A cross-sectional study of the first and second delays among women admitted to a maternity hospital with severe obstetric complications (near-miss) in Afghanistan

In Afghanistan, the majority of women continue to give birth at home because of poverty, difficult access to health facilities, or gender-based restrictions. Women are often brought into hospitals in moribund conditions after the onset of complications at home. A detailed understanding of the determinants of health care seeking delays is necessary in order to help identify strategies which could reduce the incidence of very severe complications and maternal deaths and improve foetal outcomes in complicated pregnancies. Areview of existing studies of care-seeking delays indicated that durations of care-seeking time had not been well explained because of various methodological limitations. The large majority of previous studies were descriptive and fell short in identifying contributing factors that could be eliminated by interventions while analytical studies lacked methodological rigour largely due to sample size limitations associated with rarity of maternal deaths. In this thesis, data from a hospital-based cross-sectional survey conducted among 472 women with severe obstetric complications in Afghanistan were analysed using a refined version of the conceptual framework developed by Thaddeus and Maine (1994). Three types of care-seeking delays were considered: the duration oftime from onset of symptoms to decision to seek care (or 'decision delay'), the duration from the decision to departure for health care facilities, (or 'departure delay') and variation in self-reported travel time from GIS-modelled travel time (or 'travel delay'). The study posited that delayed care-seeking would be best explained by a combination of factors including a woman's health care practice during pregnancy, her family's financial and social resources, geographical accessibility to healthcare and the types of symptoms and signs associated with each complication. It was also postulated that care-seeking delay would be among important determinants of foetal death. Regression techniques were used to identify determinants of the three types of delays, and logistic regression techniques were employed to assess the role of delays on foetal mortality. This study showed that failure to use antenatal care ('ANC') service during pregnancy was associated with an increase in decision delay. Lack of birth plans and absence of a midwife in the locality were also associated with an increased decision delay for ante- and intra-partum women. Awoman's weak relationship with her birth family was associated with an increased decision delay for complication types which did not have clear symptoms while a woman from an impoverished household appeared to experience a long decision delay when she suffered a complication with dramatic symptoms. In addition to seasonal effects, difficult geographical access to healthcare and lack of social capital were found to be positively associated with delay in departure for healthcare facilities. Multi-referrals, low household economic status, lack of community cohesion, and lack of access to vehicle were associated with an increase in travel delay. Finally, decision delay contributed to an increased risk of foetal death. The main conclusion from this work is that ANCinterventions have a significant role to play in facilitating rapid uptake of emergency care, once a complication occurs, in a setting where access to routine and emergency care is socially and geographically difficult. This in turn has implications not only for maternal but also for foetal outcomes. Future research and programmatic efforts should be directed towards understanding and exploiting the roles that social resources could play in facilitating access to emergency obstetric care