Background & Aims
Shoulder injuries are a prevalent form of musculoskeletal disorders and a common reason to seek healthcare. Health sytem level care utilization patterns for shoulder disorders have not been well described. Knowledge of health care use across different diagnostic categories is important because it informs whether ongoing management of shoulder pain is aligned with best practice recommendations. Therefore, this study of beneficiaries of the US Military Health System used routinely collected health data to describe a) usage of pharmacologic and non-pharmacologic treatments following a shoulder injury and b) determine if observed usage for shoulder injury varied whether care was delivered in a military or civilian health system.
Methods
This cohort included patients with an index outpatient shoulder disorder visit between 07/2014 and 04/2019. Patients were required to have no shoulder care in the prior 6 months and retain health system eligibility for 6-month pre- and 3-month post-index visits. Patients were excluded if the shoulder disorder was related to trauma, fracture, or amputation. Patients were then grouped into diagnostic categories: rotator cuff/sub-acromial joint syndrome (rotator cuff), acromioclavicular joint dysfunction (AC joint), glenohumeral instability (instability/dislocation), glenohumeral hypomobility, osteoarthritis (OA), non-specific shoulder disorders, and multiple diagnoses. Healthcare use was broadly classified into pharmacological (e.g. NSAIDS, opioids) and non-pharmacologic (e.g. physical therapy, acupuncture). The number and percentage of patients who received care at least once were summarized by diagnostic category and compared across military and civilian clinics.
Results
Of 1,208,473 adults with new shoulder care episodes during this time, 456,440 individuals met final cohort inclusion. The diagnostic categories ranged from non-specific shoulder pain (n=305,895, 67.0%) to rotator cuff (n = 84,731, 18.6%) to AC joint disorders (n = 4,013, < 0.1%). Overall NSAID prescription was 10.4%, ranging from 6.2% (AC joint) to 13.9% (OA and multiple diagnoses). Opioid prescription was 8.4%, ranging from 5.0% (instability/dislocation) to 11.3% (multiple diagnoses). Steroid injection was 10.2%, ranging from 3.0% (instability/dislocation) to 38.2% (multiple diagnoses). Physical therapy use was 31.0%, ranging from 20.5% (OA) to 56.1% (glenohumeral hypomobility). Acupuncture prescriptions use was only 0.4% for the cohort. Care patterns were congruent across military and civilian clinics, with notable exceptions being higher steroid injection in civilian clinics, while military clinics had higher physical therapy utilization.
Conclusions
Shoulder injuries that received care in the Military Health System appear to largely follow best practice recommendations for pain management. That is, in this cohort, there were relatively low prescription rates for opioids and steroid injections (each ? 10%), while prescriptions of certain non-pharmacologic options, like physical therapy, were higher in military clinics. The observed variability coincided with what is appropriate for a given diagnostic category. However, there were some areas where observed variability indicates opportunities for improvement (e.g. increased prescription of acupuncture). Finally, the differences between military and civilian clinics provide circumstantial evidence for system level factors that may facilitate care inconsistent with best practice recommendations (e.g. in civilian clinics higher reimbursement leading to greater use of steroid injections and copayments leading to lower use of physical therapy).
References
Ria E Hopkins, Gabrielle Campbell, Louisa Degenhardt, Suzanne Nielsen, Fiona Blyth, Milton Cohen, Natasa Gisev. Use of pharmacological and nonpharmacological treatments for chronic noncancer pain among people using opioids: a longitudinal cohort study. Pain 2022 Jun 1;163(6):1049-1059.
Hwansoo Joo, Yoon Jae Lee, Joon-Shik Shin, Jinho Lee, Me-Riong Kim, Wonil Koh, Yeoncheol Park, Yun Kyung Song, Jae-Heung Cho, In-Hyuk Ha. Medical service use and usual care of common shoulder disorders in Korea: a cross-sectional study using the Health Insurance Review and Assessment Service National Patient Sample. BMJ Open. 2017 Jul 12;7(7):e015848.
Daniel I Rhon, Tina A Greenlee, Julie M Fritz. The Influence of a Guideline-Concordant Stepped Care Approach on Downstream Health Care Utilization in Patients with Spine and Shoulder Pain. Pain Med. 2019 Mar 1;20(3):476-485.
Presenting Author
Steven George
Poster Authors
Steven George
PT, PhD
Duke University
Lead Author
Sarah Morton-Oswald MB
Duke University
Lead Author
Hui-Jie Lee PHD
Duke University
Lead Author
Maggie Horn DPT
PHD
Duke University
Lead Author
Nrupen Bhavsar PHD
Duke University
Lead Author
Daniel Rhon
Uniformed Services University
Lead Author
Topics
- Epidemiology