Proposal summaries
B3825 - Parenting and child self-esteem and depression symptoms - 05/07/2021
The aim of this study is to analyse the relationship between mothers’ and fathers’ parenting and child depression trajectories, focusing on aspects which have received limited attention in the literature. Past research has examined how parenting behaviour has influenced child wellbeing, highlighting the positive association of parental control, rejection and unsupportiveness with child depression trajectories and low self-esteem (McLeod, Weisz, & Wood, 2007; Yasmin & Hossain, 2014). Conversely, parental warmth (Del Barrio, Holgado-Tello, & Carrasco, 2016) and supportive behaviour (Juang & Silbereisen, 1999) have been identified as protective factors on that may limit the severity of depression symptoms. Nonetheless, the literature on child psychopathologies so far has prevalently focused on mothers’ parenting, or has failed to distinguish between the individual effects of mothers’ and fathers’ parenting on child outcomes (Parent, Forehand, Pomerantz, Peisch, & Seehuus, 2017). This appears to be limiting, as the role of fathers on child outcomes was found to be distinct from that of mothers (Jeynes, 2016; Connell & Goodman, 2002).
Very few papers have looked at the relationship between self-esteem and depression considering both mothers and fathers. As strong evidence has identified low self-esteem as a precursor for depression (Manna, Falgares, Ingoglia, Como, & De Santis, 2016; McClure, 2010; Orth, 2018), we hypothesise it should play an important mediating role between parenting and child depression. In addition, parental depression could be an important environmental factor to consider, as it is related to higher levels of depression in children (Kwong et al., 2019; Natsuaki et al., 2014), whilst a child’s friendships could be considered as a protective factor, since they have been found to be negatively associated with mental health difficulties (Preddy & Fite, 2012).
This study will therefore examine how both mothers’ and fathers’ parenting affect child depression. The relationship between parenting and child depression will be analysed taking into consideration the mediating effects of the child’s self-esteem. We will also examine any moderating effect of maternal and paternal depression, and a potentially protective role of friendships.
B3824 - Multivariate Genetic Study of Impulsivity and Binge Drinking through Late adolescence and Early Adulthood - 05/07/2021
Binge drinking behaviors in late adolescence and early adulthood have been shown to be important indicators of risk for developing problems with alcohol use and related consequences which may extend beyond early adulthood. Binge drinking behaviors are also thought to be influenced, in part, by impulsive personality traits such that more impulsive adolescents and emerging adults tend to binge drink more heavily, more frequently, and beyond early adulthood. However, the connection between impulsive personality and binge drinking has been clouded to some extent in past research because there is generally not a widely agreed upon “best” way to measure impulsive personality. Additionally, this measurement issue may obscure our ability to detect genetic influences on impulsive personality and binge drinking and their connection across the lifespan. The proposed study seeks to clarify the role that genetic factors might play in this relationship by using results from pre-existing genetic studies of impulsive personality and alcohol use to define genetic risk factors that reflect more succinct traits (e.g., the drive to behave impulsively in response to emotions and rewards, and the inability to stop or control these drives). These re-defined genetic risk factors will then be used to determine the extent to which genetic influences for impulsive personality predict changes in binge drinking across late adolescence and early adulthood in the ALSPAC sample.
B3823 - Genetic Covariation Between Alcohol Use Aggression ADHD and Executive Function Across Early Adolescence and Young Adulthood - 02/07/2021
The relationship between alcohol use and aggression and its public health consequences have long been established with some suggestions that executive functioning and attention-deficit/ hyperactivity disorder may play a role in this relationship. While it is generally accepted that these traits are similarly influenced by genetics, less is known about the specific genetic factors they share in common, and how this genetic overlap may change over time. The proposed project aims to use advanced genetic tools to identify biological contributions to the genetic overlap between alcohol use, aggression, ADHD, and executive function, to examine changes in their shared genetic influences across adolescence and young adulthood.
B3816 - Association between early infant and young child feeding and BMI Z score trajectory among children under 5 years old - 30/06/2021
In recent years, the rate of childhood obesity has risen rapidly in western countries. Systematic review showed that childhood obesity is associated with adult obesity and various adverse health outcomes in adulthood. In fact, overweight and obesity, and the relevant non-communicable diseases are largely preventable. Research has shown that infant and young child feeding is one of the important variables associated with early growth trajectory and later overweight and obesity. A pragmatic challenge of many previous studies is that Body Mass Index ((BMI); an indicator of obesity and overweight) is assessed at one point in time, which did not account for changes in BMI over time. The BMI Z score trajectory overcomes the limitations of static analysis and allows comparisons across ages. However, there are few studies exploring the association between infant feeding and BMI Z score trajectory in the UK. Relevant studies in other countries have different findings on the effects of complementary feeding and duration of breastfeeding on BMI Z score trajectory. It is controversial whether feeding pattern has an effect on BMI among children. No study has been found to explore the independent effect of the feeding pattern (on-demand vs. schedule) on BMI Z score trajectory among children.
B3822 - Pubertal Timing Physical Activity and Depression A Prospective Cohort Study - 28/06/2021
There is growing global concern about the rise in mental health problems in young people.
Rates of depression are known to increase after puberty, with a robust finding that early puberty in girls is linked to increased depressive symptoms. These findings support the ‘early timing’ hypothesis’, which proposes that early maturing girls find puberty more challenging as they are not yet prepared to cope with the biological, psychological and social changes. However, the relationship between pubertal timing and depression is less clear in boys. Understanding the mechanisms that explain the increase in depression during the
critical period of adolescence is vital to developing effective preventative interventions.
Levels of physical activity during adolescence might play a role in contributing to young people’s risk of depressive symptoms. There is some evidence that higher levels of physical activity are associated with lower levels of depression. Participating in physical activity may provide protection from developing depressive symptoms, for example by increasing self-esteem and improving social support. In adolescent girls there is evidence for a decline in physical activity levels at onset of puberty, particularly in those who develop earlier. In contrast, later maturing boys may have an athletic disadvantage which might discourage involvement in physical activity.
The proposed project will use repeated data on age at onset of puberty, physical activity, and depressive symptoms/depression to examine the role of physical activity as a potential mechanism that could explain the link between pubertal timing and depression in boys and girls in adolescence and young adulthood.
B3820 - Cannabis tobacco and mental illness isolating their relationship through triangulation of evidence - 23/06/2021
Cannabis and tobacco are two of the most commonly used substances worldwide. A substantial body of evidence documents an association between cannabis use and increased risk of psychotic and affective disorders. Tobacco use has also been associated with increased risk of several psychiatric outcomes. Co-use of cannabis and tobacco is a common practice, comprising concurrent co-use (i.e. use of both products in a pre-defined time-period) and co-administration (i.e. used simultaneously within the same delivery method). The high degree of overlap between these substances, and insufficient measurement in existing research, introduces complexity in accurately assessing how these substances impact on risk of subsequent mental illness. Therefore, it is possible that unmeasured tobacco exposure has confounded observed associations between cannabis and multiple psychiatric outcomes. Furthermore, the potential impact of co-use for mental health is an underexplored area. Studying populations where cannabis and tobacco are typically not administered together (e.g. USA) compared to populations where co-administration is common practice (e.g. Europe), offers an avenue through which to explore this issue. This project will explore the individual and combined roles of cannabis and tobacco in mental illness, through triangulating data from countries with differing cannabis-tobacco co-administration profiles. To further strengthen basis for causal inference, this project will also triangulate evidence across statistical and design-based approaches for studying causal effects.
B3821 - OLINK inflammation markesr in G0/G1 - 24/06/2021
As part of the large Templeton grant (B3397) investigating funds are available to generate inflammation data from blood samples. This proposal formalises the request for the relevant samples
B3818 - Epigenome-wide association study of depression and antidepressant use - 05/07/2021
At least 7 million adults in the UK are prescribed antidepressants every year. Antidepressant prescriptions in teenagers are rising. In nearly 50% of people on antidepressants the depression symptoms do not improve after the pharmacological treatment. We will investigate the biological pathways that lead to depression by looking at biomarkers in their blood DNA, as well as molecular markers that antidepressant treatments might leave on blood cells in young adults. We will assess antidepressant use in young adults and potential reasons for stopping the medications, with the ultimate aim to explore the possibility of personalized treatment options.
B3808 - The association between epigenetic prediction signatures of complex traits and health outcomes in ALSPAC - 22/06/2021
Modelling complex phenotypes using DNA methylation (DNAm) is becoming increasingly common in Epigenetic Epidemiology. This process often includes the use of weighted DNAm “scores” to differentiate between classes of categorical exposures, estimate continuous exposures and/or predict disease outcomes. The reason DNAm can do this is because it bridges the gap between your biology and your environment; one function of DNAm is that it will work to alter gene expression in response to an environmental stimulus. Accordingly, it can be thought of as a “biosocial archive”.
A recent study found that a DNAm proxy of smoking explained more variance in certain mental and physical health outcomes than someone self-reporting their smoking status. This is important because it highlights that DNAm can either augment or improve on how we define self-report traits such as smoking, and could potentially extend to other traits such as alcohol consumption, BMI and educational attainment. Further, using DNAm either in conjunction with, or instead of self-report, may allow researchers to identify targeted interventions for a variety of disease outcomes.
Accordingly, in this project we will be investigating whether DNAm proxies vs self-report smoking, BMI, alcohol consumption and educational attainment, respectively, explain more variance in cardiovascular, mental health, socioeconomic, clinical, metabolic and addiction outcomes in ALSPAC.
B3815 - Cholesterol trajectories throughout childhood and associations with growth and carotid intima-media thickeness - 28/06/2021
Cholesterol is a type of fat that is essential for our body to work well. Among its roles, it helps the body make use of hormones and vitamins. Our bodies make cholesterol, but we also get cholesterol from food. When we eat too many foods high in fat we can start to have a build-up of cholesterol in our bodies, especially in our blood vessels. We know that having too much bad cholesterol (called Low-Density Lipoprotein or LDL-cholesterol) in our blood can result in forming a layer that sticks to the inside walls of our blood vessels, making them thinner and sometimes blocking them, which can result heart disease, such as having a heart attack.
We know what levels of LDL-cholesterol are considered too high in adults, but we know very little about what normal and high levels of LDL-cholesterol are in children. We know cholesterol is important for brain development, that cholesterol may go up and down at certain ages in children, but there are very few studies of how cholesterol changes normally throughout childhood and adolescence. We also know little about how young people’s cholesterol is related to their growth and health (for example, weight and blood pressure), or whether it is related to early signs of heart disease in early adulthood.
We will answer these questions using this cohort. The results will be important for the general population and for children who have certain conditions that require medications to lower their cholesterol in childhood.
B3810 - Birth order and cord blood DNA methylation - 21/06/2021
Birth order has strong correlations with developmental and environmental aspects. Birth order is a presumed factor affecting multiple diseases including leukemia (1, 2, 3), atopy (4), type I diabetes (5), non-Hodgkin lymphoma (6) and synovial sarcoma (7). Moreover, first born subjects have unique in utero characteristics including less sufficient placentation, higher estrogen levels, and lower insulin sensitivity. Birth order might therefore play a role in shaping DNA methylation pattern and influence the initiation of the diseases mentioned above.
1. Dockerty JD, Draper G, Vincent T, Rowan SD, Bunch KJ. Case-control study of parental age, parity and socioeconomic level in relation to childhood cancers. Int J Epidemiol. 2001. doi:10.1093/ije/30.6.1428
2. Hjalgrim LL, Rostgaard K, Hjalgrim H, et al. Birth weight and risk for childhood leukemia in Denmark, Sweden, Norway, and Iceland. J Natl Cancer Inst. 2004. doi:10.1093/jnci/djh287
3. Westergaard T, Andersen PK, Pedersen JB, et al. Birth characteristics, sibling patterns, and acute leukemia risk in childhood: A population-based cohort study. J Natl Cancer Inst. 1997. doi:10.1093/jnci/89.13.939
4. Upchurch S, Harris JM, Cullinan P. Temporal changes in UK birth order and the prevalence of atopy. Allergy Eur J Allergy Clin Immunol. 2010. doi:10.1111/j.1398-9995.2009.02312.x
5. Cardwell CR, Stene LC, Joner G, et al. Birth order and childhood type 1 diabetes risk: A pooled analysis of 31 observational studies. Int J Epidemiol. 2011. doi:10.1093/ije/dyq207
6. Grulich AE, Vajdic CM, Kaldor JM, et al. Birth order, atopy, and risk of non-Hodgkin lymphoma. J Natl Cancer Inst. 2005. doi:10.1093/jnci/dji098
7. Wiemels JL, Wang R, Feng Q, et al. Birth characteristics and risk of early-onset synovial sarcoma. Cancer Epidemiol Biomarkers Prev. April 2020:cebp.0093.2020. doi:10.1158/1055-9965.EPI-20-0093
B3809 - Comprehensive Prenatal Smoking Assessment and Newborn Methylation - 21/06/2021
The detriment of maternal smoking on offspring is well known while the biological mechanism regulating the relationship are not fully understood. Epigenetic profile has emerged as a potential link between the two. Sustained maternal smoking during pregnancy, here defined as smoking past the first trimester, has been associated with deferentially methylated CpG sites in the offspring (1-6) and a dose-response effect might also be plausible (2, 4, 7).
1. Joubert BR, Felix JF, Yousefi P, Bakulski KM, Just AC, Breton C, et al. DNA Methylation in Newborns and Maternal Smoking in Pregnancy: Genome-wide Consortium Meta-analysis. Am J Hum Genet. 2016;98(4):680-96.
2. Ladd-Acosta C, Shu C, Lee BK, Gidaya N, Singer A, Schieve LA, et al. Presence of an epigenetic signature of prenatal cigarette smoke exposure in childhood. Environmental research. 2016;144(Pt A):139-48.
3. Lee KW, Richmond R, Hu P, French L, Shin J, Bourdon C, et al. Prenatal exposure to maternal cigarette smoking and DNA methylation: epigenome-wide association in a discovery sample of adolescents and replication in an independent cohort at birth through 17 years of age. Environ Health Perspect. 2015;123(2):193-9.
4. Markunas CA, Xu Z, Harlid S, Wade PA, Lie RT, Taylor JA, et al. Identification of DNA methylation changes in newborns related to maternal smoking during pregnancy. Environ Health Perspect. 2014;122(10):1147-53.
5. Rzehak P, Saffery R, Reischl E, Covic M, Wahl S, Grote V, et al. Maternal Smoking during Pregnancy and DNA-Methylation in Children at Age 5.5 Years: Epigenome-Wide-Analysis in the European Childhood Obesity Project (CHOP)-Study. PLoS One. 2016;11(5):e0155554.
6. Witt SH, Frank J, Gilles M, Lang M, Treutlein J, Streit F, et al. Impact on birth weight of maternal smoking throughout pregnancy mediated by DNA methylation. BMC Genomics. 2018;19(1):290.
7. Richmond RC, Simpkin AJ, Woodward G, Gaunt TR, Lyttleton O, McArdle WL, et al. Prenatal exposure to maternal smoking and offspring DNA methylation across the lifecourse: findings from the Avon Longitudinal Study of Parents and Children (ALSPAC). Hum Mol Genet. 2015;24(8):2201-17.
B3799 - Exploratory ML Based Longitudinal Modelling Approaches to Map Genetic-Phenotypic Multi-Trait Predictors of Fibrosis - 28/06/2021
Global Computational Biology & Digital Sciences (GCBDS) is a department in Boehringer Ingelheim’s Innovation Unit (IU) and is principally engaged in early discovery research programs. In 2020 the IU embarked on a curriculum to more deeply explore healthcare databases and biobank initiatives. The overall objective of the program is to explore advanced analytical methods to harness genomic & phenotypic data in order to better understand causative pathological mechanisms for disease which have the long-term potential to support patients.
Boehringer has a longstanding heritage in a number of unmet areas of medical need we consider our core disease areas; These include cardio-metabolic diseases, central nervous system diseases, immunology, respiratory and oncology as well as fields of active research that are currently covered by Research Beyond Borders (e.g. infectious diseases or regenerative medicines).
As part of BI’s preparatory research, a meta-analysis was undertaken to identify initiatives which focus on longitudinal multi-generational familial analysis in the field of fibrosis research (lung, liver). Two area’s of interest are of high priority (i) developing computational based method development programs focused on accurate and interpretable statistical approaches and (ii) Exploring methods to undertake trans biobank analysis on individuals from diverse ethnic backgrounds, but sharing similar phenotypic traits.
B3814 - Polygenic susceptibility in chronic air pollution exposure associated adverse respiratory health in adolescents a meta-analysis - 05/07/2021
Several studies support an association between being exposed to air pollution and having respiratory health problems in adolescence. However, only some of those who are highly exposed to air pollution develop respiratory diseases. Genetics play a role in how an individual respons to air pollution exposure as well as in the probability to develop air pollution induced respiratory health problems. We want to investigate if there is a difference in how air pollution affects our respiratory health, when comparing 'genetic low-risk' and 'genetic high-risk' groups of people.
B3811 - Alcohol Consumption And Hypertensive Disorders Of Pregnancy A Negative Control Analysis In ALSPAC - 14/06/2021
Drinking alcohol when pregnant has been shown to be associated with reduced risk of having high blood pressure disorders during pregnancy. However, given that the opposite is true outside of pregnancy, that drinking alcohol is associated with increased blood pressure and risk of stroke and heart attack, the effect in pregnancy may be a result of other factors, not alcohol itself. We aim to use the Avon Longitudinal Study of Parents and Children (ALSPAC) to compare women who drank alcohol during their pregnancy and those who didn't, to see if drinkers were more likely to develop high blood pressure disorders during pregnancy than non-drinkers. We also aim to use information on partner's alcohol use during pregnancy and mum's risk of these disorders, to try and understand if the effect seen in mum is caused by alcohol or whether it is related to other factors shared by mum and her partner. In other words, given that a partner's drinking can't have a physical effect on mum's risk of these disorders, if we see a decreased risk when comparing drinking and non-drinking partner's, we can assume that maybe there are other factors shared by both mum and partner that is causing the reduced risk, not the alcohol itself.
B3812 - Trajectories of problematic cannabis use in adolescence and their association with childhood psychological risk factors - 14/06/2021
Cannabis is the most commonly used illicit drug worldwide, with use particularly prevalent in adolescents. The number of under-18’s in treatment for cannabis as a primary problem in the UK in 2019-2020 was 11,136, 78% of all treatment entrants in this age group. Whilst this demonstrates a current clear demand for treatment of cannabis problems, the majority of people with cannabis problems do not seek out professional treatment. Further, current psychosocial treatment options lack data on their long-term efficacy, and there are no approved pharmacological treatments. One approach to tackle the global burden of cannabis use disorder (CUD) is to develop early targeted interventions, directed towards reducing the incidence of CUD. Identifying key variables that are associated with development of cannabis problems could help to ascertain which groups are most at risk. Previous research in adolescent and adult populations indicates that cannabis use profile, mental health problems and other drug use are all associated with increased risk of sustained problems related to cannabis use, compared to those with stable-low risk of cannabis problems. However, these studies either did not investigate childhood risk factors, or only assessed these retrospectively, and did not collect data specifically prior to cannabis initiation. This project aims to identify distinct trajectories of cannabis problems, and assess whether mental health and cognition in childhood (8-10 years old), prior to onset of cannabis use, represent potential modifiable risk factors for likelihood of experiencing problems with cannabis use, in order to inform early interventions to potentially reduce the incidence of cannabis problems.
B3801 - HMOX1 imputation testing and impact on outcomes - 10/06/2021
This PhD proposal aims to look at a gene, the HMOX1 gene, that is critical for managing the breakdown of haemoglobin in humans. Previous studies have shown that this gene is critical for management of infection. In animal studies, loss of function of this gene is associated with much worse outcomes in infection (more likely to die). In humans, there is much more uncertainty about the role of this gene and its importance. It has previously been shown that a small region of DNA in this gene is comprised of a dinucleotide repeat, where the two bases guanine and thymine are repeated. This is called a “GT” repeat. In this region, the DNA code is of variable length between patients, with an average of 31 repeats across the whole population, although some people have as little as 20 repeats, and some as high as 45. In previous small studies, the length of this repeat has altered the amount the gene is switched on and off (gene expression), and some studies have shown this has important impacts on lots of diseases, including in infection. Previous studies have shown those with longer repeats are more likely to be diagnosed with Type 2 diabetes, for example. Importantly, multiple studies in malaria have shown that carriage of the longer repeats is associated with much better outcomes, with a large reduction in severe malaria. However, all of the previous studies (in all clinical diseases) have been performed in small cohorts. In this proposal, we aim to measure this repeat using bioinformatic tools in ALSPAC using whole genome sequencing data. We also then plan to measure this repeat using another methodology, by imputing it from array data. These two methods will be compared to test the accuracy of repeat calling from array data. This will also be used to allow confidence in the imputation method so this can be used in other databases (e.g. UK Biobank). Subsequently, we will then look at the relationship between the repeat length and various clinical and bioinformatic measurements. In particular, we will focus on whether repeat length relates to immunological function (e.g. antibody levels, T-cell function (where measured)), clinical outcomes (both self-reported and healthcare linked), and metabolomic outcomes. Finally, we will also look at the impact of this repeat on the transcriptome, to see if gene expression truly changes with the repeat length. Further work will compare results from ALSPAC with other cohorts e.g. UK Biobank. All participants who have genetic data will be included.
B3806 - Examining the genetic architecture of depression in and around pregnancy in the ALSPAC study - 07/06/2021
Depression is a common and complex disease, influenced by both genetic and environmental factors, with heritability estimated to be from 31% to 42%. However, a recent genome-wide association study meta-analysing data on 807,553 individuals from three largest GWAS studies identified 102 independent variants and 269 genes associated with depression. This result suggests that depression is a polygenetic trait and enables researchers to use these polygenic risk scores to investigate how genetic liability for depression manifests over the life course.
Depression is more common in women during childbearing age, but it has particular importance during pregnancy and after birth due to additional potential consequences for the child. There are many women worldwide suffering from perinatal depression, and according to a meta-analysis, the prevalence of it ranges from 6.5% to 13% at different time points during pregnancy. Perinatal depression has been reported to be associated with many complications, such as hypertension, preeclampsia, and gestational diabetes for mothers and premature birth, low birth weight, fetal growth restriction for foetuses. Furthermore, some pregnant women with depression are reported to have suicidal ideation and thoughts of harming their child. It is clear that perinatal depression is important for both the mother and the child, and thus it is important to examine its aetiology in order to identify potential treatments and interventions. The role of genetics in perinatal presentation of depression is one particular path that could further enhance our understanding of the disease.
There are many studies on heritability of depression, but few about the genetic architecture of perinatal depression. This study will use the polygenetic risk score based on GWAS study on broad depression and examine the association between greater genetic liability to depression and depression across the perinatal period.
B3803 - Inflammation and Health The causes and consequences of GlycA and its role in the inflammatory system - 07/06/2021
Inflammation is known to be associated with adverse non-communicable diseases (NCDs). However, the most commonly used biomarkers in inflammation research (namely C-Reactive Protein and Interleukin-6) are potentially not valid measures of chronic and cumulative inflammation. This is because they respond to short-term, acute inflammatory changes in the body. This therefore limits our knowledge of the true association between chronic inflammation and NCDs, but also limits our understanding of the role certain exposures (such as adverse childhood experiences; ACEs) have on levels of chronic inflammation.
A novel biomarker GlycA may be a more reflective of chronic inflammation because it is not affected by short-term inflammatory changes, given that it is a composite biomarker. Therefore GlycA may be a more valid measure compared to previously used biomarkers. This project aims to use GlycA to investigate the causes and consequences of chronic inflammation more accurately, with a focus on mental health (such as depression). The project will collaborate with the work being done by Debbie Lawlor’s group investigating the association between GlycA and cardiovascular disease.
B3800 - Body Mass Index trajectories across childhood and young adulthood and risk for disordered eating in young adulthood - 07/06/2021
Disordered eating behaviours (such as fasting, purging, binge-eating and excessive exercise) are common in young adulthood, and have the potential to cause serious long-term harm. Identifying risk factors that precede the onset of disordered eating in young adulthood is important for our understanding of how disordered eating develops. Furthermore, these risk factors may be useful for identifying individuals at high risk of developing disordered eating and for designing effective treatments.
Although diagnosis for an eating disorder often depends on Body Mass Index (BMI) and this can be a barrier for treatment, there is less evidence on whether changes in BMI across childhood and adolescence precede later disordered eating. Previous research using the ALSPAC sample (Perry et al. 2021) has found 5 different BMI trajectories from 1-24 years: stable average, gradually decreasing, puberty-onset minor increase, puberty-onset major increase and persistently high BMI. Compared to those with a stable average BMI, those with a puberty-onset major increase trajectory had higher risk of a depressive episode and depressive symptoms. However it is unclear whether these BMI trajectories are differentially associated with subsequent disordered eating.
This study investigates the associations between BMI trajectories across childhood up to age 24 with disordered eating behaviours at age 24. This study will improve our understanding of how BMI can lead to the development of disordered eating behaviours.