  |
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| Journal of the New Zealand Medical Association,
04-June-2004, Vol 117 No 1195
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Factors associated with not breastfeeding exclusively among
mothers of a cohort of Pacific infants in New Zealand
Sarnia Butler, Maynard Williams, Colin Tukuitonga, Janis
Paterson
The benefits of breastfeeding (for both infant and mother)
are numerous and well documented.1–3 A
threshold level may exist for protection against infectious diseases for
infants, with greatest protection observed with exclusive
breastfeeding.2 For these protective and other
observed benefits, it has been recommended that the introduction of
complementary foods should be delayed (in most circumstances) until the infant
is approximately 6 months old.4 Thus, it is
important to encourage mothers to breastfeed exclusively over the first few
months of the infant’s life.
Although breastfeeding rates in New Zealand tend to be high
compared to other countries,5,6 exclusive
breastfeeding rates appear to reduce substantially—well before the
recommended 6 months.7 To provide opportunities
for targeted intervention, and to better inform breastfeeding programmes, it is
important to identify factors that may hinder the initiation or maintenance of
exclusive breastfeeding for different ethnic groups in New Zealand.
An earlier study identified Pacific ethnicity as being
associated with not exclusively breastfeeding at discharge from
hospital.7 Furthermore, we have previously
shown that breastfeeding rates observed in our cohort (at 6 weeks post-birth)
fall below recommended national targets.8 The
present study investigates the association between not breastfeeding exclusively
(among mothers of a cohort of Pacific infants in New Zealand) and several
maternal, sociodemographic, and infant care factors.
MethodsData were collected as part of
the Pacific Islands Families (PIF) Study; a longitudinal investigation of a
cohort of 1398 infants (11 pairs of twins) born at Middlemore Hospital, South
Auckland, New Zealand during the year 2000. Middlemore Hospital was chosen as
the site for recruitment of the cohort as it has the largest number of Pacific
births in New Zealand and is representative of the major Pacific ethnicities. It
was estimated that a cohort of 1000 would provide sufficient statistical power
to detect moderate-to-large differences (after stratification for major Pacific
ethnic groups and other key variables).
Eligibility criteria included having at least one
parent who self-identified as being of Pacific ethnicity and at least one parent
who was a New Zealand permanent resident. Thus, non-Pacific mothers were
eligible for the study in cases where the infant's father was of Pacific
descent. Detailed information about the cohort and procedures is described
elsewhere.9
Approximately 6 weeks after the birth of their child,
Pacific interviewers (fluent in English and a Pacific language) visited the
mothers in their homes. Of the 1376 mothers, 1365 were biological and 11 were
foster or adoptive mothers. Eligibility criteria were confirmed, and informed
consent was gained for their participation in an interview and for our access to
their Middlemore Hospital discharge record.
Each mother participated in a 1-hour interview (in
their preferred language) about the health and development of their child, and
family functioning. Questions regarding how the infant had been fed for the
first 6 weeks of their life were included in this assessment.
Consistent with other
studies,10,11 breastfeeding was considered
exclusive if no other milk, formula, or solids were given apart from liquids
such as water. Combination breast-and-formula feeding and formula-only feeding
were therefore considered as ‘not breastfeeding exclusively’.
Several variables were examined for associations with ‘not breastfeeding
exclusively’ via univariate and multivariate logistic regression analyses.
First, factors associated with ‘not breastfeeding
exclusively’ (at the time of discharge from hospital) were examined.
Subsequently, factors associated with not breastfeeding exclusively (at 6 weeks
post-birth) were examined for mothers who initially exclusively breastfed in
hospital. Of the 1365 biological mothers in the PIF study, infant feeding data
from hospital records were available for 1247 of the mothers (91.4% of the
mothers of the cohort). Data from these 1247 mothers and the responses based on
the first-born twin for twin pairs were used in all analyses.
ResultsNinety-six percent (n=1590) of
potentially eligible mothers of Pacific infants (who had been born between 15
March 2000 and 17 December 2000) gave consent to be visited in their homes when
their infant was 6 weeks old.
Of the 1477 mothers contacted and who met the eligibility
criteria, 1376 (93.2%) agreed to participate in the study. A more conservative
recruitment rate of 87.1% would include mothers who consented to contact and
were (a) confirmed eligible, or (b) of indeterminable eligibility due to
inability to trace. Of the 1247 biological mothers in the present study (1.7%
gave birth to twins; n = 21), 47.5% self-identified their major ethnic group as
Samoan, 16.6% as Cook Island Maori, 4.3% as Niuean, 20.8% as Tongan, 3.5% as
Other Pacific (includes mothers either identifying equally with two or more
Pacific groups, equally with Pacific and Non-Pacific groups, or with Pacific
groups apart from Tongan, Samoan, Cook Island or Niuean), and 7.4% as
Non-Pacific. The mean age (SD) of mothers was 27.9 (6.1) years; 80.7% were
living together in married or de facto partnerships, 32.8% of mothers were New
Zealand born, and 27.7% had post-school qualifications.
At the time of discharge from hospital, 1017 mothers (81.6%)
were exclusively breastfeeding. Of 23 variables examined for potential
association with not breastfeeding exclusively at discharge from hospital, 9
reached statistical significance (p<0.05); findings for these variables are
shown in Table 1.
Table 1. Numbers (row percentages) and odds ratios of
‘not breastfeeding exclusively’ at discharge from hospital by
variables attaining significance in univariate logistic regression analyses
(n=1247)
*p<0.05;
†p<0.01;
‡p<0.001.
For the categories within each variable, the numbers and
percentages of mothers who reported not breastfeeding exclusively are given
along with their respective univariate odds ratios (95% CI) indicating
likelihood of not breastfeeding exclusively.
Variables examined, but not significantly associated with
not breastfeeding exclusively, included maternal age, ethnicity, whether born in
New Zealand, social marital status, education, English fluency, cultural
alignment, parity, whether pregnancy was planned, whether they attended
antenatal classes, alcohol consumed during pregnancy, whether general
practitioners or traditional healers were seen during pregnancy, birth weight,
whether the infant was discharged at the same time as mother, household size
(persons), and annual household income.
Many variables examined for individual associations with not
breastfeeding exclusively at discharge from hospital are likely to be
interrelated. A multiple logistic regression analysis was undertaken to control
for confounding effects and enable identification of important variables able to
provide a parsimonious explanation of the data.
Five demographic variables (age, education, ethnicity,
marital status, and household income) were initially forced into the model as
control variables, and then all remaining variables were submitted to a forward
stepwise procedure (p to enter = 0.15, and p to remove = 0.20).
Table 2 demonstrates that (when adjusting for all other
variables in the final model) factors significantly associated with not
breastfeeding exclusively (p<0.05) at discharge from hospital were caesarean
delivery, not being employed prior to pregnancy, living in New Zealand for more
than 10 years, twin birth status, not seeing a midwife during pregnancy, and
smoking during pregnancy. Variables included in the model but failing to reach
significance were the five demographic control variables, and cultural
alignment.
Table 2. Adjusted odds of not breastfeeding exclusively
at discharge from hospital for variables attaining significance in a multiple
logistic regression (n=1235)§
*p<0.05;
†p<0.01;
‡p<0.001;
§Variables included in the model but
failing to reach significance were ethnicity, marital status, age, education,
household income, and acculturation.
To examine factors associated with a change from exclusive
breastfeeding in hospital to not exclusively breastfeeding by 6 weeks post
birth, data from the 1017 mothers (who were initially exclusively breastfeeding)
were also assessed via univariate and multiple logistic regression analyses.
Of the 1017 mothers who initially breastfed exclusively, 631
(62%) continued to do so at 6 weeks. In addition to the variables examined at
the time of hospital discharge, 10 variables (based on events post-discharge and
gathered at the 6-week interview) were included to identify any potential
association with not breastfeeding exclusively at 6 weeks post-birth. The 10
variables were current employment status; current smoking status; alcohol
consumed since birth; whether they had a home visit for the infant from a
midwife, traditional healer, or a Plunket nurse; use of a dummy; infant’s
feeding pattern (on demand or to a schedule); whether the infant sleeps in the
parental room at night; and use of regular childcare.
Fourteen variables reached statistical significance
(p<0.05) and findings for these variables are shown in Table 3.
Table 3. Numbers (row percentages) and odds ratios of
not breastfeeding exclusively (at 6 weeks post-birth) by variables attaining
significance in univariate logistic regression analyses (n=1017)
*p<0.05;
†p<0.01;
‡p<0.001;
§Includes mothers identifying equally with
two or more Pacific Island groups, equally with Pacific Island and non Pacific
Island groups, or with Pacific Island groups other than Tongan, Samoan, Cook
Island Maori or Niuean; ‼Univariate odds
ratio.
The multiple logistic regression analysis followed that
described previously, except the additional 10 variables gathered during the
6-week interview were added for potential inclusion during the stepwise
procedure.
Table 4 demonstrates that, when adjusting for all other
variables in the final model, factors significantly associated with not
breastfeeding exclusively (p <0.05) at 6 weeks post-birth were employment
prior to pregnancy, being currently employed, parity of five or more children,
current smoking (smoked yesterday), having a home visit from a traditional
healer, not receiving a home visit from a Plunket nurse, dummy use, the infant
not sharing the same room as the parents at night, regular childcare
arrangements, and the infant not discharged home from hospital at the same time
as the mother.
Table 4. Adjusted odds of not breastfeeding exclusively
(at 6 weeks post-birth) for variables attaining significance in a multiple
logistic regression (n=1004)§
*p<0.05;
†p<0.01;
‡p<0.001;
§Variables included in the model but
failing to reach significance were ethnicity, marital status, age, education,
and household income.
DiscussionOn the assumption that the more
important associations between the variables examined and not breastfeeding
exclusively are those identified by the multiple regression analyses, the
discussion will focus on these findings.
With the exception of maternal smoking, it appears that
different factors are associated with the initiation and maintenance of
exclusive breastfeeding among mothers of Pacific infants. Mothers who smoked
during pregnancy were almost twice as likely not to begin exclusive
breastfeeding compared to non-smokers. Similarly, mothers who reported that they
were current smokers were over twice as likely to have abandoned exclusive
breastfeeding by 6 weeks post-birth.
The negative association between maternal smoking and
breastfeeding has been reported elsewhere.12,13
Both physiological and psychosocial factors have been postulated to
explain lower rates of initiation and duration of breastfeeding among smokers
compared to non-smokers.7,14 It has also been
suggested that some smoking women may decide against breastfeeding in order to
reduce risk to their infants, unaware that not to breastfeed at all is less
optimal than smoking while
breastfeeding.15
Employment status was associated with not breastfeeding
exclusively, but the effects differed regarding initiation and maintenance of
breastfeeding. Employment prior to pregnancy reduced risk of not breastfeeding
exclusively at discharge from hospital, whereas (by six weeks) both prior and
current employment increased risk of not breastfeeding exclusively. Other
unmeasured factors associated with being previously employed may explain the
initial reduced risk, whereas the practicalities of a return to employment (or
plan to return) may explain the change from initial exclusive breastfeeding to
alternative feeding methods by 6 weeks post-birth.
Research suggests that the relationship between employment
and breastfeeding is complex, and findings can differ depending on how and when
employment status is measured. Most studies find negative associations between
employment and breastfeeding, particularly the relationship between return to
employment and shortened duration of
breastfeeding,16,17 although timing and
intensity of return to employment are facets that complicate this negative
association.17
Factors associated with an increased risk of not
breastfeeding exclusively at discharge from hospital (but not associated with
continuation to 6 weeks) included longer residency in New Zealand, twin birth
status, not seeing a midwife during pregnancy, and caesarean delivery. Mothers
who have resided in New Zealand for over 10 years were more likely not to
breastfeed exclusively at discharge from hospital compared to those who have
lived here less than 5 years.
Other studies report similar findings, with shorter length
of residence following immigration being associated with a greater likelihood of
breastfeeding.18 Longer-term exposure to the
availability of infant formulas in New Zealand may explain these differences. In
concordance with other authors,7,19 given the
obvious demands placed by multiple births, it was not surprising to find that
twin birth status increased risk of not breastfeeding exclusively.
Mothers who did not see a midwife during pregnancy may have
missed the opportunity to gain breastfeeding advice prior to delivery, possibly
influencing feeding decisions. Other studies have also found caesarean delivery
to negatively influence initiation of breastfeeding, but this is often overcome
following discharge from hospital or once feeding is
established.20,21 Hospital practices, level of
postoperative recovery, effects of postoperative drugs, and child illness are
possible factors that may impede initiation of breastfeeding following a
caesarean delivery.20–22
Factors associated with an increased risk of cessation of
exclusive breastfeeding (between hospital discharge and 6 weeks post-birth)
included higher parity, infant not discharged home at the same time as the
mother, having a home visit from a traditional healer, not receiving a home
visit from a Plunket nurse, regular childcare arrangements, dummy use, and the
infant not sharing the same room as the parent(s) at night.
Inconsistent findings have been found in the literature
between parity and initiation and maintenance of
breastfeeding.23 In contrast to a large,
nationally representative American Study,13
women in our study were less likely to continue to breastfeed exclusively if
they had five or more children. Additional pressures within the household posed
by caring for a larger family may dissuade mothers from exclusive breastfeeding.
Mothers were over three times more likely to not continue
exclusively breastfeeding if their infant was not discharged from hospital at
the same time as them. Such a finding is not unexpected as this variable implies
need for specialist neonatal care. Maintenance of exclusive breastfeeding in
these cases is likely to be compromised and has been observed
previously.7 The use of regular childcare has
similar constraints for breastfeeding due to absence from the infant.
Although it appears that newer migrants to New Zealand might
be more traditional in terms of infant feeding practices, the finding that
postnatal home visits by a traditional healer increased the risk of not
breastfeeding exclusively was somewhat unexpected and warrants further study.
Examination of other data for these participants indicated that, in most cases,
traditional massage and/or herbal treatment was sought to assist the closure of
the infant’s fontanelle. It is not known whether breastfeeding was
hampered by this traditional treatment or whether the healer variable acted as a
proxy for other influential factors inherent in the mother or the child.
The opposite effect was observed in relation to Plunket
nurses, which is encouraging news as the majority of mothers are seen by Plunket
nurses. However, approximately 16% of mothers not seen by Plunket nurses, appear
to have fallen through the gaps. Concern has been expressed previously about
timing of the commencement of well child services—including late referrals
and difficulty in locating mothers upon discharge from
hospital.24
Infant care practices were also associated with a greater
likelihood of not breastfeeding exclusively by 6 weeks post-birth. In
concordance with other studies,5,7,25 if the
infant did not sleep in the same room as the parent(s) at night, mothers were
more than twice as likely not to exclusively breastfeed. It should be noted that
approximately half of those sharing the room shared the same bed (analyses not
shown here).
Despite indications that mothers may bed-share for the
purpose of breastfeeding,5 it is unclear
whether bed or room sharing is causally related to, or a consequence of, not
breastfeeding exclusively. Avoidance of mixed health messages is, however, of
critical importance for bed sharing given that it is a common practice among
Pacific families in New Zealand,26 and that
infants may be at increased risk for SIDS, particularly when sharing a bed with
mothers who smoke.27 Instead, sharing the same
room rather than the same bed is
recommended.25
Mothers who reported their infant used a dummy were 2.5
times more likely to have changed to not breastfeeding exclusively by 6 weeks.
Although dummy-use has been linked to a shorter duration of
breastfeeding5,7,25 it remains controversial
due to inconsistent findings.28 As with bed or
room sharing, it is not known whether dummy-use is a cause, or consequence, of
reduced breastfeeding.
In conclusion, several factors were identified to be
independently negatively associated with exclusive breastfeeding at the time of
discharge from hospital and at 6 weeks post-birth. Many of these factors (eg,
maternal smoking) support previous research, while others (such as the use of
traditional healers) are more specific to the Pacific Island population and
require further investigation within the New Zealand context.
Health professionals should be alerted to circumstances that
may lead to unsuccessful establishment or maintenance of exclusive
breastfeeding. As not all risk factors are modifiable, it is important to
emphasise those that are, such as smoking. Furthermore, mothers should be
encouraged to lead healthier lifestyles and be educated on ways to reduce
possible harm to their infants15—not only
for greater success with breastfeeding, but as a preventative risk factor for
SIDS and other negative health consequences.29
The message to continue breastfeeding exclusively for the
first 6 months (despite smoking) should be reiterated—as it may provide
additional protection for the infant against respiratory
illness.30
Author information:
Sarnia Butler, Research Fellow, Pacific Islands Families: First Two Years of
Life Study, Auckland University of Technology; Maynard Williams, Senior Research
Fellow and Statistician, Auckland University of Technology; Colin Tukuitonga,
Pacific Health Research Centre, Department of Maori and Pacific Island Health,
University of Auckland—and Co-Director, Pacific Islands Families: First
Two Years of Life Study, Auckland University of Technology; Janis Paterson,
Co-Director, Pacific Islands Families Study, Auckland University of Technology,
Auckland.
Acknowledgements:
The Pacific Islands Families Study is supported by grants awarded from
the Foundation for Science, Research and Technology, the Health Research Council
of New Zealand, and the Maurice and Phyllis Paykel Trust. The authors gratefully
acknowledge the families who have participated in the study, the Pacific Peoples
Advisory Board, and the other members of the research team.
Correspondence:
Sarnia Butler, Faculty of Health Studies, Auckland University of Technology,
Private Bag 92006, Auckland. Fax: (09) 917 9877; email: sarnia.butler@aut.ac.nz
References:
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| | Current
issue | Search journal |
Archived issues | Classifieds
| Hotline (free ads) Subscribe | Contribute | Advertise | Contact Us | Copyright | Other Journals |
|