Causes of adult female deaths in Bangladesh: findings from two National Surveys
BMC Public Health volume 15, Article number: 911 (2015)
Assessment of causes of death and changes in pattern of causes of death over time are needed for programmatic purposes. Limited national level data exist on the adult female causes of death in Bangladesh.
Using data from two nationally representation surveys, the 2001 and 2010 Bangladesh Maternal Mortality Surveys (BMMS), the paper examines the causes of adult female death, aged 15–49 years, and changes in the patterns of these deaths. In both surveys, all household deaths three years prior to the survey were identified. Adult female deaths were then followed by a verbal autopsy (VA) using the WHO structured questionnaire. Two physicians independently reviewed the VA forms to assign a cause of death using the ICD-10; in case of disagreement, a third physician made an independent review and assigned a cause of death.
The overall mortality rates for women aged 15–49 in 2001 and 2010 were 182 per 100,000 and 120 per 100,000 respectively. There is a shift in the pattern of causes of death during the period covered by the two surveys. In the 2001 survey, the main causes of death were maternal (20 %), followed by diseases of the circulatory system (15 %), malignancy (14 %) and infectious diseases (13 %). However, in the 2010 survey, malignancies were the leading cause (21 %), followed by diseases of the circulatory system (16 %), maternal causes (14 %) and infectious diseases (8 %). While maternal deaths remained the number one cause of death among 20–34 years old in both surveys, unnatural deaths were the main cause for teenage deaths, and malignancies were the main cause of death for older women. Although there is an increasing trend in the proportion of women who died in hospitals, in both surveys most women died at home (74 % in 2001 and 62 % in 2010).
The shift in the pattern of causes of adult female deaths is in agreement with the overall change in the disease pattern from communicable to non-communicable diseases in Bangladesh. Suicide and other violent deaths as the primary cause of deaths among teenage girls demands specific interventions to prevent such premature deaths. Prevention of deaths due to non-communicable diseases should also be a priority.
Knowledge about causes of death is crucial for public health planning and resource allocation . However, information on causes of death is limited in most developing countries. In these countries vital registration systems are either absent or poorly developed and lack quality information. Most deaths occur at home  and thus, information on causes of death is not available, except from very few areas that have comprehensive demographic surveillance system capturing information on deaths by cause, age and sex . In order to get an estimate of cause specific mortality, these countries often rely on modeling exercises or on facility-based data, which is usually biased and incomplete for estimating causes of death.
Information on causes of death is especially important in view of the epidemiological transition that developing countries are experiencing. Although limited evidence exists, Bangladesh, a developing country in South Asia, is going through an epidemiological transition -- a shift in the mortality pattern from communicable diseases (CDs) to chronic diseases [4, 5]. Analysis of data from Matlab, a rural area in Bangladesh with a demographic surveillance system collecting information on causes of death since 1986, showed a substantial change in the mortality pattern from acute, infectious and parasitic diseases to non communicable, degenerative and chronic disease during the last 20 years .
In Bangladesh, there is a paucity of data on the causes of death at the population level. Although the Bangladesh Bureau of Statistics (BBS) maintains a nationally representative Sample Vital Registration System (SVRS) which records causes of death based on the information collected by a lay reporting system, there are reservations about the accuracy of the causes of death information from this system . Studies have used verbal autopsy (VA) as a means to identify the causes of death among children and adult population. VA is a method of assessing probable causes of death based on an interview with the next of kin or other caregivers who were present at the time of death or who are knowledgeable about the events leading up to death. VA has been used previously in Bangladesh to ascertain causes of death including maternal deaths [7–9] and childhood deaths [10–19].
VA has increasingly been used to determine causes of death, especially in settings where deaths occur outside a hospital. Recent studies have suggested that VA can provide causes of death information that, at the population level, is similar to death certification in high-quality hospitals . In this paper we examine changes in causes of adult female deaths using VA reporting on deaths from nationally representative sample of households.
The present paper uses VA data from two nationally representative surveys, the 2001 and 2010 Bangladesh Maternal Mortality Surveys (BMMS), to examine the causes of adult female death aged 15–49 years, and the changes in the patterns of causes of death during the 9 year period between the two surveys. Both surveys were large and covered approximately 100,000 and 174,000 households respectively using nationally representative samples, sufficient to detect a 20 % decline in reduction of maternal mortality with 95 % confidence and 80 % power. In both surveys samples were selected through a three-stage sampling design [20, 21]. Both surveys used the same methodology to measure maternal mortality and assign the causes of maternal and non-maternal death over time. In each survey, the household questionnaire included questions on all household deaths that happened in about four years before the survey, and recorded the age, sex and date of death of each deceased person. Deaths of women of reproductive age (15–49) were subsequently followed up for VA using an adapted version of the 2007 version of the World Health Organization (WHO) questionnaire for adult deaths . In the VA questionnaire information was collected on the signs and symptoms surrounding every death reported by the most knowledgeable person in the household. In the 2001 survey, 50 and in the 2010 survey, 60 female interviewers, each with at least a Bachelor’s degree, received a 7-days intensive training on the VA questionnaire before collecting information on the signs and symptoms surrounding every death reported by the most knowledgeable person in the household.
Cause of death was determined by independent review of the VA forms by two physicians. If they could not agree on the final cause of death, the case was then reviewed by a third physician. For a small number of cases where the three physicians agreed that the death was maternal but could not agree on a single cause of death, an expert committee of obstetricians was consulted to assign a cause of death. For each case, an underlying cause of deathFootnote 1 was coded using the International Classification of Diseases, 10th Revision (ICD-10) classification. The underlying cause of death is grouped into eight categories: maternal; infectious diseases; malignancies; diseases of the circulatory system; suicide; other violent deaths; miscellaneous, and deaths for which cause of death was impossible to assign, or for which the reviewing physicians could not agree on a single cause of death. In this analysis, deaths that occurred in the three years preceding the two surveys among women aged 15–49 years are included.
Analysis included mortality rates and proportional mortality showing age specific patterns of causes of death, place of death and health seeking behavior before death. The mortality rate is calculated as the ratio of deaths in the three years preceding the survey to the person years lived by women aged 15–49 years (person years lived by specific age groups for age specific mortality) in the same period expressed per 100,000 women aged 15–49 years. Proportional mortality (%) was calculated by dividing the number of deaths attributed to a specific cause by the total number of deaths for which a VA was carried out. The difference between the two death rates in 2001 and 2010 was compared by calculating the 95 % confidence intervals (CI), and the difference between two proportions was compared by z-test for statistical significance at 80 % power and p < 0.05. Data were analyzed using SPSS version 17.0 and Microsoft Excel 2007. As the sample was not self-weighted household weights were used to get unbiased estimates at the national level.
This paper is a secondary analysis of existing de-identified data sets. Ethical approval for the BMMS 2001 and 2010 were obtained from the Ethical Review Committee (ERC) of the Bangladesh Medical Research Council (BMRC).
In the 2001 BMMS, a total of 696 deaths were reported among women 15–49 years in the three years prior to the survey. The number of such deaths was 749 in the 2010 survey. The overall mortality rates for this group of women in 2001 and 2010 were 182/100,000 (95 % CI 161-201/100,000) and 120/100,000 (95 % CI 107-132/100,000) respectively (Fig. 1).
Table 1 ranks the number of deaths in the three years preceding the two surveys by causes of death under eight categories. The highest numbers of deaths in the 2001 survey were due to maternal causes (140) and diseases of the circulatory system (103) whereas in the 2010 survey the highest numbers of deaths observed were due to malignancy (159), diseases of the circulatory system (122) and miscellaneous causes (122).
Table 2 shows mortality rates of women by causes of death in the two surveys. In the 2001 survey the highest death rate was observed due to maternal causes (37/100,000), followed by circulatory diseases (26/100,000), malignancy (25/100,000) and infectious diseases (23/100,000), whereas in the 2010 survey, women had the highest mortality from malignancy (26/100,000) followed by death from circulatory diseases (20/100,000), maternal causes (17/100,000) and infectious diseases (17/100,000). When compared between the two surveys, the overall pattern of death rates shows a decreasing trend for almost all causes, except for deaths due to malignancy. However, none of these decreases was statistically significant except for death rates due to maternal (95 % CI 28-46/100,000 and 12-22/100,000 in 2001 and 2010 respectively) and infectious cause (95 % CI 16-30/100,000 and 7-14/100,000 in 2001 and 2010 respectively).
When the proportion of deaths by different causes were compared between the two surveys (Table 3), in the 2001 survey the main cause of death was maternal (20 %), followed by diseases of circulatory system (15 %), malignancy (14 %) and infectious diseases (13 %); in the 2010 survey, malignancy was the number one cause (21 %), followed by diseases of circulatory system (16 %), maternal causes (14 %) and infectious diseases (8 %). Between the two surveys, the relative share of deaths due to malignancy and diseases of the circulatory system increased significantly (p < 0.05) whereas the proportion of deaths due to maternal causes and infectious diseases decreased significantly (p < 0.05). The proportion of deaths due to suicide and other unnatural causes remained the same in both surveys.
As expected, cause specific death rates varied by different age groups (Table 2). The death rate from maternal causes increased beginning at age 20 and decreased after age 40. Death rates from circulatory diseases and malignancies increased sharply with age, whereas death rates from infectious diseases increased moderately with age. The suicide rate, on the other hand, was highest under the age of 30. Deaths due to other unnatural causes such as injuries, drowning, snakebites and other causes showed no clear age pattern. Both miscellaneous and unspecified and undetermined death rates increased moderately with age.
Distribution of deaths by different cause categories also varies by age groups (Table 3). While maternal deaths remained the number one cause of death among 20–34 years old in both surveys, unnatural deaths (suicide and other violent deaths) were the main causes for teenage deaths. In 2001, 28 % of teenage deaths were due to these causes, this figure increased in 2010 when 37 % of teenagers had unnatural deaths (p > 0.05). Unnatural deaths (suicide and other violent deaths) were the second leading cause of death for women aged 20–24 years after maternal deaths in both surveys; in 2001, 1 in every 5 women had suicidal deaths and an additional 7 % had other violent deaths. This figure decreased to 10 % for suicide and 9 % for other violent causes in 2010. A similar pattern of change in the causes of death is observed for 25–29 years age group for maternal and unnatural deaths. Diseases of the circulatory system and malignancies were the two leading causes of death among older women aged 35–49 years. These two diseases together accounted for 45 % of total deaths in 2001 and 50 % of total deaths in 2010 in this age group. In 2001, diseases of the circulatory system were the primary cause of death in this age group, followed by malignancies whereas malignancies ranked as the main cause in 2010 followed by diseases of the circulatory system. Women died from different types of malignancies and there was little variation in the type of malignancies between the two surveys. In 2001, among women who died from malignancies, the distribution by type malignancy was: cancers of the gastrointestinal tract, including liver (37 %), blood and lymphatic system (19 %), reproductive cancer including cancers of the reproductive organs including breast and cervical cancer (28 %) and other cancers (18 %). In 2010, women died from cancers of the gastrointestinal tract, including liver (30 %), cancers of blood and lymphatic system (28 %), reproductive cancers including cancer breast and cervical cancers (29 %) and other cancers (13 %). Under the miscellaneous category, diabetes, hepatitis and chronic respiratory diseases were the main causes of death in both surveys.
Although there is an increasing trend in the proportion of women who die in health facilities, the majority of women still die at home. In the 2010 survey, 38 % women died in a health facility compared to 26 % in 2001 (46 % increase; p < 0.05) as shown in Table 4. The increase in facility deaths was observed in all categories of deaths except in unnatural deaths.
Care seeking before death also improved (Table 5). In the 2010 survey, overall 78 % of women sought care from a health facility or pharmacy before death whereas this figure was 63 % in the 2001 survey (24 % increase; p < 0.05). While this increase was observed among all death categories except unnatural deaths, the increase was highest for deaths due to circulatory diseases (68 % increase; p < 0.05), followed by miscellaneous causes (44 %; p < 0.05) and maternal causes (29 %; p < .05). In both 2001 and 2010, public facilities were the most prominent source of care. However, in the 9 years between the two surveys, there has been a substantial increase in seeking care from private health facilities (increased from 12 to 46 %).
This paper presents, for the first time ever, provides national estimates of causes of adult female deaths and provides a unique opportunity to to examine changes in the patterns of causes of death over time. Overall, there has been a major shift in the pattern of causes of death over the 9 year period. While deaths due to maternal and infectious origins are significant contributors of adult female deaths, deaths due to non-communicable diseases (NCDs) such as circulatory diseases, malignancies, diabetes and other chronic conditions are on the rise. Unnatural deaths including deaths due to suicides are also a major killer.
While this study provides national level estimates of cause specific female mortality and the causes of death, the findings should be interpreted with caution, given the difficulties in assigning an exact cause of death using the VA method. Verbal autopsy, as a method of ascertaining causes of death has increasingly been recommended, especially when death occurs outside of health facilities and without the presence of a medically trained person to assign a cause of death [20, 23–25]. However, the validity of using VA to assess the causes of death depends on the extent and the accuracy of signs, symptoms and circumstantial evidence collected through the VA questionnaire and then on the review of this information and assignment of an appropriate cause of death by a skilled reviewer. The validity of using VA to ascertain causes of death may also vary by age group of the deceased and the cause of death .
We used an adapted version of the WHO questionnaire for adult deaths , have employed highly trained interviewers to conduct the interviews and, finally, have engaged trained physicians with previous experience in reviewing the VA questionnaires and assigning ICD-10 codes for causes of death. We also used similar methodology to collect and interpret data from both surveys so that the results are highly comparable. Even though we are unable to compare our findings with a gold standard -- death certificates, or hospital records, there is evidence from validation studies done in other settings where physician reviewed VA results were compared to the gold standard and found to have high diagnostic accuracy for diseases like cardiovascular diseases and malignancies  and other causes of adult death .
A methodological challenge in assigning an exact cause of death using the VA method is also reflected in the proportion of deaths that are being categorized as “not-classified”. In both surveys, about 10 % of deaths were categorized as “not-classified” for which it was either impossible to assign a cause of death on the basis of information collected through VA or the reviewing physicians could not agree to a single cause of death. If an appropriate diagnosis could be made for these cases, the cause specific mortality reported here might be affected to some extent. However, previous studies using similar verbal autopsy method as a means to assess the causes of death have reported either higher or similar figures for “unknown” or “unclassified” categories of death [7, 8].
Our findings are in agreement with other studies documenting NCDs as a major cause of adult death including female deaths in Bangladeshi [4, 7, 20, 30, 31]. Labrique et al.’s study conducted in a cohort of rural Bangladeshi women found NCDs as the leading cause of death among women of reproductive age (48 %), followed by pregnancy related causes (22 %), infectious diseases (17 %) and injury (9 %). Circulatory diseases accounted for the largest proportion of NCD deaths (40 %), followed by cancer (15 %) . Similarly, Alam et al.’s study conducted in another rural area of Bangladesh also found that 65 % of all female deaths aged 15–59 years were due to NCDs, followed by deaths due to communicable diseases (15 %), pregnancy related deaths (10 %) and injury related deaths (8 %) .
Similar to many other developing countries, Bangladesh is going through a shift in the disease and mortality patterns from acute infectious and parasitic diseases to non-communicable, chronic diseases. The 2010 update of Global Burden of Disease Study estimated that between 1990 and 2010, proportion of deaths due to NCDs in Bangladesh increased from 34 to 60 %, whereas contribution of communicable diseases (including maternal, neonatal, and nutritional disorders) nearly halved (56 % to 30 %) and the proportion of injury remained around 10 % (NCDs) . Given the prominence of NCDs in the mortality statistics and their potential impact on the socioeconomic development of the country, the government of Bangladesh has identified NCDs as a priority and has taken steps to strengthen activities for the prevention and treatment of these diseases. While the prevention efforts should be focused on identifying and modifying risk factors through practicing healthy life-style, early detection and treatment are also important to decrease the long term impact of these diseases.
Our results indicate that a substantial proportion of female deaths occurred due to intentional and unintentional harm, with suicide being the leading cause of death among female teenagers in both surveys. While limited information exists on the extent and the causes of suicide among the Bangladeshi population, a recent study conducted to understand the epidemiology of suicide in Bangladesh found that suicide is the fourth injury related deaths in Bangladesh and the leading cause of injury related deaths among adolescents aged 10–19 years, with slightly higher risk of suicide among females compared to males . In this study the overall suicide rate among adult females was found to be 8 per 100,000 which is lower than the overall suicide rate we found in our study (23 per 100,000 in 2001 and 16 per 100,000 in 2010). Another study conducted in Matlab, a rural area of Bangladesh, found an average annual death rate due to suicide among women 15–44 years old of 13 per 100,000 in 1982–1998 , which is also lower than the present study.
Because of methodological differences we cannot compare previous study findings with our study findings. However, suicide is an issue of growing concern in Bangladesh with some areas reporting 37 % of all deaths of women aged 15–44 years attributable to suicide . Longitudinal data collected from different demographic surveillance sites of icddr,b suggest that the percentage of deaths attributable to suicide among women, particularly among young women of reproductive age, is increasing .
We are also unable to explain the underlying causes of this high level of unnatural deaths. However, other studies have identified a number of circumstances leading to suicides among young girls and women including emotional and verbal abuse by husbands and other family members, financial hardship in the family, childlessness and rejection of marriage offer or unacceptable partner . Prevention of suicides is thus challenging and requires multifaceted approaches and adaptation of strategies to locally relevant cultural factors. A detailed investigation of these deaths, along with identification of their determinants, is needed before any programmatic actions can be taken to reduce such untimely deaths.
Using BMMS data, this paper presents the cause and age specific adult female deaths at the national level in two surveys, almost a decade apart, and comparing them between the two surveys. The sample size for the two surveys was calculated to have sufficient statistical confidence and power to estimate the national MMR and calculate the changes in MMR over the 9 year period between these surveys but not to estimate cause specific mortality or age and cause-specific mortality or to compare them between the two surveys. However, the results are strongly indicative of the decreasing trend for death rates for some diseases (maternal and infectious diseases).
There is an increasing trend in seeking care before death and in the proportion of deaths that happened in health facilities. The increase in seeking care is more marked for conditions like circulatory diseases and diseases like diabetes, chronic respiratory conditions etc. The majority of women visited public facilities for seeking care. Given that more than 70 % of the country’s population live in rural areas where the first point of care usually is public facilities that are primarily designed to provide maternal and child health care services, whether these facilities are equipped to manage the increasing number of NCD cases needs to be explored.
Bangladesh has made remarkable progress in the health and development sector in the last three decades. Bangladesh is one of the nine countries that are on track to achieve the Millennium Development Goal (MDG) 5 of reducing maternal mortality. This was made possible through improvements in the health sector as well as in non-health sectors. However, much still needs to be done to prevent untimely deaths of women. Our study provides a solid basis in support of the argument that Bangladesh is going through an epidemiological transition and NCDs are significant contributors to adult female deaths. Given the prominence of NCDs in the mortality statistics and their potential impact on the socioeconomic development of the country, the government of Bangladesh has identified NCDs as a priority area and has taken steps to strengthen activities for the prevention and treatment of these conditions. However, only 4.5 % of the national health budget in the Health Population and Nutrition Sector Development Program (HPNDSP 2011–16) has been allocated for these diseases. The Government efforts should be strengthened further to take this trend into account and to increase the proportion of the health budget for NCDs, and to use evidence-based information on how to tackle these conditions. While prevention efforts should be focused on identifying and modifying risk factors through practicing healthy life-style, early detection and treatment are also important to decrease the long term impact of these diseases.
The underlying cause of death is defined as the disease or injury which initiated the chain of morbid events leading directly to death, or the accident/act which produced the fatal injury.
Bangladesh Bureau of Statistics
Bangladesh Maternal Mortality Surveys
The International Classification of Diseases- 10th revision
Health Population and Nutrition Sector Development Program
Millennium Development Goal
Maternal mortality ratios
Sample Vital Registration System
World Health Organization
Soleman N, Chandramohan D, Shibuya K. Verbal autopsy: current practices and challenges. Bull World Health Organ. 2006;84(3):239–45.
Chandramohan D, Maude GH, Rodrigues LC, Hayes RJ. Verbal autopsies for adult deaths: issues in their development and validation. Int J Epidemiol. 1994;23(2):213–22.
Alam N, Chowdhury HR, Ahmed A, Rahman M, Streatfield PK. Distribution of cause of death in rural Bangladesh during 2003–2010: evidence from two rural areas within Matlab Health and Demographic Surveillance site. Global Health Action, [S.l.], 2014;7. ISSN 1654-9880. Available at: http://www.globalhealthaction.net/index.php/gha/article/view/25510. Date accessed: 17 Sep. 2015. doi:http://dx.doi.org/10.3402/gha.v7.25510.
Karar ZA, Alam N, Streatfield PK: Epidemiologic transition in rural Bangladesh, 1986-2006. Global Health Action, [S.l.], v. 2, jun. 2009. ISSN 1654-9880. Available at: http://www.globalhealthaction.net/index.php/gha/article/view/25510Date accessed: 17 Sep. 2015. doi:http://dx.doi.org/10.3402/gha.v7.25510
Molla A. Epidemiological Transition and Its Implications for Financing of Health Care: A Case Study in Bangladesh. Inaugural Conference of the American Society of Health Economists. TBA, Madison, WI, USA: Economics of Population Health; 2006.
Bangladesh Bureau of Statistics (BBS). Population & Housing Census: Preliminary Results,2011. Dhaka, Bangladesh: Statistics Division, Ministry of Planning; 2011.
Alam N, Chowdhury HR, Bhuiyan MA, Streatfield PK. Causes of death of adults and elderly and healthcare-seeking before death in rural Bangladesh. J Health Popul Nutr. 2010;28(5):520.
Fauveau V, Wojtyniak B, Koenig MA, Chakraborty J, Chowdhury A. Epidemiology and cause of deaths among women in rural Bangladesh. Int J Epidemiol. 1989;18(1):139–45.
Yusuf HR, Akhter HH, Chowdhury ME, Rochat RW. Causes of death among women aged 10–50 years in Bangladesh, 1996–1997. J Health Popul Nutr. 2007;25(3):302.
Chen LC, Rahman M, Sarder A. Epidemiology and causes of death among children in a rural area of Bangladesh. Int J Epidemiol. 1980;9(1):25–34.
Zimicki S, Nahar L, Sarder A, D’Souza S. Demographic Surveillance System Matlab. Vol. 13: Cause of Death Reporting in Matlab Source Book of Cause-Specific Mortality Rates 1975–1981.ICDDRB Scientific Report No. 63. Dhaka: ICDDRB, Bangladesh; 1985
D'Souza S. Mortality case study, Matlab, Bangladesh. Bangladesh: International Centre for Diarrhoeal Disease Research; 1985.
Bhatia S. Patterns and causes of neonatal and postneonatal mortality in rural Bangladesh. Stud Fam Plann. 1989;20:136–46. doi:10.2307/1966568. PMID:2734810.
Fauveau V, Wojtyniak B, Chowdhury HR, Sarder AM. Assessment of cause of death in the Matlab Demographic Surveillance System. 1991.
Snow R, Winstanley M, Marsh V, Newton C, Waruiru C, Mwangi I, et al. Childhood deaths in Africa: uses and limitations of verbal autopsies. The Lancet. 1992;340(8815):351–5.
Kamal GM, Streatfield K, Rahman S. Causes of death among children in Bangladesh. In: ClelandJ, EbanksGE, WaiL, Ali RashidMA, editors. Secondary analysis of Bangladesh Fertility Survey, 1989 data. Dhaka: National Institute of Population Research and Trainin. 1989
Salway S, Nasim SMA. Levels, trends and causes of mortality in children below 5 years of age in Bangladesh: findings from a national survey. J Diarrhoeal Dis Res. 1994;12:187–93.
Baqui AH, Black RE, Arifeen S, Hill K, Mitra S, al Sabir A. Causes of childhood deaths in Bangladesh: results of a nationwide verbal autopsy study. Bull World Health Organ. 1998;76(2):161.
Kalter HD, Gray RH, Black RE, Gultiano SA. Validation of postmortem interviews to ascertain selected causes of death in children. Int J Epidemiol. 1990;19(2):380–6.
Hernández B, Ramírez-Villalobos D, Romero M, Gómez S, Atkinson C, Lozano R. Assessing quality of medical death certification: concordance between gold standard diagnosis and underlying cause of death in selected Mexican hospitals. Popul Health Metr. 2011;9:38.
NIPORT, ORC Macro, Johns Hopkins University, ICDDR,B. Bangladesh Maternal Health Services and Maternal Mortality Survey 2001. NIPORT, ORC Macro, Johns Hopkins University, and ICDDR,B, Dhaka, Bangladesh and Calverton/Maryland, USA; 2003.
National Institute of Population Research and Training (NIPORT), Measure Evaluation and icddr,b. Bangladesh maternal mortality and health care survey 2010. Dhaka: NIPORT, MEASURE Evaluation and icddr,b; 2011.
Dongre AR, Singh A, Deshmukh P, Garg B. A community based cross sectional study on feasibility of lay interviewers in ascertaining causes of adult deaths by using verbal autopsy in rural Wardha. Online Journal of Health and Allied Sciences. 2009; 7(4).
Edmond KM, Quigley MA, Zandoh C, Danso S, Hurt C, Agyei SO, et al. Diagnostic accuracy of verbal autopsies in ascertaining the causes of stillbirths and neonatal deaths in rural Ghana. Paediatr Perinat Epidemiol. 2008;22(5):417–29. The Authors, Journal Compilation © 2008 Blackwell Publishing Ltd.
Baiden F, Bawah A, Biai S, Binka F, Boerma T, Byass P, et al. Setting international standards for verbal autopsy. Bull World Health Organ. 2007;85(8):570–1.
Weldearegawi B, Ashebir Y, Gebeye E, Gebregziabiher T, Yohannes M, Mussa S, et al. Emerging chronic non-communicable diseases in rural communities of Northern Ethiopia: evidence using population-based verbal autopsy method in Kilite Awlaelo surveillance site. Health Policy Plan. 2013;28(8):891–8.
World Health Organization. Geneva: World Health Organization; 2007. Verbal autopsy standards: ascertaining and attributing cause of death. http://apps.who.int/iris/bitstream/10665/43764/1/9789241547215_eng.pdf. Accessed 16 September 2015.
Quigley MA, Chandramohan D, Rodrigues LC. Diagnostic accuracy of physician review, expert algorithms and data-derived algorithms in adult verbal autopsies. Int J Epidemiol. 1999;28(6):1081–7.
Khademi H, Etemadi A, Kamangar F, Nouraie M, Shakeri R, Abaie B, et al. Verbal autopsy: reliability and validity estimates for causes of death in the Golestan Cohort Study in Iran. PLoS One. 2010;5(6), e11183.
Labrique A, Sikder S, Wu L, Rashid M, Ali H, Ullah B, et al. Beyond pregnancy–the neglected burden of mortality in young women of reproductive age in Bangladesh: a prospective cohort study. BJOG: An International Journal of Obstetrics & Gynaecology. 2013;120(9):1085–9.
Moesgaard Iburg K. Global, regional, and national age–sex specific all-cause and cause-specific mortality for 240 causes of death, 1990–2013: a systematic analysis for the Global Burden of Disease Study 2013. The Lancet; 2014.
Global Burden of Disease Study 2010. Bangladesh Global Burden of Disease Study 2010 (GBD 2010) Results 1990-2010. Seattle, United States: Institute for Health Metrics and Evaluation (IHME), 2013. Accessed 17 September 2015.
Mashreky SR, Rahman F, Rahman A. Suicide kills more than 10,000 people every year in Bangladesh. Archives of suicide research : official journal of the International Academy for Suicide Research. 2013;17(4):387–96.
Ahmed MK, van Ginneken J, Razzaque A, Alam N. Violent deaths among women of reproductive age in rural Bangladesh. Soc Sci Med. 2004;59(2):311–9.
Johnston HB, Naved RT. Spousal Violence in Bangladesh: A Call for a Public-health Response. J Health Popul Nutr. 2008;26(3):366–77.
This work was supported by the United States Agency for International Development (USAID) and the UK Department for International Development (UKaid). This work is based on the BMMS 2001 and BMMS 2010 surveys, which were carried out by the National Institute of Population Research and Training (NIPORT) with technical support from icddr,b, MEASURE Evaluation and USAID/Bangladesh. The authors like to acknowledge the support provided by Dr. Afsana Bhuiyan of icddr,b in formatting the graphs and tables in the manuscript and Professor Nitai Chakraborty, Department of Statistics, Biostatistics and Informatics, Dhaka University for his assistance in addressing reviewers comments on sample size and weighting of the results. The authors’ views expressed in this publication do not necessarily reflect the views of the United States Agency for International Development (USAID) or the United States Government.
This work was funded by the United States Agency for International Development (USAID) and the UK Department for International Development (UKaid). This work is based on the 2001 and 2010 BMMS which were funded by the Government of the People's Republic of Bangladesh, USAID and AusAID.
The authors declare that they have no competing interests.
All authors were involved in conceptualizing the manuscript. QN performed the data analysis, and drafted the manuscript. All authors (including SEA, KJ and PKS) participated in data interpretation and revision and finalization of the manuscript. All authors approved submission of the article in the present form. The corresponding author had full access to all of the data used in the study. All authors were the key investigators of the BMMS 2010 and SEA, KJ and PKS were lead investigators of the BMMS 2001. The BMMS 2001 and BMMS 2010 data files are also openly available.
Availability of data and materials
About this article
Cite this article
Nahar, Q., Arifeen, S.E., Jamil, K. et al. Causes of adult female deaths in Bangladesh: findings from two National Surveys. BMC Public Health 15, 911 (2015). https://doi.org/10.1186/s12889-015-2256-6
- Adult female mortality
- Causes of death
- Verbal autopsy