Table 5.

Results of all-cause mortality in patients with rheumatoid arthritis following hospitalisation for pneumonia according to disease activity and treatment with prednisolone

	No prednisolone +low CRP	No prednisolone +high CRP	Prednisolone treatment +low CRP	Prednisolone treatment +high CRP
No of patients (N)	64	310	42	288
All-cause mortality at 30 days, n (%)	4 (6.3)	41 (13.2)	1 (2.4)	43 (14.9)
Crude HR for 30-day mortality	1 (ref)	2.20 (0.78 to 6.13)	0.38 (0.04 to 3.37)	2.48 (0.89 to 6.91)
Adjusted HR for 30-day mortality	1 (ref)	2.08 (0.74 to 5.81)	0.36 (0.04 to 3.23)	2.11 (0.76 to 5.88)
All-cause mortality at 90 days, n (%)	5 (7.8)	56 (18.1)	1 (2.4)	74 (25.7)
Crude HR for 90-day mortality	1 (ref)	2.46 (0.98 to 6.14)	0.29 (0.03 to 2.55)	3.57 (1.44 to 8.83)
Adjusted HR for 90-day mortality	1 (ref)	2.35 (0.94 to 5.87)	0.29 (0.03 to 2.52)	3.09 (1.25 to 7.65)

Data are presented as HRs with 95% CIs for all-cause mortality.

Treatment with prednisolone 0–3 months prior to admission. CRP levels were measured 30–90 days prior to pneumonia admission with a value of 8 mg/L used as a division between low and high CRP levels.

Adjusted for sex, age, level of comorbidity, alcoholism and antibiotic use before admission.

CRP, C reactive protein.