Treatment with 1 G azithromycin was observed prospectively in 130 women with cervicitis (>30 polymorphonuclear leucocytes/high-powered field) enrolled in a cervicitis aetiology study of 558 women at three sexually transmitted infection clinics in Sydney, Australia. Two overlapping groups of women with cervicitis were considered: ‘cervicitis group 1’ (n = 116) excluded women with Trichomonas vaginalis and a subgroup of this, ‘cervicitis group 2’ (non-specific cervicitis) (n = 96) further excluded women with Neisseria gonorrhoea, Chlamydia trachomatis and Mycoplasma genitalium at enrolment. Testing for Chlamydia trachomatis, Mycoplasma genitalium and Trichomonas vaginalis was by PCR and Neisseria gonorrhoea by PCR and culture. Treatment outcomes were cervicitis or vaginal symptoms at follow-up. Effect on cervicitis at follow-up was also assessed after additional reported partner treatment. In ‘cervicitis group 1’ where prevalence of Mycoplasma genitalium and/or Chlamydia trachomatis was 23/116 (19.8%), azithromycin reduced cervicitis at follow-up (RR = 0.62 (95% CI 0.39–0.97) p = 0.035), but there was no significant effect in non-specific cervicitis (‘cervicitis group 2’) (RR = 0.60 (95% CI 0.35–1.01) p = 0.056). Empiric treatment did not reduce vaginal symptoms at follow-up in either group. No effect of empiric partner treatment was seen. The conclusion was that empiric azithromycin treatment of cervicitis reduces cervicitis at follow-up in populations with high prevalence of Chlamydia trachomatis and/or Mycoplasma genitalium. There are no benefits of empiric azithromycin for non-specific cervicitis or empiric partner treatment.