Koch 1967a.

Methods	Stratified random allocation***; double‐blind (A); placebo‐controlled; 18% drop out after 3 years (study duration = 3 years + 2 years post‐intervention period). Natural losses; no differential group losses.
Participants	120 children analysed at 3 years (present for entire trial period). Age range at start: 11‐12 years (average = 11). Surfaces affected at start: 19.7 DFS. Background exposure to fluoride: none reported. Year study began: 1962. Location: Sweden.
Interventions	FT versus PL (NaF group = 1000 ppm F). School use/supervised, daily, 1g applied for 2 min (non‐fluoride toothpaste provided to all for home use). Abrasive system: methacrylate polymer (acrylic).
Outcomes	3yDFS increment ‐ cl(CA)(E). Reported at 1 and 3 years follow ups (and 2 years post‐treatment). DFT. O‐DFS. MD‐DFS. BL‐DFS. Annual CAR. Secondary caries.
Notes	Participants randomised (N = 146). Baseline characteristics (DFS, DFT, SAR, TAR) 'balanced'. Clinical (VT) caries assessment by one examiner; diagnostic threshold = CA; radiographic assessment (2 postBW) used as an aid but not reported; state of tooth eruption included = E. Intra‐examiner reproducibility checks for DFS in 10% sample (icc over 0.98); reversals very small in both groups and equally common. *** Allocation concealment considered adequate by consensus.
Risk of bias
Bias	Authors' judgement	Support for judgement
Allocation concealment (selection bias)	Low risk	A ‐ Adequate