Big Data for Oral Health Knowledge: What is Big Data?

Many may be familiar with this map:

This image is in the public domain due to its age. Author died in 1858, material is public domain.

Original map made by John Snow in 1854. Cholera cases are highlighted in black. Click here for larger version.

The map depicts clusters of cholera cases in the London epidemic of 1854; the cholera cases are highlighted in black. The physician, John Snow, studied the outbreak by mapping the data from an EHR onto Google Maps and discovered that cholera is spread by contaminated water. Well kind of—it was the 19th century after all. He recorded the details about the cases by talking to residents and illustrated them as dots on a map. While some state that this was the founding event of the science of epidemiology, my question is the following: Is this an example of Big Data?

While there is probably no correct answer to this question, Dr. Snow clearly combined two disparate data sources, medical records of confirmed cases and location data, and visualized the results to gain insight. His discovery influenced public health and the construction of improved sanitation facilities. He proved that the cause of cholera stemmed from contaminated water from specific pumps, debunking the popular belief that cholera spread by breathing bad air. Dr. Snow integrated unstructured data sources to find correlations without definitive causation, as he was unable to show the bacterium Vibrio cholerae by chemical and microscope examination. Thus, my answer is the following: Yes, John Snow was a Big Data user. What do you think? One and a half centuries later, are we sufficiently integrating oral health data to make an impact on dental public health?

The preparations for the Big Data for Oral Health Knowledge; How to get the Big Answers workshop scheduled for March 14, 2015 are progressing nicely. We are excited that a few more invitees have accepted our invitation—again, just to name a few: Dr. Bill Kohn (Delta Dental), Dr. Jane Weintraub (Dean of the School of Dentistry at the University of North Carolina at Chapel Hill), Dr. Muhammad Walji (Associate Dean for Technology Services & Informatics at the School of Dentistry, University of Texas Health Science Center at Houston), Dr. Damien Walmsley (ADEE) and Dr. David DuChene (OmicronHealth).

Please feel free to share your thoughts with us.

Chris, Khalil, Lynn and Heiko

Christopher H. Fox, DMD, DMSc Executive Director International & American Associations for Dental Research

Khalil Yazdi, PhD CIO in Residence, Cloud Services Program Development at Internet2

Lynn Johnson, PhD Professor of Dentistry, School of Dentistry Associate Dean for Faculty Affairs and Institutional Effectiveness Clinical Professor, School of Information The University of Michigan

Heiko Spallek, DMD, PhD, MSBA Associate Professor, Dental Public Health Associate Dean for Faculty Affairs Executive Director, Center for Informatics in Oral Health Translational Research School of Dental Medicine, University of Pittsburgh

Big Data for Oral Health Knowledge: Toward a Complete and Sustainable Learning Health System

The preparations for the Big Data for Oral Health Knowledge; How to get the Big Answers workshop scheduled for March 14, 2015 are progressing nicely.

On January 23, 2015, I had the pleasure to attend Dr. Chuck Friedman’s talk “Toward a Complete and Sustainable Learning Health System” at the University of Pittsburgh. Chuck is the Josiah Macy Jr. Professor of Medical Education and Chair of the Department of Learning Health Sciences at the University of Michigan, he held executive positions at the Office of the National Coordinator for Health IT in the U.S. Department of Health and Human Services, including as the Deputy National Coordinator and as Chief Scientific Officer and is the former associate director for research informatics and information technology of the National Heart, Lung and Blood Institute at the NIH. At Pitt, he had served as associate vice chancellor for biomedical informatics and founding director of the Center for Biomedical Informatics.

Chuck outlined his vision about the Learning Health System (LHS), a system that can continuously and routinely study and improve itself. He pointed out that there is growing recognition that the nation requires an LHS to provide higher quality, safer, and more affordable health care–and to empower biomedical research and public health. The LHS is an ultra large-scale information infrastructure, effectively a smart grid for health that aggregates data from disparate sources, convert the data to knowledge, and disseminate that knowledge, in actionable forms, to everyone who can benefit from it. Reports from the Institute of Medicine, many journal articles, and current federal policies argue strongly for achievement of this “Big Hairy Audacious Goal” to improve individual and population health. I was amazed by his presentation that succinctly described the LHS concept, framing it as a consummate informatics challenge that invokes in equal parts people and technology. He also argued that the LHS must be viewed as an investment in infrastructure that includes, but transcends, the currently popular topic of “Big Data”.

Chuck is an excellent presenter who drew analogies between the LHS and the “moon shot” of the 1960s and the building the Panama Canal. The latter more akin to that of the LHS as it required, among several feats of engineering, the conquest of an infectious disease and creation of a new nation. Chuck concluded that solving the problems prerequisite to achieving an LHS may require a significantly new science of learning systems.

I am delighted that Chuck will present his vision of the LHS at our Workshop on March 14th stressing the imperative of the LHS and how it is currently envisioned, the features it requires to be effective and sustainable, and the key questions a science of learning systems must address. He also asks all Workshop participants to think about the following question: If we get the Big Answers from Big Data, how do we put them to use to ensure that they change practice and improve health?

We are excited that a few more invitees have accepted our invitation—again, just to drop a few names: Dr. David Clark (Behavioral and Social Sciences Research Branch, NIDCR), Dr. Ira Lamster (Professor, Department of Health Policy & Management, Dean Emeritus, College of Dental Medicine, Columbia University), Dr. Sheila Riggs (Director, Office of Community Engagement for Health, Clinical Translational Science Institute and Chair, Department of Primary Dental Care School of Dentistry, University of Minnesota) and Dr. Cecile Feldman (Dean, Rutgers School of Dental Medicine and ADEA President-Elect).

Please feel free to share your thoughts with us.

Chris, Khalil, Lynn and Heiko

Christopher H. Fox, DMD, DMSc Executive Director International & American Associations for Dental Research

Khalil Yazdi, PhD CIO in Residence, Cloud Services Program Development at Internet2

Lynn Johnson, PhD Professor of Dentistry, School of Dentistry Associate Dean for Faculty Affairs and Institutional Effectiveness Clinical Professor, School of Information The University of Michigan

Heiko Spallek, DMD, PhD, MSBA Associate Professor, Dental Public Health Associate Dean for Faculty Affairs Executive Director, Center for Informatics in Oral Health Translational Research School of Dental Medicine, University of Pittsburgh

Big Data for Oral Health Knowledge: Are we making a big mistake?

The preparations for the Big Data for Oral Health Knowledge; How to get the Big Answers workshop scheduled for March 14, 2015 are progressing nicely. The title for this blog post was paraphrased from Tim Hartford’s provocative article in The Financial Times:

Hartford summarizes in his article some of the inherent problems around Big Data. As always in technology, there is a lot of hype around a new concept and many snake oil sales people who attempt to promise easy solutions when in fact there are rarely easy ones available. It goes without saying that we need causation in medicine and dentistry and cannot rely on correlation alone. This is especially true as we need to remember that all our efforts in healthcare are ultimately geared towards the future—we are not (for the most part) historians. In other words, retrospective studies are only a means to figure out what can be done for future patients (dead people do not benefit from an autopsy). Or, to turn the argument around: We need prospective studies that allow the individual provider to determine what is the best therapy for the individual patient at hand—pure correlation without hypothesis and analysis of reasons will not get us there.

The other concerns associated with Big Data relate to the lack of randomization. Providers often wish that they could compare their patient at hand with all patients who had similar symptoms and conditions. Based on the aggregated outcomes for different therapy choices “for patients like mine” the provider could then determine which therapy is the best option for their individual patient. While this sounds like an intriguing concept, we should not forget that we cannot assume that the assignment of the treatment was randomly made like in a clinical trial—severe cases of a condition, for instance, might get preferably one of the available therapy options. Thus, results, even when using n=ALL, might not tell us which therapy is truly the best option, but only which one is most frequently applied.

How would we benefit if we had access to all patient data at all times from all places—I really mean ALL?

Lastly, we are excited that a few more invitees have accepted our invitation to the by-invitation-only workshop on March 14, 2015—again, just to drop a few names: Dr. Gregg Gilbert (PI of the National Dental PBRN), Dr. Chuck Friedman (Chair, Department of Learning Health Sciences, Medical School, Professor of Information and Public Health, University of Michigan), Dr. David Gesko (Dental Director and Senior Vice-President of HealthPartners Dental Group) and Dr. Elsbeth Kalenderian (Chair, Oral Health Policy and Epidemiology, Chief of Quality, Harvard Dental Center).

Please feel free to share your thoughts with us.

Chris, Khalil, Lynn and Heiko

Christopher H. Fox, DMD, DMSc Executive Director International & American Associations for Dental Research

Khalil Yazdi, PhD CIO in Residence, Cloud Services Program Development at Internet2

Lynn Johnson, PhD Professor of Dentistry, School of Dentistry Associate Dean for Faculty Affairs and Institutional Effectiveness Clinical Professor, School of Information The University of Michigan

Heiko Spallek, DMD, PhD, MSBA Associate Professor, Dental Public Health Associate Dean for Faculty Affairs Executive Director, Center for Informatics in Oral Health Translational Research School of Dental Medicine, University of Pittsburgh

Data Management

The preparations for the Big Data for Oral Health Knowledge; How to get the Big Answers workshop scheduled for March 14, 2015 are well under way. The majority of the invitees have already accepted the invitation to the by-invitation-only workshop, such as Dr. Kathy Atchinson (UCLA, consulting for NIDCR), Dr. John Kusiak (Senior Advisory to the NIDCR Director and Acting Deputy, who will be playing an active role in the Big Data Research initiatives for NIDCR/NIH), Dr. Mary Marazita (one of the top-funded genetics researchers in dentistry), Dr. Patrick Ferrillo (Dean at the University of Pacific and representative of IFDEA), Hiroshi Ogawa (WHO Oral Health Program), Dr. Richard Valachovic (Executive Director, ADEA), Dr. Helen Whelton (Dean, University of Leeds), Dr. Peter Polverini (former Dean, University of Michigan), and Dr. Martha Somerman (Director, NIDCR).

During our preparatory meetings, Dr. Seth Weinberg, a faculty member at the University of Pittsburgh who is involved with several large research initiatives, recently conveyed to us that he believes data sharing – which is a key step in generating big data – should start with proper data management that makes research data available to the outside world in a way that can be used by others. He sees a lack of database management capacity as one of the biggest obstacles for Big Data.

We share his concerns and have found a related 4-minute video (that we wish we had produced, but it is from NYU):

This “data management horror story” might resemble some of the past experiences researchers have faced when it comes to storage, documentation and file formats.

Stay tuned to learn here at the DIOC blog more about the workshop.

Christopher Fox, Khalil Yazdi, Lynn Johnson, Heiko Spallek

Christopher H. Fox, DMD, DMSc Executive Director International & American Associations for Dental Research

Khalil Yazdhi, PhD CIO in Residence, Cloud Services Program Development at Internet2

Lynn Johnson, PhD Professor of Dentistry, School of Dentistry Associate Dean for Faculty Affairs and Institutional Effectiveness Clinical Professor, School of Information The University of Michigan

Heiko Spallek, DMD, PhD, MSBA Associate Professor, Dental Public Health Associate Dean for Faculty Affairs Executive Director, Center for Informatics in Oral Health Translational Research School of Dental Medicine, University of Pittsburgh