Table 2.

Univariate and multivariable logistic regression models for identifying factors associated with access to maternal healthcare services (MHC) among Indigenous women in Chittagong Hill Tracts, Bangladesh, after adjusting for clustering by para

Variables	n (%)	Univariate analysis			Multivariable logistic regression
		OR	95% CI for OR	P value	Adjusted OR	95% CI for AOR	P value
Place of residence (subdistrict)
Matiranga	325 (74.2)	Ref
Khagrachhari Sadar	113 (25.8)	4.2	2.5 to 7.1	< 0.001	2.4	0.90 to 7.0	0.10
Ethnicity
Chakma	220 (50.2)	Ref			Ref
Marma	100 (22.8)	1.6	1.0 to 2.8	0.05	0.8	0.32 to 2.0	0.70
Tripura	118 (27.0)	0.50	0.34 to 0.84	0.006	0.5	0.25 to 1.3	0.20
Age
Young (15–29)	328 (74.9)	Ref			Ref
Adult (30 and above)	110 (25.1)	1.9	1.2 to 3.0	0.007	1.7	0.90 to 3.5	0.10
School attendance
Did not attend—Junior	281 (64.2)	Ref			Ref
Secondary and above	135 (35.8)	7.9	4.7 to 13.1	< 0.001	2.4	1.2 to 4.9	0.01
Household income
4000–9000	198 (45.2)	Ref			Ref
10 000–19 000	111 (25.3)	2.1	1.3 to 3.3	0.002	1.1	0.61 to 2.1	0.24
20000 and above	129 (29.5)	8.2	4.7 to 14.4	< 0.001	2.0	0.90 to 4.4	0.11
Knowledge of nearest health facilities
No	68 (15.5)	Ref			Ref
Yes	370 (84.5)	6.1	3.4 to 11.1	< 0.001	3.8	1.8 to 7.8	< 0.001
Access to media access for MHC information
No	352 (80.4)	Ref			Ref
Yes	86 (19.6)	13.0	5.5 to 30.7	< 0.001	2.3	0.76 to 7.2	0.14
Knowledge on pregnancy-related complications
No	323 (73.7)	Ref			Ref
Yes	115 (26.3)	3.5	2.1 to 5.8	< 0.001	3.0	1.5 to 5.8	0.002
Number of total pregnancies
1–2 pregnancies	310 (70.8)	Ref			Ref
3 and above pregnancies	128 (29.2)	0.50	0.31 to 0.72	0.001	0.54	0.30 to 1.0	0.05

AOR, adjusted Odds Ratio; CI, Confidence Interval; OR, Odds Ratio.