Pathology Informatics:Past, Present, and Future
Mike Hogarth, MD, FACP UCDHS Medical Dir. of Clinical RegistriesProfessor and Vice Chair, Pathology and Lab MedicineProfessor, Internal [email protected]
Overview
• What is “informatics”?• What is “medical informatics”?• A history of medical informatics• History of UCDHS informatics (in pathology)• What is “pathology informatics”?• UCDHS pathology informatics – what do we do?• Thoughts on pathology informatics and the future of
healthcare -- ?what does the future hold?
What is “Informatics”
It is the science of “information”....
What is “medical informatics”?
"The study, invention and implementation of structures and algorithms to improve communication, understanding and management of medical information." --- Homer Warmer
"Medical information science is the science of using system-analytic tools . . . to develop procedures (algorithms) for management, process control, decision making and scientific analysis of medical knowledge.” --- Ted Shortliffe
Informatics does not always require computers – Nightingale was a pioneer in medical data analytics
Notes on matters affecting the health, efficiency and hospital administration of the British army – Florence Nightingale (1858)-- demonstrated that sanitation was a significant cause of the 42.7% hospital mortality rate in Barrack Hospital, Scutari, Istanbul
Crimean War 1853-1856
Florence Nightingale
1858
Nightingale pioneered the use of data and statistics to improve health
• Developed a Model Hospital Statistical Form• 1858 – Became Fellow of the Royal Statistical Society• 1874 – Honorary Member, American Statistical
Association• 1860 – Sent letter to Intl. Statistical Congress president
advocating for uniform collection of hospital statistics so outcomes could be compared – the first model for the systematic collection of hospital data using a uniform classification of diseases (ICD)
Really, what is medical informatics...?
• A multi-disciplinary field that is focused on the science of information
• Examples of what you might see in the field• Medical decision making models• Clinical decision support• Information organization and flow• Algorithms• User interface design and usability (error causing?)• Medical image processing• Modeling of medical observations and data• Ontology and terminology systems• Methods for evaluating healthcare information systems• Development and design of clinical information systems
Informatics draws from many fields
What medical informatics is NOT...
• It is not “information technology” • It uses information technology as a tool... much like a pathologist
might use a microscope... A pathologist is not a “microscopist” but a diagnostician!
• It is not “computers in medicine”• It studies various aspects of information in healthcare, but it is
not simply the implementation of computerization of the process• Some excellent informatics has been done *without* computers!
• It is not “healthcare programming”• We often build systems to implement models of information
management but informatics is not about the building of these systems..
• Informatics uses software design/development to achieve its goals (and study its benefits/detrimental effects of such systems)
Medical Informatics is not a new field...
Robert Ledley – a pioneer in using computers in medicine
Robert Ledley -Physicist and DDS-Pioneered the use of digital computers in biology and medicine-Developed system to automate analysis of chromosomes (FIDAC)-Led the development of the Automatic Computerized Transverse Axial (ACTA) scanner...the CT scan!-National Medal of Technology
The direct application of computing to medical practice is also not new...
• Dr. Morris Collen hired by Dr. Sidney Garfield (Kaiser Permanente)
• 1952 – Became Medical Director, Kaiser, Oakland• 1961 – Heard presentation on automating clinical data
capture and processing• mid 1960’s - Implemented computerized “multi-phasic”
testing (health screening)• Kaiser was one of the first to computerize medical data
Medical informatics is not just for young healthcare professionals...
Computerized medical records are also not new...
• 1965 – El Camino Hospital (Mt. View, CA) and Lockheed collaborate on the “Technicon Systems” electronic medical record system
• 1972 – Technicon Medical Information System goes “live” in El Camino hospital
• Physician order entry• Clinical documentation
directly from health provider• Results reporting
Bill Creelman of Technicon performing training onSystem with nursing staff.
UCDHS and Medical Informatics
• 1973 – Sacramento Medical Center is bought from Sac county by UC Davis Medical School
• 1976 - Dick Walters, PhD – Faculty in Computer Science Dept. also joins faculty in Family Practice
• 1985 - Dr. Walters elected to ACMI
• co-developer of MUMPS, a programming language in clinical systems
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1130476/
1976
UC Davis Informatics Milestones
• 1985 - Dr. Cardiff (Chair, DOPLM) and Dr. Walters create the “Center for Medical Informatics”
• 1999 – Health Informatics master’s program
• 1987 – 1st “web-cast” informatics symposium featuring Dr. Don Lindberg
• 2000 – UCD DOPLM develops hybrid tele-pathology system
Medical Informatics evolves in the US
AMIA Definition of Biomedical Informatics and related fields
Medical Informatics evolves in the US
AMIA Definition of Biomedical Informatics and related fields
Pathology Informatics-- Association of Pathology Informatics (API)
• 2000 – founded with the goal of defining pathology informatics as a clinical subspecialty within pathology• Develops standards for managing pathology-related
information• Promotes education in pathology informatics
• “Pathology informatics involves collecting, examining, reporting, and storing large complex sets of data derived from tests performed in clinical laboratories, anatomic pathology laboratories, or research laboratories in order to improve patient care and enhance our understanding of disease-related processes”
UCDHS Pathology Informatics
• 2009 – Pathology Informatics formally established in DOPLM
• Mission – to apply information science to solve challenges in the discipline of pathology
So what is “pathology informatics” here?• The application of information science to support the scope of
pathology practice (or research)• What does it entail?
• Laboratory Information Systems• Managing clinical information systems supporting anatomic pathology, clinical
pathology, transfusion medicine, and molecular diagnostics• Laboratory automation
• Digital Pathology Imaging Systems• Image capture methods, diagnostic fidelity • Whole-slide scanning systems, tele-pathology systems, Image archives/path-
PACS
• Data mining (of pathology information)• Natural language processing of surgical pathology reports
• Decision support• Provide CDSS for diagnostic testing by health providers
• Analytics and Registries • to support performance/quality improvement
UCDHS Pathology Informatics: 2014-2015
• Information Management• Implement, operate/manage systems that support UCDHS anatomic pathology,
clinical pathology, transfusion medicine• Integration of LIS data with EHR• Integration of LIS data with external referral labs (ie, ARUP)• Whole slide scanning systems (in collab with health IT)
• Education• PGY Pathology informatics rotation• Health informatics certificate for pathology fellows• Health informatics master’s program (leadership, teaching courses)• Master’s of Public Health (public health informatics)
• Grants/Contracts -- ~$25M in grants/contracts since 2005• California Electronic Death Registration System (2005)• Maryland EDRS (2014)• MediCal “Big Data” analytics and EHRs (2014)• TRANSCEND Clinical Trial Information System (supporting I-SPY2 since 2009)• pSCANNER – PCORI Clinical Data Research Network (2014)• UC-ReX (founding member, 2009)• Athena Breast Health Network (2008)• Oversight of UCDHS clinical registries (2011)• UC Health Quality Improvement analytics (2014)
UCDHS Pathology Informatics
http://www.ucdmc.ucdavis.edu/pathinformatics
Pathology Informatics Staff
What does the future hold?
• Trends in healthcare1. The pursuit of the Triple Aim
Increasing focus on quality/cost (value)• DSRIP (Delivery System
Reform Incentive Payments) program
• Accountable Care Organizations (ACOs)
1. Tailored diagnostics and therapeutics
2. Wearable devices 3. “Big Data” and “Data
Science” 4. “Open Data” movement
(health.data.gov)
“The Triple Aim”
21st Century US Healthcare
The Learning Healthcare System
Sarah M. Greene, Robert J. Reid, Eric B. Larson; Implementing the Learning Health System: From Concept to Action. Annals of Internal Medicine. 2012 Aug;157(3):207-210.
TMR Transfusion Registry
Transfusion registry leverages data from the Laboratory Information System (LIS), clinical data from UC Davis Health System EHR, and administrative (coded) data
Registries a new way
A Tethered Meta-Registry
-“Meta-Registry” – All data for all registries is in one repository
-“Tethered” – routine, automated data extraction from source systems
-One data set – algorithms “tag” patients as being in one or more registries
-Automated dashboards and reports
A Tethered Meta-Registry
-“Meta-Registry” – All data for all registries is in one repository
-“Tethered” – routine, automated data extraction from source systems
-One data set – algorithms “tag” patients as being in one or more registries
-Automated dashboards and reports
Did you know? “80% of data elements in institutional registries are common across
all of them”
Did you know? “80% of data elements in institutional registries are common across
all of them”
Tethered Meta Registry Data Flow
Sepsis Registry
Mobility (ICU)
Registry
Diabetes Registry
Source Data“Tether”
EMR Reporting Database
Administrative Data
Laboratory Information
System
TMR Patient
TMR Encounters
TMR Flowsheets
TMR Procedures/Labs
TMR Medications
2.2 Million
25 Million
100 Million
57 MillionCancerRegistry
Database
TMR Cancer Diagnosis
Transfusion Registry
Cancer Registry
A glimpse into the possible future...
http://www.businessweek.com/articles/2014-07-03/hospitals-are-mining-patients-credit-card-data-to-predict-who-will-get-sick#r%3Dread
Pathology Informatics and future trends
• “Team Medicine” / Integrated care• Will require tight data integration between EHR and LIS functions• Consolidation of LIS as a core EHR component
• Computerized clinical decision support • Will require structured data from pathologists• Structured reporting (synoptic reporting) using XML CAP “electronic checklists”
(eCC XML)• We are currently collaborating with ASCO and CAP on improving structured data
capture in cancer care
• Population health management • Will require clinical data analytics and registries• Combining data from multiple sources for predictive analytics.• Creation/management of data sets (registries) to support performance
improvement within pathology, as well as externally• Analytics platform to characterize care
• Patient-centered care • Will require new ways of “reporting” information
• Patient-directed reports?• Mobile/mhealth
Questions...
Slide deck download – http://www.hogarth.org