DIAGNOSTIC WORKSHEET AND

PROGNOSTIC SHEET

Sebagai Salah Satu Tugas Dalam Mata Kuliah Evidence Based Medicine

Oleh:
dr. Muhammad Febriandi Djunaidi
04012781923005

Program Studi:
Ilmu Bedah

Pembimbing:
DR. Iche Andriyani Liberty, S.KM., M.Kes

PROGRAM PENDIDIKAN DOKTER SPESIALIS 1

FAKULTAS KEDOKTERAN
UNIVERSITAS SRIWIJAYA
2019
2
 DIAGNOSIS WORKSHEET

Citation:
Rapid Diagnosis of Childhood Pulmonary Tuberculosis by Xpert MTB/RIF Assay
Using Bronchoalveolar Lavage Fluid.

Are the results of this diagnostic study valid?

Was there an independent, blind Yes, this study use Composite Clinical
comparison with a reference (“gold”) Reference Standard (CCRS) as a
standard of diagnosis? comparison.
Was the diagnostic test evaluated in an Yes, it was.
appropriate spectrum of patients (like
those in whom it would be used in
practice)?
Was the reference standard applied Yes
regardless of the diagnostic test result?
Was the test (or cluster of tests) validated Yes
in a second, independent group of
patients?

Are the valid results of this diagnostic study important?

Yes. They are important.

SAMPLE CALCULATIONS
Target disorder
(iron deficiency anemia)
Present Absent Totals
Positive 731 270 1001
Diagnostic (< 65 mmol/L) a b a+b
test result
(serum Negative 78 1500 1578
ferritin) ( 65 mmol/L) c d c+d
809 1770 2579
Totals a+c b+d a+b+c+d

Sensitivity = a/(a+c) = 731/809 = 90%

Specificity = d/(b+d) = 1500/1770 = 85%
Likelihood ratio for a positive test result = LR+ = sens/(1-spec) = 90%/15% = 6
Likelihood ratio for a negative test result = LR - = (1-sens)/spec = 10%/85% =
0.12
Positive Predictive Value = a/(a+b) = 731/1001 = 73%
Negative Predictive Value = d/(c+d) = 1500/1578 = 95%
Pre-test probability (prevalence) = (a+c)/(a+b+c+d) = 809/2579 = 32%
Pre-test odds = prevalence/(1-prevalence) = 31%/69% = 0.45
Post-test odds = pre-test odds  LR
Post-test probability = post-test odds/(post-test odds +1)

3
OUR CALCULATIONS

1. Xpert MTB/RIF Assay

Target disorder
(Pulmonary Tuberculosis)
Totals
Present Absent

Diagnostic Positive 44 0 44
test result
(Xpert
MTB/RIF 168 207
Negative 39
assay)

Totals 83 168 251

Sensitivity = a/(a+c) = 44/83 = 53%
Specificity = d/(b+d) = 168/168 = 100%
Likelihood ratio for a positive test result = LR+ = sens/(1-spec) = 53%/0% = ~
Likelihood ratio for a negative test result = LR - = (1-sens)/spec = 47%/100% =
0.47
Positive Predictive Value = a/(a+b) = 44/44= 100%
Negative Predictive Value = d/(c+d) = 168/207 = 81.2%
Pre-test probability (prevalence) = (a+c)/(a+b+c+d) = 83/168 = 49.4%
Pre-test odds = prevalence/(1-prevalence) = 49.4%/50.6% = 0.97
Post-test odds = pre-test odds  LR  0.97 x 0.47 = 0.4559
Post-test probability = post-test odds/(post-test odds +1)  0.4559/1.4559 = 0.31

2. MTB Culture

Target disorder
(Pulmonary Tuberculosis)
Totals
Present Absent

Diagnostic Positive 24 0 24
test result
(MTB
Culture)
Negative 59 168 227

Totals 83 168 251

Sensitivity = a/(a+c) = 24/83 = 28.9%
Specificity = d/(b+d) = 168/168 = 100%
Likelihood ratio for a positive test result = LR+ = sens/(1-spec) = 28.9%/0% = ~

4
Likelihood ratio for a negative test result = LR - = (1-sens)/spec = 71.1%/100% =
0.71
Positive Predictive Value = a/(a+b) = 24/24 = 100%
Negative Predictive Value = d/(c+d) = 186/227 = 74%
Pre-test probability (prevalence) = (a+c)/(a+b+c+d) = 83/251 = 33%
Pre-test odds = prevalence/(1-prevalence) = 33%/67% = 0.49
Post-test odds = pre-test odds  LR  0.49x0.71 = 0.3479
Post-test probability = post-test odds/(post-test odds +1)  0.3479/1.3479 = 0.258

3. AFB Microscopy

Target disorder
(Pulmonary Tuberculosis)
Totals
Present Absent

Diagnostic Positive 7 0 7
test result
(AFB
micros
Negative 76 168 244
copy)

Totals 83 168 251

Sensitivity = a/(a+c) = 7/83 = 8.4%
Specificity = d/(b+d) = 168/168 = 100%
Likelihood ratio for a positive test result = LR+ = sens/(1-spec) = 8.4%/0% = ~
Likelihood ratio for a negative test result = LR - = (1-sens)/spec = 91.6%/100% =
0.916
Positive Predictive Value = a/(a+b) = 7/7 = 100%
Negative Predictive Value = d/(c+d) = 168/244 = 68.8%
Pre-test probability (prevalence) = (a+c)/(a+b+c+d) = 83/251= 3.2%
Pre-test odds = prevalence/(1-prevalence) = 3.2%/96.8% = 0.03
Post-test odds = pre-test odds  LR  0.03 x 0.916 = 0.02748
Post-test probability = post-test odds/(post-test odds +1)  0.02748/1.02748 =
0.0267

Can you apply this valid, important evidence about a diagnostic test in caring for
your patient?

Is the diagnostic test available, affordable, The diagnostic test is affordable,

accurate, and precise in your setting? accurate and precise in our setting, but
maybe it is not available in most
places.

5
 Can you generate a clinically sensible Yes.
estimate of your patient’s pre-test probability
(from personal experience, prevalence
statistics, practice databases, or primary
studies)?
 Are the study patients similar to your Yes, they are.
own?
 Is it unlikely that the disease possibilities Yes, it is.
or probabilities have changed since the
evidence was gathered?
Will the resulting post-test probabilities Yes, it will affect our management.
affect your management and help your
patient?
 Could it move you across a test- Yes, this diagnostic test is precise and
treatment threshold? accurate.
 Would your patient be a willing partner
in carrying it out? Yes, our patient will do this diagnostic
test if we explain the benefit of this
test.
Would the consequences of the test help Yes, this diagnostic test will help in
your patient? diagnosing childhood pulmonary
tuberculosis sooner so the therapy also
begin sooner.

Additional notes:

6
 PROGNOSIS WORKSHEET
Citation:
PLA1A2 platelet polymorphism predicts mortality in pre-diabetic subjects of the
population based KORA S4-Cohort
Are the results of this prognosis study valid?
Was a defined, representative sample of
patients assembled at a common (usually
early) point in the course of their disease?
Was patient follow-up sufficiently long
and complete?
Were objective outcome criteria applied
in a “blind” fashion?
If subgroups with different prognoses are
identified, was there adjustment for
important prognostic factors?
Was there validation in an independent
group (“test set”) of patients?
Are the valid results of this prognosis study important?
How likely are the outcomes over time?
How precise are the prognostic
estimates?
If you want to calculate a confidence interval around the measure of prognosis:
7
Yes, a defined and representative
sample of patients assembled at a
common point in the course of their
disease.
Yes, patient was followed up
sufficiently long (within 10 years).
No “blind” fashion in this journal.
Yes, there was. Model was adjusted for
age, sex, waist-hip ratio, blood pressure
(diastolic and systolic), cholesterol
(total, HDL, LDL), smoking status
(categorized: non-smoker, former
smoker, current smoker), alcohol intake
categorized: ≥20 g/day for women; ≥40
g/day for men), physical activity
(categorized: >1 h per week)
Yes, there was.
PLA2 and mortality significantly
correlates in non-diabetics subjects with
HbA1c values of >5.5% up to 6.5%,
including the pre-diabetic subjects.
Elevated blood glucose levels beyond the
diabetic threshold are a powerful predictor
of 30 day mortality in acute heart failure
patients, emphasizing the critical role of
the pre-diabetic state.
This study has more than 4000 subjects, so
the standard error of this study is small.
Confidence interval (CI) become more
narrower and show that this prognostic
estimates are precise.
 Clinical Measure
Proportion (as in the rate of
some prognostic event, etc.)
where:
the number of patients = n
the proportion of these patients who
experience the event = p
Standard Error (SE) Typical Calculation of CI
If p = 4261/6640 = 0.64 (or
64%) and n = 6640
 p  (1  p) / n
SE =
where p is proportion and  0.64  (1  0.64) / 6640
n is number of patients = 0.000035 (or 0.0035%)

n from your evidence: 6640 Your calculation:

p from your evidence: 4261 SE: 0.0035%
95% CI is 64% ±1.96 × 0.0035%
or 63.9932% to 64.0068%

Can you apply this valid, important evidence about prognosis in caring for your
patient?

Were the study patients similar to your Yes, it was.

own?
Will this evidence make a clinically Yes, it will make an important impact in
important impact on your conclusions my conclusions about what to offer to my
about what to offer or tell your patient? patient.

Additional notes: