**Background:** Peripherally inserted central catheters (PICCs) are widely used in China for long-term transfusion, chemotherapy, and parenteral nutrition. Proper PICC maintenance—including aseptic technique, function evaluation, connector and dressing replacement, flushing and sealing, and health education—is critical to prevent complications such as venous thrombosis, catheter-related bloodstream infection, phlebitis, occlusion, and migration. In China, nurses are responsible for PICC maintenance. While some provinces have established province-level PICC maintenance networks with unified quality standards and training, Guizhou province had not. This study aimed to investigate the level of PICC maintenance practice among nurses in Guizhou and explore its influencing factors.
**Methods:** A web-based cross-sectional study was conducted from 1 July 2022 to 30 July 2022 in 11 tertiary and 26 secondary hospitals in Guizhou province. Eligible participants were registered nurses engaged in intravenous therapy with PICC maintenance experience who were willing to participate. Using G*Power V.3.1.9.7, a sample size of 787 was required; 832 nurses completed the survey (response rate 52.1%). Data were collected via Wen Juan Xing, an electronic data collection tool. Three validated questionnaires measured PICC maintenance knowledge (19 items, score 0–19, converted to 100-mark system), attitudes (17 items, 4-level Likert, score 17–68, converted), and practice (16 items, 5-level Likert, score 16–80, converted). Cut-off points were ≥60 for knowledge (pass), ≥80 for attitudes (positive), and ≥80 for practice (acceptable). Content validity was established via two-round Delphi with 20 experts (Scale Content Validity Index: 0.94, 0.98, and 0.99 for knowledge, attitudes, and practice, respectively). Reliability was assessed among 200 nurses (Cronbach's α: 0.71, 0.95, and 0.96). Descriptive statistics, univariate analyses (t-test, Mann-Whitney, Kruskal-Wallis), and standard multiple regression were performed using SPSS V.23.
**Key Results:** Participants' mean age was 32.05±6.03 years; 92.8% were women, 77.5% had a bachelor's degree or above, 69.6% held a nurse or senior nurse title, 67.1% had >5 years of work experience, 55.6% worked in tertiary hospitals, 91.8% had PICC guidelines available, and 83.9% had prior PICC maintenance training. Mean knowledge score was 53.57±13.80 (only 29.6% passed), mean attitude score was 89.93±11.25 (68.5% positive), and mean practice score was 79.77±12.13 (60.8% acceptable). Univariate analysis showed significantly higher practice scores among nurses with higher age (p=0.022), higher professional title (p=0.006), higher educational level (p=0.002), working in tertiary hospitals (p<0.001), having PICC guidelines (p<0.001), having training (p<0.001), passing knowledge (p<0.001), and having positive attitudes (p<0.001). Multiple regression (R²=0.33, F(8,823)=50.51, p<0.001) identified three significant predictors: availability of PICC guidelines (β=0.10, p=0.002), previous training on PICC maintenance (β=0.18, p<0.001), and attitudes toward PICC maintenance (β=0.48, p<0.001). Age, professional title, educational level, hospital type, and knowledge were not significant in the final model.
**Clinical Implications:** The finding that only 60.8% of nurses had acceptable PICC maintenance practice is concerning, especially given that 83.9% reported prior training, suggesting training effectiveness needs improvement. The strong predictive role of attitudes (β=0.48) indicates that interventions targeting nurses' beliefs about the importance of PICC maintenance may be particularly impactful. The availability of guidelines was a significant predictor, supporting the implementation of accessible, evidence-based protocols. The authors recommend establishing a province-level PICC maintenance alliance in Guizhou to develop and update guidelines, provide regular training, and implement programs to enhance nurses' attitudes. Study limitations include convenience sampling (potential selection bias), self-reported practice (social desirability bias), restriction to Guizhou province (limited generalizability), and lack of organizational-level variables (e.g., workload, staffing).