It seems plausible that in a crash DAG, speeding is downstream of poor time management/organizational skills (e.g., leaving late, detouring for gas, returning for forgotten items, stress-induced errors like missing an exit). If so, behavioral interventions that improve planning and reduce time pressure could lower speeding and, in turn, crash risk.
We’re building a DAG for population-level drivers of STIs, but it’s getting very large because of behavioral, psychological, and social layers involved. In classical epidemiology, we use DAGs to identify the minimum necessary adjustment set to control confounding without opening backdoor paths.
I’m wondering whether a similar idea exists for interventions: is there a minimum necessary intervention set that maximizes prevention impact and efficiency (best balance of investment and return)? That would give us a more practical guide on designing comprehensive interventions. Any suggestion?
It seems plausible that in a crash DAG, speeding is downstream of poor time management/organizational skills (e.g., leaving late, detouring for gas, returning for forgotten items, stress-induced errors like missing an exit). If so, behavioral interventions that improve planning and reduce time pressure could lower speeding and, in turn, crash risk.
We’re building a DAG for population-level drivers of STIs, but it’s getting very large because of behavioral, psychological, and social layers involved. In classical epidemiology, we use DAGs to identify the minimum necessary adjustment set to control confounding without opening backdoor paths.
I’m wondering whether a similar idea exists for interventions: is there a minimum necessary intervention set that maximizes prevention impact and efficiency (best balance of investment and return)? That would give us a more practical guide on designing comprehensive interventions. Any suggestion?