The Evolution of Health Information Exchange



thank you so much for inviting me here today and I'm gonna be moving around a little bit just because that's the kinetic energy that I bring to every room so Markos told me that I can go as far as right here so if I start going past just make sure that y'all help help keep me true and stay on this way so I'm here to talk to you today about data right we're in the school of biomedical informatics so I'm gonna talk to you about data but I'm going to talk to you about data in the context of healthcare and specifically in the context of health information exchange I'm going to share with you a little bit about my background a little bit about my interest talk to you specifically about some research that I've done germane to health information exchange I'm going to take you through a fun exciting little use case that we're doing right now every day using 1987 technology and I'm gonna wrap it up with a bow at the end okay and I promise to keep you awake so thanks for the full house – by the way this is great I love to see that so this is a little bit about my background I'm originally a registered and licensed dietitian I have about 10 years of experience working in a clinical setting I have a master's degree in business administration with a focus on finance so every time I look at a problem in health care I look at numbers I see numbers I see spreadsheets I think about what is this gonna cost us how much is this gonna save us what kind of time are we gonna save how is this gonna save us in terms of minutes in terms of seconds in terms of dollars I think about numbers in that way so that's the first introduction to data and the way that I think about everything in terms of ROI or return on investment I have about 10 years teaching college students so I love being in front of a class I love working with college students and I love love love to teach my PhD is from the School of Public Health I'm a second student that ever petitioned to take courses here at the School of biomedical informatics and I fought fought fought to get those courses to count and now students can freely take classes back and forth and I completely claim credit for a lot of that whether or not it's true I completely claim credit and I have spent the last four years of my career as one of the founding executives with the regional health information exchange here in Houston and dr. Jim Langevin founding was our founding CEO here and now he's here with you you get the pleasure to have him here at the school so the organization of the presentation is going to be such we're gonna go over some basic definitions so for those of you who don't have a healthcare background I want to take you through a little bit of healthcare finance and why we don't finance things in a way that's tenable in the United States and just kind of make the case why we have to do better and when you are out there or you're out there right now but when you're out there after your degree why it's so important to think about cost and why it's so important to never lose sight that there is a patient that you are working towards in any kind of setting whatever kind of data you're doing if you're doing genetics if you're doing any kind of EHR technology there is a patient that you're doing this for the benefit of and it's all in the context of that patient we're going to talk about the public health significance of the research we'll look at some models of health information exchange nationally and that's from my published research we're going to look at that nifty little use case that's going to leverage some 1987 technology and I'll just make some recommendations for future research she's moving around too much no problem I was what I was muted yeah that's right you yesterday I'm gonna start over okay good all right so moving on so healthcare finance 101 we spend way too much money in the United States if you have been a member of a healthcare system in another country welcome to the United States where we will spend more money on you than any other country in the world any other country in the world with not necessarily the best outcomes when you need a hammer we will use a bulldozer every single time every single time we do have amazing health care outcomes amazing I mean you can walk across the street to MD Anderson and and miracles are made you can go across the street to Texas Children's Hospital pick a hospital miracles happen but we spend a tremendous amount of money in this country on health care almost twice as much as the next industrialized country which if you're curious is in Scandinavia in England so we spend the greatest amount per capita per person of any industrialized country nearly eighty five hundred dollars per person we are on did I lose audio again I'm getting beats up here no okay so we are on track to spend around a quarter of our gross domestic product that means what we make in the United States all the money that we make a quarter of it goes towards health care that is nearly twice as much as any other industrialized country I don't know if that weightiness kind of sits on you but we spend way too much with not the same outcomes as any other country so where is all this money going okay it's going towards technology it's going to what we're doing right now so how can we deploy this technology than most efficiently and then the best way that we possibly can for the best outcomes that's where we all come in just an interesting fact you know which state does the absolute worst in ensuring ensuring people for health and yes yes which county does the worse we're sitting right in the middle of it right now largest percentage of uninsured in the country right here in the zip code you're sitting in it right now we're living in it so this is the gaps and transitions of care ok so a typical patient in this country you're being coordinated by multiple multiple doctors so if you think about it you know you think about how many doctors you might have if you have a chronic disease let's say you have diabetes treated fine no problems you may have endocrinologist you may have an eye doctor you may have a specialist a podiatrist that you go to see for your feet you may have a general practitioner if you're also a woman you may have an OB gen that's five doctors right there and that's it everything's perfect okay if you get into an accident you're gonna go to a needy you'll have an emergency physician are they gonna be able to get all your records what about if you have any kind of behavioral health maybe you have a psychiatrist that follows you now we're up to seven this is still someone who's in pretty good health okay this is not someone who has really challenging problems this is someone who's of average pretty good health you're responsible for coordinating your own care as a patient in this country you don't have someone that's coordinating care for you you're seeing these providers that are belonging to different delivery systems they all have their own electronic health record technology that don't all necessarily talk to each other so how are these records going back and forth how are people coordinating your care how are they not duplicating things that have been done over and over and over again this can get really complicated in the case of a very sick patient or a child or your mom or your dad or your grandparents that have to get stuck over and over and over again haven't I already had that lab test why are you doing this again why am I having to pay for this again is my insurance gonna cover this these are the kinds of questions that we get asked in health care management so there's abundant research that shows this abundant research that there's gaps in care and again we're in technology we're in the data business we see this we know there's solutions that we can come up with their solutions we're working on their solutions that we can we can put in place to fix this New England Journal of Medicine States 16 physicians coordinating care for chronic conditions 16 physicians many of them in different types of systems these aren't people all that are all under the same roof Medicare patients between 40 to 70% admitted to the hospital receive services from a mean of 10 or more physicians during their stay this is one episode of care seeing 10 or more physicians they're probably not talking to each other maybe even in the same chart and in the same electronic health record system and not coordinating care with each other this is a problem from a safety aspect this is a huge problem the AHRQ reports that routinely almost 50% of patients fall through the cracks in transitions of care one in five discharge patients may have an adverse event within three weeks from missing information that was should have been given to them at their discharge this is simple this could have been dr. Smith you're gonna be discharged now you just had surgery you need to followup with your doctor in three days maybe her nervous doesn't communicate that to her maybe she was on medication when she was discharged so she didn't understand it her husband picked her up and she was like I don't know what they said to me I don't know what that was and we lost that piece of paper so she never scheduled a visit nobody ever called her to check on her so she has an adverse event very simple stuff but no one is coordinating this care so in comes collaborative technology collaborative technology that seems kind of complicated I don't really know what that is does anybody scratching their head what's collaborative yeah I don't even know what that is you know what this is this is collaborative technology so let's look at how many of you know if this is anybody use Outlook 1.1 billion of you use Outlook to schedule meetings to schedule events okay do you want to come to this meeting I'm inviting you to this meeting you're collaborating with people this is Outlook okay 1.1 million what about billion sorry what about Twitter in the middle 500 million people collaborating this is collaborative technology key harmony when we started the hie dr. Liang Kabir how's your lady beer and I we go into meetings and say it's like match.com come on people you gotta join because if everyone doesn't participate you're only going in its like match.com and you see yourself you don't want to just see yourself you've got to collaborate with other hospitals to come in a harmony 1.2 million users a month get on a hominid to find the love of their life and collaborate 1.2 million souls a month Facebook don't even get me started on Facebook 750 million active users and if someone knows this I want them to raise their hand does anybody know what this is come on students this is one of the largest Chinese dating websites out there no what do you know that come on I was hoping someone would recognize this no one knows what it is over 1 million users monthly China love cupid come on I was just hoping I would get at least one smile this is collaboration technology though it's technology that bridges the gaps where people come together so they can share information around a common goal so what would this look like in healthcare who's our common goal remember at the beginning talking about data it's not always just about the data who's at the end of the data it's the patient the common goal and healthcare technology is all about the patient so if we think of the patient as the Nexus here of care you may have a clinic clinic has electronic health record technology right so maybe they're using each clinic works there loading all their charts into eClinicalWorks all scripts greenway a prima any of these maybe you've worked with these maybe you've learned about these here in your classes maybe you have a solo doc solo practitioner or a physician out in the field maybe this is a free-standing er you know when you get it you get the flu on the weekends and you need to go somewhere this person also is using electronic health record technology you may not be talking to each other you see them during the day you see them on the weekends might be useful for them to talk to each other about you same with this one same with this one this one may be at a hospital this may be big technology epic Cerner big time technology what the big players and the and the TMC use and what the hospitals use big time technology massive databases of technology massive units of what we can mine to see what kind of diseases are we seeing what are our treatment rates what are our success rates how many congestive heart failure patients do we have right now this kind of centralized data repository information that anybody would love to get their hands on this is a patient that sits in the midle health information exchange can bridge the gaps between all all of these health information exchange using standards hl7 a little bit hl7 a little bit of CCDA a little bit of CCD a couple of different standards that are coming out can be the pipes that go from one electronic health record system to another electronic health record system matching patients using record locator services providing patient information at the point of care for physicians that are treating patients that's the cool stuff that I'm interested in so services that can typically be provided by health information exchanges or clinical integration so if you're at MD Anderson you're seeing patients from all over Houston all over Texas all over the country all over the world you need the ability to pull those records patient records that may exist in other places right so you want to get the patient's history and bring it into your your electronic health record so MD Anderson is converting over to an epic system of extremely robust electronic health record system this system will be able to talk to other systems not just epics but using health information technology interoperability systems using different kinds of standards with clinical integration we'll be able to pull records in you can know history you can know any kind of allergies you can know Prior treatments you can know everything about that patient that you need to know at the point of care when you're about to deliver chemo radiation I don't want to repeat any kind of tests maybe that they've already had anything like that without having to leave their epic system this is clinical integration medical image exchange I mentioned how health care finance it's really not viable in the United States we repeat so much we repeat MRIs we repeat cat-scans repeat tonnes of radiological exams that don't always have to be repeated they're a danger to the patient they can be a danger to the pocketbook they're not always reimbursed by insurance they're not always reimbursed by the patient who ends up paying for that who is it paying for that taxpayers do absolutely healthcare costs go up as a result for everyone when it's uncompensated no one wins no one wins in this case when things are not reimbursed and duplicated over and over and over again if you could just get those prior studies get a prior study view it if it's good enough if it's excellent wonderful the patient doesn't need to go through any additional radiation that's something medical secured direct messaging this is just a way to get physicians give physicians and clinicians the ability to send messages back and forth in a HIPAA compliant environment a lot of times we're seeing physicians and other clinicians and nurses and social workers and people that are maybe ancillary providers like social workers or other people that don't exist and live in the hospitals that they're just being sent emails about patients they're being sent charts about patients to their Gmail to their yahoo mail to their hotmail this is not HIPAA compliant you can't send a pH I through you can't send protected health information about a patient that's through a non protected way so these are all opportunities for us to send data and for us to be able to collaborate and to know how to collaborate around a patient so these are some more formal definitions of health information exchange the one that I really like is the one that's out of hims this is technology that facilitates the exchange of patient specific electronic health information it's no more fancy than that you're just taking health and patient health information from one place you're moving it to another there's a lot of other stuff you have to think about technology standards patient authorization legal rights signing contracts who gets money do people get paid for it and we'll go over some of those barriers basically this is this is all that it is the benefits are huge if you can make this all work so I just want to drill the significance of this all it's not just about saving money but our public health benefits when the outcomes increase and the cost decreases if outcomes are better that's good for everybody everybody wins cost decreases everybody wins outcomes increase patience happy things aren't duplicated everybody wins care is provided is that is that a higher level everybody wins so it's important to understand and I became very very interested in doing my dissertation to see you know not very many people have studied if collaboration technology is what is it really doing how is it evolving what are the standards who's doing it out there what are these models that exist that are out there and so I looked and at the time at in 2013 there were 206 a Chinese that existed public a Chinese that existed across the country and I studied every single one of them to look at their models to look at how they develop to look at their method of funding to look at their technology in their architecture so I'll give you a little bit of an idea of some research that I published based upon this so I looked at I pulled out 25 of 25 of the a Chinese from the largest metropolitan cities basically ones that compared to Houston since I was working in Houston living and breathing the hie and Houston so large metropolitan community-based not-for-profit h IES which was very important i wanted to focus on not-for-profit hie s because as a neutral entity they are slightly different than enterprise for-profit h IES and Colet I collected qualitative data and I broke it out into a framework basically looking at the scale what size that they covered were they small a Chinese and I found that a lot of so-called hie s were really just websites websites that existed that were maybe one hospital that had just a few transactions so I wanted large hie s that covered 20 counties several million people people that were really exchanging data in a meaningful way to see what were they doing and how were they doing it so these are the the locations of the 25 which I looked at for basically Geographic sampling so the top case study is of course on Greater Houston health connect and then the two most successful in my research health bridge in Cincinnati and then Regan Street you know in Indiana and health connect covers around 7 million people health bridge and Cincinnati Cincinnati covers 3 million in Regan Street covers around 7 million and what I found was interesting was that unlike in Houston both most health information exchanges don't even give the patient the option to say yes I want to participate in this and I want to share my data they just the patient participates the patient isn't asked so that's what we call an opt-in model so here in Houston if you visit a hospital you will be asked to sign a consent form because the hospitals in Houston in the 20 surrounding counties participate in health connect that's called an opt-in model this is a make-or-break scenario for most health information extension exchanges so as a patient you say yes I want to participate I want my information to be shared with doctors who treat me that's important to me I want them to have information on me I don't see why I wouldn't want that to be shared yes so your information is shared freely these other two don't even ask your information is just shared which they can do by law our hospitals didn't want that there are also different technical models centralized data repository Regan Strief puts everything into a giant data repository mine's it does all kind of neat nifty things to it we don't do that here in Houston we use what's called a federated or a decentralized type model it's kind of like Expedia comm if you will so Expedia right y'all got my analogy back there I can see already so Expedia doesn't duplicate the entire database of United Airlines or Continental or Southwest or Air France right they say what where do you want to go when do you want to go how many people are going so that's what you do with it with a decentralized or a federated type hie dr. Angela Ross is my patient this is her age this is her medical record number are there any other records that exist for a doctor Angela Ross health information exchange goes out queries and brings back there are six records that exist from various from various hospitals that she's she's been seen and it brings them up query and retrieve so we also found I also found that there are five factors that determine the success Meaningful Use is huge thank you thank you to the government meaningful use some of you may be working in meaningful use projects but meaningful use which states that physicians have to adopt EHRs if they're going to accept money from the federal government by a certain point they have to do certain things with it they have to use it in a meaningful way to exchange data with other providers they have to step it up each year each year each year it gets a little bit more difficult hie health information exchange is a factor in Meaningful Use eventually everyone is going to have to exchange data in a meaningful way they're gonna have to utilize this technology in some meaningful way because it is helping the patient drive clinical outcomes so that's definitely helped changes in insurance reimbursement insurance companies no longer are paying for paying for doctors to order I want the cess run again on with the cess run again I want this test run again I want this test run again I want this test run again insurance companies are saying this test has already been run you need to go find where it's been run and you need to bring it into your practice we've already paid for this test it was done yesterday across the street go and get it or tell the patient to go and get it if you're a patient have you had a cat scan if you've had a cat scan and you go and you sound on a second opinion I think I want that cat scan done again guess what you're gonna have to pay for it out of your pocket if you can get anybody to do it for you cuz your insurance company will not pay for it they're not going to pay for it and they're very very expensive so you're not gonna do it the second time so those have helped encourage HIA usage most importantly hurdles hurdles that that I find regularly so technology so when trying to bring this technology to to any company and trying to to any hospital and trying to say you know what health information exchange is a great thing dr. yang and I really think that you need to bring this on to your hospital you know Indy Anderson should really really have health information technology because it would be a really great thing okay dr. yang if he's the head of MD Anderson's technology if he's the CIO if he's anyone else he's gonna say okay I have a few questions some of them are technical how are you gonna share our data who are you gonna share the data with do I own it do I own the data what if I decide to pull the data back can I do that because that's always a question so some of the questions yes they're gonna be technical and as data experts you know they're gonna be technical some are gonna be financial so that's why we're getting into the yellow here some are gonna be financial how much is this gonna cost me is it gonna be expensive how much time am I gonna have to put my teeth staff on to this how much time is it gonna take them am I gonna have to pull them away from other projects what kind of Labor am I talking about are you sustainable what about if you go out of business then what am I going to be left with that's also a question the big thing though and the really big thing that's hard to convey is organizational who else is in this this is gonna make me more competitive somebody else gonna try and steal this from me right dr. Murthy who else is gonna participate in this what types of entities are you gonna take my data and sell it to life insurance companies are you gonna take my data and sell it to a pharmaceutical company are you gonna take my data and sell it to the hospital across the street so they know my trade secrets now I'm worried about this so what's really important that I want to convey is in health information technology in general it is about the technology true it's about data and it's very very important that we understand data we understand jargon we understand how it works we understand how it can benefit the patient the patient especially in healthcare we understand the finances what are the implications how much is it going to cost is there a return on the investment what's the labor that's going to be required but organizationally you have to understand if you are going to make the case to a CIO or if you are the CIO and a case is being made to you you have to understand the implications of the competition and of the space and of the management that's involved when making a huge decision whether it's choosing a new technology to go into your hospital or your clinic or you're selling a new technology that's going to a hospital or a clinic you have to understand the organizational landscape so summary of those findings basically federal policies have definitely helped they always help us with health information technology help us inch forward with meaningful use help us drive adoption because if we leave it up to people in health care we are so far behind it is not slick and savvy like Apple or Google I mean you can look into technology and in health care we are just dramatically behind and so many things that we do so federal policies have helped inch people along even though we think it's important here in Houston with our hie to give patients the opportunity to say yes I want to participate most people across the country don't share that view and so they have had an opt-out security model 64% of most health information exchanges don't even ask the patient unless the patient says that you're sharing my data with people and I really don't like that oh ok that's fine you don't like that then we'll take your name off the list but otherwise your data is in that's an opt-out model most everyone uses a third-party vendor that means they go out and they buy technology to do this they don't customize or don't make their own types of solutions and scalability size is important you've got to be you've got to cover a large area of patience for this really to be important so this is kind of that match.com effect the more people that collaborate the better the more that collaborate the better so just to re reiterate on the focal points of what should be important for any collaborative technology fragmented care creates problems especially for people that are uninsured people who are medically indigent so you're maybe not indigent but you're medically indigent which means you don't have healthcare insurance you don't have good coverage you maybe don't have enough money to pay for your healthcare what that means is it becomes everyone else's problem and it becomes everyone else's financial burden because if you're not paying for your healthcare everyone else who does have healthcare insurance is paying for your healthcare so it is our problem it's everyone's problem everyone who receives healthcare so our primary focus is always on improving quality access reducing costs and increasing patient safety that should always be our primary focus in healthcare in health information technology that can bridge these gaps and solve all of the problems ready for the fun use case ok so all that so it's most of the students in the back this is all gonna be for you because the because all the professors in the front are gonna know the answers ok so what do you see hl7 come on the professor's already beat you to it okay so hl7 right so hl7 was developed in I already told you 1987 awesome so it's now 2015 ok right good so we're rolling here so hl7 right so it kind of just if you've never seen it seven it kind of just looks like a lot of line and letters and everything it's super psyched when we see this kind of stuff right if you've never had an hl7 class dr. Smith is like this is so cool I'm really loving this presentation now okay so this is ADT admission discharge transfer okay stick with me here because I'm gonna break this down and everybody's gonna know what an ATT messages every time you go to the hospital and events is created an hl7 event is created an ADT message admission a discharge be transferred yes so we got a winner in the front row ok ATT so help connect in virtually all health information exchanges have real-time hl7 connections hl7 is like chatter it's constant it's chatter it's not episodic it's chatter it goes on constantly every time someone comes into the hospital admission every time they leave discharge every time they're escalated to a new level of care or moved in the hospital so maybe what does that look like practically so I was doing ok I'm in the hospital now I'm not doing so ok I moved to the ICU intensive care unit not good that's a transfer an ATT message is communicated ok chatter this is sent out back and forth got it so there's different components those are really useful within these components there's different types ok different segments so we've got evey in right event tight tells me kind of what's happening we've got the patient identifiers so here's where it starts to get good we know the name we know date of birth we know race gender stuff we can match on we can match patients this way this is how we start to know that one patient over here is the same patient over here is the same patient over here is the same patient over here this is the good stuff this is how we match across locations then we've got a PV one segment patient visit do they come through the emergency room that's really important how did they get in this hospital was it an emergency visitor was it not if you're an insurance company if you're a primary care physician you want to know how did my patient get in the hospital how come I didn't know about this I didn't admit them oh they came in through the emergency room oh okay I see I got it that makes sense who's the attending physician very good stuff insurance information are we gonna get paid and how are we gonna get paid right really good stuff okay so let's look at this again so now are you starting to see some patterns you starting to see some of that awesome so if you didn't see it before maybe now you're seeing some of this so boom now what I've got I know that somebody came in to Ben Taub I know that it's a hospital a oh one it's an admission I know that now the event type they came in through the emergency room I know that who's my patient Betty Firefox right some other stuff we know where she lives we know her next-of-kin we know some other stuff and I'm just highlighting the big things cool things dr. Gonzalez that's her attending physician important to know she's got an allergy that's really good this is what it does to her gives her hives what does she come in for preeclampsia she's pregnant and having problems this is huge stuff to an insurance company big time and her insurance Texas health plan I made something up so what can we do with this knowing this information that's constantly coming in we can start routing this information it's coming in instantly instantly as soon as it happens as soon as a person comes in so who can I tell who may want to know about this insurance company may want to know oh oh somebody came in what happened to her whom who else might want to know her PCP might want to know her primary care physician her OB Jen might want to know if the OB Jen's listed we can tell the OB Jen immediately somebody else might want to know we can start routing this back and forth we can let people know immediately something is going on because I'm gonna tell you something everything that happens in the ER is off the grid until it's processed by insurance 24 hours to 7 days to sometimes 30 days later none of these people would know about it depending on where you are it could be 30 days later before anybody would ever know that this happened to this woman if we can get on it ahead of time she can be processed as soon as she gets finished leaves another segment is gonna be created so we know follow up with Miss Betty Firefox she needs to go into her OB Jen and see what was happening so we can keep her out of the hospital so that was your ATT listen did everybody get that by the way it's pretty simple right you get more of it and 5,300 says dr. Smith so data sharing is hard this stuff is really really hard what you're doing what you're learning what you'll eventually end up doing this stuff is hard and it's not just about the technology it's not just about the data it's what we do with it it's in the context of how you're gonna do it if you're working with patients it's in the context of there's a patient at the end of this what does that mean if you're a CIO or if you're talking to a CIO it's all it's also in the context of the slide that I showed you with the finances it's good to understand and to know about that and to understand how much does this cost how much is it gonna save what's the time factor gonna be it's also important to know in the larger scheme of things of the organization what's the competitive aspect look like why am I getting resistance why are people not wanting to do this is there competitive aspect to this at all or is that something that I can use to my benefit that can actually help people want to do this but this stuff is hard and to decide how to share what to share how much to share how much not to share that's really tough stuff to figure out so in summation I think that this type of collaborative technology that can sit between healthcare providers is essential I think it's very very important I think it's dramatically understudied I think it's something that needs more of a focus we've got institutions that have spent billions with ABI millions of dollars developing their technology infrastructure but then it stops at the silos of their four walls so all of this all of this information about their patients may not leave but the patients do so we need some type of collaboration technology that extends and travels with the patient because we are all patients if you're not a patient today will be a patient at some time hopefully in a good way hopefully it's delivering a baby or something wonderful like that but academic studies and in the context of an academic environment studying this working on this learning about this working in this environment can really help guide policies decisions and implementations in the right direction on how we do this and how we implement collaboration technology in the best way I also think that there's room for longitudinal assessments of hie s and what's the true value from a financial perspective and then also from a clinical perspective are they really making an impact and if not why not and how can we put that back on the right track and I believe that basically that's my last slide because I think that yeah once again if I haven't drilled into your heads about how important cost is I'm just gonna say it again that driving down cost is very important and that in the context management policy and the leadership so so very very important in a data-driven environment and that is my final slide so it's difficult to measure what's not there what's easier to measure is the duplication aspect so it's kind of like measuring when someone didn't die it's easier to measure when someone did die it's kind of like a mortality study if you will and it kind of philosophically so there is a there is a bevy of research that's coming out now that's saying duplication of tests when it's unnecessary when the test is already there that if you duplicate the test that that is costing money now what you're asking is a slightly different question which I think is very very interesting and it would be something that could be explored yeah it's just trying to structure how would that cost avoidance how would that cost avoidance be but what we're seeing now is the Brooking Institute just actually released a study and it's it was actually published in the last of a Mia so it's the the very last one that I think came out in maybe in June or just in July so it was a single author the gentleman from the Brookings Institute and I can't remem I can't remember his name off the top man but he was saying that in western New York that if the physician was querying the HIA in western New York where they have 100% penetration so all all hospitals are feeding in to the hie so there is information there so what you're saying is what if there's not information there but when there is information there there is a nearly 50% reduction and duplication of tests so there's a huge proof for the reduction side now what you're asking of course is the other side which is a different question right but what you can look at is how many times you're not getting reimbursed from your insurance companies for duplication because that is a concrete line-item on every single hospitals budget so you have a I'm not getting reimbursed in the reimbursement the the lack of reimbursement rate varies but it could be somewhere anywhere between ten percent to twenty percent for duplication of tests so if you're not getting that back on your line item so every CT scan could be between a thousand and fifteen hundred dollars so if you're not getting that back in ten to twenty percent of the cases it it can get pretty pretty high really quickly if you were to invest a hundred thousand dollars in an hie over the course of the year the ROI is is easily made seems quite self-evident what could happen with information where does the patient again I think one of the consequences of meaning we used was providers of the patient portal and we're required to give a patient portals as a doctor somebody didn't realize it those patients have more than one doctors so hybrid seven seven seven patient portals so nobody makes an exchange how does that also evolved into the patient portal of empowerment I just want one portal for myself seven facilities and positions that I use yes so as of right now across the country we haven't seen a lot of H I used to do this successfully and because it is complicated back to my data sharing is hard piece but what you're saying is is echoed everywhere because there are portals that you know patients have this point the epic portal we've got this border we've got this work everybody has the same thing you don't want the same one so it would be an hie that's patient facing basically is what you're talking about no one has done this successfully because the authentication piece and the whole encryption piece and because it is it is utilized you have to be on the EHR to use it and because the patients aren't on an EHR and it's not an EHR type system this type of technology hasn't successfully been brought down in mass from multiple providers to a mobile app if you will that's successful people have done it and tried it but I don't think any of them have been successful to the ends that we would deploy it here in Houston across the providers that we've got across the street no one would would implement that I'm logging into some other thing well you dragged me across from the brand the system that I want them to be attached depends we had what if we paint query call back into a there was so much organizational resistance to that for that reason for members are patients so I think what has to happen if the system has to use an HIV and pull back all the data available color package with a look at paper and MD Anderson yeah I know but it would have to go the other direction individual patient is yep we have multiple positions how do I ever get there data not just their healthcare data but their social security data or other things like that so and your audience is the patient it's you know if you can get enough people on or and if you can build that interfaces correctly hl7 to bring in the data from the different organizations so in thinking about the models so what makes our community happy is you're not making a centralized data repository of all my data we have the top research institutions in the world here on famine and on Maine they do not want MD Anderson Texas Children's and everyone else along the row Memorial Hermann everyone they do not want you duplicating their entire database and putting it into a sbm i sized database okay of a ten ten story building they do not want their information collated together with everyone else's so this decentralized type of approach is the only thing that made everyone happy and will make everyone happy today for this region can you speak to the current status of the accuracy of the patient matching algorithms no they're not so probabilistic matching is is you know I mean I hate to use the phrase it's complicated again but it's kind of a junk in junk out type thing so any kind of interoperability solution that's matching across multiple systems relies on the quality of data inherent in those native systems so if I put in at my clinic Maria s Vasquez and someone else puts in Maria Vasquez then my matching algorithm is going to be knocked off or if I spell Vasquez with an S instead of with a Z my algorithm is gonna be knocked off because that heavy reliance is upon first-name lastname I mean gender date of birth of course but of 16 points in a matching algorithm you've got heavy reliance on first-name lastname you're still gonna get points off for your matching algorithm so what we like to do with our algorithm and I can only speak to the one that we use at Health Connect is we defer on the side of we don't merge anything so everything gets returned back and the physician or the clinician or whoever selecting the patient is left to say is my Maria Vasquez the one that lives on Live Oak or is my Maria Vasquez the one that lives on West Elm and make that final determination and then proceed next into the next layer but there's one step back if you're looking from your native system it's going to match on your medical record number as well so that's a really strong match unless you've got a lot of junk in your own and your own medical record so it's kind of a quality of data at your native system there's a lot of duplicates there's a lot of difference creates crucians the potential of health information exchange she said that there so what I found so my research is primarily on community based not-for-profit hie s so that's in line exactly with what you're saying and the question for anybody I guess who's listening was there are certain restrictions surrounding not-for-profit hie s and selling data and phi2 outside entities it doesn't mean that those entities don't ask and it doesn't mean that those entities don't continue to ask and ask in different in creative ways because they do but they do ask us even to D identify it and we don't do that either identifiable or de-identified so so the question was it sounds like the patient doesn't own their own data so technically the pit is your data the problem is that you you do own your own data but the problem is sometimes it's difficult to get your data does that make sense good luck getting your medical record you are entitled to it you can have a full copy by law to get your entire medical record however there's an administrative process that you would need to go through in order for you to get it and there may actually be a cost for you to get your medical record as well that's a more complicated discussion that would be glad to take up with you afterwards actually yeah yes absolutely so it's not just about the it's not just about the technology I realize some of you come from a straight technology background you're learning about healthcare some of you may come over from a stray healthcare background and you're learning about technology I know one for sure he's been working with us so it's really important to understand both components if you're coming from directly from a from a technology background it's very important to learn about the context in which you will be working so if that's in a hospital environment or you do hope to work in a clinic or a healthcare environment it's very important for you to get some exposure to that environment to get some exposure either by volunteering by visiting there's always always always ways for you to get into a healthcare type environment for you to see what that actually looks like because it's very very important for you to see the kind of setting that you will be wanting to work in and the challenges are different and it's it's it's good for you to understand the whole breadth and what's of what's expected of you because it's not just about the technology it's very it's very important to learn about the data that you'll use

3 Comments

  1. Great explanation of the content

  2. As a student in HIM, I found the information very enlightening. I will be starting to view any information I receive as the Financial information first. The information that you have including "Chatter" is no less important.

  3. Wow. Simple, straight to the point. Well informed, integrated HIE to biomedical informatics and gave straightforward examples that moved the talk along. I was also impressed with the acknowledgement of the importance of the return on investment; i.e., the payback of engaging in HIEs. Also just the right amount of evangelism . TED-worthy information.

Leave a Reply

Your email address will not be published. Required fields are marked *