How Oregon and Colorado psilocybin safety estimates change when studies measure harms beyond the dosing session

State-regulated psilocybin services in Oregon and Colorado appear to have few serious events during or immediately after dosing, but reported safety changes substantially with the measurement method. Oregon administrative data and facilitator records captured mostly mild, transient reactions and few serious events, while a 346-client cohort that surveyed clients for three months found emergency care, worsening depression and new thoughts of death or suicide that facilitator reports missed. Most participants improved on depression measures, but a small minority deteriorated. The datasets cannot be compared directly because they use different definitions, follow-up periods and populations, and none has a control group. The evidence therefore does not establish that services are safe or unsafe, or that psilocybin caused later harms. It shows that incident systems ending around the session can undercount problems emerging afterward, supporting longer-term, client-reported safety monitoring.