An HIT Moment with ... is a quick interview with someone we find interesting. Dan Nigrin, MD, MS is senior VP for information services & CIO, Division of Endocrinology & Informatics Program, Children’s Hospital Boston; assistant professor of pediatrics, Harvard Medical School; and affiliated faculty, Harvard-MIT Division of Health Sciences and Technology.
Mainstream applications always struggle with peds-specific logic. Does pediatrics really need to be that much different and are vendors doing enough to support it?
Yeah, pediatrics definitely requires specific functionality that, in the past especially, didn’t come "out of the box" for many systems on the market. Good examples include growth chart support, weight-based dosing, gestational-age based dosing, and for some centers like ours at Children’s Boston, being able to document interventions performed on fetuses while they’re still in utero!
Overall, I think that vendors have made strides in many of these areas. For example, our Cerner system now includes good functionality for weight and gestational age-based dosing and decision support to go along with them, and the growth chart functionality allows for custom, disease-specific growth charts to be loaded by the client.
But we’ve still got a way to go. Case in point: in 1997 I wrote a Web-based application called GrowthCalc to allow specialists at our institution to calculate various anthropometric values on their patients. Today, over 10 years later, it’s still used on a daily basis at institutions around the world because nothing better is out there. It’s not that my work was that fancy or special; it’s just that it fills a niche that hasn’t yet been included in the EMR systems on the market.
What are the five most promising systems or technologies being used or considered right now at Children’s?
Five? OK, you asked for it – I’ll move from the micro level, the patient, outwards to the hospital level, and then to the macro level, the region.
- Our MyChildren’s patient portal, which we are now rolling out to all of our patients. In addition to the usual stuff found in tethered patient portals (e.g. appointment requests, billing inquiries and online payment, demographic updates, secure clinician messaging), we’ve also seamlessly built in Indivo functionality to allow our patients to have a secure, portable, personally controlled health record (PCHR) that is automatically fed by our Children’s clinical systems. We’ve had discussions with eClinicalWorks, whose systems we are putting into our affiliated pediatric practice network offices, to likewise automatically feed those patient’s Indivo records with their primary care information. Most importantly, with the patient’s consent, clinicians within Children’s will have single-click access to the patient’s Indivo record from within our EMR environment so that, unlike most personal health records out there now, our clinicians will have ready access to the information that our patients are maintaining.
- iAware. Another new Cerner offering, this is an always-on system intended to be displayed at the patient’s bedside, likely in a critical care environment. Our intensive care unit clinicians had found it difficult to get a good overview of the patient when they had to click through various parts of the chart to find the bits of data they needed to synthesize the patient’s status. This new approach aggregates the key data elements, including vital signs, labs, meds, and inputs/outputs, and shows them in a very intuitive and graphically rich way. From a technology point of view, there’s nothing to it – we deployed it live in three weeks in our intermediate care unit – but from a clinician’s point of view, it’s priceless. It actually takes the data that we work so hard to collect electronically and presents it to clinicians in a useful way (what a novel idea!).
- MA-SHARE. Building on the success of the New England Healthcare EDI Network (NEHEN), which allowed New England payers and providers to exchange administrative transactions in a secure way, MA-SHARE (Simplifying Healthcare Among Regional Entities) is allowing organizations in the New England area to exchange clinical data. Our primary focuses right now are on exchange of CCD documents between organizations as well as the facilitation of ePrescribing in our region. This is a RHIO done right – a sustainable, beneficial model.
- Catalyst / i2b2 / SHRINE. Harvard University was recently awarded one of the NIH CTSA grants to further clinical and translational science across the country. A major focus of our proposal (now called Catalyst) centered on IT and its ability to tie together the various people and projects Harvard-wide. Using i2b2 querying tools developed at Harvard and now deployed at four major Harvard teaching hospitals (Beth Israel, Brigham & Women’s, Children’s, and Mass General), we also are working on SHRINE, which will allow us to execute clinical queries across these institutions. So investigators will soon be able (with IRB approval) to ask questions like, "How many patients are seen at each institution with disorder X who also have lab value Y and who are on medication Z?" Powerful stuff …
Children’s has a notable informatics training program. What influence does their scientific work have on the practical side of the healthcare IT market?
The Children’s Hospital Informatics Program or CHIP is a biomedical informatics multidisciplinary applied research and education program that’s been in place at Children’s since the mid 90s. Although its roots were definitely in clinical informatics, it is now also a leader in functional genomics, public health informatics, and personalized medicine. What’s more, its members understand that all of these things are interrelated and that their true benefit comes when they’re not looked at in isolation. For example, the genotype is worthless without phenotype information to go with it.
Examples of ways in which CHIP’s work has had influence on mainstream healthcare IT include:
- Distributed querying. Some of the earliest work from CHIP included a system called W3-EMRS, which allowed queries for a patient’s data to be distributed across multiple organizations. It was implemented first as a pilot and then successfully used at Caregroup, when it was first formed to virtually integrate the disparate EMR’s that each institution brought to the table. Similar models are now used in several RHIO efforts. In addition, this distributed query approach is now the basis for SPIN, the shared pathology information network; and SHRINE, described above.
- Secondary re-use and mining of clinical data. We realized long ago that the treasure trove of clinical data being acquired by EMR systems was largely underutilized. In the late 90s, I developed the Goldminer system at Children’s, which allowed for much easier investigation of the data stored in our systems for clinical research. This was soon followed by work at Partners in the Research Patient Data Registry (RPDR), and which in turn led to the development of i2b2. i2b2 is now implemented in many institutions nationwide, and although open source, there are commercial vendors out there who specialize in its implementation.
- Public health informatics. We’ve done quite a bit of work using existing data sources for public health related functions. Aegis performs automated, real-time surveillance for bioterrorism and naturally occurring outbreaks. It is the syndromic surveillance system for the Massachusetts Department of Public Health, enabling real-time population health monitoring. HealthMap is another CHIP project that was funded by Google.org to gather and display information from news sources around the world about infectious diseases.
- Genomics. CHIP members pioneered the use of relevance networks in the analysis of both genetic and clinical information, and they literally wrote the book on using microarrays. They continue to lead the field.
- Personally Controlled Health Records. Indivo.
How did Indivo come about and what impact will it have on healthcare?
About a decade ago, researchers in CHIP developed the open source Indivo. It was actually called PING back then. It was, essentially, the world’s first PCHR. It enables patients to own complete, secure copies of their medical records. A good analogy is that it’s like a Quicken for healthcare.
It is amazing to think how far Indivo, and the idea of putting patients in control of their health information, has come in that time. It still seems futuristic to some, but we expect PCHRs to be universally available and used in the very near future. There’s been a lot of buzz around PCHRs since Microsoft and Google announced theirs; what people may not know is that both companies’ deployments are fundamentally based on the Indivo model. There’s even a rumor out there that MS’s HealthVault actually contains some Indivo code under its hood. Both companies were present at our two Personally Controlled Health Record Infrastructure conferences we hosted at Harvard in 2006 and 2007.
As many people know, Dossia has also adopted the Indivo infrastructure, and in fact Wal-Mart just went live, offering our Indivo-based PCHR to 1.4 million employees and their dependents.
Indivo and PCHRs in general will have a major impact on healthcare. With PCHRs, patients will be able to aggregate and share almost all of the information in their medical records such as lab tests, medications, and clinical notes, which in the past has been largely inaccessible to them. We see this leading to improved communications and continuity of care with clinicians, and the ability to provide more complete and accurate information to health care providers than the current system allows.
We also see this as exciting for the biomedical research enterprise. With PCHRs, researchers may be able to recruit with patient consent hundreds, thousands, possibly millions of patients from all over the world for their studies, potentially speeding up the time it takes to bring research to the bedside.
While this is all very exciting, there is a lot of work to be done if PCHRs are to reach the full extent of their potential. In a recent New England Journal of Medicine article, my colleagues Ken Mandl and Zak Kohane call for attention and regulation as various PCHRs are developed and adopted. Without it, it’s possible that the tremendous benefits of PCHRs could be overshadowed by problems arising from the unethical and uncontrolled use of valuable medical information.
Which title have you found to be the best for impressing strangers: doctor, CIO, Harvard professor, or the guy behind the Defective Records electronic music label? How do you find the time to do all that stuff and which ones require wearing a tie?
The last, by far – the first three things are a dime a dozen! Seriously though, my music creation and record label stuff, and more recently software synthesizer development, are all great hobbies that I wish I had more time for. How do I do them all? Jack of all trades, master of none?? Oh, and about the tie – if you believe my friend and across-the-street colleague John Halamka, you don’t need a tie for any of ’em, just a black mock turtleneck!