The study investigated the factors influencing the therapeutic efficacy of nonlactating mastitis (NLM) based on hematological inflammatory markers and created a predictive model. They retrospectively analyzed 264 cases of patients with NLM from two Chinese hospitals since January 2022, divided into a model group (n=185) and a validation group (n=79) in a ratio of 7:3. In the model group, patients were further divided into a group with good efficacy (n=133) and poor efficacy (n=52). Binary logistic regression analysis showed that systemic inflammation index (SII) and C-reactive protein (CRP) significantly influence therapeutic efficacy (p < 0.05). The predictive model has the formula Logit(P) = -74.457 + (0.823X1) + (0.589X2). The area under the ROC curve (AUC) was 0.94 (95% CI: 0.823–0.966) in the model group and 0.85 (95% CI: 0.794–0.893) in the validation group, which confirms good predictive ability and clinical applicability. The conclusion of the study states that SII and CRP are the key factors of NLM effectiveness and the model based on them was successfully established.