BACKGROUND: Four types of malaria vaccine, SPf66 and MSP/RESA vaccines (against the asexual stages of the Plasmodium parasite) and CS-NANP and RTS,S vaccines (against the sporozoite stages), have been tested in randomized controlled trials in endemic areas. OBJECTIVES: To assess malaria vaccines against Plasmodium falciparum, P. vivax, P. malariae and P ovale in preventing infection, disease and death. SEARCH STRATEGY: We searched the Cochrane Infectious Diseases Group Specialized Register (April 2004), CENTRAL (The Cochrane Library Issue 2, 2004), MEDLINE (1966 to April 2004), EMBASE (1980 to April 2004), Science Citation Index (1981 to April 2004), and reference lists of articles. We also contacted organizations and researchers in the field. SELECTION CRITERIA: Randomized controlled trials comparing vaccines against Plasmodium falciparum, P. vivax, P. malariae or P. ovale with placebo or routine antimalarial control measures in people of any age receiving a challenge malaria infection. DATA COLLECTION AND ANALYSIS: Two reviewers independently assessed trial quality and extracted data. MAIN RESULTS: Eighteen efficacy trials involving 10,971 participants were included. There were ten trials of SPf66 vaccine, four trials of CS-NANP vaccines, two trials of RTS,S vaccine, and two of MSP/RESA vaccine. Results with SPf66 in reducing new malaria infections (P. falciparum) were heterogeneous: it was not effective in four African trials (Peto odds ratio (OR) 0.96, 95% confidence interval (CI) 0.81 to 1.14), but in five trials outside Africa the number of first attacks was reduced (Peto OR 0.77, 95% CI 0.67 to 0.88). Trials to date have not indicated any serious adverse events with SPf66 vaccine. In three trials of CS-NANP vaccines, there was no evidence for protection by these vaccines against P. falciparum malaria (Peto OR 1.12, 95% CI 0.64 to 1.93). In a small trial in non-immune adults in the USA, RTS,S gave strong protection against experimental infection with P. falciparum. In a trial in an endemic area of the Gambia in semi-immune people, there was a reduction in clinical malaria episodes in the second year of follow up, corresponding to a vaccine efficacy of 66% (CI 14% to 85%). In a trial in Papua New Guinea, MSP/RESA had no protective effect against episodes of clinical malaria. There was evidence of an effect on parasite density, but this differed according to whether the participants had been pretreated with sulfadoxine/pyrimethamine or not. The prevalence of infections with the parasite subtype of MSP2 in the vaccine was reduced compared with the other subtype (Peto OR 0.35, CI 0.23 to 0.53). AUTHORS' CONCLUSIONS: There is no evidence for protection by SPf66 vaccines against P. falciparum in Africa. There is a modest reduction in attacks of P. falciparum malaria following vaccination with SPf66 in other regions. Further research with SPf66 vaccines in South America or with new formulations of SPf66 may be justified. There was not enough evidence to evaluate the use of CS-NANP vaccines. The RTS,S vaccine showed promising result, as did the MSP/RESA vaccine, but it should include the other main allelic form of MSP2. The MSP/RESA trial demonstrated that chemotherapy during a vaccine trial may reduce vaccine efficacy, and trials should consider very carefully whether this practice is justified.