Training in Pathology Informatics: Lecture, Project, Rotation, Fellowship and beyond

1
Training in Pathology Informatics Lecture,
Project, Rotation, Fellowship and beyond

• Raymond D. Aller, M.D.
• Director of Informatics
• USC Pathology
• 20 September 2010

2
Acknowledgments

• Dozens have contributed I can mention only a
  few here
• Ramzi Cotran, Stan Robboy, Roger Cote, Brad
  Copeland
• Frank Elevitch,Bill Hartmann,Tom Lincoln
• Mark Tuthill, Bob Miller, John Gilbertson, Paul
  Catrou, Anil Parwani, Ul Balis
• Wes Naritoku, John Vallone,
• Alexis Carter, Stephen Hewitt, John Sinard, Brian
  Jackson

3
Evolution of training in pathology informatics

• Explorers and pioneers we have been training for
  a long time
• Training in general medical informatics programs
• Early adopters - OJT
• ABP informatics as a subspecialty -
• Spotty coverage in residency
• Sporadic fellowships

4
More recently, it gets better ...

• Ongoing fellowships
• General medicine rediscovers board certification
• Advertising for these skills

5
The explorers

• Bill Dito, Art Rapapport, George Brecher, Homer
  Warner, Phil Hicks, many others
• Self taught
• From basic data management to deeper analysis

6
The pioneers

• Mike McNeely, Don Connelly, Delane Wycoff, Art
  Krieg,
• Again, focused on facilitating lab workflow
• Some graduate degrees, MD theses, NLM
  fellowships
• Both explorers and pioneers passed knowledge to
  colleagues via national meeting seminars

7
Informatics training past

• 1970's informal rotations, occasional masters
  and MD theses Missouri, Utah, Minnesota,
  Harvard, others
• 1980's under pathology leadership, NLM sponsored
  training fellowships at several universities
  plus others
• NLM leader Don Lindberg
• Ohio State Jack Smith, Oregon Bob Beck,
  Minnesota Don Connelly, Utah Stan Huff,
  Indiana Clem McDonald

8
National pathology meetings and resources

• 1970's on Informatics seminars ASCP, CAP,
  IAP (now USCAP)
• 1983 AIMCL, University of Michigan
• 1984 Bill Dito's Informatics in Pathology
  journal
• 1987 monthly pathology informatics coverage in
  CAP Today (continues today)
• Late 1980's over half of the seminars presented
  at the biennial CAP meetings are on informatics.

9
User Group meetings

• Another important source of lab informatics
  training
• How to get more value out of a system you already
  own
• How to get past seeming roadblocks to improve
  patient care
• Many of the most relevant and practical seminars
  are in this context
• E.g. Sunquest, Kontron, Medlab, Mediware,
  Meditech, Cerner, Soft

10
Written Resources

• For many years, there was only a single textbook
  in this field (Elevitch, ABCs of LIS)
• Chapters in Henry
• Some journal articles
• CAP Today coverage
• In past five years, a number of useful books have
  appeared

11
Just in time knowledge

• Calling your friends
• Local colleagues
• Google, Wikipedia, etc
• API listserv

12
American Board of Pathology proposed subspecialty

• 1992 Letter of intent to American Board of
  Medical Specialties
• Intended to be open to diplomates of any of the
  ABMS boards
• Appointed 5-member informatics test committee
• Test committee met for a number of years
• Unfortunately, could not devise a sufficient
  number of questions testing informatics (as
  opposed to bits and bytes)

13
Formal informatics training in pathology
departments.

• 1993 the residency informatics rotation
  described
• An increasing number of departments offer
  rotations
• 1980's-1990's A few unique fellowship trainings
  Johns Hopkins, Michigan
• CAP Foundation training grants for residents to
  attend AIMCL, APIII

14
Into our own the 2000's

• A number of formal fellowships established,
  fellows trained
• The Association of Pathology Informatics
• Description of curriculum for a fellowship
• At end of decade a few fellowships fully active,
  but others lack sufficient candidates

15
How is informatics information available to
pathologists today?

• National meetings CAP, USCAP, AABB, DR/EWC,
  AACC, ASM, CLMA, PI2010
• National LIS user's groups
• CAP Today
• Occasional articles in other path journals
• Articles in general informatics journals
• Most recently online J Path Informatics
• Residency training in informatics required
• Fellowships

16
Residency rotations lacking

• Many departments/programs lack interested faculty
• In some departments, informatics does not get
  academic respect
• Competition for time in the general residency
  training
• We know how to teach surgical pathology we are
  still learning how to teach informatics

17
Compensating for inadequate training

• Residents fail to recognize the gap in
  informatics training until too late.
• Some apply for CAPF travel grants to attend
  LabInfoTech, APIII, Pathology Visions, or now
  PI2010.
• Others don't do even that much.
• Very few seem to take advantage of rotations
  available in other programs, with strong
  informatics faculty

18
Leaving value on the table ...

• One program with a well established informatics
  curriculum and faculty offered a rotation for
  residents from other programs - yet over a number
  of years, no one applied or came.
• While residents seem to understand the need for
  slide-reading skill, they seem clueless about
  managing the actual product of their department.

19
Pathologists without basic understanding

• Recent training is no defense
• Many residencies lack a meaningful rotation
• National pathology meetings offer fewer
  informatics seminars than they did 15 years ago.
• National meeting seminars they gave a party, no
  one came...

20
Most fellowships are under-subscribed

• One university offered a fellowship, it was never
  filled
• Another only had four fellows in 25 years
• Others are less than half full.

21
Program began trained so far concurrent duration advance admiss faculty

Pitt 2000 15 1 1 16 26
MGH/Part 2007 2 5 1 to 2 16 14
Henry Ford 2006 2 1 2 4 1
Johns Hopkins 1990 4 1 1 to 2 4 1
Michigan 95/08 3 1 2 4 55
22
Length of the fellowship

• Some programs structured as one year, others as
  two years
• One year
• less time away from other aspects of pathology
  practice
• May be easier to schedule
• Two years
• More time for fellow to learn the environment,
  become a productive member of the team
• Informatics projects are at least two years

23
Thoughts on structure 2 years

• Gives a better chance for the fellow to become a
  productive (net-positive-FTE) member of the team
• Lead time on position recruitment may be 18
  months
• Strongly consider including a component of
  case-based responsibilities

24
Case-based responsibilities

• Such as a rotation reading biopsies - within the
  informatics fellowship.
• to prepare our fellows for practice settings, and
  
• to maintain continuity with case-based pathology

25
Thoughts on background 1 year

• For those who already have leadership and
  management skills (may have been in practice for
  some years)
• Desire greater depth in informatics than afforded
  by their (non-)rotation in residency.
• How do we ensure that they come up to speed more
  rapidly, and take on adequate front-line
  responsibility?

26
More concerns about 1-year fellowships

• Not long enough to learn the systems and people,
  to become productive, and to have an impact.
• Not enough time to develop skills yes it gets
  them out and they can start earning, but they are
  very green mainly this relates to a mature
  perspective and business management and
  negotiation skills
• Both MGH and Henry Ford noted 2 years is just
  barely enough time to develop skills

27
How far in advance are fellows appointed?

• The more successful and highly subscribed
  programs designate the next fellows about 15
  months in advance.
• In order to compete for candidates 18-24 months
  in advance, funding must be secure a long time
  ahead
• Otherwise, the best candidates have already
  committed to another program before the
  less-funded program can even offer a slot.

28
Those programs recruiting later ..

• ... recruit only a few months before the program
  would begin.
• However, the best candidates are making a
  commitment for a period 18 months through 42
  months from now.

29
Facilitator/administrative support

• Half time, or so
• The two most successful/highly subscribed
  fellowships each have such a support person
• It appears that the others (less successful in
  recruiting) do not have such support.

30
How many faculty?

• .. at least 40 time doing informatics (service,
  research, teaching, etc)
• One, one, two, two six, 5, 14

31
Faculty roles

• Program leader/director/cheerleader
• Plus additional staff with special interests
  imaging ,
• Adjunct contributors for management, regulatory,
• Co-directors in CP, AP, molecular, imaging

32
Number of fellows -

• Many programs - zero much of the time
• Some programs, one
• Another, five

33
Funding

• To assure continuation of program
• To increase the number of fellows
• To provide additional support for faculty (e.g.,
  NLM training grant at Pitt)

34
Funding sources

• -- Intramural/practice funds
• Laboratory revenues
• NIH training grants of various types
• One program mentioned growing beyond the current
  5 - "could easily recruit and train 6 or more
  fellows next year"

35
Consistent funding

• Essential to carry forward with such a fellowship
  (can't recruit otherwise)
• Some fellowships have been hamstrung by
  inconsistent funding
• However, it is often challenging to justify
  existing front-line operational employees in the
  informatics group - the linkage between funding
  and productivity of an informatics fellow may be
  more difficult

36
General medical informatics programs

• Interaction with NLM funded general medical
  informatics fellowship?
• MGH - available, but not best option
• Some others not available

37
Making the fellowship sustainable not just
funding

• How to reach the point where faculty time
  consumed is more than made up by useful work
  product of the fellows.
• How do we get fellows up to a point where they
  can begin to do things independently (need to
  acquire a deep understanding of existing systems)
  
• I just don't have any "excess" time to spend
  with a fellow.

38
Impact on faculty time

• our current fellows are experienced, all three
  have prior experience with operations and
  research. They are a net positive
• it's nearly a wash year one very valuable year
  two

39
Avoiding faculty overwhelm

• Recruit candidates who already have
  management/leadership skills
• If you offer a 1-year fellowship, bring in people
  who can already contribute

40
Effect on faculty workload

Faculty time
Weeks --gt
41
Subjects covered in fellowships

• Management and workflow optimization
• Administrative-type activities with the
  informatics group
• However, avoid "just follow me around to
  meetings"
• Image management, analysis, image perception
• Molecular epidemiology
• Health services research

42
Programming?

• Understanding the use of tools is essential, but
  it doesn't teach pathology informatics
• for a person who already knows programming, to
  simply do another programming project is not
  productive.

43
Operational responsibilities

• "depends dramatically on their career goals and
  their experience" vs.
• Essential for a training program
• Developing new tools and approaches
• Implementing systems
• Innovative pathology reporting (online,
  graphical, interactive)
• Web-based delivery of diagnostics information

44
Fellowship projects

• Large clinical operations projects that involve
  the health system and the lab
• many! as many as possible big, small part of
  team
• goal is to get them to lead a large to medium
  size department project of their own with budget,
  staff, planning etc.
• The topics are matched to the fellows interest
  and skills

45
Ideas for projects

• may originate from their own idea
• more typically related to labs needs and
  established plans.
• As projects require capital and resources these
  just don't pop up and get done.

46
Informaticists should use what they develop

• In one case, a clinical pathologist C developed
  a cytology system for his colleagues they
  seemed to think it was fine
• Several months later, one of the partners became
  ill, and C was asked to take a rotation on
  signing out Paps
• C soon realized that the system's workflow was
  awful
• He fixed it, and his colleagues agreed it was
  much improved.

47
Implications informatics practice

• Some program directors commented on the jobs
  their fellows had taken
• Spending 50 time in informatics, and 50 time in
  a more traditional subspecialty, has several
  advantages
• Not only are systems better designed, but you
  have more credibility if you share in the
  workload
• More opportunity for interaction with clinician
  colleagues if part of your time is in a
  case-based practice.

48
Research time

• . Academic 35-70 of time is protected
  research time, depending on goals. Related
  opportunity to earn a graduate degree (MS, Ph.D.)
• Community -they have "research projects" and
  studies they get involved in. Our shop is very
  operational. I consider what they do more proving
  concepts or testing solutions versus hypothesis
  driven research
• For the pathologist interested in the community
  practice of informatics, time spent on research
  means less time learning the practicalities of
  informatics

49
Service work

• Myriad
• system maintenance,
• issues resolution,
• Training of staff and residents
• No on call per se
• Image analysis
• Diagnostic pathology (e.g., surgical pathology,
  cytopathology)
• Clinical pathology (e.g., transfusion medicine)

50
A different type of disease

• The surgical pathologist diagnoses a series of
  patient diseases
• the informaticist
• diagnoses and treats the diseases of systems
• Finds ways to make work more efficient for
  his/her colleagues

51
How do we teach

• Responsibility
• How do we structure our organization so that the
  fellow has clear operational responsibility?
• Leadership informatics has much in common with
  management both are difficult to teach to
  residents, fellows
• Delegate the various tasks

52
After fellowship ....

• What are the next career steps for a pathologist
  after fellowship?
• Several academic medical centers
• Another director of lab informatics at a large
  hospital in India
• Some full time informatics,
• more are half time informatics, half-time
  surgical pathology/etc.
• At least one has founded own company

53
Creation of new fellowships?

• Talked with at least four organizations that are
  considering starting a fellowship academic,
  government, reference lab,
• And with one well-established department that
  doesn't think it's feasible
• Why create a fellowship?
• General clinical informatics is thinking about
  board certification need to be at the table
• Recognize the need for informatics leaders in
  many organizations

54
Barriers to creating a fellowship

• Faculty time
• Funding
• Facilitator (support person)
• Finding qualified people
• Very very small candidate pool
• only one or two qualified candidates per year"
• Fifth wheel (Finding a role)
• Free space
• F ......

55
Lack of qualified candidates

• Not enough exposure in medical school, or
  residency
• Lack of training
• Lack of mentors/examples
• Most residents have no concept of informatics
  practice
• Inconsistent funding for positions year to
  year, soft money, or dependent on vagaries
• Informatics fellowships are not ACGME-accredited
• Uncertainty about job prospects marketability
• Some applicants seem to want to hide in a
  backwater

56
ACGME accreditation? But there isn't an approved
sub-specialty...

• Two types of programs accredited
• Regular subspecialty transfusion medicine,
  chemical pathology, microbiology, etc...
• Selective pathology fellowships - over 60 of them
• Many in special areas of surgical pathology
• A few in clinical pathology
• This is the area where we would apply for
  accreditation of an informatics fellowship

57
ACGME accreditation for fellowships

• Advantages
• Medicare funds ½ FTE
• Time in fellowship counts towards AP/CP Board
  time
• Time counts toward maintenance of certification
• How do we sign up?
• Pathology Residency Review Committee program
  completes and submits the Program information
  form
• RRC evaluates the program, evaluates the quality
  of the teaching.

58
Accreditation a mixed blessing?

• One large program "don't think this would be a
  good thing"
• A smaller program don't really want the
  paperwork hassle it is enough to comply with our
  own non-accredited fellowship requirements in the
  institution.

59
Accreditation 2

• - If board recognition became a reality for
  fellows, that might be a reasonable trade.
• - I just saw the 75 pages of our cytopath
  fellow had to submit to the ACGME. Major waste of
  time that had no impact on what fellows do now
  versus what they did ten years ago

60
Job prospects?

• There are ads for pathologist informaticists
  but not oodles of them
• In many cases over the decades, informatics jobs
  have been created (perhaps from something else)
  when a candidate appeared
• In other cases, a pathologist in a different role
  has evolved into the informatics role
• Even advertised positions sometime specify a
  combination of skills that fit only one person in
  the country

61
Some not optimistic

• Although I believe there should be more jobs
  available, especially as some of us older guys
  retire, and the perceived need, at least by API
  members, seems to be growing, I'm not sure those
  jobs will be supported. (Locally, when I retire,
  I really doubt if my position will survive).

62
Not optimistic 2

• Hospital IS wants to absorb the LIS which I have
  fought to keep in Pathology for the last 20
  years.
• I think Pathology may have lost the ability to
  win the battle of ownership of pathology
  informatics. Hopefully, I'm wrong.
• Currently, I sense a resurgence of energy to get
  pathology informatics back under pathology.

63
Informatics plus another subspecialty

• Maybe one idea should be to emphasize pathology
  informatics not only as a primary field, but also
  as a secondary field of emphasis when recruiting
  residents (eg, surg path/informatics,
  hemepath/informatics, molecular/informatics,
  etc).

64
Trailblazers, pioneers, settlers

• "Today's pathologists who work in imaging and
  data are like trailblazers, and the fellows will
  be the earliest settlers, but it will remain a
  rough world out there for a while, before things
  develop. I think the trailblazers enjoy the
  "risk", but I am not certain fellows appreciate
  that there are not jobs with pathology
  informatics as a job description/title out there
  yet."
• In 50 years, we will be suburbanites

65
What are candidates interested in?

• Recently, have seen increased interest from
  people that want to be directors of core labs.

66
General medical informatics programs

• In some organizations, pathology informatics
  faculty spend their time with a general medical
  informatics group
• Examples Utah, Oregon, Johns Hopkins, Ohio State

67
Advice to a department considering offering a
program

• O Ramp up your faculty and your department
  pathology informatics initiatives first, the
  fellowship will follow naturally.
• R plan money, space, time and faculty be
  aware that qualified candidates may not surface

68
Adding informatics into other fellowships

• There may be opportunities to modify a
  traditional (e.g., surgical pathology) fellowship
  year to include a high component of informatics.
• During the elective months
• And as ongoing responsibility during the
  case-based months

69
References

• There are several useful articles on this topic
  in the literature.
• Rather than trying to jam them on several slides,
  I ask that you eMail me to request a list

70
Thank you!

• I appreciate your attention
• raller _at_ usc.edu
• I welcome your questions