Table 4

Risk factors for experiencing physical violence from clients in the last 12 months among indoor-working sex workers in London significant in a multivariate analysis

	Physical violence^§						p Value^*
	n/n	%	Unadjusted		Adjusted^¶
	n/n	%	OR	95% CI	OR	95% CI
Demographic indicators
Migrant status
UK-born	26/103	25.2%	1.0		1.0
Eastern Europe	37/152	24.3%	1.0	0.53 to 1.70	0.8	0.37 to 1.86	0.65
Recruitment method
Outside of project^‡	41/181	22.6%	1.0		1.0
Project drop-in/outreach referral^†	22/74	29.7%	1.4	0.78 to 2.65	1.6	0.76 to 3.49	0.21
Sexual risk behaviours
10 or more clients in last 7 days
No	18/90	20.0%	1.0		1.0
Yes	30/97	30.9%	1.8	0.91 to 3.51	1.7	0.80 to 3.55
Don't know/no response	15/68	22.10%	1.1	0.52 to 2.45	1.0	0.41 to 2.43	0.28
Currently has non-paying sex partner
No	20/105	19.1%	1.0		1.0
Yes	41/147	27.9%	1.6	0.90 to 3.01	2.0	1.03 to 3.96	0.04
Adverse health outcomes
Alcohol audit score
Scores <3	30/100	30.0%	1.0		1.0
Scores 3+	27/137	19.7%	0.6	0.31 to 1.04	0.4	0.21 to 0.82	0.01
Used drugs during or before sex work in last 4 weeks
No	45/203	22.2%	1.0		1.0
Yes	17/45	37.8%	2.1	1.07 to 4.24	1.9	0.81 to 4.38	0.14
Ever been arrested or imprisoned
No	43/200	21.5%	1.0		1.0
Yes	19/53	35.8%	2.0	1.06 to 3.93	2.6	1.14 to 5.71	0.02

↵* p Value derived from Wald test.
↵† 10 people were recruited via outreach referrals.
↵‡ This includes cold calling, field workers' personal contacts or acquaintance or social network referrals and one recruit of unknown origin.
↵§ Physical violence is defined as reporting one or more of the following incidences: being robbed, hit, beaten, threatened, attacked with a weapon or kidnapped.
↵¶ Final model adjusted for migrant status, recruitment location, having a current sex partner, lifetime experience of arrest or prison, and alcohol audit score.