Brent Leininger, DC, MS
The Spine Journal
Volume 16, Issue 11, November 2016, Pages 1292–1304
Abstract
Background Context
Chronic neck pain is a prevalent and disabling condition among older adults. Despite the large burden of neck pain, little is known regarding the cost-effectiveness of commonly used treatments.
Purpose
This study aimed to estimate the cost-effectiveness of home exercise and advice (HEA), spinal manipulative therapy (SMT) plus HEA, and supervised rehabilitative exercise (SRE) plus HEA.
Study Design/Setting
Cost-effectiveness analysis conducted alongside a randomized clinical trial (RCT) was performed.
Patient Sample
A total of 241 older adults (≥65 years) with chronic mechanical neck pain comprised the patient sample.
Outcome Measures
The outcome measures were direct and indirect costs, neck pain, neck disability, SF-6D-derived quality-adjusted life years (QALYs), and incremental cost-effectiveness ratios (ICERs) over a 1-year time horizon.
Methods
This work was supported by grants from the National Center for Complementary and Integrative Health (#F32AT007507), National Institute of Arthritis and Musculoskeletal and Skin Diseases (#P60AR062799), and Health Resources and Services Administration (#R18HP01425). The RCT is registered at ClinicalTrials.gov (#NCT00269308). A societal perspective was adopted for the primary analysis. A healthcare perspective was adopted as a sensitivity analysis. Cost-effectivenesswas a secondary aim of the RCT which was not powered for differences in costs or QALYs. Differences in costs and clinical outcomes were estimated using generalized estimating equations and linear mixed models, respectively. Cost-effectiveness acceptability curves were calculated to assess the uncertainty surrounding cost-effectiveness estimates.
Results
Total costs for SMT+HEA were 5% lower than HEA (mean difference: −$111; 95% confidence interval [CI] −$1,354 to $899) and 47% lower than SRE+HEA (mean difference: −$1,932; 95% CI −$2,796 to −$1,097). SMT+HEA also resulted in a greater reduction of neck pain over the year relative to HEA (0.57; 95% CI 0.23 to 0.92) and SRE+HEA (0.41; 95% CI 0.05 to 0.76). Differences in disability and QALYs favored SMT+HEA. The probability that adding SMT to HEA is cost-effective at willingness to pay thresholds of $50,000 to $200,000 per QALY gained ranges from 0.75 to 0.81. If adopting a health-care perspective, costs for SMT+HEA were 66% higher than HEA (mean difference: $515; 95% CI $225 to $1,094), resulting in an ICER of $55,975 per QALY gained.
Conclusion
On average, SMT+HEA resulted in better clinical outcomes and lower total societal costs relative to SRE+HEA and HEA alone, with a 0.75 to 0.81 probability of cost-effectiveness for willingness to pay thresholds of $50,000 to $200,000 per QALY.