It all depends on how you look at the data.

Last month, citing a new state of Florida study, the New York Times reported that the "rate of deaths caused by prescription drugs was three times the rate of deaths caused by all illicit drugs combined." The story's headline, "Legal Drugs Kill Far More Than Illegal, Florida Says," reinforced the image of prescription pharmaceuticals exterminating Florida's drug users by the thousands.

Nobody denies that many psychoactive drugs—prescription or otherwise—can be deadly. But a hard look at the Florida study (PDF) and its underlying data indicates that what's killing most users in the drug-saturated state—and, by extension, in the rest of the country—is not individual drugs. The deadliest of drug-taking behaviors is the consumption of multiple drugs, or, in the lingo of the drug-abuse industrial complex, "polydrug abuse."

The study, conducted by the Florida Department of Law Enforcement (FDLE) and the Florida Medical Examiners Commission, analyzed the cases of 8,620 people 1) who died in the state during 2007, 2) whose death led to a medical examiner's report, and 3) who had one or more major drugs (including alcohol) in their bodies at the time of death. The "vast majority" of cases involved more than one drug, according to the study.

"The state's medical examiners were asked to distinguish between the drugs being the 'cause' of death or merely 'present' in the body at the time of death," the study states. Because medical examiners often attribute cause of death to multiple drugs, a single death can result in two or three drugs earning "credit" for causing the death. The report provides this disclaimer about such double- and triple-counting: "Many of the deaths were found to have several drugs contributing to the death, thus the count of specific drugs listed is greater than the number of cases."

So when the Times reports that "benzodiazepine, mainly depressants like Valium [diazepam] and Xanax [alprazolam], led to 743 deaths," it's lifting numbers directly from the report. But most of those deaths were actually polydrug deaths.

For instance, the report recorded 556 deaths caused by alprazolam and another drug (or other drugs) but just six deaths in which alprazolam was the only drug present and caused the fatality. Likewise, diazepam in combination with another drug (or other drugs) caused 171 deaths. By itself, it caused just three.

The pattern repeats for other popular pharmaceuticals used illicitly. Oxycodone (OxyContin): 664 deaths in combination, 41 alone. Hydrocodone (Vicodin): 251 deaths in combination, 13 alone. Propoxyphene (Darvon): 76 deaths in combination, nine alone.

The Times article also neglects to acknowledge that many of the deaths attributed to prescription drugs in the study were ruled suicides. Pages 34 and 35 of the study report that 19 percent of the alprazolam deaths and 21 percent of the diazepam deaths were suicides. For these individuals, the drugs were no more dangerous than the walkway along the Golden Gate Bridge. They chose to make the drugs deadly.

None of this is to endorse the recreational use of prescription drugs as safe. The obvious conclusion, though, is one that Florida authorities and the New York Times avoid—namely, that pharmaceuticals that are extraordinarily safe when taken under a doctor's direction can become wildly hazardous when combined with other drugs. Drug users should never mix their drugs!

What prevents the state of Florida and the Times from noting the obvious? Perhaps both worry they'll be accused of encouraging illicit drug use and would rather watch drug users die.