Table 3 Major algorithms used for signal detection.

From: Causes of drug-induced photosensitivity: an analysis using FDA adverse event reporting system database

Algorithms

Equation

Criteria

ROR

ROR=(a/c)/(b/d)

The lower limit of 95% CI > 1, N ≥ 3

\(95\% {\text{CI}} = e^{{\ln (ROR) \pm 1.96\sqrt {\frac{1}{a} + \frac{1}{b} + \frac{1}{c} + \frac{1}{d}} }}\)

PRR

\(PRR = \frac{{a(a + b)}}{{c(c + d)}} = \frac{{a(c + d)}}{{c(a + b)}}\)

PRR ≥ 2, χ2 ≥ 4, N ≥ 3

\({X^2}=\frac{{{{(|ad - bc| - \frac{n}{2})}^2}n}}{{(a+b)(a+c)(c+d)(b+d)}}\)

BCPNN

IC = log2a(a + b + c + d)/(a + c)(a + b)

IC025 > 0

\(95\% {\text{CI}} = {\text{E}}(IC) \pm 2\sqrt {V(IC)}\)

  1. Equation: a = The number of reports of DIPs with the adverse event of target drug. b = The number of reports of all other drugs with the adverse event of target drug. c = The number of target drug with all other adverse events. d = The number of reports of all other drugs with all other adverse events. n = a + b + c + d. ROR refers to Reporting Odds Ratios, PRR to Proportional Reporting Ratios, and BCPNN to Bayesian Confidence Propagation Neural Network. 95% CI denotes the 95% Confidence Interval, N represents the number of reports, and X² is Chi-Squared. IC stands for Information Component, with IC025 indicating the lower limit of the 95% Confidence Interval for the IC, E(IC) as the IC expectations, and V(IC) as the IC variance.
Back to article page