What is it about?
Usually, estimating the effect of treatment on continuous outcomes - such as the duration of colds - is based on comparing group differences. However, it is possible or probable that the effect of treatment is greater on colds that are long, whereas it is smaller on colds that are short. This study introduces a method to estimate back-transformed quantile treatment effect (QTE), which presents QTEs as a function of the original outcome values in the control group. As an example, this analysis indicated that in the Mossad trial (1996), 4-day colds were shortened by 2 days, whereas 15-day colds were shortened by 8 days. Over all the range, colds were shortened by 43% so that the relative effect captured the effect usefully.
Featured Image
Photo by Brittany Colette on Unsplash
Why is it important?
The QTE approach can be particularly useful in the analysis of randomized trials and other controlled studies since it enables investigation of how the treatment affects the whole distribution of a continuous outcome of untreated patients. In medicine, often a relevant question is how large the effect is on patients with a particular duration of disease, and the impact of treatment on a disease of long duration is often much more relevant than slightly shortening an already short disease. In calculating the overall average, the latter may mask substantially larger effects on the former. By comparing the entire distributions of disease durations, the new approach captures more information about the effect of the treatment than a single average.
Perspectives
It is much more useful to state that properly composed zinc gluconate lozenges may shorten 4-day colds by 2 days and 15-day colds by 8 days; rather than all colds on average by 4 days. Furthermore, when only one average estimate is used, our study indicates that the 43% effect is more useful information than the 4-day effect.
Dr Harri Hemila
Helsingin Yliopisto
Read the Original
This page is a summary of: Estimating quantile treatment effect on the original scale of the outcome variable: a case study of common cold treatments, Trials, November 2025, Springer Science + Business Media,
DOI: 10.1186/s13063-025-09265-z.
You can read the full text:
Contributors
The following have contributed to this page
