States are taking efforts to prevent patients from "doctor shopping" across their borders. (AP Photo/Toby Talbot, File)

Deaths tied to opioid painkillers, like oxycodone and hydrocodone, have doubled to about 17,000 a year over the course of a decade — now outnumbering the people killed in traffic accidents each year. One reason behind the alarming rise: "doctor shopping," when patients search around for doctors who'll prescribe them powerful medications.

In a new study published this month, researchers tried to better understand where doctor shopping is actually taking place. Most strikingly, the analysis found about one-third of apparent "doctor shopper" patients crossed state lines to obtain a prescription, which helps explain why smaller states like Rhode Island, New Hampshire and Delaware showed the highest rates of this occurring. There's a major reason why doctor shoppers are crossing borders — it's harder to track the prescriptions. Many states have drug monitoring programs in place to track prescriptions dispensed within their borders, but not across them.

"Part of the problem is that state systems all vary — they're either home-grown or operated by different vendors and they're not interoperable," said Douglas McDonald of Abt Associates, who co-authored the NIH-funded study published in Pharmacoepidemiology and Drug Safety. "There are also legal questions about what you can share about patient data."

The following map show the county-level rates of doctor shopping, according to the study:


(Pharmacoepidemiology and Drug Safety)

States know this cross-border shopping is happening, and they're starting to deal with the problem. Just this year, New Jersey entered into agreements with Delaware and New York to share prescription drug information. New England governors, whose states have some of the highest rate of prescription drug abuse, this year also agreed to share monitoring data. Some physician groups have called for the creation of a national monitoring program.

The universe of doctor shoppers is relatively low — another study from McDonald last year estimated just about one of every 143 patients prescribed a prescription painkiller in 2008 appeared to be doctor shoppers. They represented 0.7 percent of all patients with such a prescription, but bought about 4 percent of all opioids measured by weight.

Doctor shoppers tend to be in their 20s and 30s, and they're pretty large outliers when compared to other patients prescribed opioids. They see an average of 10 doctors per year, receiving 32 different prescriptions and generally paying cash for at least one prescription to avoid detection, according to McDonald's research.

McDonald goes all the way down to the county level to try to understand the variation in doctor shopping behavior. He estimated 30 percent of the inter-county variation could be explained by the prevalence of opioid prescribers — in other words, a county will have more doctor shoppers if they have more patients and more prescribers. Another 30 percent of the variation could be explained by another combination of other local factors, like residents' income, health-care utilization rates and a county's racial and ethnic makeup.

The research, based on IMS Health records of 146 million prescriptions dispensed across the country in 2008, couldn't account for the rest of the 40 percent of the county-level variation. There are, of course, other methods of obtaining these drugs illegally beyond doctor shopping — forgery, theft, and shady Internet sites, just to name a few.

The data doesn't include clinical information that would allow the researchers to determine whether apparent doctor shoppers had legitimate medical reasons for their prescriptions. It may look like a patient is trying to abuse the system, but it could also be indicative of the patient receiving poorly coordinated care. But the likelihood of patients being a doctor shopper increases as the number of prescribers increases.