Reinserting feeding tubes that are accidentally removed exposes patients to risk and consumes hospital resources. We were interested to know if using a bridle to secure tubes would be more effective than tape at preventing accidental tube removal. This was a quality improvement project with a before-and-after design. Between May 2007 and August 2007, we prospectively followed 90 tubes (50 tape, 40 bridle). Tubes were followed up daily until accidental tube removal, ICU discharge, or planned tube removal. Our primary endpoint was accidental tube removal. We compared the 2 groups on the following: (1) proportion of tubes accidentally removed; (2) rate of accidental tube removal (per 100 tube-days); and (3) Kaplan-Meier survival analysis. Survival analysis data were right-censored at ICU discharge or planned tube removal. There were no significant differences between groups in any demographics. The proportion of tubes accidentally removed was 36% (18 of 50) in the tape group and 10% (4 of 40) in the bridle group; P<.05. The rate of accidental tube removal (per 100 tube-days) was 6.4 (18 in 281 tube-days) in the tape group and 1.6 (4 in 248 tube-days) in the bridle group; P<.05. Survival analysis showed a significant difference between the groups with a log-rank test for equality of survivor function of P<.05. Using a bridle to secure feeding tubes significantly reduces the proportion and rate of accidental tube removal and results in increased tube survival by Kaplan-Meier analysis.