GP-diagnosed internalizing and externalizing problems and dropout from secondary school: a cross-sectional study

GP-diagnosed internalizing and externalizing problems and dropout from secondary school: a... Abstract Background Dropout from secondary education is a major concern in many Western countries because it is associated with later adverse consequences at the individual and societal level. Efforts have thus been made to identify precursors for dropout. The aim of the study was to examine the risk for not finishing secondary education by age 20 when mental health problems were diagnosed during general practitioner (GP) consultations. Methods National registries were linked to assess the association between GP-diagnosed internalizing and externalizing problems from the ages of 15–20 years and school dropout in a sample of 63 970 Norwegians, adjusting for health and social background factors. Relative risks (RR) were estimated by Poisson regression. Results Dropout was bivariately related to both internalizing (RR = 2.2 among girls and 1.7 among boys) and externalizing problems (RR = 2.7 for girls and 2.0 for boys), though the associations were somewhat attenuated in the adjusted analyses. After controlling for somatic comorbidity and parent education level, the absolute risk for not fulfilling secondary education by age 20 was 43% among girls and 60% among boys with internalizing problems and approximately 15% points higher with externalizing problems. The highest absolute risk for dropout was found for boys and girls who have both externalizing and internalizing problems. However, with some overlap in the confidence intervals, the added impact of internalizing problems when added to externalizing problems is uncertain. Conclusions Intervention for mental health problems by a GP could benefit adolescent education outcomes and mental well-being. Introduction The large proportion of adolescents who fail to complete secondary education is an area of great concern in many western countries. In Norway, nearly one third of students never finish upper secondary education, which is usually attended by students between the ages of 16 and 19 years. Truncating one’s educational path in this way is related to later adverse consequences on an individual and societal level, such as lack of participation in the labour market and early dependence on social insurance benefits.1,2 Mental health problems are found among 10–20% of adolescents, and studies indicate an increasing frequency.3,4 Such problems are an early risk indicator and predictor of early school departure.5–9 Externalizing mental health problems, including symptoms of ADHD and behaviour problems, have consistently shown an association with school dropout.5–8 The association between internalizing problems such as anxiety and depression and dropout from school is less consistent and not robust to adjustments for externalizing problems.7 One previous prospective register-based study found associations between early internalizing problems and school dropout, but for girls only.5 Further, most studies have not accounted for comorbidity, but some studies suggest that the association between internalizing problems and school dropout is partly accounted for by co-occurring externalizing problems.5,7 Having somatic health problems during adolescence also increases the risk for not completing higher secondary education.10,11 Undiagnosed medical problems were the most frequent health issue among students with extensive absenteeism.12 However, there is a shortage of studies that consider both somatic and psychological problems as predictors for early school departure. In studying the association between mental health problems and dropout from school, it is important to account for family socioeconomic status (SES), and especially parent education levels. Family SES is repeatedly found to be associated with adolescent mental health,13 and parent education level is a significant predictor of academic achievement and school dropout;14,15 it is strongly associated with the likelihood of their offspring completing upper secondary education.16 General practitioners (GPs) are the first line of health care for young people and are the most frequently used health services among adolescents.17 One study found more frequent GP consultations among adolescents with higher school absenteeism,17 while another have reported more frequent GP consultations to be associated with lower risk of school dropout among boys, with no such association among girls.11 Thus, GPs may be reliable informants on mental health in adolescents and may also be important for early intervention. Aims The overall aim of the current study was to assess the association between mental health problems and school dropout in a national register including 63 970 young adults. More specifically, we wanted to assess the frequency of internalizing and externalizing problems during GP consultations and to study associations with completion of upper secondary education by age 20. Further, the independent risk associated with both internalizing and externalizing problems was explored, while accounting for relevant social background factors and other health complaints. Methods Study design and setting This cross-sectional study is based on register data from 2006 to 2011 covering a national cohort of persons born in Norway in 1991. Data sources The regular GP database, containing information about all regular GPs contracted to municipalities and the identity of the patients on each GP’s list, including age and gender. The GPs send claims of fee-for-service for each patient contact to the Norwegian Health Economics Administration, collected in the KUHR database. The National Educational Database was used for information on completed education. Parent information was retrieved from the National Population Register. Statistics Norway linked the registers and provided de-identified data to use for research. Ethics The study had permission from the Data Inspectorate and all respective register owners to use these data sources for the current research purposes. Sample All inhabitants born in 1991 that where assigned to a GP in 2011 (N = 66 516) were included. There were some participants for which information about their education level (N = 2546) and their parents’ education level (N = 5859) was unavailable. This resulted in a sample size of 63 970 (48.5% girls) for the bivariate analyses, and 58 111 (47.9% girls) for the multivariate analyses. GP visits and diagnoses from claims data The number of GP visits from 15- to 20-year-olds were obtained from the KUHR database in the 2006–2011 period. For each consultation, the claims included one or more diagnoses according to the International Classification of Primary Health Care (ICPC-2). The ICPC-2 coding system is organized with chapters for ‘organ systems’ including a general chapter (A), a psychological chapter (P) and a social chapter (Z).18 Internalizing and externalizing mental health problems were defined according to specific diagnostic codes in the P-chapter of the ICPC-2 coding system. Internalizing problems are covered in the ICPC codes P01-03, P11, P27-29, P73-80, P82, P86 or P99, where anxiety and depression were the most frequent diagnoses. Externalizing problems are presented in codes P04, P15-16, P18-19, P22-23 or P81, where hyperactivity and behavioural problems were most frequent. These groups of GP diagnoses were the main predictors in the analyses. In addition, some of the most frequent diagnoses (excluding infections and reproductive health issues) used by GPs in consultations with patients in this age group were used for adjustment purposes in the regression models: sleep disturbance/insomnia (P06), tiredness/fatigue (A04), headache (N01), neck or back pain (L01-03, L18, L83-85) and asthma or allergy (R96-97). Education and dropping out Compulsory education in Norway is 10 years, with primary and lower secondary school finished at age 16. Then follows a right to a further 3 years of education in upper secondary school, also including vocational training for 1 extra year. In the current study, dropping out/not finishing upper secondary education was operationalized as not having completed 13 years of education by the age of 20 years. Statistical analyses Student’s T- and chi-squared tests were used to compare frequencies of GP contact and type of diagnoses in groups of students who had or had not completed schooling by age 20. Poisson regression models with robust estimates of variance were used to calculate relative risk (RR) for not finishing upper secondary education by age 20, with separate models for boys and girls, presenting both univariate estimates and multivariate analyses. Gender, diagnoses from GP consultations, the number of GP consultations and parent education level (completed upper secondary school or not) were the main predictors. We also calculated the number of GP consultations with nonpsychological diagnoses as a general indicator of somatic health problems, used as predictor in regression models in addition to the selected diagnoses. The multivariate models take into account the interaction between internalizing and externalizing problems, present the RR for having either internalizing or externalizing problems and the interaction term. After regression, the Stata command lincom was used to test for significances of adding internalizing to externalizing problems and vice versa, and the margins command was used to estimate the absolute risk for dropout controlled for the other variables in the model. Sensitivity analyses performed on the distribution of diagnoses and GP visits using different samples did not indicate any selection bias owing to missing parent education information. Stata Statistical Software (Release 14; College Station, TX, USA) was used for all statistical analyses. Results Among the 63 970 persons included, 69.3% of the girls and 54.0% of boys had finished upper secondary education by age 20. Girls had an average annual consultation rate of 1.6 in the total study period compared with 0.9 among boys. In both genders, approximately 10% of consultations were related to a mental health problem (table 1), and 23.4% of girls and 16.5% of boys had ≥1 consultation(s) related to a psychological diagnosis in the study period. Table 1 Overview of education by age 20 years, use of a GP from age 15 to 20 years, and parent education level among all persons aged 20 in 2011 included in the Norwegian list patient system   Girls  Boys  Proportion finished upper secondary education by age 20, %  69.3  54.0  Number of GP consultations from age 15 to 20 years, mean  9.60  5.53  Number of GP consultations with any psychological diagnosis (P in ICPC) from age 15 to 20 years, mean  0.97  0.58  Total number of GP consultations from 2006 to 2011, proportion (%) within each group:      0  4.4  9.8      1–4  26.3  43.3      5–9  30.8  30.2      ≥10  38.5  16.7  Number of GP consultations with psychological diagnosis (P in ICPC) in 2006–2011, proportion (%) within each group:      0  76.6  83.5      1–2  13.3  9.9      3–5  5.6  4.2      ≥6  4.5  2.4  Proportion (%) with ≥1 GP consultation(s) with a diagnosis of:      Internalizing psychological problemsa          Age 15–17 years  8.4  3.4      Age 18–20 years  16.5  8.5      Age 15–20 years (whole period)  20.6  10.4      Externalizing psychological problemsb          Age 15–17 years  2.6  3.6      Age 18–20 years  2.9  4.2      Age 15–20 years (whole period)  4.3  6.1      Both externalizing and internalizing (also included above)          Age 15–20 years (whole period)  2.3  2.0  Insomnia, fatigue, headache, neck or back pain and/or asthma or allergy (age 15–20 years)c  52.0  38.8  Parent education <13 years (%)      Mother  36.7  36.6      Father  34.5  34.5    Girls  Boys  Proportion finished upper secondary education by age 20, %  69.3  54.0  Number of GP consultations from age 15 to 20 years, mean  9.60  5.53  Number of GP consultations with any psychological diagnosis (P in ICPC) from age 15 to 20 years, mean  0.97  0.58  Total number of GP consultations from 2006 to 2011, proportion (%) within each group:      0  4.4  9.8      1–4  26.3  43.3      5–9  30.8  30.2      ≥10  38.5  16.7  Number of GP consultations with psychological diagnosis (P in ICPC) in 2006–2011, proportion (%) within each group:      0  76.6  83.5      1–2  13.3  9.9      3–5  5.6  4.2      ≥6  4.5  2.4  Proportion (%) with ≥1 GP consultation(s) with a diagnosis of:      Internalizing psychological problemsa          Age 15–17 years  8.4  3.4      Age 18–20 years  16.5  8.5      Age 15–20 years (whole period)  20.6  10.4      Externalizing psychological problemsb          Age 15–17 years  2.6  3.6      Age 18–20 years  2.9  4.2      Age 15–20 years (whole period)  4.3  6.1      Both externalizing and internalizing (also included above)          Age 15–20 years (whole period)  2.3  2.0  Insomnia, fatigue, headache, neck or back pain and/or asthma or allergy (age 15–20 years)c  52.0  38.8  Parent education <13 years (%)      Mother  36.7  36.6      Father  34.5  34.5  Notes: N = 31 010 girls and 32 960 boys. Corresponding to ICPC codes ((https://ehelse.no/icpc-2e-english-version).): a P01-03, P11, P27-29, P73-80, P82, P86, or P99; b P04, P15-16, P18-19, P22-23, or P81; c P06, A04, N01, L01-03, L18, L83-85, R96-97. Table 1 Overview of education by age 20 years, use of a GP from age 15 to 20 years, and parent education level among all persons aged 20 in 2011 included in the Norwegian list patient system   Girls  Boys  Proportion finished upper secondary education by age 20, %  69.3  54.0  Number of GP consultations from age 15 to 20 years, mean  9.60  5.53  Number of GP consultations with any psychological diagnosis (P in ICPC) from age 15 to 20 years, mean  0.97  0.58  Total number of GP consultations from 2006 to 2011, proportion (%) within each group:      0  4.4  9.8      1–4  26.3  43.3      5–9  30.8  30.2      ≥10  38.5  16.7  Number of GP consultations with psychological diagnosis (P in ICPC) in 2006–2011, proportion (%) within each group:      0  76.6  83.5      1–2  13.3  9.9      3–5  5.6  4.2      ≥6  4.5  2.4  Proportion (%) with ≥1 GP consultation(s) with a diagnosis of:      Internalizing psychological problemsa          Age 15–17 years  8.4  3.4      Age 18–20 years  16.5  8.5      Age 15–20 years (whole period)  20.6  10.4      Externalizing psychological problemsb          Age 15–17 years  2.6  3.6      Age 18–20 years  2.9  4.2      Age 15–20 years (whole period)  4.3  6.1      Both externalizing and internalizing (also included above)          Age 15–20 years (whole period)  2.3  2.0  Insomnia, fatigue, headache, neck or back pain and/or asthma or allergy (age 15–20 years)c  52.0  38.8  Parent education <13 years (%)      Mother  36.7  36.6      Father  34.5  34.5    Girls  Boys  Proportion finished upper secondary education by age 20, %  69.3  54.0  Number of GP consultations from age 15 to 20 years, mean  9.60  5.53  Number of GP consultations with any psychological diagnosis (P in ICPC) from age 15 to 20 years, mean  0.97  0.58  Total number of GP consultations from 2006 to 2011, proportion (%) within each group:      0  4.4  9.8      1–4  26.3  43.3      5–9  30.8  30.2      ≥10  38.5  16.7  Number of GP consultations with psychological diagnosis (P in ICPC) in 2006–2011, proportion (%) within each group:      0  76.6  83.5      1–2  13.3  9.9      3–5  5.6  4.2      ≥6  4.5  2.4  Proportion (%) with ≥1 GP consultation(s) with a diagnosis of:      Internalizing psychological problemsa          Age 15–17 years  8.4  3.4      Age 18–20 years  16.5  8.5      Age 15–20 years (whole period)  20.6  10.4      Externalizing psychological problemsb          Age 15–17 years  2.6  3.6      Age 18–20 years  2.9  4.2      Age 15–20 years (whole period)  4.3  6.1      Both externalizing and internalizing (also included above)          Age 15–20 years (whole period)  2.3  2.0  Insomnia, fatigue, headache, neck or back pain and/or asthma or allergy (age 15–20 years)c  52.0  38.8  Parent education <13 years (%)      Mother  36.7  36.6      Father  34.5  34.5  Notes: N = 31 010 girls and 32 960 boys. Corresponding to ICPC codes ((https://ehelse.no/icpc-2e-english-version).): a P01-03, P11, P27-29, P73-80, P82, P86, or P99; b P04, P15-16, P18-19, P22-23, or P81; c P06, A04, N01, L01-03, L18, L83-85, R96-97. A GP diagnosis reflecting an internalizing mental health problem was registered among 20.6% of girls and 10.4% of boys during these 6 years, while 4–6% had externalizing problems. GP-diagnosed mental health problems and dropout The 6-year incident rate of GP consultations for internalizing problems was 2–3 times higher among both girls and boys that had not finished secondary school by age 20 compared with those who had finished (table 2). Table 2 Total number of GP consultations, frequency of internalizing and externalizing diagnoses and parent education level by gender and the child’s education status by age 20 in a cohort born in 1991 followed from age 15 to 20 years   Finished upper secondary education at age 20  Girls  Boys  Yes  No  Yes  No  N (% within gender)  21 489 (69.3%)  9521 (30.7%)  17 795 (54.0%)  15 165 (46.0%)  Number of GP consultations in total from age 15 to age 20 years, mean (95% CI)  8.05 (7.95–8.14)  13.10 (12.88–13.32)*  4.79 (4.72–4.85)  6.40 (6.30–6.50)*  Number of GP consultations with any psychological diagnosis (P in ICPC) from age 15 to 20 years, mean (95% CI)  0.50 (0.47–0.53)  1.88 (1.81–1.97)*  0.22 (0.20–0.23)  1.00 (0.95–1.04)*  Number of GP consultations in 2006–2011 grouped, proportion (%) within each group:      0  5.0  3.1**  11.4  8.0**      1–4  30.3  17.2**  46.7  39.3**      5–9  33.3  25.0**  29.6  30.9**      ≥10  31.4  54.7**  12.3  21.8**  Proportion (%) with ≥1 GP consultation(s) during 2006–2011 with a diagnosis of:      Internalizing psychological problems  14.0  35.5**  5.6  16.0**      Externalizing psychological problems  1.6  10.5**  1.8  11.1**      Both internalizing and externalizing (also included above)  0.6  6.2**  0.4  3.9**      Insomnia, fatigue, headache, neck or back pain and/or asthma or allergy  48.8  59.8**  35.9  42.5**  Parent education <13 years, (%)      Mother  30.3  51.7**  27.5  47.6**      Father  29.0  47.7**  26.1  44.9**    Finished upper secondary education at age 20  Girls  Boys  Yes  No  Yes  No  N (% within gender)  21 489 (69.3%)  9521 (30.7%)  17 795 (54.0%)  15 165 (46.0%)  Number of GP consultations in total from age 15 to age 20 years, mean (95% CI)  8.05 (7.95–8.14)  13.10 (12.88–13.32)*  4.79 (4.72–4.85)  6.40 (6.30–6.50)*  Number of GP consultations with any psychological diagnosis (P in ICPC) from age 15 to 20 years, mean (95% CI)  0.50 (0.47–0.53)  1.88 (1.81–1.97)*  0.22 (0.20–0.23)  1.00 (0.95–1.04)*  Number of GP consultations in 2006–2011 grouped, proportion (%) within each group:      0  5.0  3.1**  11.4  8.0**      1–4  30.3  17.2**  46.7  39.3**      5–9  33.3  25.0**  29.6  30.9**      ≥10  31.4  54.7**  12.3  21.8**  Proportion (%) with ≥1 GP consultation(s) during 2006–2011 with a diagnosis of:      Internalizing psychological problems  14.0  35.5**  5.6  16.0**      Externalizing psychological problems  1.6  10.5**  1.8  11.1**      Both internalizing and externalizing (also included above)  0.6  6.2**  0.4  3.9**      Insomnia, fatigue, headache, neck or back pain and/or asthma or allergy  48.8  59.8**  35.9  42.5**  Parent education <13 years, (%)      Mother  30.3  51.7**  27.5  47.6**      Father  29.0  47.7**  26.1  44.9**  Notes: Register-based study, national data. * P < 0.001, Student’s T-test. ** P < 0.001, Pearson’s chi-squared (comparing ‘not finished’ to ‘finished’ upper secondary education by age 20). Table 2 Total number of GP consultations, frequency of internalizing and externalizing diagnoses and parent education level by gender and the child’s education status by age 20 in a cohort born in 1991 followed from age 15 to 20 years   Finished upper secondary education at age 20  Girls  Boys  Yes  No  Yes  No  N (% within gender)  21 489 (69.3%)  9521 (30.7%)  17 795 (54.0%)  15 165 (46.0%)  Number of GP consultations in total from age 15 to age 20 years, mean (95% CI)  8.05 (7.95–8.14)  13.10 (12.88–13.32)*  4.79 (4.72–4.85)  6.40 (6.30–6.50)*  Number of GP consultations with any psychological diagnosis (P in ICPC) from age 15 to 20 years, mean (95% CI)  0.50 (0.47–0.53)  1.88 (1.81–1.97)*  0.22 (0.20–0.23)  1.00 (0.95–1.04)*  Number of GP consultations in 2006–2011 grouped, proportion (%) within each group:      0  5.0  3.1**  11.4  8.0**      1–4  30.3  17.2**  46.7  39.3**      5–9  33.3  25.0**  29.6  30.9**      ≥10  31.4  54.7**  12.3  21.8**  Proportion (%) with ≥1 GP consultation(s) during 2006–2011 with a diagnosis of:      Internalizing psychological problems  14.0  35.5**  5.6  16.0**      Externalizing psychological problems  1.6  10.5**  1.8  11.1**      Both internalizing and externalizing (also included above)  0.6  6.2**  0.4  3.9**      Insomnia, fatigue, headache, neck or back pain and/or asthma or allergy  48.8  59.8**  35.9  42.5**  Parent education <13 years, (%)      Mother  30.3  51.7**  27.5  47.6**      Father  29.0  47.7**  26.1  44.9**    Finished upper secondary education at age 20  Girls  Boys  Yes  No  Yes  No  N (% within gender)  21 489 (69.3%)  9521 (30.7%)  17 795 (54.0%)  15 165 (46.0%)  Number of GP consultations in total from age 15 to age 20 years, mean (95% CI)  8.05 (7.95–8.14)  13.10 (12.88–13.32)*  4.79 (4.72–4.85)  6.40 (6.30–6.50)*  Number of GP consultations with any psychological diagnosis (P in ICPC) from age 15 to 20 years, mean (95% CI)  0.50 (0.47–0.53)  1.88 (1.81–1.97)*  0.22 (0.20–0.23)  1.00 (0.95–1.04)*  Number of GP consultations in 2006–2011 grouped, proportion (%) within each group:      0  5.0  3.1**  11.4  8.0**      1–4  30.3  17.2**  46.7  39.3**      5–9  33.3  25.0**  29.6  30.9**      ≥10  31.4  54.7**  12.3  21.8**  Proportion (%) with ≥1 GP consultation(s) during 2006–2011 with a diagnosis of:      Internalizing psychological problems  14.0  35.5**  5.6  16.0**      Externalizing psychological problems  1.6  10.5**  1.8  11.1**      Both internalizing and externalizing (also included above)  0.6  6.2**  0.4  3.9**      Insomnia, fatigue, headache, neck or back pain and/or asthma or allergy  48.8  59.8**  35.9  42.5**  Parent education <13 years, (%)      Mother  30.3  51.7**  27.5  47.6**      Father  29.0  47.7**  26.1  44.9**  Notes: Register-based study, national data. * P < 0.001, Student’s T-test. ** P < 0.001, Pearson’s chi-squared (comparing ‘not finished’ to ‘finished’ upper secondary education by age 20). GP-diagnosed externalizing problems were found among 1.6% of girls and 1.8% of boys when upper secondary school was finished by age 20, the corresponding figures were 10.5% and 11.1% among those who dropped out of school (table 2). For both internalizing and externalizing psychological problems, the RR for dropout was higher among girls than boys (table 3). When adjusting for other health problems and parent education level the RR for dropout associated with internalizing problems was 1.85 (95% CI: 1.78–1.92) among girls and 1.49 (95% CI: 1.45–1.54) among boys. For externalizing problems, the corresponding figures were 2.48 (95% CI: 2.32–2.66) for girls and 1.84 (95% CI: 1.78–1.90) for boys. Table 3 Associations between not finished secondary school by age 20 and internalizing and externalizing problems according to diagnoses in ≥1 GP consultation(s) during the age period of 15–20 years, adjusted for other common health problems and parent education   Girls  Boys  Crude rates  Multivariate modela  Crude rates  Multivariate modela    RR  95% CI  RR  95% CI  RR  95% CI  RR  95% CI  Internalizing problems  2.23  2.16–2.31  1.85  1.78–1.92  1.68  1.63–1.72  1.49  1.44–1.54  Externalizing problems  2.74  2.63–2.84  2.48  2.32–2.66  2.00  1.95–2.05  1.84  1.78–1.90  Internalizing × Externalizingb      0.60  0.55–0.65      0.70  0.66–0.74  Number of GP consultationsc  0  Ref.    Ref.    Ref.      Ref.  1–4  1.0  0.88–1.13  0.95  0.84–1.07  1.09  1.04–1.15  1.04  0.99–1.09  5–9  1.24  1.10–1.40  1.07  0.95–1.21  1.16  1.10–1.22  1.07  1.01–1.12  ≥10  2.0  1.77–2.25  1.44  1.28–1.63  1.41  1.33–1.48  1.17  1.11–1.23  Parent education <13 years  Mother  1.83  1.77–1.90  1.55  1.50–1.61  1.58  1.54–1.62  1.40  1.37–1.44  Father  1.72  1.66–1.78  1.43  1.38–1.48  1.54  1.50–1.58  1.37  1.33–1.40    Girls  Boys  Crude rates  Multivariate modela  Crude rates  Multivariate modela    RR  95% CI  RR  95% CI  RR  95% CI  RR  95% CI  Internalizing problems  2.23  2.16–2.31  1.85  1.78–1.92  1.68  1.63–1.72  1.49  1.44–1.54  Externalizing problems  2.74  2.63–2.84  2.48  2.32–2.66  2.00  1.95–2.05  1.84  1.78–1.90  Internalizing × Externalizingb      0.60  0.55–0.65      0.70  0.66–0.74  Number of GP consultationsc  0  Ref.    Ref.    Ref.      Ref.  1–4  1.0  0.88–1.13  0.95  0.84–1.07  1.09  1.04–1.15  1.04  0.99–1.09  5–9  1.24  1.10–1.40  1.07  0.95–1.21  1.16  1.10–1.22  1.07  1.01–1.12  ≥10  2.0  1.77–2.25  1.44  1.28–1.63  1.41  1.33–1.48  1.17  1.11–1.23  Parent education <13 years  Mother  1.83  1.77–1.90  1.55  1.50–1.61  1.58  1.54–1.62  1.40  1.37–1.44  Father  1.72  1.66–1.78  1.43  1.38–1.48  1.54  1.50–1.58  1.37  1.33–1.40  Notes: N = 27 849 girls and 29 643 boys; Poisson regression models, register based study, national data. a Multivariate model also adjusted for: insomnia, fatigue, headache, back or neck pain, asthma and allergy. b Adjusting for interaction between internalizing and externalizing in multivariate models. c Number of GP consultations in 2006–2011, psychological reasons excluded. Table 3 Associations between not finished secondary school by age 20 and internalizing and externalizing problems according to diagnoses in ≥1 GP consultation(s) during the age period of 15–20 years, adjusted for other common health problems and parent education   Girls  Boys  Crude rates  Multivariate modela  Crude rates  Multivariate modela    RR  95% CI  RR  95% CI  RR  95% CI  RR  95% CI  Internalizing problems  2.23  2.16–2.31  1.85  1.78–1.92  1.68  1.63–1.72  1.49  1.44–1.54  Externalizing problems  2.74  2.63–2.84  2.48  2.32–2.66  2.00  1.95–2.05  1.84  1.78–1.90  Internalizing × Externalizingb      0.60  0.55–0.65      0.70  0.66–0.74  Number of GP consultationsc  0  Ref.    Ref.    Ref.      Ref.  1–4  1.0  0.88–1.13  0.95  0.84–1.07  1.09  1.04–1.15  1.04  0.99–1.09  5–9  1.24  1.10–1.40  1.07  0.95–1.21  1.16  1.10–1.22  1.07  1.01–1.12  ≥10  2.0  1.77–2.25  1.44  1.28–1.63  1.41  1.33–1.48  1.17  1.11–1.23  Parent education <13 years  Mother  1.83  1.77–1.90  1.55  1.50–1.61  1.58  1.54–1.62  1.40  1.37–1.44  Father  1.72  1.66–1.78  1.43  1.38–1.48  1.54  1.50–1.58  1.37  1.33–1.40    Girls  Boys  Crude rates  Multivariate modela  Crude rates  Multivariate modela    RR  95% CI  RR  95% CI  RR  95% CI  RR  95% CI  Internalizing problems  2.23  2.16–2.31  1.85  1.78–1.92  1.68  1.63–1.72  1.49  1.44–1.54  Externalizing problems  2.74  2.63–2.84  2.48  2.32–2.66  2.00  1.95–2.05  1.84  1.78–1.90  Internalizing × Externalizingb      0.60  0.55–0.65      0.70  0.66–0.74  Number of GP consultationsc  0  Ref.    Ref.    Ref.      Ref.  1–4  1.0  0.88–1.13  0.95  0.84–1.07  1.09  1.04–1.15  1.04  0.99–1.09  5–9  1.24  1.10–1.40  1.07  0.95–1.21  1.16  1.10–1.22  1.07  1.01–1.12  ≥10  2.0  1.77–2.25  1.44  1.28–1.63  1.41  1.33–1.48  1.17  1.11–1.23  Parent education <13 years  Mother  1.83  1.77–1.90  1.55  1.50–1.61  1.58  1.54–1.62  1.40  1.37–1.44  Father  1.72  1.66–1.78  1.43  1.38–1.48  1.54  1.50–1.58  1.37  1.33–1.40  Notes: N = 27 849 girls and 29 643 boys; Poisson regression models, register based study, national data. a Multivariate model also adjusted for: insomnia, fatigue, headache, back or neck pain, asthma and allergy. b Adjusting for interaction between internalizing and externalizing in multivariate models. c Number of GP consultations in 2006–2011, psychological reasons excluded. The absolute risk for dropout with different combinations of problems, after controlling for the adjusting variables, are shown in table 4. Table 4 Absolute riska for not having finished secondary school by age 20 according to internalizing and externalizing problems in GP diagnoses in any consultation during the age period 15–20 years   Girls  Boys    Risk for drop-out, %  95% CI  Risk for drop-out, %  95% CI  Neither internalizing nor externalizing  23.1  22.6–23.7  40.3  39.7–40.9  Internalizing - not externalizing  42.7  41.5–44.0  60.2  58.4–61.9  Externalizing - not internalizing  57.5  54.8–61.1  74.2  72.0–76.4  Internalizing and externalizing  63.6  60.6–66.6  77.2  74.5–80.0    Girls  Boys    Risk for drop-out, %  95% CI  Risk for drop-out, %  95% CI  Neither internalizing nor externalizing  23.1  22.6–23.7  40.3  39.7–40.9  Internalizing - not externalizing  42.7  41.5–44.0  60.2  58.4–61.9  Externalizing - not internalizing  57.5  54.8–61.1  74.2  72.0–76.4  Internalizing and externalizing  63.6  60.6–66.6  77.2  74.5–80.0  Notes: N = 27 849 girls and 29 643 boys; register based study, national data. a Marginal effects, giving the risk estimates adjusted for somatic comorbidity and parent education (based on a Poisson regression model, see table 3). Table 4 Absolute riska for not having finished secondary school by age 20 according to internalizing and externalizing problems in GP diagnoses in any consultation during the age period 15–20 years   Girls  Boys    Risk for drop-out, %  95% CI  Risk for drop-out, %  95% CI  Neither internalizing nor externalizing  23.1  22.6–23.7  40.3  39.7–40.9  Internalizing - not externalizing  42.7  41.5–44.0  60.2  58.4–61.9  Externalizing - not internalizing  57.5  54.8–61.1  74.2  72.0–76.4  Internalizing and externalizing  63.6  60.6–66.6  77.2  74.5–80.0    Girls  Boys    Risk for drop-out, %  95% CI  Risk for drop-out, %  95% CI  Neither internalizing nor externalizing  23.1  22.6–23.7  40.3  39.7–40.9  Internalizing - not externalizing  42.7  41.5–44.0  60.2  58.4–61.9  Externalizing - not internalizing  57.5  54.8–61.1  74.2  72.0–76.4  Internalizing and externalizing  63.6  60.6–66.6  77.2  74.5–80.0  Notes: N = 27 849 girls and 29 643 boys; register based study, national data. a Marginal effects, giving the risk estimates adjusted for somatic comorbidity and parent education (based on a Poisson regression model, see table 3). The presence of either internalizing or externalizing problems increased the absolute risk of dropout for both boys and girls relative to those without such problems. The absolute risk for dropout was increased for those having both externalizing and internalizing problems (girls: 63.6%, boys: 77.2%) compared with those having only externalizing problems (girls: 57.5%, boys: 74.2%). However, the overlap in confidence intervals (most pronounced for boys) suggests that some uncertainty remains regarding the impact of having internalizing problems when externalizing problems were present. As sensitivity analyses, we ran regression models including internalizing or externalizing problems during age 15–17 years and 18–20 years, respectively. The associations with dropout were similar for boys independent of the age period used. For girls, however, the association between dropout and internalizing psychological problems reported during age 15–17 years was lower (RR 1.58, [1.51–1.65]) than for the same diagnoses reported in the later 3-year period (RR 1.76, [1.69–1.82]) (adjusted analyses, not tabulated). Discussion The aim of the current study was to investigate GP-diagnosed internalizing and externalizing mental health problems from age 15 to 20 years as predictors of school dropout by age 20 in a sample of 63 970 young adults from Norway. The dropout rate at 20 years was 31% for girls and 46% for boys in the total study population. The risk for school dropout was substantially increased for both girls and boys presenting with internalizing or externalizing problems. The associations with school dropout were attenuated somewhat, but remained significant after accounting for parent education level and other health complaints. The dropout risk for girls was 42.7% and for boys was 60.2% for internalizing problems and, respectively, 57.5% and 74.2% when externalizing problems were found in GP diagnoses. Having both kinds of problems was associated with the highest absolute risk for dropout for both genders. However, with somewhat overlapping confidence intervals, the impact of comorbid internalizing problems should therefore be interpreted with some caution. Comparison with the literature Although GP diagnoses are not indicators of population prevalence of mental health problems, the 6-year consultation prevalence for mental health problems are in line with prevalence estimates in adolescents.19 The prevalence of internalizing problems in the present study is comparable to the 10% found in a survey among 16-year-olds,20 but clearly lower than the 23% lifetime prevalence of depression among adolescents found in another study.21 Internalizing problems were relatively frequent in the current sample, especially among girls, and in line with population based studies.22 Externalizing problems were less commonly observed, but somewhat more often in boys relative to girls. The association between dropout and mental health problems is strong in the present study, in line with earlier studies.5,9 Both internalizing and externalizing mental health problems showed independent associations with school noncompletion. While the association between externalizing problems is in accordance with the literature,5–8 the current study further strengthens the importance of internalizing problems as an independent risk factor for school dropout. In the present study more than one-third of girls in the dropout group had a GP-diagnosed internalizing problem. The combination of high frequency and high associated risk for school noncompletion indicates that this might have important public health implications and could indicate that internalizing problems have a larger impact than what has been found in other studies that indicated less clear associations.7,9 In general, dropout is more frequent among boys and there is a concern for how boys manage in modern school systems because of their lower achievement compared with girls, giving the boys problems in passing exams and therefore not completing education.23 However, there may also be gender differences in how symptoms are described to GPs, and this may account for some of the gender differences in associations between dropout and health problems that are reported by GPs.24 Health complaints stabilize from the time of adolescence to adulthood25,26 and are associated with low work–life participation later in life.27 Furthermore, health complaints often co-occur with mental health problems.28 In this study, while adjusting for these health complaints in the final model, mental health problems remained a significant predictor of school dropout. This confirms the high rate of co-occurring somatic health complaints and mental health problems in this age group. Having parents who have not fulfilled upper secondary education was a strong predictor for dropout, even when adjusting for health problems, in line with the established associations between socioeconomic background and academic performance and dropout from secondary school.14,29 Strengths and limitations A main strength of the current study is that it is based on a near complete population sample and data collected from national registers. The fee-for-service financing model in the list patient system results in registration of all patient contact, and is a more reliable measure of utilization than the self-reported data commonly used in similar studies. The high coverage limits selection bias and increases external validity of the findings. The other registers used are maintained by national authorities where quality control is highly prioritized. A limitation is that the majority of claims report only one diagnosis when, in a large proportion of consultations, the GPs deal with multiple problems. Probably the reported ICPC-2 diagnoses underestimate the number of psychological issues raised in the consultations.30 In addition, the precision of diagnoses is uncertain,30 but by grouping diagnosis in internalizing and externalizing mental health problems, the impact of misclassifications should be reduced. In the present cross-sectional study, it is not possible to interpret associations as causal relationships. With the design using GP consultations from the whole age span framing the upper secondary school period, the relation in time between dropout and symptoms cannot be assessed and a temporal relationship is not certain. When timing of the diagnosis was included, there was no difference related to the time period used in association with school completion in boys, but there was a stronger association between later diagnoses among girls. Previous studies with prospective designs have found mental health problems present prior to dropout from school,10,31,32 but increases in internalizing symptoms after dropping out have also been documented.6 Future studies using longitudinal designs are needed to further elucidate the temporal order between mental health problems and school dropout. Implications The clear association found in this study between recognized mental health problems among young people in encounters with GPs and school dropout supports the policy of working for improving mental health service in primary care. However, there are challenges to recognize, follow up and properly treat mental health problems among young people.33 Under-presentation of problems seems to be common in this age group, which challenges a GP’s skills to recognize emotional problems.34,35 GPs might be in a good position to work within the mental health field and with adolescents. They may build relations through different health care contacts over time and have knowledge of local context.36 It is a paradox, however, that the majority of adolescents with a recognized mental health problem have rather few consultations with GPs where that problem seems to be central. This may be explained by the finding that GPs consider themselves as not sufficiently skilled to work with these problems.37,38 Improving the GPs’ knowledge and giving them more confidence at working with youth mental health problems have been shown to be effective33 and could be implemented and evaluated in Norway. Last, the associations between leaving school early and mental health problems are complex and the solutions may be related to solving education problems or improving the psychosocial situation of the student.39,40 There is a need for cooperation between the GP, which may be a ‘recognizer’, other health care providers and the schools. Improved cooperation is challenging because municipal health care services often are fragmented. To include the GP as a resource in a multidisciplinary work, there is a need for both a mutual interest in competencies and a clarification of each other’s roles. An increase of resources, such as money and time, is also necessary to make it possible to foster cooperation between the GP and other stakeholders. Conclusion Mental health problems seem to have a great impact on risk for school dropout in upper secondary education, and measures to reduce dropout from upper secondary education should include elements to identify, prevent and treat mental health problems. Many students consult their GPs, which indicates that GPs could be central in such programs. Acknowledgements The authors thank Magne Solheim for advice on statistical methods. Conflicts of interest: None declared. Key points Both internalizing and externalizing mental health problems are associated with dropout from upper secondary education. A large proportion of students not finishing upper secondary education at the expected age had consulted a GP about a mental health problem. GPs can be assigned a more central role in public health strategies to reduce school dropout. References 1 Sagatun A, Wentzel-Larsen T, Heyerdahl S, Lien L. Mental health in adolescence and subsequent receipt of medical benefits in young adulthood: the mediating role of upper secondary school completion. Scand J Public Health  2016; 44: 431– 8. Google Scholar CrossRef Search ADS PubMed  2 De Ridder KAA, Pape K, Johnsen R, et al.   School dropout: a major public health challenge: a 10-year prospective study on medical and non-medical social insurance benefits in young adulthood, the Young-HUNT 1 Study (Norway). J Epidemiol Commun Health  2012; 66: 995– 1000. Google Scholar CrossRef Search ADS   3 Kieling C, Baker-Henningham H, Belfer M, et al.   Child and adolescent mental health worldwide: evidence for action. Lancet  2011; 378: 1515– 25. Google Scholar CrossRef Search ADS PubMed  4 Bor W, Dean AJ, Najman J, Hayatbakhsh R. Are child and adolescent mental health problems increasing in the 21st century? A systematic review. Aust N Z J Psychiatry  2014; 48: 606– 16. Google Scholar CrossRef Search ADS PubMed  5 Sagatun A, Heyerdahl S, Wentzel-Larsen T, Lien L. Mental health problems in the 10th grade and non-completion of upper secondary school: the mediating role of grades in a population-based longitudinal study. BMC Public Health  2014; 14: 16. Google Scholar CrossRef Search ADS PubMed  6 Esch P, Bocquet V, Pull C, et al.   The downward spiral of mental disorders and educational attainment: a systematic review on early school leaving. BMC Psychiatry  2014; 14: 237. Google Scholar CrossRef Search ADS PubMed  7 Melkevik O, Nilsen W, Evensen M, et al.   Internalizing disorders as risk factors for early school leaving: a systematic review. Adolesc Res Rev  2016; 1: 245– 55. Google Scholar CrossRef Search ADS   8 Van der Ende J, Verhulst FC, Tiemeier H. The bidirectional pathways between internalizing and externalizing problems and academic performance from 6 to 18 years. Dev Psychopathol  2016; 28: 855– 67. Google Scholar CrossRef Search ADS PubMed  9 Veldman K, Bultmann U, Stewart RE, et al.   Mental health problems and educational attainment in adolescence: 9-year follow-up of the TRAILS study. PLoS One  2014; 9: e101751. Google Scholar CrossRef Search ADS PubMed  10 De Ridder KA, Pape K, Johnsen R, et al.   Adolescent health and high school dropout: a prospective cohort study of 9000 Norwegian adolescents (the Young-HUNT). PLoS One  2013; 8: e74954. Google Scholar CrossRef Search ADS PubMed  11 Homlong L, Rosvold EO, Haavet OR. Can use of healthcare services among 15-16-year-olds predict an increased level of high school dropout? A longitudinal community study. BMJ Open  2013; 3: e003125. Google Scholar CrossRef Search ADS PubMed  12 Vanneste YTM, Mathijssen JJP, Goor ILAMvd, et al.   Extensive medical absenteeism among secondary school students: an observational study on their health condition from a biopsychosocial perspective. Open J Prev Med  2015; 5: 111. Google Scholar CrossRef Search ADS   13 Reiss F. Socioeconomic inequalities and mental health problems in children and adolescents: a systematic review. Soc Sci Med  2013; 90: 24– 31. Google Scholar CrossRef Search ADS PubMed  14 Sirin SR. Socioeconomic status and academic achievement: a meta-analytic review of research. Rev Educ Res  2005; 75: 417– 53. Google Scholar CrossRef Search ADS   15 Battin-Pearson S, Newcomb MD, Abbott RD, et al.   Predictors of early high school dropout: a test of five theories. J Educ Psychol  2000; 92: 568– 82. Google Scholar CrossRef Search ADS   16 Norway S. Throughput of pupils in upper secondary education, 2010–2015: Statistics Norway; 2016 [updated 02.06.2016. Available at: https://www.ssb.no/en/utdanning/statistikker/vgogjen/aar/2016-06-02. 17 Askeland KG, Haugland S, Stormark KM, et al.   Adolescent school absenteeism and service use in a population-based study. BMC Public Health  2015; 15: 626. Google Scholar CrossRef Search ADS PubMed  18 The Norwegian Directorate of eHealth. ICPC-2e – English version Oslo2017 [updated 20.08.2017. Available at: https://ehelse.no/icpc-2e-english-version. 19 Schulte-Korne G. Mental health problems in a school setting in children and adolescents. Dtsch Arztebl Int  2016; 113: 183– 90. Google Scholar PubMed  20 Christensen KS, Haugen W, Sirpal MK, Haavet OR. Diagnosis of depressed young people—criterion validity of WHO-5 and HSCL-6 in Denmark and Norway. Fam Pract  2015; 32: 359– 63. Google Scholar CrossRef Search ADS PubMed  21 Sund AM, Larsson B, Wichstrom L. Prevalence and characteristics of depressive disorders in early adolescents in Central Norway. Child Adolesc Psychiatry Ment Health  2011; 5: 28. Google Scholar CrossRef Search ADS PubMed  22 Lundervold A, Breivik K, Posserud M-B, et al.   Symptoms of depression as reported by Norwegian adolescents on the short mood and feelings questionnaire. Front Psychol  2013; 4: 613 Google Scholar CrossRef Search ADS PubMed  23 Stoet G, Geary DC. Sex differences in academic achievement are not related to political, economic, or social equality. Intelligence  2015; 48: 137– 51. Google Scholar CrossRef Search ADS   24 Wang Y, Hunt K, Nazareth I, et al.   Do men consult less than women? An analysis of routinely collected UK general practice data. BMJ Open  2013; 3: e003320. Google Scholar CrossRef Search ADS PubMed  25 Waddell G, Burton K, Aylward M. Work and common health problems. J Insur Med  2007; 39: 109– 20. Google Scholar PubMed  26 Hestbaek L, Leboeuf-Yde C, Kyvik KO, Manniche C. The course of low back pain from adolescence to adulthood: eight-year follow-up of 9600 twins. Spine  2006; 31: 468– 72. Google Scholar CrossRef Search ADS PubMed  27 Huurre T, Rahkonen O, Komulainen E, Aro H. Socioeconomic status as a cause and consequence of psychosomatic symptoms from adolescence to adulthood. Soc Psychiatry Psychiatr Epidemiol  2005; 40: 580– 7. Google Scholar CrossRef Search ADS PubMed  28 Myrtveit SM, Sivertsen B, Skogen JC, et al.   Adolescent neck and shoulder pain—the association with depression, physical activity, screen-based activities, and use of health care services. J Adolesc Health  2014; 55: 366– 72. Google Scholar CrossRef Search ADS PubMed  29 Parr AK, Bonitz VS. Role of family background, student behaviors, and school-related beliefs in predicting high school dropout. J Educ Res  2015; 108: 504– 14. Google Scholar CrossRef Search ADS   30 Joling KJ, van Marwijk HW, Piek E, et al.   Do GPs' medical records demonstrate a good recognition of depression? a new perspective on case extraction. J Affect Disord  2011; 133: 522– 7. Google Scholar CrossRef Search ADS PubMed  31 Mojtabai R, Stuart EA, Hwang I, et al.   Long-term effects of mental disorders on educational attainment in the National Comorbidity Survey ten-year follow-up. Soc Psychiatry Psychiatr Epidemiol  2015; 50: 1577– 91. Google Scholar CrossRef Search ADS PubMed  32 Kessler RC, Foster CL, Saunders WB, Stang PE. Social consequences of psychiatric disorders. I: educational attainment. Am J Psychiatry  1995; 152: 1026– 32. Google Scholar CrossRef Search ADS PubMed  33 Verhaak PF, van Dijk M, Walstock D, Zwaanswijk M. A new approach to child mental healthcare within general practice. BMC Fam Pract  2015; 16: 132. Google Scholar CrossRef Search ADS PubMed  34 Haller DM, Sanci LA, Patton GC, Sawyer SM. Toward youth friendly services: a survey of young people in primary care. J Gen Intern Med  2007; 22: 775– 81. Google Scholar CrossRef Search ADS PubMed  35 Haller DM, Sanci LA, Sawyer SM, Patton GC. The identification of young people's emotional distress: a study in primary care. Br J Gen Pract  2009; 59: e61. 70. Google Scholar CrossRef Search ADS PubMed  36 Hafting M, Garlov I. “You may wade through them without seeing them” : general practitioners and their young patients with mental health problems. Nord J Psychiatry  2009; 63: 256– 9. Google Scholar CrossRef Search ADS PubMed  37 Hafting M, Garlov I. [General practitioner and mental diseases among children and adolescents]. Tidsskr Nor Laegeforen  2006; 126: 299– 301. Google Scholar PubMed  38 Roberts JH, Crosland A, Fulton J. “I think this is maybe our Achilles heel…” exploring GPs' responses to young people presenting with emotional distress in general practice: a qualitative study. BMJ Open  2013; 3: e002927. Google Scholar CrossRef Search ADS PubMed  39 Fortin L, Marcotte D, Potvin P, et al.   Typology of students at risk of dropping out of school: description by personal, family and school factors. Eur J Psychol Educ  2006; 21: 363. Google Scholar CrossRef Search ADS   40 Cederberg M, Hartsmar N. Some aspects of early school leaving in Sweden, Denmark, Norway and Finland. Eur J Educ  2013; 48: 378– 89. Google Scholar CrossRef Search ADS   © The Author(s) 2018. Published by Oxford University Press on behalf of the European Public Health Association. All rights reserved. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png The European Journal of Public Health Oxford University Press

GP-diagnosed internalizing and externalizing problems and dropout from secondary school: a cross-sectional study

Loading next page...
 
/lp/ou_press/gp-diagnosed-internalizing-and-externalizing-problems-and-dropout-from-40Z2CFiKIv
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of the European Public Health Association. All rights reserved.
ISSN
1101-1262
eISSN
1464-360X
D.O.I.
10.1093/eurpub/cky026
Publisher site
See Article on Publisher Site

Abstract

Abstract Background Dropout from secondary education is a major concern in many Western countries because it is associated with later adverse consequences at the individual and societal level. Efforts have thus been made to identify precursors for dropout. The aim of the study was to examine the risk for not finishing secondary education by age 20 when mental health problems were diagnosed during general practitioner (GP) consultations. Methods National registries were linked to assess the association between GP-diagnosed internalizing and externalizing problems from the ages of 15–20 years and school dropout in a sample of 63 970 Norwegians, adjusting for health and social background factors. Relative risks (RR) were estimated by Poisson regression. Results Dropout was bivariately related to both internalizing (RR = 2.2 among girls and 1.7 among boys) and externalizing problems (RR = 2.7 for girls and 2.0 for boys), though the associations were somewhat attenuated in the adjusted analyses. After controlling for somatic comorbidity and parent education level, the absolute risk for not fulfilling secondary education by age 20 was 43% among girls and 60% among boys with internalizing problems and approximately 15% points higher with externalizing problems. The highest absolute risk for dropout was found for boys and girls who have both externalizing and internalizing problems. However, with some overlap in the confidence intervals, the added impact of internalizing problems when added to externalizing problems is uncertain. Conclusions Intervention for mental health problems by a GP could benefit adolescent education outcomes and mental well-being. Introduction The large proportion of adolescents who fail to complete secondary education is an area of great concern in many western countries. In Norway, nearly one third of students never finish upper secondary education, which is usually attended by students between the ages of 16 and 19 years. Truncating one’s educational path in this way is related to later adverse consequences on an individual and societal level, such as lack of participation in the labour market and early dependence on social insurance benefits.1,2 Mental health problems are found among 10–20% of adolescents, and studies indicate an increasing frequency.3,4 Such problems are an early risk indicator and predictor of early school departure.5–9 Externalizing mental health problems, including symptoms of ADHD and behaviour problems, have consistently shown an association with school dropout.5–8 The association between internalizing problems such as anxiety and depression and dropout from school is less consistent and not robust to adjustments for externalizing problems.7 One previous prospective register-based study found associations between early internalizing problems and school dropout, but for girls only.5 Further, most studies have not accounted for comorbidity, but some studies suggest that the association between internalizing problems and school dropout is partly accounted for by co-occurring externalizing problems.5,7 Having somatic health problems during adolescence also increases the risk for not completing higher secondary education.10,11 Undiagnosed medical problems were the most frequent health issue among students with extensive absenteeism.12 However, there is a shortage of studies that consider both somatic and psychological problems as predictors for early school departure. In studying the association between mental health problems and dropout from school, it is important to account for family socioeconomic status (SES), and especially parent education levels. Family SES is repeatedly found to be associated with adolescent mental health,13 and parent education level is a significant predictor of academic achievement and school dropout;14,15 it is strongly associated with the likelihood of their offspring completing upper secondary education.16 General practitioners (GPs) are the first line of health care for young people and are the most frequently used health services among adolescents.17 One study found more frequent GP consultations among adolescents with higher school absenteeism,17 while another have reported more frequent GP consultations to be associated with lower risk of school dropout among boys, with no such association among girls.11 Thus, GPs may be reliable informants on mental health in adolescents and may also be important for early intervention. Aims The overall aim of the current study was to assess the association between mental health problems and school dropout in a national register including 63 970 young adults. More specifically, we wanted to assess the frequency of internalizing and externalizing problems during GP consultations and to study associations with completion of upper secondary education by age 20. Further, the independent risk associated with both internalizing and externalizing problems was explored, while accounting for relevant social background factors and other health complaints. Methods Study design and setting This cross-sectional study is based on register data from 2006 to 2011 covering a national cohort of persons born in Norway in 1991. Data sources The regular GP database, containing information about all regular GPs contracted to municipalities and the identity of the patients on each GP’s list, including age and gender. The GPs send claims of fee-for-service for each patient contact to the Norwegian Health Economics Administration, collected in the KUHR database. The National Educational Database was used for information on completed education. Parent information was retrieved from the National Population Register. Statistics Norway linked the registers and provided de-identified data to use for research. Ethics The study had permission from the Data Inspectorate and all respective register owners to use these data sources for the current research purposes. Sample All inhabitants born in 1991 that where assigned to a GP in 2011 (N = 66 516) were included. There were some participants for which information about their education level (N = 2546) and their parents’ education level (N = 5859) was unavailable. This resulted in a sample size of 63 970 (48.5% girls) for the bivariate analyses, and 58 111 (47.9% girls) for the multivariate analyses. GP visits and diagnoses from claims data The number of GP visits from 15- to 20-year-olds were obtained from the KUHR database in the 2006–2011 period. For each consultation, the claims included one or more diagnoses according to the International Classification of Primary Health Care (ICPC-2). The ICPC-2 coding system is organized with chapters for ‘organ systems’ including a general chapter (A), a psychological chapter (P) and a social chapter (Z).18 Internalizing and externalizing mental health problems were defined according to specific diagnostic codes in the P-chapter of the ICPC-2 coding system. Internalizing problems are covered in the ICPC codes P01-03, P11, P27-29, P73-80, P82, P86 or P99, where anxiety and depression were the most frequent diagnoses. Externalizing problems are presented in codes P04, P15-16, P18-19, P22-23 or P81, where hyperactivity and behavioural problems were most frequent. These groups of GP diagnoses were the main predictors in the analyses. In addition, some of the most frequent diagnoses (excluding infections and reproductive health issues) used by GPs in consultations with patients in this age group were used for adjustment purposes in the regression models: sleep disturbance/insomnia (P06), tiredness/fatigue (A04), headache (N01), neck or back pain (L01-03, L18, L83-85) and asthma or allergy (R96-97). Education and dropping out Compulsory education in Norway is 10 years, with primary and lower secondary school finished at age 16. Then follows a right to a further 3 years of education in upper secondary school, also including vocational training for 1 extra year. In the current study, dropping out/not finishing upper secondary education was operationalized as not having completed 13 years of education by the age of 20 years. Statistical analyses Student’s T- and chi-squared tests were used to compare frequencies of GP contact and type of diagnoses in groups of students who had or had not completed schooling by age 20. Poisson regression models with robust estimates of variance were used to calculate relative risk (RR) for not finishing upper secondary education by age 20, with separate models for boys and girls, presenting both univariate estimates and multivariate analyses. Gender, diagnoses from GP consultations, the number of GP consultations and parent education level (completed upper secondary school or not) were the main predictors. We also calculated the number of GP consultations with nonpsychological diagnoses as a general indicator of somatic health problems, used as predictor in regression models in addition to the selected diagnoses. The multivariate models take into account the interaction between internalizing and externalizing problems, present the RR for having either internalizing or externalizing problems and the interaction term. After regression, the Stata command lincom was used to test for significances of adding internalizing to externalizing problems and vice versa, and the margins command was used to estimate the absolute risk for dropout controlled for the other variables in the model. Sensitivity analyses performed on the distribution of diagnoses and GP visits using different samples did not indicate any selection bias owing to missing parent education information. Stata Statistical Software (Release 14; College Station, TX, USA) was used for all statistical analyses. Results Among the 63 970 persons included, 69.3% of the girls and 54.0% of boys had finished upper secondary education by age 20. Girls had an average annual consultation rate of 1.6 in the total study period compared with 0.9 among boys. In both genders, approximately 10% of consultations were related to a mental health problem (table 1), and 23.4% of girls and 16.5% of boys had ≥1 consultation(s) related to a psychological diagnosis in the study period. Table 1 Overview of education by age 20 years, use of a GP from age 15 to 20 years, and parent education level among all persons aged 20 in 2011 included in the Norwegian list patient system   Girls  Boys  Proportion finished upper secondary education by age 20, %  69.3  54.0  Number of GP consultations from age 15 to 20 years, mean  9.60  5.53  Number of GP consultations with any psychological diagnosis (P in ICPC) from age 15 to 20 years, mean  0.97  0.58  Total number of GP consultations from 2006 to 2011, proportion (%) within each group:      0  4.4  9.8      1–4  26.3  43.3      5–9  30.8  30.2      ≥10  38.5  16.7  Number of GP consultations with psychological diagnosis (P in ICPC) in 2006–2011, proportion (%) within each group:      0  76.6  83.5      1–2  13.3  9.9      3–5  5.6  4.2      ≥6  4.5  2.4  Proportion (%) with ≥1 GP consultation(s) with a diagnosis of:      Internalizing psychological problemsa          Age 15–17 years  8.4  3.4      Age 18–20 years  16.5  8.5      Age 15–20 years (whole period)  20.6  10.4      Externalizing psychological problemsb          Age 15–17 years  2.6  3.6      Age 18–20 years  2.9  4.2      Age 15–20 years (whole period)  4.3  6.1      Both externalizing and internalizing (also included above)          Age 15–20 years (whole period)  2.3  2.0  Insomnia, fatigue, headache, neck or back pain and/or asthma or allergy (age 15–20 years)c  52.0  38.8  Parent education <13 years (%)      Mother  36.7  36.6      Father  34.5  34.5    Girls  Boys  Proportion finished upper secondary education by age 20, %  69.3  54.0  Number of GP consultations from age 15 to 20 years, mean  9.60  5.53  Number of GP consultations with any psychological diagnosis (P in ICPC) from age 15 to 20 years, mean  0.97  0.58  Total number of GP consultations from 2006 to 2011, proportion (%) within each group:      0  4.4  9.8      1–4  26.3  43.3      5–9  30.8  30.2      ≥10  38.5  16.7  Number of GP consultations with psychological diagnosis (P in ICPC) in 2006–2011, proportion (%) within each group:      0  76.6  83.5      1–2  13.3  9.9      3–5  5.6  4.2      ≥6  4.5  2.4  Proportion (%) with ≥1 GP consultation(s) with a diagnosis of:      Internalizing psychological problemsa          Age 15–17 years  8.4  3.4      Age 18–20 years  16.5  8.5      Age 15–20 years (whole period)  20.6  10.4      Externalizing psychological problemsb          Age 15–17 years  2.6  3.6      Age 18–20 years  2.9  4.2      Age 15–20 years (whole period)  4.3  6.1      Both externalizing and internalizing (also included above)          Age 15–20 years (whole period)  2.3  2.0  Insomnia, fatigue, headache, neck or back pain and/or asthma or allergy (age 15–20 years)c  52.0  38.8  Parent education <13 years (%)      Mother  36.7  36.6      Father  34.5  34.5  Notes: N = 31 010 girls and 32 960 boys. Corresponding to ICPC codes ((https://ehelse.no/icpc-2e-english-version).): a P01-03, P11, P27-29, P73-80, P82, P86, or P99; b P04, P15-16, P18-19, P22-23, or P81; c P06, A04, N01, L01-03, L18, L83-85, R96-97. Table 1 Overview of education by age 20 years, use of a GP from age 15 to 20 years, and parent education level among all persons aged 20 in 2011 included in the Norwegian list patient system   Girls  Boys  Proportion finished upper secondary education by age 20, %  69.3  54.0  Number of GP consultations from age 15 to 20 years, mean  9.60  5.53  Number of GP consultations with any psychological diagnosis (P in ICPC) from age 15 to 20 years, mean  0.97  0.58  Total number of GP consultations from 2006 to 2011, proportion (%) within each group:      0  4.4  9.8      1–4  26.3  43.3      5–9  30.8  30.2      ≥10  38.5  16.7  Number of GP consultations with psychological diagnosis (P in ICPC) in 2006–2011, proportion (%) within each group:      0  76.6  83.5      1–2  13.3  9.9      3–5  5.6  4.2      ≥6  4.5  2.4  Proportion (%) with ≥1 GP consultation(s) with a diagnosis of:      Internalizing psychological problemsa          Age 15–17 years  8.4  3.4      Age 18–20 years  16.5  8.5      Age 15–20 years (whole period)  20.6  10.4      Externalizing psychological problemsb          Age 15–17 years  2.6  3.6      Age 18–20 years  2.9  4.2      Age 15–20 years (whole period)  4.3  6.1      Both externalizing and internalizing (also included above)          Age 15–20 years (whole period)  2.3  2.0  Insomnia, fatigue, headache, neck or back pain and/or asthma or allergy (age 15–20 years)c  52.0  38.8  Parent education <13 years (%)      Mother  36.7  36.6      Father  34.5  34.5    Girls  Boys  Proportion finished upper secondary education by age 20, %  69.3  54.0  Number of GP consultations from age 15 to 20 years, mean  9.60  5.53  Number of GP consultations with any psychological diagnosis (P in ICPC) from age 15 to 20 years, mean  0.97  0.58  Total number of GP consultations from 2006 to 2011, proportion (%) within each group:      0  4.4  9.8      1–4  26.3  43.3      5–9  30.8  30.2      ≥10  38.5  16.7  Number of GP consultations with psychological diagnosis (P in ICPC) in 2006–2011, proportion (%) within each group:      0  76.6  83.5      1–2  13.3  9.9      3–5  5.6  4.2      ≥6  4.5  2.4  Proportion (%) with ≥1 GP consultation(s) with a diagnosis of:      Internalizing psychological problemsa          Age 15–17 years  8.4  3.4      Age 18–20 years  16.5  8.5      Age 15–20 years (whole period)  20.6  10.4      Externalizing psychological problemsb          Age 15–17 years  2.6  3.6      Age 18–20 years  2.9  4.2      Age 15–20 years (whole period)  4.3  6.1      Both externalizing and internalizing (also included above)          Age 15–20 years (whole period)  2.3  2.0  Insomnia, fatigue, headache, neck or back pain and/or asthma or allergy (age 15–20 years)c  52.0  38.8  Parent education <13 years (%)      Mother  36.7  36.6      Father  34.5  34.5  Notes: N = 31 010 girls and 32 960 boys. Corresponding to ICPC codes ((https://ehelse.no/icpc-2e-english-version).): a P01-03, P11, P27-29, P73-80, P82, P86, or P99; b P04, P15-16, P18-19, P22-23, or P81; c P06, A04, N01, L01-03, L18, L83-85, R96-97. A GP diagnosis reflecting an internalizing mental health problem was registered among 20.6% of girls and 10.4% of boys during these 6 years, while 4–6% had externalizing problems. GP-diagnosed mental health problems and dropout The 6-year incident rate of GP consultations for internalizing problems was 2–3 times higher among both girls and boys that had not finished secondary school by age 20 compared with those who had finished (table 2). Table 2 Total number of GP consultations, frequency of internalizing and externalizing diagnoses and parent education level by gender and the child’s education status by age 20 in a cohort born in 1991 followed from age 15 to 20 years   Finished upper secondary education at age 20  Girls  Boys  Yes  No  Yes  No  N (% within gender)  21 489 (69.3%)  9521 (30.7%)  17 795 (54.0%)  15 165 (46.0%)  Number of GP consultations in total from age 15 to age 20 years, mean (95% CI)  8.05 (7.95–8.14)  13.10 (12.88–13.32)*  4.79 (4.72–4.85)  6.40 (6.30–6.50)*  Number of GP consultations with any psychological diagnosis (P in ICPC) from age 15 to 20 years, mean (95% CI)  0.50 (0.47–0.53)  1.88 (1.81–1.97)*  0.22 (0.20–0.23)  1.00 (0.95–1.04)*  Number of GP consultations in 2006–2011 grouped, proportion (%) within each group:      0  5.0  3.1**  11.4  8.0**      1–4  30.3  17.2**  46.7  39.3**      5–9  33.3  25.0**  29.6  30.9**      ≥10  31.4  54.7**  12.3  21.8**  Proportion (%) with ≥1 GP consultation(s) during 2006–2011 with a diagnosis of:      Internalizing psychological problems  14.0  35.5**  5.6  16.0**      Externalizing psychological problems  1.6  10.5**  1.8  11.1**      Both internalizing and externalizing (also included above)  0.6  6.2**  0.4  3.9**      Insomnia, fatigue, headache, neck or back pain and/or asthma or allergy  48.8  59.8**  35.9  42.5**  Parent education <13 years, (%)      Mother  30.3  51.7**  27.5  47.6**      Father  29.0  47.7**  26.1  44.9**    Finished upper secondary education at age 20  Girls  Boys  Yes  No  Yes  No  N (% within gender)  21 489 (69.3%)  9521 (30.7%)  17 795 (54.0%)  15 165 (46.0%)  Number of GP consultations in total from age 15 to age 20 years, mean (95% CI)  8.05 (7.95–8.14)  13.10 (12.88–13.32)*  4.79 (4.72–4.85)  6.40 (6.30–6.50)*  Number of GP consultations with any psychological diagnosis (P in ICPC) from age 15 to 20 years, mean (95% CI)  0.50 (0.47–0.53)  1.88 (1.81–1.97)*  0.22 (0.20–0.23)  1.00 (0.95–1.04)*  Number of GP consultations in 2006–2011 grouped, proportion (%) within each group:      0  5.0  3.1**  11.4  8.0**      1–4  30.3  17.2**  46.7  39.3**      5–9  33.3  25.0**  29.6  30.9**      ≥10  31.4  54.7**  12.3  21.8**  Proportion (%) with ≥1 GP consultation(s) during 2006–2011 with a diagnosis of:      Internalizing psychological problems  14.0  35.5**  5.6  16.0**      Externalizing psychological problems  1.6  10.5**  1.8  11.1**      Both internalizing and externalizing (also included above)  0.6  6.2**  0.4  3.9**      Insomnia, fatigue, headache, neck or back pain and/or asthma or allergy  48.8  59.8**  35.9  42.5**  Parent education <13 years, (%)      Mother  30.3  51.7**  27.5  47.6**      Father  29.0  47.7**  26.1  44.9**  Notes: Register-based study, national data. * P < 0.001, Student’s T-test. ** P < 0.001, Pearson’s chi-squared (comparing ‘not finished’ to ‘finished’ upper secondary education by age 20). Table 2 Total number of GP consultations, frequency of internalizing and externalizing diagnoses and parent education level by gender and the child’s education status by age 20 in a cohort born in 1991 followed from age 15 to 20 years   Finished upper secondary education at age 20  Girls  Boys  Yes  No  Yes  No  N (% within gender)  21 489 (69.3%)  9521 (30.7%)  17 795 (54.0%)  15 165 (46.0%)  Number of GP consultations in total from age 15 to age 20 years, mean (95% CI)  8.05 (7.95–8.14)  13.10 (12.88–13.32)*  4.79 (4.72–4.85)  6.40 (6.30–6.50)*  Number of GP consultations with any psychological diagnosis (P in ICPC) from age 15 to 20 years, mean (95% CI)  0.50 (0.47–0.53)  1.88 (1.81–1.97)*  0.22 (0.20–0.23)  1.00 (0.95–1.04)*  Number of GP consultations in 2006–2011 grouped, proportion (%) within each group:      0  5.0  3.1**  11.4  8.0**      1–4  30.3  17.2**  46.7  39.3**      5–9  33.3  25.0**  29.6  30.9**      ≥10  31.4  54.7**  12.3  21.8**  Proportion (%) with ≥1 GP consultation(s) during 2006–2011 with a diagnosis of:      Internalizing psychological problems  14.0  35.5**  5.6  16.0**      Externalizing psychological problems  1.6  10.5**  1.8  11.1**      Both internalizing and externalizing (also included above)  0.6  6.2**  0.4  3.9**      Insomnia, fatigue, headache, neck or back pain and/or asthma or allergy  48.8  59.8**  35.9  42.5**  Parent education <13 years, (%)      Mother  30.3  51.7**  27.5  47.6**      Father  29.0  47.7**  26.1  44.9**    Finished upper secondary education at age 20  Girls  Boys  Yes  No  Yes  No  N (% within gender)  21 489 (69.3%)  9521 (30.7%)  17 795 (54.0%)  15 165 (46.0%)  Number of GP consultations in total from age 15 to age 20 years, mean (95% CI)  8.05 (7.95–8.14)  13.10 (12.88–13.32)*  4.79 (4.72–4.85)  6.40 (6.30–6.50)*  Number of GP consultations with any psychological diagnosis (P in ICPC) from age 15 to 20 years, mean (95% CI)  0.50 (0.47–0.53)  1.88 (1.81–1.97)*  0.22 (0.20–0.23)  1.00 (0.95–1.04)*  Number of GP consultations in 2006–2011 grouped, proportion (%) within each group:      0  5.0  3.1**  11.4  8.0**      1–4  30.3  17.2**  46.7  39.3**      5–9  33.3  25.0**  29.6  30.9**      ≥10  31.4  54.7**  12.3  21.8**  Proportion (%) with ≥1 GP consultation(s) during 2006–2011 with a diagnosis of:      Internalizing psychological problems  14.0  35.5**  5.6  16.0**      Externalizing psychological problems  1.6  10.5**  1.8  11.1**      Both internalizing and externalizing (also included above)  0.6  6.2**  0.4  3.9**      Insomnia, fatigue, headache, neck or back pain and/or asthma or allergy  48.8  59.8**  35.9  42.5**  Parent education <13 years, (%)      Mother  30.3  51.7**  27.5  47.6**      Father  29.0  47.7**  26.1  44.9**  Notes: Register-based study, national data. * P < 0.001, Student’s T-test. ** P < 0.001, Pearson’s chi-squared (comparing ‘not finished’ to ‘finished’ upper secondary education by age 20). GP-diagnosed externalizing problems were found among 1.6% of girls and 1.8% of boys when upper secondary school was finished by age 20, the corresponding figures were 10.5% and 11.1% among those who dropped out of school (table 2). For both internalizing and externalizing psychological problems, the RR for dropout was higher among girls than boys (table 3). When adjusting for other health problems and parent education level the RR for dropout associated with internalizing problems was 1.85 (95% CI: 1.78–1.92) among girls and 1.49 (95% CI: 1.45–1.54) among boys. For externalizing problems, the corresponding figures were 2.48 (95% CI: 2.32–2.66) for girls and 1.84 (95% CI: 1.78–1.90) for boys. Table 3 Associations between not finished secondary school by age 20 and internalizing and externalizing problems according to diagnoses in ≥1 GP consultation(s) during the age period of 15–20 years, adjusted for other common health problems and parent education   Girls  Boys  Crude rates  Multivariate modela  Crude rates  Multivariate modela    RR  95% CI  RR  95% CI  RR  95% CI  RR  95% CI  Internalizing problems  2.23  2.16–2.31  1.85  1.78–1.92  1.68  1.63–1.72  1.49  1.44–1.54  Externalizing problems  2.74  2.63–2.84  2.48  2.32–2.66  2.00  1.95–2.05  1.84  1.78–1.90  Internalizing × Externalizingb      0.60  0.55–0.65      0.70  0.66–0.74  Number of GP consultationsc  0  Ref.    Ref.    Ref.      Ref.  1–4  1.0  0.88–1.13  0.95  0.84–1.07  1.09  1.04–1.15  1.04  0.99–1.09  5–9  1.24  1.10–1.40  1.07  0.95–1.21  1.16  1.10–1.22  1.07  1.01–1.12  ≥10  2.0  1.77–2.25  1.44  1.28–1.63  1.41  1.33–1.48  1.17  1.11–1.23  Parent education <13 years  Mother  1.83  1.77–1.90  1.55  1.50–1.61  1.58  1.54–1.62  1.40  1.37–1.44  Father  1.72  1.66–1.78  1.43  1.38–1.48  1.54  1.50–1.58  1.37  1.33–1.40    Girls  Boys  Crude rates  Multivariate modela  Crude rates  Multivariate modela    RR  95% CI  RR  95% CI  RR  95% CI  RR  95% CI  Internalizing problems  2.23  2.16–2.31  1.85  1.78–1.92  1.68  1.63–1.72  1.49  1.44–1.54  Externalizing problems  2.74  2.63–2.84  2.48  2.32–2.66  2.00  1.95–2.05  1.84  1.78–1.90  Internalizing × Externalizingb      0.60  0.55–0.65      0.70  0.66–0.74  Number of GP consultationsc  0  Ref.    Ref.    Ref.      Ref.  1–4  1.0  0.88–1.13  0.95  0.84–1.07  1.09  1.04–1.15  1.04  0.99–1.09  5–9  1.24  1.10–1.40  1.07  0.95–1.21  1.16  1.10–1.22  1.07  1.01–1.12  ≥10  2.0  1.77–2.25  1.44  1.28–1.63  1.41  1.33–1.48  1.17  1.11–1.23  Parent education <13 years  Mother  1.83  1.77–1.90  1.55  1.50–1.61  1.58  1.54–1.62  1.40  1.37–1.44  Father  1.72  1.66–1.78  1.43  1.38–1.48  1.54  1.50–1.58  1.37  1.33–1.40  Notes: N = 27 849 girls and 29 643 boys; Poisson regression models, register based study, national data. a Multivariate model also adjusted for: insomnia, fatigue, headache, back or neck pain, asthma and allergy. b Adjusting for interaction between internalizing and externalizing in multivariate models. c Number of GP consultations in 2006–2011, psychological reasons excluded. Table 3 Associations between not finished secondary school by age 20 and internalizing and externalizing problems according to diagnoses in ≥1 GP consultation(s) during the age period of 15–20 years, adjusted for other common health problems and parent education   Girls  Boys  Crude rates  Multivariate modela  Crude rates  Multivariate modela    RR  95% CI  RR  95% CI  RR  95% CI  RR  95% CI  Internalizing problems  2.23  2.16–2.31  1.85  1.78–1.92  1.68  1.63–1.72  1.49  1.44–1.54  Externalizing problems  2.74  2.63–2.84  2.48  2.32–2.66  2.00  1.95–2.05  1.84  1.78–1.90  Internalizing × Externalizingb      0.60  0.55–0.65      0.70  0.66–0.74  Number of GP consultationsc  0  Ref.    Ref.    Ref.      Ref.  1–4  1.0  0.88–1.13  0.95  0.84–1.07  1.09  1.04–1.15  1.04  0.99–1.09  5–9  1.24  1.10–1.40  1.07  0.95–1.21  1.16  1.10–1.22  1.07  1.01–1.12  ≥10  2.0  1.77–2.25  1.44  1.28–1.63  1.41  1.33–1.48  1.17  1.11–1.23  Parent education <13 years  Mother  1.83  1.77–1.90  1.55  1.50–1.61  1.58  1.54–1.62  1.40  1.37–1.44  Father  1.72  1.66–1.78  1.43  1.38–1.48  1.54  1.50–1.58  1.37  1.33–1.40    Girls  Boys  Crude rates  Multivariate modela  Crude rates  Multivariate modela    RR  95% CI  RR  95% CI  RR  95% CI  RR  95% CI  Internalizing problems  2.23  2.16–2.31  1.85  1.78–1.92  1.68  1.63–1.72  1.49  1.44–1.54  Externalizing problems  2.74  2.63–2.84  2.48  2.32–2.66  2.00  1.95–2.05  1.84  1.78–1.90  Internalizing × Externalizingb      0.60  0.55–0.65      0.70  0.66–0.74  Number of GP consultationsc  0  Ref.    Ref.    Ref.      Ref.  1–4  1.0  0.88–1.13  0.95  0.84–1.07  1.09  1.04–1.15  1.04  0.99–1.09  5–9  1.24  1.10–1.40  1.07  0.95–1.21  1.16  1.10–1.22  1.07  1.01–1.12  ≥10  2.0  1.77–2.25  1.44  1.28–1.63  1.41  1.33–1.48  1.17  1.11–1.23  Parent education <13 years  Mother  1.83  1.77–1.90  1.55  1.50–1.61  1.58  1.54–1.62  1.40  1.37–1.44  Father  1.72  1.66–1.78  1.43  1.38–1.48  1.54  1.50–1.58  1.37  1.33–1.40  Notes: N = 27 849 girls and 29 643 boys; Poisson regression models, register based study, national data. a Multivariate model also adjusted for: insomnia, fatigue, headache, back or neck pain, asthma and allergy. b Adjusting for interaction between internalizing and externalizing in multivariate models. c Number of GP consultations in 2006–2011, psychological reasons excluded. The absolute risk for dropout with different combinations of problems, after controlling for the adjusting variables, are shown in table 4. Table 4 Absolute riska for not having finished secondary school by age 20 according to internalizing and externalizing problems in GP diagnoses in any consultation during the age period 15–20 years   Girls  Boys    Risk for drop-out, %  95% CI  Risk for drop-out, %  95% CI  Neither internalizing nor externalizing  23.1  22.6–23.7  40.3  39.7–40.9  Internalizing - not externalizing  42.7  41.5–44.0  60.2  58.4–61.9  Externalizing - not internalizing  57.5  54.8–61.1  74.2  72.0–76.4  Internalizing and externalizing  63.6  60.6–66.6  77.2  74.5–80.0    Girls  Boys    Risk for drop-out, %  95% CI  Risk for drop-out, %  95% CI  Neither internalizing nor externalizing  23.1  22.6–23.7  40.3  39.7–40.9  Internalizing - not externalizing  42.7  41.5–44.0  60.2  58.4–61.9  Externalizing - not internalizing  57.5  54.8–61.1  74.2  72.0–76.4  Internalizing and externalizing  63.6  60.6–66.6  77.2  74.5–80.0  Notes: N = 27 849 girls and 29 643 boys; register based study, national data. a Marginal effects, giving the risk estimates adjusted for somatic comorbidity and parent education (based on a Poisson regression model, see table 3). Table 4 Absolute riska for not having finished secondary school by age 20 according to internalizing and externalizing problems in GP diagnoses in any consultation during the age period 15–20 years   Girls  Boys    Risk for drop-out, %  95% CI  Risk for drop-out, %  95% CI  Neither internalizing nor externalizing  23.1  22.6–23.7  40.3  39.7–40.9  Internalizing - not externalizing  42.7  41.5–44.0  60.2  58.4–61.9  Externalizing - not internalizing  57.5  54.8–61.1  74.2  72.0–76.4  Internalizing and externalizing  63.6  60.6–66.6  77.2  74.5–80.0    Girls  Boys    Risk for drop-out, %  95% CI  Risk for drop-out, %  95% CI  Neither internalizing nor externalizing  23.1  22.6–23.7  40.3  39.7–40.9  Internalizing - not externalizing  42.7  41.5–44.0  60.2  58.4–61.9  Externalizing - not internalizing  57.5  54.8–61.1  74.2  72.0–76.4  Internalizing and externalizing  63.6  60.6–66.6  77.2  74.5–80.0  Notes: N = 27 849 girls and 29 643 boys; register based study, national data. a Marginal effects, giving the risk estimates adjusted for somatic comorbidity and parent education (based on a Poisson regression model, see table 3). The presence of either internalizing or externalizing problems increased the absolute risk of dropout for both boys and girls relative to those without such problems. The absolute risk for dropout was increased for those having both externalizing and internalizing problems (girls: 63.6%, boys: 77.2%) compared with those having only externalizing problems (girls: 57.5%, boys: 74.2%). However, the overlap in confidence intervals (most pronounced for boys) suggests that some uncertainty remains regarding the impact of having internalizing problems when externalizing problems were present. As sensitivity analyses, we ran regression models including internalizing or externalizing problems during age 15–17 years and 18–20 years, respectively. The associations with dropout were similar for boys independent of the age period used. For girls, however, the association between dropout and internalizing psychological problems reported during age 15–17 years was lower (RR 1.58, [1.51–1.65]) than for the same diagnoses reported in the later 3-year period (RR 1.76, [1.69–1.82]) (adjusted analyses, not tabulated). Discussion The aim of the current study was to investigate GP-diagnosed internalizing and externalizing mental health problems from age 15 to 20 years as predictors of school dropout by age 20 in a sample of 63 970 young adults from Norway. The dropout rate at 20 years was 31% for girls and 46% for boys in the total study population. The risk for school dropout was substantially increased for both girls and boys presenting with internalizing or externalizing problems. The associations with school dropout were attenuated somewhat, but remained significant after accounting for parent education level and other health complaints. The dropout risk for girls was 42.7% and for boys was 60.2% for internalizing problems and, respectively, 57.5% and 74.2% when externalizing problems were found in GP diagnoses. Having both kinds of problems was associated with the highest absolute risk for dropout for both genders. However, with somewhat overlapping confidence intervals, the impact of comorbid internalizing problems should therefore be interpreted with some caution. Comparison with the literature Although GP diagnoses are not indicators of population prevalence of mental health problems, the 6-year consultation prevalence for mental health problems are in line with prevalence estimates in adolescents.19 The prevalence of internalizing problems in the present study is comparable to the 10% found in a survey among 16-year-olds,20 but clearly lower than the 23% lifetime prevalence of depression among adolescents found in another study.21 Internalizing problems were relatively frequent in the current sample, especially among girls, and in line with population based studies.22 Externalizing problems were less commonly observed, but somewhat more often in boys relative to girls. The association between dropout and mental health problems is strong in the present study, in line with earlier studies.5,9 Both internalizing and externalizing mental health problems showed independent associations with school noncompletion. While the association between externalizing problems is in accordance with the literature,5–8 the current study further strengthens the importance of internalizing problems as an independent risk factor for school dropout. In the present study more than one-third of girls in the dropout group had a GP-diagnosed internalizing problem. The combination of high frequency and high associated risk for school noncompletion indicates that this might have important public health implications and could indicate that internalizing problems have a larger impact than what has been found in other studies that indicated less clear associations.7,9 In general, dropout is more frequent among boys and there is a concern for how boys manage in modern school systems because of their lower achievement compared with girls, giving the boys problems in passing exams and therefore not completing education.23 However, there may also be gender differences in how symptoms are described to GPs, and this may account for some of the gender differences in associations between dropout and health problems that are reported by GPs.24 Health complaints stabilize from the time of adolescence to adulthood25,26 and are associated with low work–life participation later in life.27 Furthermore, health complaints often co-occur with mental health problems.28 In this study, while adjusting for these health complaints in the final model, mental health problems remained a significant predictor of school dropout. This confirms the high rate of co-occurring somatic health complaints and mental health problems in this age group. Having parents who have not fulfilled upper secondary education was a strong predictor for dropout, even when adjusting for health problems, in line with the established associations between socioeconomic background and academic performance and dropout from secondary school.14,29 Strengths and limitations A main strength of the current study is that it is based on a near complete population sample and data collected from national registers. The fee-for-service financing model in the list patient system results in registration of all patient contact, and is a more reliable measure of utilization than the self-reported data commonly used in similar studies. The high coverage limits selection bias and increases external validity of the findings. The other registers used are maintained by national authorities where quality control is highly prioritized. A limitation is that the majority of claims report only one diagnosis when, in a large proportion of consultations, the GPs deal with multiple problems. Probably the reported ICPC-2 diagnoses underestimate the number of psychological issues raised in the consultations.30 In addition, the precision of diagnoses is uncertain,30 but by grouping diagnosis in internalizing and externalizing mental health problems, the impact of misclassifications should be reduced. In the present cross-sectional study, it is not possible to interpret associations as causal relationships. With the design using GP consultations from the whole age span framing the upper secondary school period, the relation in time between dropout and symptoms cannot be assessed and a temporal relationship is not certain. When timing of the diagnosis was included, there was no difference related to the time period used in association with school completion in boys, but there was a stronger association between later diagnoses among girls. Previous studies with prospective designs have found mental health problems present prior to dropout from school,10,31,32 but increases in internalizing symptoms after dropping out have also been documented.6 Future studies using longitudinal designs are needed to further elucidate the temporal order between mental health problems and school dropout. Implications The clear association found in this study between recognized mental health problems among young people in encounters with GPs and school dropout supports the policy of working for improving mental health service in primary care. However, there are challenges to recognize, follow up and properly treat mental health problems among young people.33 Under-presentation of problems seems to be common in this age group, which challenges a GP’s skills to recognize emotional problems.34,35 GPs might be in a good position to work within the mental health field and with adolescents. They may build relations through different health care contacts over time and have knowledge of local context.36 It is a paradox, however, that the majority of adolescents with a recognized mental health problem have rather few consultations with GPs where that problem seems to be central. This may be explained by the finding that GPs consider themselves as not sufficiently skilled to work with these problems.37,38 Improving the GPs’ knowledge and giving them more confidence at working with youth mental health problems have been shown to be effective33 and could be implemented and evaluated in Norway. Last, the associations between leaving school early and mental health problems are complex and the solutions may be related to solving education problems or improving the psychosocial situation of the student.39,40 There is a need for cooperation between the GP, which may be a ‘recognizer’, other health care providers and the schools. Improved cooperation is challenging because municipal health care services often are fragmented. To include the GP as a resource in a multidisciplinary work, there is a need for both a mutual interest in competencies and a clarification of each other’s roles. An increase of resources, such as money and time, is also necessary to make it possible to foster cooperation between the GP and other stakeholders. Conclusion Mental health problems seem to have a great impact on risk for school dropout in upper secondary education, and measures to reduce dropout from upper secondary education should include elements to identify, prevent and treat mental health problems. Many students consult their GPs, which indicates that GPs could be central in such programs. Acknowledgements The authors thank Magne Solheim for advice on statistical methods. Conflicts of interest: None declared. Key points Both internalizing and externalizing mental health problems are associated with dropout from upper secondary education. A large proportion of students not finishing upper secondary education at the expected age had consulted a GP about a mental health problem. GPs can be assigned a more central role in public health strategies to reduce school dropout. References 1 Sagatun A, Wentzel-Larsen T, Heyerdahl S, Lien L. Mental health in adolescence and subsequent receipt of medical benefits in young adulthood: the mediating role of upper secondary school completion. Scand J Public Health  2016; 44: 431– 8. Google Scholar CrossRef Search ADS PubMed  2 De Ridder KAA, Pape K, Johnsen R, et al.   School dropout: a major public health challenge: a 10-year prospective study on medical and non-medical social insurance benefits in young adulthood, the Young-HUNT 1 Study (Norway). J Epidemiol Commun Health  2012; 66: 995– 1000. Google Scholar CrossRef Search ADS   3 Kieling C, Baker-Henningham H, Belfer M, et al.   Child and adolescent mental health worldwide: evidence for action. Lancet  2011; 378: 1515– 25. Google Scholar CrossRef Search ADS PubMed  4 Bor W, Dean AJ, Najman J, Hayatbakhsh R. Are child and adolescent mental health problems increasing in the 21st century? A systematic review. Aust N Z J Psychiatry  2014; 48: 606– 16. Google Scholar CrossRef Search ADS PubMed  5 Sagatun A, Heyerdahl S, Wentzel-Larsen T, Lien L. Mental health problems in the 10th grade and non-completion of upper secondary school: the mediating role of grades in a population-based longitudinal study. BMC Public Health  2014; 14: 16. Google Scholar CrossRef Search ADS PubMed  6 Esch P, Bocquet V, Pull C, et al.   The downward spiral of mental disorders and educational attainment: a systematic review on early school leaving. BMC Psychiatry  2014; 14: 237. Google Scholar CrossRef Search ADS PubMed  7 Melkevik O, Nilsen W, Evensen M, et al.   Internalizing disorders as risk factors for early school leaving: a systematic review. Adolesc Res Rev  2016; 1: 245– 55. Google Scholar CrossRef Search ADS   8 Van der Ende J, Verhulst FC, Tiemeier H. The bidirectional pathways between internalizing and externalizing problems and academic performance from 6 to 18 years. Dev Psychopathol  2016; 28: 855– 67. Google Scholar CrossRef Search ADS PubMed  9 Veldman K, Bultmann U, Stewart RE, et al.   Mental health problems and educational attainment in adolescence: 9-year follow-up of the TRAILS study. PLoS One  2014; 9: e101751. Google Scholar CrossRef Search ADS PubMed  10 De Ridder KA, Pape K, Johnsen R, et al.   Adolescent health and high school dropout: a prospective cohort study of 9000 Norwegian adolescents (the Young-HUNT). PLoS One  2013; 8: e74954. Google Scholar CrossRef Search ADS PubMed  11 Homlong L, Rosvold EO, Haavet OR. Can use of healthcare services among 15-16-year-olds predict an increased level of high school dropout? A longitudinal community study. BMJ Open  2013; 3: e003125. Google Scholar CrossRef Search ADS PubMed  12 Vanneste YTM, Mathijssen JJP, Goor ILAMvd, et al.   Extensive medical absenteeism among secondary school students: an observational study on their health condition from a biopsychosocial perspective. Open J Prev Med  2015; 5: 111. Google Scholar CrossRef Search ADS   13 Reiss F. Socioeconomic inequalities and mental health problems in children and adolescents: a systematic review. Soc Sci Med  2013; 90: 24– 31. Google Scholar CrossRef Search ADS PubMed  14 Sirin SR. Socioeconomic status and academic achievement: a meta-analytic review of research. Rev Educ Res  2005; 75: 417– 53. Google Scholar CrossRef Search ADS   15 Battin-Pearson S, Newcomb MD, Abbott RD, et al.   Predictors of early high school dropout: a test of five theories. J Educ Psychol  2000; 92: 568– 82. Google Scholar CrossRef Search ADS   16 Norway S. Throughput of pupils in upper secondary education, 2010–2015: Statistics Norway; 2016 [updated 02.06.2016. Available at: https://www.ssb.no/en/utdanning/statistikker/vgogjen/aar/2016-06-02. 17 Askeland KG, Haugland S, Stormark KM, et al.   Adolescent school absenteeism and service use in a population-based study. BMC Public Health  2015; 15: 626. Google Scholar CrossRef Search ADS PubMed  18 The Norwegian Directorate of eHealth. ICPC-2e – English version Oslo2017 [updated 20.08.2017. Available at: https://ehelse.no/icpc-2e-english-version. 19 Schulte-Korne G. Mental health problems in a school setting in children and adolescents. Dtsch Arztebl Int  2016; 113: 183– 90. Google Scholar PubMed  20 Christensen KS, Haugen W, Sirpal MK, Haavet OR. Diagnosis of depressed young people—criterion validity of WHO-5 and HSCL-6 in Denmark and Norway. Fam Pract  2015; 32: 359– 63. Google Scholar CrossRef Search ADS PubMed  21 Sund AM, Larsson B, Wichstrom L. Prevalence and characteristics of depressive disorders in early adolescents in Central Norway. Child Adolesc Psychiatry Ment Health  2011; 5: 28. Google Scholar CrossRef Search ADS PubMed  22 Lundervold A, Breivik K, Posserud M-B, et al.   Symptoms of depression as reported by Norwegian adolescents on the short mood and feelings questionnaire. Front Psychol  2013; 4: 613 Google Scholar CrossRef Search ADS PubMed  23 Stoet G, Geary DC. Sex differences in academic achievement are not related to political, economic, or social equality. Intelligence  2015; 48: 137– 51. Google Scholar CrossRef Search ADS   24 Wang Y, Hunt K, Nazareth I, et al.   Do men consult less than women? An analysis of routinely collected UK general practice data. BMJ Open  2013; 3: e003320. Google Scholar CrossRef Search ADS PubMed  25 Waddell G, Burton K, Aylward M. Work and common health problems. J Insur Med  2007; 39: 109– 20. Google Scholar PubMed  26 Hestbaek L, Leboeuf-Yde C, Kyvik KO, Manniche C. The course of low back pain from adolescence to adulthood: eight-year follow-up of 9600 twins. Spine  2006; 31: 468– 72. Google Scholar CrossRef Search ADS PubMed  27 Huurre T, Rahkonen O, Komulainen E, Aro H. Socioeconomic status as a cause and consequence of psychosomatic symptoms from adolescence to adulthood. Soc Psychiatry Psychiatr Epidemiol  2005; 40: 580– 7. Google Scholar CrossRef Search ADS PubMed  28 Myrtveit SM, Sivertsen B, Skogen JC, et al.   Adolescent neck and shoulder pain—the association with depression, physical activity, screen-based activities, and use of health care services. J Adolesc Health  2014; 55: 366– 72. Google Scholar CrossRef Search ADS PubMed  29 Parr AK, Bonitz VS. Role of family background, student behaviors, and school-related beliefs in predicting high school dropout. J Educ Res  2015; 108: 504– 14. Google Scholar CrossRef Search ADS   30 Joling KJ, van Marwijk HW, Piek E, et al.   Do GPs' medical records demonstrate a good recognition of depression? a new perspective on case extraction. J Affect Disord  2011; 133: 522– 7. Google Scholar CrossRef Search ADS PubMed  31 Mojtabai R, Stuart EA, Hwang I, et al.   Long-term effects of mental disorders on educational attainment in the National Comorbidity Survey ten-year follow-up. Soc Psychiatry Psychiatr Epidemiol  2015; 50: 1577– 91. Google Scholar CrossRef Search ADS PubMed  32 Kessler RC, Foster CL, Saunders WB, Stang PE. Social consequences of psychiatric disorders. I: educational attainment. Am J Psychiatry  1995; 152: 1026– 32. Google Scholar CrossRef Search ADS PubMed  33 Verhaak PF, van Dijk M, Walstock D, Zwaanswijk M. A new approach to child mental healthcare within general practice. BMC Fam Pract  2015; 16: 132. Google Scholar CrossRef Search ADS PubMed  34 Haller DM, Sanci LA, Patton GC, Sawyer SM. Toward youth friendly services: a survey of young people in primary care. J Gen Intern Med  2007; 22: 775– 81. Google Scholar CrossRef Search ADS PubMed  35 Haller DM, Sanci LA, Sawyer SM, Patton GC. The identification of young people's emotional distress: a study in primary care. Br J Gen Pract  2009; 59: e61. 70. Google Scholar CrossRef Search ADS PubMed  36 Hafting M, Garlov I. “You may wade through them without seeing them” : general practitioners and their young patients with mental health problems. Nord J Psychiatry  2009; 63: 256– 9. Google Scholar CrossRef Search ADS PubMed  37 Hafting M, Garlov I. [General practitioner and mental diseases among children and adolescents]. Tidsskr Nor Laegeforen  2006; 126: 299– 301. Google Scholar PubMed  38 Roberts JH, Crosland A, Fulton J. “I think this is maybe our Achilles heel…” exploring GPs' responses to young people presenting with emotional distress in general practice: a qualitative study. BMJ Open  2013; 3: e002927. Google Scholar CrossRef Search ADS PubMed  39 Fortin L, Marcotte D, Potvin P, et al.   Typology of students at risk of dropping out of school: description by personal, family and school factors. Eur J Psychol Educ  2006; 21: 363. Google Scholar CrossRef Search ADS   40 Cederberg M, Hartsmar N. Some aspects of early school leaving in Sweden, Denmark, Norway and Finland. Eur J Educ  2013; 48: 378– 89. Google Scholar CrossRef Search ADS   © The Author(s) 2018. Published by Oxford University Press on behalf of the European Public Health Association. All rights reserved. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

The European Journal of Public HealthOxford University Press

Published: Mar 9, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off