The influence of psychological factors on breastfeeding duration
Maxine O’Brien, Elizabeth Buikstra & Desley Hegney
Accepted for publication 1 April 2008
Correspondence to M. O’Brien:
e-mail: [email protected]
Maxine O’Brien BSc PhD
Postdoctoral Research Fellow
Centre for Rural and Remote Area Health,
University of Southern Queensland,
Toowoomba, Queensland, Australia
Elizabeth Buikstra BSc PhD
Statewide Project Officer
Centre for Rural and Remote Mental Health
Queensland, Cairns, Queensland, Australia
Desley Hegney BA PhD RN
Professor of Nursing, Director,
Research and Practice Development Centre,
University of Queensland and Blue Care,
Brisbane, Queensland, Australia
O’BRIEN M., BUIKSTRA E. & HEGNEY D.O’BRIEN M., BUIKSTRA E. & HEGNEY D. (20082008) The influence of psychological
factors on breastfeeding duration. Journal of Advanced Nursing 63(4), 397–408
doi: 10.1111/j.1365-2648.2008.04722.x
AbstractTitle. The influence of psychological factors on breastfeeding duration.
Aim. This paper reports on a study examining the relationship between women’s
psychological characteristics and breastfeeding duration, after controlling for socio-
demographic factors.
Background. The literature suggests that psychological factors may influence
breastfeeding behaviour, but studies are few. Existing evidence and the results of
phase 1 of our study were used to construct a list of psychological factors, which
were tested for their association with breastfeeding duration in the current design.
Method. Participants were postnatal inpatients in one of two regional hospitals
between October and December 2005 and they completed the initial questionnaire
within 14 days of giving birth (n = 375). Infant feeding method at 6 months and the
timing of introduction of other food(s), where relevant, were ascertained by tele-
phone interview.
Findings. Forty-four per cent of the sample showed signs of postnatal distress in the
14 days following the birth. Breastfeeding duration was statistically significantly
associated with psychological factors including dispositional optimism, breastfeed-
ing self-efficacy, faith in breastmilk, breastfeeding expectations, anxiety, planned
duration of breastfeeding and the time of the infant feeding decision. As a set, these
psychological factors were more predictive of breastfeeding duration than was the
set of socio-demographic characteristics. The duration of any breastfeeding was
uniquely predicted by faith in breastmilk, planned breastfeeding duration and
breastfeeding self-efficacy.
Conclusion. This increased knowledge of the factors influencing breastfeeding will
assist in identifying women at risk of early weaning and in constructing programmes
capable of increasing the length of time for which women breastfeed.
Keywords: breastfeeding duration, empirical research report, midwifery, psycho-
logical factors, questionnaire
Introduction
The numerous adverse consequences of replacing breastfeed-
ing with artificial baby formulas are well known (Horta et al.
2007, Ip et al. 2007). However, many mothers wean from
the breast well before the initial 6 months of exclusive
breastfeeding recommended by national and international
authorities (WHO 2001, Australian Bureau of Statistics
2003, Callen & Pinelli 2004). Previous research has demon-
strated the influence of socio-demographic factors on breast-
feeding duration (Callen & Pinelli 2004); however, these
factors are not easily modified and therefore offer little
ORIGINAL RESEARCHJAN
� 2008 The Authors. Journal compilation � 2008 Blackwell Publishing Ltd 397
opportunity to increase the duration of breastfeeding. This
has led to calls for breastfeeding researchers to focus on
identifying factors which may be open to modification (e.g.
Blyth et al. 2002).
One research area with the potential to identify modifiable
factors involves the study of psychological differences
between women and the effect of these differences on
breastfeeding behaviour. Breastfeeding research conducted
to date has demonstrated the effect of many factors associ-
ated with the human mind and its functions, including the
woman’s attitudes and plans regarding breastfeeding (Forster
et al. 2006, Scott 2006) and her confidence in her ability to
breastfeed (O’Brien & Fallon 2005), in addition to factors
such as depression (Henderson et al. 2003) and self-esteem
(Damato 2005). However, to date little research has been
performed to explore more fully the psychological charac-
teristics of women that may influence breastfeeding duration,
or just how predictive these characteristics are.
Background
The data reported in this paper were gathered during the
second phase of a two-phase study designed to identify and
explore individual psychological differences between women
and their influence on breastfeeding duration. In phase 1, we
gathered the opinions of mothers and breastfeeding clini-
cians, asking them to draw on their intimate knowledge of
breastfeeding to decide which psychological factors they
believed influence breastfeeding duration. In an attempt to
draw on all available sources of evidence, this information
was considered along with a careful review of the literature.
Taken as a whole, the literature and the phase 1 data support
a theoretical framework (Shields & Tajalli 2006) summarized
by the working hypothesis that individual psychological
differences exist between women which explain breastfeeding
duration, over and above the influence of socio-demographic
characteristics.
In selecting psychological factors to be measured in the
current study, factors thought to be of most importance by
phase 1 participants were given priority over those thought to
be of less importance. Preference was also given to factors
signalling individual psychological differences between
women, as opposed to factors of a more social nature which
may pose problems with modification as our ultimate goal.
The decision process resulted in the selection of 15 psycho-
logical factors for measurement: the woman’s self-esteem,
current life priorities, psychological reactance (motivation to
regain lost or threatened personal freedoms), adaptability,
optimism, breastfeeding self-efficacy, achievement striving,
faith in breastmilk, breastfeeding expectations, depression,
anxiety, stress, time of decision, planned breastfeeding dura-
tion and mothering self-efficacy. An association between
breastfeeding duration and five of these factors is supported
by the results of several previous studies, while four other
factors are less well established, and the remainder have
attracted little or no known research support in the past. A list
of the psychological factors included in our design and a
selection of the published research are shown in Table 1.
Socio-demographic influences previously identified in the
literature were measured and controlled during the analysis.
This list of factors included the woman’s age, education,
occupation, marital status, socio-economic status, parity,
previous breastfeeding experience, hospital where the birth
occurred, rural/remote or regional dwelling and planned
return to employment, in addition to her partner’s prefer-
ences and occupation (Fitzpatrick et al. 1994, Bourgoin et al.
1997, Bulk-Bunschoten et al. 2001, Nagy et al. 2001, Callen
& Pinelli 2004, Arora et al. 2005, Stamp & Casanova 2006,
Bolling et al. 2007).
The study
Aim
The aim of this study was to examine the relationship between
women’s psychological characteristics and breastfeeding
duration, after controlling for socio-demographic factors.
Design
A prospective survey-based design was used as the final phase
of a triangulated, mixed-method research programme, as
described above (Denzin & Lincoln 2005).
Participants
The study was conducted in a large regional centre of
Queensland, Australia, which has two hospitals with mater-
nity units, one publicly funded and one privately funded.
Women who gave birth in either of these hospitals and were
inpatients between October 2005 and December 2005 were
invited to participate in the project. Eligibility criteria
excluded 26 women who were under 18 years of age; had
insufficient English language skills to complete the question-
naire; were too ill; or whose baby was too ill, had died or was
not in their care. Of the 657 women who gave birth during
the study period, 375 completed the questionnaire within the
14 day cut-off period and returned it to the researcher,
leaving an overall participation rate of 59Æ4% (private
hospital = 75Æ7%, public hospital = 45Æ2%). The rate of
M. O’Brien et al.
398 � 2008 The Authors. Journal compilation � 2008 Blackwell Publishing Ltd
attrition at follow-up was relatively low (15%), and the use
of survival analysis and Cox regression techniques allowed all
data collected at the first interview to be included in the
major analyses.
Data collection
Questionnaires
Data for the initial interview were collected using a 123-item
self-report questionnaire arranged in three sections. The first
two sections were based on previous research (O’Brien &
Fallon 2005) and contained items designed to cover the socio-
demographic information indicated in the literature as being
relevant to breastfeeding duration and the participant’s cur-
rent feeding method (Webb et al. 2001). The third section
contained measures of the woman’s psychological charac-
teristics and space for additional qualitative comments where
desired.
Prevalidated psychometric scales were used where available.
Where such scales were not available, researcher-designed
scales were employed. These included scales measuring the
planned duration of breastfeeding and the time of the infant
feeding decision, both of which had been validated in an earlier
study (O’Brien & Fallon 2005). Additional measures of the
woman’s faith in breastmilk (relative to faith in formula),
breastfeeding expectations and maternal priorities were
constructed for use in the current study.
Faith in breastmilk was measured using a two-item scale.
The first item measured faith in breastmilk (‘If I feed my baby
breastmilk, I can relax and be sure he/she is getting everything
he/she needs’); and a second, similarly worded item, mea-
sured faith in formula. Participants were asked to respond to
the items on a six-point scale from (1) ‘strongly disagree’ to
(6) ‘strongly agree’. Scores on the two items were combined
to produce an overall score, where high scores indicated high
faith in breastmilk and low faith in formula. While Cron-
bach’s alpha is inappropriate for use with brief scales such as
this, the inter-item correlation of 0Æ33 suggested satisfactory
internal consistency for this two-item scale (Briggs & Cheek
1986).
Similarly, the woman’s expectations regarding breastfeed-
ing were measured with two items representing opposing
views of the level of difficulty they expected to encounter (e.g.
‘I don’t expect to experience any real trouble with breast-
feeding, I think it will be easy for me’). Participants
responded on a six-point scale from (1) ‘strongly disagree’
to (6) ‘strongly agree’. Responses on the two items were
combined to form a single score, with high scores indicating
more expected difficulty with breastfeeding (inter-item cor-
relation = 0Æ50).
The placement of the new baby at the top of the mother’s
priorities was considered by phase 1 participants to be
important for breastfeeding success. To measure this novel
factor, participants were asked to rate nine potentially
Table 1 Exemplars of previous research support for psychological factors selected for inclusion in the phase 2 quantitative enquiry
Factors previously established in the literature
Self-esteem Damato (2005), Papinczak and Turner (2000), Whelan and Lupton (1998)
Depression Dunn et al. (2006), Forster et al. (2006), Hatton et al. (2005), Henderson et al. (2003),
McLearn (2006)
Time of infant feeding decision O’Brien and Fallon (2005), Boulton and Landers (1999), Scott et al. (2001a, 2001b)
Planned duration of breastfeeding Blyth et al. (2004), Forster et al. (2006), Donath et al. (2004), Kronborg and Vaeth (2004),
Whaley et al. (2002)
Breastfeeding self-efficacy O’Brien and Fallon (2005), Blyth et al. (2002, 2004), Creedy et al. (2003),
Dennis and Faux (1999)
Factors with only tentative research support
Mothering self-efficacy Tarkka et al. (1999)
Achievement striving Rentschler (1991)
Stress Ueda et al. (1994)
Anxiety Papinczak and Turner (2000), Cooke et al. (2007) NB. Cooke = statistically nonsignificant result
Factors raised in previous qualitative enquiries (no known previous quantitative support)
Need for freedom Bulk-Bunschoten et al. (2001), Kirkland and Fein (2003), Kieffer et al. (1997)
Adaptability Hegney et al. (2007), Hewat and Ellis (1986), Vandiver (1997)
Breastfeeding expectations Hegney et al. (2007), Mozingo et al. (2000)
Faith in breastmilk Bottorff (1990), Hegney et al. (2007)
Novel factors included
Optimism Effect on infant feeding choice supported by Wagner et al. (2006)
Current life priorities Raised in phase 1 of the current study only
JAN: ORIGINAL RESEARCH Influence of psychological factors
� 2008 The Authors. Journal compilation � 2008 Blackwell Publishing Ltd 399
competing priorities on a scale of 1–10, with 1 being the
lowest priority and 10 being the highest. Scale items included
priorities such as this baby, husband/partner, other children,
paid work and time for myself. The scale also included an
‘other’ category. A single dichotomous variable was then
created for the analysis. This identified participants who had
assigned this baby a higher priority than any other and those
who had assigned one or more competing priorities the same
or higher priority rating compared with this baby. A list of all
instruments used to measure psychological factors is pre-
sented as Table 2.
At the 6-month follow-up, two to five items were admin-
istered during a telephone interview, depending on the
woman’s feeding method at that time. This instrument was
based on earlier research (O’Brien & Fallon 2005) and
gathered information on the woman’s feeding method at that
time and the time of introduction of other foods or the
cessation of breastfeeding where appropriate.
Procedure
The questionnaire was piloted with a small group of mothers
and subject-matter experts (n = 11) and changes made as
required. Eligible women were approached by the researcher
during their postnatal hospital stay and invited to participate
in the research by completing the initial questionnaire within
14 days and accepting a brief follow-up telephone call at
6 months postpartum. After explaining the study and
receiving signed consent, consenting participants were left
with the coded self-report questionnaire, a copy of the plain
language statement describing the study and a postage-paid
envelope. The participant was asked to place the completed
questionnaire in a locked box near the nurses’ station or
return it through the postal system.
Participants were telephoned for follow-up when their
baby was 6 months old. Participants were considered lost to
follow-up if the telephone numbers supplied were discon-
nected, or if the researcher was unable to contact them after a
minimum of five attempts.
Ethical considerations
Ethics approval was gained from the appropriate university
and health service committees. Informed consent was
obtained from all participants. Data were stored securely
according to Australian National Health and Medical
Research Council guidelines (NHMRC 2007) and all data
were kept separately from any information which could
identify participants.
Data analysis
The aim of the statistical analysis was to assess the relation-
ship between time of cessation of breastfeeding and the set of
influencing factors and to determine whether the psycholog-
ical factors under investigation made a unique contribution to
the prediction after controlling for the remaining variables
(working hypothesis). An alpha level of P < 0Æ05 was used
for all analyses, which were performed using the Windows
version of the Statistical Package for Social Sciences (SPSS
2005).
Prior to analysis, data were inspected for accuracy of entry,
missing values and the fit between their distributions and the
assumptions of the various analyses to be undertaken. The
three Depression, Anxiety and Stress Scales (DASS) sub-scales
were statistically significantly positively skewed and dis-
played statistically significant positive kurtosis. Scores on
Table 2 Instruments used to measure psychological factors
Factor Instrument used
Self-esteem Rosenberg’s Self-Esteem Scale Rosenberg (1989)
Psychological reactance Hong’s Psychological Reactance Scale – Refined Hong and Faedda (1996)
Adaptability Adaptability Scale from the International Personality Item Pool Goldberg et al. (2006)
Dispositional optimism Life Orientation Test-Revised Scheier et al. (1994))
Breastfeeding self-efficacy Breastfeeding Self-Efficacy Scale – Short Form Dennis (2003)
Depression Depression, Anxiety Stress Scale-21 Lovibond and Lovibond (1995)
Anxiety Depression, Anxiety Stress Scale-21 Lovibond and Lovibond (1995)
Stress Depression, Anxiety Stress Scale-21 Lovibond and Lovibond (1995)
Achievement striving Achievement Strivings Scale [adapted] Spence et al. (1987)
Mothering self-efficacy Self-efficacy subscale from the Parenting Sense of Competence Scale Gibaud-Wallston (1977)
Time of decision Single Likert-scale item validated during an earlier study O’Brien and Fallon (2005)
Planned duration Single item validated during an earlier study O’Brien and Fallon (2005)
Faith in breastmilk Novel scale (properties described in text)
Breastfeeding expectations Novel scale (properties described in text)
Maternal priorities Novel scale (properties described in text)
M. O’Brien et al.
400 � 2008 The Authors. Journal compilation � 2008 Blackwell Publishing Ltd
these variables were normalized using log (Depression and
Anxiety) and square root (Stress) transformations. Analyses
were run first with the transformed scores and again with the
untransformed scores and the results compared. As no
meaningful differences existed in these results, the original
scores were retained to aid interpretation. Inspection of
correlations and tolerance scores revealed no evidence of
multicollinearity in the multivariate data set and no multi-
variate outliers existed [Mahalanobis Distance v2(5 = 20Æ52,
P < 0Æ001)]. The proportionality of hazards assumption was
tested by examination of the interactions of each covariate
with a variable representing the natural logarithm of time
(using a = 0Æ004 to account for familywise error) and no
statistically significant interactions were observed.
The accuracy and precision of Cox regression survival
analysis becomes untrustworthy when the number of events
per variable becomes too low. In the case of this study,
cessation of breastfeeding/fully breastfeeding represented
such an ‘event’. Conservative guidelines calling for a mini-
mum of 10 events per variable were followed to ensure
sufficient power for the Cox regressions (Peduzzi et al. 1995).
Of the 342 participants who initiated fully breastfeeding and
provided complete data sets, 302 had ceased fully breast-
feeding (event occurred) at 26 weeks, allowing for the
inclusion of up to 30 covariates and ample power to test
the current model for the cessation of fully breastfeeding (12
covariates). Of the 351 women who initiated any breastfeed-
ing and provided complete data sets, 137 had ceased any
breastfeeding at 26 weeks (event occurred), allowing up to 13
covariates in the model. Power for the 12 covariates utilized
in this analysis was therefore adequate.
Spearman’s rho correlations and Wilcoxon (Gehan) statis-
tics were employed to test the bivariate relationship between
variables of interest and the duration of fully or any
breastfeeding. Variables statistically significantly related to
breastfeeding duration in these analyses were entered into a
sequential Cox regression equation in two steps, with all
statistically significant socio-demographic factors entered at
step 1 and statistically significant psychological factors
entered at step 2. All analyses were performed twice, first
using the data measuring the duration of fully breastfeeding
and second using data measuring the duration of any
breastfeeding.
Validity and reliability
With the exception of the three researcher-designed scales
described above, the validity and reliability of scales used to
measure psychological constructs in this study were estab-
lished during psychometric testing, with the results published
by the scale authors. Tabachnick and Fidell (2001) recom-
mend a Cronbach’s alpha coefficient of 0Æ70 or above as
signalling satisfactory internal consistency for these tests and
this statistic was exceeded with the current sample (range
a = 0Æ74–0Æ95) for all but two of these instruments.
One of the two problematic instruments was the Adapt-
ability Scale from the International Personality Item Pool
(Goldberg 1999, Goldberg et al. 2006), which achieved
a = 0Æ69 with the present sample following the deletion of
two items with inter-item correlations below 0Æ3. The second
problematic scale was the Achievement Strivings Scale
(Spence et al. 1987), which also achieved a Cronbach’s alpha
of 0Æ69 after the deletion of one item with an inter-item
correlation below 0Æ3. However, no statistically significant
association was found between scores on either of these scales
and breastfeeding duration, thereby automatically excluding
these data from the multivariate analysis.
Results
At the time of the first interview the average age of
participants’ babies was 3Æ5 days (range = 0–14). The typical
participant was married or in a cohabiting relationship (92%)
and believed that their partner preferred breast over artificial
feeding (72%). Most participants had already had at least
one baby (61%) and had no current plans regarding when
they may return to full- (88%) or part- time (63%) work. The
average participant was approximately 30 years of age
(range = 18–45 years) and had completed 13Æ6 years of
education (range = 9–22 years). Breastfeeding was defined
according to current Australian guidelines (Webb et al. 2001)
and the breastfeeding rates observed were similar to those
seen in previous Australian studies (e.g. Hegney et al. 2003).
Breastfeeding rates observed in the current study are pre-
sented in Table 3.
Results on the Depression, Anxiety and Stress Scales DASS-
21 showed that, at the time of the first interview, >44% of
the sample reported symptoms of postnatal distress in the
form of mild to extremely severe stress, anxiety and/or
depression. Without taking co-morbidity into account,
Table 3 Breastfeeding rates
Feeding method Time 1 (0–14 days) 6 months
Fully breastfeeding 82Æ8 10Æ1Complementary breastfeeding 7Æ8 44Æ2Breastfeeding (any) 90Æ6 54Æ3Artificially feeding 9Æ4 45Æ7
Feeding methods classified according to current Australian guidelines
Webb et al. (2003).
Values are expressed as %(n).
JAN: ORIGINAL RESEARCH Influence of psychological factors
� 2008 The Authors. Journal compilation � 2008 Blackwell Publishing Ltd 401
symptoms of stress were reported by approximately 26% of
the sample, 15% reported symptoms of depression and 34%
reported symptoms of anxiety. Of these, co-morbid anxiety
and depression were experienced by 12Æ1% of women in the
study.
Spearman’s Rho correlations revealed that the duration of
both any breastfeeding (BF) and fully breastfeeding (FBF) was
statistically significantly related to continuous measures of
the mother’s age (FBF and BF rs = 0Æ2), education (FBF
rs = 0Æ19, BF = 0Æ24), dispositional optimism (FBF rs = 0Æ12,
BF rs = 0Æ18), breastfeeding self-efficacy (FBF rs = 0Æ48, BF
rs = 0Æ46), faith in breastmilk (FBF rs = 0Æ49, BF rs = 0Æ39),
breastfeeding expectations (FBF rs = �0Æ36, BF rs = �0Æ30)
and anxiety (FBF rs = �0Æ12, BF rs = �0Æ16). For the
categorical variables, Wilcoxon (Gehan) statistics from actu-
arial life tables revealed statistically significant associations
(a = 0Æ05, df = 1) between the duration of breastfeeding and
the mother’s usual occupation (FBF v2 = 6Æ7, BF v2 = 6Æ5),
previous breastfeeding experience (FBF v2 = 13Æ4, BF
v2 = 15Æ7), partner’s infant feeding preference (FBF
v2 = 22Æ1, BF v2 = 28Æ9), the time of the infant feeding
decision (FBF v2 = 15Æ7, BF v2 = 20Æ6) and planned duration
of breastfeeding (FBF v2 = 38Æ5, BF v2 = 57Æ1).
To determine whether the psychological variables would
predict breastfeeding duration over and above the influence
of other related variables, thereby supporting our working
hypothesis, variables statistically significantly associated with
breastfeeding duration were entered into Cox regression
equations in two blocks for both BF and FBF. The five socio-
demographic variables entered at step 1 statistically signifi-
cantly predicted the duration of both BF [v2 (5) = 49Æ51,
P £ 0Æ001, R2 = 0Æ15] and fully breastfeeding [v2 (5) = 26Æ98,
P £ 0Æ001, R2 = 0Æ08]. The subsequent inclusion of the five
statistically significant psychological variables at step 2
significantly increased the model’s ability to predict the
premature cessation of both BF [v2 (7) = 90Æ62, P £ 0Æ001, R2
Change = 0Æ29] and fully FBF [v2 (7) = 54Æ78, P £ 0Æ001, R2
Change = 0Æ17]. Overall, the model with all variables
included moderately predicted the premature cessation of
FBF [v2 (12) = 83Æ74, P £ 0Æ001, R2 = 0Æ27] and more
strongly predicted the duration of BF [v2 (12) = 148Æ62,
P £ 0Æ001, R2 = 0Æ49] in this sample.
Alpha for the statistical significance of covariates was set at
a conservative level of P < 0Æ004 to adjust for familywise
error for 12 covariates (Tabachnick & Fidell 2001). After
controlling for all variables in the model (step 2), the
woman’s faith in breastmilk and her planned duration of
breastfeeding were reliable unique predictors of the duration
of fully breastfeeding. The working hypothesis was therefore
supported. The probability of continuing to fully breastfeed
increased by 36% (1Æ36 times) for every one-point increase in
the faith in breastmilk score (score range 1–5), and partic-
ipants were 1Æ72 times more likely to cease fully breastfeeding
prematurely if they had planned to breastfeed for 6 months
or less.
After controlling for the effect of socio-demographic
factors in the model, three psychological variables were
found to be reliable unique predictors of the duration of any
breastfeeding, providing further support for our working
hypothesis. Every one-point increase in the breastfeeding
Table 4 Results of the cox regression
predicting duration of fully breastfeedingCovariate B SESE Odds ratio 95% CI P value
Step 1 – Socio-demographic variables
Years of education �0Æ05 0Æ03 0Æ96 0Æ91–1Æ01 0Æ084
Mother’s occupation 0Æ06 0Æ17 1Æ06 0Æ77–1Æ47 0Æ720
Breastfeeding experience �0Æ24 0Æ13 0Æ79 0Æ62–1Æ01 0Æ060
Partner’s preference �0Æ49 0Æ14 0Æ61 0Æ46–0Æ81 0Æ001
Mother’s age �0Æ03 0Æ01 0Æ97 0Æ95–1Æ00 0Æ036
Step 2 – Addition of psychological variables
Years of education �0Æ05 0Æ03 0Æ95 0Æ90–1Æ00 0Æ066
Mother’s occupation 0Æ03 0Æ17 1Æ03 0Æ74–1Æ44 0Æ868
Breastfeeding experience 0Æ02 0Æ13 1Æ02 0Æ78–1Æ32 0Æ903
Partner’s preference �0Æ26 0Æ15 0Æ77 0Æ57–1Æ04 0Æ084
Mother’s age �0Æ03 0Æ01 0Æ97 0Æ94–0Æ99 0Æ007
Breastfeeding self-efficacy �0Æ02 0Æ01 0Æ98 0Æ96–0Æ99 0Æ008
Optimism 0Æ01 0Æ02 1Æ01 0Æ98–1Æ05 0Æ544
Anxiety �0Æ02 0Æ01 0Æ99 0Æ97–1Æ01 0Æ144
Faith in breastmilk 0Æ31 0Æ09 1Æ36 1Æ14–1Æ62 0Æ001
Breastfeeding expectations 0Æ01 0Æ07 1Æ01 0Æ88–1Æ17 0Æ853
Planned duration 0Æ55 0Æ14 1Æ72 1Æ31–2Æ27 <0Æ001
Time of decision �0Æ26 0Æ14 0Æ77 0Æ58–1Æ02 0Æ065
M. O’Brien et al.
402 � 2008 The Authors. Journal compilation � 2008 Blackwell Publishing Ltd
self-efficacy score resulted in a 5% increase in the likelihood
of continuing any breastfeeding to 6 months (Breastfeeding
Self-Efficacy Scale – Short Form, score range 14–70). Also,
participants were 1Æ70 times more likely to continue any
breastfeeding for each one-point increase in faith in breast-
milk score (score range 1–5) and were more than twice as
likely to cease breastfeeding prematurely if they had planned
to breastfeed for 6 months or less (odds ratio = 2Æ19). The
results of these Cox regression equations are presented in
Tables 4 and 5.
Discussion
There are several limitations to the study that should be
taken into account when attempting to draw conclusions
from its results. Firstly, while the participation rate com-
pared well with other postnatal research, previous research
suggests that important differences in the demographic
characteristics of participants and non-participants may
exist, such that younger and unmarried women are less
likely to participate (Hegney et al. 2003). The current
sample was certainly relatively homogeneous with regard to
marital status (92% married/cohabiting). It should also be
noted that women <18 years of age or with insufficient
English language skills were excluded from participating.
Particular caution in generalizing these results to young,
unmarried or non-English-speaking women is therefore
warranted. However, in acknowledging this, it should also
be noted that while most hospital-based breastfeeding
studies gather data from a single maternity unit, we included
data from two maternity units, one public and one private,
thus increasing generalizability across the Australian hospi-
tal system.
There were some measurement issues which may also have
influenced the study results. In addition to the concerns about
using the three novel scales employed in this study (i.e. Faith
in Breastmilk, breastfeeding expectations and current life
priorities), two of the previously validated instruments
displayed less than optimal reliability. These were the
International Personality Item Pool Adaptability scale (Gold-
berg et al. 2006) and the Achievement Strivings scale (Spence
et al. 1987). These factors were not statistically significantly
associated with breastfeeding duration and were excluded
from the multivariate analysis. However, further research to
establish the relationship between adaptability, achievement
striving and breastfeeding duration is warranted. Despite
these limitations, the results offer several new insights which
may be of utility to those with an interest in breastfeeding.
First, while a clinical interview is necessary to diagnose
stress, anxiety or depression accurately in individuals (Lovi-
bond & Lovibond 1995), use of the DASS-21 in this study
revealed a somewhat alarming level of potential postnatal
distress among participants, with 44Æ1% of the sample
reporting symptoms of these states. Also, while the literature
appears to focus on the incidence and effects of depression in
the postpartum period, the most common form of postnatal
distress reported in this sample was anxiety. Only 3% of the
sample revealed symptoms of depression alone, while 21Æ8%
reported symptoms of anxiety alone and a further 12Æ1% were
anxious–depressed. This is particularly important given that
Table 5 Results of the cox regression
predicting duration of any breastfeedingCovariate B SESE Odds ratio 95% CI P value
Step 1 – Socio-demographic variables
Years of education �0Æ06 0Æ04 0Æ94 0Æ87–1Æ02 0Æ114
Mother’s occupation �0Æ15 0Æ27 0Æ86 0Æ51–1Æ47 0Æ586
Breastfeeding experience �0Æ66 0Æ19 0Æ52 0Æ36–0Æ75 0Æ001
Partner’s preference �0Æ99 0Æ18 0Æ37 0Æ26–0Æ53 <0Æ001
Mother’s age �0Æ02 0Æ02 0Æ99 0Æ95–1Æ02 0Æ376
Step 2 – Addition of psychological variables
Years of education �0Æ08 0Æ04 0Æ92 0Æ85–1Æ00 0Æ050
Mother’s occupation �0Æ18 0Æ28 0Æ83 0Æ49–1Æ43 0Æ508
Breastfeeding experience �0Æ22 0Æ20 0Æ80 0Æ55–1Æ18 0Æ264
Partner’s preference �0Æ50 0Æ20 0Æ61 0Æ41–0Æ90 0Æ013
Mother’s age �0Æ03 0Æ02 0Æ97 0Æ93–1Æ00 0Æ075
Breastfeeding self-efficacy �0Æ05 0Æ01 0Æ95 0Æ93–0Æ97 <0Æ001
Optimism 0Æ03 0Æ03 1Æ03 0Æ98–1Æ08 0Æ296
Anxiety �0Æ01 0Æ02 0Æ99 0Æ97–1Æ02 0Æ620
Faith in breastmilk 0Æ53 0Æ15 1Æ70 1Æ26–2Æ29 0Æ001
Breastfeeding expectations 0Æ04 0Æ12 1Æ04 0Æ82–1Æ31 0Æ759
Planned duration 0Æ79 0Æ19 2Æ19 1Æ52–3Æ16 <0Æ001
Time of decision �0Æ14 0Æ20 0Æ87 0Æ60–1Æ28 0Æ487
JAN: ORIGINAL RESEARCH Influence of psychological factors
� 2008 The Authors. Journal compilation � 2008 Blackwell Publishing Ltd 403
higher anxiety was associated with earlier weaning, while
depression had no statistical relationship to breastfeeding
duration in this sample. The lower incidence of depression
found in this study may be because of the widespread use of the
Edinburgh Postnatal Depression Scale (Cox & Sagovsky
1987). This instrument is thought to measure both depression
and anxiety, without adequately distinguishing between these
two distinct psychological states (Brouwers et al. 2001).
In addition to depression, seven other psychological factors
included in the design revealed no statistical relationship to
breastfeeding duration in this sample. These unrelated factors
were psychological reactance (the motivation to regain lost or
threatened personal freedoms), adaptability, mothering self-
efficacy, the mother’s current life priorities, achievement
striving, self-esteem and stress. This result is surprising, given
the evidence suggesting a link between many of these factors
and breastfeeding behaviour (see Table 1). It is also worth
noting that two of these factors (the mother’s life priorities
and mothering self-efficacy) were together ranked as the most
important factor influencing breastfeeding duration by phase
1 participants.
In total, the results revealed seven psychological factors
which were statistically associated with the duration of
breastfeeding: anxiety, dispositional optimism, the timing of
the infant feeding decision, breastfeeding expectations,
planned breastfeeding duration, faith in breastmilk and
breastfeeding self-efficacy. Three of these factors have been
identified as predictors of breastfeeding duration in previous
studies, namely the time of the infant feeding decision,
breastfeeding self-efficacy (O’Brien & Fallon 2005) and
planned breastfeeding duration (Blyth et al. 2004). A
relationship between breastfeeding duration and anxiety has
also been found (Papinczak & Turner 2000). However,
measures used have often been less than ideal (Clifford et al.
2006, Forster et al. 2006), making the current results the
most decisive indication to date of a relationship between
maternal anxiety and early weaning. In the case of breast-
feeding expectations, previous qualitative research suggests
that women may experience a clash between highly idealized
antenatal expectations of breastfeeding and early problems,
leading to disillusionment and early weaning (Mozingo et al.
2000). However, we measured expectations postnatally and
found the opposite to be true. In this study, women who
believed that breastfeeding would be difficult for them
when asked in the 14 days following the birth were more
likely to wean early than those who believed it would be easy,
perhaps because of the experience of early breastfeeding
problems.
Two novel factors associated with breastfeeding duration
were also identified in this study: dispositional optimism and
faith in breastmilk. Optimism has recently been linked to
choice of infant feeding method (Wagner et al. 2006).
However, we found a statistically significant relationship
between optimism and breastfeeding duration, such that the
higher the score on dispositional optimism, the longer the
woman breastfed for. This fits well with what is known about
other, related psychological influences on breastfeeding and
the theories underlying dispositional optimism (Scheier &
Carver 1992) and supports the inclusion of this factor in
future research designs.
The second novel factor was faith in breastmilk. Score on
the Faith in Breastmilk measure was identified as one of two
reliable, unique predictors of both fully and any breastfeeding
duration in this sample. The results for fully breastfeeding
showed that the probability of continuing to fully breastfeed
increased by 36% for every one point increase in faith in
breastmilk score (score range 1–5). The relationship between
this measure and continuing any breastfeeding was even more
strongly predictive, showing that women in the sample were
70% more likely to wean prematurely for each one-point
increase in this scale score. It is possible that this measure
may tap into infant feeding attitudes measured by instru-
ments such as the 17-item Iowa Infant Feeding Attitudes
Scale (De La Mora et al. 1999). However, the brevity, self-
report format and predictive ability of the Faith in Breastmilk
scale will increase its usefulness in clinical settings. Further
research to confirm the psychometric properties of this scale
is therefore warranted.
The second reliable unique predictor of both any and fully
breastfeeding identified was the woman’s planned breast-
feeding duration. As mentioned previously, this factor has
been confirmed as an important predictor of breastfeeding
duration in the past (e.g. Blyth et al. 2004, Forster et al.
2006). When compared with women who planned to
breastfeed for >6 months, women in the current study
who planned to breastfeed for 6 months or less were 1Æ72
times more likely to cease fully breastfeeding and 2Æ19 times
more likely to cease any breastfeeding before 6 months.
While it can be argued that these findings may indicate that
women in this group are merely exercising their right to set
and adhere to their personal goals for breastfeeding, the data
suggest that the majority of these women actually fell well
short of their stated breastfeeding goals. The mean planned
breastfeeding duration for this group was approximately
23 weeks; however, they actually breastfed for an average of
only 12 weeks (10 weeks fully), almost half the originally
planned duration for the group. This suggests that planning
to breastfeed for a relatively short time may be associated
with other risk factors, such as a lack of commitment to
breastfeeding goals.
M. O’Brien et al.
404 � 2008 The Authors. Journal compilation � 2008 Blackwell Publishing Ltd
Finally, breastfeeding self-efficacy, or the mother’s confi-
dence in her ability to breastfeed, was identified as a reliable
unique predictor of the duration of any (but not fully)
breastfeeding in this sample. Every one-point increase in
breastfeeding self-efficacy score (score range = 14–70) was
associated with a 5% increase in the participant’s likelihood
of continuing any breastfeeding to 6 months. However, in
previous research which included fewer psychological factors
and a 6-week outcome measure a 6% increase per score point
was found for both any and fully breastfeeding duration
(O’Brien & Fallon 2005). This raises questions as to why
fully breastfeeding was not uniquely predicted by breastfeed-
ing self-efficacy in the current study, while any breastfeeding
was. It seems likely that this differing result may be because
of the inclusion of additional psychological factors such as
optimism, anxiety and faith in breastmilk. These factors may
have influenced the mother’s decision to introduce supple-
mental foods to the infant prematurely and thus cease fully
(but not any) breastfeeding; however, further research in this
area is warranted.
Implications for practice and research
Our data reveal that breastfeeding rates (particularly at
6 months) remain well below national targets and interna-
tional breastfeeding guidelines (WHO 2001). This means that
further efforts to increase breastfeeding duration by breast-
feeding clinicians, researchers and governments are needed.
These efforts should include further research into the
incidence and effect of anxiety on postnatal women. Clini-
cians may also find it helpful to address the issue of anxiety
with breastfeeding women. Discussing the woman’s fears and
worries, providing education on the anxiety response, and
offering help with structured problem solving and relaxation
techniques may help to alleviate the symptoms of many
women, while those with more severe symptoms should be
referred to a suitable mental health professional (WHO
2000).
The results also revealed that several psychological factors
were statistically significantly associated with the duration of
both fully and any breastfeeding. As a set, these psychological
factors were more important to the prediction of breastfeed-
ing duration than were the socio-demographic factors. These
data suggest that women who possess the right psychological
make-up for breastfeeding may be capable of overcoming the
disadvantages posed by socio-demographic factors, a finding
that may give hope to many breastfeeding advocates strug-
gling in difficult socio-demographic environments.
Finally, the importance of faith in breastmilk and planned
breastfeeding duration over and above other potential risks to
breastfeeding suggests that an increased focus on these three
distinct psychological factors by both researchers and clini-
cians is warranted.
Conclusion
The results of this research confirm that psychological factors
are important predictors of breastfeeding duration. This
knowledge has implications for both clinical practice and
future research and represents an increase in our ability to
identify women at risk of early weaning and to construct
programmes capable of increasing the length of time for
which women breastfeed.
Acknowledgements
We would like to thank the management and staff of the
Toowoomba Health Service and St Vincent’s Private Hospi-
tal, Toowoomba for their generous assistance and support.
Thanks also to the study participants, who gave up their
valuable time to discuss and subsequently participate in this
What is already known about this topic
• Only a minority of women continue to breastfeed for
long enough to meet World Health Organization
guidelines for breastfeeding duration, which recom-
mend exclusive breastfeeding for the first six months.
• Several individual psychological differences appear to
influence the duration of breastfeeding, but little
exploratory research has been performed in this area.
What this paper adds
• Duration of breastfeeding is statistically significantly
associated with levels of optimism, breastfeeding self-
efficacy, faith in breastmilk, breastfeeding expecta-
tions, anxiety, planned duration of breastfeeding and
the time of the infant feeding decision.
• As a set, individual psychological differences are more
predictive of breastfeeding duration than the known
socio-demographic determinants.
• After controlling for socio-demographic characteris-
tics, measures of individual psychological factors can
uniquely predict the duration of both fully and any
breastfeeding.
• Knowledge of the factors influencing breastfeeding will
assist in identifying women at risk of early weaning
and in constructing programmes capable of increasing
the length of time for which women breastfeed.
JAN: ORIGINAL RESEARCH Influence of psychological factors
� 2008 The Authors. Journal compilation � 2008 Blackwell Publishing Ltd 405
project. We would also like to express our gratitude to Dr
Tony Fallon for the valuable early contributions he made to
this work.
Author contributions
MOB and DH were responsible for the study conception and
design. MOB performed the data collection. MOB and EB
performed the data analysis. MOB was responsible for the
drafting of the manuscript. MOB, EB and DH made critical
revisions to the paper for important intellectual content.
MOB and EB provided statistical expertise. EB and DH
supervised the study.
References
Arora S., McJunkin C., Wehrer J. & Kuhn P. (2005) Major factors
influencing breastfeeding rates: mother’s perception of father’s
attitude and milk supply. Pediatrics 106(5), 67–72.
Australian Bureau of Statistics (2003) 4810.0.55.001 Breastfeeding
in Australia, Electronic Delivery. Australian Bureau of Statistics,
Canberra.
Blyth R.J., Creedy D.K., Dennis C.L., Moyle C., Pratt J. & De Vries
S.M. (2002) Effect of maternal confidence on breastfeeding dura-
tion: an application of breastfeeding self-efficacy theory. Birth:
Issues in Perinatal Care 29(4), 278–284.
Blyth R.J., Creedy D.K., Dennis C.L., Moyle W., Pratt J., De Vries
S.M. & Healy G.N. (2004) Breastfeeding duration in an Australian
population: the influence of modifiable antenatal factors. Journal
of Human Lactation 20(1), 30–38.
Bolling K., Grant C., Hamlyn B. & Thornton A. (2007) Infant
Feeding Survey 2005. The Information Centre for Health and
Social Care and the UK Health Department, London.
Bottorff J.L. (1990) Persistence in breastfeeding: a phenomonological
investigation. Journal of Advanced Nursing 15(2), 201–209.
Boulton J. & Landers M. (1999) The Toowoomba Children’s
Nutrition Study. Queensland Health, Toowoomba.
Bourgoin G.L., Lahaie N.R., Rheaume B.A., Berger M.G., Dovigi
C.V., Picard L.M. & Sahai V.F. (1997) Factors influencing the
duration of breastfeeding in the Sudbury region. Canadian Journal
of Public Health 88(4), 238–241.
Briggs S.R. & Cheek J.M. (1986) The role of factor analysis in the
development and evaluation of personality scales. Journal of Per-
sonality 54, 106–148.
Brouwers E.P.M., van Baarb A.L. & Pop V.J.M. (2001) Does the
Edinburgh Postnatal Depression Scale measure anxiety? Journal of
Psychosomatic Research 51(5), 659–663.
Bulk-Bunschoten A.M.W., van Bodegom S., Reerink J.D., Pasker-de
Jong P.C.M. & De Groot C.J. (2001) Reluctance to continue
breastfeeding in The Netherlands. Acta Paediatrica Scandinavica
90(9), 1047–1053.
Callen J. & Pinelli J. (2004) Incidence and duration of breastfeeding
for term infants in Canada, Unites Stated, Europe and Australia: a
literature review. Birth 32(4), 285–292.
Clifford T.J., Campbell M.K., Speechley K.N. & Gorodzinsky F.
(2006) Factors influencing full breastfeeding in a southwestern
Ontario community: assessments at 1 week and at 6 months
postpartum. Journal of Human Lactation 23(3), 292–304.
Cooke M., Schmied V. & Sheehan A. (2007) An exploration of the
relationship between postnatal distress and maternal role attain-
ment, breast feeding problems and breast feeding cessation in
Australia. Midwifery 23(1), 66–76.
Cox J.L. & Sagovsky R. (1987) Detection of postnatal depression:
development of the 10-item Edinburgh Postnatal Depression Scale.
British Journal of Psychiatry 150, 782–786.
Creedy D.K., Dennis C.L., Blyth R.J., Moyle R., Pratt J. & De Vries
S.M. (2003) Psychometric characteristics of the Breastfeeding Self-
Efficacy Scale: data from an Australian sample. Research in
Nursing and Health 26(2), 143–152.
Damato E.G. (2005) Duration of breastfeeding for mothers of twins.
Journal of Obstetric, Gynecologic, & Neonatal Nursing 34(2),
201–209.
De La Mora A., Russell D.W., Dungy C.I., Losch M. & Dusdieker L.
(1999) The Iowa Infant Feeding Attitude Scale: analysis of reli-
ability and validity. Journal of Applied Social Psychology 29(11),
2362–2380.
Dennis C.L. (2003) The Breastfeeding Self-Efficacy Scale: psycho-
metric assessment of the short form. Journal of Obstetric, Gyne-
cologic, & Neonatal Nursing 32(6), 734–744.
Dennis C.L. & Faux S. (1999) Development and psychometric testing
of the Breastfeeding Self-Efficacy Scale. Research in Nursing &
Health 22(5), 399–409.
Denzin N.K. & Lincoln Y.S. (2005) The Sage Handbook of Quali-
tative Research. Sage Publications, London.
Donath S.M., Amir L.H. & ALSPAC Study Team (2004) The rela-
tionship between maternal smoking and breastfeeding duration
after adjustment for maternal infant feeding intention. Acta Pae-
diatrica 93(11), 1514–1518.
Dunn S., Davies B., McCleary L., Edwards N. & Gaboury I. (2006)
The relationship between vulnerability factors and breastfeeding
outcome. Journal of Obstetric, Gynecologic, & Neonatal Nursing
35(1), 87–97.
Fitzpatrick C.C., Fitzpatrick P.E. & Darling M.R.N. (1994) Factors
associated with the decision to breastfeed among Irish women.
Irish Medical Journal 87(5), 145–146.
Forster D., McLachlan H. & Lumley J. (2006) Factors associated
with breastfeeding at 6 months postpartum in a group of Austra-
lian women. International Breastfeeding Journal 1(18), doi:
10.1186/1746-4358-1-18.
Gibaud-Wallston J. (1977) Self-esteem and situational stress: factors
related to sense of competence in parents. Unpublished PhD
dissertation. George Peabody College for Teachers, Rhode Island.
Goldberg L.R. (1999) A broad-bandwidth, public-domain, person-
ality inventory measuring the lower-level facets of several
five-factor models. In Personality Psychology in Europe, Vol. 7
(Mervielde I., Deary I., De Fruyt F. & Ostendord F., eds), Tilburg
University Press, Tilburg, The Netherlands, pp. 7–28.
Goldberg L.R., Johnson J.A., Eber H.W., Hogan R., Ashton M.C.,
Cloninger C.R. & Gough H.G. (2006) The International Person-
ality Item Pool and the future of public-domain personality
measures. Journal of Research in Personality 40, 84–96.
Hatton D.C., Harrison-Hohner J., Coste S., Dorato V., Curet L.B. &
McCarron D.A. (2005) Symptoms of postpartum depression and
breastfeeding. Journal of Human Lactation 21(4), 444–449.
M. O’Brien et al.
406 � 2008 The Authors. Journal compilation � 2008 Blackwell Publishing Ltd
Hegney D., Fallon T., O’Brien M.L., Laurent K., Baker S., Plank A.
& Doolan J. (2003) The Toowoomba Infant Feeding Support
Service Project: Report on Phase 1 – A Longitudinal Needs Anal-
ysis of Breastfeeding Behaviours and Supports in the Toowoomba
Region. Centre for Rural and Remote Area Health, University of
Southern Queensland, Toowoomba.
Hegney D., Fallon A.B. & O’Brien M.L. (2007) Against all odds: a
retrospective case-controlled study of women who experienced
extraordinary breastfeeding problems. Journal of Clinical Nursing
17, 1182–1192.
Henderson J.J., Evans S.F., Straton J.A.Y., Priest S.R. & Hagan R.
(2003) Impact of postnatal depression on breastfeeding duration.
Birth: Issues in Perinatal Care 30(3), 175–180.
Hewat R.J. & Ellis D.J. (1986) Similarities and differences between
women who breastfeed for short and long durations. Midwifery
2(1), 37–43.
Hong S.-M. & Faedda S. (1996) Refinement of the Hong Psycho-
logical Reactance Scale. Educational and Psychological Measure-
ment 56(1), 173–183.
Horta B.L., Bahl R., Martines J.C. & Victora C.G. (2007) Evidence
on the Long-Term Effects of Breastfeeding: Systematic Reviews
and Meta-Analyses. World Health Organisation, Geneva.
Ip S., Chung M., Raman G., Chew P., Magula N., DeVine D.,
Trikalinos T. & Lau J. (2007) Breastfeeding and Maternal and
Infant Health Outcomes in Developed Countries. Evidence
Report/Technology Assessment No. 153 (Prepared by Tufts-New
England Medical Center Evidence-based Practice Center). AHRQ
Publication No. 07-E007. Agency for Healthcare Research and
Quality, Rockville, MD.
Kieffer E.C., Novotny R., Welch K.B., Mor J.M. & Thiele M. (1997)
Health practitioners should consider parity when counselling
mothers on decisions about feeding methods. Journal of the
American Dietetic Association 97(11), 1313–1316.
Kirkland V.L. & Fein S.B. (2003) Characterising reasons for
breastfeeding cessation throughout the first year postpartum
using the construct of thriving. Journal of Human Lactation
19(3), 278–285.
Kronborg H. & Vaeth M. (2004) The influence of psychosocial
factors on the duration of breastfeeding. Scandinavian Journal of
Public Health 32(3), 210–217.
Lovibond S.H. & Lovibond P.F. (1995) Manual for the Depression
Anxiety Stress Scales. Psychology Foundation Monograph, Sydney.
McLearn K.T. (2006) Maternal depressive symptoms at 2 to
4 months post partum and early parenting practices. Archives of
Pediatrics & Adolescent Medicine 160(3), 279–284.
Mozingo J.N., Davis M.W., Droppleman P.G. & Merideth A. (2000)
‘It wasn’t working’: women’s experiences with short term breast-
feeding. American Journal of Maternal Child Nursing 25(3), 120–
126.
Nagy E., Orvos H., Pal A., Kovacs L. & Loveland K. (2001)
Breastfeeding duration and previous breastfeeding experience. Acta
Paediatrica 90(1), 51–56.
NHMRC (2007) Australian code for the responsible conduct of re-
search. In Revision of the Joint NHMRC/AVCC Statement and
Guidelines on Research Practice. National Health & Medical
Research Council and the Australian Research Council, Canberra.
O’Brien M.L. & Fallon A.B. (2005) The effect of breastfeeding self-
efficacy on breastfeeding duration. Birth Issues 14(4), 135–142.
Papinczak T.A. & Turner C.T. (2000) An analysis of personal and
social factors influencing initiation and duration of breastfeeding
in a large Queensland maternity hospital. Breastfeeding Review
8(1), 25–33.
Peduzzi P., Concato J., Feinstein A.R. & Holford T.R. (1995)
Importance of events per independent variable in proportional
hazards regression analysis II. Accuracy and precision of regres-
sion estimates. Journal of Clinical Epidemiology 48(12), 1503–
1510.
Rentschler D.D. (1991) Correlates of successful breastfeeding. Image
– the Journal of Nursing Scholarship 23(3), 151–154.
Rosenberg M. (1989) Society and the Adolescent Self-Image. Wes-
leyan University Press, Middletown, CT.
Scheier M.F. & Carver C.S. (1992) Effects of optimism on psycho-
logical and physical well-being: theoretical overview and
empirical update. Cognitive Therapy and Research 16(2), 201–
228.
Scheier M.F., Carver C.S. & Bridges M.W. (1994) Distinguishing
optimism from neutocism (and trait anxiety, self-mastery, and self-
esteem): a reevaluation of the Life Orientation Test. Journal of
Personality and Social Psychology 67(6), 1063–1078.
Scott J.A. (2006) Predictors of breastfeeding duration: evidence from
a cohort study. Pediatrics 117(4), e646–e655.
Scott J.A., Landers M.C.G., Hughes R.M. & Binns C.W. (2001a)
Factors associated with breastfeeding at discharge and duration of
breastfeeding amongst two populations of Australian women.
Journal of Paediatrics & Child Health 37(3), 254–262.
Scott J.A., Landers M.C.G., Hughes R.M. & Binns C.W. (2001b)
Psychosocial factors associated with the abandonment of breast-
feeding prior to hospital discharge. Journal of Human Lactation
17(1), 24–30.
Shields P.M. & Tajalli H. (2006) Intermediate theory: the missing
link in successful student scholarship. Journal of Public Affairs
Education 12(3), 313–334.
Spence J.T., Helmreich R.L. & Pred R.S. (1987) Impatience versus
achievement strivings in the type A pattern: differential effects on
students’ health and academic achievement. Journal of Applied
Psychology 72(4), 522–528.
SPSSSPSS (2005) Statistical Package for the Social Sciences (for Windows),
Version 14. SPSS Inc., Chicago, IL.
Stamp G.E. & Casanova H.T. (2006) A breastfeeding study in a rural
population in South Australia. Rural and Remote Health 6, 495.
Tabachnick B.G. & Fidell L.S. (2001) Using Multivariate Statistics.
Allyn and Bacon, Sydney.
Tarkka M.T., Paunonen M. & Laippala P. (1999) Factors related
to successful breast feeding by first-time mothers when the
child is 3 months old. Journal of Advanced Nursing 29(1), 113–
118.
Ueda T., Yokoyama Y., Irahara M. & Aono T. (1994) Influence of
psychological stress on suckling induced pulsatile oxytocin release.
Obstetrics & Gynecology 84(2), 259–262.
Vandiver T.A. (1997) Relationship of mothers’ perceptions and
behaviors to the duration of breastfeeding. Psychological Reports
80(3), Part–2.
Wagner C.L., Wagner M.T., Ebeling M., KatChatman K.G., Cohen
M. & Hulsey T.C. (2006) The role of personality and other factors
in a mother’s decision to initiate breastfeeding. Journal of Human
Lactation, 22(1), 16–26.
JAN: ORIGINAL RESEARCH Influence of psychological factors
� 2008 The Authors. Journal compilation � 2008 Blackwell Publishing Ltd 407
Webb K., Marks G.C., Lund-Adams M., Rutishauser I.H. &
Abraham B. (2001) Towards a National System for Monitoring
Breastfeeding in Australia: Recommendations for Population
Indicators, Definitions and Next Steps. Australian Food and
Nutrition Monitoring Unit, Canberra.
Whaley S.E., Meehan K., Lange L., Slusser W. & Jenks E. (2002)
Predictors of breastfeeding duration for employees of the Special
Supplemental Nutrition Program for Women, Infants, and Chil-
dren (WIC). Journal of the American Dietetic Association 102(9),
1290–1293.
Whelan A. & Lupton P. (1998) Promoting successful breast feeding
among women with a low income. Midwifery 14(2), 94–104.
WHO (2000) Management of Mental Disorders. WHO, Burwood,
Victoria.
WHO (2001) The Optimal Duration of Exclusive Breastfeeding.
Note for the press No. 7, April 2. WHO, Geneva.
M. O’Brien et al.
408 � 2008 The Authors. Journal compilation � 2008 Blackwell Publishing Ltd
Top Related