. Author manuscript; available in PMC: 2022 Aug 1.

Published in final edited form as: Cancer Epidemiol Biomarkers Prev. 2021 Dec 3;31(2):479–485. doi: 10.1158/1055-9965.EPI-21-0946

Table 3.

Associations Between Post-Diagnosis Aspirin Use and Quantity and Multiple Myeloma-Specific and All-Cause Mortality.

	NHS		HPFS		Pooled
	N Events/ N at Risk	HR (95% CI)¹	N Events/ N at Risk	HR (95% CI)¹	HR (95% CI)^1,2
Multiple Myeloma-Specific Mortality

Post-Diagnosis Aspirin Use
Non-users	116/140	ref	83/111	ref	ref
User	88/125	0.72 (0.51, 1.02)	34/60	0.45 (0.29, 0.71)	0.61 (0.46, 0.79)
Post-Diagnosis Number of 325-mg Tablets/Week
Non-user	116/140	ref	83/111	ref	ref
1 to <6	58/80	0.75 (0.51, 1.11)	15/19	0.62 (0.34, 1.14)	0.70 (0.51, 0.96)
≥6	18/29	0.58 (0.33, 0.99)	16/34	0.39 (0.22, 0.73)	0.49 (0.33, 0.74)
P-trend³		0.05		0.003	0.0005

All-Cause Mortality

Post-Diagnosis Aspirin Use
Non-users	125/140	ref	105/111	ref	ref
User	100/125	0.70 (0.50, 0.97)	53/60	0.53 (0.36, 0.77)	0.63 (0.49, 0.80)
Post-Diagnosis Number of 325-mg Tablets/Week
Non-user	125/140	ref	105/111	ref	ref
1 to <6	67/80	0.74, 0.51, 1.06)	17/19	0.56 (0.32, 0.98)	0.67 (0.49, 0.90)
≥6	20/29	0.57 (0.34, 0.95)	29/34	0.51 (0.31, 0.83)	0.56 (0.40, 0.80)
P-trend³		0.03		0.008	0.002

Abbreviations: HR: Hazard Ratio; CI: Confidence Interval; NHS: Nurses’ Health Study; HPFS: Health Professionals Follow-Up Study.

Cox proportional hazards model adjusted for age at multiple myeloma diagnosis (years), calendar year of diagnosis (<2000 vs. ≥2000), post-diagnosis body mass index and number of comorbidities, and pre-diagnosis aspirin use (user vs. non-user).

Pooled models were stratified by cohort (sex).

P-values for trend tests modeled as an ordinal variable using the mid-point of each category of the respective variable in Cox proportional hazard models.