LAUSR

gold standard for assessing the effectiveness of treatments. If conducted properly, randomized controlled trials allow investigators to control for many types of bias. For ethical reasons, however, performing a randomized controlled trial to study the effectiveness of treatments in intoxications may not be possible. In addition, the small number of cases each year limits prospective studies, as illustrated by an open-label trial investigating the effectiveness of silibinin, which was terminated after 10 years and did not provide conclusive results [6]. Due to these limitations, most data on the treatment of intoxications comes from real-world data, such as case series and case reports. In our study [4], we performed a systematic review of Amanita phalloides intoxications published since 1975. We agree with the authors that our approach has a risk of bias and confounding due to the non-experimental design. During data analysis, a multivariable analysis was considered to control for this. In the end, however, this analysis was not possible due to the high percentage of missing data, as shown in Table 1 of our article. Propensity scoring, as suggested, would be an alternative to multivariable analyses to control for confounding in non-experimental studies [7]. In this method, cases and controls are matched to control for confounders and thus reduce bias. As for multivariable analysis, again, too many variables are missing to perform such an analysis properly. In our opinion, performing such an analysis would introduce new forms of bias. However, we agree with the authors that this is an important limitation of our study. Given that randomized controlled trials may not be possible, real-world data remains an important source of information on the effectiveness of treatments in intoxicated patients. To improve the availability, comparability and reusability of case report data, we suggest the use of a standard format to report these cases, which follows the CARE guidelines [8]. In this way, analyses such as those performed in our study can be improved, leading to a more accurate interpretation of treatment outcomes so that better guidelines for management are produced. Disclosure statement