SEARCH

SEARCH BY CITATION

Keywords:

  • maternal health;
  • obstetric complications;
  • care seeking;
  • Bangladesh
  • santé maternelle;
  • complications obstétriques;
  • recours aux soins;
  • Bangladesh
  • salud materna;
  • complicaciones obstétricas;
  • búsqueda de cuidados;
  • Bangladesh

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Objective  Seeking care from a basic or comprehensive facility in response to obstetric complications is a key behaviour promoted in safe motherhood programmes. This study examined definitions of care seeking for maternal health complications used by families in rural Bangladesh, and the frequency and determinants of locally-defined care seeking practices.

Methods  We conducted 24 semi-structured qualitative interviews with women who had recently given birth to characterize care seeking behaviours in response to perceived complications. Based on these findings, a quantitative household questionnaire was developed and administered to 1490 women, half of whom reported a ‘serious or very serious’ complication during their last pregnancy and/or delivery (n = 769; 52%), and were included in the quantitative analysis.

Results  Informants described three care seeking patterns in qualitative interviews: (i) sending a family member to purchase treatment to administer in the home; (ii) sending for a provider to treat the woman in the home and (iii) taking the woman outside the home to a facility or provider's office. The quantitative survey revealed that most women sought care for ‘serious’ complications (86%), with 42% seeking multiple sources of care. The majority of women purchased a treatment to administer at home (68%), while 20% brought a provider to the home. Thirty per cent of women were taken to a provider or facility.

Conclusions  Families generally seek care for complications, but care seeking does not correspond to definitions used by maternal health programmes. Local definitions of care seeking must be considered in intervention design so that promotion of care seeking increases for facility-based care for life-threatening emergencies rather than unintentionally increasing the use of home-based treatments of little medical value for prevention of mortality.

Objectif  Le recours aux soins dans un service de base ou spécialisé pour répondre à des complications obstétriques est un comportement clé encouragé dans des programmes pour la sûreté de maternité. Cette étude a examiné les définitions de recours aux soins pour les complications de santé maternelles utilisés par les familles en zone rurale au Bangladesh, la fréquence et les déterminants des pratiques «locales» en matière de recours aux soins.

Méthodes  Nous avons mené 24 entretiens qualitatifs semi structurés avec des femmes ayant récemment donné naissance afin de caractériser des comportements de recours aux soins pour répondre à des complications perçues. Basé sur ces observations, un questionnaire quantitatif sur les ménages a été développé et présentéà 1490 femmes. La moitié d'entre elles ont rapporté une complication «sérieuse ou très sérieuse» au cours de leur dernière grossesse et/ou accouchement (n = 769, 52%) et ont été incluses dans l'analyse quantitative.

Résultats  Les informateurs ont décrit trois modes de recours aux soins dans les entretiens qualitatifs: 1) envoyer un membre de famille pour l'achat d'un traitement à administrer à domicile, 2) envoyer chercher un praticien pour traiter la femme à domicile et 3) amener la femme hors du domicile pour être vue dans le cabinet d'un praticien ou dans un service. L'enquête quantitative a indiqué que la plupart des femmes ont fait recours aux soins pour des complications «sérieuses» (86%), avec 42% d'entre elles ayant recherché multiples sources de soin. La majorité des femmes ont acheté un traitement à administrer à domicile (68%) alors que 20% ont fait appel à un praticien à venir à domicile. 30% des femmes ont été amenées chez un praticien ou dans un service.

Conclusions  Les familles font recours aux soins en général pour des complications, mais les recours aux soins ne correspondent pas aux définitions utilisées par des programmes de santé maternelle. Des définitions locales de recours aux soins devraient être considérées dans la conception des interventions afin de promouvoir l'augmentation du recours aux soins dans les services pour des urgences vitales plutôt que d'augmenter involontairement l'utilisation de traitements à domicile qui ont peu de valeur médicale pour la prévention de la mortalité.

Objetivo  La búsqueda de cuidados como respuesta a complicaciones obstétricas, en una instalación básica o integral, es un comportamiento clave promovido en programas para una maternidad segura. En este estudio se examinaron definiciones de búsqueda de cuidados para complicaciones en la salud materna utilizados por familias en Bangladesh rural, y la frecuencia y los determinantes de prácticas de búsqueda de cuidados localmente definidos.

Métodos  Realizamos 24 entrevistas semi-estructuradas a mujeres que habían dado a luz recientemente, con el fin de caracterizar los comportamientos de búsqueda de cuidados como respuesta a la percepción de complicaciones. Basados en estos hallazgos, se desarrolló un cuestionario cuantitativo para los hogares, y se administró a 1,490 mujeres, la mitad de las cuales reportaron complicaciones ‘serias o muy serias’ durante su último embarazo y/o parto (n = 769; 52%), y se incluyeron dentro del análisis cuantitativo.

Resultados  Los informantes describieron tres patrones de búsqueda de cuidados en las entrevistas cualitativas: 1) enviar a un miembro de la familia a comprar el tratamiento para ser administrado en casa; 2) enviar a buscar a un proveedor que trate a la mujer en casa; y 3) llevar a la mujer fuera del hogar a un centro sanitario o al despacho de un proveedor. La encuesta cuantitativa reveló que la mayoría de las mujeres buscó cuidados para complicaciones ‘serias’ (86%), con un 42% buscando múltiples fuentes de cuidados. La mayoría de las mujeres compraron tratamiento para ser administrado en casa (68%), mientras un 20% trajo el proveedor a casa. 30% de las mujeres fueron llevadas hasta un proveedor o centro sanitario.

Conclusiones  Generalmente las familias buscan cuidados frente a las complicaciones, pero la búsqueda de cuidados no se corresponde con las definiciones utilizadas por programas de salud materna. Las definiciones locales de búsqueda de cuidados debe considerarse en el diseño de una intervención, de manera que la promoción de búsqueda de cuidados aumente en las emergencias potencialmente fatales y basados en un centro sanitario, en vez de un aumento no intencionado del uso de tratamientos realizados en el hogar, con poco valor médico para la prevención de la mortalidad.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

The Millennium Development Goal (MDG) for maternal mortality is ‘to reduce the maternal mortality ratio by three-quarters by the year 2015’ from 1990 levels (United Nations 2004; Khan et al. 2006). Achieving this ambitious goal requires strengthening of preventive interventions at the community level, ensuring high-quality basic and comprehensive obstetric care, and promoting timely care seeking from these facilities for maternal emergencies (Starrs 1997). Care seeking is in many ways the cornerstone of efforts to reduce maternal mortality, and yet an evidence base for how to best promote care seeking in different settings is lacking (Miller et al. 2003; Stanton 2004).

The concept of ‘care seeking’ often has been defined in narrow terms in maternal health, with ‘care’ denoting services provided by professionals with appropriate life-saving skills, and ‘seeking’ denoting transfer of the woman from the home to a health facility. Local definitions of care seeking can differ considerably from definitions used by programmes, making it difficult both to plan interventions to promote care seeking, and to accurately measure the impact of these interventions. These definitions need to be incorporated in the design of household surveys to track progress toward achieving the MDG for maternal mortality, and taken into account in the design of interventions to promote care seeking.

In a recent national survey on maternal health in Bangladesh, the majority of women reported complications during pregnancy and childbirth, but few reported that they sought care from medically trained providers in health facilities, even if they perceived the complication to be life-threatening (National Institute of Population Research and Training (NIPORT), ORC Macro, Johns Hopkins University, & ICDDR, B 2003). Most women reported accessing care in the home or seeking care from sites other than health facilities (National Institute of Population Research and Training (NIPORT), ORC Macro, Johns Hopkins University, & ICDDR, B 2003).

This paper examines: (i) definitions of care seeking for maternal health complications used by families in rural Bangladesh; (ii) the frequency and determinants of locally-defined care seeking practices and (iii) the implications of local definitions of care seeking for monitoring maternal health interventions and strategies to promote care seeking.

Methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

We present findings from semi-structured interviews and a household survey in three sub-districts of Sylhet District in north-eastern Bangladesh: Beanibazar, Zakiganj and Kanaighat. These data were collected as part of the Projahnmo study, a cluster-randomized trial from mid-2003 to December 2005 testing a package of maternal and neonatal interventions to reduce neonatal mortality in these three sub-districts (Baqui & Ahmed 2004; Winch et al. 2005). The data in this paper are from a sub-study in 2005 on the effects of the interventions on maternal morbidity.

Qualitative interviews

Twenty-four in-depth semi-structured interviews were conducted with women who had given birth 6–12 months prior to the interview. Women were identified based on a 2004 household survey for Projahnmo and were included in the sampling frame if they reported at least one complication. The selected sample was categorized into one of six strata based on distance to a comprehensive obstetric facility. Four women were randomly selected from each stratum. In the in-depth semi-structured interviews, women were asked about their most recent birth experience, any complications experienced during pregnancy, childbirth and the post-partum period, care seeking behaviours during each of these phases and costs of treatments. If the woman reported more than one complication in a given phase of the birthing process, she was asked to select the ‘most serious’. Care seeking behaviours were ascertained for the ‘most serious’ complications within each phase that a given woman had experienced. Notes and audio recordings from the interviews were transcribed and translated from the local language into English. Analysis included several rounds of manual coding of the transcripts and discussions with the research team using the constant comparative method to synthesize the data (Boychuk Duchscher 2004). Based on these qualitative findings, a series of questions was developed to quantitatively measure each of the three local patterns of care seeking identified.

Quantitative household survey

A pregnancy surveillance system was established in 2005 for the maternal morbidity sub-study within the Projahnmo study area. Female interviewers went to each household to identify women with a live or stillbirth 2–16 weeks prior to the visit. Women identified through the pregnancy surveillance system (n = 1490) were administered a structured questionnaire to ascertain background information, socioeconomic status, birth history, complications and care seeking behaviours. Women were prompted on 27 specific complications during each phase of the birthing process (pregnancy, childbirth and post-partum). If a woman reported more than one complication, she was asked to specify the complication which occurred last. Women were then asked about perceptions of the severity of that condition (not at all serious, slightly serious, serious or very serious so she thought she might die) and subsequent care seeking behaviours. Women who reported their last condition to be serious or very serious were included in the analysis (n = 769; 51.6%).

Women were asked if someone purchased a medicine/treatment to administer in the home, and if yes, what type of treatment was purchased. They were then asked if anyone came to the home to treat the problem, and if yes, what type of health provider was brought to the home. Finally, women were asked if they went outside the house to treat their problem, who decided to seek care, and what type of care was sought. All responses were unprompted. They were pre-tested and feedback was reviewed with the study team. These questions were further tested and refined prior to initiating data collection.

Data analysis

Data were reviewed for accuracy, consistency and completeness and entered into a database using Visual FoxPro version 9.0 (Microsoft Corp., Redmond, WA, USA). Errors in data collection and entry were identified using range and consistency checks. Descriptive analysis of care seeking behaviours among women who reported serious complications was conducted using stata version 8.0 (Stata Corp. Inc., College Station, TX, USA).

Ethical approval

Ethical approval for this study was obtained from the Committee on Human Research at the Johns Hopkins Bloomberg School of Public Health and The Ethical Review Committee at ICDDR,B in Dhaka, Bangladesh. All participants gave informed consent prior to enrolment.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Qualitative semi-structured interviews

Background characteristics of the sample are provided in Table 1. Women (n = 24) reported a total of 90 complications during pregnancy, childbirth and the post-partum period, with 61 complications judged by the women to be ‘serious’. Care was sought for 90% of these ‘serious’ complications. The most common complications reported were lower abdominal pain, fever, prolonged labour, uterine prolapse, bleeding and swollen legs.

Table 1.   Per cent distribution of background characteristics among women with perceived ‘serious’ complication by type of study and by characteristic, Sylhet District, Bangladesh 2005
CharacteristicSemi-structured interviews n (%)Household survey n (%)
Mean age (SD)26 years 27 years (5.8)
Mean parity (SD) 3 children  3.8 children (2.3)
Education
 None14 (58.3)275 (35.8)
 Primary 4 (16.7)282 (36.7)
 Secondary or more 6 (25.0)212 (27.6)
Sub-district
 Beanibazar (more affluent)12 (50.5)360 (46.8)
 Zakiganj (more rural/poorer)10 (41.7)208 (27.1)
 Kanaighat (more rural/poorer) 2 (8.3)201 (26.1)
Location of last delivery
 Home20 (83.3)656 (85.3)
 Facility 4 (16.7)113 (14.7)
Total24 (100)769 (100)

Three distinct patterns of seeking care emerged from the interviews: (i) sending a male family member to purchase treatment to be administered in the home; (ii) sending for a provider to treat the woman in the home and (iii) taking the woman outside the home to a provider based in a health facility or private office. Of the three patterns, only the third one involved taking the woman outside the home for care. Women often reported seeking multiple sources of care (40%). There was no apparent relationship between type of complication and pattern of care seeking.

Sylhet District is characterized by a pluralistic healthcare system (Parkhurst et al. 2005). There are a variety of traditional and medically trained providers who are consulted for maternal complications. The traditional or informal sector is comprised of spiritual healers, ‘village doctors’ (gram daktar), homeopaths, traditional birth attendants and pharmacies where a variety of medications are sold. Homeopaths are trained in a 3–4 year government programme and provide treatments, such as sugar, sugar balls and tonics. Village doctors and most operators of pharmacies have no or limited formal training, although both were commonly referred to as ‘doctors’ in interviews. Both village doctors and operators of pharmacies provide a range of allopathic medicines and injections. The formal health system is also quite diverse, with a mixture of private facilities, government facilities and doctors’ offices. Providers with medical qualifications to manage obstetric complications include doctors, nurses and midwives.

The most prevalent pattern of care seeking was purchasing medicines or other treatments to administer in the home. Typically, a male family member was sent to a provider close to home, described the woman's symptoms and purchased the medicine or treatment offered. Family members sought treatments from a variety of trained and untrained providers; common treatments included allopathic medicines from operators of pharmacies and from medically trained providers, treatments from spiritual healers (amulets, blessed water) and homeopathic medicines. The median cost of these treatments was US $1, including fees, medication and transportation costs. Families often purchased only partial amounts of medications based on their ability to pay. A woman with post-partum fever stated:

I did not go to any doctor, my husband went to pharmacy and discussed with him regarding my health problem. Then he gave my husband some medicine. He bought 10 to 20 Taka's [US $0.15–30] of medicine. I took this medicine for only 3 to 4 days. Then I was cured. (Mother of four children, near comprehensive obstetric care facility)

Relatively few women reported receiving care from a provider in the home. Providers, most often spiritual healers, homeopathic doctors and operators of pharmacies, were brought to the home if the woman was unable to easily travel. Women also reported bringing medically trained providers to the home. One woman with bleeding during pregnancy reported:

When I noticed that my raw blood was discharging, then I told my husband and he went to a moulana [spiritual healer] and brought him to the home. He gave me spiritual water but I was not cured… my husband again went to the moulana and he… did rocter ban marche [black magic regarding my blood]… the moulana gave me 4 amulets, spiritual water, spiritual oil… after that I was cured. (Mother of four children, far from comprehensive obstetric care facility)

Women also described leaving the home to seek care in a health facility or a provider's home/office. A variety of providers was consulted, including operators of pharmacies, spiritual healers, homeopaths and medically trained doctors. Seeking this type of care was influenced by type of illness, reputation of provider, the woman's previous experiences and the family's ability to pay. The median cost of seeking treatment outside the home was US $11, including provider fees, medication and transportation costs. One woman explained:

At third day of pain [labour pain] I was taken to the hospital… During my last visit to the hospital, the doctor suggested to my mother, mother in law, and husband that next time, when I have ‘that pain’, they must take me to the hospital. They were also scared because the first baby was dead during delivery… After arriving in the hospital, they pushed saline and injection to increase the pain. I went in the evening to the hospital and in the night I delivered the child. (Mother with one child, near comprehensive obstetric care facility)

Household survey

Of the 1490 respondents, 886 reported at least one complication during pregnancy, childbirth or after delivery. Thirteen per cent of women reported their complication to be not at all serious or slightly serious; while 86.8% (n = 769) perceived their complication to be serious or very serious and were included in the analysis. Background characteristics of the sample are provided in Table 1.

Of the 769 women, 19.5% reported one complication and 80.5% reported multiple complications (Table 2). Of women with serious complications, 85.6% reported seeking care and 42.0% reported seeking multiple sources of care.

Table 2.   Among women with perceived ‘serious’ complications*, percentage of type of complication, Sylhet District, Bangladesh 2005 (n = 769)
Reported serious complication%n
  1. *All complications occurred during pregnancy, childbirth and/or after delivery.

  2. †Gastrointestinal problems includes gastritis, oral thrush, abdominal pain, hyperemesis, vomiting, diarrhoea/dysentery.

  3. ‡Urinary tract infection includes pain in urination, stop urination, urine infection, excess urination.

  4. §Includes retained placenta and upward displacement of the placenta.

  5. ¶Other includes weakness, vaginal discharge (non-foul smelling), respiratory problems, foetal movement change, anorexia, chest pain, dizziness, anaemia, premature rupture of membranes, general oedema, leg pain, itching body, stitch pain, piles, scabies, pox, tineasis, bleeding per leg and acne.

Severe headache47.6366
Blurred vision36.8283
Post-partum pain/cramping pain34.7267
Swelling of leg34.2263
Irregular labour pain during delivery36.5281
Lower abdominal pain24.4188
Cramping of hands/legs19.6151
Swelling of face/hand16.0123
Bleeding13.3102
High fever/temperature13.1101
Gastrointestinal problems†6.056
Labour pain for more than 12 h5.643
Symptoms of urinary tract infection‡5.542
High blood pressure4.434
Convulsions3.225
Loss of consciousness3.124
Foul smelling discharge2.620
Torn uterus1.814
Vaginal tear/tear in birth canal during delivery or after delivery1.713
Retained placenta§1.310
Hand or leg prolapsed during delivery0.86
Obstructed delivery due to placenta or big head of baby0.43
Prolapsed uterus0.11
Others¶11.891
Total 769

The most common pattern of care seeking was bringing medicine and/or treatment to the home (67.7%). Women and their families tended to seek this type of treatment for post-partum/cramping pain (90.9%), bleeding (70.2%), symptoms of infection (71.4%) and prolonged/obstructed labour (77.3%) (Table 3). Almost half of the women reported taking tablets/capsules to address the complication (44.5%), while few women reported receiving an injection to speed up labour (2.1%) (Table 4).

Table 3.   Among women with perceived ‘serious’ complications*, per cent who sought care by type of care and type of last complication, Sylhet District, Bangladesh 2005 (n = 769)
ComplicationBrought treatment to home (%)Brought provider to home (%)Sought care outside home (%)Total n
  1. *All complications occurred during pregnancy, childbirth and/or after delivery.

  2. †Pre-eclampsia/eclampsia includes severe headache, blurred vision, high blood pressure, swelling of hand, puffy face, convulsions, vertigo.

  3. ‡Complications of labour and delivery includes hand or leg prolapsed during delivery, ‘irregular’ labour pain.

  4. §Sepsis includes foul smelling discharge, high fever.

  5. ¶Prolonged/obstructed labour includes labour pain >12 h, obstructed labour due to placenta or big head of baby.

  6. **Includes retained placenta and upward displacement of the placenta.

  7. ††Gastrointestinal problems includes gastritis, oral thrush, abdominal pain, severe abdominal pain, hyperemesis.

  8. ‡‡Other includes prolapsed uterus, torn uterus, vaginal tear, weakness, vaginal discharge (non-foul smelling), respiratory problems, foetal movement change, anorexia, chest pain, dizziness, anaemia, premature rupture of membranes, general oedema, leg pain, itching body, stitch pain, piles, scabies, pox, tineasis, bleeding per leg and acne.

Symptoms of pre-eclampsia/eclampsia†50.519.242.499
Complications of labour and delivery‡62.567.240.367
Swelling/cramping of leg47.69.555.663
Symptoms of sepsis§71.417.923.256
Bleeding70.227.727.747
Symptoms of urinary tract infections57.116.742.942
Cramping/post-partum pain90.93.03.033
Prolonged/obstructed labour¶77.359.154.522
Lower abdominal pain64.711.870.617
Loss of consciousness62.562.518.716
Retained placenta**40.060.020.05
Gastrointestinal problems††50.00.050.02
Other‡‡74.69.121.3342
Total67.719.930.3769
Table 4.   Among women with perceived ‘serious’ complications, per cent distribution who sought care by type of care, Sylhet District, Bangladesh 2005 (n = 769)
Type of care%n
  1. *Includes trained and untrained traditional birth attendants.

  2. †Includes doctor, nurse, midwife, paramedic, female health visitor and medical officer.

  3. ‡Includes Sylhet Medical College Hospital, Upazila (sub-district) Health Complex, Maternal Child Welfare Centre.

  4. §Includes Family Welfare Centre.

  5. ¶Includes home of homeopathic doctor, quack doctor, herbal medicine doctor and village doctor.

Type of treatment brought to home
 Tablet/capsule44.5342
 Homeopathic6.852
 Saline/injection through saline6.651
 Intramuscular injection to speed labour2.116
 Spiritual1.713
 Other6.147
 No treatment brought to the home32.2248
Type of provider brought to home
 Village doctor7.356
 Traditional birth attendant*5.341
 Medically trained provider†3.225
 Health assistant0.43
 Female cleaner (Maid)0.43
 Projahnmo community health worker0.32
 Herbalist0.11
 Other2.922
 No provider brought to home80.2616
Location of care outside the home
 Comprehensive obstetric care facility‡11.085
 Home of medically trained provider†8.767
 Basic obstetric care facility§2.519
 Home of untrained provider¶2.721
 Home of traditional birth attendant*0.32
 Satellite clinic/vaccination centre0.32
 Other4.837
 No care sought outside home69.7536
 Total100%769

Few women reported that a provider was brought to the home to treat the complication (19.9%). Providers brought to the home most often treated complications of labour and delivery (67.2%) and loss of consciousness (62.5%) (Table 3). Village doctors (7.3%) and traditional birth attendants (5.3%) were most common, with few families bringing a medically trained provider to the home (3.2%) (Table 4).

Thirty per cent of women sought care outside the home in a health facility or in a provider's office/home. This type of care was most prevalent for lower abdominal pain (70.6%), swollen legs/cramping of legs (55.6%), prolonged/obstructed labour (54.5%) and symptoms of pre-eclampsia/eclampsia (42.4%) (Table 3). The most common locations for care seeking outside the home were basic or comprehensive obstetric care facilities (13.5%) and the home of a medically trained provider (8.7%). Few women sought care outside the home from untrained providers (Table 4).

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Promoting recognition of complications and appropriate care seeking behaviours is a key component of safe motherhood programmes, especially in settings, such as Bangladesh where home-based birth is the norm. In a household survey (n = 769), we found that almost all women (86%) who reported ‘serious’ complications sought some type of care for these conditions, and most often (68%) a family member brought treatment to the home. Occasionally (20%) a provider was summoned to the home, and less than one-third (30%) sought care outside the home with a provider based in a health facility or private office. These findings are similar to a recent national survey. Among women who perceived ‘life-threatening’ complications, 61.8% sought care, with 32.2% seeking care at home and 29.6% seeking care outside the home (National Institute of Population Research and Training (NIPORT), ORC Macro, Johns Hopkins University, & ICDDR, B 2003). Although home-based care by non-trained providers may not directly cause harm to women, this type of care is of limited utility as it cannot prevent mortality from life-threatening complications, such as haemorrhage, eclampsia, obstructed labour, sepsis or complications of abortion, and thus delays effective treatment (Starrs 1997; Sibley et al. 2004; World Health Organization 2005). These findings have important implications in terms of social and cultural issues, healthcare delivery and access, and safe motherhood programmes and research.

Social and cultural issues:

In this study, the majority of women sought care for perceived serious complication at home from unqualified providers. There are several advantages to home-based care in terms of cost, type of treatment and local understandings of illness causation. In the qualitative semi-structured interviews, the median cost of home-based treatments was significantly less expensive (US $1) than care received in health facilities and/or provider's offices/homes (US $11). In addition, families were able to purchase partial amounts of medications, decreasing their financial burden. Further, families had control over both the type of treatment provided and the amount of money they spent. Some women preferred homeopathic treatments that are perceived to be weaker and more easily tolerated during pregnancy, similar to perceptions regarding homeopathic treatment for sick newborns in Bangladesh (Ahmed et al. 2001; Winch et al. 2005).

Home-based treatment allows the family to protect the woman from perceived threats, such as possession by malevolent sprits and the evil eye. In Sylhet, as in other parts of Africa, Asia and Latin America, pregnant women are deemed susceptible to malevolent spirits (bhut, upri) that can set upon them and cause ill health and even death, to the woman and her newborn (Blanchet 1984; Jeffrey et al. 1989; Afsana & Rashid 2000; Van Hollen 2003; Winch et al. 2005). By receiving spiritual water, oil and amulets in the home, women attempt to reduce their exposure to these perceived threats.

Healthcare delivery and access:

Some women reported seeking care outside the home (30.3%), the majority with medically trained providers (22.5%). Women preferred private offices of doctors, and this may be because these offices offered additional privacy or care perceived as better in quality. In the qualitative interviews, women who sought this type of care were from the more affluent part of the study area (Beanibazar sub-district) and had higher education. Education is an important factor in seeking care from medically trained providers (Chakraborty et al. 2003). Women cited the importance of the reputation of the provider as well as their previous experiences as important influences in seeking care outside the home. As noted in the case of Ayesha (Box 1), she chose to seek care with a medically trained doctor because her sister had previously died in childbirth, and she had the economic means to seek care in a private health facility. Some women did receive allopathic medicine or brought a medically trained provider to the home. These patterns of care seeking allow the women to receive some of the benefits of professional care while maintaining control over costs, type of treatment and limiting exposure to spirits and other perceived risks. These patterns have also been observed in India (Van Hollen 2003).

Table Box 1. 
Ayesha is 27 years old, lives in the central part of the study site, closer to the main road and to health facilities, and has 8 years of education. Her husband works in the Middle East, and she returned to her natal home for childbirth. This was her first pregnancy, and she gave birth at a private clinic in the town of Sylhet. During pregnancy, Ayesha went to the health facility for regular antenatal care check-ups. At 7 months of pregnancy, she reported pete bedna (strong pain in her abdomen) and was taken to the government health facility where she was referred to a private clinic. Ayesha was examined by a female doctor and given saline and an injection. The next morning, the doctor suggested a Caesarean section because Ayesha was 7 months pregnant, but Ayesha and her family refused. Ayesha's child was overdue. Her sister went to the government health facility and was told to wait two more days. After 2 days, the doctor suggested going to a private clinic where Ayesha received saline and an injection from a female physician. When labour did not start, a Caesarean section was performed. Ayesha had no problems during the post-partum period.

Safe motherhood programmes

These findings have important implications for safe motherhood programmes in terms of definitions and measurement. Families in rural Bangladesh go to great lengths to seek care for perceived obstetric complications, but these care seeking behaviours are not always appropriate and may cause delays in receiving life-saving care. Formative research is essential to elucidate local definitions of ‘care seeking’ and to understand social and cultural barriers to seeking care from medically trained providers. These local definitions need to be considered during programme design to deter interventions that may inadvertently increase care seeking in the home when their recommendations about the types of care to be sought for which complications are non-specific. Furthermore, programme evaluation activities need to incorporate these local definitions into questionnaires to more accurately monitor programme progress.

Research

Additional research is needed to design more effective programme messages around care seeking. In this study, 44.5% of women reported taking tablets/capsules at home for perceived complications. Some women reported receiving saline injections at home (6.5%), with a few receiving an ‘injection to speed up labour’ (2.1%). Further research is needed to understand the types of medicines purchased, for which complications and the training level of providers administering these treatments.

Safe motherhood programmes work under the assumption that recognition of ‘serious’ complications will prompt care seeking from basic or comprehensive obstetric care facilities (Thaddeus & Maine 1994; Nachbar et al. 1998; Yassin et al. 2003; Fronczak et al. 2005). In this study, although 87% of women reported a ‘serious’ or ‘very serious’ complication, the majority sought care at home, as illustrated in the case of Fatima (Box 2). Further research using both qualitative and quantitative methods is needed to more clearly understand women's perceptions of ‘severity’, as well as triggers for and barriers to subsequent care seeking behaviours.

Table Box 2. 
Fatima is 30–35 years old, lives in a more remote part of the study area, and has no education. She has been pregnant seven times and has four living children. Her last delivery was at home with a dhonni/dhoroni (traditional birth attendant). During pregnancy, she suffered from amasha (dysentery). Her husband first consulted a moulana (spiritual healer) who prescribed an amulet to protect Fatima from upri (literally wind /spirits ‘from above’). When the dysentery persisted, her husband brought an operator of a pharmacy to the home to administer i.v. saline. Finally, Fatima's husband accompanied her to a pharmacy (a relative) for further treatment, and the condition was cured. During childbirth, Fatima reported kacha rocoto (raw blood discharge) 3 days before delivery, and after the baby was born, her uterus, placenta and intestinal coil came out together. Once the placenta was delivered, the dhonni (TBA) told Fatima's husband to bring medicine from the ‘doctor.’ Her husband went to the pharmacy (the same as the one used to treat the dysentery) to purchase medicine. The traditional birth attendant replaced the uterus and intestinal coil, and the medicine helped with the pain. Fatima did not report any specific problems during the post-partum period.

This study may be limited by several factors. First, these findings are based on women's self-reports of ‘serious’ complications which tend to over- or under-estimate medically-diagnosed complications (Stewart & Festin 1995; Danel et al. 1996; Ronsmans et al. 1997; Seoane et al. 1998; Fortney & Smith 2000; Sloan et al. 2001). Although self-reports cannot be used to accurately estimate incidence or prevalence of complications, they are useful in relation to care seeking behaviours (Fortney & Smith 2000; Yassin et al. 2003). Secondly, women who reported multiple complications in the household survey were asked about the ‘last complication’ in regards to care seeking behaviours. This may have biased our findings as complications other than the last complication may have been more serious or prompted different care seeking behaviours. Other questions on ‘the most serious complication’ were pre-tested, but were too difficult for study respondents. Finally, this analysis only included women with ‘serious’ or ‘life-threatening’ complications, and care seeking behaviours for less serious conditions were excluded.

Women suffer needlessly from maternal morbidity and mortality in Bangladesh and in many resource poor settings. There is a need to incorporate these recommendations into safe motherhood policies and programmes to more accurately document care seeking for perceived complications in an effort to achieve MDG5.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

We thank the many individuals in Sylhet District of Bangladesh who gave their time generously and the field and data management staff of Projahnmo. We thank the Bangladesh Ministry of Health and Family Welfare colleagues at the sub-district, district and central levels and the members of the Shimantik executive committee for their valuable help and advice. The critical innovative inputs of Projahnmo Maternal Morbidity Study Group members are acknowledged. Projahnmo Maternal Morbidity Study Group (in alphabetical order): Jahiruddin Ahmed, Saifuddin Ahmed, Nabeel Ashraf Ali, Ahmed Al-Kabir, Arif Billah Al-Mahmud, Tariq Anwar, Nazma Begum, Robert E Black, Mohiuddin Chowdhury, Milan Krishna Das, Zafar Ahmad Hakim, AKM Fazlul Haque, Daniel Hossain, Syed Moshfiqur Rahman, Qazi Sadequr Rahman, Mathuram Santosham, Ashrafuddin Siddik. This study was supported by the Office of Health, Infectious Diseases and Nutrition, Global Health Bureau, United States Agency for International Development (USAID) (award HRN-A-00-96-90006-00. The opinions expressed herein are those of the authors and do not necessarily reflect the views of USAID.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • Afsana K & Rashid SF (2000) Discoursing Birthing Care: Experiences from Bangladesh. University Press Limited, Dhaka, Bangladesh.
  • Ahmed S, Sobhan F, Islam A & Barkat-e-khuda (2001) Neonatal mortality and care-seeking behaviour in rural Bangladesh. Journal of Tropical Pediatrics 47, 98105.
  • Baqui A & Ahmed S (2004) Community-Based Interventions to Reduce Neonatal Mortality in Bangladesh – Projahnmo I Baltimore. Johns Hopkins University, ICDDR,B, Dhaka, Bangladesh.
  • Blanchet T (1984) Meanings and Rituals of Birth in Rural Bangladesh. United Press Limited, Dhaka, Bangladesh.
  • Boychuk Duchscher JE (2004) Grounded theory: reflections on the emergence vs. forcing debate. Journal of Advanced Nursing 48, 605612.
  • Chakraborty N, Islam MA, Chowdhury RI et al. (2003) Determinants of the use of maternal health services in rural Bangladesh. Health Promotion International 18, 327337.
  • Danel I, Ponce de Leon R & Lazado P (1996) Validation Study of Responses to the Maternal Morbidity Module of the 1994 Ecuadorian Reproductive Health Survey. In: Seminar Report, Seminar on Innovative Approches to the Assessment of Reproductive Health, Manila, 24 27 September 1996 . International Union for the Scientific Study of Population, Paris, France.
  • Fortney JA & Smith JB (2000) Measuring maternal morbidity. In: Safe Motherhood Initiatives: Critical Issues (eds MBerer & TKSRavindran) Reproductive Health Matters Special Issue, London, pp. 4350.
  • Fronczak N, Antelman G, Moran AC et al. (2005) Delivery-related complications and early postpartum morbidity in Dhaka, Bangladesh. International Journal of Gynecology and Obstetrics 91, 271278.
  • Jeffrey PM, Jeffrey R & Lyon A (1989) Labour Pains and Labour Power: Women and Childbearing in India. Zed Press, London.
  • Khan KS, Wojdyla D, Say L et al. (2006) WHO analysis of causes of maternal death: a systematic review. Lancet 367, 10661074.
  • Miller S, Sloan NL, Winikoff B et al. (2003) Where is the ‘E’ in MCH? The need for an evidence-based approach in safe motherhood. Journal of Midwifery Womens Health 48, 1018.
  • Nachbar N, Baume C & Parekh A (1998) Assessing Safe Motherhood in the Community: A Guide to Formative Research. JSI/MotherCare, Washington, DC.
  • National Institute of Population Research and Training (NIPORT), ORC Macro, Johns Hopkins University, & ICDDR, B (2003) Bangladesh Maternal Health Services and Maternal Mortality Survey 2001. NIPORT, ORC Macro, Johns Hopkins University, and ICDDR,B, Dhaka, Bangladesh and Calverton, MD.
  • Parkhurst JO, Penn-Kekana L et al. (2005) Health systems factors influencing maternal health services: a four-country comparison. Health Policy 73, 127138.
  • Ronsmans C, Achadi E, Cohen S et al. (1997) Women's recall of obstetric complications in south Kalimantan, Indonesia. Studies in Family Planning 28, 203214.
  • Seoane G, Castrillo M & O'Rourke K (1998) A validation study of maternal self reports of obstetrical complications: implications for health surveys. International Journal of Gynecology and Obstetrics 62, 229236.
  • Sibley L, Sipe TA & Koblinsky M (2004) Does traditional birth attendant training improve referral of women with obstetric complications: a review of the evidence. Social Science & Medicine 59, 17571768.
  • Sloan NL, Amoaful E, Arthur P et al. (2001) Validity of women's self-reported obstetric complications in rural Ghana. Journal of Health Population and Nutrition 19, 4551.
  • Stanton CK (2004) Methodological issues in the measurement of birth preparedness in support of safe motherhood. Evaluation Review 28, 179200.
  • Starrs A (1998) The Safe Motherhood Action Agenda: Priorities for the Next Decade. Report on the Safe Motherhood Technical Consultation. Colombo, Sri Lanka. 18 23 October 1997 . Safe Motherhood Inter-Agency Group/Family Care International, New York.
  • Stewart MK & Festin M (1995) Validation study of women's reporting and recall of major obstetric complications treated at the Philippine General Hospital. International Journal of Gynecology and Obstetrics 48(Suppl), 0S530S66.
  • Thaddeus S & Maine D (1994) Too far to walk: maternal mortality in context. Social Science & Medicine 38, 10911110.
  • United Nations (2004) Millennium Development Goals. http://www.un.org/millenniumgoals/ (accessed 24 July 2004).
  • Van Hollen CC (2003) Birth on the Threshold: Childbirth and Modernity in South India. University of California Press, Berkeley, CA.
  • Winch PJ, Alam MA, Akther A et al. (2005) Local understandings of vulnerability and protection during the neonatal period in Sylhet district, Bangladesh: a qualitative study. Lancet 366, 478485.
  • World Health Organization (2005) The World Health Report 2005: Make Every Mother and Child Count. World Health Organization, Geneva, Switzerland.
  • Yassin K, Laaser U & Kraemer A (2003) Maternal morbidity in rural upper Egypt: levels, determinants, and care seeking. Health Care for Women International 24, 252267.