Highlights

Focus on CHIP

Uncovering Patterns for Advancing Healthcare

CHIP case study

The Boston Children's Hospital Informatics Program (CHIP) is a multidisciplinary applied research and education program at Boston Children's Hospital. Under the direction of Dr. Isaac Kohane, 22 CHIP faculty investigators work at the intersection of information science and clinical and biomedical discovery.

CHIP's faculty include distinguished leaders in the fields of personally controlled medical records; coordination and analysis of dispersed medical record information; and the aggregation, analysis and display of large, complex datasets.

IndivoHealth

IndivoHealth™ lays the foundation for the realistic application of Personally Controlled Health Records (PCHRs), enabling patients to own complete and secure copies of their medical records--and integrating health information across sites of care and over time. IndivoHealth is a project of the CHIP, based on a decade of development by Drs. Isaac Kohane, Kenneth Mandl, Dan Nigrin, and many others.

The software application platform is built to public standards and is made available as open-source code for wide adoption. Using this platform, Indivo has already developed and deployed PCHR systems in real-life settings. The latest and largest deployment was announced in September 2008, in collaboration with the Dossia consortium, Walmart is rolling out an Indivo-based PCHR system to all of its 1.4 million employees and their dependents.

References

Mandl KD, Kohane IS. Tectonic shifts in the health information economy N Engl J Med. Apr 17 2008;358(16):1732-1737.

Mandl KD, Simons WW, Crawford WCR, Abbett JM. Indivo: a personally controlled health record for health information exchange and communication BMC Medical Informatics and Decision Making 2007, 7:25 doi:10.1186/1472-6947-7-25

For More Information See: http://www.indivohealth.org/ [Press Release]

Intelligent Histories

Dr. Ben Reis has developed a longitudinal screening software tool for the prediction of future diagnoses.

Identifying warning signs for life-threatening medical events from slow developing chronic conditions is hampered by the scattering of personal medical information across multiple care providers and by the massive amount of information in the medical history presented to the physician at any patient visit. By aggregating medical records and applying Bayesian models of longitudinal medical histories, the resulting "Intelligent Histories" are able to assess an individual's risk of a future diagnosis. Tested on a de-identified, state-wide claims database covering 6 years of inpatient hospitalizations, observation admissions, and emergency department encounters, Intelligent Histories produces sensitive, specific and timely identification of patients at high risk for a specific future diagnosis, months or years in advance of the actual diagnosis by a physician.

The analysis has a broad application to predict outcomes for slow-to-develop medical indications, and could provide a new window to implement preventive intervention, avoiding costly and life-threatening complications.

HealthMap

Developed by CHIP investigators Dr. John Brownstein and Clark Freifeld, HealthMap provides a unified and comprehensive view of the current global state of infectious diseases and their effects on human and animal health.

The HealthMap application filters and synthesizes data from a variety of internet sources, ranging from conventional news media reports to internet blogs, in multiple languages, then extracts the relevant disease-specific information tied to its location in the world. In essence, HealthMap provides a bird's-eye view of global health, with interactive maps and color-coded alerts of infectious "hot spots," and is able to sound warnings about outbreaks well before they are reported by public health sources. "We've traced the earliest reports of the SARS virus back to Internet chat rooms where people were talking about a problem going on in Guangdong Province in China," notes Brownstein.

Additional development of HealthMap is funded through a $3 millon grant from Google.org, to combine HealthMap with the ProMED-mail global alert system for health specialists run by the International Society of Infectious Diseases. Together, ProMed-mail and HealthMap aim to identify infectious disease hot spots, detect outbreaks earlier, and initiate quick responses.

A view of HealthMap is freely available at HealthMap.org, integrated into Google's GoogleMap viewer.

Access to raw data feeds of the HealthMap content for customized use is available through a licensing agreement. Also, applications are in development for the intelligent extraction of relevant information from diverse Internet sources, and the display of geopositional data correlated with a wide array of other extracted intelligence.

References:

Surveillance Sans Frontières: Internet-Based Emerging Infectious Disease Intelligence and the HealthMap Project. Brownstein JS, Freifeld CC, Reis BY, Mandl KD (2008) PLoS Med 5(7): e151

HealthMap: Global infectious disease monitoring through automated classification and visualization of Internet media reports. Freifeld CC, Mandl KD, Reis BY, Brownstein JS (2007) J Am Med Inform Assoc.

HealthMap: the development of automated real-time internet surveillance for epidemic intelligence. Brownstein JS, Freifeld CC (2007) Euro Surveill 12: E071129 071125.

HealthMap: Internet-based emerging infectious disease intelligence. Brownstein JS, Freifeld CC, Reis BY, Mandl KD (2007) In: Institute of Medicine, editor. Infectious Disease Surveillance and Detection: Assessing the Challenges - Finding Solutions. Washington, DC. pp. 183-204.

For More Information See: http://www.healthmap.org/ [Press Release]

i2b2

i2b2 (Informatics for Integrating Biology and the Bedside) was born out of frustration on the part of clinicians and researchers with the slow speed of translational medicine. Despite the vast amounts of genomic data and validated research results, treatments and cures based on this information are slow to develop. Dr. Isaac Kohane directs i2b2, an NIH-funded National Center for Biomedical Computing based at Partners HealthCare Systems, that is developing a health information system designed to enable the medical community to apply what is already known about medicine and individual patients to the development of diagnostics and therapeutics.

The i2b2 Center is developing a scalable framework of informatics applications to bridge clinical research data and the vast data banks arising from basic science research in order to better understand the genetic bases of complex diseases. The platform is structured as a “Hive” of smaller software units, developed separately, that are designed to work together to answer many different scientific and medical questions. This knowledge will lead to the design of targeted therapies for individual patients with diseases having genetic origins.

The i2b2 Hive has standardized the way clinical data is gathered within the participating medical centers, making such data available for rapid and powerful searches. For example, using simple queries, de-identified patient data can be mined to show side effects and adverse events related to a drug much more quickly than it is currently. On the scientific side, a researcher has access to all the information that is known about a gene or protein of interest, as well as information about potentially useful research tools and models that could be used to study it.

The i2b2 system is customizable for a variety of uses, including the development of smarter, cheaper clinical trials and post-market surveillance by pharmaceutical companies, and the monitoring of patient safety and quality of care by hospitals.

For More Information See: http://www.i2b2.org/

For More Information on the Childen's Hospital Informatics Program: http://www.chip.org

Children's HospitalTerms and Conditions
Site Designed and Developed by Genuine