Transcript board review : ebm & prevention

BOARD REVIEW : EBM &
PREVENTION
Jesse C James
Q & A: Harrisons I-1
• A physician is deciding whether to use a new test
to screen for Disease X. The prevalence is 5%.
Sensitivity 85% Specificity 75%. In a population
of 1000, how many with X will be missed in
screening?
• A 50
• B 42
• C 8
• D 4
• E 10
Q&A
• What is this question asking?
How many people w disease have negative test?
What is the expected number of false negatives?
%FN=(1-sensitvity) or #FN= Prevalent cases-TP cases
• Prevalence = number of people with disease
Prevalence = 5% * 1000 = 50
• Sensitivity= 85% with disease will be True Positives
Sens= .85*50 = 42.5 True Positives
FN = 50-42.5= 7.5 FN
Sensitivity
• The percentage of persons with the disease of
interest who have positive test results
= True-post/(True-post + False-neg) x 100 (HUH??)
= True positives/All pts w disease (OK!!!)
Tests that are very sensitive are clinically useful to
rule out a disease…because if you had it, you
would be positive
IDEPENDENT of the prevalence of disease
Specificity
• The percentage of persons without the
disease who have negative test results
= True-neg/(True-neg + False-post)
= True-neg/All patients with disease
The more specific the more likely persons without
the disease of interest will be excluded.
Both sensitivity and specifity are independent of
prevalence
Q&D
• PREVALENCE: Prob (+ disease)
• SENSITIVITY: Prob (True Positive) given
disease
• SPECIFICITY: Prob (True Negative) given no
disease
• PPV: Prob (+disease) given positive test
• NPV: Prob (-disease) given negative test
Q & A: Harrisons I-1
• A physician is deciding whether to use a new test
to screen for Disease X. The prevalence is 5%.
Sensitivity 85% Specificity 75%. In a population
of 1000, how many with X will be missed in
screening?
• A 50
• B 42
• C 8
• D 4
• E 10
ANSWER
•
•
•
•
The answer is C. 8
Prevalent cases 50
True positives=Prev Cases*Sens= 50*.85= 42.5
False negatives=Prev Cases-TP=7.5
2x2 Table
Q&A Harrison’s I-2
• How many patients were told erroneously that
they have disease based on this test?
• A.713
• B.505
• C.237
• D.42
• E.8
ANSWER
• What is the question asking?
How many without disease have a positive test?
How many false positives are expected?
Total without disease – TN cases= # False positives
• Prevalence = number of people with disease
People w/o disease = 1000-(5% * 1000) = 950
• Specificity= 75%of people without disease will be TN
cases
Spec(TN) cases= .75*950 = 712.5 TN cases
FPcases = 950-712.5= 237 FP cases
ANSWER
• The answer is C. 237
Q&A: Harrisons I-3
• Drug X is investigated in a meta-analysis for its
effect on mortality after a myocardial
infarction. It is found that mortality drops
from 10 to 2% when administered. What is the
ARR?
• A. 2%
• B. 8%
• C. 20%
• D. 80%
ANSWER
• ANSWER IS B 8%
• ARR = Difference between risk in Exposed
(Exposed Event Rate) vs risk in Control
(Control Even Rate)
• ARR= CER-EER
• 10%-2%=8%
• The RELATIVE RISK REDUCTION=ARR/CER=
8%/10%=80%
Q&A: Harrison’s I-4
• How many patients will have to be treated
with drug X to prevent one death?
• A. 2
• B. 8
• C. 12.5
• D. 50
• E. 93
ANSWER
• The answer is C. 12.5
• The NNT is equal to the inverse of the ARR
• 1/ARR= 1/8% = 12.5
Q&A: Harrison’s I-6
• Which of the following regarding CAD in women
is true?
A. Death rates for CAD for men and women have
been increasing over the last 30 years.
B. The most common initial symptom of heart
disease in women is angina.
C. Women with AMI are more likely than men to
present with VTach.
D. Women in all age groups have lower mortality
from MI than males do.
ANSWER
• The answer is B.
• Mortality from CAD has been increasing among
women for the last 30 years and among men it
has been decreasing.
• According to Framingham study angina is the
most common presentation of coronary heart
disease in women.
• Females w MI are more likely than males to
present w cardiogenic shock and males are more
likely to present w Vtach
Q&A: MKSAP14 FIM-4
• 23 yo woman evaluated for migraine occurring fives times
per month wants to restart prophylactic medication
regimen. The pt has already responded poorly to Betablockers and amitriptyline in the past. She wonders
whether gabapentin will help.
Which of the following sources is most likely to provide
reliable information for answering this patient’s question?
A. Randomized controlled trial
B. Case report
C. Case controlled studies
D. Systematic reviews
E. Drug information inserts
Study Types
• Case report/Case series: a description of single or multiple
anecdotal presentation and management of diseases
• Case-control: typically identify patients with an outcome
and then make retrospective survey of exposures. Matches
cases with disease to controls without disease and check
for exposures.
• Cohort study: prospective study that follows groups (the
cohorts) divided by exposure to measure risk of
development of outcome over time
• Systematic review: a literature assessment that identifies a
question, eliminates/includes sources, appraises results
and synthesizes/describes conclusions.
ANSWER
• The answer is D: systematic review
• Systematic reviews and meta-analyses are provided summaries of
voluminous information from multiple publications of original
research. They can show where a body of literature has both
consistencies and controversies and be generalized more reliably
across populations.
• Single RCTs rarely provide definitive answers to broad clinical
questions due to the difficulty in generalizing to a single patient and
the probability of differences being due to chance alone (type I
error).
• Case reports are clinical observations that should never be used to
draw clinical conclusions.
• Case controlled studies are not randomized and prone to bias and
typically describe the association between and exposure and an
event.
Q&A: MKSAP14 FIM-7
• 19 yo F w RLQ abd apin and fever. Abd tender to palpation
w/o rebound/guarding. Pelvic exam normal. From
experience the probability of acute appendicitis is 50%. A
positive abdominal CT has a likelihood ratio of 13.3 for the
diagnosis of acute appendicitis. If the CT scan is positive
for appendicitis what is the approximate increase in
probability that the patient has appendicitis?
A. 5%
B. 15%
C. 30%
D. 45%
E. 60%
LIKELIHOOD RATIOS
• LR: considers both the sensitivity and specificity
combined into a single measure of diagnostic
effectiveness
• +LR= sensitivity/(1-spec)
• Proportion of pts with disease that test positive versus
the prop of pts w/o disease who test positive
• The greater the magnitude of a LR, the more useful the
test is for increasing the probability for confirming a
target disease.
– Positive LR of 2, 5, and 10 increase the probability of
disease by 15%, 30% and 45% respectively
ANSWER
• The answer is D: 45%
MKSAP14 FIM13
• 55 yo w cough and malaise that occurs multiple times
per year. He has 40 pack-year history. On exam he is
afebrile HR, RR, BP wnl. Cardiopulmonary exam is
normal, chest clear w/o consolidation/wheeze and
heart regular. Which of the following is the most
appropriate initial smoking cessation management
step during this visit?
• A. Recommend nicotine gum
• B. Provide a clear, personalized message to the patient
• C. Refer the patient to behavioral modification
• D. Prescribe bupropion
ANSWER
• The answer is B. Provide a clear, personalized
message to the patient
• Although buproprion, nicotine replacement, and
behavioral therapy are appropriate adjunct
interventions, however, using these without
adequately assessing the patients readiness for
behavioral change is premature.
• Brief interventions for as few as 1-3 min have
been shown to result in an increased number of
patients who quit and abstain from cigarette
smoking.
MKSAP GIM6
• 25 yo woman evaluated during for routine exam,
nonsmoker, social alcohol drinker, denies illicit drug
history. She has had 3 sexual partners and is in a
serious monogamous relationship. She has no history
of STIs and takes OCP for pregnancy prevention. She
has a scheduled Pap smear. Which of the following is
most appropriate for this patient?
• A. Encourage sunscreen use
• B. Prescribe multivitamin with folic acid
• C. Screen for Chlamydia
• D. Measure fasting plasma glucose
ANSWER
• The answer is C: Screen for Chlamydia
• CDC recommends annual screening for sexually active
women aged 25 and younger at increased risk for infection
– New or multiple sexual partners
– History or current symptoms of STI
– History of unprotected intercourse
• Age <25 is strongest predictor in men and women
• Benefit of counseling to patients w high sun exposure is
unknown
• A multivitamin w folic acid is recommended for pregnant
women to prevent neural tube defects but is not necessary
in this patient she is not pregnant and on OCPs.
• FPG is not recommended for routine screening without risk
factors for DM.
MKSAP 14 GM37
• 45 yo AAM is evaluated for concerns about prostate
cancer. A close friend was recently diagnosed w
extensive disease and has a poor prognosis. The
patient asks if he should have a screening test for the
disease. He has once per night nocturia and no
hesitancy, freq, or dribbling. Which is the most
appropriate plan of action?
• A. PSA measurement
• B. PSA and DRE
• C. Transrectal US
• D. Random biopsies
• E. Shared Decision making
ANSWER
• The answer is E: Shared Decision Making
discussion risk and benefits.
• USPSTF recommends shared decision making
on the potential risk and benefits of screening
for prostate cancer.
• The PPV for PSA is 30%. Of the men with
positive test, only a third actually have
prostate cancer.
MKSAP 63
• 22 yo nursing school grad is evaluated for preemployment. She has hx of SLE and no recollection of
having chickenpox and her varicella titer is negative.
Which is the most appropriate recommendation?
• A. No vaccination
• B. Single vaccination (shortened series), clear for work
• C. Single vaccination, delay work 4 weeks
• D. Two-dose vaccination series over 6 weeks, clear for
work
• E. Two dose vaccination series over 6 weeks, delay
work for 4 weeks.
ANSWER
• The answer is E: Two dose vaccination series
over 6 weeks, delay work for 4 weeks.
• Due to her age and occupation she is at high risk
for VZV infection an should be immunized w the
usual two doses. The vaccine is recommended
for all adults with no evidence of immunity.
• The live vaccine can cause shedding in the 4
weeks following injection so she should avoid
patients who might be sick from exposure to her.
• A single dose is not advised for anyone
MKSAP14 HO-70
• 30 yo woman G3P2 evaluated for routine exam, two years
had bilat tubal ligation. Took OCPs for 5 years between
births. Healthy and w/o complaints. FH sig for maternal
cousin diagnosed w OvarianCa at age 48. No other FH Brst
or Ovrn Ca. Exam normal. Has friend of Askenazi Jewish
descent w strong FH. Pt wants to know what she can do to
reduce her own risk Brst/Ovrn Ca. Which of the following
would be most appropriate?
• A. Routine Ca screening
• B. Prophylactic bilat ooporectomy
• C. CA-125 measurement
• D. Restart OCPs
• E. Routine daily vitamins
ANSWER
• The answer is A: routine screening.
• Pt not a high risk for cancer so only age and
gender appropriate screening are indicated.
• Pts friend most likely harbors BRCA1 gene
women with this abnormality are encouraged
to use shared decision-making to consider
prophylactic oophorectomy,
chemoprophylaxis or routine CA-125
screening.
MKSAP HO-84
• 59 yo woman is evaluated during a routine exam. Her
family history includes a sister who was recently diagnosed
with advanced stage-ovarian cancer. The remainder of her
medical and family history is noncontributory. Physical
exam normal. Pt is concerned about her risk for developing
ovarian cancer and ask what routine screening methods
will decrease her risk. Which of the following is the most
appropriate recommendation for ovarian cancer screening
in this patient?
• A. No Screening test
• B. Serum CA-125
• C. Transvaginal US
• D. Doppler ovarian exam
Answer
• The correct answer is A.
• No screening tool has been shown to decrease
ovarian cancer mortality in general or in at risk
risk populations.
• No clinical review organizations recommend
routine ovarian ca screening.
• Predictive models developed show the
available screening tools would have at most a
small benefit.
GOOD LUCK GUYS!!!