Background: Although there are available studies that determine the risk factors for preterm birth, they presently do not allow the prediction of risk in individual patients in daily practice. Thus, developing and validating the risk prediction score for preterm birth guide caregivers in promptly providing the treatment choice for individual patients and be more cost-effective by identifying high- risk patients who will benefit most from certain interventions. Objective: To develop and validate a risk score for the prediction of preterm birth using maternal characteristics. Method: A retrospective follow-up study was conducted on March (1- 30) 2021 at Felege Hiwot comprehensive specialized hospital. The sample size was determined by assuming 10 events per predictor, based on this assumption total sample size was1308. Data were collected using a semi- structured checklist through chart review. Data were coded and entered into Epidata, version 3.02, and was analyzed by using R statistical programming language version 4.0.4 for further processing and analysis. Bivariable logistic regression was done to identify the relationship between each predictor and preterm birth. Variables with (p < 0.25) from the bivariable analysis were entered into a backward stepwise multivariable logistic regression model, and significant variables (p < 0.05) were retained in the multivariable model. Model accuracy and goodness of fit were assessed by computing the area under the ROC curve (discrimination) and calibration plot (calibration) respectively. Result: The incidence of preterm birth was 13.4%. Residence, gravidity, hemoglobin < 11 mg/dl, early rupture of membranes, antepartum hemorrhage, and pregnancy-induced hypertension were predictors of preterm birth. These predictors were all included in the model; the AUC was 0.786 with a sensitivity of 75.14 % and speci¦city of 67.46%. At the threshold scores of 3. The model calibration test had a p-value of 0.492. Conclusion and recommendation: This study showed the possibility of predicting preterm birth using maternal characteristics during pregnancy. Thus, using this model could help to identify pregnant women at a higher risk of having a preterm birth to be linked to a center that has the necessary facilities for corticosteroid administration, antibiotic treatment in the event of infection, and other services. Key words, Prediction Model, Preterm birth, Ethiopia