Transcript feeling good about - University of Southern California

Information Skills
for the Clinician
or “feeling good about not knowing everything”1
Evans Whitaker, M.D., M.L.I.S.
Keck School of Medicine
Norris Medical Library
Los Angeles, CA 90089-9130
[email protected]; (323) 442-1128
1stolen
from Slawson, D.C., Shaughnessy, A.F., Bennett, J.H. (1994). Becoming a medical information
master: feeling good about not knowing everything. Journal of Family Practice, 38(5), 505-517.
ASK QUESTIONS!
SHAMELESS LIBRARY PLUG:
Call, email, IM your local medical librarian early and often!!
Go to Norris homepage -- http://www.usc.edu/nml.
Click on the HELP link in right upper corner to reach us.
Introduction
No EBM today…previous groups felt they knew
enough
• Pick up (3) handouts and (1) evaluation form
• Time is cut by 15 minutes – we will move briskly
• As a long-time clinician I use a practical approach to
getting information – whatever works as long as the
quality of the information is good
• I will not spend time reviewing sources from1st and 2nd
year
• At the end of the session please complete the
evaluation forms -- so we can continue to
improve!
• Thanks in advance
•
I believe I need more training in
Evidence-Based Medicine…
0%
. ..
m
to
o
I’v
e
ha
d
t..
.
I’v
e
ha
d
ab
ou
0%
gr
ee
0%
Di
sa
0%
ee
0%
Ag
r
5.
re
e
4.
Ag
3.
ng
ly
2.
Strongly Agree
Agree
I’ve had about the
right amount
Disagree
I’ve had too much
already!
St
ro
1.
How comfortable are you with EBM?
bl
e
0%
un
co
m
fo
rta
Ve
ry
fo
rta
bl
e
so
0%
Un
co
m
ta
Co
m
fo
r
0%
So
-
0%
bl
e
0%
le
5.
ta
b
4.
fo
r
3.
co
m
2.
Very comfortable
Comfortable
So-so
Uncomfortable
Very
uncomfortable
Ve
ry
1.
I search MEDLINE (PubMed/Ovid)…
lu
e.
..
On
c
ei
n
ab
m
ea
0%
Ne
ve
r
0%
on
th
0%
On
c
*The second full moon in a
calendar month, happens about
once every 2.5 years
0%
or
e.
..
4.
rm
3.
yo
2.
Weekly or more
frequently
Once a month
Once in a blue
moon
Never
W
ee
kl
1.
I use MeSH when I search
MEDLINE (Ovid, PubMed)…
0%
t’s
M
eS
H?
0%
W
ha
0%
Ne
ve
r
0%
So
m
et
im
es
4.
s
3.
ay
2.
Always
Sometimes
Never
What’s MeSH?
Al
w
1.
MeSH






Medical Subject Headings
Collects all synonyms for a term/ concept
under one agreed-upon (by NLM) term
Example: heart attack, MI, myocardial
infarction
Searches using MeSH retrieve more articles
than those using title or text words
Scope Note and Tree help find right term
Expand, Focus, Subheadings, and Limits
are all much less important
When I search MEDLINE…
0%
no
tc
on
fid
n.
..
en
ts
ea
r
m
ch
i
at
e..
.
0%
du
se
fu
l
et
hi
fin
lly
et
im
es
fin
ds
om
Is
om
Ia
m
h.
.
w
fin
d
ill
w
0%
ng
us
ef
ul
0%
Iu
su
a
4.
nt
I
3.
co
nf
ide
2.
I am confident I will find what I
want (or know the information
doesn’t exist)
I usually find something useful
I sometimes find useful
material, but almost always
wonder what I am missing.
I am not confident searching
Ovid/PubMed, I’ll stick with
UpToDate!
Ia
m
1.
Preferred source for clinical information?
er
0%
Ot
h
ip
ed
i
a
0%
W
ik
LI
NE
0%
ED
M
oD
at
Up
T
ed
ic
ar
)
es
sM
Ac
c
0%
e
0%
in
e
0%
ch
ol
7.
0%
(S
6.
le
5.
in
e
4.
ed
ic
3.
Go
og
2.
eMedicine
Google (Scholar)
AccessMedicine
UpToDate
MEDLINE
Wikipedia
Other
eM
1.
Introduction 2

Today’s Session:
•

Discuss an approach to finding and using
information in the course of your clinical
practice
Objectives:
•
o
o
At the end of the session you will
have heard one way to address this issue
have had a little practice with the resources.
Introduction 3

The Problem:
•
o
Too many questions
15/d or 2 per 3 patients seen = number of questions
generated in a clinical day (Covell, 1985)
• Not enough time
o Little time to search for information
• Expanding universe
o 2 year doubling time on biomedical information

How to solve the problem?
•
Have adequate baseline subject specific
knowledge
• Know your information resources
• Develop good technology and information skills.
Introduction 4
•
Of those 15 questions per clinical day
a few are urgent - the results will change the care of
your patient
o other questions can wait for answers
• Today we will focus on urgent questions.
o require the ability to find valid, relevant answers in 1-5
minutes
• The other questions
o write them down for later
o (you will not remember them otherwise)
o
Where do we get this quick information?
Point-of-Care Evidence Pyramid
(with apologies to the EBM Evidence Pyramid)
Predigested
Primary Medical
Literature
UpToDate, ACP Pier, Essential Evidence, Clinical Evidence
Ovid EBM Databases, National Guideline Clearinghouse, TRIP
database, SUMSearch
“Raw” Primary Literature
Opinions of Experts
“Curbside Consult”
Opinion of Colleagues
--varying levels of experience and expertise
*
OvidSP Basic Search
A Very Traditional Source
Another Very Traditional Source
Evidence at Clinical Speed

From Y1 and Y2 (will not repeat today)
 UpToDate
 ACP
Pier
Clinical Information Tools
 Essential Evidence
 Clinical Evidence
 National Guideline Clearinghouse

New today
 GoogleScholar
 Scirus.com
 Ovid
Basic Search
 TRIP Database
Google Scholar
•
•
•
•
The library world’s “elephant in the room”*
Good resource, like all “tools” it has its “best uses”
Pros:
o Scholar limits Google searches to “scholarly sites”
o Familiar interface
o Can link to USC resources (change “Scholar
Preferences”)
o "Sensitive" - searches full text of articles
Cons:
o Wide range of validity, currency, relevance –
requires careful screening of materials
o Exclusive users miss the most specific, powerful
medical information tools
o Not "specific" - includes articles not truly related to
topic
Google Scholar
•
What is it good for?
o
o
To begin exploration of a new topic area
To find search terms
•
Limitations
o
o
Too much information to sort
Validity, currency, relevancy vary widely
•
Thing to know:
o
Set Scholar Preferences for Library to "University
of Southern California" on your home computer
(full text links)
Scirus.com
Massive database of “scientific websites”
Product of Elsevier
Material is better screened than Google
Scholar, but still variable in quality
• Same uses as Google Scholar
• Can set library preferences like GS
•
•
•
OvidSP Basic
•
•
•
o
o
•
o
o
•
o
o
o
Rapid “Google-like” searches
“Include related terms” must be checked
≥500 hits is the norm
“Good ones” in first 20 to 30 hits
Nothing good?  reword your search or go to Advanced
For this class only use Ovid training account:
http://ovidsp.ovid.com
username: sci001; password: medical
Create a personal account
save searches
have new articles from saved searches sent to you
annotate articles
OvidSP Basic Search
•
•
Search for “laparoscopic colectomy”
“Complete reference” of a good article
•
•
•
“MeSH Subject Headings” -- can be used to
construct a search similar to Advanced
Ovid Search
Example
Ovid Basic
•
•
can combine with AND or OR
can use limits
TRIP Database (Turning Research Into Practice)
tripdatabase.com
•
•
•
•
•
o
o
o
•
Free EBM database
Information pulled from multiple sources
Lacks full text links
Keyword searches
Contains:
EBM information
patient handouts
medical images
Try - “Ottawa ankle rules” in EBM for X-ray
decision rule
 Hands
on exercise 1: Lumps
Two simple clinical scenarios follow…
Copies of the scenarios are on your desks.
Lumps 1
•
•
•
•
•
•
34 y.o. woman G1 P1 L1 (1.5 year old girl) noticed a lump in her
left breast while doing a self breast exam.
Menarche 12.5Y
Her last period was three weeks ago, cycles are regular 28d. No
hormonal medications.
FMHx: Maternal aunt died from premenopausal breast cancer.
PE: Your exam confirms a 1.5 cm smooth, mobile non-tender
nodule in the upper outer quadrant of the left breast. Otherwise
breast and axillary exam are normal for age.
What do you do? You have two or three minutes while your patient
dresses.
• You do not have a subscription to UpToDate, your books are still in
a box at home, your partners all took vacation for 2 weeks when
you started work, and your parents are not doctors.
Lumps 2
Same clinical setting as “Lumps 1”
67 y.o. male in for routine exam. He has no health
concerns, no current symptoms, no significant past
medical history. His wife “made him come in.”
• PE: Incidentally discovered on exam is a 1.2 cm nontender nodule low in the anterior neck, slightly to the
right of midline. You think this is located in his thyroid
gland. You point it out to your patient and he states he
never noticed anything, and in fact isn’t sure he feels it
now.
• What do you do? Same information dearth as before.
•
•
In-class Assignment 1
•
•
Divide into groups of 2-4
One half of room gets neck nodule, other half
gets breast lump
• No UpToDate, use electronic sources at Norris
Medical Library
• You have 6 minutes to come up with a plan.
• I want to know
o
o
o
o
•
o
The plan
What resources you drew on to make your plan
How reliable is your information?
What else would you like to know?
Each group prepare to present...
Pick a spokesperson, we’ll discuss afterwards.
Points of Discussion -- Breast

Breast
Patient is premenstrual, aunt’s hx does not
appreciably increase risk. SBE finding most common
reason for visit to PCP for breast lump evaluation.
• Waiting until 3-10d post next menses is reasonable,
although this kind of waiting and re-exam is only
specifically mentioned in one guideline I found.
• Persistent nodule merits aspiration, consideration of
US, Mammography, referral to Surgery (or OB/GYN,
depending on community). Breast cyst aspiration is
potentially a primary care procedure. "Triple-test"
• Answers found in UTD, EE, ACP Pier, NGC, MultieBook search, etc.
•
Points of Discussion -- Thyroid

Thyroid
•
•
•
•
•
•
•
4-7% of population have thyroid nodules (Mazzaferri,
E.L.,1993)
Increased cancer risk in those >60 y.o.
Overall cancer risk in a given nodule is <5%.
Women constitute >75% of those with thyroid nodules;
men have higher risk of cancer in a given nodule
This patient is asymptomatic
Bottom line is that FNA needs to be performed by
experienced practitioner. Many recommend TSH or
ultrasound before FNA. That is debated
Answers found in multiple sites. The answers
consistent with one another
Other things…philosophical
•
It is OK to tell your patient that you don’t
know the answer...
•
•
•
Tell them you will investigate and get back to
them
Do not forget to follow up!!
Set an agenda for yourself to get back in
touch with your patient and follow
through on it.
• Some clinicians re-schedule patients a
day to a week later to “guarantee”
closure.
Hands-on interlude 2: Blood
Similar to the Lumps exercise. Stay in your same groups.
Try using different resources this time, continue to stay away
from UpToDate!
Blood

Two problems:
First
•
•
50 y.o. man in for routine exam requests PSA
testing. Father with prostate cancer at 75 y.o.
Prostate exam (DRE) normal today. Wants
free and total PSA.
What do you tell him? What is your advice
based on?
Blood 2

Second
•
•
•
47 y.o. woman requesting BRCA1/2 genetic
testing. She is a mother of 2 and healthy.
Mother and older sister with breast cancer -both before menopause. Neither family
member has been genetically tested.
Do you order the test?
Would you order his PSA?
Yes, he wants it and
evidence supports the
decision.
2. Yes, he wants it, “the
patient is always
right”.
3. Unsure.
4. No, evidence does not
support ordering it.
1.
0%
up
po
..
do
es
n
“t
he
t,
en
ce
ts
i
No
,
ev
id
an
ew
Ye
s,
h
0%
ot
s
pa
Un
su
re
.
...
de
nc
ev
i
nd
ta
ts
i
an
ew
Ye
s,
h
0%
tie
nt
. ..
0%
Discussion -- PSA
“Contentious issue”
The science is unclear  discuss pros/cons
with your patient  informed decision
• What about cost aspects?
•
•
•
•
•
•
If not clearly effective should we spend national
healthcare dollars here?
In the US the “haves” get what we want
The “have nots” don’t get basic medical care
Can we continue in this mode?
Would you order BRCA testing
on your patient?
0%
0%
0%
Un
su
re
3.
No
2.
Yes
No
Unsure
Ye
s
1.
Discussion – BRCA testing
This woman has a high risk family history –
raised her lifetime risk from US average of
~11% to ~25%
• She is at increased risk of carrying a BRCA
mutation – 5% have the mutation
• If she has BRCA mutation risk is ~65-80%
• The consensus seems to be that she should be
referred for genetic counseling
•
•
•
•
To understand the implications of testing
(insurance, cost, etc.)
To consider asking her family members to be tested
To consider whether knowing the result changes
Changing topics….
•
PDA’s … an opinionated view
Do you now use a PDA or
“smart phone”?
0%
Is
oo
n.
..
0%
No
0%
bu
t
3.
No
,
2.
Yes
No, but I soon plan to
No
Ye
s
1.
PDA users...what do you use?
in
d
–W
r
..
s.
ow
alm
-P
dr
0%
Ot
he
0%
OS
0%
PD
A
Go
o
gl
e
An
Sd
bi
an
O
Sy
m
0%
oi
d
ev
i..
.
on
iP
h
0%
PD
A
0%
e
0%
er
ry
..
8.
0%
Bl
ac
kb
7.
o.
6.
,n
5.
ne
4.
ho
3.
ar
tp
2.
Smartphone, not Blackberry or
iPhone
Blackberry
iPhone
SymbianOS device
Google Android
PDA - PalmOS
PDA – Windows Mobile or PPC
Other
Sm
1.
Favorite medical software for
your PDA?
er
0%
Ot
h
Pi
er
0%
AC
P
e
oD
at
...
0%
Up
T
http://www.usc.edu/e_resources/hsl/lists/sub_1
27.php
0%
vid
What other medical PDA software
do you recommend?
What do you like about it?
0%
lE
5.
Es
se
nt
ia
4.
te
s
3.
cr
a
2.
Epocrates
Essential Evidence
UpToDate
ACP Pier
Other
Ep
o
1.
Assumptions
One gadget is better than two
Smartphone
 Availability of medical software is
Blackberry, iPhone
driven by demand for the device
are ascendant
Browsing capability is important
As above
Choose a company with a future
As above
Blackberry
•True keyboard (except Storm)
•3G connectivity
•microSD card storage 1-2GB
•Medical applications now,
many more to come
•Bold $299 (with current
discount) and $40/mo. phone,
$30/mo. Data
•Several service providers
•“What you need”
vs.
iPhone
•Touch screen “keyboard”
with “intelligent fill”
•3G connectivity
•8 vs. 16 GB (iTunes fxn.)
•Medical applications few at
present, with unclear plans
to encourage development
in the future
•iPhone 3G 16GB $299,
with same mo. Costs
•Must switch to AT&T
•“What you want”
Review of Reviews




Reviewers complained about slow typing/
high error rate with iPhone vs. BlackBerry
Reviewers felt that all other aspects seemed
similar (and good)
Noted on recent poll that 79% iPhone users
were “very satisfied” or better, vs. 54% of
BlackBerry users
Reviewers felt that Apple planned to tightly
control software development – unknown
future for new medical products. BlackBerry
software development moving rapidly.
Wrap Up
Strategy…1
• Bookmark Norris home page - starting point for all
information seeking during your USC career
• First
• UpToDate
• (Off-campus? …
• Use ACP Pier, Essential Evidence, Clinical Evidence)
• Enter one central concept; scan results; select best match.
 Interesting Study:
• ACP Pier, InfoRetriever (now Essential Evidence), UpToDate,
•
FirstCONSULT, and DISEASEDEX were compared (Campbell, 2006).
The 18 participant were each given 3 clinical questions, and given 3 minutes
to answer each question. Results showed that users found significantly more
answers with UTD than other products. They heavily preferred UTD interface,
and felt more confident about the results from UTD!
Strategy…2
•
If you are not finding what you want in the
clinical information tools…
•
Second Option
•
OvidSP Basic Search
•
•
•
OR
Consider EBM sources
•
•
You may be able to find a paper, scan it for
information within a short time frame.
Tripdatabase.com OR Clinical Evidence
Third Use GoogleScholar, Scirus.com for
failures of more specific biomedical sources.
Thanks for your attention!!
•
Please fill out evaluations before you
leave!
• Feel free to contact me if you have
information questions.
References
•
•
•
•
•
•
•
•
•
•
•
•
•
Adair, R.F., Holmgren, L.R., (2005) Do drug samples influence resident prescribing behavior? A randomized trial.
American Journal of Medicine, 118, 881-884.
Allison, J.J., Kiefe, C.I., Weissman, N.W., Carter, J., Centor, R.M. (1999). The art and science of searching MEDLINE to
answer clinical questions: Finding the right number of articles. Intl J Tech Assess in Health Care, 15(2), 281296.
Barry, H.C., Ebell, M.H., Shaughnessy, A.F., Slawson, D.C., Nietzke, F. (2001). Family physicians’ use of
medical abstracts to guide decision making: style or substance? Am Board of FP, 14(6), 437-442.
Campbell, R. (2006). An evaluation of five bedside information products using a user-centered, task- oriented approach.
J Med Lib Assoc, 94(4), 435-440.
Covell, D.G., Uman, G.C., Manning, P.R. (1985). Information needs of office practitioners: are they being met? Annals of
Internal Medicine, 103, 596-599.
Guyatt, G., Rennie, D. (2002). Users’ Guides to the Medical Literature: Essentials of Evidence-Based Clinical Practice.
Chicago: AMA.
Greenhalgh, T. (2006). How to read a paper: The basics of evidence-based medicine, 3rd ed. Malden, MA:
Blackwell.
Mazzaferri, E.L. (1993). Management of a solitary thyroid nodule. N Engl J Med, 328, 553-559.
Michaud, G., McGowan, JL., van der Jagt, R., Wells, G., & Tugwell, P. (1998). Are therapeutic decisions supported by
evidence from health care research? Archives of Internal Medicine158(15),1665-1668.
Shaughnessy, A.F., Ebell, M.H., Slawson, D.C. (2008). Information mastery: Basing care on the best
available evidence. In Essentials of Family Medicine, 5th ed. Philadelphia: Wolters.
Slawson, D.C., Shaughnessy, A.F., Bennett, J.H. (1994). Becoming a medical information master: feeling good about not
knowing everything. Journal of Family Practice, 38(5), 505-517.
Slawson, D.C. (2005). Teaching evidence-based medicine: should we be teaching information management
instead? Acad Med, 80(7), 685-689
Wilczynski, N.L., Walker, C.J., McKibbon, K.A., Haynes, R.B. (1995). Reasons for the loss of sensitivity and
specificity of methodologic MeSH terms and textwords in MEDLINE. Proc Annu Symp Comp Appl Med Care,
19, 436-440.