The Omega Study's high participation and retention rates reduced the likelihood that selection bias and missing data influenced our estimates of association.
For example, Molitor et al. (2006) documented the effect of missing exposure data at the individual level and determined that the methodology chosen to account for this missing data influences the conclusions regarding the observed health effects.
The presence of missing data could influence the interpretation of the results.
Our previous report on the changes of muscle strength in the same population suggested that differential follow-up rate or nonrandom missing data might influence the results biased to the null (6).
However, it is not likely that the missing data have influenced the major conclusions.
Analyses done to account for missing data did not influence the study findings or conclusions.
Many clinical data sets contain substantial amounts of missing data, but the influence of missing data on the inclusion frequency of prognostic variables used to select a model is hardly recognized [ 37].
Although analyses were conducted using pairwise complete data, results from analyses repeated on a subset of the study population with complete data indicate that missing data did not influence our results.
The way we handled missing data might have influenced the estimates or inflated the effects for variables with only few data points and long periods of interpolation (Supplementary Appendix pp. 16 18).
In our population, we did not observe any association with the explored parameters; however, the relevant proportion of missing data could have influenced such finding masking the real effects in terms of increased/reduced hazard of adverse events occurrence.
