Type, Content, and Source of Social Support Perceived by Women during Pregnancy: Evidence from Matlab, Bangladesh.

2011·0.34 MB·English

by  Edmonds, Joyce K.

J HEALTH POPUL NUTR 2011 Apr;29(2):163-173 ©INTERNATIONAL CENTRE FOR DIARRHOEAL
ISSN 1606-0997 | $ 5.00+0.20 DISEASE RESEARCH, BANGLADESH
Type, Content, and Source of Social Support
Perceived by Women during Pregnancy: Evidence
from Matlab, Bangladesh
Joyce K. Edmonds1, Moni Paul2, and Lynn M. Sibley3
1Center for Research on Maternal and Newborn Survival, Emory University, Atlanta, GA, USA, 2Reproductive Health Unit,
Public Health Sciences Division, ICDDR,B, GPO Box 128, Dhaka 1000, Bangladesh, and 3Center for Research on
Maternal and Newborn Survival, Emory University, Atlanta, GA, USA
ABSTRACT
Specific and contextualized data on social support during distinct health events are needed to improve
social support interventions. This study identified the type, content, and source of social support perceived
by women during pregnancy. In-depth interviews with 25 women, aged 18-49 years, living in Matlab,
Bangladesh, were conducted. The findings demonstrated that women perceived, the receipt of eight distinct
types of support. The four most frequently-mentioned types included: practical help with routine activities,
information/advice, emotional support and assurance, as well as the provision of resources and material
goods. Sources varied by type of support and most frequently included-—mothers, mothers-in-law, sisters-
in-law, and husbands. Examples depicting the content of each type of support revealed culturally-specific
issues that can inform community-based social support interventions.
Key words: Cross-sectional studies; Pregnancy; Qualitative studies; Reproductive health; Retrospective
studies; Social support; Stress; Bangladesh
INTRODUCTION (7). However, research on social support has over-
whelmingly focused on developing countries in a
Social support is broadly defined as the resources
Western context employing generic measures that,
provided by other persons (1) and can be concep-
despite their psychometric properties, lack rele-
tualized as the function of one’s network (2,3). So-
vance to a particular population (3). Little is known
cial support is considered one of the mechanisms
about the precise nature of social support in South-
through which social networks are thought to affect
East Asian contexts during pregnancy. This paper
health (4). Yet, social support is not a uni-dimen-
presents findings of a qualitative study designed
sional, fixed set of resources, rather it is situation-,
to explore the specific dyadic characteristics (type,
issue- and context-specific (5,6). There is a need to
content, and source) of women’s perceived social
operationally define the concept of social support
support during pregnancy in Matlab, Bangladesh.
in a way that adheres to meanings prescribed by
people with direct experience of the phenomena Pregnancy is typically considered a vulnerable pe-
under study. Thus, qualitative work that explores riod for women. One important risk factor affect-
social support in distinct cultural settings during ing maternal well-being is lack of social support
specific life-events is an important first step in elu- (8). Low social support is associated with low birth-
cidating the influence of social support networks weight (9,10), poor labour progress (11), preterm
on health behaviours and subsequent outcomes labour (8,12), neural tube defects (13), depression
and anxiety (11,14). In addition, social support net-
Correspondence and reprint requests should be
addressed to: works affect pregnancy-related health behaviours
Dr. Joyce K. Edmonds and lifestyle habits, including use of prenatal and
Nell Hodgson Woodruff School of Nursing delivery services and dietary habits (15,16). The as-
Emory University
sociation between social support and pregnancy
1520 Clifton Road NE
outcomes is complex, involving the psychological
Atlanta, GA 30322
and biological (i.e. hormones and immune media-
USA
Email: [email protected] tors) response to life-events and stress.
Social support during pregnancy in Bangladesh Edmonds JK et al.
Overall, the evidence suggests that social support is in cities and abroad. Kinship networks, however, re-
protective during pregnancy. Pregnant women need main a universal aspect of the social structure, and
the support of family members, friends, and health division of labour is predominantly gender-based
professionals. However, two systematic Cochrane in rural villages with exceptional patterns among
reviews reported no benefit of interventions pro- the poorest and most educated women.
viding ‘additional’ support during pregnancy either
Matlab has a unique maternal healthcare infra-
for reducing premature birth (17) or for preventing
structure due to the renowned efforts of the Ma-
postpartum depression (18). One explanation for
ternal and Child Health and Family Planning Pro-
these negative findings is that the intervention tri-
als did not account for the functional, qualitative gramme (MCH-FP) of ICDDR,B serving one half of
aspects of social support that are grounded in expe- the Demographic Surveillance System (DSS) area
rience or a specific context (8). The interventions established in the early 1960s—now called Health
may not have provided the needed type of support and Demographic Surveillance System (HDSS).
by available, accessible and trustworthy individuals The Matlab health service area (MHSA) currently
who share common experiences. Wellman suggests provides healthcare coverage to approximately
that scholars interested in the implications of social 113,660 people, of whom 31,527 are women aged
support on health “must take into account the var- 15-49 years through the ICDDR,B’s four health sub-
ied nature of social support and of those who pro- centres and town hospital (21). The service area,
vide it” (19). He contends that investigators should where the study was conducted, consists of four
consider the aetiology of support and seek a more blocks and 67 villages.
fine-grained understanding of the type of relations
Methods
that produce particular kinds of support. The de-
velopment of culturally-tailored interventions de- A retrospective, cross-sectional research design was
pends on the specificity of such data. For example,
used. In-depth, semi-structured interviews using
Norbeck et al. based their effective social support
a prepared interview guide were conducted with
intervention to prevent low birthweight among
a non-random, sample of 25 community-dwell-
African American pregnant women on qualitative
ing women from the population of interest during
work that identified a woman’s mother and male
August 2008. The population of interest was Ban-
partner as critical sources of emotional support
gladeshi women aged 18-49 years residing in the
(20). As a first step in enhancing the relevance and
MHSA, who had an uncomplicated pregnancy and
validity of social support interventions, this study
delivery, resulting in a livebirth between 26 May
sought to identify and characterize the functions
2008 and 10 August 2008. The total sample (n=25)
of social support during pregnancy as perceived by
was purposively divided between women who
women in a developing-country setting.
had delivered in the home (n=12) and in a facili-
MATERIALS AND METHODS ty (n=13). The sample was selected to be as repre-
sentative as possible of the population of interest.
Study site Collection of data concluded after 25 interviews
because the respondents were converging on con-
The study was conducted in Matlab, Bangladesh, a
sensus, and the central limit theorem states that
rural subdistrict located 55 km southeast of the capi-
normal approximation is typically good with
tal city Dhaka. The vast majority of the population
25 observations or more. Data collection occurred
is Muslim, and the remainder is Hindu, Buddhist
no more than three months after the index birth.
or Christian. Islam is the main religion, and almost
The sample was drawn from the maternal and
all residents speak in Bangla, the official national
child programme database and recruited using
language. The principal economic activities are ag-
an active household strategy. This is a process in
riculture and fishing. Remittances from men who
which members from the eligible population are
migrate to the city or abroad for work opportunities
located and invited to participate in the study by
are another source of income for households in the
research staff. The staff travel within a community
area. The primary modes of transportation with-
household-by-household assisted by sample rosters
in Matlab are by foot, rickshaw, country boat, or
stratified at the village level.
three-wheeled taxi. Patrilineal and virilocal family
is the traditional type of family-unit found in rural
Collection of data
Bangladesh, although changes to kinship organiza-
tion are occurring due to the rapid socioeconomic The primary data-collection technique was individ-
transitions, including employment opportunities ual in-depth interviews guided by Spradley’s ethno-
164 JHPN
Social support during pregnancy in Bangladesh Edmonds JK et al.
graphic interviewing techniques supplemented by summaries for each individual respondent to ex-
participant-observation (22). A bilingual Bangla- tract an initial understanding of the data. Line-by-
deshi research officer, who had accrued five years line content analysis was conducted searching for
of experience conducting in-depth interviews and examples of support functions, which were high-
focus groups among villages in the study area, di- lighted and later coded manually. Individual data
rectly asked each participant a series of closed and units (words, phrases, and sentences) representa-
open-ended questions. Close-ended questions tive of support functions were considered from
asked about sociodemographic and reproductive the emic perspective (using the respondent’s own
health characteristics. Open-ended questions asked words and descriptors to explain the phenome-
about the people (relatives) who provided respon- non) and used for generating codes for categories
dents with support during pregnancy. Specific cul- of support. The data were compared across cases
turally-relevant probes generated responses about to find confirmations, contrasts, or contradictions.
the types of support provided by each relative men- The support categories found were systematically
tioned. Initially, the probes were derived a priori expanded, developed, and summarized with a fo-
from a widely-cited functional definition of social cus on examples representing the content of each
support by House who defined it as an interperson- category. The sources of support for each respon-
al transaction involving one or more of the follow- dent were organized by support category using
ing: emotional concern, instrumental aid, infor- data from the kinship matrix. Frequency ranking
mation, and appraisal (23). The probes evolved as determined the most salient categories of support
specific examples of support emerged. In the later provided to respondents during their pregnancy.
stages of interviewing, questions were based on the Sources of support for the four most frequently-
analysis of early information to find contradictory, mentioned support categories were tabulated
confirmative and alternative examples. Responses across interviews according to what a respondent
were recorded using a matrix of the most common mentioned. Results of the analysis were reviewed
kin and non-kin relations and their local Muslim and corroborated with the research officer who
and Hindu terms as described in Aziz’s seminal conducted the interviews.
work on kinship in Bangladesh (24). A list of rela-
Ethical approval
tions categorized by type of support was generated
after each interview.
The Institutional Review Board of the Emory
University and the Ethical Review Committee of
One to two interviews were conducted per day. All
interviews were conducted in the home. The av- ICDDR,B both approved the study. Verbal informed
erage duration of the interviews was 70 [standard consent of all women who volunteered to partici-
deviation (SD) 9.9] minutes. The research officer pate was obtained following standard disclosure
and the Principal Investigator debriefed after each procedures.
day of interview for approximately 45 minutes
RESULTS
per interview. All interviews were audio-taped us-
ing an Olympus Digital DS-2300 recorder, and the The characteristics of the sample are presented in
field notes served as a back-up method of retriev- Table 1. The mean age of the female participants
ing data. Each interview was transcribed verbatim was 25.7 (SD 6.6) years. All the respondents were
and translated, assisted by a transcription software. married. Ninety-two percent were Muslims, with a
Two research officers, bilingual in Bangla and Eng- Hindu minority (8%). The average educational
lish conducted the transcription and translation. attainment was 6 (SD 4) grades, indicating comple-
The research officer who conducted the interviews tion of primary schooling. The average number of
transcribed these in Bangla; the other one trans- livebirths, including the index birth, was 2.4 (SD
lated the Bangla transcription into English. Finally, 1.3, range 1-6). In response to a question about
the research officer who conducted the interviews how well the respondent could read or write, 12
verified the completed English transcripts together (48%) did so with ease, six (24%) with difficulty,
with the Principal Investigator. and seven (28%) not at all. Over half (52%) of the
women reported not having regular exposure to
Analysis of data
the media (e.g. watch television or listen to radio at
Sociodemographic and reproductive health charac- least once a week). Asset scores, a proxy for socio-
teristics were analyzed in the SPSS software (version economic status, ranged from 1 to 5, with 1 being
16.0) using simple frequencies (25). The transcripts the lowest wealth quintile and 5 being the highest.
were read repeatedly together with the field-note The average asset score was 3.3 (SD 1.4). The mean
Volume 29 | Number 2 | April 2011 165
Social support during pregnancy in Bangladesh Edmonds JK et al.
niment outside the homestead. Specific examples
Table 1. Descriptive characteristics of the study
depicting the content of each type of support are
sample (n=25)
presented in Table 2. The four most frequently-
Total sample mentioned support categories in rank order were:
Sample characteristics
No. % (a) practical help with routine activities, (b) infor-
Age (years) mation and advice, (c) emotional support and as-
surance, and (d) resources and material goods.
18-21 8 32
22-25 5 20
Sources of support
26-29 4 16
≥30 7 28 In total, 19 unique relations were perceived as sourc-
Parity es of support. The average number of supportive re-
1 8 32 lations mentioned was 9.96 (SD 1.9, range 6-15).
2 5 28 All the relations were described using descriptive
kin terminology, with two exceptions—commu-
3 3 20
nity health research workers (CHRWs) and neigh-
≥4 8 8
bours. CHRWs are employed by ICDDR,B and visit
Education attainment
households once every two months to collect de-
Never attended 5 20
mographic data and provide health-promotion and
1-5 grades (primary) 5 20
preventive services. Neighbours are often extended
6-10 grades (secondary) 14 56
family living in adjacent baris (homesteads). The
11-12 grades 1 4
most frequently-mentioned supportive female kin-
Literacy ship relations were: mother, sister, mother-in-law,
Easily 12 48 husband’s sister, husband’s brothers or cousins and
With difficulty 6 24 their wives, and wives of husband’s uncles. In the
Not at all 7 28 Bangladeshi context, wives of husband’s brothers
Regular exposure to media or cousins are referred to as jals, a kinship term that
12 48
Yes translates into sister-in-law. The most frequently-
13 52
No mentioned supportive male kinship relations were
husband, father, and father-in-law. CHRWs and
Religion
neighbours were the two most frequently-men-
Muslim 23 92
tioned supportive non-kin relations.
Hindu 2 8
Asset score Type of support by source
1 (lowest) 3 12
The hierarchy of relations varied by type of sup-
2 5 20
port provided. The percentages of the respondents
3 5 20
reporting the four top ranked types of support by
4 7 28
source are shown in Figure 1-4. Responses indicated
5 (highest) 5 20
that mothers, mothers-in-law, and husbands were
Type of household the most multiplex of relations, providing all four
Marital 18 72 categories of support. Multiplex is a social network
Nuclear 5 20 term meaning multiple types of support among the
Natal 2 8 same set of people.
household-size was 7.8 (SD 2.5). The large majori- During the interviews, support categories were
ty (72%) of the women in the sample lived with discussed in relation to the type of activities they
the husband and husband’s kins, characteristic of compromised. De-identified quotations from the
patrilocal kinship systems. transcripts were used for illustrating the content
of the top four ranked categories of support and
Type and content of support
key themes. In the quotes, the words of the re-
Women perceived eight types of support during spondents are presented in format text and those
of the interviewer in bold.
their pregnancy period. These included practical
help with routine activities, information and advice,
Practical help with routine activities
emotional support and assurance, resources and
material goods, logistic communication, prayer and The vast majority of the examples involved practi-
spiritual rituals, nutritional support, and accompa- cal help with routine activities or the provision of
166 JHPN
Social support during pregnancy in Bangladesh Edmonds JK et al.
Table 2. Description of types of support with examples of content
Type of support Content examples
Practical help with Milked cows, prepared fishes
routine activities Cleaned and strained rice
“Brought water from tubewell for cooking in big pots”
Fed other children
Washed clothes
Went to market when husband not in the home
Helped with heavy work during harvest season, husking paddy
Made fire for cooking
Information and advice “Do not go to roof to hang clothes”
“When ill, take rest”
“Give prayer five times a day”
“Take food properly”
“Do not work with cold water”
“Do not stay in field at night because of evil spirits”
“Keep a match with you when you go out at night”
“Call husband to do heavy work”
“Abide by father-in-law and mother-in-law, words they say”
Emotional support and “Do not worry” “Chinta koro na”
assurance Inquired about how she is feeling
“Do not be tense”
Gossiped
“If you have a problem, I will go with you to facility”
Talked with her when she feels unsure
Massaged body with mustard oil
Tell me if you have any physical problems, inform of discomfort or changes
Monetary and material Paid for urine test to confirm pregnancy
goods Bought and delivered medications
Paid rickshaw fare to clinic
Paid costs for antenatal care
Paid kobiraj for amulets and rituals
Saved money in case of complication
Logistic communication Went out to locate dai
Informed others in household of pregnancy and labour pains
Phoned mother to come to marital home when labour started
Went to find rickshaw-puller or country boatman
Prayer and spiritual Gave holy blow and gave blessed water
rituals Brought amulets from kobiraj
Brought blessed molasses to increase labour pains
Nutritional support Gave milk and tea during labour pains
Brought special foods, such as grapes and guavas, from bazzar upon request
Made foods she likes when sick
Accompaniment Accompanied and assisted with bathing in pond
Accompanied at night to latrine
Accompanied to facility or met at facility
Stayed at facility during labour
skill, labour and time resources in relation to regu- culture and animal husbandry, and childcare (26).
larly-performed household duties. The examples They work for long hours, and the work is physi-
were consistent with descriptions of women’s pro- cally demanding (i.e. carrying water from the well
duction in which rural Bangladeshi women work and husking the paddy). Mothers-in-law, moth-
within or near the household in food processing, ers, and sisters-in-law were the top three sources of
food preparation, household maintenance, horti- practical help.
Volume 29 | Number 2 | April 2011 167
Social support during pregnancy in Bangladesh Edmonds JK et al.
Fig. 1. Sources of practical/routine help
50
45
s 40
t
n
e 35
nd 30
o 25
p
s 20
e
f r 15
o
% 10
5
0
M other-in-labawnd’sM obtrhoetrher’s wife Husband HSuisstbeaursnsbd’as nsid’sst eursncle’s Fwaitfeher-in-lawNeighbour Father C HR W
us H
H
CHRW=Community health research worker
Fig. 2. Sources of information and advice
70
ts 60
n
e 50
d
n
o 40
p
s 30
e
r
f 20
o
% 10
0
nd’s brother’s wifeHusband M other C HM RotWher-in-law HuSsisbtaebnradn’sd ’ssi sutenrcle’s wifeNeighbour FaFtathehrer-in-law
ba us
us H
H
CHRW=Community health research worker
She (sister-in-law) did not let me separate cooked cooking. When she saw me cooking for a long
rice from the liquid—vater mar gala—it is con- time, she told me to ‘get up ….’ (R01)
sidered a heavy work since the hot rice-pot needs
to be lifted and held for sometime. She filled the M: “Who used to help you in your daily
pitchers—she did everything when I had abdo- work?” R: “There are my mother and bhabi (wife
men pain. (R16) of my brother) living in my house. Bhabi used
to do household works. When I went for a bath,
My husband used to help me with things I
my mother looked after me so that I did not fall
could not do. For example, carrying water for
down.” (R09)
cows. (R23)
The impression gleaned from the interviews was
Shashuri (mother-in-law) cooked, filled up pitch-
that women’s level of daily work was reduced when
ers, and she helped. (R23)
residing in or visiting their natal homes. Reduction
M: “You have two jals (sisters-in-law). How did of workload was described as a clear incentive for
they help you?” R: “She helped, helped me in natal home-visits during pregnancy, not just for de-
168 JHPN
Social support during pregnancy in Bangladesh Edmonds JK et al.
Fig. 3. Sources of emotional support
50
45
s
t 40
n
e 35
d
n 30
o
p 25
s
e 20
r
of 15
% 10
5
0
band’Ms otbrhoetrher’s wife HHususbbaanndd’s sisters SistersNeighbMooutusrhbear-nid’ns- launwcle’s wife C HR W FatFahterher-in-law
us H
H
CHRW=Community health research worker
Fig. 4. Sources of resources and material goods
45
40
s
t 35
n
de 30
n
o 25
p
s 20
e
r
f 15
o
% 10
5
0
Husband M oMtohterher-in-Flaatwher-in-law Father HSuissbtaberansnd’ds’ ss isbtreourtsshbear’ns d’sw ifuencle’s wife C HR WNeighbour
us H
H
CHRW=Community health research worker
livery as traditionally dictated by the sociology of helped me. They brought for me whatever food
first births in Bangladesh. Pregnant women likely I could eat. I used to vomit instantaneously af-
benefit from the close social and physical support ter eating. I could not keep the food inside. They
that comes with spending time in their natal home, used to prepare what I wanted to eat. (R03)
a point that is illustrated in the following quotes:
It was more work during the harvest time
(in marital home). I went to my father’s house
In my father’s house, they did not let me do any
(natal home). (R01)
work. I ate and slept. They did not even let me hang
my mosquito net. My mother or sister did this for For one woman, the anticipated level of pregnancy
me. My sisters also washed clothes for me. (R21) and postpartum support in her natal home influ-
enced the final decision about the place of delivery.
I did not work in my father’s house. My younger
sister and mother did all the work. (R22) My husband preferred to take me (to deliver) at
the (natal) home in village because more care
When I was in my mother’s house, my sisters could be provided there. (R01)
Volume 29 | Number 2 | April 2011 169
Social support during pregnancy in Bangladesh Edmonds JK et al.
Information and advice Emotional support and assurance
Information and advice refer to feedback about Emotional support and assurance refer to the provi-
the women’s condition or situation. Sisters-in- sion of re-assurance and sympathetic listening that
law (68%), husbands (64%), mothers (56%), and results in feelings of comfort and security. Women
CHRWs (56%) were the most frequently-men- described emotional support as chatting about
tioned sources of information and advice. Place of their comforts, discomforts, and the progression of
delivery was a common topic as evidenced from their pregnancies. They referred to the people who
the following quotes. consoled them and whom they confided in about
concerns about pregnancy and childbirth. Mothers
She (CHRW) told that home-delivery would
(48%), sisters-in-law (48%), and husbands (44%)
be bad. She told me to go to Matlab on the ex-
were the three most frequently-mentioned sources
pected date. She told me that it would be good
of emotional support:
to go to Matlab. If your baby is born in Matlab,
it will get good treatment. You and your baby … she (jal) told me, ‘Do not get afraid, nothing will
both will be safe and sound. (R19) happen.’ When I felt pain she told me, ‘Do not
worry. Nothing will happen.’ (R10)
She (CHRW) told me, ‘do not do heavy work’.
She told me not to go to father’s house (natal M: “Who consoled you?” R: “I used to chat with
home) and to stay here. ICDDR,B’s hospital is my husband.” (R08)
nearby and told me to stay here. (R10)
Neighbours also provided emotional support and
M: “You talked with your niece about your assurance:
comfort and discomfort. What did she say?”
R: She told me to go to ICDDR,B. Here in our area My neighbour Farida Apa used to tell me, ‘do not
everyone goes to ICDDR,B. Home-delivery does be scared.’ She encouraged me. She told me, ‘You
not take place usually. (R11) will get what Allah offers you’. (R08)
Other kinds of advice were more culture-specific Notably, only 4% of the respondents reported that
and reflect the persistence of traditional practices their mothers-in-law provided them with emo-
surrounding mobility during pregnancy. Pregnant tional support during pregnancy. This low ranking
women are advised to not go out during the after- is consistent with the traditional dominant role
noon or evening for fear that they might become a mother-in-law has over her daughter-in-law in
possessed by free-ranging spirits thought to cause patrilocal kinship systems (28). The imbalance of
poor pregnancy outcomes. These evil spirits are authority and power contributes to an indifferent
believed to catch a woman when she travels alone or even hostile relationship.
outside her residence, especially during early morn-
Resources and material goods
ing, noon, and at dusk (27).
Resources and material goods refer to the sharing
She (sister-in-law) forbade me to move in a clumsy
or provision of goods, money, or tools. Women
way. She told me to be careful about movement
frequently mentioned husband’s participation in
and timing of movement. M: “Please explain
securing monetary resources for healthcare, foods
to me.” R: There are also fears of evil spirits.
with increased nutritional value, transportation,
You cannot go out at noon, evening, dawn, and
and other related expenses. Husbands were the pri-
night. Evil spirits will catch you. (R25)
mary source of resources and material goods men-
… They (aunts) talked to me. They told me how to tioned by 45% of the respondents, followed by
move and what was good for me. They forbade me to mother’s resources mentioned by 10% and moth-
go anywhere. They told me to be careful in my move- er-in-law’s resources mentioned by 8% of the re-
ment. They forbade me to go anywhere at noon. spondents. There was less variability in the sources
They told me to stay always in my room. (R06) providing resources and material goods compared
to the other most frequently-mentioned types of
My mother told me to be careful. She forbade me
support. The provision of monetary support by
to do heavy work. She forbade me to go outside
husbands and mothers was captured in the follow-
at evening. I took fire (match box) with me wher-
ing quotes:
ever I went at noon or at night, at the time of
Khon. (R23) My husband used to ask me how I was. He que-
170 JHPN
Social support during pregnancy in Bangladesh Edmonds JK et al.
ried my condition. He told me to let him know is a precautionary intervention aimed at prevent-
what I needed. He gave me money. (R18) ing spirit attraction, particularly in the later stages
of pregnancy. Previous research in Bangladesh has
I took money from my mother (natal home). I
described restrictions of activity linked to beliefs in
did not take it from my marital house. (R10) supernatural sprits (29,30). With respect to health,
researchers have noted that restrictions of mobility
DISCUSSION
can reduce actual and potential access to needed
The study provided current examples of the type, medical services (31,32). Interventions that seek to
content, and source of social support during preg- increase healthcare access should consider the re-
nancy grounded in a specific context of rural Ban- ceipt of messages that act as deterrents to women’s
mobility.
gladesh. The data indicate that women in Matlab
perceive the receipt of eight distinct types of sup-
The documentation of the sources of emotional
port and, on average, have 10 supportive relations
support and assurance revealed the unexpected
in their networks. Although the sources varied by
role husbands play in supporting their wives. Typi-
type of support provided, the role of marital and
cally, women are thought to provide substantially
natal kins was prominent in all the narratives. Cer- more emotional support than men due to their
tain relationships were of particular importance and increased empathy (19,33). Therefore, it is not
include the most frequently (overall)-mentioned surprising that mothers and sisters-in-law are the
sources—mothers, mothers-in-law, sisters-in-law, top sources mentioned by the women. In contrast,
and husbands. All of these relations are kin-based men are thought to lack emotional expressiveness
and, with the exception of women’s mothers, result and their contribution to emotional support often
from marriage. The types of support were discussed downplayed. However, our findings suggest that
in terms of the helping activities they comprised. women perceive that their husbands do provide
Specific findings relevant to future research and so- some degree of emotional support and assurance
cial support interventions are discussed below. during pregnancy. A recent study by Singh and
Ram in rural Ahmadnagar, India, also showed that
Practical help with routine daily activities was the a substantial proportion of men provided a notable
most acknowledged type of support among the level of support to their wives during pregnancy
participants and appears to be more forthcoming (34). Further investigation of the benefits asso-
in a woman’s natal home. Women reported receiv- ciated with galvanizing spousal support during
ing help with milking cows, bringing water from pregnancy is recommended.
the tubewell, washing clothes in the pond, clean-
Specific to the Matlab setting, CHRWs are instru-
ing and straining paddy, and childcare. In develop-
mental in providing information and advice, par-
ing-county settings where economic resources are
ticularly about where to deliver. Yet, although all
scarce, non-monetary support of this kind may rep-
the participants reported being visited by a CHRW
resent the most reliably-available type of support.
during their pregnancies, only 60% mentioned
Women likely value practical help with routine
them as sources of support. Aside from these
activities because it allows time for the extra rest
specially-trained laywomen, the respondents did
recommended during pregnancy and may, in turn,
not list health professionals in their networks,
decrease the stress associated with poor pregnan-
suggesting that pregnant women do not consider
cy outcomes. Injury (i.e. back-pain, falls) associ-
professionals as sources of support. This finding
ated with heavy household work may also be pre-
corroborates other research that shows that non-
vented. Interventions that solicit practical support
professionals are the preferred and prominent so-
from the family during pregnancy and promote a
cial support providers (6). As social support appears
women’s ability to spend time with her natal kins
to be a lay resource, interventions that encourage
may foster maternal well-being. Further studies are women to use and enhance personal networks are
needed to establish causal links between providing warranted.
practical help in the context of a women’s daily life
with improvements in pregnancy outcomes. A behaviour-change communication and social
mobilization programme in Bangladesh provides
The transcripts revealed the persistence of tradi- some evidence that improvements in lay social sup-
tional advice regarding constraints on movement port during pregnancy can be achieved. One of the
during pregnancy. The advice is reflective of the be- programme goals was to generate or increase the
lief that pregnancy is a period of increased vulnera- support provided by husbands and mothers-in-law
bility to supernatural spirits. Decreased movement during pregnancy. The provision of positive sup-
Volume 29 | Number 2 | April 2011 171
Social support during pregnancy in Bangladesh Edmonds JK et al.
port was highlighted through community mobili- ACKNOWLEDGEMENTS
zation efforts and a mass-media campaign. Family
The study is a collaboration between the Center
members were encouraged to make available regu-
for Research on Maternal and Newborn Survival,
lar and nutritious foods, allow women to take rest,
Emory University, USA and the International Cen-
and avoid heavy work. Social support was presented
tre for Diarrhoeal Disease Research, Bangladesh,
as a social custom to be followed during pregnancy.
funded through the Woodruff Health Sciences
Post-evaluation data suggest that positive changes
Center Foundation grant to the Center for Research
occurred in terms of the behaviour and attitudes of
on Maternal and Newborn Survival, Nell Hodgson
mothers-in-law towards their daughters-in-law and
Woodruff School of Nursing, Emory University.
husbands towards their wives. Health workers were
In addition, a grant (No. F31NR010650) from the
encouraged to continue community mobilization
National Institute of Nursing Research further sup-
through group discussions about the need for the
ported the first author. The content is solely the
supportive role of husbands, mothers-in-law, and
responsibility of the authors and does not necessari-
other members of the family during pregnancy
ly represent the official views of the National In-
(35). Replication of this approach deserves further
stitute of Nursing Research, the National Institutes
consideration.
of Health, or the Woodruff Health Sciences Center
Limitations Foundation.
The findings must be considered within the scope The authors thank Allison Moran who facilitated
of limitations of the study. As with all research in- local approvals to conduct the study.
volving self-reported interview responses, there is
the potential for recall bias. Further, the study de- REFERENCES
sign did not allow for analysis of the need for sup-
1. Cohen S, Syme S. Issues in the study and application
port, the timing of support in relation to the stages
of social support. In: Cohen S, Syme S, editors. Social
of pregnancy, and satisfaction with the support re-
support and health. San Fransico: Academic Press,
ceived. The study also measured perceived versus
1985:3-22.
received support. Perceived support, rather than
2. Pescosolido BA. Illness careers and network ties: a
received support, is most closely linked to health
conceptual model of utilization and compliance. Adv
outcomes (36). Despite the fact that the study was
Med Sociol 1991;2:164-84.
carried out in Matlab, an area with sustained ma-
ternal and child-health programmatic inputs (e.g. 3. McDowell I. Measuring health: a guide to raing scales
CHRW’s home visitation), the findings are likely and questionnaires. 3rd ed. New York, NY: Oxford
reflective of women’s support functions during University Press, 2006. 768 p.
pregnancy in other areas of Bangladesh—not in the 4. Berkman LF, Glass T, Brissette I, Seeman TE. From so-
sense of statistical generalization but in the sense cial integration to health: Durkheim in the new mil-
of transferability of knowledge in ethnographic lennium. Soc Sci Med 2000;51:843-57.
work.
5. Agadjanian V. Informal social networks and epidem-
ic prevention in a third world context: cholera and
Conclusions
HIV/AIDS compared. Soc Networks Health 2002;8:201-
Specific and contextualized data from this study 21.
provide possible insights to researchers studying
6. Finfgeld-Connet D. Clarification of social support. J
social support and programme staff developing
Nurs Scholarsh 2005;37:4-9.
interventions in this population. The potential
7. Adams AM, Madhavan S, Simon D. Measuring social
to influence health outcomes positively depends
networks cross-culturally. Soc Networks 2006;28:363-
on predicting which supportive functions will be
76.
most effective to a particular kind of stressor (37).
It is likely that, in pregnancy, only certain types or 8. Elsenbruch E, Benson S, Rücke M, Rose M, Duden-
sources of support may be helpful, such as practical hausen J, Pincus-Knackstedt MK et al. Social support
help by members of a women’s naturally-occurring during pregnancy: effects on maternal depressive
social network. Formative qualitative research that symtoms, smoking and pregnancy outcome. Hum
details the nature of such support networks in dis- Reprod 2006;22:869-77.
tinct cultural settings is an important first step in 9. Da Costa D, Dritsa M, Larouche J, Brender W. Psycho-
establishing effective social support interventions social predictors of labor/delivery complications and
for pregnant women. infant birth weight: a prospective multivariate study.
172 JHPN