Internalising symptoms and body dissatisfaction:
Untangling temporal precedence using cross-lagged models in two cohorts
Helen Sharpe, Praveetha Patalay, Miranda Wolpert
Evidence Based Practice Unit, University College London and the Anna Freud Centre
Corresponding author: Dr Helen Sharpe, 21, Maresfield Gardens, London, NW3 5SU, UK, +44 (0) 207 443 2263, h.sharpe@ucl.ac.uk
BACKGROUND
YOUNGER BOYS
• Cross sectional studies demonstrate that body dissatisfaction and
internalising symptoms have a tendency to co-occur and are both overrepresented in girls compared to boys.1,2
• However, it is not clear whether body dissatisfaction typically precedes
internalising symptoms or vice versa.3,4
• Existing literature provides theoretical and empirical support for both
possibilities, but is limited in two ways:
.65***
Internalising
symptoms:
age 8-9
Body
dissatisfaction:
age 8-9
.38***
Internalising
symptoms:
age 11-12
• Cross lagged models were estimated in MPlus 7 using weighted least squares
means and variance adjusted estimation (WLSMV).
• Imputation for missing values was based on full information maximum
likelihood (FIML). Complete case analysis produced similar results.
• Items of the internalising scale were included in the model to derive the latent
internalising factor score at each time-point.
Internalising
symptoms:
age 12-13
.25***
.09**
Body
dissatisfaction:
age 10-11
-.04NS
Body
dissatisfaction:
age 8-9
Body
dissatisfaction:
age 9-10
.41***
n = 2691, RMSEA = .043 (.042 .045), CFI = .91, TLI = .91, χ2 (525) = 3175.73
.69***
Internalising
symptoms:
age 13-14
.04
.36***
Body
dissatisfaction:
age 11-12
.48***
.51***
Internalising
symptoms:
age 12-13
.22***
NS
n = 2757, RMSEA = .038 (.036, .039), CFI = .91, TLI = .91, χ2 (525)=2576.93
*** p < 0.001, ** p < 0.01
Internalising
symptoms:
age 13-14
.06NS
.31***
.10***
.08**
Body
dissatisfaction:
age 13-14
.71***
.34***
-.001NS
Body
dissatisfaction:
age 12 -13
.63***
Internalising
symptoms:
age 11-12
.14***
-.02NS
.47***
Body
dissatisfaction:
age 10-11
OLDER GIRLS
.30***
• OLDER COHORT: 5,981 adolescents aged 11-12 years at baseline
ANALYSIS
.63***
.13***
-.02NS
OLDER BOYS
• YOUNGER COHORT: 5,485 children aged 8-9 years at baseline
• Internalising symptoms: Me and My School Questionnaire - Emotional
Difficulties Subscale (10 items, e.g., “I worry a lot”) 5
• Body dissatisfaction: “I like the way I look” (Never/Sometimes/Always)
.33***
Internalising
symptoms:
age 10-11
.28***
n = 2794, RMSEA = .042 (.041, .043), CFI = .90, TLI = .90, χ2 (525) = 3106.05
• Two longitudinal cohorts in England followed annually over three years
MEASURES
.44***
.74***
Internalising
symptoms:
age 9-10
.24***
-.01NS
Body
dissatisfaction:
age 9-10
.61***
.08**
.12***
NS
YOUNGER GIRLS
Internalising
symptoms:
age 8-9
.28***
-.03
DESIGN
»» Drawn from 37 state secondary schools
»» 54% female
»» 78% White, 12% Asian, 6% Black, 4% Other/Unclassified
»» 19% socio-economically deprived
Internalising
symptoms:
age 10-11
.30***
METHODS
»» Drawn from 138 state primary schools
»» 49% female
»» 73% White, 15% Asian, 5% Black, 7% Other/Unclassified
»» 23% socio-economically deprived
.74***
Internalising
symptoms:
age 9-10
.33***
.15***
»»No study has simultaneously tested the two temporal hypotheses within the same model.
»»The studies focus almost exclusively on early adolescents resulting in little being known about development from pre-adolescence and across puberty.
PARTICIPANTS
RESULTS
Body
dissatisfaction:
age 11-12
.38***
Body
dissatisfaction:
age 12 -13
.59***
Body
dissatisfaction:
age 13-14
n = 3224, RMSEA = .039 (.038, .040), CFI = .93, TLI = .92, χ2 (525)=3086.83
RMSEA = Root Mean Square Error of Approximation; CFI = Comparative Fix Index, TLI = Tucker-Lewis Index
CONCLUSIONS
• Body dissatisfaction in those aged 8 - 10 is driven by internalising
symptoms, both in boys and girls.
• From age 11, internalising symptoms continue to drive body
dissatisfaction in boys, whereas, in girls body dissatisfaction becomes the
more salient driver of later internalising symptoms.
• This finding may reflect gender-specific risk profiles as young people
transition through puberty.
• Preventative interventions would benefit from adopting a developmentally
sensitive approach that takes into account gender differences in risk
pathways.
REFERENCES
1.Ohring et al. (2002). Girls’ recurrent and concurrent body dissatisfaction: Correlates and
consequences over 8 years. International Journal of Eating Disorders, 31(4), 404-415.
2.Neumark-Sztainer et al. (2002). Ethnic/racial differences in weight-related concerns
and behaviors among adolescent girls and boys: Findings from Project EAT. Journal of
Psychosomatic Research, 53(5), 963-974.
3.Paxton et al. (2006). Body dissatisfaction prospectively predicts depressive mood and
low self-esteem in adolescent girls and boys. Journal of Clinical Child and Adolescent
Psychology, 35(4), 539-549.
4.Quick et al. (2013). Prospective predictors of body dissatisfaction in young adults 10year longitudinal findings. Emerging Adulthood, 1(4), 271-282.
5.Deighton et al. (2013) The development of a school-based measure of child mental
health. Journal of Psychoeducational Assessment, 31(3), 247-257