These differences indicate that the remaining severity classification discrepancies between the VSS and the CSS may be due, not only to the severity threshold chosen, but also to the differences in individual item scoring. In order to obtain equivalent severity cutoffs between the two scoring systems, item cutoffs should be reconsidered. While FG-4592 solubility dmso better consistency between severity score cutoffs could be achieved, due to the differences in items included in each scoring system and because the

CSS is affected more by missing a symptom than the VSS (i.e. CSS does not provide a point score for the number of diarrhea episodes until two episodes have occurred and for the duration of vomiting until 2 days of vomiting have passed), it is unlikely that

the severity scores would ever identify the exact same proportions of Proteases inhibitor severe disease in any population. Weaknesses of this post-hoc analysis included that the trials were designed to capture moderate to severe cases and, as explained in the main efficacy manuscript for Africa [8], despite common case capture methods, success in capturing cases differed between sites and regions. The challenges in capturing and scoring cases for the Mali site are described in this supplement [28]. Despite this, scoring distributions for the VSS and the CSS appeared normal in each region. Additionally, diary cards were not used to collect symptoms at home in these trials and, depending on healthcare seeking behaviors, the average time from symptom onset to clinic assessment varied by participant and site, thus leaving

some sites more dependent on parental recall than others and allowing episode severity to develop further before seeking treatment at a healthcare facility. Larger discrepancies were identified between the two scoring systems in Asia as compared to Africa; the scoring systems, originally developed for use in middle- to high-income countries, did not perform similarly others across low-income regions. For the CSS, this may be due to differences between regions in interpretation and understanding of subjective items, like behavior and temperature duration. For the VSS, this may be due to differences in rehydration and hospitalization patterns between regions. It was also observed that, based on the number of participants enrolled at each site, some sites captured an increased number of cases as compared to other sites which may have been due to differences in medical facility utilization by site, indicating a challenge of running any multi-center trial and trying to ensure that case capture methods are identical, regardless of cultural differences in health care seeking behaviors.