The Influence of a Guideline-Concordant Stepped Care Approach on Downstream Health Care Utilization in Patients with Spine and Shoulder Pain.

BACKGROUND

Stepped care approaches are emphasized in guidelines for musculoskeletal pain, recommending less invasive or risky evidence-based intervention, such as manual therapy (MT), before more aggressive interventions such as opioid prescriptions. The order and timing of care can alter recovery trajectories.

OBJECTIVE

To compare one-year downstream health care utilization in patients with spine or shoulder disorders who received only MT vs MT and opioids. The secondary aim was to compare differences based on order and timing of opioids and MT.

DESIGN

Retrospective observational cohort.

METHODS

Patients with an initial consultation for a spine or shoulder disorder who received at least one visit for MT were included. Person-level data from the Military Health System Management and Reporting Tool (M2) database were aggregated by a senior health care analyst at Madigan Army Medical Center. Groups were created based on the order and timing of interventions provided. Outcomes included health care utilization (medical costs and visits) over the year following initial consultation. Control measures included metabolic, mental health, chronic pain, sleep, and substance abuse comorbidities, as well as prior opioid prescriptions. Generalized linear models with gamma log links were run due to the heavily skewed nature of cost data.

RESULTS

From 1,876 unique patients with spine or shoulder disorders receiving MT, 1,162 (61.9%) also received prescription opioids. Mean one-year costs in the MT-only group ($5,410, 95% confidence interval [CI] = $5,109 to $5,730) were significantly lower than in the MT+opioid group ($10,498, 95% CI = $10,043 to $10,973). When patients had both treatments, mean one-year costs in the MT-first ($10,782, 95% CI = $10,050 to $11,567) were significantly lower (P = 0.030) than opioid-first ($11,938, 95% CI = $11,272 to $12,643), and MT-first had a significantly lower mean days' supply of opioids (34.2 vs 70.9, P < 0.001) and mean number of unique opioid prescriptions (3.1 vs 6.5, P < 0.001).

CONCLUSIONS

MT alone resulted in lower downstream costs than with opioid prescriptions. Both the order of treatment (MT before opioid prescriptions) and the timing of treatment (MT < 30 days) resulted in a significant reduction of resources (costs, visits, and opioid utilization) in the year after initial consultation. Clinicians should consider the implications of first-choice decisions and the timing of care for treatment choices utilized for patients with spine and shoulder disorders.