Table 4.

Results of all-cause mortality following hospitalisation for pneumonia: a subanalysis of patients with rheumatoid arthritis (RA) and their preadmission treatments for RA with patients receiving methotrexate as a reference

RA therapy		Mortality at 30 days n (%)	HRs 30-day mortality		Mortality at 90 days n (%)	HRs 90-day mortality
	N		Crude	Adjusted*		Crude	Adjusted*
All	1220	165 (13.5)			243 (19.9)
Methotrexate monotherapy	143	16 (11.2)	1.0 (ref)	1.0 (ref)	23 (16.1)	1.0 (ref)	1.0 (ref)
Any prednisolone†	526	71 (13.5)	1.23 (0.72 to 2.12)	1.16 (0.68 to 2.01)	121 (23.0)	1.48 (0.95 to 2.31)	1.43 (0.91 to 2.22)
Any biologics	46	1 (2.2)	0.19 (0.02 to 1.42)	0.35 (0.05 to 2.66)	1 (2.2)	0.13 (0.02 to 0.94)	0.21 (0.03 to 1.56)
Other csDMARDS as monotherapy	63	8 (12.7)	1.17 (0.50 to 2.73)	1.24 (0.53 to 2.92)	10 (15.9)	1.01 (0.48 to 2.12)	1.09 (0.52 to 2.29)
Combination therapy (csDMARDs)	48	1 (2.0)	0.18 (0.02 to 1.34)	0.27 (0.36 to 2.06)	3 (6.2)	0.37 (0.11 to 1.22)	0.56 (0.17 to 1.87)
No RA medication	394	68 (17.3)	1.60 (0.93 to 2.76)	1.49 (0.86 to 2.57)	85 (21.6)	1.40 (0.89 to 2.23)	1.35 (0.85 to 2.14)

Patients categorised according to therapy registered within 1 year prior to pneumonia admission.

Each patient was assigned to one group using the hierarchy described in the text and shown in online supplementary figure 1.

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

†Prescriptions for prednisolone 0–3 months prior to admission.

csDMARDs, conventional synthetic disease-modifying antirheumatic drugs.