This hospital based comparative study was conducted in a populous cities of Sindh, Pakistan. The study aimed to assess prevalence of consanguinity and to determine association of consanguinity with premature births. The 962 recently delivering mothers were selected through convenience sampling. The socio-demographic predictors of premature births were sorted by bivariate analysis. The logistic regression modelling was done to compare odds ratios at 95% confidence intervals for the association between consanguinity and premature birth. The prevalence of consanguinity was 57.38%; the 47.82% of pregnancies in first cousin couples resulted in premature births (odds ratio 4.34). Younger maternal age i.e. < 25 years, addiction to betel nuts and inadequate antenatal care (Odds ratios 1.90, 2.89 & 1.84 respectively) were identified as potential statistically significant confounders. The adjusted odds ratios of premature births among first and second cousin couples were 4.22 & 1.62 i.e. P < 0.01 and P < 0.05, respectively. The study highlights an increase in preterm births in consanguineous couples.