Breastfeeding and offspring’s compassion and empathy in adulthood: No association

Breastfeeding and offspring’s compassion and empathy in adulthood: A study with an over 30‐year follow‐up. Aino I. L. Saarinen et al. Scandinavian Journal of Psychology, December 10 2019. https://doi.org/10.1111/sjop.12600

Abstract: This study investigated whether breastfeeding predicts offspring’s dispositional compassion and empathy from early adulthood to middle age. The parents of the participants (N = 1,394) of the Young Finns study answered questions about breastfeeding in 1983, and the participants’ compassion and empathy were evaluated in 1997‒2012 (participants were aged 20‒50 years). Breastfeeding did not predict the course of compassion or empathy in adulthood at the age of 20‒50 years. The associations remained non‐significant, when adjusted for age, gender, socioeconomic factors, and a wide range of characteristics of the family environment (including mother’s gestational age; premature birth; birth weight; number of other children at home; parental mental disorder; parental relationship status; parental postnatal smoking; parental postnatal alcohol use; parenting behavior; and child’s externalizing behavior). In conclusion, breastfeeding seems not to predict offspring’s compassion or empathy in adulthood. The findings may present a hopeful perspective for children growing up with non‐breastfeeding caregivers.

DISCUSSION
To the best of our knowledge, this is the first study to
investigate whether breastfeeding predicts offspring’s compassion
or empathy in adulthood. Our results revealed that neither having
received breastfeeding, nor the duration of breastfeeding
predicted the course of compassion or empathy from ages 20 to
50, that is from early adulthood to middle age. The associations
of breastfeeding with compassion and empathy remained nonsignificant
over the whole follow-up, when adjusted for age and
gender, when adjusted also for socioeconomic factors in
childhood and adulthood, and when additionally adjusted for a
wider range of characteristics of the child, home environment,
and parenting (i.e., mother’s gestational age; premature birth;
birth weight; number of other children at home; parental mental
disorder; parental relationship status; parental postnatal smoking;
parental postnatal alcohol use; parenting behavior; and child’s
externalizing behavior). Moreover, the effect of breastfeeding on
compassion and empathy remained non-significant also in the
multiple imputed dataset, indicating that the non-significant
associations were not accounted for by some attrition biases over
the follow-up. Taken together, our study provided evidence that
breastfeeding does not predict the development of dispositional
compassion or empathy of the offspring in adulthood.
It has been suggested that the previous null results regarding
breastfeeding and socioemotional outcomes might be at least
partly accounted for by a limited range of control variables
(Jansen, de Weerth & Riksen-Walraven, 2008). That is, there
might be some selection bias between breastfeeding and nonbreastfeeding
mothers with regard to other qualities. For
example, non-breastfeeding mothers may be more career-oriented
(Stewart-Knox, Gardiner & Wright, 2003) and more likely able
to provide a favorable home environment with regard to
socioeconomic factors. In the present study, however, we
controlled for a relatively comprehensive range of the
characteristics of the child and home environment. Nevertheless,
all the associations of breastfeeding with offspring’s compassion
or empathy remained non-significant. Hence, this implies that
breastfeeding may not predict offspring’s compassion or
empathy in adulthood indirectly via other qualities of the family
environment.
Another explanation for the previous results, which have
found no link for breastfeeding to offspring’s socioemotional
development in childhood, might potentially be that compassion
may not be fully developed yet in early childhood. Several
studies have investigated socioemotional outcomes among
children aged as young as 1–3.5 years (e.g., Borra et al., 2012;
Oddy et al., 2011). It has been suggested that there might be
cognitive requirements for being able to feel compassion: for
example one must be able to imagine himself or herself to the
same situation and to imagine how it would feel like (Cassell,
2002). Our study demonstrated that there exist no association of
breastfeeding with compassion after childhood and cognitive
maturation.
Compassion and empathy are generally regarded as
cornerstones of one’s socioemotional development. Deficits in
compassion or empathy constitute a central feature of a range of
psychiatric disorders, for example, antisocial personality disorder
and paranoia (APA, 2013). Hence, our results, which showed no
association of breastfeeding with empathy or compassion,
tentatively suggest that non-breastfeeding does not predispose the
child to socioemotional deficits or to psychiatric disorders through
them. This is in line with a previous study demonstrating that
breastfeeding is not linked with violent behavior in adulthood
(Caicedo, Gonc alves, Gonz alez & Victora, 2010).
In our study, there were some limitations that are necessary to
be taken into consideration. Breastfeeding was evaluated by
asking parents whether the child had received breastfeeding and
for how many months breastfeeding had occurred. It was not
assessed, however, whether the child had been exclusively or
non-exclusively breastfed, that is whether the child had received
also complementary food or other nutritions during breastfeeding.
However, breastfeeding has been commonly assessed this way in
previous studies (e.g., Belfort et al., 2016; Narvaez et al., 2013).
Moreover, such studies that have differentiated between exclusive
and non-exclusive breastfeeding have found no differences
between the effects of exclusive vs. non-exclusive breastfeeding
on socioemotional outcomes in childhood (e.g., Kramer et al.,
2011; Lind et al., 2014).
Moreover, breastfeeding was evaluated retrospectively.
Previously, the length of recall period is found to positively
correlate with recall bias (Horta et al., 2013). In this study, we
aimed to minimize potential recall bias by asking parents about
breastfeeding in 1983 (i.e., in the first follow-up measurement).
Further, it has been highlighted that the duration of breastfeeding
should be checked from many sources if possible (Horta et al.,
2013). We advised the parents to check this information from the
child’s personal record cards obtained from well-baby clinics.
Finally, there is evidence that high parental socioeconomic status
is related to underestimation of the duration of breastfeeding
(Huttly, Barros, Victora, Beria & Vaughan, 1990). In this study,
we controlled for a variety of childhood covariates, including
parental socioeconomic status.
Finally, the rate of non-breastfed children was comparatively
low in the present study (7.5%). Also, previous studies have
found that the rate of non-breastfed children varies between 0.5
and 26% in Europe (Clements et al., 1997; H€ornell, Aarts,
Kylberg, Hofvander & Gebre-Medhin, 1999; Michaelsen et al.,
1994). Hence, no firm conclusions can be made about the effects
of non-breastfeeding on the course of compassion and empathy.
Nevertheless, there was a higher frequency of participants who
had been breastfed over a comparatively short time period (e.g.,
4 months) and they did not differ from others in their compassion
or empathy levels. Overall, previous studies have suggested that
the potential association of breastfeeding with compassion and
empathy might go indirectly via close and supportive parent-child
relationship (Bystrova et al., 2009; Cernadas et al., 2003) and
secure attachment (Britton et al., 2006). In the light of previous
evidence, that kind of mediating socioemotional factors are
unlikely to develop over a short period of breastfeeding.
The present study had a variety of substantial strengths. First, we
had a prospective follow-up of over 30 years and a comparatively
large population-based sample consisting of six different age
cohorts. Second, our data enabled us to investigate the course of
dispositional compassion and empathy from ages 20 to 50, that is,
from early adulthood to middle age. Third, we used multilevel
models for repeated measurements that have comparatively strong
statistical power (Gelman & Hill, 2006), so that also modest
associations between the study variables could be obtained. Fourth,
we replicated all the analyses in a multiple imputed dataset in order
to ensure that the results were not accounted for by certain attrition
biases in our sample. Finally, we could control for a range of
potential confounding variables, such as age, gender,
socioeconomic factors in childhood and adulthood, and also a
range of characteristics of the child and family environment. Taken
together, our data provided exceptional possibilities to investigate
whether breastfeeding predicts the course of compassion and
empathy of the offspring in adulthood.
Suboptimal breastfeeding is common not only in developing
countries (Huffman et al., 2001; M€uller & Krawinkel, 2005) but
also in Western countries. In the United States, for example, only
about one third of mothers continue breastfeeding the infant for
the first 6 months (McDowell, Wang & Kennedy-Stephenson,
2008), although the WHO has recommended exclusive
breastfeeding for the first 6 months of life (WHO & UNICEF,
2003). There are a variety of reasons for suboptimal
breastfeeding, including experiences of restricted freedom,
conflicts between motherhood and pursuing a career, sexualityrelated
feelings, and perceived social isolation (Stewart-Knox
et al., 2003; Van Esterik, 2002). Importantly, many mothers are
also forced to introduce formula-feeding because of hormonal or
nutritional reasons (Goldman, Hopkinson & Rassin, 2007; Jansen
et al., 2008). Still, however, breastfeeding is considered even as
an aspect of morality or a measure of the quality of motherhood
in some populations (Lee, 2007). Consequently, a review
concluded that mothers who introduce formula-feeding frequently
experience feelings of guilt, uncertainty and failure (Lakshman,
Ogilvie & Ong, 2009). In many cases, formula-feeding mothers
have to defend their feeding methods when visiting child care
services, which may lower their trust toward health-care
professionals (Lee, 2007).
In the context of child health-care services, it is necessary to
clarify the precise arguments for breastfeeding for mothers. There
is a great amount of evidence that breastfeeding is linked with
better somatic health of the child, including properties from blood
pressure to respiratory diseases and allergies (e.g., Friedman &
Zeiger, 2005; Owen et al., 2002; Wold & Adlerberth, 2002). On
the other hand, previous meta-analyses have concluded that
breastfeeding does not predict higher intelligence or cognitive
abilities of the offspring (e.g., Horta, Loret de Mola & Victora,
2015; Walfisch, Sermer, Cressman & Koren, 2013). Moreover,
most previous studies have found no evidence for a link for
breastfeeding to prosocial behavior, sociability, or emotional
difficulties in childhood (e.g., Kramer et al., 2008; Lind et al.,
2014; Oddy et al., 2011; Tanaka et al., 2009). In line with these
findings, the present study showed that breastfeeding does not
predict compassion or empathy of the offspring in adulthood. Our
findings may present a hopeful and encouraging perspective for
children growing up with non-breastfeeding caregivers.
This study was supported financially by the Academy of Finland (M.H.,
grant numbers 308676 and 258578); Signe and Ane Gyllenberg
Foundation (M.H.); the Jenny and Antti Wiguri Foundation (L.P.-R.); and
Mannerheim League for Child Welfare’s Research Foundation and Finnish
Cultural Foundation (E.O.). The Young Finns Study has been financially
supported by the Academy of Finland: Grants 286284, 134309 (Eye),
126925, 121584, 124282, 129378 (Salve), 117797 (Gendi), and 41071
(Skidi); the Social Insurance Institution of Finland; Competitive State
Research Financing of the Expert Responsibility area of Kuopio, Tampere
and Turku University Hospitals (grant X51001); the Juho Vainio
Foundation; the Sigrid Juselius Foundation; the Yrj€o Jahnsson Foundation;
the Paavo Nurmi Foundation; the Finnish Foundation of Cardiovascular
Research and Finnish Cultural Foundation; the Tampere Tuberculosis
Foundation; the Emil Aaltonen Foundation; and Diabetes Research
Foundation of Finnish Diabetes Association. The funding source had no
role in study design, data collection, data analysis, data interpretation,
writing of the report, or in the decision to submit the article for publication.