RA, n = 22,624 | General Population, n = 22,579 | OR (95% CI) ††, RA vs controls | aOR (95% CI) ‡, RA vs controls | |
---|---|---|---|---|
Compliance per patient*, %, mean (SD) | 72.3 (37) † | 70.4 (38) † | ||
Compliance per period**, n (%) | 34,804/48,724 (71.4) | 36,067/51,081 (70.6) | 1.05 (1.02–1.09) § | 1.05 (1.02–1.09) §§ |
Compliance per period before 2001‡‡, n (%) | 10,287/16,958 (60.7) | 11,291/18,294 (61.7) | 0.96 (0.92–1.01) ¶ | 0.89 (0.85–0.94) §§ |
Compliance per period after 2001‡‡, n (%) | 24,519/31,766 (77.2) | 24,776/32,787 (75.6) | 1.10 (1.06–1.15) §§ | 1.10 (1.06–1.15) §§ |
↵* Measured as the mean % of eligible periods per patient with ≥ 1 PG test.
↵** Measured with periods as the unit of analysis (i.e., as the no. periods with ≥ 1 PG test divided by the no. eligible periods in each cohort) and corresponding percentage.
↵† Difference between compliance in RA and general population, statistically significant p < 0.0001, using the Mann-Whitney U test.
↵†† Unadjusted OR from generalized estimating equation (GEE) model estimating the odds of receiving a PG test within a period in the RA cohort relative to general population, measured in separate models for the periods before and after 2001.
↵‡ Adjusted OR from multivariable GEE adjusting for sex, age, and Romano comorbidity score measured at start of period, and start year of period (as a continuous variable), measured in separate models for the periods before and after 2001.
↵‡‡ Compliance per period was stratified according to whether the date at the beginning of the period was before or after January 1, 2001.
↵§ p = 0.0003.
↵§§ p < 0.0001.
↵¶ p = 0.1184. DM: diabetes mellitus; RA: rheumatoid arthritis; aOR: adjusted OR; PG: plasma glucose.