Table 4

Several sensitivity analyses to test the robustness of our findings

			Adjusted HR (95% CI)*
Sensitivity analyses	SLE (n)	IR†	Past statin use	Recent statin use	Current statin use	Current statin use	Current statin use
						≤1 year	>1 year
1. Restrict to SLE patients with at least two medical records for SLE	83	0.2	2.11 (0.97 to 4.63)	1.83 (0.75 to 4.46)	0.80 (0.42 to 1.52)	1.22 (0.58 to 2.56)	0.53 (0.24 to 1.16)
2. Restrict to SLE patients who were referred to a rheumatologist or received at last one prescription of azathioprine, cyclophosphamide, cyclosporine or methotrexate and/or received at least two prescriptions of corticosteroids or (hydroxy)chloroquine after the first-time diagnosis of SLE	121	0.3	2.37 (0.98 to 4.77)	1.74 (0.76 to 4.01)	1.19 (0.71 to 2.01)	1.54 (0.81 to 2.93)	1.02 (0.57 to 1.82)
3. Restrict to SLE patients with a minimum of two medical records for diagnosis of SLE at least 2 months apart but within a 2-year span.	44	0.1	3.85 (0.94 to 10.58)	2.72 (0.79 to 9.42)	1.13 (0.46 to 2.78)	1.87 (0.66 to 5.34)	0.69 (0.23 to 2.08)
4. Exclude 2 years after initiation of statin treatment	139	0.5	0.90 (0.50 to 1.63)	0.81 (0.37 to 1.82)	0.38 (0.23 to 0.63)	0.26 (0.13 to 0.53)	0.43 (0.26 to 0.72)
5. Shift the event (SLE) date exactly 2 years before the date of the first-time diagnosis of SLE	140	0.4	1.07 (0.54 to 2.13)	0.61 (0.23 to 1.65)	0.74 (0.46 to 1.20)	0.58 (0.31 to 1.06)	0.84 (0.51 to 1.39)

*Adjusted for age, sex, practice, smoking, cardiovascular diseases, hyperlipidaemia, hypertension, diabetes and use of non-steroid anti-inflammatory drugs.
†Incidence rate is calculated for each sensitivity analysis by dividing the number of events by the person time within each given recency of use.
IR, incidence rate (per 10 000 person-years); SLE, systemic lupus erythematosus.