Fitting a Model for the Determinants of Breast Cancer in Women Using Meta-Regression

Meta-regression analysis is a robust statistical framework which is use to examine the relationship between breast cancer determinants and the disease outcomes by combining evidence from multiple studies. This paper aims to fit a model of the determinants of breast cancer using random effect model. The effect size index was odd ratio and data was sourced via Pubmed, Science Direct, Medline, Rechargegate and Google scholar. The random-effects model was employed for the analysis. odd ratio was used as a measure of the association of breast cancer determinants in women. It was formed that the reported I^2statistic is 0.23(23%), which suggest no heterogeneity using the categorization of Higgins et al (2003). In other words, 28% of the variability in the residuals is still attributed to the between-study variation, whereas only 72% is attributed to the within-study variation. The adjuste R^2statistic is 99.06(99%) which assess the proportion of between-study variance explained by the covariates, here ninety-nine per cent (99%) of the between-study variance is explained by the covariates. The model test which is the X^2statistic is 9.59 with p-value of 0.0479 which test that all the coefficients other than the intercepts are equal to zero based on the chi-square X^2 distribution with p-1 degree of freedom. This paper concluded that on the basis of the data presented and analyzed, it could be noted that the collated results of the secondary studies show that menarche, menopause and family history of breast cancer were strong risk factor for breast cancer while ever breast was not. This paper recommended that menarche, menopause and family history of breast cancer are strong risk factor for breast cancer and it should be included in sensitization and also client counselling.