OBJECTIVES: This study aimed to examine the association of payer status with length of stay in patients hospitalized for injury. METHODS: A total of 1,020 hospitalization records related to injury in 2006 National Hospital Ambulatory Medical Care Survey-Emergence Department Visit Data (NHAMCS-ED) were analyzed with national weight. A general linear model (GLM) was used to explore the association between payer status and length of stay, controlled for predisposing factors: age, gender, race; illness factors: intent type, pain, length of ED visit, total number of procedures in ED; and enabling factors: ownership of the hospital, and whether or not the hospital received any Medicaid disproportional share program funds in 2005. Multiple comparisons of least squared means using the self-pay group as reference group were adjusted based on Bonferroni’s method. Length of stay was logarithmically transformed to depress its variance. RESULTS: The weighted sample size is 3,119,552. For 33,202 patients whose primary expected sources of payment was worker compensations, the mean length of stay was 5.2 days (95% CI: 2.4-11.2); 4.9 days (95% CI: 4.6-5.3) for 1,046,655 Medicare patients; 4.8 days (95% CI: 2.6-5.7) for 18,445  whose expected source was no charge/charity; 4.3 days (95% CI: 3.9-4.7) for 619,854 with Medicaid/SCHIP; 3.9 days (95% CI: 3.6-4.2) for 806,346 with private insurance; 2.9 days (95% CI: 2.6-3.2) for 357,032 self-pay patients. The GLM fit well (F=6.33, p<0.0001, R2=0.16) in this study. Effects of payer status, age, race and total number of procedures were statistically significant (all p-values<0.005). As to multiple comparisons, patients with either of Medicaid/SCHIP and private insurance stayed in hospital significantly longer than self-pay patients (1.4 days, 95% CI: 1.1-1.7, p<0.01; 1.3 days, 95% CI: 1.1-1.6, p<0.01, respectively). CONCLUSIONS: Payer status was significantly associated with the length of stay for inpatients with injury, indicating a significant variation due to health care access.