I'm going to put aside the fancy methodology of CausalImpact and Bayesian time-series models and suggest that you have a fundamental question of generalizability of your findings. What is the goal of this project? Are you trying to learn about how a population of people (all males?) respond to the intervention? Or, are you trying to learn how 1 specific male responded to 1 specific situation?

It seems reasonable that your 1 participant experienced an intervention that elicited some kind of change in their HR and HRV measures. You might consider if there are other explanations (diet, sleep, exercise, etc) and you might ultimately feel confident in saying that a likely cause of change in HR and HRV was the change in living conditions.

But, you only have 1 participant worth of data. Do you feel confident in generalizing to a population of all males based on the data of 1 male? Probably not. You just have no way of knowing if the change in living condition has an impact beyond your 1 subject until you measure more subjects.