Nosten 1996.

Methods	Randomized controlled trial Generation of allocation sequence: valid complex scheme requiring balance of pre‐specified characteristics; randomization within households, then evaluated and repeated 3 times to achieve age/sex balance Allocation concealment: not used Blinding: double blind Inclusion of all randomized participants: 1221/1348 (90.6%) participants received all 3 doses Length of follow up: 15 months after third dose
Participants	1348 children of Karen ethnicity Inclusion criteria: age 2 to 15 years Exclusion criteria: serious underlying illness; pregnancy; splenectomy; epilepsy; anaemia (packed cell volume (PCV) < 25%); uncontrolled asthma; thalassemia with anaemia; any other condition deemed to increase risk
Interventions	1. 3 doses of SPf66 vaccine (MPS, San Diego, CA) adsorbed onto aluminium hydroxide (2 mg in 0.5 mL for children ≥ 5 years; 1 mg in 0.25 mL for children < 5 years) subcutaneously on days 0, 30, and about 180 2. Engerix‐B recombinant hepatitis B vaccine subcutaneously on same schedule Those with parasites at cross‐sectional surveys before first and third vaccine doses were treated with artesunate/mefloquine; children sick and parasitaemic on day of vaccination were also treated and not vaccinated until free of symptoms
Outcomes	1. First episode of symptomatic Plasmodium falciparum malaria (case definition: slide positive ‐ any density ‐ for P. falciparum (alone or mixed with Plasmodium vivax) together with at least 1 of the following: oral temperature > 38 ºC or history of fever within 3 days; chills; headache; myalgia; arthralgia; nausea) 2. Severe malaria (case definition: World Health Organization (WHO) definition or parasitaemia > 4%) 3. Symptomless malaria, detected at cross‐sectional surveys (case definition: positive slide with no symptoms in preceding 4 days or subsequent 14 days 4. Time to first episode of symptomatic P. falciparum infection 5. SPf66 seroconversion
Notes	Location: Shoklo refugee camp, in North‐West Thailand, where malaria transmission is low, unstable, seasonal, and highest in May to September and November to January Date: 1993 to 1995 Method of surveillance: active follow up: case detection (daily surveillance and cross‐sectional surveys every 2 to 3 months); passive follow up by case detection