Researchers at Columbia University’s Mailman School of Public Healthlooked at the frequency of nonmedical prescription opioid use and the risk of heroin-related behaviors and found that past-year heroin use rose among individuals taking opioids like oxycontin, and these increases varied by race and ethnicity. The most significant rise in heroin use was among Hispanics and non-Hispanic whites, where the rate of heroin use for the latter group increased by 75 percent in 2008-2011 compared to earlier years. Findings are online in the journal Drug and Alcohol Dependence.
Nonmedical prescription opioid use is defined as using a substance that is not prescribed or taking a drug only for the experience or the feeling it caused. The research, led by Silvia Martins, MD, PhD,associate professor of Epidemiology at the Mailman School of Public Health, sheds light on the racial and ethnic differences in trends of nonmedical opioid and heroin use over time.
Using data from the National Survey on Drug Use and Health, a large nationally representative household sample of 67,500 people, and self-reported heroin use within the last 12 months, the researchers examined the change in patterns of past-year non-prescription drug and heroin use between 2002-2005 and 2008-2011 across racial and ethnic groups. The study also looked at the association between past year frequency of both, heroin-related risk behaviors, and exposure to heroin availability. For those who had endorsed using heroin in the past, participants were also asked how they administered the drug.
In 2008-2011, the risk of past-year heroin use, ever injecting heroin, past-year heroin abuse or dependence, and the perception of availability of heroin increased as the frequency of nonmedical opioid use increased for all race and ethnicities, but particularly for non-Hispanic Whites.
“We found that individuals endorsing past year non-prescription opioids who also use heroin are likely to be in more advanced stages of their drug use,” said Martins. “The individuals tend to use prescription opioids as a substitute for heroin when heroin is unavailable, to augment a heroin-induced ‘high,’ to ‘treat’ withdrawal symptoms, and to curb heroin use.”
Regarding frequency of use, for Hispanics, increases were significant only among those using opioids about 1-29 days in the past year. Among blacks and whites, significant increases in the rate of heroin use were observed among those using prescription opioids more frequently (100-365 days) in the past year.
Consistent with earlier research, with the exception of Hispanics, frequent prescription opioid users of all race and ethnicities and heroin users were at increased risk of ever injecting heroin and of past year heroin abuse and/or dependence. “This is alarming and raises concern since injection drug use among prescription opioid users can contribute to the spread of HIV, as recently reported in Southern Indiana, as well as of Hepatitis C,” says Martins.
“It is possible that Hispanics who were frequent opioids users—more than 100 days per year—and who were also likely to use heroin in previous years, are now only able to use prescription opioids 1-29 days a year in the past 12 months due to recent constraints in the prescription opioids market,” observed Martins.
“The noteworthy increase in the annual rate of heroin abuse or dependence among non-Hispanic Whites parallels the significant increase in nonmedical opioid use during the last decade and the growing number of heroin overdose deaths described for this race and ethnic group in recent years,” said Martins. “Overall, our results suggest a connection between opioid and heroin use and heroin-related adverse outcomes at the population level, implying that frequent nonmedical users of prescription opioids, regardless of race or ethnicity, should be the focus of public health efforts to prevent and mitigate the harms of heroin use.”
The study was funded by the National Institute of Drug Abuse, National Institutes of Health (grants K01DA030449, R03DA037770, and R01DA037866), and the Eunice Kennedy Shriver National Institute of Child and Human Development (grant HD020667). The authors report no conflicts of interest.
