BACKGROUND: In Latin America, motels rent rooms for commercial and non-commercial sex. We investigated the impact of providing health-education material and condoms on condom use in Managua, Nicaragua. METHODS: In a randomised controlled trial, in 19 motels, we gave condoms on request, made them available in rooms, or gave condoms directly to couples, with and without the presence of health-education material in the rooms. In a factorial design we assessed condom use directly by searching the rooms after couples had left. FINDINGS: 11 motels were used mainly by sex workers and their clients and eight mainly for non-commercial sex. 6463 couples attended the motels in 24 days. On 3106 (48.0%) occasions, at least one used condom was retrieved. Condom use was more frequent for commercial sex than for non-commercial sex (60.5 vs 20.2%). The presence of health-education material lowered the frequency of condom use for commercial sex (odds ratio 0.89 [95% CI 0.84-0.94]) and had no effect on use for non-commercial sex (1.03 [0.97-1.08]). Condom use increased for commercial (1.31 [1.09-1.75]) and non-commercial sex (1.81 (1.14-2.81) if condoms were available in rooms. Directly handing condoms to couples was similarly effective for commercial sex but less effective for non-commercial sex (1.32 [1.03-1.61] vs 1.52 [1.01-2.38]). INTERPRETATION: In Latin America, motels are key locations for promoting the use of condoms. Making condoms available in rooms is the most effective strategy to increase condom use, whereas use of health-education material was ineffective. These findings have important implications for HIV-prevention policies. [Authors]