For me, both epidemio and gene flow were the place i could potential lose a mark. the question for epidemiology was a two parter so both design and the justify component. Also idk if this is a mark but i mentioned it was a cohort study - depends if hsc markers mark for this but adding this would have elevated the response. for the design component going over all the steps u would do - so i did a numbered format. for example, like gathering the people from each age group and doing a good cross-section of society. then i just did like 9 steps in total going from gathering people to collecting data to presenting data. for the justify part is when the reliability, accuracy and validity stuff came in. Personally, i regret not doing this but i didnt split my justification into these criteria clearly. I just wrote like 2 pages of justification without a clesr structure. But i went over like general evaluative stuff like time of study, number of people, regular checkups, systematic bias, random bias. I also mentioned that despite the data not showing trends specific to each region A,B,C this would be data collected for internal use. In regards to ur controls - as long as you mentioned like how ur design effectly eliminates eroneous variables you should be good. But since ur jsyt justifying i dont think they will mark you off if you dont name drop reliability, validity and accuracy. Also independt and dependent shouldnt be an issue in epidemiology studies and the question basically specified them anyway so should be chill - i didnt specifcy them as well in my response.