I'd offer case-mix as a simple explanation of the findings. In other words, when lots of cardiologists in a hospital leave for a conference, elective procedure

I'd offer case-mix as a simple explanation of the findings. In other words, when lots of cardiologists in a hospital leave for a conference, elective procedures are deferred until their return... which means that emergency cases constitute a larger proportion of the reduced number of cases that do come into the hospital during the conference. It is completely reasonable to believe that emergency cases have a higher mortality rate than elective cases. This would raise the mortality rate, but not the mortality count. The same effect would probably be seen over long holidays, when people tend not to schedule elective cases.