Skip to main content

Table 1 Association between patient- and physician-level characteristics and having a spinal X-ray or CT/MRI within 6 months after an index visit for non-persistent lower back pain in Alberta between April 1st, 2011 and March 31st, 2012 (N = 97,740)

From: Measuring the frequency and variation of unnecessary care across Canada

  X-ray CT/MRI
Characteristic n Rate per 100a Odds Ratio (95% CI)b Rate per 100a Odds Ratio (95% CI)b
Patient-level
Age
 18–44 [ref] 49,623 24.6 1.00 3.2 1.00
 45–64 35,736 31.7 1.37 (1.32–1.41)* 5.8 1.66 (1.55–1.77)*
 65–84 11,063 39.8 1.92 (1.83–2.01)* 6.7 1.85 (1.69–2.02)*
 85+ 1318 39.9 1.98 (1.76–2.22)* 3.6 0.97 (0.71–1.31)
Sex
 Male [ref] 46,167 29.2 1.00 5.0 1.00
 Female 51,573 29.0 0.96 (0.94–0.99)* 4.2 0.81 (0.76–0.86)*
Income Quintile
 1st – lowest [ref] 21,598 28.6 1.00 3.7 1.00
 2nd 21,028 29.1 1.02 (0.98–1.06) 4.4 1.13 (1.03–1.24)*
 3rd 19,732 29.3 1.03 (0.98–1.07) 4.7 1.15 (1.05–1.26)*
 4th 18,969 29.2 1.03 (0.99–1.08) 4.9 1.19 (1.09–1.31)*
 5th – highest 15,625 30.2 1.05 (1.01–1.10)* 5.3 1.25 (1.14–1.38)*
Rurality
 Urban [ref] 80,951 28.7 1.00 4.1 1.00
 Rural 16,275 31.6 1.05 (0.99–1.10) 7.0 1.38 (1.27–1.49)*
Physician-level
Annual LBP patient volume
 50+ [ref] 54,978 27.2 1.00 3.6 1.00
 50 42,762 31.6 1.16 (1.09–1.22)* 5.8 1.44 (1.30–1.58)*
Specialty
 Specialist [ref] 1018 27.7 1.00 10.3 1.00
 General practitioner 96,722 29.1 1.19 (0.97–1.47) 4.5 0.53 (0.41–0.70)*
Compensation model
 Non-FFS 1878 28.8 1.00 6.1 1.00
 FFS 95,862 29.1 1.21 (1.06–1.37)* 4.5 1.02 (0.81–1.29)
  1. Note: * P ≤ .05; CT/MRI = having either a CT or MRI; CI = confidence interval; LBP = lower back pain; FFS = fee-for-service; [ref] = reference category
  2. arate is the observed imaging rate (# tests/n) at index visit + 6 months for each patient in denominator
  3. bobtained via multivariable regression with adjustments for all patient and physician characteristics listed in table (in addition to patient comorbidity indicator and physician years in practice - both statistically insignificant) using GEEs to account for possible clustering within physicians. Patient comorbidity indicator represented whether patient had history of any of the following comorbidities: coronary artery disease, congestive heart failure, atrial fibrillation, other cardiac arrhythmia, cardiac valvular disease, renal disease, previous cardiovascular disease, peripheral vascular disease, venous thromboembolism, COPD, diabetes, hypertension, and/or asthma