Skip to main content
. 2024 Sep 20;15:1463560. doi: 10.3389/fphar.2024.1463560

TABLE 1.

Detailed formulas for disproportionality analysis.

Methods Formula Signal standard
ROR ROR = ad/bc
ROR 95%CI = eln (ROR)±1.96 (1/a+1/b+1/c+1/d)0.5
a≥3
lower limit of ROR 95%CI > 1
MHRA PRR = a (c + d)/c (a+b)
PRR 95%CI = eln (PRR)±1.96 (1/a-1/(a+b)+1/c-1/(c + d))0.5
a≥3
lower limit of PRR 95%CI > 1
BCPNN IC = log2 [a (a+b + c + d)]/[(a+b) (a+c)]
γ = γij [(N+α) (N+β)]/[(a+b+αi) (a+c+βj)]
E (IC) = log2 [(a+γij) (N+α) (N+β)]/[(N+γ) (a+b+αi) (a+c+βj)]
V(IC)=(log2)−2{(N-a+γ-γij)/[(a+γij) (1 + N+γ)] +
(N-a-b+α-αi)/[(a+b+αi) (1 + N+α)] +
(N-a-c+β-βi)/[(a+b+βi) (1 + N+β)]}
SD = 2(V(IC))0.5
IC-2SD = E (IC)-2SD
a≥3
IC-2SD > 0
MGPS EBGM = a (a+b + c + d)/[(a+c) (b + d)]
EBGM05 = eln (EBGM)±1.96 (1/a+1/b+1/c+1/d)0.5
a>0
EBGM05 > 2

Note: γ, γij are the Dicichlet distribution parameter; αi, α, βj, β are Beta distribution parameter; SD, is the standard deviation; IC-2SD, is the lower limit of IC, 95% CI; hypothesis α = β = 2, γij = βj = αi = 1; ROR: reporting odds ratio; MHRA:medicines healthcare products regulatory agency; PRR: proportional reporting ratio; BCPNN: bayesian confidence propagation neural network; MGPS: Multi-Item Gamma Poisson Shrinker; EBGM: empirical bayesian geometric mean; EBGM05: lower limit of EBGM, 95%CI., 95% CI: 95% Confidence Interval.