Biostats

Question 1

Sensitivity vs. Specificity

Answer 1

Sensitivity = TP / (TP + FN)

• the probability that a patient WITH a disease will test positive for that disease.
• SnOUT- high sensitivity rules OUT a disease
• important for screening tests

Specificity = TN/ (TN + FP)

• the probability that a patient withOUT a disease will test negative for that disease
• SpIN- high specificity rules IN a disease
• important for confirmatory tests

Question 2

effect of specificity on PPV

Answer 2

higher specificity means that there are fewer false positives, meaning the positive predictive value is higher.

Question 3

effect of sensitivity on NPV

Answer 3

higher sensitivity means there are fewer false negatives, meaning the negative predictive value is higher

Question 4

when is the confidence interval not statistically significant?

Answer 4

if it crosses the null value (1)

Question 5

Positive likelihood ratio

Answer 5

sensitivity/ (1-specificity)
probability of the patient with disease testing positive, divided by the probability of a patient without the disease testing positive

TP/FP rate

Question 6

Negative likelihood ratio

Answer 6

1-sensitivity / specificity
probability of a patient with the disease testing negative, divided by the probability of a patient without the disease testing negative

FN rate/ TN rate

Question 7

Number needed to treat - formula?

Answer 7

1/ARR (absolute risk reduction)