Table 2.

Definition of the estimands for the primary analyses

Attribute	Definition
Treatments	The primary analyses are based on the following comparisons: (a) Arm 3 vs Arm 1—that is, 3HP+routine treatment support vs 3HR+routine treatment support (b) Arm 5 vs Arm 1—that is, 1HP+routine treatment support vs 3HR+routine treatment support (c) Arm 2 vs Arm 1—that is, 3HR+additional treatment support vs 3HR+routine treatment support If comparison (a) shows 3HP improves adherence compared with 3HR, then additional treatment support will be formally tested for 3HP by comparing Arm 4 vs Arm 3—that is, 3HP+additional treatment support vs 3HP+routine treatment support. Additional treatment support will be similarly assessed for 1HP.
Population	Adults aged 16–65 years diagnosed with LTBI and eligible for LTBI treatment.
Endpoint	Adequate treatment adherence, defined as taking ≥90% of allocated doses within the allowable time frame.
Intercurrent events	The main intercurrent events and how they will be handled in the estimand are as follows: Failure to collect all prescriptions—composite and treatment policy strategies lead to same estimated effect. Early treatment discontinuation for any reason including adverse event(s) and active TB: a treatment policy strategy will be used, that is, the participant is considered to have stopped treatment regardless of the occurrence of the intercurrent event.
Population-level summary measure	Risk ratio for adequate treatment adherence comparing the relevant arms.

LTBI, latent tuberculosis infection.