Hospital Information Systems: Where weve come from and where were going

1
Hospital Information SystemsWhere weve come
from and where were going
Jonathan Pell, M.D.

Assistant Professor, Hospital Medicine
IS
Physician Liaison

University of Colorado at Denver and Health
Sciences Center
Tuesday Morning Conference Denver Veteran
Affairs Medical Center January 20th 2009
2
Objectives

• What is a Hospital Information System (HIS) and
  why should I care?
• Brief history of hospital HISs
• Problems with development of HIS
• Barriers to clinician adoption of new
  technologies
• Barriers to hospital adoption of HIS
• Potential future directions for HISs

3
Government employee
4
An Hour in the Life of a Hospitalist

• Starting your 7pm-7am shift and get sign-out from
  4 daytime teams (8-10 patients each)
• ED calls you with a new admission
• Nurse calls about pt Xs headache 30min later
• Finally get to the ED to admit patient
• Get back to the floor and sign orders

5
History of Computers
First microprocessors and PCs late 1970s
Punch card data processing 1890
Wireless computers late 1990s
General purpose computers 1950
First minicomputer late 1960s
World Wide Web early 1990s
First digital computer 1940
6
Original Hospital Information Systems (HIS)

• 1962 Initiated by Bolt, Beranek and Newman and
  carried out by Octo Barnett at MGH
• Funded by NIH whose biggest concern was not
  enough MD input

7
(No Transcript)
8
Other HIS Pioneers

• Warner at Latter Day Saints hospital, Utah
• Collen at Kaiser Permanente, California
• Wiederhold at Stanford University

9
Progression of Computer Use in Hospitals
10
One System for all?

• Departmental systems became feasible in 1970s
• Departmental systems develop tailored to
  specific application areas
• No common databases or database systems
• Best of breed theory begins to develop

11
What makes up a HIS of today

• Admission, discharge, and transfer system (ADT)
• Electronic Medical Record (EMR)
• Picture Archiving and communication (PACS)
• Pharmacy
• Labs (including microbiology, pathology)
• Billing and Scheduling
• Active patient data systems (ER, Med/surg, OR,
  ICU)

12
(No Transcript)
13
Electronic Health Record (EHR) Needs

• Accessible
• Secure
• Acceptable to clinicians
• Acceptable to patients
• Integrated with both patient specific and patient
  nonspecific information

14
Data that goes into an EHR
Clinician orders
Patient Demographics and billing
Patient phone calls
Patient specific lists -problem list
-medication list
Prescriptions and medications administered
Labs, microbiology, pathology, and radiology
results
Active patient information
-Vital signs -Is and
Os
Clinician visit notes -ER visits
-Hospitalization
summaries
Procedure Reports
15
Problem Lots of forms of Data

• Free text
• Lists of text (problem lists)
• Numbers with titles and error ranges (labs)
• Images in multiple forms (ECG,CXR)
• Multiple note formats
• Text with numbers (prescriptions)
• Trends of numbers (in hospital vitals, labs)

Shortliffe, EH (2006)
16
What do we want coming out of an EHR?
Clinician orders
Patient Demographics and billing
Patient phone calls
Patient specific lists -problem list
-medication list
Prescriptions and medications administered
Labs, microbiology, pathology, and radiology
results
Active patient information
-Vital signs -Is and
Os
Clinician visit notes -ER visits
-Hospitalization
summaries
Procedure Reports
17
And MoreEHR Functional Components

• Clinical Decision Support clinical system,
  application or process that helps health
  professionals make clinical decisions to enhance
  patient care defined by HIMSS
• Integrated view of patient data
• Clinician Order Entry
• Access to Knowledge Resources
• Integrated communication and reporting support
• E-prescription when patients are discharged

18
How do solve the multiple data form problem?

• Original Solution- Substitution
• Display information we already have on computer
  screen
• What we need- Transformation
• Rethink how we obtain patient information and
  manage patients
• Understand computer technology to change how we
  think about patient data use

19
How Physicians Enter Data

• Transcription- dictated or written notes
• Filling out structured encounter forms
• Direct data entry

20
The Informatics World Solution Coding

• Problem You cant put the art of medicine into
  code (at least not easily)
• Coding Systems
• ICD-9 (International Classification of Disease)
• SNOMED (Systemized Nomenclature of Medicine)
• CPT (Current Procedural Terminology)
• LOINC (Laboratory Observations, Identifiers,
  Names, and Codes)
• Arden Syntax medical decision logic

21
Lost in Translation
Amount given 60meq, Site Medication
administered P.O., Correct patient, time,
route, dose and medication confirmed prior to
administration. Patient advised of actions and
side-effects prior to administration,
Allergies confirmed and medications reviewed
prior to administration. (1926 CK1)
Follow Up Decreased symptoms. (2129
DVB) ORDERS BMP BASIC METABOLIC PANEL by TAI
for BA on Wed Dec 31, 2008 1806 Status Done
by System Wed Dec 31, 2008 1858. PHOSPHORUS
SERUM/PLASMA by TAI for BA on Wed Dec 31, 2008
1806 Status Done by System Wed Dec 31, 2008
1858. CBC COMPLETE HEMATOLOGY PROFILE by TAI
for BA on Wed Dec 31, 2008 1806 Status Done
by System Wed Dec 31, 2008 1824. MAGNESIUM
SERUM by TAI for BA on Wed Dec 31, 2008 1806
Status Done by System Wed Dec 31, 2008
1858. CT BRAIN by TAI for BA on Wed Dec 31,
2008 1808 Status Cancelled by System Wed
Dec 31, 2008 1820. XR SHOULDER 3 VIEW
INCLUDING AXILLARY by TAI for BA on Wed Dec 31,
2008 1815 Status Cancelled by System Wed Dec
31, 2008 1820. MR BRAIN by CK1 for CK1 on Wed
Dec 31, 2008 2043 Status Cancelled by
System Wed Dec 31, 2008 2107. XR CHEST PA LAT
by CK1 for CK1 on Wed Dec 31, 2008 2104 Status
Done by System Wed Dec 31, 2008 2214.
22
Narrative Text vs Coded Data

• Narrative PMedHx
• DMII diagnosed 10 yrs ago now on insulin with
  last A1c 10.6 (12/15/08) suspectedly due to poor
  medication compliance
• Chronic renal insufficiency secondary to diabetes
  with 1g proteinuria and baseline creatinine 2.1
  (12/15/08)
• Coded PMedHx-
• 250.42 (DM 2 uncontrolled with renal
  complications)

23
Benefits

• Text
• Easy to document and interpret
• Comprehensive and fully customizable
• Good for individual patient care
• Coded Data
• Aggregate analysis
• Well defined for billing
• Information system friendly

24
Data-Interchange Standards

• International Standards Organization (ISO)s Open
  Standards Institure (OSI) seven levels required
  for data exchange
• HL7 (Health Level 7) - Data interchange
• Digital Imaging Communications in Medicine
  (DICOM) for PACS
• National Council for Prescription Drug Programs
  (NCPDP) - pharmacy
• ASTM 1238 lab information interchange

25
Partial Solutions

• Extensive Interface Engine hardware, software
  ,and support
• At a minimum, difficult interfaces result in
  steep learning curves and structural
  inefficiencies in task performance. At worst,
  problematic interfaces can have serious
  consequences in patient safety

Lin at al Applying human factors to the design of
medical equipment. J. of Clin. Monitoring and
Computing.14(4) 253-263.1998.
26
Transfer of patients between different systems

• Medications dropped from lists
• Redundant admission orders written
• Documented patient information from previous
  system lost or difficult to interpret
• Orders dropped on transfer
• Medications mistakenly given twice

27
Database standards
28
Single Vendor or Best of Breed

• Few single vendors out there
• Epic
• Meditech
• Cerner
• McKesson
• GE/IDX
• No longer best of breed in each department

29
Who is looking at the big picture?

• HIMSS- Health Care Information and Management
  Systems Society
• IHE- Integrating the Healthcare Enterprise
• CCHIT-Certification Commission for Healthcare
  Information Technology
• HITSP- Healthcare Information Technology
  Standards Panel

30
HITSP Programs of work topics

• Lab results reporting
• Bio-surveillance
• Consumer empowerment
• Emergency Responder-HER
• Quality
• Medication management
• Personalized Healthcare
• Consultations and transfers of care
• Immunizations and response
• Patient-provider secure messaging
• Remote monitoring

31
Clinician Barriers to IT system implementation
and change
32
Clinician Barriers to IT system implementation
and change

• Clinician prefer computer use for consultation
  but do not like data entry
• Opposed to extra effort unless clear benefit
• Do not like the inflexibility
• Disrupts time for the clinician patient encounter
  
• Clinicians dont like change

Mcdonald et al 1992.
33
What do Clinicians Care About

• Does it have the information we are used to
  having
• What is its usability
• Learnability
• Efficiency
• Memorability
• Minimization of Errors
• Satisfaction

Nielson 1993
34
IT Industry Response

• More code devoted to Graphic User Interface
• Understanding needs of different users
• Understanding workflow
• Budgets spent on usability increasing
• Implementation budgets increasing

35
What do hospitals care about?

• Cost reduction
• Productivity enhancement
• Quality Improvement
• Competitive Advantage
• Regulatory Compliance

36
2008 HIMSS Leadership Survey
37
National Level

• The Computer-Based Patient Record An Essential
  Technology for Health Care -IOM report in 1991
  and revised in 1997
• National commitment of 50 billion dollars over 5
  years toward electronic health record for all?
• IT czar in Washington
• RHIOs and Potential for a National Health
  Information Infrastructure (NHII)

38
NHII

• Idea first raised in 2001 by the National
  Committee on Vital and Health Statistics
• Distributed system of databases using standards
  for access
• Benefits in
• Cost of Care
• Compliance with national guidelines
• Public health notification
• Research

39
Physician Visit of the Future

• Patient physician interaction is voice
  recognition recorded into standard history format
• Physical exam is performed and commented on by
  device peripherals
• Physician uses Tablet PCs or PDAs to review
  vitals, radiology, labs, and clinician notes,
  etc.
• All physician orders are entered through the
  device and incorporated into note for plan
• E and M billing recommendations made and verified
• All this information could be viewed by itself
  and in aggregate from anywhere securely

40
Whats Happening at UCH

• Evaluating use of a single vendor-Epic
• Single database and interface system
• CPOE
• Decision support
• Customized user views of patient information
• CORHIO participation

41
References
Barnett, GO. History of Medical Informatics
Proceedings of ACM conference on History of
medical informatics .Bethesda, Maryland, United
States, 43 49, 1987. Barnett, GO. Computers
and Patient Care N. Eng. J. of Med.1968. 269
1321-1327. Nielson 1993 Usability Engineering.
Boston, Academic Press. Mcdonald, C.J. et al The
Regenstrief medical record system 20 years of
experience in hospitals, clinics, and
neighborhood health centers. MD Computing. 9
(1992) 206-217. Lin at al Applying human factors
to the design of medical equipment. J. of Clin.
Monitoring and Computing.14(4) 253-263.1998. van
Ginnekan, AM. The computerized patient record
balancing effort and benefit. Int. J. of Med.
Informatics. 65 (2002) 97-119. Shortliffe, EH
(2006) Biomedical Informatics Computer
Applications in Health Care and Biomedicine 3rd
Edition. New York. Springer