make sure that the patients are on board with that and that requires a lot of stakeholder agreements that are hard to come by once you have it it works pretty well there's always been a problem with sustainability I just reviewed a book on interoperability just not a lot to say about sustainability what works and what's required and that we know that health information are expensive it requires a lot of coordination on IT a lot of 24 by 7 monitoring pretty challenging stuff we believe that the focus on health information exchange pulling together distributed data is the wrong approach and that what we're trying to do is to repeat what goes on in the paper world so provider I mean a patient comes to see a clinician with his or her electronic health record and right now with hie the idea is you go to the index and say where are the records stored you go out to each of those stores and you pull it back you assemble it you bring it back to the clinician who then sees the patient and puts the data back in the HR and a cycle goes through the problems are that it's real time record reconciliation you get multiple records come in at the same time and you have to avoid overwhelming clinician with too many duplicate data points and that's hard to do it also works best with a patient identifier nationally of course which we don't have there's always the challenge that you're going to miss data and you won't collect the full group of data you also have a problem with the data coming from the electronic health records being assembled and then moved on to the provider so for each data element it's being transmitted twice which means this is more of a security risk searching for research is more difficult because you have a distributed model and that so far we haven't found a new broadly applicable way to make this sustainable now this was a simulation paper where they put in a failure rate of being able to collect data from a node a node would be electronic health record of an office or a hospital that sort of thing and they set that at various levels but for this chart it's at failing five out of ten thousand that's a ninety-nine point nine five percent success rate most would say that's pretty good for retrieving it then they looked at well how many times would a patient be seen in a year and then how many providers is typical for patients to see in a given year nessa get older of course that number can go 7 10 and above so what they showed was in simulation that you're going to have a lot more failures when you have a distributed model then when you have a centralized model where all the data are in one place now the next case domini well why don't we just work where there's not all the data well Billy as mob who founded the health record bank proposed and wrote about it that really with electronic health records just having only half the medication list probably does it make it worthwhile in order to take the trouble to health information exchange you have to be reasonably assured did you have most of the data before those data and effort to get to it becomes worthwhile again an argument for putting all the data in one place so that one of the challenges of many in health information exchanges is that if a patient's going to record that they have concerns about their privacy or they have requests for it they're going to have to register that each organization across the country that has their record and then there's a chance that they forgot one or that they have to update it they have to go back and update them all because this is being pulled from each of those sources at the time that the patient seen at their next provider the challenge with annotated data or correcting data that's difficult when you have a distributed model so they can get back to age issue and correct problems that they see on the ohsu my chart or equivalent but it's hard doing that across the country and then you don't propagate the corrections although the next time they pull from you your record will send the correct record to that clinician across the state across the country right now mostly participation health information exchange is voluntary and it's troublesome they doctors have to pay interface costs and these are reasons why you don't get full participation some places around the country house but it's not widespread so I'm just going to review graphically some of those things that we know about health information exchange currently and this is the following slides are about a review done and turned in March of 2016 about the state health information exchange and now of course their private health information exchanges whose data may only be partially represented on this chart but the idea here is that this is turned in 2016 directed exchange you can see that most states have directors name that means direct trust that means like a secure email we have clinicians secure address and you send a note as needed or when we crossed it so most states have that interesting not I'd oh but then when we look at curry base exchange so the eHealth exchange or Sequoia or carry quality or common well then it's a it's a little different not so many place to that you see Oregon doesn't have a widespread query-based change except what we have with epic now this is looking at acute care hospitals participating in health information exchange so according to American Hospital Association the 54 acute care hospital four hundred Q care hospitals the United States doing about 37 million hospitalization and you can see the number of hospitals doing some kind of health informatics change has come up dramatically since these programs are put in place starting in 2011 but you still see that only about 2,100 hostels out of 5400 through 2013 were participating in health information exchange and most of that was query and all bench a lot of that is epic doing query to other epic facilities so the the dark blue is directed exchange the light blue is query now when we go to ambulatory facilities the situation reverses its still dramatically increasing good for that 23 230,000 physician offices 15,000 urgent care centers 925 million patient visits a year ambulatory and you see directed in the case of physician office way outweighs query and it's nice to see that going up but we still only have thirty thousand offices out of 245,000 sites participating so we're still a long way away from having a sort of coverage we need and again the details here aren't important well the dark blue represents 80 to 100 percent and we're talking about hospitals exchanging clinical care summary tears 2011 years 2014 well look at that look who's at the top of the heat for exchanging clinic here at summaries Washington Oregon now and kudos to us for getting that done we're doing a good job with that i tribute a lot of that the concordance around epic but it's still a good job by our state's and our clinicians to get that done now this is part out same report looking at how do you evaluate cumulatively health information exchange and what they've chosen to do is what they've chosen to do is say it can be represented by seven different facilities hospitals exchanging clinical care summaries with ambulatory providers and so and the dark blue was what was the case in 2010 and that light gray is 2014 so you can see a dramatic improvement that sixty-six percent of hospitals now exchange clinical care summaries with ambulatory providers not all ambulatory providers it's just that they do do it and then the next one is exchanging clinical care summaries with other hospitals again about sixty percent not bad I'm choosing not to focus on la
results both showing results and viewing results ambulatory and ordering lab orders and viewing resorts that's kind of old style here for organ we did that a long time ago mostly and physicians prescribing e-prescribing chr i think that the clinical summaries is really what we're talking about that has the core part of the data that are crucial for informing care both real-time and in analysis now this chart is about the dark blue is here it's the hospital to use health information exchange and those that use of meaningful use certified electronic health record so you see ninety-five percent of hospitals use meaningful use certified EHRs seventy-six percent of hospitals engage in health information exchange here's the same thing for physician offices seventy-eight percent are using certified electronic health records but only thirty-eight percent using health information exchange still not bad I mean thirty-eight percent but look at the number of patient visits according to this study number of patient visits in we're health information exchange data was available at the visit so the unit analysis of the visit were way way behind in terms of providing data for care so we've done a good job with health information exchange but we have a long way to go and I would propose we're not on the right track so we have distributed architecture that we've been talking about the information is left in place at each provider organization you get it when you need it it's inefficient it has its error-prone it doesn't scale it's harder to protect privacy it's impractical to search for research or application processes in a centralized architecture the ideas should bring it together in one place in one account like a bank account and that the comprehensive records over time as they're collected are then available to the next provider so health record banks the concepts been around for a long time near list one that we know about was written about it was Peter Sullivan's from MIT talking about the guardian angel it has a long paper talking about that value that health record banks or the equivalent can bring to the process of care Jonathan gold Marion ball wrote a paper in 2007 for the IBM systems journal Marion balls been a big advocate of this for a long time it's a secure repository of health records that patient or consumer decides who gets to look at the record and for how long and is aware of everybody who does look at their records I think that idea the metaphor of a bank makes sense the idea of we know most of us don't keep all our money on our person in our house we trust somebody like a bank which are mostly trustable notwithstanding the recent problems at Wells Fargo but that's that's just leadership and you still add 100,000 employees are doing a great job I'd say banks are generally trustful and that the idea is you collect deposits like your financial pauses put them in a place and then you direct how they spend you don't just go to wells fargo and say we'll just buy whatever you think i need you say i want you to send that amazon that's a weight issue that to the valet whatever it is we're doing the same thing with your data now HIPAA is clear although I would say most providers don't know that let's read what it says about the individual we're an individual request and electronic copy of protected health information that a covered entity maintains electronically which is pretty common health course the covered entity must provide the individual with access to the information in the requested electronic form and format if it is readily producible in that form and format in terms of a designated Krusty HIPAA goes on to say and individual also has a right to direct the covered entity to transmit the protected health information about the individual directly to another person or entity designated by the individual pretty clear but that's the sort of thing they're talking about 21st century cures past December 2016 also confirms their intention of a single longitudinal record section 4006 of this 996 page bill says the Secretary shall use existing authorities to encourage partnerships between health information exchange organizations and networks and healthcare providers health plans and other appropriate entities with the goal of offering patients access to their electronic health information a single longitudinal format that is easy to understand secure and may be updated automatically they go on the secretary that means secretary of health and human services in consultation with the National Coordinator shall promote policies that ensure that a patient's electronic health information is accessible to that patient and the patient's designee in a manner that facilitates communication with the patient's health care providers and other individuals including researchers consistent with such patients consent so I think the the intent of both laws is clear although I would say that most individual providers are unaware of how in how purposeful this is and what demand if a patient walked in could put on the practice these are our principles of our alliance each patient's records should be functionally stored in one place but not all records in the same place I'll come back to that each patient should control access to his or her own medical records medical records should be stored under patient controlled by a trusted organization there was an article in The Wall Street you know a couple weeks ago that talked about patients during all the records on their mobile phone as if they could go around and collect it maybe that will come about someday but it puts the onus on individuals we think with the health record thank you don't personally have to ferry check among the payment sources we think that you're going to get a trust or organization to collect your data too so I think was a little misguided in that wall street journal but we'll get back for that so with the centralized model the clinician looks up the patient's records if it's to get the update the in one transmission the health record sends back a complete summary of the patients so far the patient has encounter with clinician and the clinicians EHR is updated and the health bank is updated pretty simple and that the patient can control it and by the patient having all the data in one place it makes a club possible that the patient could purchase an application that would run against their data and say well am i at risk of such and such and should I change my eating and should i change my exercise should I change my sleeping answer for all those yes well but nonetheless the idea is based on their data because we know that when recommendations are based on personal data it means a lot more than just say well you should wear your seat belt where your wife's coloma you should exercise more when they say you know we've looked at your genetics and your family history and your prior problems and you particularly would benefit by extra sleep or avoiding such and such or including such and such in your diet that sort of thing now we'll come back to the business model but it has been proposed that we pay providers for making deposits in the health record bank and again it depends on the business model we'll get to that again so the things that are in favor of electronic health record banks are that most records at providers are already electronic we couldn't say that in 2006 we've already talked about that consumers have a right to their record we have standards such as the CDA and consolidated CDA that work for filling a health record bank with appropriate data you don't need a national patient identifier because if I walk into my office for my physician and I and they say where do you want to represent I just pull out a card they scan it I'm not talking about uploading it to the car I just say scan and there's okay and then I go home just like from home depot I go and buy something and I get a paper receipt if I want and it's in my email when I get home seven stead of being in my email it's in my health record bang it could be that easy that smartphones can help on that and we'll talk about computer security on these so again we talked about the patient depository we have to be ready for immediate access for real-time care we need to prevent the loss of all the data from an intrusion i mean anthem with some 80 million records laws we can have that we're talking about real healthcare data and then we also need to make sure that these data stores allow themselves to be used for medical research if the consumer so wishes so specifically getting into security and then we'll get into searching but talking about security now we don't want to jeopardize all the patient data by a loss of a single decryption key a loss of a given credentials any single file that a person may have or that the searcher gets access to all the data so those are security requirements and that we believe that in that it's been established a long time ago this is from a study in 1976 @ centralized data are more secure and that you don't have much multiple transmissions as you do with distributed data but the challenge with centralized records is they have all the data in one place and so one break-in can potentially get to lots of data and we don't want that so there's this perception that data in the cloud are not secure and that is one of the challenges for all kinds of performance that we have in our society but particularly for health records act so the idea that has been proposed and there are other ideas that the idea of storing each patient's record in a separate place where you still need to passwords so that you don't get 12 passwords to all the database I mean a public-private encryption key you don't get that key doesn't get you access to all the records that get you access to one record and the idea and this here now we're dealing with a different part about financial bank the security deposit if you've ever stored stuff in there you bring your key in from home and someone from the bank comes in and they both unlock it and then they leave you in in the room to count your gold but the idea here is that each patient's record would be under separate patient password and the system password and that if you designate a clinician take care of room care of you and look at your record that you would give her of access to your key she would know what your key is but she gets her key through your key and a listed cap and then when you leave that position you would draw her access to your record and she loses her key to your record so that's the idea of storing it all in one place but under individual password so this has written up in journal biomedical informatics he goes through it in detail so the idea is that each patient's records stored in a separate location although all essentially the challenge is that when you do searching for research it can take a little longer and will address that so the idea that he wanted to show that dr. jasna polana to show is that searching isn't a big problem and here's why so if you have one record and you're just looking for one record it you retrieve the record you decrypt it you examine it and then you go on to the next record and if you've done all and records you exit and you're done so I'll take n iterations but if if you have two processors then you do it parallel and then the time it takes is n over 2 and then of course you get the idea that if you have K processes which is a common way to do computing now that you have n over K and that in a cloud environments wouldn't be an inconceivable have a thousand or ten thousand processors to do searches across 10 or 30 million people so what he did there was to take a look at again simulated along the x-axis you have the number of parallel search server is going from 250 to 10,000 these are commodity servers and then along the z axis you have the population involved and then the variable is or the dependent variables how long does it take to produce a search record where you do a search speed of 25 records per second you can see that with 5 even 10 million patients in there that at 25 records for second searching that you're still out at 30 minutes or so which is pretty reasonable for research this is search is not for when you're getting care there has to be immediate active searches when you're doing research and you can afford to wait a little bit of time so what we're doing is trading off the security of having a single person's record under a single password versus taking a little longer to do the search and then on this slide he adjust and now what we've said is let's fix the population to 10,000 and say what is the effect of speed on the outcome and you can see that if you can do 50 records per second you can even do 10 million with the less than 10 minutes recovery time so the point is is that this exceeds to a widespread commodity search with multiple servers so we balanced the security of individual record under single password versus ability to search all the records using distributed processing so just got a couple slides looking over the comparison of health information exchanges with health record banks so we've already talked about health information exchanges as are many to many scattered distribution health record bank you get it all in one place send it to one place many-to-one to send it and then all 121 to retrieve it for that visit well with hie you have many records coming in at the same time you have to manage that and reconcile them with health record bank it's already been reconciled that our job at the health record bank is to make the deposits as they come reconcile it with the records are already in place and then you get one summarized record at the time of care now if you have health information exchanges without a repository you have no place for the patients are result put together but a health record bank the very essence of it is that all the patients deposits are in one place with health information where I talked about you need a lot of trusted stakeholders to come together make agreements with a health record bank once it's established and once the patient gives it the trust and once they should go to their providing that would you send me a copy then you you're beyond having have multiple stakeholders degree on it health information exchanges are viewed by some as being the providers record clearly with the health record bank gets a consumers copy of their professional health record and they own it they get decide where to put it then you have genomic and imaging data again more kinds of healthcare data they're scattered all over and especially with with genomics you're going to want them under your control and be able to add to it and if you want to get one more panel to add two years and do it all it's in terms of even a retail process meaning not for necessary for clinical care you're going to want to store it with the rest of your data because interpreting those genomics make sense only in the sense of all the other habits and all the other history that you have now when a consumer moves with a health information exchange they may have a way to send the data but especially if it's scattered data they don't have a central repository it's going to be hard for them update whereas moving with the health record bank it's not an issue in the same way that even if your bank doesn't have a retail location as a small town that you move to you can still do your banking with that original bank nothing changes we've talked about consumers choosing apps on time on a mobile app or a whole math but probably a mobile app to run against their data hard to do against hie where you don't really own the data you don't really have the data under your management but easy to do with a health record bank if there's an error you corrected in your health record bank and then the next person that goes to look at it is that you're going to see for care gets the updated copy easy for consumers to control access and then again as we get to patient generated help it easier to do with your to your own records and to do it to all the providers that you might see but there are problems with a centralized model who's the data stored who governs ac
ess who controlled access individual records who controls search so we have health record Bank Alliance proposes that communities get together with a non-profit board of consumers and providers to say what are the rules from managing this health record bank you can have a private party for of an organization actually doing the storage but we would propose that because these are sensitive that the initial health record banks be managed by community groups who get together and they set the rules and then they hire out the actual technical storage of their data another problem with centralized accesses we have a single point of failure so if a health of bank of help us health record bank loses their internet connectivity because of a DNS attack then you're you're without and that's a problem that we haven't allowed to figure out ways it is to have copies of your data in multiple places so that one out each may not affect it so severely and we talked about risk of the whole data set being compromised ok this is a proposed business model a health record bank according to these data haven't been put together yet but the idea is well what is it what does it work through what's it look like financially so this is based on many spreadsheets putting together how many technical people does it take to manage incoming data to reconcile it to manage the storage that sort of thing so at a million subscribers they're figuring that their cost of sixteen dollars per person per year but look at that there's six dollars to manage the storage and ten dollars per person ear to pay providers to create deposits so if we if we take that out we're talking at six for six dollars per person for a million six million dollars about the cost of most health information exchanges right now now in terms of revenue sponsorships meaning that people that will pay because they believe that they can offer an application to the patient that the patient will want and so they may sponsor the data applications this would be paid for by consumers for example the emergency department access alert means what happens if I fall off my bike show up at the emergency department to see Andy he doesn't know who I am or knows who I am but I know my record that burn see dogs get automatic access to health record bank and everybody my family and friends co gets the email it says hey Gibson showed up in the emergency department emergency doctor they are open the record meaning that there's the break the glass access they have good access to my full records and there will be a price to pay if that's inappropriate and meanwhile my family knows that that doctor looked at preventive care reminders are things that we believe that patients will pay for to help them manage their care medication reminders again you can do preventive care reminders and medication reminders without your health record bank but these are just examples of applications that could run against all of your data in one place research is a big deal but I don't know what the value of our data are but I would think that if you had a million people I in a given area with their complete data and half of them say yeah you can use my data for research inappropriately identified or in a de-identified way and oh by the way is in a pay me for that that it could be pretty valuable saying what is the effectiveness of drugs out in the marketplace when they're being used so this does not assume any health care cost savings so about health insurance company said this could be helpful for managing populations that they might pay for some of the health record bank the freemium model means that you get basic applications you get your records no charge to you and that you can buy up if you want some of these other applications and then there could be advertising or you could buy out the advertising by by paying for it in the same way your kindle cheaper if you take the advertising it's thirty dollars more if you take your kindle with without advertising just some ideas and so running that business model through ultimately getting at 2 million 2.8 million consumers by month 48 break-even was at 16 months and substantial return on investment of in the third and fourth years again these are just models based on number of people they felt would be necessary to run a health record bank for three million people so it was tried in arizona 2010 2011 and it fails and i got to give credit to Billy a snob Ted short left is is the it's a chair of the board of health record banking alliance in a long term proponent of it he actually lived as part of a year in Arizona and they wrote up that they fail doing a health record bank in Arizona 2010 I remember 2010 high-tech American Recovery and Reinvestment Act was just getting started electronic health records were not in as whitey use as they are now nonetheless they start out with four hundred thousand dollars capitalization they had a community nonprofit that did the governance and set the rules and then they paid a for-profit health record Bank company to do the technical piece they did initially billboard and radio and then they went to Facebook and Twitter and they got some uptake by consumers they also went to large physician groups and did in person marketing and the physicians were signing up to get paid for making the positive stealth record Bank of course in the early months and years that's not a lot of payment but over time the idea that physicians at scale would get three or dollars per month for contributing data to health record bank and this is not onerous you just say you use your same EHR interface you just put a new address in it as you would with direct they were also looking at initially a ninety nine dollar fee to consumers they did surveys and they they can they survey consumers and consumers that I pay between 100 and 150 dollars to start this process and that about twenty five or thirty percent of people said that they would be willing to pay at that level for this so again that's well less than seventy-five percent will less than a hundred percent but people will know pay some they start out ninety nine dollars what they turned out they only got ten consumers to sign up so it failed and one of the things in going back is physicians were reluctant to ask patients to pay for something that's going to cost them money directly so our feeling going forward is we don't charge the patient because that's a non-starter one of our one of our challenges is that we don't believe that patients really care about their records until they get sick or until it family member gets sick or injured I'm sorry under the question was until somebody asked for sure so we don't care until someone asked for it and then we wish that we had it oh I have a friend who I went to high school with college with we went to different medical schools he was an anesthesiologist he had pneumothoraces I had pneumothoraces and he was in Hawaii he had his lung scrubbed from it but it could always come back he thought he was getting it on a weekend you think I'm going to go to kawaii urgent care now I'm going to have this complicated history about new authority so he was scrambling why he was having chest pain scrambling to put his records together so that when he appeared in urgent carry data records well again as you point out until you need it until someone asks for it you don't care about it but then you do care about it all in a hurry so we believe that the idea is make this so simple using existing standards that is sort of like would you like a copy of today's visit sent so you just do it automatically yeah it just collects there and then when you need it you know I've already got that covered doesn't cost me anything and we're ready to go so what a policy changes do we believe are needed to make health record banks work I counted to number one is you got to require clinicians to do what HIPAA and 21st century cures already require which is send it but instead of making it when the patient asks just make it routine now will that be politically public feasible well this is the challenge of my job and of this concept of can we convince them to put in an apology saying you need to send it it's not it's obligated now now I talked to Western of the quality health network in Grand Junction Colorado and there are health information exchange with a federated model but he says that even though it's a federated model which each organization maintains their own data when you make a query it acts like it's centralized it's that fast and I said well do you have a patient portal and says you know we did but then our and they have here's an example one of the successful health information exchanges western Colorado 150,000 patients 1,100 providers 130 offices 26 hospitals 70,000 queries per month the database so there and they have ninety-four percent participation among their providers that's the sort of participation needed to make all this work so i would say they've got what they need in quality health network for real-time care research is another thing patient controls another thing but if we overlap if we overlap health information exchanges with the repository with repository and a health record bank I think there's a big overlap but the idea is because you have a patient portal on he says well we wanted that but then the doctors screen because they get meaningful use dollars by showing that they have five percent of their patients actually using that if the patients are going to the hie portal instead of the individual office coral they lose their credit for Meaningful Use so they had to turn it off so again that that's a change that we need to make in the in the requirements to make sure that that worked all right so we need to relax it in that case but it's our contention we don't need new standards for all the information they're not perfect but they're good enough to get going even if we have text blobs coming over for family history maybe we can do natural language processing to pre digest it for clinicians that's another talk we'll do that another day but the idea is let's just get a copy of what's being collected anyway sent to the patient under their control so we've already talked about the challenge some of the challenge most consumers probably don't care about their health care records there's an argument about whether boomers care more about whether Millennials care more we can have that discussion I would say that most clinicians don't know that the patient gets a copy it's a full electronic copy if they ask for it they're pretty much used to putting their portal up there and saying that that's good I think one of the doubt one of the headwinds reruns with the health record bank is that there have been a lot invested in state health information exchanges and for people to say you know we're going to augment that that requires I just think its growth but other people would perceive that as a failure and I think that's the reason why they go along with health record bank we'll see what the new administration the new Congress and does I believe that some vendors believe that they are going to offer that personal health record with a woman's name to their patients but i think that there's reasons to do it away from electronic health record vendors but they may want to do that I'd be buying if one of the epic Cerner allscripts meditech wanted to do a consumer record and we put up the characteristics and evaluate and say does it meet our principles of what we want with health record by insurance companies may feel some loss of power on one hand made me feel lots of power that somebody else is collecting data like the patient that might be more powerful the other point if once you move to value-based payment and may say hey if I can get all the clinical data and the payment data and we engage the patients in their care maybe they can become less expensive to manage and will do better and maybe accountable care organizations would be interested in this of course as with banking we had a hundred years to set up banking regulations but there needs to be audit and it needs to be regulation but that's not inconceivable that could happen and then genomics and imaging data those kind of costs that I add there weren't looking for storage costs where you have genomics and imaging data especially imaging data but I think storage costs will continue to come down I don't see that as it made a big a big hill to get over but it is a consideration so who should hold sales record bank government I'm not sure that most consumers are going to be happy with the government hosting that insurance companies I think consumers feel a certain distrust with some insurance some parts of some insurance companies hospitals I think have competitive issues they're not ideal physicians are working hard just to get to carry down I'm not sure they haven't wherewithal to do this accountable occurring rescission they could do this especially when the value of all the data becomes clear when you truly go at value-based care and we're not there yet financial banks I don't want throw that out I think that there are banks that are doing medical banking that's a different content from health record banking medical banking is where banks you the inexpensive technology manage the copay and the point of service collection in the deductible and all that and cerner ran a medical bank for their employees that made it very easy the whole thing was reconciled all the transaction taken care of as in the patient presented for care and then community organization we've talked about because if we could get a place like Portland OR Oregon to say just send a copy of what you're sending to each other send it to a health record bank Portland be a great place to do this so my last content slide then we'll go to questions is blockchain what about watching lots of hype about watching Harvard Business Review just had to write up just a couple weeks ago in a January edition about watching I think they got it right I think that blunt I mean I don't think watching is a distributed ledger mean it's a great place to say here's what happened here's a transaction having a certain date time and things where we count on somebody else like a county government to store who bought houses who recorded their title to a house so you don't get at them adverse claims on your house title or identities who now we count on the state to do driver's license but the idea was wandering great place to store where no one person is in charge of facts and transactions but I would say it's not a good place to store the content of medical records it's a good place to store not a patient was seen by a given provider on a different given date time and then some doctor in New York contribute says I saw Gibson in New York when he was visiting in April 29th and April 30 sight throughout and for some reason I had to see Andy the moonscape arm around a bad year on April thirtieth and that Andy's the emergency department records it in Portland the other burns department recorded in New York ultimately the block chain nodes are brought up to speed and then the next person on Monday April or May first would see exactly what order was and they know where to go for my record not required trail traffic bank but I think that the idea is that blotching is not a place to store records there might be a pointer to your records so that it it's too expensive and too complex to store the content of the record but it may have something to do with keeping a secure agreed upon record uncontroverted will record of who saw boom when okay in summary we've talked about health record banks and their data are required for safe care and population health management we don't need anymore standards to do it hippo re says that per view download and transmit and I went and looked at it a lot this week to make sure it says view download and transmit to our portal that doesn't transmit it's not adequate to meet the letters wall with HIPAA Lord 21st century cures if the patient has four well I think helped record banks have a lot of the value of health information exchanges but add even more in terms of patient control the ability to upload your own data the abilit
to control of the auditing the sense of privacy over your record and we talked about about HR BS don't exist in that name yet in this country although I'm trying to establish contact with health information exchanges that are run a lot in the same way I think rhode island has one I think my health of my health place in aqua Florida has that quality health network and western Colorado may have a lot of that and that there's also not yet confirmed health record business model but I think that they have a chance of being sustainable based on their value to consumers and the applications and research that will pay for health record so let's all we have time for some questions and love to hear what you're thinking thank you here so since I'm sitting in for bill today I'll ask the first question because fill out these guys this is great really fascinating stuff I have a couple of questions a lot a couple of areas that seem like they're their substantial challenges yeah one is that it sort of seems like it's screaming for some kind of inexpensive killer app in this space that you bring the consumer in with the freemium model but you're really dependent on a significant fraction of consumers being willing to pay for the updated app and what might that be and is there data backing up what that would be in the second the second question and maybe maybe I just didn't see it the way what you talked about it does seem like there's some complexity with a lot of health providers having arrangements with the health record bank in order to you know be assured of the HIPAA compliance and the financials and all that so while it may be less complex than than a health information exchange in all the n-by-n kind of things it does seem like you know a health record Bank of Oregon and Washington having relationships with all the health providers in Oregon Washington seems a daunting task but maybe you could talk a little bit more about that no I think it is a daunting task I think we're deep in the end zone now I do believe that things are better now than they were ten years ago in terms of the killer app I don't think we know what that is yet i spoke with john casillas on tuesday he established the medical banking project he promotes out all over the world and that's medical banking where you're handling micro payments and payments for healthcare we talked about he agrees the challenge is finding that application that will make all this payoff you know I don't exactly at their there's a company called daca do that does a health score and they take an enormous history I started it but it takes a take some time to complete where they ask you about all your habits and everything that's happened to you ask you about your numbers a blood pressure cholesterol which your exercise what kind of exercise you should do is stress level your status of your relationship pretty personal information and then they give you a score between zero and thousands and it changes day to day and then of course for those that are competitive I'm not in that sense but for those that are competitive now they're doing rollator their buds about whether the score is going up and down at they sleep more to this quarter anyway I'm not saying that to kill around but it's intriguing because it's based about me and what happened to me and in fact what I do in the past and what do I appear to be at risk at I'd also say as those genomic markers come in one by one over time now we're going to over find out things I mean all these things that you thought like white can some people smell in their urine that asparagus and other people well I don't know exactly what number but they found the gene for that or or why is it that the end we can eat so much fat and I mean the several genes that I mean all these things that we y can tell my wife hate some spices that her whole family hates what's the one in Mexican food starts with the scene cilantro she hates on for me I don't really taste it that much but her whole family brothers sisters all of them I'm sure there's a gene for an out cilantro is insignificant but as the genomics come in and combined with our habits and our current functioning I think that will be valuable is it a killer app demarco out to tell and I don't know the answer that in terms of the complexity about building yeah it's complex but it's not like we have to start over with the standards we just it's already in the law that it's required I think that we need to get a demonstration side up and the idea openly would be we might start with a community but I think ultimately you might look to a national organization like the visa did with with credit card payments is that yeah you're part of the visa network and you just send it there rule ferreted out not not fair will move your data out to the particular bank that yours but you just upload when you do your visit in addition of paying for it you also just upload your record to the health bank that you have an account with and you go online echeck and see events turn well agree it's a big deal and that's part of my job is to figure out create awareness look at what we need to do policy wise and on the hill and say how are we going to bring this about so no it's complex but it's not like we have to make up a lot of new stuff go ahead so thanks very much and I first wanted to say that things like this like health record banks I think are going to be incredibly important but as you mentioned there's a lot of difficulties with it a couple that I wanted to ask you about is one is you know why for example do you think that it should be a privatized situation and you know why should we or why do you think that customer or consumers are more likely to trust a private in this organization or a private company to hold all of this data more so than they would dress for example a local or federal government to I wanted to highlight one possibility in terms of like you're bringing up killer apps or reasons that consumers would do this because it seems like a big hurdle is how do you get people to actually be interested in this whole thing to begin with and that's you know maybe it should be monetized for the consumers instead of being monetized for the company that is doing all of this in the sense that consumers already we give up all of our health data we give up all this information and companies like insurance companies and other corporations actually market are our information out as you know advertisements or other health indices or whatever you know this kind of a project would would definitely allow consumers essentially take control of that and then the last thing that I just want to mention really quick is health providers already have access so will prevent health providers from doing an end run on this process essentially because HIPAA does give them the right to access whatever care whatever information they need about their patients that involves direct care so there's nothing that would force them to go to these things in order to access information taking a third question first they go there because over time it would be more complete and they could go to one place and get it all so that that's why they would do even though they get it where ever they would i SPECT these hie s will continue especially the vendor Beiste cherries will continue the second one in terms of monetizing absolutely we believe that the monetization data will be very valuable and I just don't know what the dollar amount of that is but no the in the consumer would have the right to make that decision so different from where your data taken out you're not involved this would have to be expressly saying you get to choose here's the value equation here's the day they're going to use in here is how much you can pay for it so we think motivations a big deal it will capitalize it to begin with because of the network effect but over the long term on ization porn and the next one up for the next iteration is talk I will research the surveys on who consume consumers trust and I just proffered that they're more likely to trust a trusted private organization than government but I'm just property not now based on my experience but I don't have the data and all will work that up so how about you and then you you we believe regions and communities are the best absolutely that's the place care is mostly regional this idea that I'm in New York and I get hurt or something that's less than 1% of it or even less than that so no we totally agree its community based its regional base and that's where we would start that's a place to show it the long run it could be national but no we agree it starts with it starts with the community and that's where we build the critical mass rural yeah I mean you could see that Eastern Oregon with a few hospitals over there and the multiple providers they could still benefit but you're right I urban areas may benefit the most of that but it's a model that I mean we look at ed AE emergency department information exchange and pre manage and how that's taken off and all the hospitals in Washington Oregon do that I mean you can see what community can do and if it the city is that the state is it to States getting together either the antimatter so okay so does that mean we can start at your in Portland one more question I think we have time for thank you yep you okay let's stay in touch on that oh I in terms of putting all this together I think it's an issue of trust and it makes a lot of sense what you say and the issue is again me I need to bring myself up to speed on what consumer sentiment is on this but I think it's an issue of trust whether we could put all those data parts together what right agree that's a challenge if it'll do it at their clinician physician advises up to I mean it goes way up now maybe that's weak you know thank you very much for the attention thanks for the ideas
