Survival Guide for a Pathologist Informatician Working in a Large Healthcare System

1
Survival Guide for a Pathologist Informatician
Working in a Large Healthcare System

• Bruce A. Friedman, M.D.
• Active Emeritus Professor of Pathology
• University of Michigan Medical School
• Ann Arbor, Michigan
• Email bfriedma_at_umich.edu
• Blog www.labsoftnews.com
• Lab infoTech Summit www.labinfotech.com

2
Major Topics to be Presented in Lecture

• Define politics, power, organizational
  hierarchies, status of informatics in
  healthcare deliver systems pathology depts.
• Discuss history of LISs their relationship to
  pathology departments and to lab personnel
  supporting these systems
• Explain how the emergence of EMRs in large
  healthcare systems is affecting pathology
  informatics as career path
• Show how the early health model and full service
  diagnostics radically alter ground rules for
  healthcare pathology/labs
• Given this historical context for healthcare IT,
  discuss optimal career strategy for pathology
  informaticians for next decade

3
Nature of Politics in a Large Hospital Systems
Relevance for Pathology/Labs

• Politics tightly linked to power exercised by
  members of an organization power determines
  who/what allocates resources
• Political power in hospital system accrue to MDs
  who generate revenue resources i.e., perform
  procedures with high margins
• Pathology/lab medicine not a locus of power in
  large healthcare delivery networks because their
  primary role is to serve clinicians
• Labs viewed as a cost-center rather than
  revenue-center, even with a successful lab
  outreach program contributing to bottom line
• Increasing commoditization of lab testing
  reference labs drive down the cost of testing by
  competing with each other on price

4
Problem-Solving as an Additional Source of
Political Power in Healthcare Systems

• Textbooks assert that political power flows from
  the ability to address solve the strategic
  problems facing an organization
• Such a rule of thumb is probably more relevant
  in for-profit organizations where premium placed
  on innovation profits
• Because of complex payer system in U.S.,
  attention of healthcare executives focused on
  reimbursement financials
• In hospitals, executive management suite
  MD-proof little real input from physicians other
  than standards of clinical practice
• Innovations by MDs applauded by administrators
  when result in high-margin popular procedures
  (e.g., medical imaging)

5
Power Hierarchy in Pathology/Lab Medicine
Pathologists Emphasize Surgical Pathology

• On power scale in hospital, clinical departments
  rank higher than pathology/ radiology power
  hierarchy also exists within pathology
• In community hospitals, pathologists have largely
  exited from CP, leaving in hands of Ph.D.s and
  medical technologists to manage
• This migration prompted by reimbursement issues
  (CP largely commoditized) and the natural
  inclinations of pathologists
• Trend less evident in academic pathology
  departments with greater demand for esoteric
  testing popularity of lab outreach
• Net result of this set of hierarchies is that AP
  has more political power than CP informatics
  viewed as a lesser niche within CP

6
History of IT Politics in Healthcare Systems
Increasing Influence of the CIO in LIS Selection

• Several decades ago, most computer applications
  in hospitals focused on patient management (PM)
  and patient billing (PA)
• Mainframe managers reported to healthcare
  system CFOs lab/radiology allowed to run their
  own systems autonomously
• In the early 90s, CIOs began to appear in
  hospitals clinical applications (i.e., EMRs)
  became much more highly valued
• Shorter leash placed on the previously autonomous
  LISs RISs, often by rolling up their budgets
  with central IT group
• C-level execs took greater interest in selection
  of LISs often selected clinical software suites
  from favorite EMR vendors

7
Attitude of Pathology Departments Toward
Informatics A Long History of Ambivalence

• For about four decades, LIS vendors have provided
  us with turnkey systems ongoing support
  provided with contracts
• Individual systems were customized based on
  departmental needs this required assistance of
  vendor personnel per contract
• Net result was informatics units staffed by med
  techs who understood their system in-depth but
  little broad IT knowledge
• Selected hospitals and academic centers broke out
  of mold, developing LISs from scratch or
  extensively modified stock LISs
• As a result, LISs in pathology departments often
  regarded as mere tools, albeit essential,
  expensive, and cantankerous ones

8
History of LISs and LIS Vendors Care and Feeding
of the Vendor Turnkey Systems

• For past 25 years, only one LIS decision was
  required of. leadership select company/product
  then desired modules
• LIS purchase decisions generally made by
  department on the basis of optimizing operations
  of high-volume labs like chemistry
• Specialized AP systems have always existed
  purchased to supplement the workhorse LIS that
  served the clinical labs
• In most departments, in-house LIS support limited
  provided mainly by med techs trained to support
  that particular system
• LIS functionality enhancements limited to
  vendors development pipeline of apps this
  controlled by marketing considerations

9
Defining Best-of-Breed for Ancillary Systems
Such as LISs and RISs

• Best-of-breed defined as superior functionality
  of healthcare software products ( e.g., LIS)
  compared to others in their class
• Sound logic behind principle theoretically, a
  specialized vendor can and should develop a
  superior overall product
• Not impossible for large vendors with broad suite
  of products to achieve excellence across entire
  product line (requires )
• This superior functionality relatively easy to
  measure by knowledgeable potential puchaser by
  scrutiny of the product
• Vendors of healthcare software suites may provide
  some lesser quality modules emphasize
  integration of entire suite

10
Are Pathology/Lab Medicine Captives of the
Vendors of Their Turn-Key Systems?

• Most of the features of LISs has been deployed by
  the vendors based on suggestions from their lab
  customers
• All such software goes live in hospital lab
  environment must meet most of the labs
  business technical requirements
• Labs no more captives than for-profit companies
  are controlled by companies such as Microoft
  or Oracle
• Relevant question in this context is whether lab
  personnel can/should develop more of their LIS
  software themselves
• This is an Impractical goal for most lab
  personnel given their training, resources, and
  unforgiving regulatory environment

11
Suggestions for Enhanced Lab Software Development
Model in the Future

• Pathology informaticians growing increasingly
  sophisticated in their ability to design/develop
  software with higher functionality
• Need for different software development model
  between vendors and academic pathology
  departments than in past
• Such departments could function as both
  developmental testing labs for new software
  modules marketed by vendors
• This new relationship would include development
  grants revenue-sharing modules must integrate
  with other products
• Different set of incentives than before
  departments worked with vendors for access to an
  LIS customized to their needs

12
Bioinformatics Has Gained Prestige as Discipline
in Research Environment

• Gains in genomics proteomics have fueled
  intense interest in bioinformatics critical
  discipline in research environment
• Many academic pathology departments have elevated
  informatics to division level parallel with AP
  and CP
• This momentum has enhanced the prestige of
  applied informatics which is often blended with
  research informatics
• Quality of pathology informatics improved by
  influx of MDs with solid backgrounds in
  engineering and computer science
• Similar changes also occurring in radiology,
  surgery, and internal medicine many more skilled
  clinical informaticians

13
Interest in EMRs Fueled by Quest for Greater
Efficiency and Error Reduction in Hospitals

• During last decade, much greater interest in
  deployment of hospital EMRs partly driven by
  efficiency/safety concerns
• Hospital executives hiring informaticians to
  guide their deployment of large scale,
  system-wide, EMR deployments
• All of major healthcare software vendors (Cerner,
  GE, McKesson, Siemens) are competing for this
  lucrative market
• RIS/PACS market also very hot new interest also
  in clinical image management systems (CIMSs)
  serving total enterprise
• LIS market viewed by major vendors as mature and
  less lucrative also being roiled by the
  emergence of middleware

14
The EMR Conundrum Packing More Complex Data into
a Smaller Receptacle

• Todays major EMRs inadequately architected to
  integrate the increasingly complex information
  generated in labs radiology
• EMR vendors placing limits on data passed to them
  from LISs/RISs to simplify data integration
  reporting complexity
• CIOs placing limits on scale complexity of EMR
  projects in order to increase the chance of their
  successful implementation
• Problem can be viewed as a shrinking EMR
  receptacle while data from pathology rapidly
  expanding in volume/complexity
• Vendors will probably not re-architect their EMRs
  because of sunk development costs CIOs
  continuing to buy their products

15
New Opportunities for Pathology Informaticians
within Pathology Depts.

• Not unusual in past for pathology informaticians
  to be selected by CIOs to work on clinical
  projects within hospital central IT groups
• Promoted on basis of visibility within hospitals
  also lack of other MD talent speaks to
  productive long history of IT in pathology
• This career path will be now be less common
  because IT-trained internists now available to
  guide EMR deployments
• EMR installations also less challenging more of
  a training exercise to convince clinicians/nurses
  to lower expectations
• Because molecular dx reporting more complex, new
  challenges in communicating/explaining such
  information to test-ordering MDs

16
Early Health Model Full Service Diagnostics
as New Healthcare Models

• Early health model defined as the detection of
  diseases in their pre-symptomatic pre-clinical
  stage concept promoted by GE
• Molecular medicine/full service diagnostics,
  promoted by Siemens, offer benefits to patients
  by enabling the following
• Early detection and treatment of disease,
  resulting in decreased morbidity and mortality
  and possibly lower costs
• Monitoring treatment efficacy by the selection of
  optimal individual therapy for each patient
  (personalized medicine)
• More effective screening programs for disease and
  determination of genetic predisposition to
  disease
• Determining prognosis of disease, particularly
  malignant neoplasms with varying scrambled
  cellular genomes

17
Why Predictions about Early Health Model Makes
Various Healthcare Players Nervous

• Third party payers like insurance companies
  U.S. government have hands full paying for
  current clinically manifest diseases
• Physicians not trained to diagnose pre-clinical,
  pre-symptomatic disease not used to managing
  patients with no complaints
• Big pharma nervous because their products on
  the market validated by controlled clinical
  trials with symptomatic patients
• Health system CEOs nervous because they control
  expensive real estate designed to care for
  sickest operated of patients
• In contrast, EHM is very appealing to mature
  healthcare consumers want to know what will
  happen to them in future

18
Who Will Derive Benefit from Emergence of Early
Health Model and Molecular Medicine?

• Healthcare consumers benefit understand
  intuitively that earlier detection of disease
  should yield better health outcomes for them
• GE and Siemens have made 23B investment in IVD
  and molecular diagnostics in the pursuit of their
  strategic vision
• GE Medical paid 9.0B for Amersham (bioscience
  medical dx) recently failed to acquire Abbotts
  IVD/POCT business
• Siemens paid 5.3B for Bayer Diagnostics, 1.9B
  for Diagnostic Products, and 7B for Dade
  (pending)
• Clearly, both GE Medical and Siemens are now
  aggressively pursuing integrated diagnostics
  portfolio through acquisitions
• If movement toward a diagnostic-centric model
  successful, biomarker measurement informatics
  will be major pivot points

19
Is the Early Health Model a Sure-Fire Means to
Reduce the Cost of Healthcare?

• Claim being made that early health model may
  reduce healthcare costs because diseases will be
  detected earlier
• Proponents assert that early care is less
  expensive than late care, which often requires
  inpatient stays chronic care
• I am suspicious of this claim, at least at this
  early point, because the dx and rx costs of this
  new model unknown
• More sophisticated imaging and proteomic testing
  cost of personalized therapy may offset
  potential savings of early dx
• Will reserve judgment but irrelevant because EHM
  will emerge or not based on more important
  factors such as efficacy

20
Relevance of EHM to Merger/Convergence of
Radiology, Pathology, and Lab Medicine

• Both EHM and molecular dx stress role of
  labs/path medical imaging only available
  processes to arrive at pre-clinical dxs
• Radiology moving technically/intellectually
  closer to lab medicine research in molecular
  imaging using biomarkers
• Idea being floated that labs/pathology medical
  imaging should merge into a Department of
  Diagnostic Medicine
• If idea catches on, need closer coordination of
  health screening for to promote dx accuracy
  waste reduction
• Merger of two depts. would yield greater
  political power in health systems would deliver
  90 of critical dx-px data

21
Possible Workflow and Handoffs Between Labs/Path
and Radiology Under the EHM

• Impractical to view medical imaging as basis for
  routine health screening with mammography as
  possible exception
• In terms of convenience, capital costs, and
  safety, only reasonable approach to screening is
  use of serum biomarkers
• Multi-organ biomarker screening will consist of
  periodic general mega-panels with follow-up with
  organ-disease panels
• Analysis of panels will required computerized
  algorithms FDA proposing to intervene with
  regulatory control of IVDMIAs
• Suspicious results from large/focused biomarker
  panels will prompt targeted medical imaging or
  molecular imaging

22
Future of Medicine and, Therefore, of Pathology
and Lab Medicine The Three Ps

• Elias Zerhouni, NIH Director, has describe the
  future of healthcare delivery as being based on
  three P's
• Predictive
• Personalized
• Preemptive
• Predictive relates to determining accurately the
  future outcome of a disease similar to prognosis
  for tumors
• Personalized relates to treatment, meaning
  turning to individualized treatment,
  particularly for malignancies
• Preemptive means deterring/preventing a disease
  from developing or ameliorating the effects of
  the disease

23
The In-Vitro Diagnostic Continuum in Labs/Path
Criticality for Healthcare Delivery

• In parallel with the three Ps, can also refer to
  the in-vitro medical continuum occurring in
  pathology labs
• Screening
• Early detection
• Prognosis assessment
• Individual therapy
• Treatment efficacy
• In-vivo/in-vitro models intersect/reinforce each
  other pathology and clinical labs play critical
  role in diagnostic-centric model
• Given volume complexity of data managed in
  clinical labs, unthinkable that pathology
  informatics will not play critical role

24
Predictions for Future of Medical Informatics
Based on Preceding Discussion

• Dominated by clinician/informaticians working
  with hospital central IT groups to deploy
  integrated EMRs
• C-level hospital executives will increasingly
  intervene in the selection and adoption of
  integrated software suites
• Hospital computing/EMRs driven by financial
  issues limitations on project scope to enhance
  system success
• Such projects intriguing from an organizational
  political perspective but generally not
  technically challenging
• Purely academic informaticians continue to work
  on projects not tethered to practical healthcare
  demands

25
Predictions for Future of Pathology Informatics
Based on Preceding Discussion

• Proteomics/genomics exploding, catapulting labs
  pathology informatics to forefront of healthcare
  delivery
• Surgical pathology, using tissue marker
  quantification, will expand beyond its
  morphology-based heritage
• Digital pathology will radically convert
  pathology and improve quality of diagnoses
  therapeutic interventions
• Informatics will serve as a critical bridge to
  enhance translational research link the bench
  to the bedside
• LIS vendors pathology personnel will develop
  mutually remunerative satisfying developmental
  relationships

26
Take Home Summary Suggested Steps for Survival
as a Pathologist Informatician

• Help elevate organizational status of informatics
  to level of a division for political visibility
  to ensure adequate funding
• Blend oversight over research applied
  informatics for better communication more
  stimulating environment
• Maintain cordial relationship with CIO/central IT
  group avoid heavy organizational obligations
  with EMR projects
• Pursue applied lab software projects with
  vendors, modules must be compatible with their
  architecture
• Demonstrate the value of informatics to elevate
  it from the tool status to critical component
  of departmental success