Table 3.

Hazard ratios estimating the relative risk of death from COVID-19 associated with baseline hormone concentration.^§

Model	Q1 (lowest)	Q2	Q3	Q4	Q5 (highest)	P-value (term)
Total testosterone (nmol/L)
Events per quintile	112	94	86	68	78
n per quintile	31 992	32 169	32 081	31 933	31 789
Model 1#	1.37 (1.07–1.75)	1.08 (0.85–1.41)	0.93 (0.78–1.14)	0.87 (0.79–0.98)	ref.	<0.001
Model 2##	0.84 (0.65–1.12)	0.82 (0.63–1.10)	0.80 (0.66–1.00)	0.82 (0.75–0.93)	ref.	0.008
SHBG (nmol/L)
Events per quintile	76	79	87	86	110
n per quintile	31 980	32 121	32 054	31 982	31 827
Model 1#	1.61 (1.23–2.07)	1.27 (0.98–1.64)	1.06 (0.86–1.30)	0.93 (0.85–1.02)	ref.	<0.001
Model 2##	1.01 (0.77–1.34)	0.94 (0.72–1.24)	0.89 (0.71–1.11)	0.87 (0.79–0.97)	ref.	0.008
cFT (pmol/L)
Events per quintile	137	110	76	66	49
n per quintile	31 738	32 051	32 126	32 073	31 976
Model 1#	1.13 (0.86–1.5)	0.93 (0.7–1.24)	0.89 (0.73–1.12)	0.91 (0.84–1.09)	ref.	0.004
Model 2##	0.86 (0.65–1.17)	0.81 (0.61–1.10)	0.83 (0.68–1.07)	0.89 (0.82–1.06)	ref.	0.248

^§Hazard ratios calculated for the medians of testosterone within each sample quintile (Q1–Q5), relative to the median for Q5. Quintile boundaries: testosterone: (nmol/L) Q1/2 9.0, Q2/3 10.8, Q3/4 12.5, and Q4/5 14.8 or (ng/dL) Q1/2 259, Q2/3 311, Q3/4 360, and Q4/5 427; SHBG: (nmol/L) Q1/2 25.8, Q2/3 33.1, Q3/4 40.5, and Q4/5 50.8; cFT: (pmol/L) Q1/2 171, Q2/3 201, Q3/4 230, and Q4/5 268. ^#Model 1 included terms for testosterone and age and region, with time modelled as a 3-level stratification factor plus an interaction of region with time (see Methods). ^##Model 2 included model 1 terms + ethnicity (White vs not White), living with partner, educational attainment, alcohol consumption, smoking status, diet (red meat: high vs low vs none), physical activity, BMI, waist circumference, cholesterol, time blood sample collected, blood type, Townsend Index quintile, diabetes, hypertension, angina, atrial fibrillation, COPD, renal impairment, liver disease, thyroid disease, and use of lipid medications (a proxy for hyperlipidemia), glucocorticoids, opioids, and anticonvulsants, with the number of medications included as a proxy for overall comorbidity status. Continuous variables are modelled using restricted cubic splines (see Methods).