Awor et al. BMC Pregnancy and Childbirth (2023) 23:101 https://doi.org/10.1186/s12884-023-05420-z RESEARCH © The Author(s) 2023. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativeco mmons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Open Access BMC Pregnancy and Childbirth Prediction of pre-eclampsia at St. Mary’s hospital lacor, a low-resource setting in northern Uganda, a prospective cohort study Silvia Awor 1* , Benard Abola 2 , Rosemary Byanyima 3 , Christopher Garimoi Orach 4 , Paul Kiondo 5 , Dan Kabonge Kaye 5 , Jasper Ogwal‑Okeng 6 and Annettee Nakimuli 5 Abstract Background Pre‑eclampsia is the second leading cause of maternal death in Uganda. However, mothers report to the hospitals late due to health care challenges. Therefore, we developed and validated the prediction models for prenatal screening for pre‑eclampsia. Methods This was a prospective cohort study at St. Mary’s hospital lacor in Gulu city. We included 1,004 pregnant mothers screened at 16–24 weeks (using maternal history, physical examination, uterine artery Doppler indices, and blood tests), followed up, and delivered. We built models in RStudio. Because the incidence of pre‑eclampsia was low (4.3%), we generated synthetic balanced data using the ROSE (Random Over and under Sampling Examples) package in RStudio by over‑sampling pre‑eclampsia and under‑sampling non‑preeclampsia. As a result, we got 383 (48.8%) and 399 (51.2%) for pre‑eclampsia and non‑preeclampsia, respectively. Finally, we evaluated the actual model perfor‑ mance against the ROSE‑derived synthetic dataset using K‑fold cross‑validation in RStudio. Results Maternal history of pre‑eclampsia (adjusted odds ratio (aOR) = 32.75, 95% confdence intervals (CI) 6.59— 182.05, p = 0.000), serum alkaline phosphatase(ALP) < 98 IU/L (aOR = 7.14, 95% CI 1.76—24.45, p = 0.003), diastolic hypertension ≥ 90 mmHg (aOR = 4.90, 95% CI 1.15—18.01, p = 0.022), bilateral end diastolic notch (aOR = 4.54, 95% CI 1.65—12.20, p = 0.003) and body mass index of ≥ 26.56 kg/m 2 (aOR = 3.86, 95% CI 1.25—14.15, p = 0.027) were inde‑ pendent risk factors for pre‑eclampsia. Maternal age ≥ 35 years (aOR = 3.88, 95% CI 0.94—15.44, p = 0.056), nullipar‑ ity (aOR = 4.25, 95% CI 1.08—20.18, p = 0.051) and white blood cell count ≥ 11,000 (aOR = 8.43, 95% CI 0.92—70.62, p = 0.050) may be risk factors for pre‑eclampsia, and lymphocyte count of 800 – 4000 cells/microliter (aOR = 0.29, 95% CI 0.08—1.22, p = 0.074) may be protective against pre‑eclampsia. A combination of all the above variables predicted pre‑eclampsia with 77.0% accuracy, 80.4% sensitivity, 73.6% specifcity, and 84.9% area under the curve (AUC). Conclusion The predictors of pre‑eclampsia were maternal age ≥ 35 years, nulliparity, maternal history of pre‑ eclampsia, body mass index, diastolic pressure, white blood cell count, lymphocyte count, serum ALP and end‑ diastolic notch of the uterine arteries. This prediction model can predict pre‑eclampsia in prenatal clinics with 77% accuracy. Keywords Risk prediction, Uterine artery Doppler indices, Maternal history, Blood tests, Pre‑eclampsia, Uganda, Africa *Correspondence: Silvia Awor s.awor@gu.ac.ug Full list of author information is available at the end of the article