The United States spends more money on the care of back and neck pain than any other health condition. Despite this, the cost-effectiveness for many recommended treatments is unclear. Our primary objective for this project was to estimate the cost-effectiveness of spinal manipulative therapy (SMT), supervised exercise therapy (ET), and home exercise and advice (HEA) for spinal pain in the U.S.
The researchers analyzed cost and clinical outcome data from eight randomized trials conducted in the U.S. using an individual participant data meta-analysis approach. They calculated cost-effectiveness from the societal and healthcare perspective of various comparisons between SMT, ET, and HEA. Incremental cost-effectiveness ratios (ICERs) were calculated using quality-adjusted life years as the main outcome.
The 8 trials included a total of 1803 participants and 1488 (83%) provided complete data. Incremental cost-effectiveness ratios and probabilities of cost-effectiveness varied substantially between studies; thus, the reseaarchers did not conduct meta-analysis and report findings from individual trials.
Cost-effectiveness findings were favorable for SMT compared to HEA for acute neck pain (ICERs below $50k/QALY) and when added to HEA for chronic back-related leg pain and chronic neck pain in older adults (better outcomes and lower costs). However, SMT was not likely cost-effective compared to HEA for chronic back pain in adults or when added to HEA for older adults (higher costs and worse outcomes).
Findings for SMT were favorable when compared to ET in adults with chronic back pain and when added to ET for chronic neck pain in adults (better outcomes and lower costs) and chronic back pain in adolescents (ICERs below $50k/QALY). However, SMT is not likely cost-effective when compared to ET for chronic neck pain in adults (ICERs below $70k/QALY for exercise) and findings were inconsistent across outcomes in older adults with chronic back pain.
Finally, ET may be cost-effective compared to HEA for adults with chronic neck pain (ICERs largely between $100-$200k/QALY), but not for chronic back pain or when added to HEA for older adults with chronic neck or back pain (higher costs and worse outcomes).
The authors concluded that overall based on willingness to pay thresholds of $50-$200k/QALY, there was moderate to high probability that spinal manipulation is cost-effective relative to HEA for neck pain and back-related leg pain, but not for chronic back pain. There was also moderate to high probability spinal manipulation was cost-effective relative to exercise therapy for chronic back pain but findings were mixed for neck pain and more favorable in older adults. Cost-effectiveness findings for exercise therapy were mostly not favorable relative to less intensive home exercise programs as costs were higher, and outcomes were often worse.
The authors admit that their analyses have several limitations: Randomized clinical trials are often designed to detect important differences in disease-specific clinical outcomes that are most likely to be impacted by the treatments assessed (e.g., pain severity, disability). Important measures for assessing cost-effectiveness include general health outcomes like changes in QALYs, healthcare use, and missed work. These measures were collected alongside disease-specific measures, but the trials were not powered to detect important differences in cost-effectiveness outcomes. Participants self-reported their use of healthcare and medications along with number of missed workdays. We did not have access to administrative data for healthcare use or costs. While access to administrative data would have reduced potential measurement error for these variables, it is not without limitations due to the high variability in coverage and re-imbursement policies for healthcare procedures across insurance products in the U.S. Costs for reduced productivity due to spinal pain included missed work in and outside of the home, but costs due to reduced productivity while still at work (i.e., presenteeism) were not included. This is an important limitation as costs due to reduced productivity while at work consistently account for a large proportion of total costs in spinal pain burden of illness studies. Finally, all studies were conducted in the U.S. with resources valued using U.S. prices and findings are not likely generalizable to populations or healthcare systems in other countries.
The authors stress that additional studies are needed to assess the cost-effectiveness of these approaches relative to medical care, the most common treatment approach in the US , as well as other guideline recommended treatments such as massage, acupuncture, mindfulness-based stress reduction, tai chi, yoga, and cognitive behavioral therapy
In view of these limitations and the fact that just 8 trials could be included, the relatively firm comclusions are surprising, in my view. To me, much of the data look unconvincing, somewhat random, inconsistent and implausible. could it be that the authors were trying to generate and emphacize positive results? After all, most of them are affiliated to the “Integrative Health and Wellbeing Research Program Earl E. Bakken Center for Spirituality & Healing, University of Minnesota”!
3 Responses to Cost-effectiveness of spinal manipulation, exercise, and self-management for spinal pain
Leave a Reply
This site uses Akismet to reduce spam. Learn how your comment data is processed.
I have looked at this CE issue for ~ 20 years now and have concluded that cost-effectiveness of alternative treatments is a miasma. A chiropractic association (California?) hired a Harvard economist or 2 to run a cost effectiveness of chiro treatment for low back pain. Firstly, when I requested it, I was told it was proprietary. When I finally found a copy, it was unconvincing. The assumptions were shaky. As I recall, the it was non-surgical low back pain only, the costs of chiropractic visits was assumed to be $20, and the number of treatment visits was very restricted. When considering natural history and placebo effects for acute low back pain, it was entirely plausible that the only cost effective treatment for acute low back pain is the control group, with reassuring phone call or other application of inexpensive psychosocial support.
Researchers need to do more than mere hypothesis testing of treatment versus control – they need to evaluate if the control condition is almost identical in its effectiveness over time. Chiropractors need to be experts in diagnosis and referral, instead they use spinal manipulation for everything. Spinal manipulation is neither necessary nor sufficient for outcomes of low back. Muscles are usually the cause of pain.
I have looked at CE of SCAM not in so much depth, but I agre wholeheartedly!
Michael Menke, your comment “Muscles are usually the cause of pain”, shows an incredible ignorance of conditions. They are referred to as “neuro-musculo-skeletal” problems. If you ignore the nerves and the joints, and only concentrate on the muscles then you are missing two of the three causes of problems.