Student Assessment: New Ideas and Old Basics Louis Pangaro, MD Professor and Vice-chair for Educational Programs Department of Internal Medicine Uniformed.

Student Assessment:New Ideas and Old Basics

Louis Pangaro, MD

Professor and Vice-chair

for Educational Programs

Department of Internal Medicine

Uniformed Service University of the Health Sciences

26 September 2007

2

“clinical assessment”

• By teachers (house staff and faculty)

• On clinical rotations

• Based on words = descriptors

• Using words = descriptive

• In vivo (vs. in vitro end of course/year)

3

What’s old?

• Suspicion of grades by teachers

• Teachers reluctance to be direct (honest?)

• Belief that grading by teachers is subjective

• Lawsuits about low grades• Jamieson, Guidebook for Clerkship Directors, 2005.

4

What’s basic?

• We have an obligation

• Fairness

• Mentoring – The Hippocratic Oath

• Professionalism– Duty and expertise (Pellegrino)

5

How it looked in 1987

• lack of meaningful comments by evaluators

• insufficient definition of evaluation criteria

• too much inter-observer variability

• late submission of evaluations

• delay in feedback to students

Tonesk X, Buchanan RG. J Med Ed. 1987

6

How it looked in 2001

• “Areas of weakness in current [clinical] evaluation models include psychometric properties associated with the tools, namely their questionable reliability and validity.”

• Turnbull, International Handbook of Research in Medical Education, 2002.

7

How it looks in 2007

• “Constructive criticism is hard to come by..”

• “…Candor is at least as painful to the provider as to the recipient…”

• “…faculty members feel uncomfortable inflicting pain even in a good cause like student improvement.”

• HMS Student Handbook, 2006-2007

8

focus

• The emotional issues for teachers and learners in the grading process.

• Dealing with barriers to candor.

9

• Quality: minimizing unwarranted variation in physician performance.– Wennberg

• Faculty Development: minimizing unwarranted variation in teacher performance.

10

What’s new

• A believe that quality methods can apply to teaching.

• Search for more rigor.

• “Best Evididence Med Ed” (BEBM)

• “Med Ed Research Certificate”

• MPH, MHPE programs

11

Today: describing success in clinical evaluations

12

New focus on evaluation in clincial setting:

• Attention to professional traits (Papadakis)

• Lawsuits by patients

13

Context of new methods

• Objective Structured Clinical Exams

• 3600 assessments

• Portfolios

• Descriptive vocabularies– To get more inter-rater agreement– To calibrate differences between levels of

performance

14

FOCUSED PHYSICAL EXAMINATION

• [Norm] • Exam is generally appropriate in scope

and technique. Identifies major abnormalities and pertinent normal findings, only occasionally missing elements. Exam linked to history. Appropriate for level of training.

• HMS Clerkship Grading Form

15

FOCUSED PHYSICAL EXAMINATION (2)

• [Poor]

• Consistently misses important findings and often does not make appropriate connection between history and physical. Often uses faulty or inappropriate technique. Not organized or thorough.

16

FOCUSED PHYSICAL EXAMINATION (3)

• [Excellent]

• Exam is consistently superior. Uncovers subtle and important findings, incorporating advanced techniques where appropriate. Exceptionally organized and thorough, even on difficult cases.

17

American Board of Internal Med

1 2 3 4 5 6 7 8 9

Minimal acceptable

Average and 2 SDs

18

ACGME Outcomes Project

• New emphasis on outcomes (vs

curriculum)

• “long-term effort designed to

emphasize educational outcome

assessment in residency programs

and in the accreditation process.”

19

ACGME: Outcomes: a cultural shift

• Designing Curriculum is no longer enough

• Results must be demonstrated.

• A shift from process to product.

• Content expertise not enough; need pedagogic expertise

20

ACGME “Competencies”(1999)

• Medical Knowledge

• Interpersonal & communication skills

• Professionalism

• Patient Care

• System-based Practice

• Practice-based learning & Improvement

21

the ACGME has spoken !

22

CanMEDS (2005)

• Medical Expert

• Communicator

• Collaborator

• Professional

• Manager

• Health Advocate

• Scholar

23

“R.I.M.E. Scheme”

– Reporter

– Interpreter

– Manager/Educator

Pangaro, Academic Med, 1999

24

Acceptance (1) Ob-Gyn

“The RIME method is a valid, logistically feasible and acceptable way of assessing medical student clinical performance…

…..minimizes disadvantages of descriptive evaluation, and maximizes the opportunity for accurate observations and helpful feedback.”

APGO UME Taskforce:Espey et al, Am J Ob Gyn, 2007

25

Fairness

Consistency

Expectations

26

Principle 1 : Fairness to society :

valid (not arbitrary) and sensitive to detect marginal performers

to students : know what’s expected, timely feedback

to teachers : know what observations to makeprotected (legally, emotionally)

27

Principle 2 : Consistency

Reliability- Within teacher

- Between teachers- In same rotation or across blocks

- Between sites in same clerkship

- Between disciplines

- suitable for high-stakes decisions

28

strategy

• Simplicity leads to acceptance and use

• Acceptance to consistency• Consistency to fairness

29

Fairness & Consistency

Reliability &Validity =

Stability of measurement and strength of inferences from

observations

30

“ Old” beliefs in our culture

• Grading by teachers is subjective, Examinations are objective

• Measurement > description, Numbers are > words

31

Framing the question more simply:

• What do we expect of students?

• Can we get all teachers to have the same expectations, and apply them consistently?

• [A question of words and of conceptual frameworks]

32

Educational Goals

Curriculum

Evaluation

Feedback / Grading

Depends on expectations

community

33

Teacher/program

Learner

Content – Goals (Patients)

The goal: progressive independence of the learner

after SFDP

34

Frames of reference for expressing goals

1. Analytic

2. Developmental

3. Synthetic

35

goals for education (generic):

• attitudes/behavior• skills• knowledge

“KSA”

36

Bloom’s Taxonomies

Cognitive Domain

Psychomotor Domain

Affective Domain

37

1. Analytic expression of Goals

– “ana-lytic”: takes the learner “apart”

– into domains, categories

– “attitude”, “skills”, “knowledge”

– considered separately

– generic terms

– useful for discrete assessments

38

Curricular Goals: KNOWLEDGE The School of Medicine will ensure that before graduation a student will have

demonstrated, to the satisfaction of the faculty, the following: • The capacity to recognize the limitations in one’s knowledge and clinical skills and

to make a commitment to engage in lifelong learning • Knowledge of the normal structure and function of each of the major organ

systems of the body and the current basic scientific mechanisms operative at the systemic, cellular, and molecular levels

• Knowledge of the various causes (genetic, developmental, metabolic, toxic, microbiologic, immune, psychosocial, neoplastic, traumatic, and degenerative) of illnesses and diseases

• Knowledge of the altered structure and function of the body and its major organ systems that are seen in various illnesses and diseases

• Knowledge of the scientific method in establishing the causation of disease and efficacy of traditional and non-traditional therapies

• Knowledge of health care policy and the economic, psychological, social, and cultural factors that affect health and health care delivery

• Knowledge of the most frequent clinical, laboratory, radiographic, and pathologic manifestations of common as well as life threatening diseases

• Knowledge about relieving pain and ameliorating the suffering of patients • Knowledge of the epidemiology of diseases and the systematic approaches useful

in promoting health • Knowledge of techniques of patient education and counseling in basic lifestyle

changes/prevention • Knowledge of and approaches to reduce the psychological and physical risks and

stresses of the practice of medicine

39

Curricular Goals: SKILLS • The School of Medicine will ensure that before graduation a student will have

demonstrated, to the satisfaction of the faculty, the following: • The ability to obtain an accurate medical history and the ability to perform both a complete

and an organ specific examination, including a mental status examination (See appendix A) • The ability to perform routine technical procedures (See appendix B) • The ability to interpret the results and be aware of the indications, complications, and

limitations of commonly used diagnostic procedures (See appendix C) • The ability to demonstrate knowledge of theories and principles that govern ethical decision

making • The ability to reason deductively and inductively in solving clinical problems • The ability to construct appropriate differential diagnoses and treatment plans for patients

with common conditions, both acute and chronic, including medical, psychiatric, and surgical conditions, and those requiring short- and long-term rehabilitation

• The ability to recognize patients with immediate life threatening conditions regardless of etiology, and to institute appropriate initial therapy

• The ability to recognize and outline an initial course of management for patients with serious conditions requiring critical care

• The ability to communicate effectively, both orally and in writing, with patients, patients’ families, colleagues, and others with whom physicians must exchange information in carrying out their responsibilities

• The ability to select appropriate tests for detecting patients at risk for specific diseases and to determine strategies for responding appropriately

• The ability to retrieve, critically review, and effectively utilize biomedical information from electronic databases and other resources for solving problems and making decisions that are relevant to the care of individuals and populations

• The ability to evaluate the economic, psychosocial, and cultural factors that impact the health of patients and families and to incorporate these into assessment and treatment plan

40

Curricular Goals: ATTITUDES • The School of Medicine will ensure that before graduation

a student will have demonstrated, to the satisfaction of the faculty, the following:

• A commitment to advocate the interests of one’s patients • Compassionate treatment of patients, and respect for their

privacy and dignity • Honesty and integrity and dutifulness in all interactions with

patients, their families, colleagues, and others with whom physicians interact

• An understanding of, and respect for, the roles of other health care professionals, and the need to collaborate with others in caring for patients and promoting health

• A commitment to provide care to patients who are unable to pay and to advocate for access to health care for members of underserved populations

41

2. Developmental Terms:

• “Novice”

• “Advanced learner”

• “Expert”

can identify absence of pulse

can distinguish specific arrythmias

can manage ventricular fibrillation

42

Dreyfus and Dreyfus

Novice Advanced beginner Competent performance Proficient performance Intuitive expert Master

Mind Over Machine (1986)

students

residents

facutly

43

•time-line, progression

included levels of function

•essential for multi-year

training

•although the terms (“novice”,

“master”…) remain generic

Developmental vs. Analytic

44

Goals: ACGME “Competencies”

• Medical Knowledge

• Interpersonal & communication skills

• Professionalism

• Patient Care

• Practice-based learning

• System-based Practice

analytic? developmental?

What about the last three??

45

A cube to encompass competence

Medical Knowledge

Patient Care

Interpersonal skills

Professionalism

Practice-based learning/improv.System-based Practice

UME --> GME CME

clinichospital

school

46

Resistance is futile

47

Educational Goals

Curriculum

Evaluation

Feedback / Grading

Depends on expectations

community

48

• “syn-thetic” -putting the learner back together

• “K S A” are all required, integrated• terms are a bit less generic, more

behavioral

3. the “Synthetic”framework

49

“R.I.M.E. Scheme”

– Reporter

– Interpreter

– Manager/Educator

Pangaro, Academic Med, 1999

50

a framework to classify level of function

• rudimentary reporting: “My patient has

a fever, cough and a bad rash - it’s

vesicular or pustular ….”

• rudimentary interpreting: “I think it

might be due to chicken pox or

herpes.”

51

• rudimentary manager/educator:

“I’d consider a smear of the

fluid and a chest x-ray . …We

might observe or treat with

acyclovir, …but I’m not sure. I’ll

have to look this up.”

52

Alternative model R.I.M.E.

• Framework that is

– developmental

– behavioral•student can visualize, framework for observer

– “synthetic”

53

“RIME”

• Reporter: takes ownership of getting the facts on every patient (“what”?)

• Interpreter: takes ownership of thinking and explaining (“why”?)…

• Manager: takes ownership of planning with patient (“how?)…

• Educator: takes ownership of developing and sharing expertise…

54

Matrix: transition to higher expectations

I II III IV PGY1 PGY2/3 Practice

EDUCATOR

MANAGER  

INTERPRETER

REPORTER

I = introduced R = repetition P = proficiency

I R R R P

I R P

I R P

I R P

M

M

M

M

M = mastery in practice

55

Duty and Expertise (Pellegrino)

• Each RIME “level” is a way of asking, does the student fulfill that promise

• Making a diagnosis, not “giving” it.

• “Objective”

• Does it affect teacher ratings?

56

Example: Construct Validity Grade Distributions Univ. of Utah

0

10

20

30

40

50

60

0 0.5 1 1.5 2 2.5 3 3.5 4

Numeric

Low Pass Pass HP HonorsLow Pass Pass HP Honors

Battistone Acad. Med. , 2001

57

Grade Distributions Univ. of Utahafter RIME methods

0

10

20

30

40

50

60

0 0.5 1 1.5 2 2.5 3 3.5 4

RIMENumeric

OO R I M E R I M E

Battistone Acad. Med. , 2001

58

Case using RIME:

W.O., student, “presents” Mrs. Jones:

• 45yo woman with acute lower back pain

• gives detailed description clinical picture suggestive of acute lumbar strain

•  through physical examination, – blood pressure 130/80, heart rate 80 – left-sided para-spinal tenderness L2 – L5

59

Case using RIME (continued)

• while student is writing up findings in the patient’s record, you interview and examine patient.

• Mrs. Jones asks: “Doctor, can you take my blood pressure since no one has?”

• at what “RIME” level is this student?

60

Is the student ready for more responsibility – yes or no?

RIME is a razor.

61

Complimentary Approaches

Analytic (domains)

• Attitude

• Skills

• Knowledge

Synthetic (“steps”)• (non-reporter)• Reporter• Interpreter• Manager/

EducatorACGME “Professionalism”

62

The RIME rhythm is familiar:

…….S.0…..

….A…………

….P………….

H&P………….

Assessment..

Plan………….

Reporter

Interpreter

Manager/

Educator

63

X X X X XX X X

Reliable

X X X X XX

X X X X X XX X

Valid

Reliability versus validity

64

reliability

• Computed estimate of whether an assessment tool is testing a single construct.

• A reliability of 0.8 is considered needed for high stakes decisions (80% signal).

65

Typical Reliabilities

exam type alpha

• “shelf”-100 MCQ 0.75 - .8

• Step 1 USMLE 0.9 - .95

• OSCE (6 stations) 0.5 - 0.6

• OSCE (12 stations) 0.7 - 0.9

66

Third-year performance Level = Grade

• (Observer)Low Pass

• Reporter Pass

• Interpreter High Pass• Manager/ Honors

Educator

D 1.0

C 2.0

B 3.0

A 4.0

67

USU Medicine clerkship

(students) Reliability

n = 467 0.83

Roop, Amer J Med, 2001

68

X X X X XX

X X X X X XX X

Valid

validity

What can you infer from your evaluation? Are you measuring what is important?

69

EDP Evaluation System

• RIME Vocabulary

• Formal Evaluation Sessions– Sit down with teachers every few weeks– Noel, J Med Ed, 1987 (detecting marginal

students)– “Frame of reference training”

70

Content Validity: Detecting Deficiencies in Professionalism

% of professionalism domains rated less than acceptable

5.5

11.4

5.7

11.910.2

17.7

0

5

10

15

20

Ambulatory# Ward+

DI

Checksheet DI

Written DI

Eval Session DI

Hemmer, Academic Medicine, 2000.

71

Predictive Validity: Sensitivity in predicting Internship Problems

(PGY1 supervisor surveys)

75

8

53

9

0

10

20

30

40

50

60

70

80

MedicineGPA s Medicine

percent

Low Ratings Bad Comment

USU classes of 86 - 93

Lavin, Academic Medicine 1998

72

Acceptance (2): Medicine

• 93% teachers’ evaluations

• 81% NBME subject examination

• 42 % RIME vocabulary

• 32% OSCE

• 22% direct observation (mini-CEX)

Hemmer, Teach Learn Med, 2007

73

Limitations of our system

• RIME is sometimes taken as developmental scheme.

• The analytic model is very strong, and RIME is used for cognitive growth only.

• Wanting to skip the Evaluation Sessions - they take time!

74

Basics Innovations

• Clinicians are good diagnosticians

• We are mentors for our students

• We promise society duty and expertise

• Descriptors provide patterns and exemplars

• Descriptive frameworks allow behavioral feedback

• Synthetic frameworks make this simpler

75

strategy

• Simplicity leads to acceptance and use

• Acceptance to consistency• Consistency to fairness

76

Thanks for coming!

[email protected]