The purpose of this recent investigation was to evaluate the association between chiropractic utilization and use of prescription opioids among older adults with spinal pain … at least this is what the abstract says. The actual paper tells us something a little different: The objective of this investigation was to evaluate the impact of chiropractic utilization upon the use of prescription opioids among Medicare beneficiaries aged 65 plus. That sounds to me much more like trying to find a CAUSAL relationship than an association.
Anyway, the authors conducted a retrospective observational study in which they examined a nationally representative multi-year sample of Medicare claims data, 2012–2016. The study sample included 55,949 Medicare beneficiaries diagnosed with spinal pain, of whom 9,356 were recipients of chiropractic care and 46,593 were non-recipients. They measured the adjusted risk of filling a prescription for an opioid analgesic for up to 365 days following the diagnosis of spinal pain. Using Cox proportional hazards modeling and inverse weighted propensity scoring to account for selection bias, they compared recipients of both primary care and chiropractic to recipients of primary care alone regarding the risk of filling a prescription.
The adjusted risk of filling an opioid prescription within 365 days of initial visit was 56% lower among recipients of chiropractic care as compared to non-recipients (hazard ratio 0.44; 95% confidence interval 0.40–0.49).
The authors concluded that, among older Medicare beneficiaries with spinal pain, use of chiropractic care is associated with significantly lower risk of filling an opioid prescription.
The way this conclusion is formulated is well in accordance with the data. However, throughout the paper, the authors imply that chiropractic care is the cause of fewer opioid prescriptions. For instance: The observed advantage of early chiropractic care mirrors the results of a prior study on a population of adults aged 18–84. The suggestion is that chiropractic saves patients from taking opioids.
It does not need a lot of fantasy to guess why some people might want to create this impression. I am sure that chiropractors would be delighted if the US public felt that their manipulations were the solution to the opioid crisis. For many months, they have been trying hard enough to pretend this is true. Yet, I know of no convincing data to demonstrate it.
The new investigation thus turns out to be a lamentable piece of pseudo research. Retrospective case-control studies can obviously not establish cause and effect, particularly if they do not even account for the severity of the symptoms or the outcomes of the treatment.