Access the full text.
Sign up today, get DeepDyve free for 14 days.
Loading next page...
/lp/springer-journals/gendered-pathways-of-internalizing-problems-from-early-childhood-to-d7TOmq3Y7e
References (0)
References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.
- Publisher
- Springer Journals
- Copyright
- Copyright © The Author(s) 2020
- ISSN
- 0091-0627
- eISSN
- 1573-2835
- DOI
- 10.1007/s10802-020-00623-w
- Publisher site
- See Article on Publisher Site
Abstract
Despite trends indicating worsening internalizing problems, characterized by anxiety and depression, there is dearth of research examining gender differences in developmental trajectories of internalizing problems from early childhood to adolescence. Drawing on the UK Millennium Cohort Study (n = 17,206, 49% female), this study examines trajectories of parent-reported, clinically-meaningful (reflecting the top 10%) internalizing problems from ages 3 to 14 years and their early predictors and adolescent outcomes. Group-based modelling revealed three trajectories when examining boys and girls together, but there were significant gender differences. When examining boys and girls separately, four trajectories were identified including two rela- tively stable trajectories showing either high or low probabilities of internalizing problems. An increasing trajectory was also found for both boys and girls, showing an increasing probability of internalizing problems which continued to rise for girls, but levelled off for boys from age 11. A decreasing trajectory was revealed for boys, while a moderate but stable trajectory was identified for girls. Boys and girls in the increasing and high probability groups were more likely to report a number of problematic outcomes including high BMI, self-harm, low mental wellbeing, depressive symptoms, and low educational moti- vation than the low group. Girls on the increasing trajectory also reported more cigarette and cannabis use and early sexual activity at age 14 compared to girls on the low trajectory. Findings suggest that intervention strategies take a systemic view, targeting not only internal feelings, but also behaviours potentially associated with later negative outcomes. . . . . . Keywords Internalizing problems Developmental trajectories Early childhood Adolescence Group-based modeling Adolescent outcomes Internalizing problems, characterized by anxiety and depres- (characterised by sadness, loss of interest and energy, and sion, represent one of the most common forms of child psy- low self-esteem), for children aged 5 to 15 years living in chopathology, with a higher prevalence in girls than boys England, rising from 3.8% in 2004 to 5.8% in 2017 (Mental (Green et al. 2005). Studies of school-age children using Health of Children and Young People in England, 2018). parent-reports as well as diagnostic tools have shown a strong, Internalizing problems in childhood and adolescence are positive association between depression and generalized anx- strongly predictive of later difficulties including co-morbid iety, suggesting that these can be classified according to a mental health problems, disrupted social relationships, sub- single internalizing disorder (Achenbach 1991; Moffitt et al. stance abuse, and reduced educational performance (Dekker 2007; Sterba et al. 2007). Recent data show a population-level et al. 2007;McLeodet al. 2016; Measelleetal. 2006). Despite increase in internalizing problems, including anxiety disorders evidence demonstrating the effectiveness of early intervention (characterised by fear and worry) and depressive disorders (Burt et al. 2016;Tothet al. 2016), internalizing problems in children and adolescents often are left undiagnosed and un- treated (Public Health England 2016). Given the possible neg- ative consequences as well as the potential for early interven- * Leslie Morrison Gutman tion, it is critical to understand the development of internaliz- l.gutman@ucl.ac.uk ing problems from an early age. This study identifies sub- groups with distinct longitudinal profiles of parent-reported University College London, 1-19 Torrington Place, Londo6n WC1E internalizing problems from ages 3 to 14 years and investi- 7HB, UK gates how early predictors and adolescent outcomes differen- UCL Institute of Education, 20 Bedford Way, London WC1H 0AL, tiate these trajectory groups, assessing gender differences. The UK 704 J Abnorm Child Psychol (2020) 48:703–718 identification of diverse developmental trajectories from early There are documented gender differences in the prevalence childhood to adolescence has important clinical implications rates, developmental course, precursors, and consequences of for prevention and treatment approaches by providing insight internalizing problems (Zahn-Waxler et al. 2008). Consistent into the pathways leading to different subtypes of internalizing gender differences in internalizing trajectories are thought to difficulties. emerge in adolescence, with girls reporting higher mean levels and a sharper increase in problems compared to boys (Leve et al. 2005). A number of hypotheses have been put forth to Trajectories of Internalizing Problems explain these gender differences including dispositional char- acteristicssuchasgirls’ heightened reactivity and rumination A developmental psychopathology framework emphasizes styles and socialization experiences such as parents’ expecta- elucidating variation in the age of onset and developmental tions for daughters to be more prosocial and submissive than course of normative and psychopathological development, re- sons (Zahn-Waxler et al. 2000). These risk factors have the vealing continuities and discontinuities among diverse path- potential to lead to greater internalizing problems in the face of ways (Cicchetti and Rogosch 2002). This framework views challenges in early adolescence (Nolen-Hoeksema and Girgus development as an active dynamic process that can diverge 1994). Developmental models that test for possible gender depending on children’s individual characteristics and envi- differences will help elucidate whether there are distinct path- ronmental contexts, showing unique patterns of change for ways for boys and girls and, if so, whether there are differ- different subgroups. Group-based trajectory modelling has en- ences in their level of severity and/or age-related rates of abled a more heterogeneous specification of developmental change. Only one study examined gender-specific trajectories pathways than a variable-oriented approach, allowing the ex- of internalizing problems from ages 2 to 11 years for both amination of questions relevant to developmental psychopath- males and females (Sterba et al. 2007). This study found three ological theory (Cicchetti and Rogosch 2002). This person- trajectories: high, low, and decreasing/increasing, which de- centred approach has advanced research into externalizing and creased until around age 6 and then steadily increased. antisocial behaviour, highlighting that individuals follow dis- Although the number, prevalence, and predictive validity of tinct pathways from early childhood and through adolescence the trajectories were similar for boys and girls, there were (e.g., Gutman et al. 2019; Hyde et al. 2015). Most of the statistically significant gender differences in the initial values research examining the development of internalizing prob- and rates of change. Girls were classified in the high group lems has investigated the average longitudinal course, twice as often as were boys; while boys were twice as likely to neglecting heterogeneity. Yet, an examination of subgroups be in the decreasing/increasing group, highlighting potential with varying levels of severity and rates of change may illu- gender differentiation in trajectories of internalizing minate different etiological and predictive relationships psychopathology. (Bauer & Curran, 2003). There are several limitations of the available literature base. A handful of studies have identified developmental trajec- All of these studies relied on samples which were gathered tories of internalizing problems from childhood to adoles- before the millennium, and only one study examined gender cence (Fanti and Henrich 2010; Korhonen et al., 2014; differences. The generation born after the millennium has faced Letcher et al. 2009;Nivard et al., 2017; Sterba et al. 2007), unique challenges, including the emergence of social media as a showing between three and six trajectory groups. Using di- prominent pastime for adolescents. The increased accessibility verse measures including the Child Behavior Checklist and time spent on social media has raised new concerns about (CBCL) and DAWBA, all of these studies identified both high adolescents’ mental health (Twenge et al. 2018). Furthermore, and low trajectories, showing stable levels of either high or there is a documented population-level increase in internalizing low internalizing problems, respectively, from childhood to distress, particularly for girls (Collishaw et al. 2010;Finket al. adolescence. Of these, the studies examining data from child- 2015; Gutman et al. 2018a), highlighting the importance of hood through middle or late adolescence revealed both in- assessing gender differences in internalizing trajectories for creasing and decreasing groups, where children show more recently born nationally-representative samples, allowing either an increase or decrease, respectively, from childhood conclusions drawn at the population-level. In addition, none of to adolescence (Korhonen et al., 2014; Letcher et al. 2009; these studies have examined gender differences in internalizing Nivard et al., 2017). One of these studies also examined ex- problems from toddlerhood to mid-adolescence. An examina- ternalizing scores and modeled their co-occurrence across tion of internalizing problems from early childhood would childhood and adolescence, showing an association particu- show the emergence of gender differences in internalized pa- larly when problems started early (Nivard et al., 2017). thology (Sterba et al. 2007), while following their course into Overall, these studies suggest that there is heterogeneity in adolescence would enlighten our understanding of their diver- the pathways of internalizing problems from childhood to gence across development. Drawing on the Millennium Cohort adolescence. Study (MCS), a nationally representative sample of children J Abnorm Child Psychol (2020) 48:703–718 705 born in the UK in 2000–2001, this study fills these research (Raudsepp and Kais 2019). However, there is little or no research gaps through the identification of distinct trajectories using examining the role of social media use in gendered pathways of parent-reported internalizing problems from ages 3 to 14 years. internalizing problems. Internalizing problems may also be related to the timing of puberty (Patton et al. 2008), with some evidence showing a Early Predictors and Adolescent Outcomes stronger association for females than males (Lewis et al. 2015; Patton et al. 2008;Negriff andSusman 2011; Ullsperger and A secondary aim is the examination of early predictors and Nikolas 2017). Higher BMI has also been shown to be a risk adolescent outcomes of internalizing problems trajectories. factor of internalizing problems, particularly for adolescent girls According to the developmental psychopathology approach, (Dockray et al. 2009;Richardsonetal. 2006). However, there is diverse developmental trajectories are distinguished by differ- little research examining the role of puberty and BMI in associ- ent risk etiologies and associated outcomes. This provides ation with heterogeneous trajectories of internalizing problems external validation by assessing whether membership in a par- for boys and girls from early childhood to adolescence. ticular trajectory can be predicted by and predict measures As a means of external validation, measures of adolescent- other than those used to create the trajectory groups (von reported mental wellbeing and depressive symptoms are in- Eye and Bergman 2003). cluded as outcomes. Further associations among parent- In terms of early factors, this study examines those factors reported trajectories of internalizing problems and parent- that have been shown to predict heterogeneity in the develop- reported mental health difficulties, including conduct prob- ment of internalizing problems from toddlerhood through ad- lems, peer problems, and hyperactivity, are examined. olescence including parental psychopathology, socio- Lastly, based on research suggesting that positive school ad- economic disadvantage, low birthweight, and smoking in justment and better parent-child relations are associated with pregnancy (Fanti and Henrich 2010;Nivardet al., 2017; recovery from elevated internalizing trajectories, adolescent- Shore et al. 2018). We extend previous findings by examining reported measures of educational motivation and parent-child both paternal and maternal psychopathology. relationships are included as outcomes (Letcher et al. 2009). Given that there is little evidence concerning how trajecto- ries of internalizing problems from childhood to adolescence may be related to adolescent outcomes, this study explores this association among a number of relevant adolescent out- Current Study comes including problematic behaviours, mental and physical health, and relationships. For problematic behaviours, includ- Drawing on the Millennium Cohort Study (MCS), a nationally ing alcohol, cigarette, substance abuse, and early sexual activ- representative sample of children born in the UK in 2000– ity, there is evidence showing an association between these 2002, this study addresses research gaps through the identifi- behaviours and depressive symptoms (e.g., Chaiton et al. cation of distinct trajectories of parent-reported internalizing 2009; Costello et al. 2008; Danzo et al. 2017;Skogen etal. problems from ages 3 to 14 years and the examination of early 2016), but less is known about gender differences. In line with predictors and adolescent outcomes. Unlike studies that ex- the “gender paradox of co-morbidities” (Loeber and Keenan amine gender differences in internalizing trajectories where 1994), girls may be less likely to engage in delinquent behav- males and females are grouped together, this study tests iour than boys, but when they do, they may be more likely to whether the intercepts and slopes of heterogeneous pathways be depressed and anxious (Zahn-Waxler et al., 2008). differ according to gender. If statistically significant differ- Examining their associations with trajectories of internalizing ences emerge, then distinct gendered trajectories will be iden- problems from early childhood to adolescence will provide a tified. This is important as some pathways may be identified better understanding of how heterogeneous groups experience for one gender, but not for the other. Despite the advantages of different manifestations of later problematic behaviours. estimating gender-specific trajectories, trajectories may be There has also been increasing attention on the negative out- identified that are not clinically meaningful (Gutman et al., comes associated with social media in adolescence (Best et al. 2018b), e.g., a high trajectory group of boys that has relatively 2014;Strasburgeretal. 2013; Woods and Scott 2016), with some lower levels of internalizing problems than girls. To remedy indication that the mental health of girls may be vulnerable to its this, age-based bandings using national norms from England use (Booker et al. 2018). Although a moderate significant asso- (reflecting the top 10%), shown to strongly predict later inter- ciation have been found between social media and depressive nalizing diagnoses, are used (Meltzer et al. 2000). Measures of symptoms in young people, most of these studies are cross- mental health problems tend to be highly skewed, with most sectional or of a limited duration (Barry et al. 2017;McCrae individuals in the lowest category. Therefore, a clinically rel- et al. 2017). There is recent evidence that increasing use of social evant measurement of internalizing problems may be better media is associated with increasing depressive symptoms in girls able to detect diverse but meaningful developmental patterns, 706 J Abnorm Child Psychol (2020) 48:703–718 providing an understanding of the pathways leading to clini- 2000 and January 2002 (Joshi and Fitzsimons 2016). The cally diagnosable internalizing disorder. survey was sampled in a complex clustered and disproportion- n terms of the number of trajectories, the existing literature ately stratified design. The clusters were electoral wards, and described above suggests that four distinct trajectories may be the strata oversampled areas of high child poverty, minority identified for males and females. Both low and high trajectories ethnic populations in England and the three smaller countries are expected, with a higher prevalence of females in the high of the UK. Data are so far available from six sweeps of inter- group (Sterba et al. 2007). Developmental change in boys and views with the families. The first survey, MCS1 (child age girls is also expected for better and for worse. In line with other 9months) wasinthe fieldmainlyin 2001, fieldwork for studies (Korhonen et al., 2014; Letcher et al. 2009;Nivardetal., MCS2 (age 3 years) was mainly during 2004, for MCS3 2017), a decreasing group, who initially present as having a high (age 5 years) mainly during 2006, and for MCS4 (age 7 years) or moderate probability of internalizing problems but shows a mainly during 2008. MCS5 (age 11 years) collected data decrease over time, as well as an increasing group, who show a mainly in 2012 when the cohort children were in their last rising probability of poor internalizing health during the transi- year of primary school. MCS6 (age 14 years) collected data tion to adolescence and beyond, may be identified. However, mainly in 2015 when they were in secondary school. girls may show an increasing probability of internalizing prob- Informants were overwhelmingly mothers (more than 95%). lems earlier (around age 10), as compared to boys (Kelly et al. The number of families who have been interviewed at least 2016). Further, in line with prevalence rates in adolescence once is 19,243, including 692 families in England who were (Public Health England 2016), there may be a higher prevalence not recruited until MCS2. If these cases are counted, the initial of females than males in the increasing group. response rate was 71%. In this study, the sample included one As shown in previous studies (Fanti and Henrich 2010; child per family, excluding children who were the second or Nivard et al., 2017; Sterba et al. 2007), it is expected that early third in sets of twins and triplets. Group-based trajectories risk factors including maternal and paternal psychopathology, were based on 17,880 children (girls = 8765; males = 9115) maternal smoking in pregnancy, and socio-economic disad- with parent ratings of internalizing problems in at least two vantage are associated with the high or decreasing trajectories, surveys. in comparison to the low group. Maternal psychopathology may be more strongly related to the higher problem groups for girls than boys (Zahn-Waxler et al. 2000). Given the explor- Measures atory nature of the adolescent-reported outcomes, there are no firm expectations regarding their associations, although prob- Internalizing Problems lematic behaviours may show a stronger association with the high and increasing trajectories compared to a low group, Internalizing Problems were assessed with the emotional particularly for girls in light of the “gender paradox of co- problems subscale of the Strengths and Difficulties morbidity” (Loeber and Keenan 1994). Girls on the high or Questionnaire (SDQ) (Goodman 1997, 2001), completed by increasing trajectories may also be more likely to have a high the parent. The SDQ is a screening questionnaire with exten- BMI and use social media compared to boys on these path- sive psychometric support (www.sdqinfo.com). In the MCS, ways (Booker et al. 2018; Dockray et al. 2009; Richardson construct, convergent, discriminant, and predictive validity et al. 2006). As a means of external validation, those on the have been established for the SDQ subscales, showing good high or increasing pathways may also be more likely to report internal reliability, ranging from 0.75 to 0.79 at ages 3, 5, and lower mental wellbeing and more depressive symptoms com- 7 for emotional problems (Croft et al. 2015). At ages 11 and pared those on other trajectories. Lastly, in light of the co- 14, alphas were 0.71 and 0.73, respectively. The questionnaire morbidity among parent-reported mental health problems for assesses emotional problems in the past 6 months using five children and adolescents, it is expected that those on the prob- items including “many fears, easily scared”, “often unhappy, lematic pathways will show higher levels of parent-reported down-hearted or tearful”, and “many worries, often seems conduct problems, peer problems, and hyperactivity com- worried” (0 = not true, 1 = somewhat true, 2 = certainly true). pared to those on the low pathway. These scores are totalled with a range of 0 to 10, with parents reporting a mean score of 2.04 (SD = 2.14) at age 3, 1.40 (SD = 1.61) at age 5, 1.54 (SD = 1.77) at age 7, 1.87 (SD = Method 2.00) at age 11, and 2.05 (SD = 2.14) at age 14. To ensure that these levels are clinically meaningful, SDQ bandings were Study Sample used based on externally given UK norms at each age (Meltzer et al. 2000), where 10% in that reference sample with MCS is a nationwide longitudinal study following children the highest scores were considered to be at high risk of emo- born in all four countries of the UK between September tional problems (0 = not high risk; 1 = high risk). Using those J Abnorm Child Psychol (2020) 48:703–718 707 SDQ bandings in this sample, 9.22% (SD = 0.30) of the chil- yes). Finally, young people were asked how often they argued dren were considered to be high risk of conduct problems with with their mother and father (1 = hardly ever, never; 2 = less than a mean score for the totalled emotional problems subscale of once a week; 3 = more than once a week; 4 = most days). 4.91 (SD = 1.25) at age 3, 5.61% (SD = 0.23) with a mean Three additional measures were taken from the young per- score of 5.81 (SD = 1.16) at age 5, 7.64% (SD = 0.27) at age sons’ self-completed questionnaire. Mental wellbeing was 7 with a mean score of 5.95 (SD = 1.22), 11.13% (SD = 0.31) assessed using a measure developed for the youth survey of with a mean score of 6.14 (SD = 1.31) at age 11, and 13.76% the British Household Panel Study in the 1990s (Taylor et al. (SD = 0.34) with a mean score of 6.24 (SD = 1.42) at age 14. 2010). This consists of a six-item scale including questions about their satisfaction with different areas of their life, includ- Early Predictors ing schoolwork, appearance, family, friends, school, and life as a whole. Responses were on a 1 (completely happy) to 7 All early predictors were measured when the child was (not at all happy) scale. The mean of responses was calculated 9 months-old. These include: race/ethnicity (0 = White for children’s overall wellbeing score, and responses were British; 1 = Black and Minority Ethnicity (BME), teenage reverse coded so that a higher score represented higher mother (1 = mother 19 years or younger at the child’s birth; wellbeing (alpha = 0.86). 0 = older than 19 years), low birthweight (1 = less than 2.5 kg; For depressive symptoms, the shortened-version of the 0 = other), single parent families (1 = single parent; 0 = two-par- Moods and Feelings Questionnaire (MFQ) was used. As a ent family), parental education (1 = no qualifications or qualifi- screening tool for depression, this measure consists of 13 de- cations below General Certificate of Secondary Education scriptive phrases about how they had been acting and feeling (GCSE) level; 0 = qualifications at or above GCSE level), pa- recently (Angold et al. 1995), such as: “I felt miserable or rental income (1 = lowest income quintile; 5 = highest quintile), unhappy”, “Ididn’t enjoy anything at all”,and “I felt so tired whether they lived in social housing (1 = yes; 0 = no), and I just sat around and did nothing” (1 = not true, 2 = sometimes, whether their mother smoked during pregnancy (1 = yes, 0 = 3 = true), and the mean of responses was used , with higher no). scores representing more negative feelings (alpha = 0.93). Maternal and paternal depressive symptoms (alpha = 0.72 To assess low educational motivation, the following question for mothers; 0.66 for fathers) were also measured using a 9- responses were combined: “How often do you try your best at item count variable as reported in Johnson et al. (2015)derived school?”, “How often do you find school interesting?” (reverse- from the (24 item) Malaise Inventory (Rutter et al. 1970). coded), “How often do you feel unhappy at school?”, “How Mothers and fathers answered such questions as “everything often do you get tired at school?”, “How often do you feel school gets on my nerves” and “I often feel miserable or depressed” is a waste of time?”,and “How often difficult to keep mind on (1 = yes, 0 = no). Mothers and fathers with a score of 5 or more work at school?” Responses ranged from all of the time (1) to were considered at risk of depression (Rodgers et al. 1999). never (4), and the mean of responses was calculated (alpha = 0.75). Adolescent-Reported Outcomes Parent-Reported Outcomes A number of single items at age 14 years were analysed, taken from the young persons’ self-completed questionnaire. Items in- Parent-reported conduct problems, peer problems, and hyper- clude: early menarche for females (0 = age 11 or older; 1 = before activity at age 14 were assessed by the SDQ (Goodman 1997, age 11); how many times they had an alcoholic drink in the last 2001). Alphas are 0.64, 0.63, and 0.78 respectively. The ques- 12 months (0 = never, 1 = 1–2times,2=3–5times, 3=6–9 tionnaire assesses mental health problems in the past 6 months times, 4 = 10–19 times, 5 = 20–39 times or 6 = 40 times or using five items for each subscale. Example questions include more); how often they smoked cigarettes (0 = never, 1 = only “often lies or cheats” for conduct problems, “rather solitary, tried smoking, 2 = used to smoke, 3 = sometimes smoke, 4 = tends to play alone” for peer problems and “easily distracted, usually 1–6 cigarettes a week, 5 = usually more than 6 cigarettes concentration wanders” for hyperactivity. SDQ bandings a week); how often they smoked cannabis (0 = never, 1 = 1–2 based on externally given UK norms at each age were times, 2 = 3–4times, 3=5–10 times, 4 = 10+ times); used (Meltzer et al. 2000), where 10% in that reference sample whether they had a high BMI (0 = other; 1 = 85th percentile or with the highest scores were considered to be at high risk of higher); how many hours they spent on social networks per week mental health problems (0 = not high risk; 1 = high risk). (0 = none, 1 = less than half an hour, 2 = half an hour to 1 hour, 3=1to2hours,4=2 to 3hours, 5=3 to5hours,7=6 to Statistical Analyses 7 hours, 7 = 7 hours or more), and whether they ever self- harmed (0 = no; 1 = yes). They were also asked if they had en- Group-based trajectory analysis in STATA TRAJ (Jones and gaged in any sexual activity in the past 12 months (0 = no; 1 = Nagin 2013) was used to identify discrete groups of children 708 J Abnorm Child Psychol (2020) 48:703–718 following similar progressions of internalizing problems as a problems was similar amongst girls and boys in most age function of age measured in months at each interview. Group- groups, at age 14, girls were significantly more likely to have based trajectory modelling is a specialized form of finite mix- internalizing problems compared to boys. Girls were also ture modelling (see Nagin 2005; Nagin and Odgers 2010). more likely to have a low birthweight. In terms of Full Information Maximum Likelihood (FIML) estimated adolescent-reported outcomes, girls were more likely to report the model parameters, thereby including every case with at smoking tobacco, self-harming, and spending time on social least two parental ratings (Schafer and Graham 2002). media. Girls also reported lower mental wellbeing, more de- Binary logit distribution was specified as internalizing prob- pressive symptoms, and more arguments with their mothers lems are considered a dichotomous variable (e.g., whether than boys, while boys reported lower educational motivation clinically meaningful or not). To establish the best fitting so- compared to girls. Parent-reported conduct problems, peer lution, a range of fit indicators was examined, including the problems, and hyperactivity were all higher for boys than lowest absolute Bayesian Information Criterion (BIC) (Nagin girls. 2005), the average posterior probability of group membership (0.70 being acceptable), and a close correspondence between Trajectories of Internalizing Problems the estimated probability of group membership and the pro- portion assigned to that group based on the posterior proba- Group-based trajectory analysis was first run with both boys bility of group membership. To assess whether gender differ- and girls together. Models with three to five trajectories with ences were evident in the intercept and slope of the trajecto- linear to quadratic functional forms were examined. The three- ries, gender and time-varying gender by age covariates were group, quadratic model fit the data best. The BIC score for the included in the model (Jones and Nagin 2013). three group, quadratic model (−18,404.5) had the absolute In order to account for the complex clustered and stratified lowest score compared to the four (−18,626.37) and five survey design of MCS, svy in STATA was used in the follow- (−18,486.06) group, quadratic models. The mean posterior ing stages of the analyses. First, gender differences in inter- probability scores ranged from 0.78 to 0.82 for the three- nalizing problems, early predictors, and adolescent outcomes trajectory model, with a mean of 0.80, indicating that most were assessed using univariate regressions for each predictor children fit their assigned trajectory well. Figure 1 depicts the on gender. For significant differences, the effect size using probability of clinically relevant internalizing problems for the Cohen’s d is reported. Then, the proportions and standard three trajectory groups from ages 3 to 14 years, along with the deviations of the early predictors and adolescent outcomes estimated proportion in each group. The predicted and ob- by the assigned trajectory group were examined (see served means were close, indicating a good fit of the model. Tables 2 and 3). To do this, univariate regressions were run There were low (65.6% estimated; 66.4% actual), high (9.2% for each factor on trajectory group status and then post-hoc estimated; 8.5% actual), and increasing (23.5% estimated; tests were conducted to compare all possible pairwise differ- 25.1% actual) probability groups. Gender differences in the ences among the four groups using the Bonferroni correction. intercept and slope of these trajectories were tested using gen- Sampling weights reflecting the MCS design were used in der, time-varying gender by age (linear slope), and time- the group-based trajectory modelling and subsequent analyses varying gender by age-squared (quadratic) covariates. These to correct for disproportionate sampling. The sampling findings revealed significant differences in the intercept, line- weights reduce the apparent size of cells populated by ar, and quadratic slopes, where p < 0.0001, for the high and oversampled strata, such as minority ethnic populations and increasing probability groups. Thus, group-based trajectory increase the apparent size of strata with under-sampled cases. analysis was run for boys and girls, separately. For the subsequent analyses, attrition weights were applied to For girls, the four-group, quadratic model fit the data best. restore the social profile of the whole cohort. The MCS survey The BIC score for the four group, quadratic model (−9840.27) team has developed attrition weights to correct for biases due had the absolute lowest score compared to the three to non-response (Hansen 2014). (−9955.57) and five (−9852.91) group, quadratic models. The mean posterior probability scores for girls ranged from 0.72 to 0.78 for the four-group trajectory model, with a mean Results of 0.74, indicating that most girls fit their assigned trajectory well. Figure 2 depicts the probability of clinically relevant Gender Differences in Internalizing Problems, internalizing problems for the four trajectory groups in girls Predictors and Outcomes from ages 3 to 14 years, along with the estimated proportion in each group. The predicted and observed means were close, Results for girls and boys are presented separately, and effect indicating a good fit of the model. The low problem group sizes for statistically significant differences are shown (see (55.2% estimated; 56.5% actual) displayed a near zero prob- Table 1). Although incidence of parent-reported internalizing ability of internalizing problems from ages 3 to 14. The J Abnorm Child Psychol (2020) 48:703–718 709 Table 1 Gender differences among internalizing problems, early predictors and adolescent outcomes Measures Boys Girls F-test Cohen’sd Mean(SD) 95% CI Mean(SD) 95% CI Internalizing problems Probability at age 3 0.09(0.28) (0.08–0.10) 0.09(0.28) (0.08–0.09) F(1, 389) = 0.12 Probability at age 5 0.05(0.23) (0.05–0.06) 0.06(0.23) (0.05–0.06) F(1, 389) = 0.05 Probability at age 7 0.08(0.28) (0.08–0.09) 0.08(0.27) (0.07–0.08) F(1, 389) = 1.55 Probability at age 11 0.12(0.32) (0.11–0.13) 0.12(0.33) (0.11–0.13) F(1, 389) = 1.44 Probability at age 14 0.12(0.33) (0.11–0.13) 0.18(0.38) (0.16–0.19) F(1, 389) = 67.85*** −0.15 Early predictors BME background 0.11(0.31) (0.09–0.13) 0.12(0.32) (0.09–0.14) F(1, 389) = 1.06 Teenage mother 0.03(0.17) (0.02–0.03) 0.03(0.17) (0.02–0.03) F(1, 389) = 0.17 Low birth weight 0.06(0.24) (0.06–0.07) 0.07(0.26) (0.07–0.08) F(1, 389) = 11.05*** −0.05 Single parent 0.15(0.35) (0.13–0.16) 0.14(0.35) (0.13–0.15) F(1, 389) = 0.85 Parent low qualifications 0.19(0.39) (0.18–0.20) 0.18(0.38) (0.17–0.19) F(1, 389) = 0.35 Income quintile 3.00(1.41) (2.92–3.08) 3.00(1.42) (2.92–3.07) F(1, 389) = 0.13 Social housing 0.23(0.42) (0.21–0.25) 0.24(0.43) (0.22–0.26) F(1, 389) = 0.17 Maternal smoking in pregnancy 0.22(0.42) (0.21–0.24) 0.21(0.41) (0.19–0.22) F(1, 389) = 3.69 Maternal depressive symptoms 0.08(0.27) (0.07–0.08) 0.07(0.26) (0.07–0.08) F(1, 389) = 3.31 Paternal depressive symptoms 0.05(0.21) (0.04–0.05) 0.05(0.22) (0.04–0.06) F(1, 389) =1.41 Self-reported adolescent outcomes Early menarche n/a n/a 0.24(0.43) (0.22–0.25) n/a How many times had an alcoholic drink 0.90(1.31) (0.85–0.96) 0.85(1.24) (0.80–0.90) F(1, 389) = 4.98 How often do you smoke 0.30(0.87) (0.27–0.33) 0.39(1.01) (0.35–0.42) F(1, 389) = 16.92*** −0.08 How often smoke cannabis 0.12(0.57) (0.10–0.14) 0.10(0.50) (0.09–0.12) F(1, 389) = 2.79 Ever self-harmed 0.09(0.29) (0.08–0.10) 0.22(0.41) (0.21–0.24) F(1, 389) = 356.80*** −0.35 High BMI 0.16(0.37) (0.15–0.17) 0.16(0.36) (0.14–0.17) F(1, 389) = 0.00 Hours spent on social networks 3.01(2.15) (2.92–3.11) 4.05(2.07) (3.97–4.13) F(1, 389) = 752.28*** −0.51 Engaged in sexual activity 0.03(0.18) (0.03–0.04) 0.03(0.17) (0.03–0.04) F(1, 389) = 0.36 Mental wellbeing 5.62(1.04) (5.58–5.65) 5.26(1.16) (5.21–5.30) F(1, 389) = 280.09*** 0.31 Depressive symptoms 1.34(0.38) (1.32–1.35) 1.55(0.51) (1.53–1.57) F(1, 389) = 657.26*** −0.48 Low educational motivation 2.90(0.48) (2.89–2.92) 2.86(0.53) (2.83–2.88) F(1, 389) = 20.77*** 0.08 How often argue with mother 2.75(1.04) (2.71–2.78) 2.91(1.06) (2.87–2.94) F(1, 389) = 86.35*** −0.17 How often argue with father 2.44(1.04) (2.39–2.48) 2.45(1.03) (2.41–2.49) F(1, 389) = 4.36 Parent-reported adolescent outcomes Conduct problems 0.15(0.36) (0.14–0.17) 0.11 (0.32) (0.10–0.12) F(1, 389) = 29.80*** 0.08 Peer problems 0.20(0.40) (0.19–0.22) 0.16 (0.37) (0.15–0.18) F(1, 389) = 18.86*** 0.07 Hyperactivity 0.15(0.36) (0.14–0.17) 0.08 (0.27) (0.07–0.09) F(1, 389) = 93.68*** 0.22 F-tests were conducted using svy in STATA, which reports design degrees of freedom for a complex clustered and stratified survey design. **p <0.01; ***p <0.001 increasing group (16.6% estimated, 16.7% actual) demon- high probability of close to 0.40 at age 3, increasing strated a near zero probability in early childhood and then until age 11, reaching more than 60% . showed an increase from ages 5 to 14, rising to more For boys, the final model meeting the selection criteria than 50%. A moderate group (21.7% estimated; 20.8% also included four quadratic trajectories. The BIC score actual) followed a probability of above 0.20 from age 3, for the four-group model (−9394.62) is lower compared decreasingto0.10fromage 5and remainingfairly sta- to the three (−9398.60) and five (−9409.03) group ble until age 14, when there was a slight increase to models. The mean posterior probability scores ranged almost 0.20. In the high group, a small percentage of from0.72to0.88for thefour-trajectorymodel,with a girls (6.5% estimated; 6.1% actual) showed a relatively mean of 0.80, indicating that most boys fit their assigned 710 J Abnorm Child Psychol (2020) 48:703–718 Fig. 1 Trajectory groups of internalizing problems, both boys and girls. Note. Shown are estimated trajectories (lines), observed group means at each age (markers) and estimated group percentages trajectory well. Figure 3 depicts the probability of clini- showing a low probability from age 3, increasing sharply cally relevant internalizing problems for the four trajecto- from ages 7 to 11, and levelling off to around 30% from ry groups in boys from ages 3 to 14, along with the esti- age 11. The high group (11.3% estimated; 10.5 actual) mated percentage in each group. The predicted and ob- displayed a high probability from ages 3 to 14 (around served values had a high level of correspondence, indicat- 50%), showing a steady increase up to age 7, then a slight ing a good fit of the model. The low problem group decline from ages 11 to 14. (59.1% estimated; 60.05% actual) showed an almost zero probability of internalizing problems from ages 3 to 14. Early Predictors and Adolescent Outcomes There was a decreasing group (12.6% estimated; 12.3% actual), which displays a high probability at age 3 (close Table 2 presents the mean differences in early risk factors and to 40%), declining sharply to near zero by age 7 and adolescent- and parent-reported outcomes among trajectory remaining low thereafter. There was also a moderately, groups for girls. Girls in the high probability group generally increasing group (17.1% estimated; 17.7% actual) showed more early risks than girls in the low group, with the Fig. 2 Girls’ trajectory groups of internalizing problems. Note. Shown are estimated trajectories (lines), observed group means at each age (markers) and estimated group percentages J Abnorm Child Psychol (2020) 48:703–718 711 Fig. 3 Boys’ trajectory groups of internalizing problems. Note. Shown are estimated trajectories (lines), observed group means at each age (markers) and estimated group percentages increasing and moderate groups showing intermediate levels high or increasing groups were more likely to report cig- of some early risks. There were no significant differences arette use, self-harm, high BMI, low mental wellbeing, among the groups for having a teenage mother and paternal and low educationalmotivationcomparedtoboysinthe psychopathology. The moderate group was disproportionally low group and depressive symptoms compared to boys from BME backgrounds compared to the low and increasing in the low or decreasing groups. The increasing group groups. For the adolescent outcomes, girls in the increasing reported more arguments with their mother than the low andhighgroupsweremorelikelytoreportself-harm,lower group. Parents of boys in the increasing or high groups mental wellbeing, more depressive symptoms, lower educa- reported that their sons showed more conduct problems, tional motivation, and more arguments with their mother com- peer problems, and hyperactivity compared to those in the pared to girls in the low or moderate groups, and high BMI low or moderate groups. There were no significant differ- compared to girls in the low group. Parents of girls in the ences among the groups in alcohol use, smoking cannabis, increasing or high groups reported that their daughters showed social media use, sexual activity, and arguing with their more conduct problems, peer problems, and hyperactivity father. compared to parents of girls in the low or moderate groups, while parents of girls in the moderate group reported that their daughters had more peer problems compared to those in the Discussion low group. Girls in the increasing group further reported more early sexual activity than girls following the low or moderate There is a dearth of recent research examining gender pathways, more cigarette and cannabis use than girls follow- differences in pathways of internalizing problems from ing the low pathway, and more alcoholic use than girls follow- early childhood to adolescence. An understanding of clin- ing the moderate pathway. There were no significant differ- ically meaningful pathways for boys and girls born ences among the groups in early menarche, social media use, around the millennium is important for intervention pur- or arguments with their father. poses, in order to target high risk children during critical Table 3 presents the mean differences in early risk fac- points in their development. Using evidence from a cur- tors and adolescent- and parent-reported outcomes among rent, nationally representative UK cohort study, following trajectory groups for boys. Boys in the high group gener- the lives of over 17,000 children born in 2000/2, this ally showed more early risks than the low group, with the study identifies distinct trajectories of internalizing prob- increasing and decreasing groups showing moderate early lems for boys and girls from ages 3 to 14 years. Although risks, for the most part. There was an overrepresentation initial findings revealed three pathways of internalizing of boys from BME backgrounds in the high and decreas- problems when both genders were examined together, sig- ing groups. No significant differences were shown for nificant gender differences were shown in the intercepts having a teenage mother and paternal psychopathology. and slopes of the high and increasing trajectories. When In terms of adolescent-reported outcomes, boys in the examining boys and girls separately, four trajectories were 712 J Abnorm Child Psychol (2020) 48:703–718 Table 2 Mean differences in early predictors and adolescent outcomes by trajectory group for girls Variables Trajectory group F-test Low Increasing Moderate High Mean SD Mean SD Mean SD Mean SD Early predictors a a b ab BME background 0.11 0.32 0.10 0.30 0.19 0.39 0.14 0.35 F(3, 387) = 4.38** Teenage mother 0.03 0.16 0.03 0.17 0.04 0.19 0.04 0.21 F(3, 387) = 1.20 a b ab ab Low birth weight 0.07 0.25 0.11 0.31 0.09 0.29 0.07 0.26 F(3, 387) = 4.55** a ab bc c Single parent 0.13 0.34 0.15 0.35 0.18 0.39 0.25 0.43 F(3, 387) = 9.17*** a b ab b Parent low qualifications 0.17 0.37 0.21 0.41 0.21 0.41 0.28 0.45 F(3, 387) = 8.96*** a b c c Income quintile 3.10 1.41 2.83 1.38 2.47 1.38 2.48 1.33 F(3, 387) = 38.14*** a ab b c Social housing 0.22 0.41 0.26 0.44 0.32 0.47 0.42 0.49 F(3, 387) = 18.98*** a b b b Maternal smoking in pregnancy 0.19 0.39 0.24 0.43 0.27 0.44 0.30 0.46 F(3, 387) = 7.57*** a b bc c Maternal depressive symptoms 0.05 0.23 0.12 0.33 0.14 0.35 0.20 0.40 F(3, 387) = 23.96*** Paternal depressive symptoms 0.05 0.21 0.07 0.25 0.08 0.28 0.10 0.30 F(3, 387) = 3.56 Self-reported adolescent outcomes Early menarche (before 11 years old) 0.23 0.42 0.24 0.43 0.27 0.45 0.27 0.45 F(3, 387) = 1.18 ab a b ab How many times had an alcoholic drink 0.86 1.25 0.94 1.27 0.68 1.12 0.82 1.22 F(3, 387) = 3.41** a b ab ab How often do you smoke cigarettes 0.33 0.93 0.60 1.26 0.41 1.01 0.46 0.98 F(3, 387) = 6.03*** a b ab ab How often smoke cannabis 0.08 0.46 0.20 0.65 0.10 0.44 0.10 0.55 F(3, 387) = 3.28** a b a b Ever self-harmed 0.18 0.39 0.37 0.48 0.17 0.38 0.36 0.48 F(3, 387) = 24.03*** a b ab b High BMI 0.14 0.35 0.20 0.40 0.16 0.37 0.23 0.42 F(3, 387) = 4.91** Hours spent on social networks 4.08 1.95 4.01 2.40 3.82 2.23 4.22 2.13 F(3, 387) = 1.38 a b a ab Engagedinsexual activity 0.03 0.16 0.06 0.24 0.02 0.15 0.03 0.16 F(3, 387) = 2.98** a b a b Mental wellbeing 5.39 1.09 4.76 1.28 5.31 1.20 4.84 1.20 F(3, 387) = 53.64*** a b a b Depressive symptoms 1.49 0.48 1.79 0.58 1.46 0.46 1.73 0.54 F(3, 387) = 49.15*** a b a b Low educational motivation 2.91 0.51 2.69 0.55 2.87 0.55 2.69 0.56 F(3, 387) = 33.75*** a b a b How often argue with mother 2.86 1.03 3.09 1.10 2.79 1.10 3.16 1.17 F(3, 387) = 10.75*** How often argue with father 2.43 0.99 2.56 1.15 2.36 1.08 2.57 1.22 F(3, 387) = 2.57 Parent-reported adolescent outcomes a b a b Conduct problems 0.07 0.25 0.23 0.42 0.13 0.34 0.28 0.45 F(3, 387) = 37.54*** a b c b Peer problems 0.08 0.28 0.38 0.49 0.19 0.39 0.45 0.50 F(3, 387) = 92.70*** a b a b Hyperactivity 0.04 0.20 0.18 0.38 0.08 0.27 0.25 0.43 F(3, 387) = 37.17*** F-tests and post-hoc analysis were conducted using svy in STATA, which reports design degrees of freedom for a complex clustered and stratified survey design. Post-hoc analyses using Bonferroni’s method identified significant pairwise comparisons (p < 0.05) between groups, shown when group means do not share any similar superscripts. *p <0.05, ** p < 0.01, ***p <0.001 identified including two relatively stable trajectories and low educational motivation than the low group. showing either high or low probabilities of internalizing Girls, but not boys, on the increasing trajectory also re- problems. An increasing trajectory was also found for ported more cigarette and cannabis use and early sexual both boys and girls, showing an increasing probability activity at age 14 than girls following the low pathway. of internalizing problems which continued to rise for girls, These findings suggest that the course of internalizing but levelled off for boys from age 11. A decreasing tra- problems varies for boys and girls with distinct manifes- jectory was revealed for boys, while a moderate but stable tations of risk. trajectory was identified for girls. Significant early risk factors and adolescent outcomes differed among the tra- Trajectories of Internalizing Problems jectory groups. Boys and girls in the increasing and high probability groups were more likely to report high BMI, As other trajectory-group studies of internalizing problems self-harm, low mental wellbeing, depressive symptoms, have shown (Fanti and Henrich 2010; Korhonen et al., 2014; J Abnorm Child Psychol (2020) 48:703–718 713 Table 3 Mean differences in early predictors and adolescent outcomes by trajectory group for boys Variables Trajectory group F-test Low Increasing Decreasing High Mean SD Mean SD Mean SD Mean SD Early predictors a a b b BME background 0.11 0.31 0.10 0.31 0.17 0.37 0.16 0.36 F(3, 387) = 5.30** Teenage mother 0.03 0.16 0.03 0.18 0.07 0.25 0.04 0.20 F(3, 387) = 2.72 a a ab b Low birth weight 0.06 0.24 0.06 0.24 0.07 0.26 0.12 0.32 F(3, 387) = 5.30** a a ab b Single parent 0.14 0.34 0.15 0.36 0.18 0.38 0.25 0.43 F(3, 387) = 11.65*** a ab b b Parent low qualifications 0.18 0.38 0.20 0.40 0.26 0.44 0.26 0.44 F(3, 387) = 6.62*** a ab b c Income quintile 3.07 1.40 2.94 1.45 2.69 1.39 2.35 1.32 F(3, 387) = 41.72*** a b bc c Social housing 0.21 0.41 0.28 0.45 0.31 0.46 0.38 0.48 F(3, 387) = 20.95*** a ab ab b Maternal smoking in pregnancy 0.21 0.41 0.26 0.44 0.27 0.44 0.31 0.46 F(3, 387) = 7.11*** a b bc c Maternal depressive symptoms 0.06 0.24 0.11 0.31 0.15 0.36 0.17 0.38 F(3, 387) = 24.36*** Paternal depressive symptoms 0.05 0.21 0.05 0.22 0.07 0.26 0.06 0.24 F(3, 387) = 1.30 Self-reported adolescent outcomes How many times had an alcoholic drink 0.92 1.30 0.89 1.44 0.86 1.29 0.78 1.26 F(3, 387) = 1.12 a b ab b How often do you smoke 0.25 0.76 0.43 1.03 0.45 1.12 0.55 1.28 F(3, 387) = 6.99*** How often smoke cannabis 0.10 0.53 0.15 0.59 0.13 0.60 0.23 0.78 F(3, 387) = 1.93 a b ab b Ever self-harmed 0.08 0.27 0.16 0.36 0.08 0.27 0.14 0.35 F(3, 387) = 6.90*** a b ab b High BMI 0.14 0.35 0.25 0.43 0.15 0.36 0.22 0.41 F(3, 387) = 7.02*** Hours spent on social networks 3.06 1.98 2.96 2.12 2.95 1.97 2.70 3.29 F(3, 387) = 0.71 Engaged in sexual activity 0.03 0.17 0.04 0.21 0.03 0.18 0.04 0.20 F(3, 387) = 0.38 a b ab b Mental wellbeing 5.68 1.02 5.38 1.05 5.53 1.02 5.39 1.09 F(3, 387) = 14.72*** a b a b Depressive symptoms 1.31 0.36 1.45 0.43 1.30 0.33 1.46 0.44 F(3, 387) = 17.34*** a b ab b Low educational motivation 2.96 0.45 2.84 0.51 2.91 0.43 2.83 0.53 F(3, 387) = 10.28*** a b ab ab How often argue with mother 2.72 1.02 2.98 1.04 2.74 1.06 2.75 1.21 F(3, 387) = 5.73*** How often argue with father 2.42 1.01 2.60 1.09 2.36 1.02 2.41 1.31 F(3, 387) = 2.95 Parent-reported adolescent outcomes a b a b Conduct problems 0.10 0.30 0.33 0.47 0.12 0.33 0.39 0.49 F(3, 387) = 54.72*** a b a b Peer problems 0.13 0.33 0.47 0.50 0.15 0.36 0.51 0.50 F(3, 387) = 87.61*** a b a b Hyperactivity 0.10 0.30 0.34 0.47 0.12 0.32 0.36 0.48 F(3, 387) = 50.69*** F-tests and post-hoc analysis were conducted using svy in STATA, which reports design degrees of freedom for a complex clustered and stratified survey design. Post-hoc analyses using Bonferroni’s method identified significant pairwise comparisons (p < 0.05) between groups, shown when group means do not share any similar superscripts. *p <0.05, ** p < 0.01, ***p <0.001 Letcher et al. 2009;Nivard et al., 2017; Sterba et al. 2007), whichexaminedtrajectoriesupto age 11 (Sterba etal. findings revealed both high and low problem groups. As ex- 2007). As these data extend from early childhood to ado- pected, the low-problem group had a slightly higher preva- lescence, they may be better able to capture the nuances lence of boys than girls (59% compared to 55%). In line with of these diverse pathways, as well as identify when gen- Sterba et al. (2007), a high problem group was revealed der differences emerge in development. What these data for both genders, showing an early-onset in childhood. demonstrate are a group of males and females, with a high Although Sterba et al. (2007) found higher prevalence and persistent probability of internalizing problems from rates for females than males in the high group, this study an early age. Males are especially at high risk of being in found the opposite. Unexpectedly, the prevalence rate was this group, which may represent the preponderance of higher for boys than girls (11.3% versus 6.5%) in the high males in this cohort with special educational needs and group. This difference may be due to the longer age range co-morbid mental health problems, more generally of the current study, in comparison to the earlier study, (Gutman et al. 2015). 714 J Abnorm Child Psychol (2020) 48:703–718 Unlike other studies which found a higher prevalence of Henrich 2010;Nivard et al., 2017;Sterba et al. 2007), both girls in the increasing group (Nivard et al., 2017), this study genders on the high pathway experienced more early risks, found that boys and girls had a similar prevalence in a clini- including parents with lower education and income, living in cally meaningful increasing pathway (17.1% and 16.6%, re- social housing and with a single parent, and having a mother spectively), but each showed a somewhat different trajectory who smoked in pregnancy and reported more post-natal de- shape. From a near zero probability, girls in this group showed pressive symptoms than those in the low group. Boys in the an onset at age 5, increasing to almost 60% at age 14; whereas high group were also more likely to have a low birthweight boys in this group showed a later onset at age 7, increasing to and BME background than boys in the low group. Both boys 30% by age 14. Thus, adolescent girls showed almost twice and girls in this group had worse adolescent outcomes, includ- the likelihood of having severe internalizing problems com- ing a high BMI, self-harm, low mental wellbeing, more de- pared to boys in this group. Similarly, Sterba et al. (2007) pressive symptoms, and low educational motivation com- found that the increasing group of girls reached a higher level pared to those on the low pathway, highlighting of internalizing problems, in comparison to the same group of the educational, mental, and physical health risks for this males at age 11. For both genders, the increase shown at age group. Parents also reported higher probabilities of adolescent 11 likely coincides with the onset of puberty. For girls, the conduct problems, peer problems, and hyperactivity than the probability of severe internalizing problems continued to rise, low or decreasing pathways. There were a few gender differ- reaching levels close to the high group by age 14. For boys, ences. Boys were more likely to report smoking cigarettes, the probability seemed to level off around age 11. This sug- while girls were more likely to report arguing with their moth- gests that boys in this group show increasing but moderate er, but not their father, than the low group. Nevertheless, un- vulnerability to internalizing problems, coinciding with the like studies examining pathways of depressive symptoms transition into secondary school. Girls, on the other hand, (Costello et al. 2008; Danzo et al. 2017; Skogen et al. 2016), may become more susceptible to internalizing problems in the high group did not report drinking more alcohol or using mid-adolescence in line with recent data (Mental Health of more cannabis than the low group. Children and Young People in England, 2018), culminating Similar to previous studies, the increasing group were more in high-risk group of adolescent girls. likely to have mothers with post-natal depressive symptoms The findings revealed a decreasing group for males, show- than the low group (Nivard et al., 2017;Sterba etal. 2007). ing a high probability of internalizing problems, close to 40% Boys on this pathway were also more likely to live in social at age 3, which plunged to near zero levels thereafter. This housing, while girls on the increasing trajectory were more suggests that there is a group of males who show severe inter- likely to experience social disadvantage, in terms of low pa- nalizing problems early in childhood, maturing out of these rental income and educational qualifications, have mothers internalizing difficulties once they reach school age. As who smoked during pregnancy, and have a low birthweight discussed below, this group showed no evidence of higher compared to the low group. Both boys and girls on the in- externalizing behaviours in adolescence compared to the low creasing pathway reported worse adolescent outcomes than group. Girls presented a moderate group, where they began those on the low pathway, including high BMI, self-harm, with a moderate probability, showing a mild dip in childhood, low mental well-being, more depressive symptoms, and low with a slight increase from ages 11 to 14, coinciding with the educational motivation; while parents reported higher proba- pubertal transition. These girls show moderate probability of bilities of conduct problems, peer problems, and hyperactivity internalizing problems throughout childhood and adoles- compared to the low or decreasing/moderate pathways. Boys cence, hovering between 10% and 20%. This trajectory is on this trajectory reported greater conflict with their mothers similar to the decreasing/increasing trajectory shown in than the low group, while girls reported more cigarette and Sterba et al. (2007), which was hypothesised to be more sen- cannabis use than the low group, more early sexual activity sitive to environmental stressors and sensitive periods than than the low or moderate groups, and more alcoholic use than those in the elevated, stable trajectory. In contrast to previous the moderate group. These findings contribute to our under- studies suggesting that gender differences in internalizing standing of the possible gender differences in both the etiolo- problems begin in adolescence (Leve et al. 2005), these find- gy and outcomes of the increasing pathway, indicating that ings indicate that gender differences may emerge for distinct girls on the increasing trajectory are not only distinguished trajectories in early childhood, in addition to those surfacing in by having greater early social disadvantage compared to boys, adolescence. but are also more vulnerable to poor behavioural outcomes in adolescence, which are likely to cascade into future difficulties Early Predictors and Adolescent Outcomes (Haller et al. 2010). The decreasing group, for boys, and the moderate group, for Early predictors and later adolescent outcomes distinguished girls, were more socially disadvantaged in terms of parental income and living in social housing, and were more likely to these trajectories. As other studies have shown (Fanti and J Abnorm Child Psychol (2020) 48:703–718 715 have mothers who reported post-natal depressive symptoms defined internalizing construct may be more stable over time than the low group. For girls, the moderate group was also more than more distinct variations within this domain, such as sep- likely to have a single parent and mother who smoked during aration anxiety and social anxiety, which may show different pregnancy, while the decreasing group, for boys, was more patterns of change over the course of development, with po- likely to live in a household with low educational qualifications. tential variation between genders (Carter et al. 2010; These two groups also included a higher proportion of BME McLaughlin and King 2015). Third, as in all longitudinal children compared to the low or increasing trajectories. Few studies, there was the problem of missingness in the data studies have examined the role of ethnicity in predicting inter- due to non-response for certain items or for a whole wave of nalizing trajectories from early childhood to adolescence, espe- data collection. This problem was addressed using MCS attri- cially with an ethnically diverse population sample, so we have tion weights and FIML estimation as implemented in STATA little information on how this finding might compare to previ- to adjust the likelihood function so that each case contributes ous studies. Given their relatively low levels of internalizing information on the variables that are observed. Fourth, group- problems in adolescence, both of these groups were similar to based trajectory analysis only provides a descriptive summary the low group in terms of the adolescent-reported outcomes. of a potential underlying typology in pathways. The fit indi- Parents of girls, however, reported that their daughters had a cators provide some guidelines about the number of types to higher probability of peer problems than the low group, which select, and the final selection is based on consideration of may highlight difficulties with social relationships. parsimony, interpretability, BIC statistics, and average poste- In line with recent research (Booker et al. 2018), this study rior probability of group membership. Individuals are dis- found that girls were more likely to use social media. However, cretely assigned to the best-fitting subgroup, despite some social media use was not linked to trajectories of internalizing degree of imprecision in group membership. Lastly, only a problems for any of the groups, for either gender. This finding subset of adolescent-reported outcomes was assessed and out- may reflect recent research showing that moderate social media comes in late adolescence and adulthood are not yet available. use does not predict changes in depressive symptoms, but rather increasing, excessive screen and media use relates to increasing depressive symptoms, highlighting that this relationship may be Conclusions bidirectional (Houghton et al. 2018; Raudsepp and Kais 2019). Specific technology-based behaviours, such as social compari- This study offers insights into the development of internaliz- son and feedback seeking, have also been shown to be associ- ing problems for children and adolescents born in the new ated with depressive symptoms, suggesting a more nuanced millennia. For boys and girls, there are two developmental approach to the study of adolescents’ media use (Nesi and trajectories demonstrating a high risk of clinically meaningful Prinstein 2015). Early menarche was also not a risk factor for internalizing problems: a high pathway, exhibiting a high girls, supporting recent research showing that menarche status probability of internalizing problems from early childhood to is not associated with worsening depression (McGuire et al. adolescence and an increasing pathway, showing a heightened 2019). Rather, increases in depressive symptoms seem to be probability of internalizing problems before and during the associated with physical changes that emerge early in the pu- pubertal transition. Given the recent attention placed on the bertal transition for early maturing girls, along with anticipatory internalizing problems of girls, one notable finding is the ele- concerns about social rejection. vated percentage of boys on the high pathway, which is 1.75 times greater than the percentage of girls on a similar trajec- tory. Of further importance is the apparent increasing risk of Limitations internalizing problems for girls in adolescence, while this risk seems to level off for boys. Most concerning are the high There are a number of limitations to consider. First, internal- levels of mental and physical health problems facing those izing problems were assessed on parental reports only, raising on the high or increasing pathways. For example, in each of the problem of informant and methodological biases. It is also these two groups, more than one-third of the girls reported possible that parental reporting differed based on the gender of engaging in self-harm, which is twice the proportion of girls the child, contributing to potential biases. Second, the extent in the low or moderate groups, and approximately one-quarter of our analyses is limited by the measures included in the of the boys and girls are overweight or obese in each of these multi-purpose longitudinal survey of a national cohort, many two groups compared to about 15% in the lower problem of which relied on a parsimonious measurement strategy. The groups. Girls on the increasing pathway reported an especially use of SDQ, as a clinical screening tool, may also be a limi- alarming level of problematic behaviours in adolescence in- tation. Although the SDQ (Goodman et al. 2000)ispredictive cluding early sexual activity and more cannabis and cigarette of depression and other internalizing diagnoses, the trajecto- use compared to the low group, confirming the “gender para- ries themselves are not clinical. Furthermore, this broadly dox of co-morbidity” (Loeber and Keenan 1994). Overall, 716 J Abnorm Child Psychol (2020) 48:703–718 and Psychopathology,1–50. https://doi.org/10.1002/ these findings highlight that intervention strategies take a sys- 9781119125556.devpsy409. temic view, targeting not only internalizing emotions, but also Carter, A. S., Godoy, L., Wagmiller, R. L., Veliz, P., Marakovitz, S., & behaviours associated with health and well-being to circum- Briggs-Gowan, M. J. (2010). Internalizing trajectories in young vent the possibility of negative outcomes emerging in later boys and girls: the whole is not a simple sum of its parts. Journal of Abnormal Child Psychology, 38(1), 19–31. adolescence and adulthood. Chaiton, M. O., Cohen, J. E., O'Loughlin, J., & Rehm, J. (2009). A systematic review of longitudinal studies on the association between Compliance with Ethical Standards depression and smoking in adolescents. BMC Public Health, 9(1), 356–367. https://doi.org/10.1186/1471-2458-9-356. Conflict of Interest The authors declare that there is no conflict of Cicchetti, D., & Rogosch, F. A. (2002). A developmental psychopathol- interest. ogy perspective on adolescence. Journal of Consulting and Clinical Psychology, 70(1), 6–20. https://doi.org/10.1037/0022-006X.70.1. Ethical Approval The surveys were granted ethical clearance for the 6. Millennium Cohort Study by National Health Service Multi-Centre Collishaw, S., Maughan, B., Natarajan, L., & Pickles, A. (2010). Trends Research Ethics Committees (MREC). For MCS1 this was the committee in adolescent emotional problems in England: a comparison of two based in the South West, for MCS2 and MCS3 the London MREC and national cohorts twenty years apart. Journal of Child Psychology for MCS4 and MCS5, the Northern and Yorkshire MREC. and Psychiatry, 51(8), 885–894. https://doi.org/10.1111/j.1469- 7610.2010.02252.x. Costello, D. M., Swendsen, J., Rose, J. S., & Dierker, L. C. (2008). Risk Consent to Participate Informed consent was obtained from all individ- and protective factors associated with trajectories of depressed mood ual participants included in the study. from adolescence to early adulthood. Journal of Consulting and Open Access This article is licensed under a Creative Commons Clinical Psychology, 76(2), 173–183. https://doi.org/10.1037/ Attribution 4.0 International License, which permits use, sharing, adap- 0022-006X.76.2.173. tation, distribution and reproduction in any medium or format, as long as Croft, S., Stride, C., Maughan, B., & Rowe, R. (2015). Validity of the you give appropriate credit to the original author(s) and the source, pro- strengths and difficulties questionnaire in preschool-aged children. vide a link to the Creative Commons licence, and indicate if changes were Pediatrics, 135(5), e1210–e1219. https://doi.org/10.1542/peds. made. The images or other third party material in this article are included 2014-2920. in the article's Creative Commons licence, unless indicated otherwise in a Danzo, S., Connell, A. M., & Stormshak, E. A. (2017). Associations credit line to the material. If material is not included in the article's between alcohol-use and depression symptoms in adolescence: ex- Creative Commons licence and your intended use is not permitted by amining gender differences and pathways over time. Journal of statutory regulation or exceeds the permitted use, you will need to obtain Adolescence, 56,64–74. https://doi.org/10.1016/j.adolescence. permission directly from the copyright holder. To view a copy of this 2017.01.007. licence, visit http://creativecommons.org/licenses/by/4.0/. Dekker, M. C., Ferdinand, R. F., Van Lang, N. D., Bongers, I. L., Van Der Ende, J., & Verhulst, F. C. (2007). Developmental trajectories of depressive symptoms from early childhood to late adolescence. Journal of Child Psychology and Psychiatry, 48,657–666. https:// doi.org/10.1111/j.1469-7610.2007.01742.x_2007. Dockray, S., Susman, E. J., & Dorn, L. D. (2009). Depression, cortisol References reactivity, and obesity in childhood and adolescence. Journal of Adolescent Health, 45(4), 344–350. https://doi.org/10.1016/j. Achenbach, T. M. (1991). Integrative guide for the 1991 CBCL/4–18, jadohealth.2009.06.014. YSR, and TRF profiles. Department of Psychiatry, University of Fanti, K. A., & Henrich, C. C. (2010). Trajectories of pure and co- Vermont. occurring internalizing and externalizing problems from age 2 to Angold, A., Costello, E. J., Messer, S. C., Pickles, A., Winder, F., & age 12: findings from the National Institute of Child Health and Silver, D. (1995). The development of a short questionnaire for Human Development study of early child care. Developmental use in epidemiological studies of depression in children and adoles- Psychology, 46(5), 1159–1175. cents. International Journal of Methods in Psychiatric Research, 5, Fink, E., Patalay, P., Sharpe, H., Holley, S., Deighton, J., & Wolpert, M. 237–249. (2015). Mental health difficulties in early adolescence: a comparison Barry, C. T., Sidoti, C. L., Briggs, S. M., Reiter, S. R., & Lindsey, R. A. of two cross-sectional studies in England from 2009 to 2014. (2017). Adolescent social media use and mental health from adoles- Journal of Adolescent Health, 56(5), 502–507. https://doi.org/10. cent and parent perspectives. Journal of Adolescence, 61,1–11. 1016/j.jadohealth.2015.01.023. Bauer, D. J., & Curran, P. J. (2003). Distributional assumptions of growth Goodman, R. (1997). The strengths and difficulties questionnaire (SDQ). mixture models: implications for overextraction of latent trajectory Journal of Child Psychology and Psychiatry, 38,581–586. classes. Psychological methods, 8(3), 338–363. Goodman, R. (2001). Psychometric properties of the strengths and diffi- Best, P., Manktelow, R., & Taylor, B. (2014). Online communication, culties questionnaire. Journal of the American Academy of Child & social media and adolescent wellbeing: a systematic narrative re- Adolescent Psychiatry, 40(11), 1337–1345. https://doi.org/10.1097/ view. Children and Youth Services Review, 41,27–36. https://doi. 00004583-200111000-00015. org/10.1016/j.childyouth.2014.03.001. Goodman, R., Ford, T., Simmons, H., Gatward, R., & Meltzer, H. (2000). Booker, C. L., Kelly, Y. J., & Sacker, A. (2018). Gender differences in the Using the strengths and difficulties questionnaire (SDQ) to screen associations between age trends of social media interaction and well- for child psychiatric disorders in a community sample. British being among 10-15 year olds in the UK. BMC Public Health, 18(1), Journal of Psychiatry, 177, 534–539. https://doi.org/10.1192/bjp. 321–323. https://doi.org/10.1186/s12889-018-5220-4. 177.6.534. Burt, K. B., Douglas Coatsworth, J., & Masten, A. S. (2016). Green, H., McGinnity, A., Meltzer, H., Ford, T., & Goodman, R. (2005). Competence and psychopathology in development. Development Mental health of children and young people in Great Britain, 2004. J Abnorm Child Psychol (2020) 48:703–718 717 Gutman, L. M., Joshi, H., Parsonage, M., & Schoon, I. (2015). Children McCrae, N., Gettings, S., & Purssell, E. (2017). Social media and depres- of the new century: Mental health findings from the millennium sive symptoms in childhood and adolescence: a systematic review. cohort study. London: Centre for Mental Health. Adolescent Research Review, 2(4), 315–330. https://doi.org/10. 1007/s40894-017-0053-4. Gutman, L. M., Joshi, H., Parsonage, M., & Schoon, I. (2018a). Trends in parent-and teacher-rated mental health problems among 10-and 11- McGuire, T. C., McCormick, K., Koch, M. K., & Mendle, J. (2019). year-olds in Great Britain: 1999–2012. Child and Adolescent Mental Pubertal maturation and longitudinal trajectories of depression dur- Health, 23(1), 26–33. https://doi.org/10.1111/camh.12179. ing early adolescence. Frontiers in Psychology, 10, 1362. Gutman, L. M., Joshi, H., Parsonage, M., & Schoon, I. (2018b). Gender- McLaughlin, K. A., & King, K. (2015). Developmental trajectories of specific trajectories of conduct problems from ages 3 to 11. Journal anxiety and depression in early adolescence. Journal of Abnormal of Abnormal Child Psychology, 46(7), 1467–1480. Child Psychology, 43(2), 311–323. Gutman, L. M., Joshi, H., & Schoon, I. (2019). Developmental trajecto- McLeod, G. F. H., Horwood, L. J., & Fergusson, D. M. (2016). ries of conduct problems and cumulative risk from early childhood Adolescent depression, adult mental health and psychosocial out- to adolescence. Journal of Youth and Adolescence, 48(2), 181–198. comes at 30 and 35 years. Psychological Medicine, 46(7), 1401– Haller, M., Handley, E., Chassin, L., & Bountress, K. (2010). 1412. https://doi.org/10.1017/S0033291715002950. Developmental cascades: linking adolescent substance use, affilia- Measelle, J. R., Stice, E., & Hogansen, J. M. (2006). Developmental tion with substance use promoting peers, and academic achievement trajectories of co-occurring depressive, eating, antisocial, and sub- to adult substance use disorders. Development and stance abuse problems in female adolescents. Journal of Abnormal Psychopathology, 22(4), 899–916. Psychology, 115(3), 524–538. https://doi.org/10.1037/0021-843X. Hansen, K. (2014). Millennium cohort study: A guide to the data sets. 115.3.524. London: UCL Institute of Education http://doc.ukdataservice.ac.uk/ Meltzer, H., Gatward, R., Goodman, R., & Ford, T. (2000). The mental doc/7464/mrdoc/pdf/mcs_guide_to_the_datasets_020214.pdf. health of children and adolescents in Great Britain.London: HM Hyde, L. W., Burt, S. A., Shaw, D. S., Donnellan, M. B., & Forbes, E. E. Stationery Office. (2015). Early starting, aggressive, and/or callous–unemotional? Moffitt, T. E., Harrington, H., Caspi, A., Kim-Cohen, J., Goldberg, D., Examining the overlap and predictive utility of antisocial behavior Gregory, A. M., & Poulton, R. (2007). Depression and generalized subtypes. Journal of Abnormal Psychology, 124(2), 329. anxiety disorder: cumulative and sequential comorbidity in a birth Houghton, S., Lawrence, D., Hunter, S. C., Rosenberg, M., Zadow, C., cohort followed prospectively to age 32 years. Archives of General Wood, L., & Shilton, T. (2018). Reciprocal relationships between Psychiatry, 64(6), 651–660. trajectories of depressive symptoms and screen media use during Nagin, D. (2005). Group-based modeling of development. Cambridge: adolescence. Journal of Youth and Adolescence, 47(11), 2453–2467. Harvard Univ. Press. Johnson, J., Atkinson, M., & Rosenberg, R. (2015). Millennium Cohort Nagin, D. S., & Odgers, C. (2010). Group-based trajectory modeling in Study: Psychological Developmental and Health Inventories Centre clinical research. Annual Review Clinical Psychology, 6,109–138. for Longitudinal Studies. http://www.cls.ioe.ac.uk/librarymedia% Negriff, S., & Susman, E. J. (2011). Pubertal timing, depression, and 5Cdocuments%5CGuide%20to%20Psychological% externalizing problems: a framework, review, and examination of 20Inventories%20in%20MCS%282%29.pdf gender differences. Journal of Research on Adolescence, 21(3), Jones, B. L., & Nagin, D. S. (2013). A note on a Stata plugin for estimat- 717–746. https://doi.org/10.1111/j.1532-7795.2010.00708.x. ing group-based trajectory models. Sociological Methods & Nesi, J., & Prinstein, M. J. (2015). Using social media for social compar- Research, 42(4), 608 –613. https://doi.org/10.1177/ ison and feedback-seeking: gender and popularity moderate associ- ations with depressive symptoms. Journal of Abnormal Child Joshi, H., & Fitzsimons, E. (2016). The millennium cohort study: the Psychology, 43(8), 1427–1438. making of a multi-purpose resource for social science and policy. Nivard, M. G., Lubke, G. H., Dolan, C. V., Evans, D. M., Pourcain, B. S., Longitudinal and Life Course Studies, 7(4), 409–430. https://doi. Munafò, M. R., & Middeldorp, C. M. (2017). Joint developmental org/10.14301/llcs.v7i4.410. trajectories of internalizing and externalizing disorders between Kelly, Y., Zilanawala, A., Sacker, A., Hiatt, R., & Viner, R. (2016). childhood and adolescence. Development and Psychopathology, Report: early puberty in 11-year-old girls: millennium cohort study 29(3), 919–928. findings. Children & Young People Now, 23,35–35. https://doi.org/ Nolen-Hoeksema, S., & Girgus, J. S. (1994). The emergence of gender 10.1136/archdischild-2016-310475. differences in depression during adolescence. Psychological Korhonen, M., Luoma, I., Salmelin, R. K., Helminen, M., Kaltiala-Heino, Bulletin, 115(3), 424. R., & Tamminen, T. (2014). The trajectories of child’s internalizing Patton, G. C., Olsson, C., Bond, L., Toumbourou, J. W., Carlin, J. B., and externalizing problems, social competence and adolescent self- Hemphill, S. A., & Catalano, R. F. (2008). Predicting female depres- reported problems in a Finnish normal population sample. School sion across puberty: a two-nation longitudinal study. Journalofthe Psychology International, 35(6), 561–579. American Academy of Child & Adolescent Psychiatry, 47(12), Letcher, P., Smart, D., Sanson, A., & Toumbourou, J. W. (2009). 1424–1432. https://doi.org/10.1097/CHI.0b013e3181886ebe. Psychosocial precursors and correlates of differing internalizing tra- Public Health England. (2016). The mental health of children in England. jectories from 3 to 15 years 1. Social Development, 18(3), 618–646. London: Public Health England https://assets.publishing.service. Leve, L. D., Kim, H. K., & Pears, K. C. (2005). Childhood temperament gov.uk/government/uploads/system/uploads/attachment_data/file/ and family environment as predictors of internalizing and external- 575632/Mental_health_of_children_in_England.pdf. izing trajectories from ages 5 to 17. Journal of Abnormal Child Raudsepp, L., & Kais, K. (2019). Longitudinal associations between Psychology, 33(5), 505–520. problematic social media use and depressive symptoms in adoles- Lewis, A. J., Kremer, P., Douglas, K., Toumborou, J. W., Hameed, M. A., cent girls. Preventive Medicine Reports, 100925. Patton, G. C., & Williams, J. (2015). Gender differences in adoles- Richardson, L. P., Garrison, M. M., Drangsholt, M., Mancl, L., & cent depression: differential female susceptibility to stressors affect- ing family functioning. Australian Journal of Psychology, 67(3), LeResche, L. (2006). Associations between depressive symptoms and obesity during puberty. General Hospital Psychiatry, 28(4), 131–139. https://doi.org/10.1111/ajpy.12086. 313–320. https://doi.org/10.1016/j.genhosppsych.2006.03.007. Loeber, R., & Keenan, K. (1994). Interaction between conduct disorder and its comorbid conditions: effects of age and gender. Clinical Rodgers, B., Pickles, A., Power, C., Collishaw, S., & Maughan, B. Psychology Review, 14(6), 497–523. (1999). Validity of the malaise inventory in general population 718 J Abnorm Child Psychol (2020) 48:703–718 samples. Social Psychiatry and Psychiatric Epidemiology, 34(6), Twenge, J. M., Joiner, T. E., Rogers, M. L., & Martin, G. N. (2018). 333–341. Increases in depressive symptoms, suicide-related outcomes, and Rutter, M., Tizard, J., & Whitmore, K. (1970). Education, health and suicide rates among US adolescents after 2010 and links to increased behavior. London: Longmans. new media screen time. Clinical Psychological Science, 6(1), 3–17. Schafer, J. L., & Graham, J. W. (2002). Missing data: our view of the state Ullsperger, J. M., & Nikolas, M. A. (2017). A meta-analytic review of the of the art. Psychological Methods, 7(2), 147–177. https://doi.org/10. association between pubertal timing and psychopathology in ado- 1037//1082-989X.7.2.147. lescence: are there sex differences in risk? Psychological Bulletin, Shore, L.,Toumbourou, J. W.,Lewis,A.J., &Kremer, P. (2018). 143(9), 903–938. https://doi.org/10.1037/bul0000106. Longitudinal trajectories of child and adolescent depressive symp- von Eye, A., & Bergman, L. R. (2003). Research strategies in develop- toms and their predictors–a systematic review and meta-analysis. mental psychopathology: dimensional identity and the person- Child and Adolescent Mental Health, 23(2), 107–120. https://doi. oriented approach. Development and Psychopathology, 15(3), org/10.1111/camh.12220. 553–580. Skogen, J. C., Knudsen, A. K., Hysing, M., Wold, B., & Sivertsen, B. Woods, H. C., & Scott, H. (2016). # Sleepyteens: social media use in (2016). Trajectories of alcohol use and association with symptoms adolescence is associated with poor sleep quality, anxiety, depres- of depression from early to late adolescence: the Norwegian longi- sion and low self-esteem. Journal of Adolescence, 51,41–49. tudinal health behavior study. Drug and Alcohol Review, 35(3), https://doi.org/10.1016/j.adolescence.2016.05.008. 307–316. https://doi.org/10.1111/dar.12350. Zahn-Waxler, C., Klimes-Dougan, B., & Slattery, M. J. (2000). Sterba, S. K., Prinstein, M. J., & Cox, M. J. (2007). Trajectories of inter- Internalizing problems of childhood and adolescence: prospects, nalizing problems across childhood: heterogeneity, external validity, pitfalls, and progress in understanding the development of anxiety and gender differences. Development and Psychopathology, 19(2), and depression. Development and Psychopathology, 12(3), 443– 345–366 10.10170S0954579407070174. Strasburger, V. C., Hogan, M. J., Mulligan, D. A., Ameenuddin, N., Zahn-Waxler, C., Shirtcliff, E. A., & Marceau, K. (2008). Disorders of Christakis, D. A., Cross, C., et al. (2013). Children, adolescents, childhood and adolescence: gender and psychopathology. Annual and the media. Pediatrics, 132(5), 958–961. Review of Clinical Psychology, 4,275–303. https://doi.org/10. Taylor, M. F., Brice, J., Buckand, N., & Prentice-Lane, E. (2010). British 1146/annurev.clinpsy.3.022806.091358. household panel survey user manual volume a: Introduction, tech- nical report and appendices. Colchester: University of Essex. Publisher’sNote Springer Nature remains neutral with regard to jurisdic- Toth, S. L., Petrenko, C. L., Gravener-Davis, J. A., & Handley, E. D. tional claims in published maps and institutional affiliations. (2016). Advances in prevention science: a developmental psycho- pathology perspective. Development and Psychopathology,1–59.
Journal
Journal of Abnormal Child Psychology – Springer Journals
Published: May 10, 2020