Statistical Analysis
Descriptive statistics were reported as frequency (%) for categorical
variables. We implemented a disproportionality analysis to evaluate the
correlation between the queried drugs and the ADR of priapism. Reporting
odds ratio (ROR) and the 95% confidence intervals (CI) were calculated
according to Stricker and Tijssen. ROR is a homolog to the odds ratio
for case-control studies and corresponds to the exposure odds among
cases of priapism over the odds of exposure among non-cases:
ROR = (a/b) × (c/d); a: drug and ADR of interest, b: drug of
interest with other ADRs, c: other drugs with ADR of interest,
and d: other drugs and ADRs.
For a drug of interest, cases (a) were considered the reports of
“priapism” according to MedDRA, while non-cases (b) were all other ADR
reports during the same period. The analysis was then stratified for the
age groups of the consumers. An ROR was considered significant if the
number of drug-ADR combinations was three or more. Statistical analysis
was performed using Stata 13 (StataCorp, TX, USA). The design and
reporting of the study followed the STROBE guidelines and checklist for
observational studies (see appendix).