Hours-of-service (HOS) regulations control the maximum daily drive time, workday hours, and work week (period) hours for commercial motor vehicle (CMV) drivers. The regulations also include periods of off-duty time that drivers must take before beginning a work shift, referred to as shift-restart methods in this study. In the 2005 regulations, the shift-restart methods included taking at least 10 consecutive hours off duty or in the sleeper berth (10+ hour restart), taking at least 34 consecutive hours off duty or in the sleeper berth (34+ hour restart), and a sleeper berth provision (SBP). The SBP requires one period of at least 8 (but less than 10) consecutive hours spent in the sleeper berth plus a period of at least 2 (but less than 10) consecutive hours spent in the sleeper berth, off duty but not in the sleeper berth, or a combination of off-duty time spent in and out of the sleeper berth. The purpose of this project was to examine the usage of shift-restart methods and the relationship between shift-restart methods and driver safety performance in a naturalistically collected driving data set. The data used for this study were collected by the Virginia Tech Transportation Institute (VTTI) in the Naturalistic Truck Driving Study (NTDS) and developed into a hybrid data set of naturalistically collected video data and activity register data that accurately detail the participating CMV drivers’ driving and non-driving activities.(7) With the activity register data, researchers determined which restart method drivers used before beginning a new work shift: 10+ hour restart, 34+ hour restart, or the SBP. The proportion of shifts preceded by SBP breaks was significantly higher for drivers who reported taking medications regularly versus those who did not and also for drivers with longer average delivery distances. The number of years of CMV driving experience had a significant inverse relationship with the proportion of total shifts with SBP breaks. A mixed-effect negative binomial model with a logarithmic link function was used to model safety-critical event (SCE) rate at the shift level, controlling for the driver. The SCE rates in shifts following an SBP break were found not to be statistically different from those in shifts following 10+ hour or 34+ hour restart breaks. Odds ratios were also used to assess the risks associated with each of the three shift-restart methods. The 10+ hour restart and 34+ hour restart methods were found not to be significantly different. However, both the 10+ hour restart and 34+ hour restart methods were associated with significantly higher risk than the SBP. This project serves to enhance the understanding of the current HOS regulations and the impact that these regulations have on drivers, a topic of significant concern in the CMV community. Drivers have different preferred break usage patterns. The use of the SBP in the current study does not appear to be associated with a decrement in safety performance. Future efforts should look into how the usage of shift-restart methods has changed under the new regulations, which went into effect on July 1, 2013, and modified the driving limits, on-duty time limits, and rest break requirements.