Table 3. Bivariate and multivariate logistic regression of factors associated with self-medication.

Variable	Total	Self-medication	CRUDE		ADJUSTED
	n	Yes (n (%))	OR (95% CI)	p- value	OR (95% CI)	p- value
Age group
18–35	222	124 (55.9)	1(baseline)		1 (baseline)
36–60	68	46 (67.6)	1.65(0.93–2.93)	0.086	1.73(0.86–3.50)	0.126
61–78	10	04 (40.0)	0.52(0.14–1.91)	0.331	0.64(0.32–1.67)	0.554
Sex
Male	160	91 (56.9)	1(baseline)		1 (baseline)
Female	140	83 (59.3)	1.10(0.69–1.75)	0.673	1.09(0.68–1.78)	0.705
Marital status
Married	139	80 (57.6)	1(baseline)		1 (baseline)
Unmarried	161	94 (58.4)	1.03(0.65–1.63)	0.884	1.49(0.80–2.75)	0.206
Educational level
No formal education	17	09 (52.9)	1(baseline)		1 (baseline)
Primary	111	70 (63.1)	1.51(0.54–4.23)	0.426	1.45(0.46–4.51)	0.522
Above primary	172	95 (55.2)	1.09 (0.40–2.97)	0.856	1.02(0.32–3.25)	0.978
Occupation
Employed	16	13 (81.2)	1(baseline)		a	a
Unemployed	284	161 (56.7)	0.30 (0.08–1.08)	0.066
Income (TZS†)
<180,000	287	166 (57.8)	1(baseline)
>180,000	13	08 (61.5)	1.16(0.37–3.65)	0.792	a	a

†TZS = Tanzanian shillings (1 United States Dollar = 2, 230 TZS).

^a Multivariate estimates were too small to be interpreted and therefore omitted.