Unfortunately, the graphs are incomplete there needs to be a longer term study three years is not enough time.
More indepth scientific research is necessary.
I referenced one earlier-
Check this out from NHTSA's
Drug and Alcohol Crash Risk: A Case-Control Study, (
emphasis added)-
"Conclusions
The study confirmed previous research indicating alcohol is a greater contributor to crash risk than drugs ... When age, gender, race/ethnicity, and alcohol consumption are taken into account,
there was no significant contribution of drugs to crash risk. This finding seems to contradict previous studies (Asbridge, Hayden, & Cartwright, 2012; Blows et al., 2005; Hels et al., 2011) that indicate a statistically significant contribution of drugs to crash risk, even if sometimes small or moderate.
However, the strength of this study lays in its rigorous methodology, stringent data collection procedures, controlled case-control matching, comprehensive laboratory testing, and sophisticated statistical analyses...
Additionally, because drug classes affect driving skills differently, overall crash risk estimates may underestimate the contribution of certain drugs to specific types of crashes. The role of THC may differ in its crash risk profile than stimulants. The results indicate that alcohol remains the main contributor to crash risk. Drugs other than alcohol, and when combined with alcohol was not a significant factor in crash risk."