Table 3.

Matched logistic regression results for the different risk factors for vaginal fistula symptoms

Risk Factors	Bayesian matched logistic regressions adjusting for outcome misclassification
	OR (95 % CrI)
Being able to read	0.87 (0.77-0.99)
Post-primary education	0.90 (0.76-1.06)
Female genital mutilation	1.04 (0.82-1.30)
Short stature (<150 cm)	1.31 (1.02-1.68)
Intimate partner sexual violencea	2.13 (1.60-2.86)
Young age at 1st intercourse (<14 years)b	1.41 (1.16-1.71)
Young age at 1st birth (<14 years)c	1.39 (1.04-1.82)
Problem with permission to seek care	1.20 (0.99-1.47)

OR = Odds ratio; 95 % CI = 95 % Confidence Interval; 95 % CrI = 95 % Credible Intervals.

Statistically significant results at the alpha = 0.05 level are bolded.

The matched logistic regression models adjust for the following covariates: age, literacy status (except for ‘Being able to read’ and ‘Post-primary education’), location of residence (urban/rural), gravidity status (except for ‘Young age at 1 ^st birth’), religion, and country (country fixed effects were omitted from the Bayesian regressions for ‘Young age at 1 ^st birth’ and ‘Problem getting permission to seek healthcare’).

^aAmong married and/or ever married women (or those in a union).

^bAmong sexually active women.

^cAmong primi/multiparous women.