Table 2.

Predictors studied across the articles reviewed,^a by study outcome and results

Predictor	Closure		Residual incontinence		Any incontinence
	Influence	No influence/ inconclusiveb	Influence	No influence/ inconclusive	Influence	No influence/ inconclusive
Patient Characteristics
S. haematobium		22
Age at fistula repair		7		18, 7		9, 16, 19, 25
Age at fistula occurrence						9, 19
Duration of fistula		10, 7		10, 18, 7		15, 16, 19c
Parity				18		9, 16, 19, 25
Number living children						19
Mode of delivery				18		19
Days in labor						19
Education						19
Literacy						19
Place of delivery						19
Fistula characteristics
Etiology						16
Number of fistulas				18		19
Fistula size		6, 7	6, 18	7, 21		9, 16, 19, 25
Scarring	17	6, 8	6d, 18	8	9, 19
Bladder size	17		18
Bladder neck / circumferential fistula					19
Extent of urethral involvement / circumferential fistula	17	6, 7	6, 18	7	19	9
Ureteric involvement		7		7		9
Other fistula location (“low”, juxtacervical, etc)					19	9, 25
Combined vvf/rvf		7		18	19
Previous repair				18		9, 16, 25
Peri-operative factors
Abdominal vs vaginal surgical route		12				16, 20e
Catheter for 10, 12, 14 days		13		13
Single (vs. double) layer closure		17
Relaxing incision					19
Martius fibrofatty flap/graft		11, 14		11		19
Martius graft (vs fibrin glue)						25
Antibiotic prophylaxis		23		23
Single dose (vs extended antibiotics)	24

^aArticles indicated by reference number

^bFactors evaluated in an observational study using bivariate analyses only, not significant (p>.05) in bivariate analysis of RCT data or not significant (p>.05) in multivariate analysis of observational data

^cOutcome confirmed via personal communication with primary author

^dCategorized as part of “special considerations” in the Goh classification system

^eOutcome examined confirmed via personal communication with primary author