Background: Cetuximab significantly improved survival outcomes of recurrent or metastatic head and neck squamous cell carcinoma (R/M HNSCC) when combined with chemotherapy. The aim of this study was to explore clinical outcomes, prognostic factors, and risk stratification model for R/M HNSCC patients who received cetuximab-containing regimens based on a real-world, multicenter Taiwanese patient cohort. Methods: This is a retrospective study involving 12 oncology institutions in Taiwan. All R/M HNSCC patients who received cetuximab-containing regimens from January 2017 to December 2020 were included in this study. Prognostic factors were evaluated by univariate/ multivariate analysis. The factors that showed significant differences (p < 0.05) were selected to establish the prediction model. The receiver operating characteristic (ROC) curve was obtained to select cut-off values as a reference for continuous variables. The risk score system incorporated both continuous and categorical factors. The score was determined according to hazard ratio. Results: A total of 818 R/M HNSCC patients were included in this study. Patient characteristics were as following: median age, 56 years; performance status (PS) 0/1/≧2, 16.0%/70.4%/13.3%; oral/ oropharynx/ hypopharynx/ larynx/ others, 51.5%/17.5%/19.7%/6.4%/5.0%; stage at initial diagnosis (AJCC8), 0/I/II/III/IVA/IVB/IVC/unknown, 0.2%/5.9%/8.1%/7.1%/ 35.5%/18.0%/6.7%/18.6%; locoregional recurrence/ distant metastasis/ unknown, 38.4%/ 56.0%/ 5.6%; site of distant metastasis, lung/ distant lymph node/ bone/ liver/ skin/ brain, 60.3%/ 37.1%/ 18.8%/ 6.3%/ 8.1%/ 3.3%; cetuximab-PF/ cetuximab-non-PF regimen, 56.6%/ 43.4%. The median overall survival (mOS) was 10.0 months (95% confidence interval [CI] 9.1-10.9 months). Multivariate analysis disclosed poor prognostic factors on OS, including poor PS, smoking history, R2 resection of primary surgery, present distant metastasis at bone, cetuximab combined with non-PF regimen. In addition, HB and neutrophil to lymphocyte ratio (NLR) were shown to have significant difference between treatment responders and non-responders (inc. SD pts). Risk-stratification model was established including factors: PS, smoking history, bone metastasis, hemoglobin level, and NLR. The mOS of the three risk groups stratified from the prediction model were 13.0/7.0/4.0 months (p < 0.001). Conclusions: Poor prognostic factors for R/M HNSCC treated with cetuximab-based regimens includes poor PS, smoking history, R2 resection of primary surgery, bone metastasis, and non-PF regimen. In this study, the risk-stratification model for cetuximab-based treatment was established using some of the identified prognostic factors help to predict the overall survival for R/M HNSCC patients.