Table 3.

The results of meta-analysis, meta-regression, and publication bias of studies conducted on the prevalence / OR of gynecologic cancer in PCOS patients

Outcomes	Number of observations	#I2%	&Publication bias	£Pooled OR (95%CI), (95%PI)	@Meta-regression
Endometrial cancer	${}^{\$}$ 6+3	$>$ 50	0.015*	1Insufficient observations
PCOS	${}^{\$}$ 7+3	$>$ 50	0.004*
NON-PCOS	${}^{\$}$ 6+3	$>$ 50	0.039*	${}^{\$}$ 2.2 (1.03, 4.7)* (0.33, 9.5)	20.71 (0.10), 62.89%
Ovarian cancer	6	$<$ 50	0.950
PCOS	7	$>$ 50	0.764		Insufficient observations
NON-PCOS	5	$>$ 50	0.051	1.3 (1.0, 1.8)* (0.74, 1.65)		1.0 (0.07), 0.00%
Breast cancer	7	$<$ 50	0.452		1.0 (0.10), 0.00%
PCOS	${}^{\$}$ 8+4	$>$ 50	0.015*
NON-PCOS	7	$>$ 50	0.652	1.1 (0.87, 1.4) (0.89, 1.98)	0.98 (0.09), 23.98%
Overall	${}^{\$}$ 19+3	$>$ 50	0.045*			0.99 (0.07), 82.46%
PCOS	${}^{\$}$ 22+10	$>$ 50	0.000*
NON-PCOS	${}^{\$}$ 18+9	$>$ 50	0.000*	${}^{\$}$ 1.4 (1.0, 1.9)* (0.57,4.5)	0.98 (0.09), 68.29%
${}^{\#}$ I-square (I2) was used to assess heterogeneity, ${}^{\&}$ Beggs' test was run to assess publication bias, *Significant level was considered at p $<$ 0.05, ${}^{\$}$ Trim and fill correction method was applied, ${}^{\pounds }$ To estimate pooled odds ratios (ORs), we applied the “Meta-prop” random effect method. The 95% prediction interval (95% PI) was estimated to evaluate clinical significance as compared to statistical significance for the pooled odds ratio, ${}^{@}$ Exponential (beta regression coefficient) (Std. Err., I-squared residuals %): 1- age adjusted, 2- PCOS diagnostic criteria adjusted, PCOS: Polycystic ovary syndrome