“Krebs Study” Shows Opioids are Safe

The Opioid Information Thread – Inspire.com

The infamous “Krebs study” is the one that came out on March 6 and found opioids no more effective than non-opioids.

A scientifically brilliant forum user at Inspire.com, Seshet, has been analyzing the study and its impact.  He dug around in this pile of horse manure until he found the pony:

The study inadvertently also shows that opioids aren’t nearly as addictive and dangerous as they are made out to be by the same people who ran the study.

Here are excerpts of Seshet’s insightful posts about this study from “The Opioid Information Thread” discussion:  

Posts started on Mar 6, 2018

Vox is reporting on a new study by Erin Krebs and colleagues under the title: “Finally, proof: opioids are no better than Tylenol for treating some chronic pain“.

Excerpts:

“Do opioids help patients with chronic pain in the long run? Are they worth all that risk? The answer, according to her newly published JAMA study, is a resounding “no.””

“Krebs is the lead author on the first randomized trial, comparing chronic pain patients on prescription opioids (like morphine, hydrocodone, and oxycodone) to patients on non-opioid painkillers (like acetaminophen or Tylenol, naproxen, or meloxicam), measuring their pain intensity and function over the course of a year.”

“For the paper, which she co-authored with colleagues in Minnesota and Indiana, she recruited 240 patients at Veteran Affairs primary care clinics who had experienced severe chronic back pain, or hip or knee osteoarthritis, for at least six months.”

This study is getting a lot of play, just as it did in spring 2017 when the results were released (we had a big discussion about it here back then). And just as happened back then, the media coverage is missing a couple of key points:

First, the study only looked at chronic back back and knee or hip osteoarthritis, two conditions that opioids are not generally thought of as good for and NSAIDs are generally thought of as the best choice for. So it’s not surprising that NSAIDs did better.

Second, the study didn’t look at any other forms of pain, in particular those for which NSAIDs are widely thought not to work and opioids are regarded as a potentially useful med (e.g.: small fiber neuropathy).

Also, media reports ignore the fact that not everyone can tolerate NSAIDs (or opioids) for a year, some people have underlying conditions that make one med a better choice than the other, and that back pain and knee osteoarthritis often get better over time (something that doesn’t happen with other disorders).

All of this can be found in a PNN article about the Krebs study from last May:

Opioids vs. NSAIDs for Chronic Pain

Red Lawhern has a really good comment here, so don’t just read the article.

The above article gives a technical critique and points out the many problems with this study.

NPR has a more nuanced view of the Krebs study:
https://www.npr.org/sections/health-shots/2018/03/06/590837914/opioids-dont -beat-other-medications-for-chronic-pain

But they also don’t mention that Krebs’ findings won’t translate to neuropathic pain disorders.
Source: https://www.vox.com/science-and-health/2018/3/6/17082590/opioids-tylenol-ch ronic-pain-study

—  Another post on Mar 6, 2018

The Krebs study is getting a lot of attention. I have a copy of the actual study, and the results are more complicated than media coverage makes them seem.

Two points:

First, the non-opioid arm of the study used a wide variety of non-opioid meds:

“In the nonopioid medication group,

• step 1 was acetaminophen (paracetamol) and nonsteroidal anti-inflammatory drugs (NSAIDs).
• Step 2 included adjuvant oral medications (ie, nortriptyline, amitriptyline, gabapentin) and topical analgesics (ie, capsaicin, lidocaine).
• Step 3 included drugs requiring prior authorization from the VA clinic (ie, pregabalin, duloxetine) and tramadol.“

But tramadol is an opioid. (https://en.wikipedia.org/wiki/Tramadol)

It is a Schedule IV controlled substance that acts as a partial opioid agonist and SNRI. So this may mess with the results of the Krebs study a bit. It’s not clear how much, since the details of prescribing during the study are not given.

Second, the study necessarily excluded certain populations:

“Patients on long-term opioid therapy were excluded. Other reasons for exclusion included contraindications to all drug classes in either group, including class-level opioid contraindications (eg, active substance use disorder), and conditions that could interfere with outcome assessment (eg, life expectancy <12 months).”

This means that the patient population that was used in the study is not necessarily representative of the population in general. Some people may be intolerant to NSAIDs, allergic to acetaminophen, or intolerant to opioids (or any of the many other meds used in the study).

The Krebs study is generally quite good. The observations above weaken the results a bit, but these are common issues in clinical research. It’s the over-interpretation of the study in the media and soon by policymakers and regulators that is the problem.

Krebs study: https://jamanetwork.com/journals/jama/article-abstract/2673971?redirect=tru e

—-  Another post on Mar 6, 2018

The LA Times is reporting on the Krebs study about opioids versus non-opioid for musculoskeletal pain. This excerpt is very interesting:

“However, hospitalizations and emergency room visits to deal with pain medications were similar in both groups, as were rates of drug misuse.”

Lead author Erin Krebs, MD, is quoted in US News as saying: “If they don’t work better than less risky drugs, there’s no reason to use opioids given “their really nasty side effects — death and addiction,” said lead author Dr. Erin Krebs, a physician and researcher with the Minneapolis Veterans Affairs Health Care System.”

(https://www.usnews.com/news/news/articles/2018-03-06/prescription-opioids-f ail-new-rigorous-test-for-chronic-pain)

So where are the addictions?

Here’s the twist: the study seems to show opioids not superior to non-opioids, but it also shows that all these people were using opioids for a whole year and didn’t show signs of addiction.

Inadvertently, I’m sure, the study shows that many of the fears about addiction are unfounded, that pain medications aren’t “heroin pills” and don’t bring out any unusual behavior in patients.

Essentially, the study shows that opioids are safe even as it questions their efficacy for these particular pain syndromes.

A group of people (234 total, about half in the opioid arm and half in the non-opioid arm of the study) was put on opioids for a year and monitored.

According to the study, there was no particular problem with misuse, abuse, or addiction.

According to Krebs in the US News interview, there are nasty side effects like addiction, but this wasn’t seen in her study.

Maybe this suggests that opioids can be used reasonably safely in medical practice. Krebs and her colleagues managed to do it.

I am not suggesting that opioids should be used over less risky medications or that they should be used without due caution and monitoring.

And the Krebs study doesn’t directly evaluate the risks of misuse, abuse, or addiction.

But the study did closely track those things. And it didn’t find any evidence that such problems occurred.

I find it interesting that a “dangerously addictive substance” didn’t cause any problems of misuse, abuse, or addiction in the study subjects during the year of the study.

Source: http://www.latimes.com/science/sciencenow/la-sci-sn-opioid-painkillers-no-b etter-20180306-story.html

—  Another post on Mar 6, 2018

MedPageToday has a look at the Krebs study on opioids versus non-opioids for knee OA and low back pain. In particular, Krebs explains how tramadol (an opioid) got into the non-opioid group…

“Isn’t tramadol an opioid? I asked lead author Dr. Erin Krebs that very question. She reminded me that this trial started in 2010:

“This was before all the concerns about opioid overdose and addiction and back then a big concern was is it ethical to deprive patients of opioids if they fail all these non-opioid medications.“

This clearly implies that nowadays, there is no concern about the ethics of forcing patients to suffer their pain even when relief is available.

“Times have certainly changed. But regardless, only 13 patients in the non-opioid group ever required escalation all the way to tramadol.”

So this sort of makes sense, in particular at an ethical level. The Krebs study was very good about making sure that pain management and quality of life were maintained for all the subjects.

But 13 patients in the non-opioid group getting tramadol represents around 10% of that group. In my view, this weakens the statistical validity of the outcomes in the study.

Moreover, there were lots of concerns in 2010 about overdose and addiction. The CDC declared the epidemic back then. FDA got Purdue to produce an abuse-deterrent formulation of OxyContin. And Krebs herself was motivated to do this study out of concern for the risks of addiction and death when opioids are used.

The Krebs study is still a good study and its results are useful. It’s pretty clear that people with knee OA and idiopathic low back pain shouldn’t preferentially get opioids since there are other good options.

But that’s all the study really says. Everything else is over-interpretation.

Source: https://www.medpagetoday.com/blogs/themethodsman/71556

— Post on Mar 7, 2018

A couple more thoughts on the Krebs study.

Although opioid analgesics did not perform any better than non-opioid analgesics, the people in the opioid-arm of the study remained at the same dose of medication during the study period and they did not experience opioid-induced hyperalgesia.

These are inadvertent findings.

They were not a part of what Krebs and colleagues were specifically trying to measure.

As a result, these are observations and not conclusions of the study.

But that is in fact what happened (or really, didn’t happen). Over 100 people were put on opioid therapy for a year, and none of them showed any signs of dose escalation or opioid-induced hyperalgesia.

The Krebs study provides a very detailed survey of what happens when people are put on opioid therapy.

A lot of what is claimed about dose escalation and opioid-induced hyperalgesia didn’t happen at all. I find this very interesting.

—- Post on Mar 8

The Krebs study is getting even more play today. Here are some of the latest headlines:

“Opioids No Better Than Over-The-Counter Meds for Chronic Pain”
(https://www.geek.com/science/opioids-no-better-than-over-the-counter-meds-f or-chronic-pain-1733200/)

“Opioids No Better Than Over-The-Counter Pain Meds For Chronic Pain”
(http://boston.cbslocal.com/2018/03/07/opioids-painkillers-study-chronic-pai n/)

“For Arthritis Pain, Tylenol Works as Well as Opioids”
(https://www.nytimes.com/2018/03/07/well/live/for-arthritis-pain-tylenol-wor ks-as-well-as-opioids.html)

Some of these articles just say “chronic pain” and don’t mention that the Krebs study looked specifically at knee and hip OA and idiopathic low back pain.

They don’t mention that the non-opioid arm of the study involved a long list of possible meds (including tramadol, which is an opioid).

They don’t discuss the risks of the non-opioid meds.

They don’t mention that some people cannot take NSAIDs (or other classes of meds) because of health problems.

Most importantly, the Krebs study says nothing one way or another about other chronic painful disorders.

For the latest happenings and concerns about our opioid pain medication, I encourage you to keep an eye on the Inspire.com discussion, The Opioid Information Thread, where Seshet posts everything new he finds.