Background/Purpose:

Patients suffering from ankylosing spondylitis (AS) often receive a delayed or mis-diagnosis, as this condition is frequently confused with mechanical back pain from other causes. If AS is not properly identified, patients can receive inadequate care, delaying disease-specific treatment, which may contribute to decreased physical functioning, quality of life, excessive use of narcotic analgesics, and disease progression. This study looks at regional variation in AS detection and examines whether regions that are better able to detect AS also are less likely to provide inappropriate care.

Methods:

A large commercial claims database covering the years 2008 through 2013 was used to measure geographic variation in AS and back pain detection and potentially inappropriate health resource utilization. All adults with 12 months of continuous eligibility in the dataset were included. Measured detection was defined as the prevalence of AS (ICD-9 codes: 720.0) and measured detection for back pain was defined as (720.1-720-.9, 721.x, 722.x, 723.0-723.8, 724.x, 739.1x-739.4x, 846.x, 847.x). We measured the share of patients with back pain who used narcotic analgesics and corticosteroid injections as a proxy for unnecessary healthcare utilization, as well as medical and pharmacy costs within each metropolitan statistical area (MSA) using generalized linear models adjusting for age, gender, and comorbidities. Geographic variation across MSAs was evaluated using standard deviation (SD), interquartile range (IQR) and coefficient of variation (CV).

Results:

In the data, 21,215,151 patients were diagnosed with back pain or AS.  Nationally, 37 per 100,000 (0.037%, SD: 0.023%, IQR: 0.024%-0.043%) individuals were diagnosed with AS.  A majority of MSAs (74.9%, 293 of 391 MSAs) had AS measured detection below the average in the data, and all MSAs (391 of 391) had AS measured detection below the 0.55% US AS prevalence estimate from the 2009-2010 National Health and Nutrition Examination Survey (NHANES).  The CV for AS measured detection (0.631) was almost 2.5 times higher than the CV for measured detection of back pain alone (0.256).  Based on our regression model, moving from an MSA at the 10^th percentile of AS measured detection to one at the 90^thpercentile would decrease the share of patients using narcotic analgesics (-1.0%) and steroids (-7.9%), but would increase spending on disease-modifying antirheumatic drugs (+14.4%).  The net effect would be a decrease in both total pharmacy costs (-7.1%) and total costs (-2.1%).

Conclusion:

Detection of AS varies dramatically across MSAs in the US and is significantly below prevalence estimates from large national surveys. Further, areas with higher rates of AS measured detection were less likely to use narcotic analgesics and steroids, suggesting that patients with back pain from AS who are not accurately diagnosed with AS may be receiving inadequate care. Increased patient and provider awareness regarding the treatment of inflammatory conditions—such as AS—that cause back pain is needed to ensure appropriate and timely care.

« Back to 2015 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/regional-variation-in-measured-detection-of-ankylosing-spondylitis/