Determinants of Consanguinity and Inbreeding Coefficient F in Dir Lower District, North-West Pakistan: A Multivariate Approach

Consanguinity is common in many countries and the rates of Consanguineous Unions (CU) and Inbreeding Coefficient F (ICF) have been observed in various isolated as well as metropolitan populations across the globe (1). Studies on the dynamics of consanguinity in transitory and fragmented populations are however, scarce (2). The Pashtun tribal belt living on both sides of the Durand Line at Pakistan-Afghanistan border, are the victim of 'war-on-terror' due to the international counter terrorism strategy. The combat, Talibanization and military operations rendered 1.9 million individuals as Internally Displaced Persons (IDPs), and had a devastating effect on the well-being, and socioeconomic and demographic foundations of Pashtuns populations dwelling across the NorthWestern territories of Pakistan (3). The nature of this demographic transition remains unexplored. To this end, as an indicator of the socio-biological composition of the Pashtun tribes we have studied consanguinity and ICF in the population of Dir Lower district, which borders the turbulent territory at the NorthWest of Pakistan and is a victim of postwar consequences and natural disasters like recurrent floods. Through a cross-sectional study, 1250 random individuals were enrolled and data were obtained on marital union types and socio-demographic variables. Collectively, CU and non-consanguineous unions were observed to be 46.2% and 53.8%, respectively , and ICF was calculated to be 0.0249. Across the seven tehsils of Dir Lower, highest rate of consanguinity was witnessed in Samar Bagh (57%), followed by Timergara (55%), and Balam-bat (54%) (Fig. 1). The rate of consanguinity was higher in subjects originating from rural areas compared to their urban counterparts (46% vs. 39%). Among various Pashtun tribes, consanguinity was observed to be highest in Mashwani (57%) and Uthman Khels (51%). Further, the rate of CU was witnessed to be significantly higher in the literate individuals compared to the illiterate group (49% vs. 28%; P<0.0001). Within the literate sample, CU and ICF were generally rising with increasing literacy level. Among the occupational groups, CU were highest in subjects engaged in sales/business (50%; ICF=0.0273), and skilled manual jobs (48%; ICF=0.0267). In the average monthly income categories the prevalence of consanguinity was rising with increasing income (P=0.0029). According to their family types, the highest rate of CU was observed in individuals having 'extended family' (50%; ICF=0.0267) (P<0.0001). For the multivar-iate logistic regression analyses, a stepwise likelihood ratio was estimated and the independent parameters were considered one-by-one and only the significant variables were retained. In the final