You are proposing an experiment that would be almost impossible to run. Little if any nutrition research is actually run like a traditional science experiment

You are proposing an experiment that would be almost impossible to run. Little if any nutrition research is actually run like a traditional science experiment (i.e. do something specific and observe results). Unfortunately outside of the prison camps on North Korea it is impossible to control the diet (or behaviors) of enough people for long enough to get useful results. Instead we observe people uncontrolled behavior (almost never directly but through questionnaires) then observe health outcomes (usually not directly) or in this case something (telomere length) believed to be associated with health outcomes. Attempts are then made to adjust for all these potential problems using additional data (usually questionnaires) - usually difficult to say how effective. These neccesary compromises introduces a whole host of potential problems that have contributed to all the confusion over what is good/bad nutrition. I've given up on having much confidence in medical research until it is repeatedly confirmed in additional studies - which may never happen.