State programs differ widely; nuanced approach permits comparison of rates of participation and analysis of unintended social costs.

A study published in Health Affairs found that while 14 of the nation's 24 medical marijuana programs were essentially nonmedical in practice, they enrolled more than 99 percent of overall participants. Fewer than one percent were enrolled in "medicalized" programs that adhere to accepted professional standards in medicine.

More than one million people participate in medical marijuana programs in the United States.

The study analyzed the extent to which medical marijuana laws and program regulations incorporate accepted medical practice, good pharmaceutical manufacturing practices as established by the FDA, and restrictions on controlled substances. The authors evaluated the strength of the association between the medicalization of each marijuana program and its enrollment rate, and assessed state-specific characteristics such as the number of physicians per capita, the burden of terminal disease, medical marijuana costs, and rates of recreational marijuana use.

The study found that laws initiated before 2009, primarily those in western states passed by voter initiative, depart the most from the medical model. The more recent programs in the midwest and northeast typically required years of work at the state level to move from the initial passage of a law to full implementation of a state-licensed manufacturing and dispensary system.

For researchers, these findings are a clear indication that data from states with medical marijuana programs cannot simply be compared to states lacking such programs. For policy makers, data that differentiate among the programs is key to interpreting the effects of medical marijuana laws.

"The extent to which states regulate their medical marijuana programs appears to have a striking impact on how many people actually use each program," noted Mark Olfson, MD, professor of psychiatry at Columbia University Medical Center (CUMC) and senior author of the paper. "High enrollment rates in less regulated medical marijuana programs raise the possibility that these programs may inadvertently attract recreational users."

"This is the first study to use a nuanced approach to analyzing these data," said Silvia Martins, MD, PhD, associate professor of epidemiology at Columbia's Mailman School of Public Health and co-author. "These results illuminate the medical and social issues that can result from the existing wide variance in U.S. medical marijuana programs at the state-level. Future studies need to further explore in greater detail variations in these medical marijuana programs across time."

"The new findings raise questions of why doctors are involved in non-medical programs in the first place," added lead author Arthur Robin Williams, MD, MBE, a fellow in the Department of Psychiatry at CUMC. "Building on this study we will look for associations between more restrictive regulations and public safety - including rates of recreational marijuana use, diversion to adolescents, emergency department visits and drug treatment admissions."