Objective To identify high-risk factors for healthcare-associated infection (HAI) in patients in intensive care units (ICUs), and develop a quick response (QR) code-based APP prediction tool. Methods Information of inpatients in general ICUs of three hospitals in Guizhou Province from January to December 2024 were collected. Risk factors were analyzed with a logistic regression model. QR code-based APP was constructed and validated. Results A total of 1 782 patients in general ICUs of three hospitals in Guizhou Province in 2024 were included in the analysis, out of which 410 were HAI cases, and the incidence of HAI was 23.01%. Multivariate logistic regre-ssion analysis results of HAI in ICU inpatients showed that regional gross domestic product (GDP) ≥58 685 Yuan, performing pathogen culture during this hospitalization, history of diabetes mellitus, history of cancer, length of hospital stay ≥7 days before infection, and duration of persistent fever ＞5 days before infection were independent risk factors for HAI in ICU patients (all P<0.05). The discrimination of the model (area under the receiver operating characteristic curve ［AUC］ of 0.841), calibration (Brier score of 0.129), and clinical effectiveness (net benefit of 11.4% when the risk threshold was 5%-74%) all performed well. Conclusion The QR code-based APP prediction tool is of great significance for scientific research transformation and precise HAI control.