Table 3.
Secondary analyses of adjusted HRs of major adverse cardiovascular events estimated by Andersen-Gill model given for overall and cardiovascular risk patients
| Overall patients | Cardiovascular risk patients | |||||||
| HR | LCL | UCL | P-value | HR | LCL | UCL | P-value | |
| csDMARD as reference category | ||||||||
| Reference: csDMARDs | 1.00 | 1.00 | ||||||
| JAKi | 0.73 | 0.46 | 1.18 | 0.20 | 0.82 | 0.50 | 1.34 | 0.44 |
| TNFi | 0.65 | 0.40 | 1.05 | 0.08 | 0.61 | 0.36 | 1.03 | 0.06 |
| Other bDMARDs | 0.56 | 0.35 | 0.88 | 0.01 | 0.62 | 0.39 | 1.00 | 0.048 |
| (II) Adjustment for covariables | ||||||||
| Reference: TNFi | 1.00 | 1.00 | ||||||
| JAKi | 0.84 | 0.47 | 1.48 | 0.54 | 0.71 | 0.38 | 1.31 | 0.27 |
| Other bDMARDs | 0.74 | 0.44 | 1.24 | 0.25 | 0.74 | 0.43 | 1.26 | 0.26 |
| csDMARDs | 1.28 | 0.78 | 2.08 | 0.33 | 1.19 | 0.72 | 1.97 | 0.50 |
| Treatment episode start≥2021 (vs.<2021) | 0.79 | 0.40 | 1.54 | 0.49 | 0.89 | 0.46 | 1.75 | 0.74 |
| Number of prior csDMARDs | 1.19 | 1.02 | 1.40 | 0.03 | 1.24 | 1.05 | 1.46 | 0.01 |
| Number of prior TNFi | 1.00 | 0.77 | 1.29 | 0.99 | 0.92 | 0.71 | 1.20 | 0.55 |
| Number of prior other b/tsDMARDs | 1.14 | 0.98 | 1.32 | 0.09 | 1.18 | 1.00 | 1.38 | 0.049 |
| Corticosteroid dose<10 mg/d (vs 0 mg/d) | 1.39 | 0.95 | 2.04 | 0.09 | 1.29 | 0.86 | 1.95 | 0.22 |
| Corticosteroid dose≥10 mg/d (vs 0 mg/d) | 1.56 | 0.85 | 2.86 | 0.15 | 1.35 | 0.72 | 2.53 | 0.35 |
| Age 50–64 years (vs.<50 years) | 1.93 | 0.73 | 5.14 | 0.19 | ||||
| Age 65–74 years (vs.<50 years) | 3.21 | 1.17 | 8.85 | 0.02 | 1.65 | 1.05 | 2.58 | 0.03 |
| Age≥75 years (vs.<50 years) | 6.78 | 2.43 | 18.95 | < 0.01 | 3.40 | 2.08 | 5.55 | < 0.01 |
| Male sex (vs female sex) | 1.74 | 1.19 | 2.55 | < 0.01 | 1.73 | 1.16 | 2.59 | 0.01 |
| Log CRP (per point) | 1.21 | 1.03 | 1.43 | 0.02 | 1.23 | 1.03 | 1.46 | 0.02 |
| % of full physical function (per 10%) | 0.96 | 0.87 | 1.06 | 0.43 | 0.94 | 0.85 | 1.04 | 0.22 |
| CDAI≥22 (vs.<22) | 0.95 | 0.64 | 1.42 | 0.81 | 0.88 | 0.59 | 1.31 | 0.53 |
| Number of cardiovascular comorbidities* | 1.58 | 1.38 | 1.80 | < 0.01 | 1.45 | 1.23 | 1.70 | < 0.01 |
| Current smokers (vs non-smokers)† | 2.12 | 1.39 | 3.24 | < 0.01 | 0.94 | 0.59 | 1.49 | 0.78 |
| Enrolling institution: rheumatology clinic (vs private praxis) | 1.24 | 0.80 | 1.93 | 0.33 | 1.30 | 0.82 | 2.06 | 0.27 |
| (III) Cox proportional hazard model‡ | ||||||||
| Reference: TNFi | 1.00 | 1.00 | ||||||
| JAKi | 0.96 | 0.49 | 1.91 | 0.92 | 0.79 | 0.38 | 1.64 | 0.54 |
| Other bDMARDs | 0.72 | 0.35 | 1.51 | 0.38 | 0.82 | 0.40 | 1.71 | 0.61 |
| csDMARDs | 1.82 | 1.00 | 3.29 | 0.049 | 1.55 | 0.85 | 2.82 | 0.16 |
In the Andersen-Gill model inverse probability weighting was applied and sensitivity analysis (I) was additionally adjusted for prior DMARD treatments. In sensitivity analysis (II), all covariates used of inverse probability of treatment weights were additionally added to the regression model.
*Comprises hypertension, coronary heart disease, diabetes mellitus, hyperlipoproteinemia, stroke and obesity.
†Comprises never smokers, previous smokers and patients with missing smoking status.
‡Patients were censored at first event or therapy switch.
bDMARD, biologic disease-modifying antirheumatic drug; CDAI, Clinical Disease Activity Index; CRP, C reactive protein; csDMARD, conventional synthetic disease-modifying antirheumatic drug; JAKi, Janus kinase inhibitor; LCL, lower confidence limit; RR, relative risk; TNFi, tumour necrosis factor inhibitor; UCL, upper confidence limit.