ontreatment

On-treatment, often written as on-treatment or "on therapy," refers to the period during which a patient is actively receiving a specified medical intervention, such as a drug or procedure. It begins at the start of treatment and ends at discontinuation, a major dose change, a therapeutic switch, or an interruption beyond a predefined gap. The term is used to distinguish periods of active exposure from times when the patient is not receiving the treatment.

In clinical trials, on-treatment analyses focus on outcomes that occur while participants are adherent to the assigned therapy. Analyses may censor follow-up at the point of discontinuation or non-adherence, or when a treatment change occurs. This approach is often contrasted with intention-to-treat analyses, which include all randomized participants regardless of subsequent adherence, preserving randomization but potentially diluting the estimated treatment effect.

In observational studies, defining the exposure window as on-treatment requires careful measurement of adherence, dosing, and treatment gaps. Researchers may use time-on-treatment or time-varying covariates to reflect changing exposure. Censoring and handling of missing data are important considerations, especially if discontinuation is related to the risk of the outcome being studied.

Strengths of on-treatment analysis include the ability to estimate effects under actual exposure and to detect safety signals that arise during therapy. Limitations include potential bias from nonrandom discontinuation, reduced generalizability to all users, and the possibility that outcomes are influenced by factors associated with ongoing treatment rather than the treatment itself.

See also: intention-to-treat, per-protocol analysis, adherence, persistence, time-on-treatment.