HOW TO READ AN ARTICLE ABOUT A DIAGNOSTIC TEST

Chitkara MB, Boykan R, Messina C

Stony Brook Long Island Children’s Hospital

WHAT IS EVIDENCE - BASED MEDICINE (EBM)?

The translation of medical research into clinical practice

Integration of best research evidence with clinical experience and patient values

Knowing how to use clinical literature to ensure optimal patient care

EBM is about USING, not doing, research

MOVE PAST THE P VALUE

We are looking for CLINICAL SIGNIFICANCE

STATISCAL SIGNIFICANCE (p ≤ 0.05) is nice…but not always feasible

GOALS FOR EBM CURRICULUM

Present a general approach to use clinical reading time more effectively

Demonstrate techniques that boil down EBM into simple, usable concepts

We want you to GET IT

HOW TO PRACTICE EBM?

Step 1: Frame your patient care question Step 2: Search and find the evidence Step 3: Validate the evidence Step 4: Evaluate the evidence Step 5: Apply the evidence

CASE SCENARIO

You are doing an elective in the Pediatric GI clinic. An 8 year old male presents with complaints of recurrent epigastric chest pain and dyspepsia. He is otherwise well-appearing. You are concerned about H. Pylori infection, but you are reluctant to recommend endoscopy in this otherwise well child. You ask your attending about other non-invasive options, and she tells you to look it up and present your findings at their Journal Club this afternoon.

ANSWERABLE CLINICAL QUESTION(PICO)

Patient – In a child w/ symptomatic GERD

Intervention – Can stool antigen testing be used

Comparison – In place of invasive procedures

Outcome - To detect the presence or absence of H. Pylori infection?

SEARCH AND YOU WILL FIND

You use your lunch break to hit the library. Your PUBMED/MeSH database search yields a

promising article. Prospective study of 103 children undergoing

endoscopy for recurrent abdominal pain (Iranikhah A et al. Iran J Pediatr, Apr 2013; 23(2):138-142).

EBM BIG THREE QUESTIONS

Is this study VALID?

What are the RESULTS?

Are the results APPLICABLE to my patient?

DIAGNOSTIC TEST – VALIDITYPRIMARY GUIDES

Was the “gold standard” applied to all patients?

Was there an independent, blind comparison to reference standard?

Was the diagnostic test evaluated in an appropriate spectrum of patients (like those in whom we would use it in practice)?

DIAGNOSTIC TEST: VALIDITYSECONDARY GUIDES

Work-up or Verification Bias: Did the results of the test being evaluated influence the decision to perform the reference standard?

Were the methods for performing the test described in sufficient detail to permit replication?

DIAGNOSTIC TEST: RESULTS

Do the authors present Likelihood Ratios? If not, is the data needed to calculate the

Likelihood Ratios included? How do I calculate a Likelihood Ratio?

DIAGNOSTIC TEST: RESULTS

Start with Sensitivity and Specificity

Sensitivity: The ability of the test to detect diseased people from a diseased

population Specificity: The ability of a test to

detect healthy people from a healthy population

DIAGNOSTIC TEST: RESULTS

Likelihood Ratios indicate by how much a given diagnostic test result will raise or lower the pretest probability of the target disorder

LR (+)The probability that the patient has a

true positive test, rather than a false positive

LR (-)The probability that the patient has a

true negative test and not a false negative

DIAGNOSTIC TEST: RESULTS

Are the results Clinically Significant?

Sensitivity = a /a+c

Specificity = d / b + d

Positive Predictive Value = a / a + b

Negative Predictive Value = d / c + d

LR(+) = [a / (a+c)] / [b /(b + d)]

LR (-) = [c / (a + c)] / [d / (b+d)]

a+b+c+db + da + c

c+ddcTest Result (-)

a + bbaTest Result (+)

TotalsDisease AbsentDisease Present

Stool Antigen Tests for H. Pylori in Children

Disease Present

Disease Absent

Totals

Test pos

35 4 39

Test neg

6 58 64

41 62 103

Sen: 35/41 = 85.4%

Spec: 58/62 = 93.5%

LR+: 35/41 ÷ 4/62=13

LR-: 6/41 ÷58/62=0.16

Pre-test probability for this study = 40%

Iranikhah A et al. Iran J Pediatr, Apr 2013; 23(2):138-142

HOW ARE LIKELIHOOD RATIOS USED Know your Pre-Test Probability (PTP) Varies from patient to patient PTP may be considered disease prevalence

for your population Check first few sentences of article

introduction to see if authors describe their disease prevalence

Might have to use personal clinical judgment Exerts a major influence on the diagnostic

process

THE FAGAN NOMOGRAM

TP 40%

DIAGNOSTIC TEST: RESULTS

LR = 1: post-test probability is exactly the same as pre-test probability

LR > 1 increases the probability that the target disorder is present

LR < 1 decreases the probability that the target disorder is present

LR = 8 means that it is 8 times more likely that a positive test is a true positive than a false positive.

LIKELIHOOD RATIOS

LR > 10 or < 0.1 generate large changes from pre-test to post-test probability

LR = 5 - 10 or 0.1 - 0.2 generate moderate shifts pre-test to post-test

LR = 2 – 5 or 0.5 – 0.2 generate small, but sometimes important changes in probability

LR = 1 – 2 or 0.5 – 1 are rarely important shifts

DIAGNOSTIC TESTLIKELIHOOD RATIO VS PREDICTIVE VALUE

Prevalence = all study pts with disease / all pts in study

Likelihood Ratio is prevalence-independent

Predictive Value is wholly prevalence-dependent

Prevalence is often higher in studies compared to routine practice due to selection bias.

DIAGNOSTIC TEST – APPLICABILITY

Test Properties may change with a different mix of disease severity or a different distribution of competing conditionsWhen patients with the target disorder all have severe disease, the LR’s will move away from a value of 1 (sensitivity increases)

When patients without the target disorder have competing conditions that mimic the test results of patients who do have the target disorder, the LRs move toward one, and the test appears less useful

DIAGNOSTIC TEST – APPLICABILITY

Test Threshold – probabilities below which a clinician would dismiss a diagnosis and order no further tests

Treatment Threshold – probabilities above which a clinician would consider the diagnosis confirmed, and would stop testing

DIAGNOSTIC TEST – APPLICABILITY

When the probability of the target disorder falls between the test and treatment thresholds, further testing is mandated

Once test and treatment thresholds are determined, the post-test probabilities have direct treatment implications

CLINICAL BOTTOM LINE What are the desires and expectations of my

patient? Will this test result in better outcomes? Will this test change my management of my

patients?

PRACTICE CASE REFERENCES

Case 1: Poehling KA et al. Accuracy and Impact of a Point-of-Care Rapid Influenza Test in Young Children with Respiratory Illnesses. Arch Ped Adol Med, July 2006; 160Case 2: Janguoo A et al. Is urinary 5-hydroxyindoleacetic acid helpful for early diagnosis of acute appendicitis? Am J Emerg Med, 2012; 30:540-544.Case 3: Gaydos CA et al. Performance of the Abbott RealTime CT/NG for Detection of Chlamydia trachomatis and Neisseria gonorrhoeae. J Clin Micro, 2010; 48(9):3236-3243.

EBM REFERENCES/RESOURCES Jaeschke R, Guyatt G, Sackett DL. User’s Guide to the

Medical Literature. How to use an Article About a Diagnostic Test. A. Are the Results of the Study Valid? JAMA, 1994; 271(5):389-391.

Jaeschke R, Guyatt G, Sackett DL. User’s Guide to the Medical Literature. How to use an Article About a Diagnostic Test. B. What Are the Results and Will They Help Me in Caring for My Patients? JAMA, 1994; 271(9):703-707.

Iranikhah A, Ghadir MR, Sarkeshikian S, Saneian H, Heiari A, Mahvari M. Stool Antigen Tests for the Detection of Helicobacter Pylori in Children. Iran J Pediatr, 2013; 23(2):138-142.