By Ed Coghlan
There is blistering criticism of a Vanderbilt University study released this week about opioid use.
The report said patients who use opioids had a 64% higher risk of dying within six months of starting treatment compared to patients taking other prescription pain medicine.
The study centered around 45,000 Medicaid patients in Tennessee over a 13 year period ending in 2012.It’s the study and the patients that they surveyed that has many crying foul.
One is Terri Lewis, PhD, who studies the health care system in the United States. Dr. Lewis is also a resident of Tennessee, is the mother of an intractable chronic pain patient who lives there and understands the state’s system very well.
She believes there are obvious limitations to the Vanderbilt study and she outlined a number of them.
Here’s her rebuttal:
- This was a retrospective study that relied on data coded on death certificates to ascertain all causes of mortality (coroners don’t have to be medical professionals in TN) – there is no indication that autopsies were conducted to ascertain the relationship of coded cause of death to preexisting conditions for which treatment was addressed.
- The data between two groups (opioids and others) were tracked over a number of days. Right off the top, there is variability introduced by unknown conditions associated with chronic pain, the course of illness is unknown, the limitations of the medical health care provided, the limitations of the information that was collected, and the detail on codes that were billed for.
- Prescription of long-acting opioids for chronic non-cancer pain, compared with anticonvulsants or cyclic antidepressants, was associated with a significantly increased risk of all-cause mortality, including deaths from causes other than overdose, with a modest absolute risk difference. ‘Association’ is not correlation or causation. One cannot conclude, based on the limitations, that there is any real relationship at all that is meaningful from which conclusions should be drawn about treatment decisions that apply to either this population or the population at large.
- Cardiac death is the leading cause of death in Tennessee. It takes years to develop and occurs due to poverty, poor diet, community acquired stress, smoking, exposures to a a variety of illness.
Dr. Lewis concluded: “Claiming that Opioids contribute to the COD seems a little over the top to me. While one may make the claim that in this particular cohort, opiates are more frequently associated with deaths among those who died with heart disease, one cannot make the claim that opiates contribute in any way to heart disease as a cause of death. The information simply doesn’t exist to support this claim. One could just as easily conclude that opioids were prescribed to treat pain associated with heart disease.”
This is a perfect example of the need to understand the limitations of health data when evaluating the soundness of a publication.
Dr. Lewis also sent links to some data to support her point of view:
Leading causes of death in Tennessee (link)
To see who is eligible for Tennessee Medicaid, click here.