Table 3. Adjusted log-binomial risk factors associated with 48 deaths among 288 persons with TB–COVID-19 in 2020, 24 US jurisdictions*.

Characteristic	No. (%) deaths‡	uPR (95% CI)	aPR (95% CI)
Timing of TB and COVID-19 diagnoses†
>90 d	9 (8.6)	Referent	Referent
<90 d	39 (21.3)	2.5 (1.3‒4.9)	2.3 (1.1‒4.8)
Age at TB diagnosis, y
<44	<5	Referent	Referent
45‒64	16 (18.0)	6.7 (2.0‒22.3)	5.6 (1.6‒19.8)
65‒74	11 (23.4)	8.7 (2.6‒29.9)	8.6 (2.4‒31.3)
75‒85	13 (40.6)	15.2 (4.6‒50.0)	12.6 (3.5‒45.7)
>85	5 (62.5)	23.3 (6.8‒80.5)	25.0 (6.9‒91.1)
Immunocompromising condition, non-HIV§
No	39 (14.6)	Referent	Referent
Yes	9 (45.0)	3.1 (1.8‒5.4)	2.7 (1.1‒6.4)

*Model adjusted for other variables in the table. NTSS outcomes updated as of September 23, 2022. Five with missing outcome; there were no missing values for days between diagnoses, age, or immunocompromising condition (non-HIV). Puerto Rico and Los Angeles County excluded (remainder of California included) because of incompleteness of outcomes data. aPR, adjusted prevalence ratio; TB, tuberculosis; TB–COVID-19, diagnosed with both TB and COVID-19 within 180 days; uPR, unadjusted prevalence ratio. †Days between TB diagnosis and COVID-19 diagnosis, regardless of which was diagnosed first. †Percentage is the number of persons with TB–COVID-19 with the characteristic who died divided by the total number of persons with TB–COVID-19 with that characteristic. §Non-HIV immunocompromising condition attributable to medical conditions, such as hematologic or reticuloendothelial malignancies (e.g., leukemia, Hodgkin’s lymphoma, carcinoma of the head or neck), or immunosuppressive therapy, such as prolonged use of high-dose adrenocorticosteriods (e.g., prednisone), but not including immunocompromising condition related to HIV infection.