Biostatistics

Question 1

Positive Predictive Value (PPV)

Answer 1

The positive predictive value (PPV) of a diagnostic test is the probability (i.e. likelihood) that an individual truly has the disease given a positive test. PPV is equal to the number of individuals who have the disease and who test positive (i.e. true positives [TP]) divided by the total number of individuals with a positive test result (TP + false positives [FP]):

PPV = TP / (TP + FP)

Predictive values depend on the prevalence of the disease in the study population; as the disease prevalence increases, PPV increases and NPV decreases, and vice versa.

Question 2

Negative Predictive Value (NPV)

Answer 2

NPV is equal to TN / (TN + FN) and describes the proportion of individuals with a negative test result who really do not have the disease.

NPV = TN / (TN + FN)

Question 3

Sensitivity

Answer 3

The proportion of individuals with a known positive condition for which the test result is positive. It is an intrinsic measure of the test’s ability to correctly identify individuals with the disease.

Screening > Rule Out.

Sensitivity = TP / (TP + FN)

Question 4

Specificity

Answer 4

Have Diagnosis > Rule In.

Specificity = TN / (TN + FP)

Question 5

Phase - I Clinical Trail

Answer 5

• Assesses safety and pharmacokinetics.
• Phase I is usually conducted with a small number (eg, 20-80) of healthy participants.
• It often is performed in a strictly controlled environment with extensive biochemical and physiologic monitoring.

Question 6

Phase - II Clinical Trail

Answer 6

• Assesses preliminary efficacy and optimal dosing.
• Phase II is conducted with a small number of participants who have the condition of interest, and patient outcomes are often compared to historical controls.
• It sometimes is called a pilot study.

Question 7

Phase - III Clinical Trail

Answer 7

• Compares treatment to standard of care.
• Phase III is conducted as a large, randomized, controlled trial (in contrast to this study, which is a surveillance study with no control group) to better assess treatment response and safety.
• It may include analysis of treatment effects in selected subsets of the target patient population.

Question 8

Phase - IV Clinical Trail

Answer 8

• Phase IV (post marketing) trials are conducted with drugs that have already been approved for use in order to study long-term effectiveness and better characterize uncommon or delayed adverse effects.
• They are typically designed to identify small treatment effects (i.e. a high power study with large numbers of participants) in diverse patient populations, and the results may be used to refine the use of drugs in clinical practice.

Question 9

Case-Control Study

Answer 9

• A case-control study is designed by selecting individuals with a particular disease (cases), individuals without that disease (controls), and then evaluating previous exposure status.
• 1 | 0
• Diseased (Cases) | Non-Diseased (Controls)
• Compares risk factor frequency

Question 10

Randomized Clinical Trial

Answer 10

• A randomized clinical trial directly compares >2 treatments.
• Usually, the subjects are randomly assigned to experience a specific exposure (eg, a medication) or no exposure (eg, placebo) and are then followed to assess for the outcome of interest (eg, disease).

Question 11

Cross-sectional Study

Answer 11

• In a cross-sectional study, exposure and outcome are measured simultaneously at a particular point in time (“snapshot study”).
• In other study designs, a certain time period separates the exposure from the outcome.
• Compares disease prevalence

Question 12

Prospective Cohort Study

Answer 12

• A group of people followed over a period of time
• Risk factors +/-
• Check disease incidence

Question 13

Retrospective Cohort Study

Answer 13

• Review past records
• Risk factors +/-
• Compare disease incidence

Question 14

Hardy-Weinberg analysis

Answer 14

• p + q = 1 (Allele Frequency)p = normal allele frequency
  q = mutant allele frequency
• p² + 2pq + q² = 1 (Phenotypic Frequency)p² = frequency of normal individuals
  2pq = carrier frequency
  q² = frequency of diseased individuals
• √q² = q
  (Calculating mutant allele frequency from disease prevalence)√q² = disease prevalence
  q = mutant allele frequency

Question 15

Number Needed to Harm (NNH)

Answer 15

The number of people who must be exposed to a treatment to cause harm to 1 person who otherwise would not have been harmed is known as the number needed to harm (NNH).

NNH = (1 / ARI)

ARI is the difference in the rate (risk) of the adverse event (AE) between the treatment group and the control group

ARI = (Rate AE treatment − Rate AE control)

Question 16

False Negative (FN)

Answer 16

People who have the disease but test is Negative

FN = (1 - Senstivity) x No of Pt’s with Disease

Question 17

True Positive (TP)

Answer 17

People who have the disease & also test positive

TP = Senstivity x No of Pt’s with the disease