Decision Making in Childhood and Eating Disorder Symptoms in Adolescence

This research, funded by the Medical Research Foundation and the Medical Research Council uses longitudinal data from the Millennium Cohort Study to investigate how decision making skills in childhood is associated with eating disorder symptoms in adolescence

 

Project Aims

Cross-sectional studies which look at a group of people affected by an illness at one point in time have shown us that decision making might be important for understanding eating disorders. This project aims to be the first ever to use longitudinal data from the Millennium Cohort Study to understand how decision making, measured in childhood and adolesence at ages 11 and 14, might be related to the presence of eating disorder symptoms in adolescence at ages 14 and 17.

This work is important because understanding the factors involved in the early stages of eating disorder symptoms will enable us to develop strategies to help young people to develop resilience to these serious mental illnesses.

We were one of seven recipients of £2.7 million of Medical Research Foundation and the Medical Research Council funding for new research into eating disorders and self-harm.

We began this project in March 2019 and completed it in July 2021. You can watch my talk where I discuss the overall findings of the work here.

 

Our Study

Study Participants

  • We’ve modelled data from 11.303 boys and girls from all four UK countries who were born in 2000-2001. These young people have been followed up on 7 occasions so far.

Measures

  • We’ve looked at their scores on the Cambridge Gambling Task at ages 11 and 14. This task measures decision making under conditions of risk. We examined five outcome measures from this task:

    1) Risk taking: the mean proportion of the current points total that the participant choses to risk on trials when the most probable colour was selected;

    2) Quality of decision-making: the mean proportion of trials where the most likely colour outcome was selected;

    3) Deliberation-time: the mean time taken to decide which colour of box is hiding the token;

    4) Risk-adjustment: the extent to which, on trials where a larger proportion of boxes are a certain colour, participants bet a higher proportion of their points;

    5) Delay-aversion: the time participants are prepared to wait in order to place a higher or lower bet.

  • We looked at the presence of eating disorder symptoms at ages 14 and 17. We’ve measured this through items assessing body dissatisfaction, the intention to lose weight, restriction of nutritional intake and the use of driven exercise to influence weight/shape and weight itself.

Data Analysis

  • We adjusted our models for key confounders associated with risk and exposure, such as gender, ethnicity, family poverty, IQ (measured at age 5 from three subscales of the British Ability Scales), pubertal status at age 11 (breast growth or menstruation or hair on body for females, and voice change or facial hair or hair on body for males), internalising and externalising symptoms at age 11 measured using the mother-rated Strengths and Difficulties Questionnaire and objectively measured exercise activity measured using accelerometers.

Progress so Far

  • We planned three publications from this work, but the data were so rich and informative that we’ve written six! We will link to the published data as the study progresses.

  • We’ve set up this website and will be hosting online events to share and discuss this information with all stakeholders interested in the topic of eating disorders

Interested in being part of our online events on this topic? Please get in touch and we’ll add you to the mailing list. We’ll only send you information about these events and nothing else.

Key Findings

1) Decision making in childhood predicts eating disorder symptoms in adolescence.

Those with better quality decision making were 34% less likely to show an intention to lose weight (b=-0.40, OR=0.66, p<0.05) and 34% less likely to be overweight (b=-0.41, RRR=0.66, p<0.05). Those with higher risk taking were 58% more likely to report dietary restriction (b=0.45, OR=1.58, p<0.05) and 46% more likely to report excessive exercise (b=0.38, OR=1.46, p<0.05).

In the complete cases sample, higher risk adjustment scores were associated with a 47% increased risk of underweight (b=0.39, RRR=1.47, p<0.05), and better quality of decision making was associated with a 46% lower risk of overweight (b=-0.60, RRR=0.54, p<0.05).  

What does this mean?

Less advantageous decision making in childhood predicts ED symptoms in adolescence and represents a prevention target.

You can read the article on this work which has been published in the Journal of Developmental & Behavioral Pediatrics.

You can watch my talk on this research which I presented at the Academy for Eating Disorders International Conference on Eating Disorders 2020 here.

2) Higher risk taking was associated with a higher likelihood of being in the eating disorder risk group and better quality of decision making was associated with a lower likelihood of being in the eating disorder risk group

We used latent class analysis to examine whether there were clusters of eating disorder symptoms developing in adolesence.

Rather than identifying several types of eating disorder in the cohort, this analysis revealed two groups within the cohort:

1) A group at lower risk of ED symptoms, who were more likely to be of ‘average’ weight, according to the UK90 Growth Charts, with minimal disordered attitudes and behaviours in relation to eating and weight;

2) A group at higher risk of eating disorder symptoms, who had more body dissatisfaction, a desire to lose weight, were using dietary restriction and exercise to influence weight and were more likely to be overweight.

Logistic regression models showed that, after adjustment for confounding, higher risk taking scores were associated with a 60% greater probability of being in the higher risk group (b=0.47, OR=1.60, p<0.01), and higher scores on quality of decision making were associated with a 30% lower probability of being in the higher risk group (b=-0.34, OR=0.70, p<0.05).

What does this mean?

Moderating risk taking and improving decision making are targets for ED prevention.

You can read the article on this work which was published in Frontiers in Psychology.

Keep reading below for more of our findings.

Decision Making is Important in Understanding how Eating Disorders Develop

What else have we learnt from this project?

 

3. Risk taking improves less between childhood and adolescence (between the ages of 11 and 14) in those who are in the eating disorder risk group at 14.

Although decision making skills improved in all participants between childhood and adolescence, as expected in typical development, one component of decision making, the risk taking variable, improved less in those who went on to show eating disorder symptoms.

This was because the data showed a smaller reduction in risk taking was associated with a higher probability of being in the higher risk group at age 14.

This work is has been published in an open access paper in European Psychiatry.

I’ve also written about this part of the project here in this blog post and you can watch my talk on this research here.

What does this mean?

This is evidence for a neurodevelopmental aetiology for eating disorders.

4. Decision making difficulties mediate the association between poor emotion regulation and eating disorder symptoms in adolescence

Here, we sought to understand more about decision making skills and whether they might help to explain an association between emotion regulation skills and eating disorder symptoms. For example, it might be that better emotion regulation skills protect against the later development of an eating disorder and one mechanism through which they achieve this is that they help people to make advantageous decisions, especially under uncertainty and risk.

In addition to the Cambridge Gambling Task and eating disorder symptom variables, we also included data from the Child Social Behaviour Questionnaire in our models. This measure is completed by the participant’s mother at ages 3, 5 and 7 and has two subscales:

1) Independence and Self-Regulation

2) Emotional Dysregulation

We modelled these data using latent growth curve analysis and the individual predicted values of the intercept (set at baseline, 3 years) and the slope (rate of annual change) were then used in the mediation analysis.

We found evidence for mediation by three measures of decision making at age 11 (poor quality of decision-making, delay aversion and low risk-adjustment) in the association between the Emotional Dysregulation subscale across ages 3 to 7 and eating disorder symptoms at age 14 even after adjustment for relevant covariates.

There was no evidence for an the association between the Independence and Self-Regulation subscale and eating disorder symptoms.

What does this mean?

Emotion regulation during childhood may be relevant for the future onset of eating disorder symptoms via its association with decision making skills. Helping children to develop strong emotion regulation skills may be one way of building resilience to eating disorder symptoms in adolescence and beyond.

You can read more about this work which has been published in an article in Psychological Medicine.

5. Decision Making in Childhood age at age 11 and adolescence at age 14 predicts the presence of eating disorder symptoms in adolescence at age 14 AND 17.

New data from the Millennium Cohort Study were made available in July 2020. These data contain responses collected from the cohort at age 17. We extended our models to include these new data.

We found that those at risk for eating disorder symptoms at age 14 were likely to continue to demonstrate these symptoms at age 17, suggesting that eating disorder symptoms aren’t usually transient events that go away.

We found that lower quality of decision-making at age 14 was associated with being in the high risk group at both age 14 and 17. Delay aversion at age 11 and lower risk taking at age 14 were associated with being in the low risk group at both age 14 and 17.

What does this mean?

This is evidence that disadvantageous decision making in childhood predicts the presence of eating disorder symptoms throughout adolescence.

This work will be reported in an article in BMJ Open (currently in press).

6. The association between emotion dysregulation in childhood and eating disorder symptoms in late adolescence is mediated by decision making in middle adolescence

Seeking to develop the findings discussed in (4), we extended our mediation model to include eating disorder symptoms at age 17.

We’re working on these models at the moment, so check back soon for the findings!