Throughout my adult life, I've had many opportunities to speak with television, radio, and print journalists. I may not always receive good press, but I almost always receive balanced press.
I've learned several lessons along the way
-It's important not to endorse any product or service. I'm always careful to present my experience in the context of a case study or objective observation. I avoid conflicts of interest by never accepting gifts, travel, or meals from vendors.
-It's important to speak as an individual and not as an organizational representative. I remove my badges and eliminate any organizational logos from the visual field. I emphasize that my comments are personal opinions and do not necessarily reflect the views of any corporation I work for.
-It's important to speak clearly, succinctly, and deliver an unambiguous message. Everyone should be able to understand a 30 second elevator speech about the technology I'm discussing.
-It's important to use personal stories, analogies, and lay language. I often describe my own experience with healthcare and the ways technology would improve my wellness or my family's care coordination.
-It's important to give honest answers. Occasionally, reporters have an agenda and try to put words in your mouth. State your beliefs without being led to a conclusion by redirecting a bad question into an relevant, thoughtful answer.
In the recent iPad2 release (I'm at 11 minutes 44 seconds into the presentation), here are the points I conveyed:
I have watched clinicians (including myself) using tablet style devices. They enhance clinician productivity because of their portability (they fit in a white coat pocket), long battery life, and ease of disinfection (alcohol wipes on the screen do not damage the display). They untether clinicians from laptops and carts, are easy to read, and provide the high definition graphics needed for clinical imaging interpretation.
The nature of the form factor makes it easier than laptops for clinicians to share medical information with patients (such as explaining conditions by retrieving images), improving communications.
My quotes were
“Sometimes doctors are overwhelmed with data. What we’ve tried to do on the iPad is to give doctors at the point of care the tools they need at the exact moment the doctor can make the difference.”
“We’re finding with the iPad that doctors are spending more time with patients. In fact, doctors are engaging patients by showing them images, showing them data on the screen. So it’s empowered doctors to be more productive, and it’s also brought doctors and patients together.”
Some have called this an endorsement. I've followed my own guidelines and described my experience using the device and the behavior of other clinicians I've observed . I did not describe any plan to purchase the devices, nor have I ever received any free iPad products or services.
In a world of YouTube, social media, and blogging, everything I do and say should be considered public. Hopefully, by following my own moral compass and my guidelines for working with the media, I can share my experiences for the benefit of all without compromising my own objectivity.
Thursday, 10 March 2011
Wednesday, 9 March 2011
Meaningful Use 2 and 3 Do It Yourself Presentation
Just as with Stage 1, it's likely that you'll be presenting the proposed Stage 2 and 3 Meaningful Use criteria to your stakeholders and boards.
Here's a comprehensive comparison of Stage 1, 2 and 3, but you'll likely want the Cliff's Notes version for your presentation.
Here's the presentation I'm using in my lectures which highlights the major changes.
Feel free to use it without attribution!
I hope it saves you time.
Here's a comprehensive comparison of Stage 1, 2 and 3, but you'll likely want the Cliff's Notes version for your presentation.
Here's the presentation I'm using in my lectures which highlights the major changes.
Feel free to use it without attribution!
I hope it saves you time.
Tuesday, 8 March 2011
The Open Science Grid
The Open Science Grid (OSG) is an open consortium of science and research communities, university and Department of Energy (DOE) laboratory IT facilities, and software developers. Together they have built and are now operating a broad distributed high throughput computing cyber infrastructure in the US. Staff are funded by the DOE SciDAC-2 program and NSF.
The OSG is designed to be effective for job runs of between one hour and a few of days, jobs that can be check-pointed, jobs that require management of large scale data movement and storage, and ensembles of jobs that can effectively run across a large number of resources.
On a typical day, the OSG supports 1.2 million CPU hours of computation, the launch of more than half a million jobs, and the transfer of more than a quarter of a petabyte of data, across more than sixty sites. Over the past year the Harvard Medical School SBGrid group has used more than 6 million CPU hours.
20% of the usage of OSG Is non-physics – across more than five different scientific disciplines. The Large Hadron Collider (LHC) uses 50% and the existing Tevatron experiments at Fermilab, Laser Interferometer Gravitational Wave Observatory (LIGO), STAR, and other physics experiments use the final 30%.
The Open Science Grid provides the US contribution to the World Wide LHC Computing Grid – most recently presented by Bob Jones.
OSG provides an engagement effort that helps new users, ranging from individuals and small groups to large communities. It engages teams to help them adapt their software (and their culture) to use a distributed set of computing and storage resources which they don’t manage directly themselves.
The OSG is built on a set of underlying principles of distributed high throughput computing. Professor Miron Livny, the lead of the Condor project at the University of Wisconsin-Madison, serves as Technical Director.
OSG partners with the NSF TeraGrid and new XD program. It has a long and productive history of collaboration with peer European projects, continuing with the European Grid Initiative (EGI-InSPIRE) and European Middleware Initiative (EMI). It works closely with ESNet and Internet2 in understanding the networking needs of high performance computing communities, testing and integration of advanced networks.
The OSG brings distributed high throughput computing services into and across campuses themselves by working with groups of faculty and researchers to leverage local and remote resources. OSG is currently working with communities at Clemson, Nebraska, Notre Dame, Purdue and Wisconsin-Madison on a prototyping effort that includes enabling the formation of local partnerships and dynamic access to shared resources using campus identities.
Here's an architecture diagram that shows the scope of OSG. The image above illustrates the usage over the past few years.
OSG is a very worthwhile application of technology - a grid computing initiative that captures millions of unused CPU hours for general use.
Monday, 7 March 2011
The Major Themes of HIMSS 2011
I was in Japan during HIMSS this year, but I asked Jeff Blair, an nationally known informatician from the Lovelace Clinic Foundation, to summarize the key themes from the events he attended. He declared HIMSS 2011 to be the year of the healthcare information exchange with the following points:
"Advances
*There was a better understanding of the capabilities and limitations of NwHIN Direct and NwHIN Exchange. In particular, it seemed as if most attendees now look at NwHIN Direct as a near-term solution and at NwHIN Exchange as having the capabilities to address Stage 2 and Stage 3 of meaningful use.
*A lot of progress has been made during 2010 by independent HIE networks and state HIE networks to develop their strategies, plans, and resources.
*The federal government, including ONC and CMS, has really ramped up to move forward with all of the HITECH initiatives and they are pushing forward on all of these initiatives at the same time.
*Many HIE networks are reporting that they have been able to expand connections to health care provider stakeholders in their communities, and that community support for HIE network services has been growing as HIEs can deliver more services.
*The vendors that provide solutions for HIE networks have matured and are able to more clearly articulate the capabilities of their components, as well as the synergies their components have with other HIE components.
*A lot of progress has been made to define the requirements for provider directories.
*The HIMSS exhibit floor was massive and impressive. There are many new information technologies that have become available and it will be interesting to see how quickly health care can adopt these new technologies (including cloud computing, mobile health, Twitter, etc.) to address health care challenges.
Constraints
*There is greater awareness that the CDA/CCD is not at the level of plug-and-play compatibility, and that work needs to be done to tighten the constraints to limit optionality to get closer to plug-and-play.
*The resources of HIE networks are now stretched thin trying to participate in all of the ONC initiatives, including conferences, committees, communities of practice, workgroups, etc.
*As ONC and CMS roll out more and more funding opportunities, HIE networks are finding that the resources of their local provider partners are also running thin to participate and/or support these funding opportunities.
*It is now clear that the development of provider directories in each state will be major projects, not just an additional component of HIE networks.
*More specifically, it is now clear that provider directories will involve entity-level provider directories (ELPD) and individual level provider directories (ILPD); that the ELPD and the ILPD will need to interact with each other; that there will be many more users of the provider directory than just HIE networks, which means that there will be several different use cases that will need to be created and addressed; and finally, that existing sources of listings of providers at the national, state, and professional association levels will all be needed to create a complete provider directory to support the meaningful use initiatives for Stage 2 and Stage 3.
*Many state HIE networks have lost their State HIE Coordinators, public health supporters, or Medicaid supporters due to the turnover caused by the state elections in November of 2010. It may take several more months to re-establish relationships with new state administrators."
Thanks to Jeff for this thoughtful summary! I'll see you all at HIMSS next year.
"Advances
*There was a better understanding of the capabilities and limitations of NwHIN Direct and NwHIN Exchange. In particular, it seemed as if most attendees now look at NwHIN Direct as a near-term solution and at NwHIN Exchange as having the capabilities to address Stage 2 and Stage 3 of meaningful use.
*A lot of progress has been made during 2010 by independent HIE networks and state HIE networks to develop their strategies, plans, and resources.
*The federal government, including ONC and CMS, has really ramped up to move forward with all of the HITECH initiatives and they are pushing forward on all of these initiatives at the same time.
*Many HIE networks are reporting that they have been able to expand connections to health care provider stakeholders in their communities, and that community support for HIE network services has been growing as HIEs can deliver more services.
*The vendors that provide solutions for HIE networks have matured and are able to more clearly articulate the capabilities of their components, as well as the synergies their components have with other HIE components.
*A lot of progress has been made to define the requirements for provider directories.
*The HIMSS exhibit floor was massive and impressive. There are many new information technologies that have become available and it will be interesting to see how quickly health care can adopt these new technologies (including cloud computing, mobile health, Twitter, etc.) to address health care challenges.
Constraints
*There is greater awareness that the CDA/CCD is not at the level of plug-and-play compatibility, and that work needs to be done to tighten the constraints to limit optionality to get closer to plug-and-play.
*The resources of HIE networks are now stretched thin trying to participate in all of the ONC initiatives, including conferences, committees, communities of practice, workgroups, etc.
*As ONC and CMS roll out more and more funding opportunities, HIE networks are finding that the resources of their local provider partners are also running thin to participate and/or support these funding opportunities.
*It is now clear that the development of provider directories in each state will be major projects, not just an additional component of HIE networks.
*More specifically, it is now clear that provider directories will involve entity-level provider directories (ELPD) and individual level provider directories (ILPD); that the ELPD and the ILPD will need to interact with each other; that there will be many more users of the provider directory than just HIE networks, which means that there will be several different use cases that will need to be created and addressed; and finally, that existing sources of listings of providers at the national, state, and professional association levels will all be needed to create a complete provider directory to support the meaningful use initiatives for Stage 2 and Stage 3.
*Many state HIE networks have lost their State HIE Coordinators, public health supporters, or Medicaid supporters due to the turnover caused by the state elections in November of 2010. It may take several more months to re-establish relationships with new state administrators."
Thanks to Jeff for this thoughtful summary! I'll see you all at HIMSS next year.
Friday, 4 March 2011
Cool Technology of the Week
In a world focused on green energy, conservation, and working more efficiently, why do we run escalators when no one is using them?
In Japan, they do not.
Above is a video from the Narita airport near Tokyo which shows the escalators to the gates. They run at a very slow speed, just enough to overcome the inertia of starting them up. When a passenger walks near the escalator platform, the rate increases and the escalator runs at full speed until 30 seconds after the passenger leaves the escalator. Here's a link to another example from the Seoul airport.
The end result is that energy is only expended when its needed.
It's this kind of creative thinking - using energy only when its needed - that defines our green power future.
On demand escalators - that's cool.
That's cool!
In Japan, they do not.
Above is a video from the Narita airport near Tokyo which shows the escalators to the gates. They run at a very slow speed, just enough to overcome the inertia of starting them up. When a passenger walks near the escalator platform, the rate increases and the escalator runs at full speed until 30 seconds after the passenger leaves the escalator. Here's a link to another example from the Seoul airport.
The end result is that energy is only expended when its needed.
It's this kind of creative thinking - using energy only when its needed - that defines our green power future.
On demand escalators - that's cool.
That's cool!
Thursday, 3 March 2011
A Perfect Weekend in Kyoto
On Sunday, we started the day with a great Japanese breakfast of Yudofu (boiled Tofu), pickles, rice, and tea at our ryokan in Central Kyoto, the Watazen. After breakfast, we walked south to the Buddhist temples of Higashi and Nishiki Honganji. We purchased coils of our favorite incense from Kungyoku-do, near Nishiki Honganji. From there, we walked east, crossing the Kamagawa river and passing Sanjusangen-do, the temple where samurai used to practice archery (you can still find the arrow holes in the beams). From there, we climbed Kiyomizu-yama on our way to the temple of Kiyomizu-dera, which is surrounded by pottery shops. We bought a pair of rustic tea cups, and a simple sake cup and bottle set made of fine clay at the Asahido pottery shop.
From there, we walked north through the park of Maruyama-koen and the shrine of Yasaka-jinja, then up Shijo-dori to buy incense at another great incense store adjacent to the Kyoto Craft Centre.
We walked back to the Kamagawa river and north to Sanjo-dori, passing our favorite rice cracker shop, Funahashi-Ya. We stopped for lunch at the conveyor belt sushi restaurant Musashi which has numerous vegetarian options.
Afterwards, my wife and daughter explored the shopping streets of Teramachi dori/Shinkyogoku, while I walked to north Kyoto and the temple of Ginkaku-ji. Just to the right is an unmarked path that leads up Daimonji, a 1500 foot peak overlooking Kyoto. At the top, I met a mountain biker named Yoshi, who gave me directions for a 5 mile ridge walk along the mountains between Kyoto and Lake Biwa, ending at the temple of Nanzenji, the starting point for the Philosopher's Walk. From there, I walked back to our ryokan and we went to dinner at our favorite tofu restaurant on Sanjo dori near the Kamagawa river. A great day with remarkable incense, pottery, and foods to take back to Boston.
On Monday after breakfast, we walked the Nishiki market street, with its wonderful pickles, tofu, yuba (tofu skin), and roasting chestnuts. We purchased dried yuba to prepare great stews back in Boston. From there, we returned to the Teramachi shopping street to visit our favorite tea shop, Horaido. The owner made us fine Gyokuro tea and provided precise instructions so we could make it at home. The key is to use 105F water for the first cup, creating a sweet. concentrated, viscous tea. We purchased Gyokuro and Sencha tea, as well as some cherry bark tea caddies to store our tea. If you visit me in my office at Harvard or BIDMC, I'll brew you a cup.
From there, we walked to the Imperial Palace and the textile areas of northwest Kyoto. It's always great to find clothing dyed with indigo at shops like Aizen Kobo.
From there we walked east and south, making our way back to the Gion, the Geisha quarter with its wooden shop fronts, antiques, and plum blossoms blooming over the canals lined with tea houses (pictured above) . We walked the antique shops of Shinmonzen, browsing in our favorite shop, Yagi, where I found several old Shakuhachi which the shop owner let me play. I purchased an old hammered incense scoop which I'll use for Koh-do, the incense ceremony.
At sunset, we wandered past the secretive lanes and alleys where Geisha and Maiko entertain their clients at chaya teahouses, with conversation and music. It was truly a magical moment.
We packed our treasures and readied for the commute home - 26 hours from point to point. Of all the cities in the world outside of the US, Kyoto is the one I am always reluctant to leave.
Wednesday, 2 March 2011
Freeing the Data
I'm keynoting this year's Intersystems Global Conference on the topic of "Freeing the Data" from the transactional systems we use today such as Enterprise Resource Planning (ERP), Customer Relationship Management (CRM), Electronic Health Records (EHR), etc. As I've prepared my speech, I've given a lot of thought to the evolving data needs we have in our enterprises.
In healthcare and in many other industries, it's increasingly common for users to ask IT for tools and resources to look beyond the data we enter during the course of our daily work. For one patient, I know the diagnosis, but what treatments were given to the last 1000 similar patients. I know the sales today, but how do they vary over the week, the month, and the year? Can I predict future resource needs before they happen?
In the past, such analysis typically relied on structured data, exported from transactional systems into data marts using Extract/Transform/Load (ETL) utilities, followed by analysis with Online Analytical Processing (OLAP) or Business Intelligence (BI) tools.
In a world filled with highly scalable web search engines, increasingly capable natural language processing technologies, and practical examples of artificial intelligence/pattern recognition (think of IBM's Jeopardy-savvy Watson as a sophisticated data mining tool), there are novel approaches to freeing the data that go beyond a single database with pre-defined hypercube rollups. Here are my top 10 trends to watch as we increasingly free data from transactional systems.
1. Both structured and unstructured data will be important
In healthcare, the HITECH Act/Meaningful Use requires that clinicians document the smoking status of 50% of their patients. In the past, many EHRs did not have structured data elements to support this activity. Today's certified EHRs provided structured vocabularies and specific pulldowns/checkboxes for data entry, but what do we do about past data? Ideally, we'd use natural language processing, probability, and search to examine unstructured text in the patient record and figure out smoking status including the context of the word smoking such as "former", "active", "heavy", "never" etc.
Businesses will always have a combination of structured and unstructured data. Finding ways to leverage unstructured data will empower businesses to make the most of their information assets.
2. Inference is possible by parsing natural language
Watson on Jeopardy provided an important illustration of how natural language processing can really work. Watson does not understand the language and it is not conscious/sentient. Watson's programming enables it to assign probabilities to expressions. When asked "does he drink alcohol frequently?", finding the word "alcohol" associated with the word "excess" is more more likely to imply a drinking problem than finding "alcohol" associated with "to clean his skin before injecting his insulin". Next generation Natural Language Processing tools will provide the technology to assign probabilities and infer meaning from context.
3. Data mining needs to go beyond single databases owned by a single organization.
If I want to ask questions about patient treatment and outcomes, I may need to query data from hundreds of hospitals to achieve statistical significance. Each of those hospitals may have different IT systems with different data structures and vocabularies. How can a query a collection of heterogenous databases? Federation will possible by normalizing the queries through middleware. For example, data might be mapped to a common Resource Description Framework (RDF) exchange language using standardized SPARQL query tools. At Harvard, we've created a common web-based interface called SHRINE that queries all our hospital databases, providing aggregate de-identified answers to questions about diagnosis and treatment of millions of patients.
4. Non-obvious associations will be increasingly important
Sometimes, it is not enough to query multiple databases. Data needs to be linked external resources to produce novel information. For example, at Harvard, we've taken the address of each faculty member, examined every publication they have ever written, geo-encoded the location of every co-author, and created visualizations of productivity, impact, and influence based on the proximity of colleagues. We call this "social networking analysis"
5. The President's Council of Advisors on Science and Technology (PCAST) Healthcare IT report will offer several important directional themes to will accelerate "freeing the data".
The PCAST report suggests that we embrace the idea of universal exchange languages, metadata tagging with controlled vocabularies, privacy flagging, and search engine technology with probabilistic matching to transform transactional data sources into information, knowledge and wisdom. For example, imagine if all immunization data were normalized as it left transactional systems and pushed into state registries that were united by a federated search that included privacy protections. Suddenly every doctor could ensure that every person had up to date immunizations at every visit.
6. Ontologies and data models will be important to support analytics
Part of creating middleware solutions that enable federation of data sources requires that we generally know what data is important in healthcare and how data elements relate to each other. For example, it's important to know that an allergy has a substance, a severity, a reaction, an observer, and an onset data. Every EHR may implement allergies differently, but by using common detailed clinical model for data exchange and querying we can map heterogeneous data into comparable data.
7. Mapping free text to controlled vocabularies will be possible and should be done as close to the source of data as possible.
Every industry has its jargon. Most clinicians do not wake up every morning thinking about SNOMED-CT concepts of ICD-10 codes. One way to leverage unstructured data is to turn it into structured data as it is entered. If a clinician types "Allergy to Pencillin", it could become SNOMED-CT concept 294513009 for Pencillins. As more controlled vocabularies are introduced in medicine and other industries, transforming text into controlled concepts for later searching will be increasingly important. Ideally, this will be done as the data is entered, so it can be checked for accuracy. If not at entry, then transformations should be done as close to the source systems as possible to ensure data integrity. With every transformation and exchange of data from the original source, there is increasing risk of loss of meaning and context.
8. Linking identity among heterogenous databases will be required for healthcare reform and novel business applications.
If a patient is seen in multiple locations how can we combine their history together so they get the maximum benefit of alerts, reminders, and decision support? Among the hospitals I oversee, we have persistent linkage of all medical record numbers between hospitals - a master patient index. Surescripts/RxHub does a realtime probabilistic match on name/gender/date of birth for over 150 million people in real time. There are other interesting creative techniques such as those pioneered by Jeff Jonas for creating a unique hash of data for every person, then linking data based on that hash. For example John, Jon, Jonathan, and Johnny are reduced to one common root name John. "John" and the other demographic fields are then hashed using SHA-1. The hashes are compared between records to link similar hashes. In this way, records about a person can be aggregated without ever disclosing who the person really is - it's just hashes that are used to find common records.
9. New tools will empower end users
All users, not just power users, want web-based or simple to use client server tools that allow data queries and visualizations without requiring a lot of expertise. The next generation of SQL Server and PowerPivot offer this kind of query power from the desktop. At BIDMC, we've created web-based parameterized queries in our Meaningful Use tools, we're implementing PowerPivot, and we're creating a powerful hospital-based visual query tool using I2B2 technologies.
10. Novel sources of data will be important
Today, patients and consumers are generating data from apps on smart phones, from wearable devices, and social networking sites. Novel approaches to creating knowledge and wisdom will source data from consumers as well as traditional corporate transactional systems.
Thus, as we all move toward "freeing the data" it will no longer be ufficient to use just structured transaction data entered by experts in a single organization, then mined by professional report writers. The speed of business and the need for enhanced quality and efficiency is pushing us toward near real time business intelligence and visualizations for all users. In a sense this mirrors the development of the web itself, evolving from expert HTML coders, to tools for content management for non-technical designated editors, to social networking where everyone is an author, publisher, and consumer.
"Freeing the data" is going to require new thinking about the way we approach application design and requirements. Just as security needs to be foundational, analytics need to be built in from the beginning.
I look forward to my keynote in a few weeks. Once I've delivered it, I'll post the presentation on my blog.
In healthcare and in many other industries, it's increasingly common for users to ask IT for tools and resources to look beyond the data we enter during the course of our daily work. For one patient, I know the diagnosis, but what treatments were given to the last 1000 similar patients. I know the sales today, but how do they vary over the week, the month, and the year? Can I predict future resource needs before they happen?
In the past, such analysis typically relied on structured data, exported from transactional systems into data marts using Extract/Transform/Load (ETL) utilities, followed by analysis with Online Analytical Processing (OLAP) or Business Intelligence (BI) tools.
In a world filled with highly scalable web search engines, increasingly capable natural language processing technologies, and practical examples of artificial intelligence/pattern recognition (think of IBM's Jeopardy-savvy Watson as a sophisticated data mining tool), there are novel approaches to freeing the data that go beyond a single database with pre-defined hypercube rollups. Here are my top 10 trends to watch as we increasingly free data from transactional systems.
1. Both structured and unstructured data will be important
In healthcare, the HITECH Act/Meaningful Use requires that clinicians document the smoking status of 50% of their patients. In the past, many EHRs did not have structured data elements to support this activity. Today's certified EHRs provided structured vocabularies and specific pulldowns/checkboxes for data entry, but what do we do about past data? Ideally, we'd use natural language processing, probability, and search to examine unstructured text in the patient record and figure out smoking status including the context of the word smoking such as "former", "active", "heavy", "never" etc.
Businesses will always have a combination of structured and unstructured data. Finding ways to leverage unstructured data will empower businesses to make the most of their information assets.
2. Inference is possible by parsing natural language
Watson on Jeopardy provided an important illustration of how natural language processing can really work. Watson does not understand the language and it is not conscious/sentient. Watson's programming enables it to assign probabilities to expressions. When asked "does he drink alcohol frequently?", finding the word "alcohol" associated with the word "excess" is more more likely to imply a drinking problem than finding "alcohol" associated with "to clean his skin before injecting his insulin". Next generation Natural Language Processing tools will provide the technology to assign probabilities and infer meaning from context.
3. Data mining needs to go beyond single databases owned by a single organization.
If I want to ask questions about patient treatment and outcomes, I may need to query data from hundreds of hospitals to achieve statistical significance. Each of those hospitals may have different IT systems with different data structures and vocabularies. How can a query a collection of heterogenous databases? Federation will possible by normalizing the queries through middleware. For example, data might be mapped to a common Resource Description Framework (RDF) exchange language using standardized SPARQL query tools. At Harvard, we've created a common web-based interface called SHRINE that queries all our hospital databases, providing aggregate de-identified answers to questions about diagnosis and treatment of millions of patients.
4. Non-obvious associations will be increasingly important
Sometimes, it is not enough to query multiple databases. Data needs to be linked external resources to produce novel information. For example, at Harvard, we've taken the address of each faculty member, examined every publication they have ever written, geo-encoded the location of every co-author, and created visualizations of productivity, impact, and influence based on the proximity of colleagues. We call this "social networking analysis"
5. The President's Council of Advisors on Science and Technology (PCAST) Healthcare IT report will offer several important directional themes to will accelerate "freeing the data".
The PCAST report suggests that we embrace the idea of universal exchange languages, metadata tagging with controlled vocabularies, privacy flagging, and search engine technology with probabilistic matching to transform transactional data sources into information, knowledge and wisdom. For example, imagine if all immunization data were normalized as it left transactional systems and pushed into state registries that were united by a federated search that included privacy protections. Suddenly every doctor could ensure that every person had up to date immunizations at every visit.
6. Ontologies and data models will be important to support analytics
Part of creating middleware solutions that enable federation of data sources requires that we generally know what data is important in healthcare and how data elements relate to each other. For example, it's important to know that an allergy has a substance, a severity, a reaction, an observer, and an onset data. Every EHR may implement allergies differently, but by using common detailed clinical model for data exchange and querying we can map heterogeneous data into comparable data.
7. Mapping free text to controlled vocabularies will be possible and should be done as close to the source of data as possible.
Every industry has its jargon. Most clinicians do not wake up every morning thinking about SNOMED-CT concepts of ICD-10 codes. One way to leverage unstructured data is to turn it into structured data as it is entered. If a clinician types "Allergy to Pencillin", it could become SNOMED-CT concept 294513009 for Pencillins. As more controlled vocabularies are introduced in medicine and other industries, transforming text into controlled concepts for later searching will be increasingly important. Ideally, this will be done as the data is entered, so it can be checked for accuracy. If not at entry, then transformations should be done as close to the source systems as possible to ensure data integrity. With every transformation and exchange of data from the original source, there is increasing risk of loss of meaning and context.
8. Linking identity among heterogenous databases will be required for healthcare reform and novel business applications.
If a patient is seen in multiple locations how can we combine their history together so they get the maximum benefit of alerts, reminders, and decision support? Among the hospitals I oversee, we have persistent linkage of all medical record numbers between hospitals - a master patient index. Surescripts/RxHub does a realtime probabilistic match on name/gender/date of birth for over 150 million people in real time. There are other interesting creative techniques such as those pioneered by Jeff Jonas for creating a unique hash of data for every person, then linking data based on that hash. For example John, Jon, Jonathan, and Johnny are reduced to one common root name John. "John" and the other demographic fields are then hashed using SHA-1. The hashes are compared between records to link similar hashes. In this way, records about a person can be aggregated without ever disclosing who the person really is - it's just hashes that are used to find common records.
9. New tools will empower end users
All users, not just power users, want web-based or simple to use client server tools that allow data queries and visualizations without requiring a lot of expertise. The next generation of SQL Server and PowerPivot offer this kind of query power from the desktop. At BIDMC, we've created web-based parameterized queries in our Meaningful Use tools, we're implementing PowerPivot, and we're creating a powerful hospital-based visual query tool using I2B2 technologies.
10. Novel sources of data will be important
Today, patients and consumers are generating data from apps on smart phones, from wearable devices, and social networking sites. Novel approaches to creating knowledge and wisdom will source data from consumers as well as traditional corporate transactional systems.
Thus, as we all move toward "freeing the data" it will no longer be ufficient to use just structured transaction data entered by experts in a single organization, then mined by professional report writers. The speed of business and the need for enhanced quality and efficiency is pushing us toward near real time business intelligence and visualizations for all users. In a sense this mirrors the development of the web itself, evolving from expert HTML coders, to tools for content management for non-technical designated editors, to social networking where everyone is an author, publisher, and consumer.
"Freeing the data" is going to require new thinking about the way we approach application design and requirements. Just as security needs to be foundational, analytics need to be built in from the beginning.
I look forward to my keynote in a few weeks. Once I've delivered it, I'll post the presentation on my blog.
Subscribe to:
Posts (Atom)
