It seemed like a good idea at the time. Make OxyContin pills harder to crush or dissolve and fewer people will abuse them--use them for purposes other than pain relief. But what the change actually seems to have accomplished is to give a big boost to heroin trafficking.

OxyContin has long been a favorite of people who take opioids for pleasure instead of for pain relief. Designed to release its medication (oxycodone) slowly over time, each pill contain a large amount of opioid. Crush it up or dissolve it in water and you've got one extremely potent narcotic.

Most previous abusers of OxyContin didn't cease their abuse after OxyContin was reformulated, they merely shifted their drug of choice -- to heroin.

In an effort to make the pills less attractive to abusers, OxyContin was reformulated in August of 2010. The new pills are much harder to crush or dissolve. This was supposed to cut down on their potential for abuse. And according to a letter published in the New England Journal of Medicine, the tactic has worked. But the cure may be worse than the disease.

Researchers at Washington University in St. Louis have been tracking patients entering drug abuse treatment programs for nearly three years, with a particular focus on changes occurring since OxyContin was reformulated. What they found is that most previous abusers of OxyContin didn't cease their abuse after OxyContin was reformulated, they merely shifted their drug of choice -- to heroin.

Heroin is much more dangerous to users than prescription opioids. It's a street drug that varies greatly in potency, which can lead to death from overdoses. It's also cut with a variety of substances, some benign and others harmful, such as strychnine or ketamine. But a thriving heroin trade doesn't just affect the drug abusers in a community; it affects every single community resident in many ways, most of them bad.

Which do you think is more toxic to your community, illicit OxyContin use or a brisk heroin trade?

The researchers surveyed over 2,500 patients entering drug treatment centers in 39 states from July 2009 through March 2012. And 103 of these patients agreed to more comprehensive telephone or online surveys.

The use of OxyContin as the primary drug of abuse decreased from 35.6% before the release of the new formulation to just 12.8% in the first quarter of 2012. In that same time period, the number of people who reported using heroin at least once in the past month nearly doubled, from around 10% to 20%. And these changes appear to be linked. In the words of one patient: "Most people that I know don’t use OxyContin to get high anymore. They have moved on to heroin [because] it is easier to use, much cheaper, and easily available." And at the same time, many law enforcement officials across the country are reporting an uptick in heroin use.

This isn't the first time an attempt to stem illegal drug use has had unintended consequences. Spraying of Mexican marijuana fields with the herbicide paraquat in the mid-1970s was a major factor in the rise of the domestic U.S. marijuana industry.

Paraquat was also sprayed in national forests in Georgia, Kentucky and Tennessee in 1983 as anti-drug tactic on behest of the Reagan administration. And while there are no confirmed reports of ill health effects among smokers of poisoned marijuana or to other people exposed to the spray, paraquat is highly toxic to humans and has also been linked to development of Parkinson's disease.

The researchers think that drug abuse strategies that focus on controlling supply simply don't work. They point out that all of the 103 respondents who completed the longer surveys said they would replace OxyContin with some other drug. And they seem to have had little trouble doing so.

Makers of opioids and other prescription drugs that are prone to abuse have been considering following OxyContin's lead and reformulating their products to make them more difficult to abuse. The results seen so far with OxyContin suggest that that may not necessarily be such a good idea.

The letter appears in the New England Journal of Medicine (NEJM).