Calculates Proportional Reporting Rate ("PRR") with confidence intervals, used in disproportionality analysis.
Arguments
- obs
Number of reports for the specific drug and event (i.e. the observed count).
- n_drug
Number of reports with the drug of interest.
- n_event_prr
Number of reports with the event in the background.
- n_tot_prr
Number of reports in the background.
- conf_lvl
Confidence level of confidence or credibility intervals. Default is 0.95 (i.e. 95 % confidence interval).
Value
A tibble with three columns (point estimate and credibility bounds). Number of rows equals length of inputs obs, n_drug, n_event_prr and n_tot_prr.
Details
The PRR is the proportion of reports with an event in set of exposed cases, divided with the proportion of reports with the event in a background or comparator, which does not include the exposed.
The PRR is estimated from a observed-to-expected ratio, based on similar to the RRR and IC, but excludes the exposure of interest from the comparator.
$$\hat{PRR} = \frac{\hat{O}}{\hat{E}}$$
where \(\hat{O}\) is the observed number of reports, and expected \(\hat{E}\) is estimated as
$$\hat{E} = \frac{\hat{N}_{drug} \times (\hat{N}_{event} - \hat{O})}{\hat{N}_{TOT}-\hat{N}_{drug}}$$
where \(\hat{N}_{drug}\), \(\hat{N}_{event}\), \(\hat{O}\) and \(\hat{N}_{TOT}\) are the number of reports with the drug, the event, the drug and event, and in the whole database respectively.
A confidence interval is derived in Gravel (2009) using the delta method: $$\hat{s} = \sqrt{ 1/\hat{O} - 1/(\hat{N}_{drug}) + 1/(\hat{N}_{event} - \hat{O}) - 1/(\hat{N}_{TOT} - \hat{N}_{drug})}$$
and $$[\hat{CI}_{\alpha/2}, \hat{CI}_{1-\alpha/2}] = $$ $$[\frac{\hat{O}}{\hat{E}} \times \exp(Q_{\alpha/2} \times \hat{s}), \frac{\hat{O}}{\hat{E}} \times \exp(Q_{1-\alpha/2} \times \hat{s})]$$
where \(Q_{\alpha}\) denotes the quantile function of a standard Normal distribution at significance level \(\alpha\).
Note: For historical reasons, another version of this standard deviation is sometimes used where the last fraction under the square root is added rather than subtracted, with negligible practical implications in large databases. This function uses the version declared above, i.e. with subtraction.
References
Montastruc J, Sommet A, Bagheri H, Lapeyre-Mestre M (2011). “Benefits and strengths of the disproportionality analysis for identification of adverse drug reactions in a pharmacovigilance database.” British Journal of Clinical Pharmacology, 72(6), 905--908. doi:10.1111/j.1365-2125.2011.04037.x , https://doi.org/10.1111/j.1365-2125.2011.04037.x.
Gravel C (2009). “Statistical Methods for Signal Detection in Pharmacovigilance.” https://repository.library.carleton.ca/downloads/jd472x08w.
Examples
prr(
obs = 5,
n_drug = 10,
n_event_prr = 20,
n_tot_prr = 10000
)
#> # A tibble: 1 × 3
#> prr2.5 prr prr97.5
#> <dbl> <dbl> <dbl>
#> 1 117. 250 534.
# Note that input parameters can be vectors (of equal length, no recycling)
pvda::prr(
obs = c(5, 10),
n_drug = c(10, 20),
n_event_prr = c(15, 30),
n_tot_prr = c(10000, 10000)
)
#> # A tibble: 2 × 3
#> prr2.5 prr prr97.5
#> <dbl> <dbl> <dbl>
#> 1 150. 333. 742.
#> 2 94.7 167. 293.