Public Health and Regional Informatics

Public Health andRegional Informatics

Mark FrisseNovember 18, 2008

see: http://sites.google.com/a/mfrisse.com/www/home/2008-11-18-bmif-300-lecture

what we will cover

• What is public health?• What is regional informatics?• What are the common themes?• What are the challenges?• What are the research and service

opportunities?

differences and similarities

• populations, not individuals• prevention more than diagnosis and

treatment• government more than providers• same! my claim is that the informatics

issues required to address public health are the same as those for many other pressing clinical problems

the textbook answersthree functions• Assessment involves monitoring and tracking

the health status of populations.• Policy development utilizes the results of

assessment activities in concert with local values and culture to recommend interventions and policies that improve health status.

• Assurance refers to the mission public health agencies have to assure constituents that services necessary to achieve agree-upon goals are provided.

big deals

• unsuccessful siege of the Assyrians against Jerusalem (701 BCE)

• guns, germs, and steel• the great influenza• HIV• drug-resistant TB, Staph, other stuff• immunizations

more big deals

• lack lung in miners• asbestos workers• back injury and other occupational-

related disorders - $2.3 billion dollars• fluoridated water changed dentistry• seat belts• high fat foods• tobacco

my view

• relationships people have with one another

• coordinated systems of prevention, detection and care

• analysis and presentation of signals• interventions

reporting

• federal & stateissues• completeness• accuracy• efficiency• latency• privacy and confidentiality

federal (examples)AIDSAmebiasisAnthraxAseptic meningitisAsthma, work-relatedBotulismChancroidChlamydia trachomatis genital infectionCholeraCoccidioidomycosisCryptosporidiosisCyclosporiasisDenguerDiphtheria

HIV InfluenzaKawasaki DiseaseLegionellosisLeptospirosisListeriosisLyme diseaseLymphogranuloma venereum (LGV)MalariaMeasles, totalMeningococcal diseaseMumpsNeurosyphilisNon-gonococcal urethritis (NGU)

PIDPertussisPlaguePoliomyelitisRabiesSalmonellosisShigellosisSilicosisSmallpoxSpinal cord injuryStrep. pneumoniae, invasive disease < 5 years Streptococcal disease, invasive, group AStreptococcal toxic-shock syndromeStreptococcus

pneumoniae, drug-resistantSyphilisTetanuToxic-shock syndrome (other than streptococcal)ToxoplasmosisTrichinellosisTuberculosisTularemiaTyphoid feverTyphus feverVancomycin-resistant Staphylococcus aureus (VRSA)Varicella

not all are infections

• Head injury• Lead poisoning• Motor vehicle injury• Sudden infant death syndrome• Suicide

tennessee

ph-1600

completeness

• automatic reporting of health conditions may lead to 4x the number of incidents

• this means 4x as much work for public health professionals…

• unless…you can give them access to a community-based record

cool people and projects

• McMurray, Kohane, Mandl• Grannis and Overhage• Mostashari• Wagner

SPIN (McMurray et. al.)

SPIN features

• is self-scaling, voluntary and hence may be applicable to a national network

• employs a distributed approach to data storage that they argue minimizes breach and hence protects privacy.

• maintains institutional participation because of the autonomy relegated by a distributed approach.

• provides oversight and transparency

cdc public health informatics grid• a need for wide distribution of public health data.• rapid growth of public health data.• cultural, social and political impediments to data

sharing.• significant and chronic financial constraints.• a dynamic and complex environment - global in scale.• an environment containing many redundant systems,

as well as application and data silos.• an environment with a wide variety of complex

requirements (disease surveillance, alerting, event detection, etc).

surveillance: retail-style

new ways

analysis: signal vs. noise

• Analysis• Case detection algorithms• Time-series methods• Combining multiple signals• Spatial and spatial-temporal

clustering • Modeling

markle principles• openness and transparency.• accountability and oversight• individual participation and control• purpose specification and minimization• collection limitation• use limitation• data integrity and quality• security safeguards and controls• legal and financial remedies for violations

cholera

whosissick.org

memphis

• Baptist Memorial Health Care Corp. (4 facilities)

• Christ Community Health (4 primary care clinics)

• Methodist Healthcare (7 facilities including Le Bonheur Children’s Medical Center)

• The Regional Medical Center (The MED)

• Saint Francis Hospital & St. Francis Bartlett (Tenet Healthcare)

• St. Jude Children’s Research Hospital

• Shelby County/Health Loop Clinics (11 primary care clinics)

• UT Medical Group (300+ clinicians)

• Memphis Managed Care/TLC (MCO)

The MidSouth eHealth Alliance

After 18 months of operation• Total # of encounter records: 3.9 million• Total # of patients: 1,050,000• Total # of patients with clinical data: 930,000• Monthly Encounter Data: 140,000• Monthly ICD-9 admission codes (Chief complaints): 34,000• Monthly labs: 2,400,000 • Monthly microbiology reports: 26,000• Monthly chest x-ray reports: 35,000• Comprehensive privacy agreements• Costs to participants less that $50,000 per hospital• Overall annual operating cost – under $3 million

ArchitectureHealth Care

Entity InternalSystems

Vaults Regional Index

Volunteer eHealth Initiative Data Bank

Identifier Information- Patient Identifier numbers- Facility identifier- Patient name- Date of birth- Gender- Social security number

Data- Demographics- Lab- Orders

“Pharmacy”

Identifier Information- Patient Identifier numbers- Facility identifier- Patient name- Date of birth- Gender- Social security number

Data- Demographics- Lab- Orders

Clinic

Identifier Information- Patient Identifier numbers- Facility identifier- Patient name- Date of birth- Gender- Social security number

Data- Demographics- Lab- Transcribed reports- Pharmacy- Orders

Hospital

Identifier Information- Patient Identifier numbers- Facility identifier- Patient name- Date of birth- Gender- Social security number

Data- Demographics- Lab- Transcribed reports- Pharmacy- Orders

Person 1CompositeInformation

Link 1

Link n

Person 2CompositeInformation

Link 1

Link n

Person 3CompositeInformation

Link 1

Link n

::

Person nCompositeInformation

Link 1

Link n

“Laboratory”

Record Locator Service

Record Access Service

Parsing/Integration Engine

Publish Data

Publish Data

Publish Data

Printer

FAXServer

WebUser

Exchange receives data & manages data transformation

• Mapping of Data• Parsing of Data• Standardization of Data• Queue Management

Data is published from data source to the exchange• Participation Agreement• Patient Data• Secure Connection• Batch / Real-Time

Organizations will have a level of responsibility for management of data

• Issue Resolution• Data Integrity• Entities are responsible

for managing their Data

Data bank compiles and aggregates the patient Data at the regional level• Compilation Algorithm• Authentication

• Security• User Access

Use• > 400 users• Low in ED (< 5%)• Growing use in safety

net clinics• hospitalists usage low• Increasing connectivity to ambulatory

sites• Reduces redundant tests; impacts

care

visualization?

public health / HIE

issues• completeness• accuracy• efficiency• latency• privacy and confidentiality

completeness

• more data but more ways of managing information at the point of decision-making

accuracy

• clear data integrity checks because the data are the same used for clinical care

efficiency

• data are collected “on the margin.”• you no longer have separate

systems, you have one, single, amorphous system whose use is dictated by need and authorization

• everything becomes a marginal cost

latency

• detection? none…nada…zip• only the time it takes the brain to

process and the system to intervene

privacy - agreements• openness and transparency.• accountability and oversight• individual participation and control• purpose specification and minimization• collection limitation• use limitation• data integrity and quality• security safeguards and controls• legal and financial remedies for violations

the real lesson• our health care system is broken• our health care system is fragmented• wherever you go - be it personal health, pay-

for-performance, public health, information exchange, or public policy - you face the same issues

• a unified approach based on a very simple, extensible technical and policy framework seems, in my mind, to be the only way informatics can help enable the health care system we all want and need.