Table 3:

Robust Logistic Regression of Non-hormone Injection (i.e., “fillers”) Use on Participant Sociodemographics, Social Networks, and Internet Use (n = 249)

	Non-Hormone Injections
	Adj. OR	95% CI	Sig.
Social Networks and Internet Use
Hormone Use by Network Alters (1 unit increase = 10% increase)	1.18	0.86;1.60	p = .301
Use of Internet to Find Trans Friends	1.43	0.65;3.16	p = .375
Age (in years)
	1.08	1.05;1.11	p < .001
Racial/Ethnic Identity
African American/Black	ref. cat.
Hispanic/Latina	4.34	1.54;12.26	p = .006
Non-Black/Non-Hispanic	1.02	0.31;3.35	p = .970
HIV Status
HIV Positive	0.85	0.38;1.92	p = .694
Homelessness
Yes	1.58	0.39;6.43	p = .521
Health Insurance
None	1.21	0.49;3.00	p = .688
Income
Greater than the Sample Average	1.46	0.71;2.99	p = .305
Sex Work as Main Source of Income	2.11	0.99;4.54	p = .054
Social Network Variables Interaction Effect
Hormone Use by Network Alters * Use of Internet to Find Trans Friends	0.74	0.48;1.14	p = .176

Wald χ²₍₂₂₎ = 38.28; p < 0.0001; Pseudo R² = 0.2033