As the arguments "individual_cv" and "mean_epg" are measures of the epidemiological context, these are often co-varied in simulations. However, "individual_cv" currently lives with the other parameters pertaining to the survey design. As in real life, the value of "individual_cv" is correlated with "mean_epg", it would be more convenient to be able to co-vary these two (i.e., without being forced to consider all possible combinations of the two, as is now the case).