Risk factors for reported obstetric complications and near misses in rural northwest Bangladesh: analysis from a prospective cohort study.

BACKGROUND: In rural Bangladesh, more than 75% of all births occur at home in the absence of skilled birth attendants. Population-based data are lacking on the burden and risk factors for obstetric complications in settings with low rates of institutional delivery. We sought to describe the prevalence of reported complications and to analyze risk factors for obstetric complications and near misses, using data from a representative, rural setting of Bangladesh. METHODS: This study utilized existing data on 42,214 pregnant women enrolled in a micronutrient supplementation cohort trial between 2007 and 2011 in rural northwest Bangladesh. Based on self-report of complications, women were categorized as having obstetric complications, near misses, or non-complicated pregnancies using definitions modified from the World Health Organization. Multivariable multinomial regression was used to analyze the association of biological, socioeconomic, and psychosocial variables with obstetric complications or near misses. RESULTS: Of enrolled women, 25% (n = 10,380) were classified as having at least one obstetric complication, 2% (n = 1,004) with reported near misses, and 73% (n = 30,830) with non-complicated pregnancies. Twelve percent (n = 5,232) reported hemorrhage and 8% (n = 3,259) reported sepsis. Of the 27,241 women with live births or stillbirths, 11% (n = 2,950) reported obstructed labor and 1% (n = 328) reported eclampsia. Biological risk factors including women's age less than 18 years (Relative Risk Ratio [RRR] 1.26 95%CI:1.14-1.39) and greater than 35 years (RRR 1.23 95%CI:1.09-1.38), history of stillbirth or miscarriage (RRR 1.15 95%CI:1.07-1.22), and nulliparity (RRR 1.16 95%CI:1.02-1.29) significantly increased the risk of obstetric complications. Neither partner wanting the pregnancy increased the risk of obstetric complications (RRR 1.33 95%CI:1.20-1.46). Mid-upper arm circumference <21.5 cm increased the risk of hemorrhage and sepsis. CONCLUSIONS: These analyses indicate a high burden of obstetric morbidity. Maternal age, nulliparity, a history of miscarriage or stillbirth, and lack of pregnancy wantedness were associated with increased risk of obstetric complications. Policies to address early marriage, unmet need for contraception, and maternal undernutrition may help mitigate this morbidity burden in rural Bangladesh.

Screening utility, local perceptions, and care-seeking for reported jaundeesh among respondents lacking signs of icterus in rural Bangladesh.

In rural Bangladesh, acute viral hepatitis presents a significant burden on the public-health system. As part of the formative work for a large epidemiologic study of hepatitis E in rural Bangladesh, we sought to identify local terms that could be used for population-based screening of acute viral hepatitis. Exploration of the local term jaundeesh for screening utility identified a high burden of reported jaundeesh among individuals without symptoms of icterus. Recognizing that local perceptions of illness may differ from biomedical definitions of disease, we also sought to characterize the perceived aetiology, care-seeking patterns, diagnostic symptoms, and treatments for reported jaundeesh in the absence of icteric symptoms to inform future population-based studies on reported morbidities. We conducted a cross-sectional survey among 1,441 randomly-selected subjects to identify the prevalence of reported jaundeesh and to test the validity of this local term to detect signs of icterus. To characterize the perceived aetiology and care-seeking patterns for jaundeesh among the majority of respondents, we conducted in-depth interviews with 100 respondents who self-reported jaundeesh but lacked clinical signs of icterus. To describe diagnostic symptoms and treatments, in-depth interviews were also performed with 25 kabirajs or traditional faith healers commonly visited for jaundeesh. Of the 1,441 randomly-selected participants, one-fourth (n=361) reported jaundeesh, with only a third (n=122) reporting yellow eyes or skin, representative of icterus; Jaundeesh had a positive predictive value of 34% for detection of yellow eyes or skin. Anicteric patients with reported jaundeesh perceived their illnesses to result from humoral imbalances, most commonly treated by amulets, ritual handwashing, and bathing with herbal medicines. Jaundeesh patients primarily sought folk and spiritual remedies from informal care providers, with only 19% visiting allopathic care providers. Although the local term jaundeesh appeared to have limited epidemiologic utility to screen for acute symptomatic viral hepatitis, this term described a syndrome perceived to occur frequently in this population. Future population-based studies conducting surveillance for acute hepatitis should use caution in the use and interpretation of self-reported jaundeesh. Further study of jaundeesh may provide insight into the appropriate public-health response to this syndrome.

Care-seeking patterns for fatal non-communicable diseases among women of reproductive age in rural northwest Bangladesh.

BACKGROUND: Though non-communicable diseases contribute to an increasing share of the disease burden in South Asia, health systems in most rural communities are ill-equipped to deal with chronic illness. This analysis seeks to describe care-seeking behavior among women of reproductive age who died from fatal non-communicable diseases as recorded in northwest rural Bangladesh between 2001 and 2007. METHODS: This analysis utilized data from a large population-based cohort trial in northwest rural Bangladesh. To conduct verbal autopsies of women who died while under study surveillance, physicians interviewed family members to elicit the biomedical symptoms that the women experienced as well as a narrative of the events leading to deaths. We performed qualitative textual analysis of verbal autopsy narratives for 250 women of reproductive age who died from non-communicable diseases between 2001 and 2007. RESULTS: The majority of women (94%) sought at least one provider for their illnesses. Approximately 71% of women first visited non-certified providers such as village doctors and traditional healers, while 23% first sought care from medically certified providers. After the first point of care, women appeared to switch to medically certified practitioners when treatment from non-certified providers failed to resolve their illness. CONCLUSIONS: This study suggests that treatment seeking patterns for non-communicable diseases are affected by many of the sociocultural factors that influence care seeking for pregnancy-related illnesses. Families in northwest rural Bangladesh typically delayed seeking treatment from medically certified providers for NCDs due to the cost of services, distance to facilities, established relationships with non-certified providers, and lack of recognition of the severity of illnesses. Most women did not realize initially that they were suffering from a chronic illness. Since women typically reached medically certified providers in advanced stages of disease, they were usually told that treatment was not possible or were referred to higher-level facilities that they could not afford to visit. Women suffering from non-communicable disease in these rural communities need feasible and practical treatment options. Further research and investment in adequate, appropriate care seeking and referral is needed for women of reproductive age suffering from fatal non-communicable diseases in resource-poor settings.