In this study we found that half of the participants with borderline low IQ (i.e. IQ between 70 and 85) at 19 years of age met the criteria for a psychiatric diagnosis, compared with only one in ten in the reference group. This implied an almost five times higher odds for a psychiatric disorder for participants with borderline IQ, compared to the reference group. ADHD and anxiety were the most common diagnoses among those with borderline IQ. Thus, our results suggest that young adults with borderline IQ have increased risk of mental health problems. The higher scores on the Autism Questionnaire in this group should be interpreted with caution since the absolute differences were small and since the questionnaire is recommended for people with IQ above 85.
The results showed a higher occurrence of ADHD in the borderline IQ group according to the K-SADS P/L interview, while when the participants rated their own symptoms by completing the ADHD-RS, there were no significant differences between the groups. The different results may be explained by the fact that the K-SADS interview conducted by an experienced child- and adolescent psychiatrist, allows for elaborative questions, which may increase the validity of the findings, while the ADHD-RS is a subjective assessment of various symptoms completed on a fixed questionnaire. Such differences between self-perceived symptoms and symptoms assessed by others have been described in several papers [13, 39].
It is unlikely that our main findings are due to chance, as indicated by the very low p-values for most of the differences between the groups. The groups did not differ significantly in scores on the self-reported ADHD-RS, however, it should be noted that the differences in attention deficit and total score were close to statistically significance (0.05–0.1).
A limitation of the present study is that it was completed in a convenience sample where participants born prematurely or with low birth weight at term were overrepresented. However, when we adjusted for birth weight group, the estimated odds ratio for mental health disorder among individuals with borderline IQ was still considerably increased. Other potential confounders, such as mother’s age and SES did not affect the association between low IQ and mental health disorders. Regarding the results obtained through the AQ questionnaire, a significant limitation is that the Norwegian version of this instrument has not been properly validated in a Norwegian population, and moreover, that the English version is validated for individuals with IQ above 85. Thus, although highly statistically significant, the small differences in median values between the low IQ and the reference group could have been caused by interpretation differences between the two groups.
An important strength of this study is that an experienced neuropsychologist and an experienced child and adolescent psychiatrist performed the assessments. They were blinded to earlier test results and also to each other’s test assessments, and to birth weight group. Also, the comprehensive assessment of mental health problems using the K-SADS P/L semi-structured interview, and the restrictive definition of mental health problems may be considered to strengthen the study. IQ score was calculated using a full WAIS. Thus it is unlikely that our findings are due to misclassification of cognitive functioning or mental health problems.
Our findings are consistent with several other studies [9, 40]. However, most previous studies have examined the risk for mental disorders among individuals with intellectual disability (IQ < 70) [10, 41, 42]. The Isle of Wight study found three to four times higher occurrence of psychiatric disease among participants with low IQ, defined as an IQ score below 70 compared to participants with normal IQ [40]. Similar results were reported by Borthwick-Duffy et al. [42], Strømme and Diseth [41] and Emerson [10]. The latter author also found a seven times higher risk for psychiatric disease among children with assumed intellectual disability. Emerson did not measure intellectual disability in the standard way, but defined intellectual disability as having learning disabilities, going to special schools or having mental age < 80% of chronological age. Emerson also found increased risk for any emotional disorders (OR 2.4, CI: 1.6–3.7), with post-traumatic stress disorder (PTSD) (OR 4.8, CI 1.1–20.9) being one of the largest contributors. Other findings included increased risk of generalized anxiety disorder (OR 2.6, CI 1.0–7.4) and depression (OR 1.7, CI 0.6–4.7) in the low IQ group [10].
Lamanna et al. found intellectual disability to be a risk factor for ADHD. They found that 22% of the participants with ADHD had coexisting intellectual disability [43]. Fernell E. et al. found that among adolescents diagnosed with ASD, 38% had intellectual disability [44].
The main findings of the present study may be considered consistent with all these studies. However, the rather novel finding in our study is that participants with borderline IQ seem to have as high a risk for psychiatric diagnoses as earlier studies have found among participants with intellectual disability. Our results may also be in keeping with one of the few studies in persons with low IQ defined as IQ below 85 [9]. In that study Koenen et al. found that high IQ protected against anxiety, social phobia and PTSD, while low IQ was a risk factor for anxiety, schizophrenia and depression [9].
There are at least two principally different plausible interpretations of the association between borderline IQ and the presence of a psychiatric disorder. First, IQ may be seen as an innate, largely genetically determined, and more or less constant trait [12], which however, may be modified by early environmental influences and/or global insults to the brain. A predisposition for different psychiatric disorders may also be inherited, i.e. be genetically determined, but the risk for psychiatric disorders may further be modified by early mal-development of the brain, such as seen in children born very preterm. Thus, it is possible that the association between below average IQ and psychiatric diagnosis do have some common cause. This connection is not unlikely, as neuro-developmental symptoms and disorders were present in our sample. This may partly explain the association in our population, since a large proportion of the individuals were born very preterm, and since others and we have shown that very preterm born individuals have high risk for cognitive and mental disorders, that may last into adulthood [13, 14, 27, 45]. Moreover, psychiatric disorders may be precipitated during childhood and adolescence by bullying, low self-esteem due to specific learning difficulties and physical deficits, or by other factors such as major life events; divorce, loss of a parent etc. [46].
An additional interpretation may therefore be that individuals with borderline IQ are more vulnerable to mental stress than individuals with average IQ. Thus, they may be more likely to develop a mental disorder as a consequence of their low IQ. This interpretation may be supported by the fact that the association between borderline IQ and mental health problems persisted when we adjusted for birth weight group, although residual confounding cannot be excluded. Moreover, this second interpretation would be in line with the notion that individuals with low IQ have reduced cognitive reserve capacity. Such reduced capacity may result in poor coping strategies, which can lead to mental health problems such as anxiety and depression [9]. Cognitive impairments will affect the individuals’ functioning in daily life and cause learning difficulties, while deficits in executive functions and attention may result in diagnoses such as ADHD/ADD. Poor academic and social coping strategies may lead to mental health disorders such as anxiety and depression [15, 17]. Problems with expressing their needs may lead to behaviour problems [9, 17]. This may become more visible as the individual gets older, especially when they start school and have to face increased demands from the environment.
The apparent dose-response relationship, whereby none of the participants with IQ above 115 had a psychiatric diagnosis, and the high ORs for diagnoses in the borderline IQ group compared to the reference group would be in line with the criteria proposed by Doll and Hill to indicate causal relationship [47]. Moreover, this interpretation may also be consistent with a timing aspect, since it may be argued that IQ is mainly determined in early childhood, while emotional disorders develop later in life [48].
Most likely the basis for the association between borderline IQ and psychiatric diagnosis is a combination of an early genetic and/or environmental influence on brain development leading both to low IQ as well as to increased risk for psychiatric disorders, and the higher vulnerability for mental stress among persons with limited cognitive reserves.
Clinical implications
Regardless of whether low IQ is in the causal chain or if low IQ and mental disorders have a common aetiology, our findings have some important clinical implications. One implication is that kindergartens, schools and the society should pay particular attention to children and youths diagnosed with borderline IQ, being aware that these individuals are at increased risk of developing or having a psychiatric disorder. This may especially involve ADHD and anxiety disorders, and the risk of having more than one disorder. It is of particular importance to be aware that these children, often having learning difficulties, are also more likely to be bullied and develop low self-esteem [46]. However, even without early signs of delayed development or specific learning difficulties, individuals with borderline IQ may experience problems keeping up with the demands at school and in social interaction, as they grow older. Thus, schools should provide special tutoring and extra care to potentially prevent development of psychiatric disease also in children with borderline IQ, but without intellectual disability.
