Biostatistical And Pharmacoeconomic Measures - NAPLEX

$NNH=\frac{1}{ARI}$. Takes the reciprocal of absolute risk increase to find patients needed for one adverse outcome.

$NNT=\frac{1}{ARR}$. Takes the reciprocal of absolute risk reduction to determine patients needed for one beneficial outcome.

$ARR=CER-EER$. Subtracts the experimental event rate from the control event rate to find the absolute benefit of treatment.

$\frac{\text{existing cases}}{\text{total population}}$. Represents the fraction of the population affected by the condition at a specific point in time.

$\frac{\text{new cases}}{\text{person-time at risk}}$. Quantifies the rate of new cases per unit of person-time, accounting for varying durations of risk exposure.

$\frac{\text{new cases in period}}{\text{population at risk at start}}$. Measures the proportion of initially at-risk individuals who develop the condition over the specified time frame.

$\frac{TP}{TP+FN}$. Divides true positives by all actual positives to assess the test's ability to detect the condition.

$OR=\frac{a\times d}{b\times c}$. Uses the cross-product ratio of a contingency table to estimate relative odds in case-control studies.

$RRR=\frac{ARR}{CER}=\frac{CER-EER}{CER}$. Divides the absolute reduction by the control rate to express proportional risk decrease due to intervention.

$RR=\frac{\text{risk in exposed}}{\text{risk in unexposed}}$. Divides the probability of the outcome in the exposed group by that in the unexposed to assess association strength.

$\frac{TN}{TN+FN}$. Divides true negatives by all negative results to determine the probability of no disease given a negative test.

$\frac{TP}{TP+FP}$. Divides true positives by all positive results to determine the probability of disease given a positive test.

$ICER=\frac{C_1-C_0}{E_1-E_0}$. Divides the difference in costs by the difference in effectiveness to assess value of the new intervention.

Chi-square test (or Fisher exact if expected counts are small). Tests independence of categorical variables; Fisher exact is used for small sample sizes to ensure accuracy.

$1.5$. Divides exposed risk by unexposed risk to quantify the relative increase in outcome probability.

$\$4{,}000\ \text{per QALY}$. Divides the $4,000 cost difference by the 1 QALY gain to find cost per additional quality-adjusted life year.

$0.80$. Divides true positives by total actual positives to measure the test's detection rate.

$LR^+=\frac{\text{sensitivity}}{1-\text{specificity}}$. Divides sensitivity by the false positive rate to indicate how much a positive result increases disease likelihood.

$SEM=\frac{SD}{\sqrt{n}}$. Divides the standard deviation by the square root of sample size to estimate mean variability.

$\frac{TN}{TN+FP}$. Divides true negatives by all actual negatives to evaluate the test's accuracy in ruling out the condition.

$z=\frac{x-\mu}{SD}$. Subtracts the mean from the value and divides by standard deviation to standardize for normal distribution.

$\bar{x}\pm^1.96\times SEM$. Adds and subtracts 1.96 times the standard error from the mean for a 95% confidence range in large samples.

$LR^-=\frac{1-\text{sensitivity}}{\text{specificity}}$. Divides the false negative rate by specificity to indicate how much a negative result decreases disease likelihood.

Independent (two-sample) $t$ test. Compares means assuming normality and independence to detect significant differences between groups.