CHAPTER 3.1 100 models or that sex disaggregation of results should be the norm in cardiovascular research.48 Study limitations The strength of this study lies in large sample sizes and access to individual patient data from both trial and observational datasets. There are also potential limitations of the study. The harmonized criteria selected to define the RCT-eligible population were chosen based on data-availability and commonality between trials. Therefore, the percentage of patients eligible for trial inclusion is likely overestimated but allows for fairer comparisons between the RCT population and real-world. The RCTs involved in this study were selected based on the availability of data from industry partners. However, a comparison of baseline characteristics with contemporary trials does not show meaningful differences (Supplementary table 4). Combining RCTs can always present a source of heterogeneity in participant characteristics due to different investigational drugs being studied, trial phases and study countries. However, our sensitivity analyses in the CAD and SR subgroups support that exclusion criteria differences between the RCTs do not affect the main conclusion. Pooling of the placebo and treatment arms does not allow extrapolation of the mortality rates and risk from this study; however, pooling does not explain the sex differences seen in these results which was the main research question and conclusion of these results. Lastly, registries are regarded a fair representation of real-world patients with considerable depth of clinical data, although there can be some differences in characteristics and treatment practices between patients who were and were not enrolled in the registries. We also acknowledge that the trial and real-world populations differed on geographical location, healthcare systems and time of data collection.
RkJQdWJsaXNoZXIy MjY0ODMw