The development and validation of an obstetric triage acuity index: a mixed-method study

Background:There is uncertainty about the applicability of developed country triage scales in the developing population. Objective:To develop and validate an obstetric triage scale in Iran. Methods:This exploratory sequential mixed-method study was conducted on experienced gynecologists, emergency medicine, midwives, and nurses as evaluators, and pregnant or parturient women as sample. Development of obstetric triage acuity index was done using qualitative study and literature review. Thirty-two evaluators in two rounds evaluated the clinical impact score, content validity ratio, and content validity index. The tool performance was compared with a standard reference in 415 women. The association of urgency levels with hospitalization, resource utilization, and emergency department length of stay were assessed in 599 women. Five midwives independently double-triaged 234 convenience women to estimate inter-rater reliability. Results:Out of 191 items in the initial draft, 9 (4.7%) items in the face validity, 33 (18.1%) in the first, and 5 (3.2%) in the second content validity round were removed because not meeting the expected thresholds. Nine items added based on evaluators' suggestions. Finally, total items had a kappa above 0.74, and S-CVI/Ave, S-CVI/UA were 0.99 and 0.86, respectively. There was a significant correlation between the two tools (r = 0.661,p < 0). Level 1 mothers had a higher chance of using >= 2 resources (OR = 16.4, 95% CI = 2.06-129.98) and hospitalization (OR = 18.0, 95% CI = 3.96-81.51) compared to the level 5, after adjusting for gestational age/postpartum, and referral method. Sensitivity, specificity, and accuracy of the scale in predicting hospitalization were 74.8% (95% CI = 69.73-79.49), 80.8% (95% CI = 73.60-86.74), and 76.7% (72.67-80.48), respectively. The over and under-triage rates were 6.1% and 17.1%, respectively. The absolute agreement inter-raters was 86.8%, and the linear weighted kappa was 0.86 (95% CI = 0.81-0.91). Conclusion:The designed OB triage scale showed good validity and reliability in the tertiary teaching hospital. Studying this tool at lower levels, non-educational, and public hospitals are recommended for external validity testing.