Another new study on opioids for chronic pain has just been released and the Australian media has picked it up and is running with sensationalistic anti-opioid headlines. Again.
I’ve seen articles today in The Conversation and the Sydney Morning Herald. Predictably, the media, and the authors of the study, are reporting once again that opioids are dangerous and not effective for chronic pain, despite the study data not actually showing that.
Both headlines are alarmist, especially the Conversation, leading with “Opioids ease osteoarthritis pain only slightly. Their deadly risks need to be weighed against any benefit”.
“Deadly risks”. You would think just one look at an opioid and you die!
The Sydney Morning Herald is a little more measured with “Opioids give osteoarthritis sufferers small benefit but greater risks“.
If you only read the headline, you would think that opioids are not effective for osteoarthritis pain. Which is what the study’s authors want you to believe. But you would be wrong. This is fallacious and biased reporting on the part of the researchers and the media.
So, what does the study actually say?
First, let’s start with the study itself. Let’s examine the quality of the evidence. It’s a systemic review/meta-analysis. A systematic review summarizes the existing, published research on a particular topic, in a methodical, rigorous, manner. It’s basically a study of all the existing evidence for a specific question, essentially providing a summary conclusion.
Why do Systemic reviews? Given the enormous amount of research on any given medical topic, it’s impossible for clinicians to keep up with all the latest data or every study. Systemic reviews are meant to review all of the available information, in a very specific way to answer a very specific question.
Which sounds great. But there are several steps in completing a systematic review. These include developing the clinical question, searching for all available literature, study selection, assessment of study quality, data extraction, data analysis, interpreting the results, implications for practice and further research, and finally publishing the review. At any of all these steps biases can come into play. Sometimes the data is cherry-picked to show what the study authors want it to show. Sometimes the study selection itself is biased. Sometimes the authors only select the studies that support the result they want to “prove” and exclude those which don’t. And sometimes the authors only report on the parts of the study that support the intended conclusion and don’t mention the parts that don’t. Systemic reviews are complex and sometimes the review is just not well done.
So let’s look at this study. Every study has limitations and they are generally declared in the study. Immediately I can see that this systemic review is using evidence that is “low” or “very low” quality.
What makes evidence low quality?
Studies can be rated either “high”, “moderate”, “low” or “very low” quality. There is a criterion that researchers use to rate the quality of evidence. It’s called Grading of Recommendations, Assessment, Development and Evaluations or GRADE. Studies are assessed for poor quality design, bias, inconsistency of results, imprecision and publication bias, and rated accordingly.
All of the studies used to draw conclusions in this study on pain relief and quality of life are of low quality. Even worse, the studies used to draw conclusions on safety and serious side effects are of very low quality.
Would you use low or very low-quality evidence to make decisions? If you’re buying a new car and it’s rated as low or very low quality in all its reviews, would you buy it? I doubt it.
More importantly, do you want your doctor treating you based on low or very quality evidence? What if you need heart surgery. Or brain surgery. Do you want the surgeon who uses low-quality evidence to guide their practice?
No, you’d look for the surgeon who only uses high-quality evidence to guide their treatment decisions. That’s how science is supposed to work. You would not want any medical treatment decisions made based on low or very low-quality evidence. That’s pure common sense.
So immediately, this study is worthless. Meaningless. Adds nothing to our understanding of the treatment of the chronic pain of osteoarthritis with opioids. Low or very low-quality evidence means more research is required and low-quality evidence should not be used to draw conclusions that guide treatment decisions.
The cynic in me says the only reason to use low-quality evidence and publish is to get the media headlines, safe in the knowledge that most journalists are unlikely to read the entire study. Journalists will interview the researchers, publish some great quotes and bam! We have more evidence of the perils of opioids. Headlines, clicks, mission accomplished. Everybody wins…except chronic pain patients.
But wait, there’s more. There are many more problems with the design and quality of this study, not to mention the way the authors have represented their findings to the media.
One of the authors of the study, Dr Abdel Shaheed, is reported in the Sydney Morning Herald as saying the adverse events in this study were “dizziness, nausea, vomiting, as well as sedation.” Yet she continues on to say:
“We also know the risk of long-term dependency, overdose and death are also factors in opioid medicine, especially with long-release forms which are often prescribed for osteoarthritis patients.”
Her comments were extremely pejorative, meant to imply that this study found dependency, overdose, and death were common adverse events. In truth, the study showed the opposite. There were NO instances of long-term dependency, no overdoses, and no deaths reported in any of the studies. Which is inconvenient, but Dr Shaheed still found a way to insert them into the media reporting and imply they were a major issues found in the study. This is extremely deceptive and unethical. Her biases are clearly showing.
A more factual conclusion to draw from this study would be that long-term dependence, overdose and death are exceedingly rare, because they did not occur at all in this systemic review and meta-analysis. But that does not suit the authors, and the media’s, sensationalist, anti-opioid narrative, so this very obvious conclusion is omitted.
If you read the study and all of the supplementary materials you will find that the actual adverse events that were evaluated were constipation, nausea, vomiting, dry mouth, diarrhoea, fatigue, pruritis (itching), somnolence (sleepiness), dizziness and headache. In the reporting they have replaced “somnolence, which is sleepiness, with “sedation” which sounds much more scary, doesn’t it?
All of these “adverse events” are extremely minor, which points to opioids being safe and well-tolerated. But again, the authors chose not to say that, instead spitting out that old chestnut about addiction, overdose and death. Fear-mongering.
Most alarmingly the study has been picked up by the Medical Journal of Australia (MJA) and published in their “Insight+” magazine. This is an online publication aimed at providing practicing doctors with the most current medical information and studies. The MJA’s twitter profile describes them as “Australia’s leading peer-reviewed general medical journal“.
The Insight+ article’s headline reads: “Stop prescribing opioids for osteoarthritis, say experts”.
Yes, the experts say this, but the evidence does NOT say this. Those that read only the headline will be left with the impression that opioids should not be prescribed for osteoarthritis pain.
For those who actually read the whole article, nowhere will they find mention that the systemic review is based on “low quality: or “very low quality” evidence or that the Insight+ article does not concur with the study’s actual results.
Few doctors are going to read the study itself, and the supplemental information. They do not have time. They trust the publications that are designed to provide the latest evidence and they trust in the integrity of the MJA. But in this case, their trust is vastly misplaced.
Where it really gets infuriating is where it reports on low-dose opioids and their pain-reliving abilities. In this study, pain scores were rated out of one hundred. A reduction in pain scores of anything less than 10 points was rated as a “very small” improvement, 10-19 points was “moderate” improvement and a 30 plus point reduction was rated as a “large” improvement in pain.
The study showed, and the authors reported, that opioids produced only a very small reduction in pain.
In the media articles, Dr says they “discovered the actual benefit of opioid painkillers for treating osteoarthritis specifically was far less than previously thought.”
“Our findings indicate that opioids provide pain relief similar to that of paracetamol and their benefit is almost half that achieved by NSAIDs.”
These quotes are only partially true, because the study also found that opioids in combination with simple analgesics provided excellent pain relief.
From the study:
“…combinations of low dose opioids with simple analgesics may have beneficial synergistic effects. For example, the 95% confidence intervals for the pain relief achieved by low dose codeine with ibuprofen and low dose oxycodone with acetaminophen included large effects (more than 30 points). “
The study found that patients taking combination treatment of low dose opioids and simple analgesics experienced pain improvement scores greater than 30 points, and therefore experience a large improvement in their pain!
Yet nowhere in any of the media reports am I reading this finding. It has been ignored, simply not reported on. The authors of the study have purposefully buried this finding because it does not suit their narrative. The journalists have quoted the authors and not bothered to read the study or the source data.
This is so deceptive as to be criminal to my mind. Unethical does not cover it. A lie of omission is still a lie.
Next problem: This study is being used as evidence that opioids are not safe and effective for long-term chronic pain. But the longest study period was 16 weeks! Once again there is no long-term evidence, so they fill the void with made-up stuff and sensationalistic cliches.
No conclusion can be drawn about long-term improvements or adverse events from this study because this was a short to medium-term study. And yet all articles warn of the dangers of long-term opioid use, and how ineffective they are long term.
More deceptively, the authors and the journalists draw flawed conclusions from the data, ignoring the more obvious conclusions. The study states:
“As many as 25% of people with osteoarthritis who commence opioid analgesic therapy are still taking them one year later.”
The authors imply that this means patients are addicted, despite the studies finding no evidence of dependence.
The authors of this and all studies completely ignore the most logical reason that a patient would continue a therapy – that is, that they are benefitting from that therapy.
In this case, if a patient is still taking opioid pain-relieving medications a year later this would be strong evidence that the opioid pain medications are providing pain relief and/or improved quality of life. If the medication were ineffective, the patient would cease taking the medication. But again, this most logical conclusion is not reported anywhere, in favour of myth building and playing into fear-based anti-opioid rhetoric.
The moral of this story is, once again, that you must read the entire study, and not rely on the headlines nor the conclusions that the researchers or media are pushing. Unfortunately, it’s clear that some researchers are inherently biased and massage their data to support their intended conclusions and they are not above selective reporting to “prove” their point. Cherry-picking the data and only reporting on the part of the study that supports the idea you want to promote, may not be lying exactly, but it is obscuring the truth. In science and in life, most would agree that this is deceptive and unethical.
I’m a strong proponent of science because the scientific method is the best method we have to increase our knowledge. Science is fluid and constantly being updated and built upon. Old ideas are sometimes discarded completely when we learn more. But science can also be manipulated and statistics can be massaged in ways to say whatever a clever statistician would like them to say. As the saying goes there’s “lies, damn lies and statistics”. Science is done by flawed human beings who sometimes have an axe to grind, and a vested interest in a specific outcome. Flawed human beings can be manipulated, or they can just be not very good at their job.
Unfortunately, these cherry—picked conclusions from low and very low-quality studies have revealed no useful data about the effectiveness or safety of opioids for chronic pain. Yet the media is once again reporting far and wide about the dangers of opioids, ignoring the thousands of people for whom these medications are life-changing, even life-saving. Reinforcing the idea that opioids are far too dangerous and do not work for chronic pain.
Thousands of chronic pain patients who have been taking opioids to manage their chronic pain for years or even decades beg to disagree. But patients’ voices are not being heard and every day patients are being force tapered off of safe and effective pain relief, forced to suffer unnecessarily.
It’s media reporting like this that is causing this to happen.
Worse, GPs are reading the publication of this study on websites directed at doctors and may base their treatment decisions on this false reportage. And patients who are in terrible pain will be denied access to a treatment that may be very effective and safe for them.
Thankfully there are many excellent doctors out there who have years of clinical experience with real-world data, who know that opioids are a safe and effective treatment for some patients who live with severe chronic pain. These doctors continue to care for these patients, monitoring them and prescribing appropriate pain relief for severe, daily chronic pain despite ever-increasing pressure not to do so. I am incredibly grateful to these doctors, and I am gratified that you will find some of them in the comments section, decrying the idea that opioids are not safe and effective.
It’s a shame that the journalists and publications involved didn’t read the actual study or the supplementary data. Had they done so they would have realised that they are reporting the same old lies and exaggerations that are not in any way supported by the data in this study. By trusting the study authors, who clearly wanted to show that opioids are ineffective and unsafe, they have done a huge disservice to chronic pain patients, particularly those who suffer the very severe pain of osteoarthritis, and for which there are few effective treatments.
These articles will contribute to more people being denied a safe, effective treatment for their pain, and greater stigma against chronic pain patients who use opioid pain relieving medications. This should weigh heavily on the journalist’s conscience. Journalistic integrity should demand that the facts be checked, that the actual study be read. But apparently, a click-bait headline is of much greater importance.
And no where does it say the Panadol long term rots away your liver and eventually wnd up in icu with liver failure. Like happened to me.
Long term opaite therapy for me 15 years has never had me end up in hospital, but is the worlds longest and most effective pain killer and is one of the safest options for relieving pain.
I’m so sorry that happened to you, Dean. I hope you’re doing Ok now. But yes, good ol’ panadol is not ‘safe’ either. Long term it can cause liver damage and many people overdose on it every year. I will write an article on it, thanks for pointing this out!
When I initially commented I clicked the “Notify me when new comments are added” checkbox and now each time a comment is added I get four emails with the same comment. Is there any way you can remove people from that service? Thanks!
I’ll look into it and figure out how to unsub you, sorry bout that!
Hello there! Would you mind if I share your blog with my twitter group? There’s a lot of people that I think would really enjoy your content. Please let me know. Thanks
Yes, please do share, I’d appreciate that! thanks 🙂