Don’t Blame Addicts for the Painkiller Overdose Epidemic

Rehab Helps Thousands of Addicts Quit. It Can Help You, Too. 1-888-341-7785

If we were really serious about preventing death from drug overdose we should make it our number one priority to teach people that drug mixing will kill them. We would recognize the following facts:

  • The vast majority of people who die of opioid painkiller poisoning (so-called “overdose”) are pain patients who are taking their medication as prescribed.
  • Only a minority of people who die of opioid painkiller overdose are non-medical users and only a small fraction of them meet criteria for opioid dependence.
  • Over 90 percent of people who die from a so-called opioid “overdose” actually die from drug mixing.
  • Although evidence based addiction treatment has value in its own right, it is NOT the primary tool we need to fight drug poisoning deaths.
  • The primary tool we need to fight drug poisoning deaths is a massive PR campaign on all media from TV to newspapers to internet to teach people that if they mix opioids with alcohol, benzodiazepines, or cocaine they are at a high risk for death.
  • The population most at risk of death are pain patients taking high dose opioid prescriptions, next after that are new young suburban heroin users who don’t know the dangers of mixing heroin with alcohol, benzodiazepines, or cocaine.
  • Calling drug mixing deaths by the misnomer “overdose” misleads the public and causes more drug poisoning deaths.

There is a popular myth constantly repeated in the media that rates of opioid addiction and non-medical use are soaring and that only addicts and recreational drug users die of overdose. Of course the press wants us to believe that “addicts” deserve this death because “addicts” are bad people. However, this myth is belied by the actual data from the National Survey on Drug Use and Health (NSDUH) as we can see in Figure 1 and Table 1 below. Note that this is the data for opioid painkillers only and does not include heroin, which I will discuss separately. (Note: all mortality data in this article is from CDC WONDER; drug use data is from SAMHDA.)


Although we see an increase in all four categories – non medical painkiller use, painkiller use disorder, painkiller dependence, and painkiller abuse – between the years 2000 and 2002, the lines are relatively flat between 2002 and 2013. In particular, we see that the rate of painkiller use disorder is the same in 2013 as it was in 2002: 0.7 percent. Non medical use decreased slightly going from 4.7 percent in 2002 to 4.2 percent in 2013. We also see a trade off in abuse and dependence rates – the increase in rates of dependence between 2002 and 2013 are offset by a drop in the rate of abuse.

Let’s compare this to rates of opioid painkiller overdose deaths seen in Figure 2.

As we can see when we compare Figure 1 with Figure 2, there are no increases in rates of painkiller use disorder or non medical painkiller use which would correspond to the increase in rates of opioid painkiller overdose deaths. However, when we compare Figure 2 with Figure 3 we see a close correspondence between number of prescriptions and number of painkiller overdose deaths, both of which have roughly doubled between 2000 and 2013.

Even more telling is what we see when we compare the number of high dose opioid prescriptions with the number of opioid painkiller overdose deaths by age group and sex. Figure 4 gives the number of opioid painkiller overdose deaths by age group and sex for 2013. Figure 5 gives the number of high dose opioid prescriptions by age and sex for 2013 (data from Express Scripts).


When we compare the curves in Figure 4 and Figure 5 we see almost perfect correspondence between the number of high dose opioid painkiller prescriptions and opioid painkiller poisoning deaths. Males show a bimodal distribution with peaks at around the mid 30s and around age 50 for both overdose death and high dose prescriptions. Women show a peak at around their early 50s for overdose death and high dose prescriptions.

If we compare the rates of opioid painkiller poisoning death by age and gender with the rates of non medical painkiller use, painkiller use disorder, and painkiller dependence, however, we see no correspondence between them. These are given in Figures 6, 7, and 8 (data from SAMHDA).



As we see from Figure 4, male opioid painkiller poisoning deaths show a bimodal distribution with peaks at 34 years of age (138.63 deaths per million) and 50 years of age (139.3 deaths per million). This does not correspond at all for the data we see for non medical use, use disorder, or dependence. For males, non medical use peaks at 12.6 percent for 20 year olds then drops down to 7.3 percent for 30 to 34 year olds, 4.4 percent for 35 to 49 year olds, and 1.7 percent for 50 to 64 year olds (Figure 6). Likewise, male painkiller use disorder peaks at 2.2 percent for ages 22 to 25 then drops down to 1.9 percent for 30 to 34 year olds, 0.8 percent for 35 to 49 year olds, and 0.5 percent for 50 to 64 year olds (Figure 7). Finally, male opioid painkiller dependence peaks at 2.0 percent for ages 24 to 29 then drops to 1.6 percent for 30 to 34 year olds, 0.4 percent for 35 to 64 year olds (Figure 8).

For females, opioid painkiller poisoning deaths peak at age 52 at 122.9 deaths per million (Figure 4). However, female non medical opioid painkiller use peaks at 8.7 percent for 22 to 23 year olds and drops to 2.4 percent for 50 to 64 year olds (Figure 6). Use disorder in females shows a bimodal distribution with peaks at 1.5 percent for 20 year olds and 1.4 percent for 30 to 34 year olds. This drops to 0.2 percent for 50 to 64 year olds (Figure 7). Dependence in females also shows a trimodal distribution with peaks at 1.1 percent at age 20, 1.1 percent at ages 22 to 23 and 1.2 percent at ages 30 to 34. At ages 50 to 64, when painkiller deaths are peaking, dependence drops to 0.1 percent (Figure 8).

In other words, high rates of opioid painkiller poisoning deaths correspond to high rates of high dose prescribing and to medical use; they do not correspond to high rates of non medical use, use disorder, or dependence.

The media needs to stop promoting the image of a typical opioid overdose as a junkie dying with a needle stuck in his/her arm; this type of overdose death is quite rare. If our goal is to save lives we need to recognize that overdoses most frequently occur when someone is taking a high dose opioid prescription according to doctor’s orders, but drinks alcohol on top of it or takes benzodiazepines with it, even the benzodiazepines may also be taken according to doctor’s orders. There may be an assumption that because you didn’t die the first time you mixed your prescription with alcohol that it is safe. However, just because you didn’t die the first time doesn’t mean you won’t die the 40th or 100th time. Drug mixing is deadly; referring to it as “overdose” misleads the public and puts them at risk.

The media needs to stop promoting the image of a typical opioid overdose as a junkie dying with a needle stuck in his/her arm; this type of overdose death is quite rare.-Kenneth Anderson

This is not a new message nor is it a new phenomenon. As early as 1972, Brecher published conclusive evidence that the majority so-called heroin overdose deaths are actually caused by drug mixing, not by large doses of opiates. Darke and Hall (2003) also found that drug mixing was the primary factor at work in so-called heroin overdose deaths; changes in the purity of heroin were only a minor factor. In fact the purity of heroin in New York City peaked in 1995 at 69.4 percent pure; it has dropped to 41.3 percent pure in 2013 in spite of New York Post scare stories about “superheroin.”

Maybe now that white middle class people in the suburbs are dying we will be able to convince the media to drop the erroneous stereotypes and listen to the message that drug mixing kills. This is not to minimize the huge importance of naloxone programs and evidence based drug treatment such as methadone and buprenorphine: these are essential lifesavers too. But above all it is paramount that we get out the message about the dangers of mixing alcohol, benzodiazepines, cocaine, or other opiates on top of opioid painkillers or heroin.

Prescription drug monitoring programs have not reduced opioid overdose, they have only driven opioid users to take up heroin. And heroin is far more dangerous than the prescription painkillers it is replacing because of its low therapeutic index. In other words, the fatal dosage for heroin is too close to the dosage that gets you high. This makes it even more likely that people will die if the mix alcohol or other drugs with heroin than if they mix them with painkillers. Naive, young, white, suburban, middle class heroin users have no idea how deadly it is to mix heroin with alcohol or benzodiazepines or cocaine. Whereas inner city users in places like New York City transmit this information via a grapevine of user networks and needle exchange programs which has kept rates of heroin overdose in New York City from skyrocketing the way they have in most of the US. Figure 9 shows the trends of heroin poisoning death rates in New York City which can be compared with those for the nation in Figure 2.

This is a direct result of prescription drug monitoring programs cutting off opioid painkiller users who are showing signs of dependence and sending them out in search of heroin instead.-Kenneth Anderson

Prescription drug monitoring programs have been an abysmal failure at reducing opioid poisoning deaths and opioid use disorders. Slight reductions in painkiller poisoning deaths have been more than compensated for by the major increases in heroin overdoses we see in Figure 2. Heroin overdose death has traditionally been associated with men in their 40s. However, the implementation of prescription drug monitoring programs over the past decade or so has also been resulted in a major shift so that now most heroin overdose deaths are occurring with young people in their 20s, as we see in Figure 10 and Figure 11. This is a direct result of prescription drug monitoring programs cutting off opioid painkiller users who are showing signs of dependence and sending them out in search of heroin instead. This is the worst policy we could have adopted. The sane thing to do would be for doctors to maintain dependent patients on their painkiller of choice and teach them how to use it without dying. Unfortunately, the law sends doctors to prison for this.

In conclusion, if we are really serious about reducing opioid related deaths in the U.S. we need to implement the following:

  • A wholesale educational campaign through all media from TV to internet to newspaper to teach people that even if they take painkillers according to doctor’s orders, drug mixing can kill them.
  • An equally broad media campaign to warn all non medical opioid users including heroin users of the dangers of drug mixing.
  • To stop calling drug mixing deaths “overdoses.”
  • Maintenance prescriptions on one’s opioid of choice, whether that be Oxycontin or what have you.
  • Evidence based opioid treatment such as methadone or buprenorphine on demand.
  • Over the counter naloxone for all.

It is time for America to get over the “blame the addict” meme and recognize the real facts surrounding opioid poisoning deaths. Only by acknowledging and accepting the truth can we develop effective strategies to stem the tide of the polydrug poisoning epidemic we are currently experiencing.

The views and opinions expressed in this article are those of the authors and do not necessarily reflect the official policy or position of Rehabs.com. We do believe in healthy dialogue on all topics and we welcome the opinions of our professional contributors.

What Are Your Thoughts on this Topic?

  • http://AddictionMyth.com/ AddictionMyth

    OK wow lots of good stuff here and thank you again for some real science not lies and propaganda. Overdose deaths are not caused by ‘addiction’ and people who abuse drugs should not be sent to ‘immediate treatment’ as Dr Andrew Kolodny of Phoenix House screams. However you will not solve the deaths problem until you understand that these are intentional suicides not accidental overdoses. Therefore, publicizing the means of mortal overdose will only make the problem worse. Instead, we must inform the public that addiction isn’t a disease and therefore overdose deaths are intentional. You can’t use the ‘I had a drug problem’ excuse. As well the 12 Step cults that bully people into suicide must be exposed. And government must stop teaching children that if they use drugs they may become addicted according to the ‘swiss cheese model’ and die. Yes this is actual official NIDA propaganda. Kids believe it!

    Please think carefully before making huge public policy recommendations.

  • http://www.addictioncapetown.blogspot.com/ Shaun Shelly

    This is a great article Kenneth. It resolves a lot of the questions I have had about the reports out of the US. I don’t understand why this is labelled controversial? Surely it makes a lot of sense and you have provided the data to back it up.

  • http://aarmedwithfacts.wordpress.com/ Juliet.roxspin

    Thank you as always Kenneth for your straight-forward, no holding back opinions based on such excellent facts. Many have been lost to prescribed remedies. I can’t help but think of my own mother, actually, and never will know what she was on! Continue this great work and for the lack of a better phrase, bless you for all the good you do!