Article type
Abstract
Background
The COVID-19 pandemic demanded real-time evidence to inform decision-making. We conducted an LSR to evaluate the safety and effectiveness of COVID-19 vaccines administered to pregnant people. The great amount of evidence, the number of outcomes, and the subgroups of interest allow a large number of meta-analyses. Therefore, it is useful to have an interactive tool that allows tailored meta-analyses by selecting filters according and subgroups for each outcome.
Objectives
To present the evidence map and the tool developed for interactive real-time meta-analyses using the shiny R library and developmental challenges
Methods
We describe the online evidence map and the main features of the tool developed for interactive real-time meta-analyses using the shiny R library.
Results
The evidence map and the meta-analysis tool are available at https://www.safeinpregnancy.org/living-systematic-review/. The evidence map is automatically generated from a Redcaps database (Figure 1a). From the values selected in the menu, dynamic texts are generated with interpretations, graphs, and tables that summarize the information. The greatest difficulties of programming in R are the long list of filters to perform the meta-analysis (random-effects model) and the need for conditional panels and monthly input updating. The filters available for comparing studies include type of outcome; outcome; subgroup; type of vaccine; schema received; pregnancy trimester; dominant variant; and effect measure analyzed. Once the values of each filter/variable are chosen (Figure 1b), the outputs for the selected outcome are
• number of studies (plus links to studies) reporting adjusted measures and number studies that were included in the meta-analysis;
• countries of residence of patients;
• forest plot using the R meta package with the following information: oby study (country, number of patients, first author, effect measure [95% CI], weight, quality of the study) and oby subgroup (combined effect [95% CI] and I2);
• text with the summary of filters chosen by the user; and
• summary table with information on all studies in the meta-analysis.
Conclusions
The presented interactive tool is useful for health decision-makers because it allows them to obtain relevant and specific evidence according to their specific needs of information regarding the effects of COVID-19 vaccines during pregnancy.
The COVID-19 pandemic demanded real-time evidence to inform decision-making. We conducted an LSR to evaluate the safety and effectiveness of COVID-19 vaccines administered to pregnant people. The great amount of evidence, the number of outcomes, and the subgroups of interest allow a large number of meta-analyses. Therefore, it is useful to have an interactive tool that allows tailored meta-analyses by selecting filters according and subgroups for each outcome.
Objectives
To present the evidence map and the tool developed for interactive real-time meta-analyses using the shiny R library and developmental challenges
Methods
We describe the online evidence map and the main features of the tool developed for interactive real-time meta-analyses using the shiny R library.
Results
The evidence map and the meta-analysis tool are available at https://www.safeinpregnancy.org/living-systematic-review/. The evidence map is automatically generated from a Redcaps database (Figure 1a). From the values selected in the menu, dynamic texts are generated with interpretations, graphs, and tables that summarize the information. The greatest difficulties of programming in R are the long list of filters to perform the meta-analysis (random-effects model) and the need for conditional panels and monthly input updating. The filters available for comparing studies include type of outcome; outcome; subgroup; type of vaccine; schema received; pregnancy trimester; dominant variant; and effect measure analyzed. Once the values of each filter/variable are chosen (Figure 1b), the outputs for the selected outcome are
• number of studies (plus links to studies) reporting adjusted measures and number studies that were included in the meta-analysis;
• countries of residence of patients;
• forest plot using the R meta package with the following information: oby study (country, number of patients, first author, effect measure [95% CI], weight, quality of the study) and oby subgroup (combined effect [95% CI] and I2);
• text with the summary of filters chosen by the user; and
• summary table with information on all studies in the meta-analysis.
Conclusions
The presented interactive tool is useful for health decision-makers because it allows them to obtain relevant and specific evidence according to their specific needs of information regarding the effects of COVID-19 vaccines during pregnancy.