Experience of social harms among female sex workers following HIV self-test distribution in Malawi: results of a cohort study

Table 4 The association between HIV testing in the previous week and physical abuse in the current week, using GEE

Variable	Category	Reports of physical abuse (reports/total weeks reported) (%)^a	Univariable analysis OR (95% CI)	p-value	Multivariable analysis OR (95% CI)^b	p-value
Test type in previous week	No test	136/1083 (12.6)	1	0.34	1	0.57
	Clinic-based test	4/47 (8.5)	0.92(0.42—2.00)		0.70(0.28—1.71)
	Self-test	15/100 (15.0)	1.67(0.88—3.19)		1.29(0.76—2.18)
	Clinic test & Self-test	1/27 (3.7)	0.51(0.12—2.12)		0.66(0.20—2.19)
Sexual encounters in the current week	0–9	50/593 (8.4)	1	< 0.01	1	< 0.01
	10–20	81/399 (20.3)	2.56(1.68—3.90)		2.56(1.68—3.92)
	> 20	25/265 (9.4)	2.50(1.43—4.36)		2.51(1.44—4.39)
Age range (years)	16—25	90/621 (14.5)	1	0.41
	26—35	60/512 (11.7)	0.78(0.37—1.63)
	> = 36	6/124 (4.8)	0.31(0.05—1.91)
Educational Attainment	Primary or less	111/826 (13.4)	1	0.34
Educational Attainment	Secondary or higher	45/431 (10.4)	0.68(0.31—1.49)	0.34
Received material goods or payment in exchange for sex	No	51/405 (12.6)	1	0.79
	Yes	82/651 (12.6)	0.89(0.39—2.05)	0.79
	Missing	23/201(11.4)

^aA total of 156 reports of physical abuse were recorded in 1257 weeks of data collection
^bOnly variables showing an association with the outcome in univariable analyses were included in the multivariable model in which odds ratios were adjusted test-type in the previous week and number of sexual encounters in the current week

ISSN: 1471-2334