Activity/Partner
The CHANGE Project assisted Family Care International's
(FCI) Skilled Care Initiative in Kenya to develop and
adapt a set of behavior-based qualitative research tools
to investigate use of skilled childbirth care. The Skilled
Care Initiative aims to improve access to and increase
use of skilled care during and after childbirth.
Many of the research results from Western Kenya confirm
findings that are common to other rural areas in Kenya
and in the developing world. There were also locally-specific
findings that helped FCI design a behavior change intervention
strategy for the project area and that contributed new
dimensions to the current understanding of careseeking
behaviors for childbirth.
Background
Increasing women's use of skilled care during and after
childbirth has been identified as the most critical
intervention to reduce maternal deaths.
Consequently, it is important to distinguish between
deliveries by a skilled provider and births assisted
by traditional birth attendants (TBAs), even formally
trained TBAs. Maternal survival programs now measure
the proportion of women who are "delivered by a
skilled provider" rather than those who give birth
with a trained attendant. Countries no longer include
TBA-assisted births official statistics.
At the local level, however, this shift to skilled
care has achieved moderate success. Increasing use of
skilled care has enormous implications for women and
families as well as the TBAs. It usually means that
women would have to leave their homes to give birth
in a facility and that TBAs would link women to skilled
care rather than attend births themselves. Qualitative
research in some countries has documented a resistance
to change even when the rationale for the change, such
as increased safety, is well understood.
As a result, there is growing interest in trying different
approaches to designing and implementing behavior change
interventions to support increased use of skilled childbirth
care.
Objectives
The CHANGE Project partnered with FCI to:
- Design a locally specific and appropriate behavior
change intervention to increase skilled careseeking;
- Test the acceptability of TBAs and increasing use
of skilled providers during and after childbirth;
- Test and refine tools that could be adapted for
rapid development of behavior-based programs to increase
use of skilled care; and
- Test the feasibility and effectiveness of using
community interviewers to conduct basic behavioral
research that interviewers with medical or academic
training usually do.
Setting
Like other rural areas in Kenya, many women in Homa
Bay give birth at home, assisted by an unskilled attendant.
In 1998, only 38% of women in Nyanza Province, delivered
with a skilled provider, compared to 44% nationwide
and 76% of women in Nairobi, Kenya's capital. Kenya's
estimated maternal mortality ratio (MMR) is 590.
The CHANGE Project worked with FCI in the Homa Bay
district of Nyanza Province, Kenya. Homa Bay is an ethnically
homogeneous area populated by approximately 300,000
of the three million Kenyans belonging to the Western
Nilotic Luo group.
Methods
Interviewers conducted more than 85 interviews in Homa
Bay, including focus group discussions (FGDs), in-depth
interviews (IDIs) and complication narratives. The group
complication narratives involved all of the people who
participated in making decisions about careseeking and
caregiving for a woman who recently experienced life-threatening
obstetric complications. Interview participants included
women who gave birth within the previous six months,
elder female family members and other family influentials,
TBAs, community and religious leaders and skilled providers.
The CHANGE Project designed interview
guides based on a set of recommended behaviors.
The guides go beyond standard questions about knowledge,
attitudes and practices to explore how acceptable it
would be to "reposition" or change the role
of TBAs to link women to skilled care. The guides also
encourage participants to contribute their own ideas
for interventions to increase skilled careseeking.
Use of Community Interviewers
A unique element of this research was the use of community
interviewers. Often people with medical or academic
training perform this type of research. In this case,
the research team felt that the advantages of using
interviewers from project communities outweighed any
disadvantages.
The research team recruited a total of 21 interviewers
from the local community. They worked in pairs -one
to interview and one to record - supervised by a "coach."
The team was impressed with the performance of the
community interviewers. They demonstrated rapid gains
in confidence and quality of interviewing and recordkeeping.
Although the community interviewers had not conducted
research before this activity, nearly all of them had
recent personal experience with childbirth so were deeply
committed to the program. To help the interviewers learn
and practice skills, the team increased the length of
training from three days to five days.
Each interviewer kept a journal of their experiences.
These journal entries, along with the notes from trainers
and community research coordinators, provide a unique
perspective on community-implemented research.
Analyzing Results
Information from the transcripts was organized into
groups according to behaviors, and each group was then
analyzed to identify key behavioral implications. Results
of the complication narratives were "mapped"
to visually present the findings (sample
grid). The detailed results were used to formulate
a behavior-based BCI strategy (strategy
worksheet) and to develop content for program materials,
messages, strategies and activities.
Key Findings and Program Implications
These discussions with women, families, community members,
TBAs and skilled providers in Homa Bay District have
enriched the local knowledge base on maternity care
and global understanding of factors that influence skilled
careseeking.
- There was a high level of knowledge about danger
signs of obstetric complications and emergencies even
among male partners and religious leaders. The research
documented the local terms and beliefs about major
obstetric complications.
- Although most women and families made some advance
preparations for the arrival of a new baby, these
are mostly small savings for routine costs such as
the purchase of baby clothes. Household and community
birth preparedness activities were uncommon. Women
explained that a primary reason for seeking antenatal
care (ANC) was to get an ANC card; the card was seen
as a "passport to skilled care." Many women
recounted long delays or unwillingness of hospital
staff to provide both routine and emergency obstetric
care if women arrived at health care facilities without
such a card. The typical or "universal"
access barriers to skilled care -- cost, distance
and transport -- were found to be important constraints
in Homa Bay as well, particularly in the most remote
areas.
- The complication narratives documented how some
serious obstetric complications continued into the
days and weeks after birth. Early postpartum careseeking
was non-existent. The importance of observing women
and newborns during the first weeks after birth to
detect complications and the need for early postpartum
care were not widely acknowledged.
Important additional local factors influencing skilled
care use during normal deliveries:
- One of the most common reasons women did not seek
skilled care was because they did not get reliable
estimates of their delivery dates from TBAs or skilled
providers. Women believed that a reliable due date
would help them prepare for use of skilled care. In
response, CHANGE developed a simple pregnancy
calculator for women and community groups to estimate
due dates.
- Women and families also shared how the poor attitudes
and behaviors of skilled providers influenced their
decisions about where to give birth. Even skilled
providers acknowledged that their behavior toward
clients was a significant barrier to use of skilled
care.
- Women also described other provider-related factors
that contributed to low quality of care, particularly
long delays for required emergency obstetric care.
Even if women did manage to reach skilled care in
a timely manner, there were frequently life-threatening
delays receiving care once they arrived.
- Women clearly expressed factors that influenced
their preferred source of childbirth care. The kindness
and "caring" care provided by TBAs stood
in stark contrast to the characteristics ascribed
to facility-based care by skilled providers and overwhelmingly
motivated women to continue delivering with TBAs.
- Nonetheless, women and families maintained an overall
faith in the technical competence of skilled providers
despite documentation of widespread and serious inadequacies
in quality of care. There is a need to investigate
the importance of elements of quality care as measured
by "community-perceived" factors versus
Western "evidence-based" standards.
Local factors influencing skilled care use during complications
and emergencies:
- Even after a complication or emergency was recognized,
some women and families did not seek skilled care
first because they believed it would not solve certain
problems. Careseeking frequently involved multiple
unskilled sources before skilled care. For example,
many of the complication narratives documented lengthy
"detours" to consult indigenous religious
and prayer groups and advise against use of modern
medical care even in extreme obstetric emergencies.
The beliefs and attitudes toward skilled childbirth
care among such religious "sect" leaders
and their followers should be further investigated.
- Many of the slow decisions and detours were a result
of "cultural mismatches" between local understanding
about causes of obstetric complications and Western
medical beliefs. The most intriguing among these "locally-perceived"
obstetric complications were "rairu." There
was a strong and widespread community belief that
rairu could not be successfully treated by Western
medicine or modern medical practitioners. There was
an equally strong belief among skilled providers that
rairu did not exist.
- Other cultural beliefs documented in the interviews,
including the efficacy of "pot medicine"
by TBAs, also influenced timely careseeking. Behavior
change interventions need to address the divergent
views and negotiate practices acceptable to all groups.
- Traditional Luo measurements of time does not correspond
to Western timekeeping conventions. Luo babies are
often named to reflect birth circumstances and many
examples of this are evident in Dhlouo language. For
example, many children are named Odoch (breech birth),
Oyoo (born on the way), Apoya (sudden onset of labor)
or Oliech (pulled out). These distinct local cultural
elements could be incorporated into behavior change
messages and materials along with specific local obstetric
terms to increase acceptability and effectiveness.
- The complication narratives documented how some
serious obstetric complications continued into the
days and weeks after birth. Early postpartum careseeking
was non-existent; the importance of observing women
and newborns during the first weeks after birth to
detect complications and the need for early postpartum
care were under appreciated.
- Skilled providers asked for more attention to the
excessive physical demands of their work and the psychological
demands placed on them by the difficult conditions.
Acceptability of "TBA Linkworker"
Concept
- Women, families and TBAs in Homa Bay were open to
changing the role of TBAs to serve as links to skilled
care providers. Participants identified "conditions
of acceptability" that would affect their acceptance
of TBA linkworkers. For example, they talked about
the need for compensating TBAs who agree to link instead
of deliver care.
- The research identified individuals who were central
to setting community norms and spreading information.
Key people from social networks and illness networks
could provide a focus for community behavior change
interventions and spread new ideas quickly.
Lessons Learned
Test new ideas with community members:
Past qualitative research has shown that families and
communities are often unwilling to accept new behaviors
that would require them to alter firmly established
childbirth traditions. In Homa Bay District however,
community members were willing to increase skilled careseeking.
Many eloquently expressed realistic, constructive suggestions
for change and a willingness to try to improve skilled
careseeking behaviors, despite the difficult circumstances
that exist in Homa Bay District for both clients and
providers.
Complication narratives: with a woman
or family of a woman who experienced an obstetric emergency
in the previous six months efficiently generate information
to design locally appropriate, research-based behavior
change interventions. If a program has resources to
conduct only one type of research, the CHANGE Project
recommends complication narratives. Any program working
to improve maternal and newborn survival should use
this method to understand current practices in their
cultural context.
Community interviewers bring unique contributions
to the research activity: The research team
extensively debated whether or not to use community
interviewers before making a decision. The benefits
of a local community interviewer team knowledgeable
about the benefits, barriers, and use of skilled obstetric
care in their district outweigh the consequences of
any missing data. Many of the community interviewers
expressed a willingness and commitment to continue and
assist their communities develop and implement community
interventions to increase use of skilled care.
Allow extra time to locate participants: Finding
women who had given birth within the past six months
required more advance planning than was anticipated.
Women who gave birth in facilities could not be located
through review of hospital records. Many of these women
and women who gave birth at home had to be identified
by community members. For example, in Homa Bay, assistant
chiefs maintain records of all births in the community.
Adaptations and Future Use of CHANGE Tools
and Approaches
Tools (Danger Signs
Plus)
Report
>> Maternal Survival Toolkit
Home |
