Mental disorders in childhood or adolescence and subsequent comprehensive school achievement

Introduction

Childhood is a critical period for mental and emotional development (Malik & Marwaha, 2023), and mental disorders during childhood can have a significant impact on long‐term educational achievements (Askeland et al., 2022; Lawrence, Dawson, Houghton, Goodsell, & Sawyer, 2019; Perry, Dunne, McFadden, & Campbell, 2008; Ringbom et al., 2022). Mental disorders diagnosed in childhood and adolescence, including psychotic disorders (e.g. schizophrenia) (Dickson et al., 2020; Glahn, Bearden, Bowden, & Soares, 2006), internalizing disorders (e.g. depression) (Esch et al., 2014; Maurizi, Grogan‐Kaylor, Granillo, & Delva, 2013; Pagerols et al., 2022), externalizing disorders (e.g. conduct disorder) (Galéra, Melchior, Chastang, Bouvard, & Fombonne, 2009; Liu, Huang, Kao, & Gau, 2017; McLeod, Uemura, & Rohrman, 2012; Sayal, Washbrook, & Propper, 2015), and neurodevelopmental conditions (e.g. autism spectrum disorder) (Galéra et al., 2009; McLeod et al., 2012; Sayal et al., 2015), have been associated with lower school achievement in adolescence. Children with mental disorders also exhibit significantly reduced school attendance rates (Lawrence et al., 2019), have a heightened risk for school expulsion (Perry et al., 2008) and suspension (Kang‐Yi et al., 2018), are more likely to drop out of high school (Askeland et al., 2022; Breslau, Lane, Sampson, & Kessler, 2008), and often experience delayed educational progression (Tempelaar et al., 2014).

Most prior studies on childhood mental disorders and school achievement have used small samples, retrospective designs, and focused on single‐disorder effects (Galéra et al., 2009; Glahn et al., 2006; Maurizi et al., 2013; Pagerols et al., 2022). To our knowledge, only two large‐scale studies on the topic have been conducted. First, a Danish registry study found that most mental disorders diagnosed before age 16 were associated with lower school achievement in ninth‐grade standardized exams (Danish and mathematics) (Dalsgaard et al., 2020). Male students with a mental disorder had lower mean grades than their female counterparts overall, but female students diagnosed with anxiety, attention‐deficit/hyperactivity, or developmental disorders performed worse than their male counterparts (Dalsgaard et al., 2020). Second, a Norwegian registry study found that only certain mental disorders, including intellectual disabilities, personality disorders, and suicide attempts, diagnosed between ages 11 and 15 were linked to poorer school achievement by the end of compulsory education (Nordmo et al., 2022). The negative associations between these diagnoses and school achievement were slightly stronger among female students (Nordmo et al., 2022). These mixed findings highlight the need for further population‐based research to clarify the role of specific mental disorders in shaping school achievement and to better understand gender differences. It is also noteworthy that the Danish study relied solely on standardized exam scores, while the Norwegian study excluded early‐onset diagnoses before age 11. Addressing these gaps, the current study used summative grades from the final three years of comprehensive school to better reflect overall school achievement, and included diagnoses from age 1 through graduation, thereby capturing both early‐ and late‐onset disorders.

Furthermore, the role of age at onset has not been examined in detail in previous studies. Mental disorders typically emerge in childhood or adolescence (Kessler et al., 2007), with a peak onset around age 14.5 (Solmi et al., 2022). Traditional childhood‐onset disorders, such as intellectual disabilities and developmental disorders, generally have the earliest onset (McGrath, Al‐Hamzawi, Alonso, et al., 2023; Solmi et al., 2022). While prior investigations have established associations between early‐onset mental disorders and poorer school achievement during adolescence (Breslau et al., 2009; Leach & Butterworth, 2012), the differential impact of disorder onset age remains underexplored. A meta‐analysis found no overall differences in school achievement by age at onset (Riglin, Petrides, Frederickson, & Rice, 2014), but findings differ in studies that assess specific time points of mental disorder onset. For example, a longitudinal study found externalizing problems at age 3 predicted school achievement at age 12 (Agnafors, Barmark, & Sydsjö, 2021), while another study found externalizing and internalizing problems at age 6 were not associated with academic achievement at age 17 (Breslau et al., 2009). These inconsistencies suggest that the relationship between mental disorder and school achievement may depend on the disorder type and the specific timing of onset. A comprehensive understanding of how age at onset influences school achievement remains lacking.

This population‐based nationwide cohort study examined the association between childhood mental disorders and subsequent school achievement at the end of comprehensive education and whether these associations differed between (a) male and female students, and (b) age at onset of mental disorders. First, we hypothesized that mental disorders diagnosed during childhood or adolescence would be associated with lower school achievement. Second, we did not formulate a hypothesis regarding gender differences due to inconsistent findings in prior research. Third, given the limited existing evidence, we did not propose a specific hypothesis concerning the role of age at onset in school achievement.

Methods

Results

Table 1 presents the descriptive information of the study population. Between age 1 and graduation (the year one turns 16), 16.41% of students were diagnosed with at least one type of mental disorder.

Students with a mental disorder had lower comprehensive school achievement than those without, with a mean difference of −0.41 [95% CI, −0.41 to −0.40], as shown in Figure 2. This pattern was consistent across all mental disorders except for eating disorders, which were associated with slightly higher school achievement compared with students without a diagnosis, 0.06 [95% CI, 0.04–0.08]. The largest school achievement difference was observed for substance use disorders, −0.74 [95% CI, −0.76 to −0.72], while the smallest was seen in those with eating disorders, 0.06 [95% CI, 0.04–0.08].

Additionally, the association between having at least one mental disorder diagnosis and school achievement was more pronounced in females than in males (female: −0.45 [95% CI, −0.46 to −0.45] vs. male: −0.37 [95% CI, −0.38 to −0.37]). The greatest gender difference was observed in intellectual disabilities (females: −0.65 [95% CI, −0.70 to −0.60] vs. males: −0.16 [95% CI, −0.20 to −0.12]), whereas the smallest difference occurred in mood disorders (females: −0.54 [95% CI, −0.56 to −0.53] vs. males: −0.48 [95% CI, −0.50 to −0.46]).

As shown in Figure 3, the later the onset of a mental disorder, the lower the overall school achievement, with the greatest association observed in those diagnosed at ages 13–17, −0.53 [95% CI, −0.54 to −0.52], followed by ages 9–12, −0.43 [95% CI, −0.44 to −0.42], ages 5–8, −0.26 [95% CI, −0.27 to −0.25], and ages 1–4, −0.17 [95% CI, −0.18 to −0.15]. This age‐related pattern was observed for substance use disorders, mood disorders, intellectual disabilities, developmental disorders, and childhood‐onset disorders (see Table 2 for numerical estimates). Furthermore, both females and males showed a clear age‐related pattern (i.e., greater association in older age groups) in terms of substance use disorders, mood disorders, intellectual disabilities, childhood‐onset disorders, and any mental disorder diagnosis. However, for anxiety disorders, there was only a clear age‐related pattern for males but not for females.

In the comorbidity analysis, students with diagnoses in multiple diagnostic categories showed progressively lower GPA compared with those without any diagnosis (−0.38 [95% CI, −0.39 to −0.37] for one diagnostic category, −0.50 [95% CI, −0.51 to −0.49] for two, and − 0.62 [95% CI, −0.64 to −0.60] for three or more) (see Table 3). The pattern was similar in both genders. In addition, females with diagnoses in multiple categories experienced a greater decrease in school achievement than males (−0.54 [95% CI, −0.56 to −0.53] vs. −0.53 [95% CI, −0.54 to −0.51] for two categories, and − 0.68 [95% CI, −0.71 to −0.66] vs. −0.60 [95% CI, −0.63 to −0.56] for three or more categories).

Discussion

This nationwide cohort study on comprehensive school achievement in children and adolescents with and without mental disorders yielded four main findings. First, all mental disorders, diagnosed in childhood or adolescence, except for eating disorders, were associated with significantly lower comprehensive school achievement. Second, the association between mental disorders and school achievement was more pronounced in females than in males. Third, the later the onset of a mental disorder, the greater was its association with school achievement. Fourth, higher mental disorder comorbidity was associated with lower GPA.

We found that students with a mental disorder had lower overall school achievement than those without, across nearly all diagnostic categories, except for eating disorders. The associations were most notable for substance use disorders. These findings align with previous studies focusing on single mental disorders (Esch et al., 2014; Galéra et al., 2009; Keen, Webster, & Ridley, 2016; Maurizi et al., 2013) as well as studies examining multiple mental disorders (Dalsgaard et al., 2020; Nordmo et al., 2022). As reported in earlier studies, substance use has been associated with executive functioning deficits, including difficulties with attention, working memory, and both verbal and nonverbal processing (Jovanovski, Erb, & Zakzanis, 2005; Kalechstein, De La Garza, Mahoney, Fantegrossi, & Newton, 2007), all of which are important for academic success. Interestingly, students diagnosed with eating disorders demonstrated higher school achievement compared with peers without a diagnosis, a finding also supported by a previous study (Dalsgaard et al., 2020). Eating disorders have been linked to high levels of perfectionism (Bulik et al., 2003; Disorder O compulsive et al., 2014), a trait often associated with better academic performance (Bills, Greene, Stackpole, & Egan, 2023; Madigan, 2019). Our findings regarding comorbidity are also in line with previous studies showing that psychiatric comorbidity has been associated with worse prognosis and outcomes than a single diagnosis (Copeland, Wolke, Shanahan, & Costello, 2015; Plana‐Ripoll et al., 2020).

A notable gender difference emerged in the association between mental disorders and school achievement, with female students showing stronger associations than males, overall and across each diagnosis. Our findings differ from those reported in the Danish study on the overall gender difference (Dalsgaard et al., 2020), which may be because we used different measures of academic achievement. However, it aligns with previous research suggesting that female students are more academically vulnerable to intellectual disabilities, anxiety disorders, attention‐deficit hyperactivity disorders, and other developmental disorders (Dalsgaard et al., 2020; Nordmo et al., 2022). The present study extends this pattern to other diagnoses, including substance use disorders, psychotic disorders, and mood disorders. Several mechanisms may account for the heightened academic vulnerability among girls. First, female students are more sensitive to academic pressure, which can amplify the cognitive and emotional burden of mental disorders (Montolio & Taberner, 2021). Second, it may partly reflect a floor effect in male students, whose lower baseline performance, particularly with early‐onset disorders, may limit observable declines (Elwood & Murphy, 2002). Although female students generally had slightly higher average GPA than males in our data, the stronger associations observed among females are unlikely to simply reflect these baseline differences as we used standardized mean differences. Rather, the results suggest that mental disorders may be linked to greater shifts in academic performance among female students. Third, female students with mental disorders are often perceived as less impaired than males, causing their difficulties to be overlooked or underestimated (Sellers, Maughan, Pickles, Thapar, & Collishaw, 2015). Consequently, they may receive less academic or psychological support despite substantial impairments, underscoring the need for greater awareness and targeted interventions. In addition, analyses of comorbidity also indicated that females with diagnoses in multiple categories experienced a greater decrease in school achievement than males. This finding suggests that greater diagnostic complexity may partly contribute to the stronger associations observed among females.

To our knowledge, this is the first nationwide cohort study to examine the association between age at onset of mental disorders and comprehensive school achievement. We found that the association between mental disorders and school achievement was more pronounced among students with later‐onset disorders, especially during adolescence, across diagnostic categories. Several mechanisms may help explain this pattern. First, earlier identification of symptoms and sustained intervention may buffer the long‐term academic consequences of early‐onset disorders. Second, later‐onset disorders may emerge concurrently with increased academic demands and developmental transitions, potentially amplifying their disruptive impact on school performance. Notably, the adolescent period—marked by critical maturation of prefrontal cortical functions essential for executive control (Konrad, Firk, & Uhlhaas, 2013; Page & Coutellier, 2018)—may render youth particularly vulnerable to new‐onset psychopathology during this time. Third, hospital inpatient care may also contribute to lower school achievement by disrupting schooling through absences and difficulties maintaining academic engagement, although hospital schooling is commonly provided in Finland. This may be particularly relevant when hospitalization occurs closer to the end of compulsory education, potentially contributing to the larger associations observed for later‐diagnosed disorders. Fourth, students diagnosed at younger ages and attending special schools or special classes are likely to follow an individual learning plan, which could partly contribute to the observed age‐related patterns, particularly for intellectual disabilities and developmental disorders. However, as only a small proportion of students are educated in such settings (Statistics Finland, 2025), the overall impact on the findings is likely to be limited. Although registry studies have often used age at diagnosis as a proxy for age at onset (Suokas, Niemi, Gutvilig, et al., 2024), it is also important to consider that our measure reflects age at diagnosis rather than true age of symptom onset. Some students diagnosed later may have experienced prolonged periods of unrecognized or unsupported difficulties, leading to cumulative academic disadvantage by the time of diagnosis. These findings underscore the need to disentangle age at onset from age at diagnosis in future research and to ensure timely recognition and support for students whose symptoms may emerge—or be identified—during academically critical periods. Together, the findings highlight the importance of timely identification and intervention, not only for early‐onset disorders but also for those diagnosed later, particularly during critical academic transition periods.

Conclusions

Students with mental disorders had lower school achievement by the end of comprehensive school, with the greatest disparity observed in substance use disorders. Notably, associations were stronger in females than in males across all disorders, suggesting a potential gender‐related difference in the academic impact of mental disorders. Additionally, later‐onset mental disorders were associated with a more pronounced decline in school achievement, highlighting the need for academic and psychological support for adolescents experiencing mental health challenges during critical educational years.

Author contributions

Y.L.—Conceptualization, methodology, visualization, writing—original draft; R.N.—Conceptualization, methodology, formal analysis, data curation, writing—review and editing; M.G.—Conceptualization, methodology, formal analysis, data curation, writing—review and editing; M.E.—Conceptualization, writing—review and editing; M.F.—Conceptualization, writing—review and editing; C.H.—Conceptualization, supervision, project administration, funding acquisition, writing—review and editing.

Ethical considerations

According to Finnish law, informed consent/assent was not required for register‐based studies; therefore, informed consent was waived. The study was approved by the Ethics Committee of the Finnish Institute of Health and Welfare (approval number THL/22/6.02.01/2019; approval date: 15 January 2019). Data linkage was authorized by Statistics Finland (TK‐53‐1696‐16) and the Finnish Institute of Health and Welfare.

Supporting information

Data availability statement

Data for this study are property of Finnish Institute of Health and Welfare and Statistics Finland. The data are available from these authorities, but restrictions apply, please see: www.findata.fi.