Abstract Objective: To investigate the risk factors for pulmonary infection during chemotherapy in patients with small cell lung cancer (SCLC) and to develop and validate a risk prediction model. Methods: Patients with SCLC who received chemotherapy at Jiangsu Province Hospital from April 2020 to March 2022 were retrospectively enrolled as the training cohort (n = 251), and a prospective cohort of SCLC patients was consecutively included as the validation cohort (n = 112). According to follow-up outcomes, patients were divided into pulmonary infection and non-infection groups. In the training cohort, univariate and multivariate logistic regression analyses were performed to identify independent risk factors and to construct a scoring model. The discriminative ability of the model was evaluated using the receiver operating characteristic (ROC) curve, and calibration was assessed using calibration curves. The model was compared with previously reported nomogram models and further validated in the validation cohort. Results: Multivariate logistic regression analysis showed that smoking history, pleural effusion, hoarseness, single-agent chemotherapy, and albumin <35 g/L after the first cycle of chemotherapy were independent risk factors for pulmonary infection in patients with SCLC (all P < 0.05). Based on these five variables, the SCLC-PIR scoring model was established, assigning 1 point for smoking history, 2 points for pleural effusion, 2 points for hoarseness, 1 point for single-agent chemotherapy, and 2 points for post-chemotherapy albumin <35 g/L. In the training cohort, the area under the ROC curve (AUC) was 0.870 (95% CI: 0.818~0.922), with an optimal cutoff value of 5 points, yielding a sensitivity of 71.7% and a specificity of 89.4%. Internal validation using bootstrap resampling (1,000 iterations) demonstrated good agreement between predicted and observed risks. The AUCs of two previously reported nomogram models were 0.607 and 0.637, respectively, both lower than that of the present model. In the validation cohort, the predictive performance of the SCLC-PIR model remained stable, with an AUC of 0.896 (95% CI: 0.832~0.961). Conclusion: The SCLC-PIR scoring model, based on smoking history, pleural effusion, hoarseness, single-agent chemotherapy, and post-chemotherapy albumin <35 g/L, shows good discrimination and calibration in predicting pulmonary infection during chemotherapy in patients with SCLC, and may be useful for early identification of high-risk patients.