STATS ARTICLES 2012
ProPublica fumbles painkiller story; Seattle Times scores touchdown
Rebecca Goldin, Ph.D and Cindy Merrick. Jan 31, 2012
ProPublica leads attack on painkillers by misusing statistics; But Seattle Times investigation uncovers really shocking numbers
Last November, the CDC’s Morbidity and Mortality Weekly Report (MMWR) included data on the sharp increase between 1999 and 2008 in deaths attributable to overdose of prescription opioid pain relievers (OPRs). These increased OPR-related deaths correspond to an overall increase in accidental drug poisoning deaths – accounting for far more deaths than any decreases seen in fatal overdoses of other drugs. In 2008, roughly 40 Americans died every day from overdose involving OPRs.
Although the new data was widely reported – NPR, ABC News, The Chicago Tribune, The Washington Post, and more all covered it – somehow, the splash at first seemed muted. Until, that is the non-profit investigative journalism outfit ProPublica decided to lambast prescription opioid use and administration with a shocking but misleading statistic. “The news about narcotic painkillers is increasingly dire,” said ProPublica. “Overdoses now kill nearly 15,000 people a year — more than heroin and cocaine combined.” A version of the ProPublica story was published in the Washington Post.
ProPublica quotes the CDC’s director, Dr. Thomas Frieden saying, “Right now, the system is awash in opioids, dangerous drugs that got people hooked and keep them hooked.”But neither ProPublica or, for that matter, Dr. Frieden report what the actual CDC’s data says.
There is a subtle but misleading switch from the CDC noting deaths “involving” OPRs to claiming that only OPRs were responsible for the deaths. The CDC data do not distinguish among those who died of overdoses with OPRs alone and those who died of overdoses of a cocktail of drugs including OPRs and either other prescription medications or illicit drugs. The distinction here means the difference between someone who has become accidentally addicted through legitimately prescribed medicine and someone who is abusing painkillers along with other drugs; in other words, a drug addict.
The CDC reports deaths attributed to drug overdose at 36,450 per year, with OPRs involved in 14,800 of these deaths. Yet the method for collecting data – statements on death certificates, leave room for enormous confusion about what substance, and whom, to blame for these deaths.
ProPublica, like many news sources, confuses the use of opioids under the supervision of a doctor, and their use more generally. In a large community sample coming from the National Survey on Drug Use and Health data (NSDUH) on people 18 and over in 2006-2008, a study published last year in the Archives of Internal Medicine found that 69 percent of those who use OPRs for nonmedical purposes had no physician source at all.
A full 80 percent of those who misused OPRs had a source other than a physician. While too much free-wheeling prescription is feeding a black market on these drugs, the widespread use of OPRs is clearly not a result of doctors casually prescribing them, but rather that people who want OPRs can find them outside the medical world, or through prescriptions intended for other people. In fact, this study found that opioid analgesic “abuse or dependence” (which consists of patients taking the drugs for the fun/feeling of the drug) was associated with having a prescribing doctor only in people 50 and over, and even among users in this group, 48.5 percent did not have a physician who prescribed it to them.
In furthering the idea of doctors going wild, ProPublica cites the example of “Dr. William Hurwitz, a pain doctor in Virginia who had been convicted on 50 counts of drug trafficking… [and was] accused of prescribing a single patient as many as 1,600 Roxicodone pain pills in one day. Hurwitz allegedly had prescribed that patient alone more than 500,000 pills between July 1999 and October 2002.”
It claims the pharma-backed American Pain Foundation filed a friend-of-the court brief supporting Dr. Hurwitz, which suggests that pharma has a more-than-medicinal interest in this kind of prescribing, which seems, on the face of it, medically implausible.
What ProPublica doesn’t note is a remark made by the presiding judge in Dr. Hurwitz’s appeal. “‘The amount of drugs Dr. Hurwitz prescribed struck me as absolutely crazy,’ U.S. District Judge Leonie Brinkema told the AP. But after hearing testimony from both sides, ‘I totally turned around on that issue… The mere prescription of huge quantities of opioids doesn’t mean anything.’”
Who is overdosing?
The questions of who is overdosing on OPRs, and why, are central to interpreting these data. But this is not a question that ProPublica is particularly interested in, given what it sees as evidence of carelessness and corruption on the part of physicians and the pain med companies. ProPublica points to the profits to be had by companies making OPRs, as if their profits alone indict the drug. It may be the case, of course, that profits are driving pharmaceutical fat cats into setting up informational websites that are merely fronts to cover “Big Pharma” interests. Money being made, however, is a separate issue from whether patients have a medical need for the product.
Unfortunately, or perhaps fortunately, just about every drug sold with a brand name in the United States has an informational website funded directly or indirectly by a company that stands to profit from health consumers choosing their products. It only seems natural that pharmaceutical companies would want to advocate for patients who want their products, and that those advocating for patients would ask these companies to foot the bill. It’s not a hard pill to swallow.
More relevant to the public interest is a study last year published in Journal of American Medical Association (JAMA) which found that, among people who are given prescriptions, the risk of overdose was directly related to the maximum prescribed daily dose of opioids. These ratios varied among different groups, reaching as high as HR=11.00 (95% CI, 4.42-32.56) for cancer patients, and as low as HR = 4.54 (95% CO, 2.46-8.37) in substance use disorder patients. But this study was not about the majority of overdose deaths: those without a prescription. As the National Institute for Drug Abuse (NIDA) notes, many overdoses result from people taking the pills in non-prescribed ways, such as crushing and snorting them.
But are there, underneath the surface, a large number of addicts dying from the opiates distributed in the legal market? According to the JAMA study, approximately 0.04 percent of patients (i.e. 2 in 5,000) overdosed from their prescriptions in 2004-2008. While small, the death rate is high enough to merit some concern. This may be a messy, uninspired story of mostly white, middle-of-their-lives Americans, to whom “addiction” used to mean something a whole world away, not medically endorsed pain relief, turning into a slow slide ending with unconsciousness in an emergency room.
What about teen users?
Typically, a pronounced increase in abuse of a specific drug can be measured by an acute effect on the youngest users. But in the case of OPRs, the increase in deaths related to overdoses has left kids less harmed than older adults. According to The National Survey on Drug Use and Health (NSDUH), the average age for first-time use of OxyContin is 22.8 years old among people ages 12-49.
Bolstering the view that kids don’t have the problems that adults do, the 2010 Monitoring the Future Survey reported that the use of Vicodin (hydrocodone/paracetamol) has gone down from last year’s 9.7 percent of 12th graders reporting having used the drug in a nonmedical situation in the past 12 months, to 8 percent reporting the same for 2010. However, reported abuse of OxyContin has remained constant, at about 5 percent of kids reporting use of this drug in the past year. OxyContin’s use has remained roughly constant since 2001. The overall proportion of 12th graders who tried any illicit drug other than marijuana was 17 percent in 2010, and continues a period of decline since about 2001, when it was about 20 percent.
Who is at fault?
The problem with tackling prescription opioid abuse is that there isn’t a single culprit. Doctors may be partially at fault. States regulate prescription drugs inconsistently, and one study (Swedlow, et al.) found that, among a group of Californian doctors in Oakland, CA, “3% of physicians account for 62% of OPR prescribed.” At the same time, however, the DEA’s crackdown and prosecution of doctors who prescribed pain medication may be the root cause of this skewed statistic, causing patients in pain to search for doctors willing to prescribe.
One extraordinary three-part journalistic investigation that has shed new light on the problem appeared recently in the Seattle Times, which looked at how Washington State pushed people who had state-subsidized health care (e.g., Medicaid) towards methadone instead of much more expensive painkillers like OxyContin. Even though methadone accounts for less than 10 percent of the narcotics prescribed, the drug, says the paper, was responsible for more than half of painkiller overdose deaths in the state. Unlike OxyContin, which has a short half-life, methadone does not quickly dissipate from the body; and depending on how much is taken, and whether it is taken with other medication, it can depress the respiratory system. Patients can fall asleep and never wake up. As the paper notes, the state has repeatedly downplayed this risk while saving millions of dollars.
With $72.5 billion per year in health care costs being attributed to the abuse of OPRs, and in light of the CDC research showing that “Medicaid populations are at greater risk of OPR overdose than non-Medicaid populations,” the Seattle Times investigation has national significance.
But anyone who wants to understand pain treatment in the U.S. today needs to understand the way law enforcement and the media have often acted to create a feedback loop in the promotion of drug “epidemics,” especially with regard to the previous OxyContin “epidemic” in the early part of the last decade. The science and information that gets omitted in these panics is itself a source of alarm, as noted in Reason, The New York Times Magazine, by the Cato Institute – and by STATS. For anyone who might believe that there hasn’t been extraordinary injustice committed in the fight against painkiller abuse, the case of paraplegic and multiple sclerosis sufferer Richard Paey, (CBS 60 Minutes) now fully pardoned, (Reason) is a warning.
There are also serious problems with data collection. Deaths from OPRs are almost certainly underreported, because many certificates in drug deaths do not specify the type of drug, and respondents to health surveys are likely to underreport nonmedical (unprescribed) OPR use.
It can be argued that our population will always maintain a certain percentage of drug abusers, which remains relatively steady among the vicissitudes of specific substances being abused in any particular era. However, the CDC report brings to light a serious and growing problem above and beyond those “usual suspects.” Despite the good news regarding high school students, as a percentage of the population, more Americans are dying than ever before due to drug poisoning. In 1999, the rate of drug overdose deaths was 4 in 100,000 Americans. By 2004 this number had risen to 6.7 in 100,000, and by 2008 to 12 in 100,000.
As reported by the MMWR, the tripling of overdose deaths in this time period, from 1999-2008, tracks closely with the period of nearly quadrupled growth of OPR sales. By 2008, the annual overdose death rate from just prescription drugs had quadrupled from 1999. By 2009, the abuse treatment rate was nearly six times that of 1999. On the other hand, the NSDUH found in 2008 that among people who used opioids for nonmedical purposes, only about 3 percent are dependent a year later.
Unfortunately, this sad, messy, uncomfortable problem is the new addiction reality – and we need the media to reflect this lest more harm be done by crackdown on physicians and pain treatment than good. In this, the Seattle Times series has made a truly vital contribution.