fpsyg-08-01361
August 7, 2017
Time: 15:7
# 3
Rozenblat et al.
5-HTTLPR and Disordered Eating
a practical perspective, this also allows for the collection of far
larger samples compared to studies using case-control designs,
which is a key consideration in genetic association research
(
Duncan and Keller, 2011
).
To further test preliminary findings from
Rozenblat et al.
(2017)
in a homogenous sample, the present study used data
from 650 participants who provided DNA in the Australian
Temperament Project (ATP), a population based cohort study
that has followed a representative sample of around 2000
participants from infancy to adulthood since 1983. The first aim
was to assess the direct effects of depressed mood, emotional
control, sexual abuse, and parental physical punishment on
adolescent drive for thinness and bulimic behaviors. The second
aim was to examine the extent to which the relationships
between these factors and eating pathology were moderated by
5-HTTLPR. Results of this study constitute an important step
toward accumulating evidence regarding whether genetic factors
may moderate the influence of psychological and environmental
risk factors on EDs.
MATERIALS AND METHODS
Participants
Australian Temperament Project participants were initially
recruited in infancy (4–8 months) in Victoria in 1983, using
a stratified random sampling framework, via maternal and
child health centers in urban and rural locations. The first
survey included 2,443 infants (48.0% female), with 16 surveys
completed to date. The present study involved a sub-set of 650
participants (50.2% female) who had completed the 11th survey
at age 15–16 years, providing information on drive for thinness
and bulimic behaviors, depressive symptoms and emotional
control, and who also provided a saliva sample for genotyping
at this time (
N = 567), or in their early 30s (N = 83, in 2015
as part of a separate sub-study). Of the 650 participants, 467
participants also provided retrospective information regarding
sexual abuse and parental physical punishment in the 14th survey
at age 23–24, and were included in the respective analyses. To
avoid issues related to genetic heterogeneity, 23 participants who
self-identified as non-Caucasian were excluded prior to forming
the sample. The final analysable sample had a higher proportion
of participants from the highest SES quartile than the original
sample (for full socio-demographic information, see Table 1).
Due to missing data, the final sample comprised 643 participants
for the depression analyses and 649 for the emotional control
analyses. Parents and adolescents provided written informed
consent for each survey wave and for the collection of saliva
samples. The data collection was approved by the Australian
Institute of Family Studies Ethics Review and carried out
in accordance with the latest version of the Declaration of
Helsinki.
Measures
Disordered Eating
Drive for thinness and bulimic behaviors were assessed at age
15–16 via the Eating Disorder Inventory-2 (EDI-2;
Garner, 1991
)
Drive for Thinness and Bulimia scales. The Drive for Thinness
scale consists of 7-items measuring participants’ desire to lose
weight or fear of weight gain (e.g., “I am preoccupied with the
desire to be thinner”). Internal consistency in the current sample
was
α = 0.92. The Bulimia scale consists of 7 items measuring
bulimic behaviors, including binging and purging (e.g., “I stuff
myself with food”), with Cronbach’s
α = 0.74 in the current
sample. For details of scoring and some minor modifications
made for an Australian context, refer to
Krug et al. (2016)
.
Psychological Stress Exposures
Depression was assessed at the same time-point as disordered
eating via the Short Mood and Feelings Questionnaire (SMFQ)
(
Angold et al., 1995
), a 13-item subscale derived from the original
33-item questionnaire. The SMFQ is intended as a screening
measure for children and adolescents that queries depressive
symptoms according to DSM-III criteria (
American Psychiatric
Association [APA], 1980
; e.g., “I feel miserable or unhappy”), with
responses provided on 3-point scale (
rarely/never, sometimes,
often/always). Participants with missing data on five or more
items were excluded from analyses (
α = 0.83 in current sample).
Emotional control measured participants’ capacity to control
their emotions and was also assessed at age 15–16 using an
ATP-devised measure consisting of 10-items (e.g., “I am able to
keep my feelings under control” and “I am able to calm down
if I am feeling nervous”) rated on a 6-point scale from
never
to
always. This measure has been previously used in studies
examining internalizing problems (
Toumbourou et al., 2011
),
with
α = 0.70 in the present sample.
Sexual and Physical Stress Exposures
(Retrospective)
A number of retrospective indicators were used at age 23–24
to assess sexual abuse and parental physical punishment during
childhood and adolescence. Sexual abuse was based on a ‘yes’
response to the questions: “You had a sexual experience with a
person who was not a family member prior to 16” and a follow
up ‘no’ response to the question “Was this consensual?”, or, a ‘yes’
response to the question “A family member did, or tried to do,
sexual things to you.”
Mild-to-moderate parental physical punishment was based
on a ‘yes’ response to the question “Your parent/s used harsh
physical treatment (e.g., smacking, hitting) to discipline you,” and
severe parental physical punishment was based on an additional
‘yes’ response to a follow up question, “Did you ever suffer effects
that lasted to the next day or longer (e.g., bruising, marking, pain,
soreness)?”, creating two distinct severity categories.
5-HTTLPR Genotyping (Moderation Variable)
Following the 11th survey, DNA for 567 participants was isolated
using Qiagen QIAamp kits from buccal epithelial cells via cotton
swabs, with further details described in
Jorm et al. (2000)
. Saliva
samples for an additional 83 participants were collected following
the 16th survey in 2015 using Oragene saliva pots or tubes and
analyzed at the Australian Genomics Research Facility (AGRF),
Adelaide, SA, Australia. Genotype frequencies were similar in the
original and 2015 samples. For all samples,
5-HTTLPR genotype
Frontiers in Psychology | www.frontiersin.org
3
August 2017 | Volume 8 | Article 1361