Educational impairment in social anxiety disorder and post-traumatic stress disorder

Resumen

ABSTRACT Background and aims Psychiatric disorders have generally been linked to academic undearchievement, but previous studies focusing on the association of social anxiety disorder (SAD) and post-traumatic stress disorder (PTSD) and educational outcomes had methodological limitations. Population-based administrative and health registers are often used for research purposes. In this sense, it is important that the diagnostic codes included in these registers are valid and reliable in order to conduct good quality epidemiological studies. This thesis aimed to validate the diagnostic code for SAD – the diagnostic code for PTSD had been previously validated – and, subsequenly, explore the association between SAD and PTSD with objective indicators of educational attainment across the lifespan by using the Swedish population-based registers. Methods The ICD-10 code for SAD (F40.1) was validated using chart review methods. Positive predictive values (PPV) and agreement between two raters (using Cohen’s kappa) were calculated. Associations between SAD and PTSD and educational outcomes were studied by means of two population-based birth cohort studies of all Swedish individuals born between 1973 and 1997 and followed up until 2013. The exposed individuals were those with registered diagnosis of SAD or PTSD in the National Patient Register. The educational outcomes under study were: elibility to acces upper secondary school, finish upper secondary education, start a university degree, obtain a university degree, and finish postgraduate education. Logistic regression models, adjusted by relevant covariates, tested the association between SAD or PTSD and the educational outcomes.The role of psychiatric comorbidities was also studied. Sibling analyses controlled for familial factors shared by full siblings. Results A total of 81% of the reviewed files to validate the code of SAD were considered to be 'true positive' cases (PPV=0.81). Inter-rater agreement regarding the presence or absence of SAD was substantial (κ=0.72). The register-based studies examining the association between SAD and PTSD with educational achievement included cohorts of 2,238,837 and 2,244,193 Swedish-born individuals, respectively. Individuals diagnosed with SAD were less likely to pass all subjects in the last year of compulsory education and less likely to be eligible for a vocational (adjusted odds ratio [aOR]=0.31) or an academic program (aOR=0.52) in upper secondary education, finish upper secondary education (aOR=0.19), start a university degree (aOR=0.47), obtain a university degree (aOR=0.35), and finish postgraduate education (aOR=0.58), compared to unexposed individuals. Similarly, a diagnosis of PTSD was associated with lower odds of achieving each of the assessed educational milestones during the study period, including 82% lower odds of finishing compulsory education (aOR=0.18), 87% lower odds of finishing upper secondary education (aOR=0.13), 68% lower odds of starting a university degree (aOR=0.32), and 73% lower odds of obtaining a university degree (aOR=0.27). In both epidemiological studies, the results remained largely unchanged when psychiatric comorbidities were taken into account. The sibling analyses showed still statistically significant but attenuated estimates, indicating that part of the observed associations could be explained by factors shared by siblings. Conclusion The diagnostic code for SAD in the Swedish National Patient Register is valid and reliable. Individuals with SAD or PTSD, as recorded in this register, are consistently less likely to achieve all educational milestones across the lifespan, over and above a number of confounders such as psychiatric comorbidities and familial factors. Early detection and intervention in these psychiatric disorders is warranted in order to alliviate the long-term adverse effect on academic performance.