Worldwide there has been an increase in the prevalence of overweight and obese in women with unprecedented upsurge in developing countries. Tanzania is among the countries reported to have high prevalence of overweight and obese among women of reproductive age. Studies from developed countries have suggested an association between maternal obesity and complications during pregnancy, delivery and postpartum period for both mother and offspring. The latter may not easily be translated to actual reality in developing countries considering high prevalence of diseases and poverty in these settings including Tanzania. Understanding the magnitude and predictors of maternal obesity will assist to draw attention of public health community for potential controlling interventions. Highlighting maternal morbidities and adverse fetal outcomes associated with maternal obesity will assist to guide clinicians by taking extra precautions when managing obese pregnant women. Objectives This study aims to determine predictors of maternal obesity and its associated maternal morbidities and fetal outcomes among women who delivered at KCMC Hospital from 2000 to 2015. Methodology A registry based retrospective cohort study design will be adopted using the KCMC maternally linked data from KCMC medical birth registry. This study will include women of reproductive age who attended their 1st antenatal clinic (ANC) visit during 1st trimester of pregnancy, and women who had delivered singleton babies at KCMC hospital from 2000 to 2015. Women who had missing either weight or height and have live birth less than 28weeks gestation will be excluded. vii Data will be analysed using STATA version 13.0. Univariate analysis will be summarized by frequency, proportion and percentages for categorical variables and measures of central tendency with respective measures of dispersion will be used for continuous variables. Both bivariate and multivariable analysis will be computed using stepwise Logistic regression and generalized linear method by estimating odds ratios and relative risk with respective 95% CI to control for confounding and effect modification. A p-value of less or equal to 5% will be considered as statistically significant.