Abstract Given the non-trivial cost of reproduction for males and substantial variation in female quality, males have been predicted to show mating bias as an evolved strategy. Using a large outbred population of Drosophila melanogaster, we test this prediction and show that males may adaptively bias their mating effort in response to the infection status of females. Given a simultaneous choice between females infected with pathogenic bacteria and sham infected females, males preferentially mated with the latter, who had a higher reproductive output compared to infected females. This may provide evidence for pre-copulatory male mate choice. Assessment of the reproductive behaviour ensured that the observed pattern of mating bias was not due to differences in receptivity between females infected with pathogenic bacteria and sham infected females. Further, there was no evidence for post-copulatory male mate choice measured in terms of copulation duration.