Objective: To identify factors that predict condom use and sexual practices of brewery employees who had recently been on short- and longterm migration in the course of their jobs in Nigeria. Design: A cross-sectional survey design was adopted to investigate the relationship between sexual practices and condom use of migrant employees, and to determine the predictors of condom use at last intercourse during migration. Setting: The study was conducted in a large multinational brewery in Nigeria with operational locations in various parts of the country. Method: A sample of 936 male employees was selected consecutively in the brewery in 2002. They were independently invited to respond to a questionnaire consisting of items that covered background characteristics, condom use and sexual experiences. Results: The findings showed that 40 per cent of participants reported they were using alcohol occasionally prior to sex and 54 per cent reported that they either received or paid money for sex. Forty-seven per cent of men indicated that they used condoms at their last intercourse during migration, while 62 per cent had had one to three sexual partners during their last migration. The odds of using a condom at last intercourse were significantly lower for married than for single participants (odds ratio, 0.57). Compared with participants without the intention to use a condom in next intercourse during migration, those with the intention were significantly more likely to report condom use (odds ratio, 9.7). The odds of condom use were significantly higher for participants who perceived that a condom prevents HIV transmission than those who either did not have the perception or were not sure (odds ratio, 0.30-1.80). Conclusion: Steps are needed to promote condom use among brewery employees on migration. Intervention programmes aimed to enhance condom use among prospective migrant employees may be more effective if they provide opportunities for the employees to think about the advantages of condoms in sexual encounters under migration conditions. (Contains 5 tables.)