Background Applying the Australasian Triage Scale to pregnant women presenting to emergency departments (EDs) is difficult as the descriptors may not reflect the urgency of the obstetric condition. This study aimed to examine whether condition-specific algorithms and triage education improved triage assessment and documentation of pregnant women presenting to the ED. Method Algorithms with a decision aid for triage with minimum agreed descriptors were developed to triage two pregnancy conditions (pre-eclampsia and antepartum haemorrhage). Triage documentation was then audited before (n=50) and after (n=50) a triage education programme which introduced algorithms for both conditions. Significant differences were examined using χ2 test with significance set at p<0.05. Results The quality of documentation of specific clinically significant symptoms of pre-eclampsia improved considerably, including the presence of headache from 58% pre-education to 80% posteducation (p=0.002), visual disturbances from 58% to 90% (p<0.001), epigastric pain from 24% to 80% (p=0.002) and the presence of fetal movements from 62% to 90% (p=0.001). Documentation of descriptors for vaginal bleeding >20 weeks gestation improved for estimation of blood loss from 54% to 86% (p<0.001), patient 'appearance' from 32% to 62% (p=0.003) and, importantly, descriptions of patient's own assessment of their well-being from 8% to 28% (p=0.009). Conclusion The introduction of triage education and condition-specific decision aids for triage markedly improved triage assessment and documentation. The application of algorithms may reduce clinical risk resulting from suboptimal triage of pregnant women presenting to EDs.