Adjusting risk factors in spontaneous abortion by multiple logistic regression.

A cross-sectional case-control study was performed to identify some obstetric and gynaecologic factors that can influence spontaneous abortion. Statistical and epidemiologic analyses were done by multiple logistic regression to adjust OR through the beta coefficient. A dicotomized outcome variable, representing spontaneous abortion, and different independent variables, representing distinct medical factors, were designed. The analysis was carried out with a personal computer and an appropriate statistic package. The variables representing age over 35 and previous spontaneous abortions were shown to be risk factors, adjusted for the rest of variables. The variables representing parity and late menarcheal age lost significance when they were adjusted with multiple logistic regression.