Cellphone use while driving is one of the major concerns in traffic safety, and numerous researches found that cellphone use while driving can increase traffic collision risk. With growing popularity of cellphone use, many states in the United States enacted the law banning handheld cellphone use while driving in recent years. However, there is a debate on effectiveness of the law. In this study, we analyze the six-year collision data between 2006 and 2010 in the state of California to examine the timing of a significant change in the trend of cellphone related collisions. We adopt the turning point analysis technique without imposing a prior belief on whether or when such a change has occurred. Both the frequentist and Bayesian approaches are applied to four different groups, including the all cellphone collision group and three subgroups characterized by cellphone usage. The result shows that the turning points coincides with the timing of enforcement of the handheld law in California, except for the subgroup containing hands-free related collisions only. We applied the same method to two confounding factors including driving under influence and CD/radio use, and find that the turning points do not agree with the cellphone related collisions. Although a more comprehensive set of confounding factors needs to be considered to establish a causal relationship between the handheld law and cellphone related collisions, coincidence between the handheld law and the turning point suggests that the law should be considered as one of the primary factors in cellphone related collisions reduction.