The existing epidemiological research on long COVID has suffered from overly broad case definitions and a striking absence of control groups, which have led to distortion of risk. The unintended ...