Consanguineous marriages are strongly preferred in much of West and South Asia. This paper examines the prevalence and sociodemographic correlates of consanguineous unions in Pakistan using local and national data. Information from 1011 ever-married women living in four multi-ethnic and multi-lingual squatter settlements of Karachi, the main commercial centre of the country, are compared with data from the national 1990/91 Pakistan Demographic and Health Survey (PDHS), based on information provided by 6611 women. Both sets of results indicate that approximately 60% of marriages were consanguineous, over 80% of which were between first cousins. The mean coefficients of inbreeding (F) in the present generation were 0.0316 and 0.0331 for the Karachi and PDHS data respectively. In both surveys the prevalence of consanguineous unions appeared to be unchanged over the past three to four decades. Consanguineous unions were more common among women who were illiterate or had only primary level education, were first or second generation migrants from rural areas of Pakistan or, in the PDHS, lived in rural areas, and whose parents were also consanguineously married.
PIP: This study describes the nature and extent of consanguineous marriages in an urban slum in Karachi, Pakistan. Data were obtained from a survey conducted among a sample of ever married women aged 15-49 years. Findings are compared to findings from the Demographic and Health Survey (DHS) among a nationally representative weighted sample. Findings confirm, in both surveys, that consanguineous marriage was common among most subgroups and in both urban and rural areas in Pakistan. Consanguinity was 58.7% in the Karachi survey and 62.7% in the DHS. 83.6% of consanguineous marriages in the Karachi survey, and 80.4% in the DHS, were between first cousins. There was little evidence of a preference for patrilateral marriages. The mean coefficient of inbreeding in the children of the present generation was 0.0316 in the Karachi study and 0.0331 in the DHS. Actual levels are probably much higher than indicated. 69.9% of consanguineous women in the Karachi study, and 79.2% in the DHS study, were women with no formal schooling. Few women were employed in the formal or informal sectors. Consanguinity was practiced by all 3 religious groups and in urban and rural areas. Punjabis had the lowest prevalence of consanguinity. Muhajirs, who were mostly from Bihar state, had the lowest prevalence among Muslim subgroups. Consanguinity occurred in nuclear and extended households and among a range of socioeconomic groups, with the exception of rural DHS data. Findings suggest that consanguinity is a social concern for the well being of the family and daughter. Economic factors may be more prevalent in rural areas.