Table 2.

Factors associated with acceptance of provider-initiated HIV testing among pregnant women attending ANC in public health facilities of Assosa town, Northwest Ethiopia, 2014 (N = 386)

Variables	PITC acceptance		COR (95 % CI)	AOR (95 % CI)	P value
	Yes	No
Residence
Rural	77	11	1.88 (0.94–3.74)	4.04 (1.24–13.11)	0.020
Urban	235	63	1.00	1.00
Ethnicity
Berta	78	11	2.24 (1.09–4.60)	3.51 (1.29–9.56)	0.049
Oromo	53	12	1.39 (0.68–2.86)	2.10 (0.89–4.93)
Others	48	9	1.68 (0.76–3.72)	1.43 (0.59–3.43)
Amhara	133	42	1.00	1.00
Occupation
Merchant	29	3	3.8 (1.10–13.20)	4.43 (1.18-16.68)	0.019
Farmer	49	9	2.15 (0.97–4.76)	1.80 (0.51–6.34)
Employed	105	18	2.30 (1.24–4.26)	2.15 (1.08–4.30)
Student	25	3	3.29 (0.94–11.48)	6.00 (1.45–24.75)
Housewife	104	41	1.00	1.00
Attitude towards PITC
High favourable	256	53	1.81 (1.01–3.24)	1.57 (1.08–6.25)	0.048
Less favourable	56	21	1.00	1.00
Stigmatization
No	99	25	2.34 (1.04–5.28)	3.54 (1.23–10.16)	0.019
Low	191	36	3.14 (1.45–6.79)	4.04 (1.52–10.72)
High	22	13	1.00	1.00
Planned to disclose results to partner after testing
Yes	297	56	6.36 (3.03–13.37)	14.85 (4.60–47.94)	<0.001
No	15	18	1.00	1.00
Perceived the pre- test counselling service
Good	278	52	3.46 (1.88–6.38)	4.23 (2.01–8.89)	<0.001
Poor	34	22	1.00	1.00
Partner reaction to positive result
Positive	211	37	2.09 (1.25–3.49)	1.7 (0.91–3.36)	0.097
Negative	101	37	1.00	1.00

N.B. The assumptions for the application of multivariate logistic regression analysis was fulfilled by using Hosmer and Lemeshow test and the model was adequately fitted (P = 0.747)

For explanatory variables having more than 2 categories, the overall significance of P value used