RESUMO
INTRODUCTION: Surgical site infections (SSIs) remain the major cause of morbidity and mortality in the postoperative period and are important surgical and hospital quality indicators. In this context, our study aims to identify SSIs associated risk factors and to develop a predictive model. METHODOLOGY: 2521 patients who underwent surgery, between June 2018 and May 2019, in four surgery departments, at the Taza Provincial Hospital (Morocco) were diagnosed for SSI according to the standards of the Center for Disease Control and Prevention. The SSIs' risk factors were assessed by univariate statistical analysis and logistic regression using the Scikit Learn function of Python. RESULTS: The average age of the studied population was 35 ± 1 years. The overall SSI incidence was 6.3% (17.95%, 6.86%, 6.67% and 3.16% respectively in child, female, male and gynaecological-obstetrical surgeries. The univariate statistical analysis has shown a highly significant (p < 0.001) and a very significant (p < 0.01) relationship between SSIs and almost all risk factors; and the logistic regression model has revealed a strong association between SSI and people who have had previous surgery, urinary catheter, antibiotic use duration, co-morbidity, American Society of Anesthesiologists (ASA) score, duration of intervention, emergency preoperative and postoperative durations, service, specialty and age range. The prediction score exceeds 96% which justifies our model's quality. CONCLUSIONS: SSIs are generally frequent among postoperative patients. Therefore, pre-operative preparation, post-operative surveillance and the environment quality of the wards are necessary to reduce SSI rates in the hospital.