If the purpose of the analysis is to determine whether the preventive measure is effective and not overwrought, an excess deaths analysis is a good one because it helps infer what the toll would be in absence of the measure.
If the purpose is to attribute cause of death, it’s just a heuristic. But a useful one when other methodologies are (rightly or wrongly) suspect.
