In the current study, we evaluated the association between IDA at 9–18 months and ADHD during childhood in a large Israeli cohort (n = 58,471) of healthy children insured by CHS.
Our findings showed no association between IDA at age 9–18-months and ADHD development during childhood, thus rejecting our hypothesis.
The IDA rate in our cohort was 3.8%, which is comparable to the country's IDA rates. According to CHS data, IDA rates in the general pediatric population in the country between 2005 and 2013 were 5–8%, while the rate in 2014 was about 3.4% [25]. Our findings reflect a downward trend in the prevalence of IDA in Israel over time, which is likely due to increased awareness of iron supplementation in babies and IDA prevention amongst the general population in Israel. The prevalence of IDA in 9-18-month-old children was higher among those with low compared to medium or high socioeconomic status. IDA prevalence was higher among orthodox Jews (a population characterized by large numbers of children and relatively low socioeconomic status) than among Arabs (who had medium rates of IDA) and secular-traditional Jews (who had the lowest rates of IDA). These findings are consistent with the literature, and indicate that lower socioeconomic groups have a greater prevalence of anemia [26]. A possible explanation is that parents from low socioeconomic status are less aware and are less likely to treat IDA. Preventive public health actions such as parental guidance, medical and social follow-up to support a balanced and iron-rich diet for mother and baby, and daily treatment with iron drops for infants according to the Ministry of Health’s protocol, may reduce the prevalence of IDA in these groups and improve public health in Israel.
Iron is a key component in brain development, particularly in infancy, and in brain function and dopaminergic activity regulation, all of which may be linked to the pathogenesis and symptoms of ADHD [15, 27, 28]. Animal models showed very clear irreversible abnormalities resulting from ID during pregnancy and infancy; however, in humans, the effects of ID in the first two years of life on cognitive and mental development in childhood are unclear [29].
Contrary to our hypothesis, we observed that IDA rate in the ADHD group (3.4%) was 0.6% lower than in the control group (4.0%), a difference that was small in magnitude yet statistically significant, due to the large sample size. Sensitivity analysis, in which the analysis was repeated in four random case-control matched samples, each with an adequate sample size to answer the primary outcome, showed higher rates of IDA, with statistical significance, in the ADHD group than in the control group only in two of four random samples. Comparing the rates of moderate ID anemia (Hb < 10gr/dl), significant between-group differences were not found (0.9% in the ADHD group vs. 1.0% in the control group). According to these results, ID is not a significant factor in the etiology of ADHD. Genetic factors, brain function and structure, acute or cumulative exposures to environmental toxins, prematurity, low birth weight, epilepsy, in utero brain damage, and severe head injury during life are among the factors that may be involved in the development of ADHD [30].
In the stratified analyses, significant differences in IDA rates between the ADHD and the control groups were found only among boys, among children of low socioeconomic status, and among secular-traditional Jews. All these statistically significant differences in IDA rates were small in magnitude (the maximal difference in IDA rate was 0.7%). The statistically significant findings among boys and among secular and traditional Jews, but not in the other stratified subgroups, may be attributed to the substantially larger proportion of boys compared to girls, and of secular-traditional Jews compared to the other cultural background subcategories. However, the significantly lower IDA rates in the ADHD compared to the control group, among those with low but not higher socioeconomic status, cannot be attributed only to large sample size. This is because the largest socioeconomic category was medium socioeconomic status, comprising about 46% of the sample. A plausible explanation for the results is a larger gap between parents with high compared to low awareness to their children's health in the low socioeconomic group. Parents with higher awareness are both more compliant with the recommendation of iron supplementation in babies and more aware of ADHD symptoms. They are more apt to seek evaluation and treatment, and hence their offspring may be expected to show a higher rate of diagnosed /treated ADHD and a lower rate of IDA. In contrast, parents with lower awareness may have less awareness to the importance of iron supplementation and may also be less aware of ADHD symptoms. Hence, their offspring may be expected to show lower rates of diagnosed /treated ADHD and higher rates of IDA. Similar differences in awareness may also help explain the results in the entire study cohort.
To the best of our knowledge, this is the first study that evaluated the association between IDA in childhood and the development of ADHD. This highlights the need for more research on the implications of IDA in childhood on cognitive and mental development later in life. A main strength of this study is the high reliability of the data, derived from the database of CHS. The exposure parameters (diagnosis of IDA) were recorded from a database of medical records, and not based on persons’ memory. Hence, a memory bias, which is common in case-control studies, was avoided. Furthermore, the definition of ADHD for inclusion in the study group required a diagnosis of ADHD, and receipt of a prescription by a medical specialist.
There are several methodological limitations to this study. First, the IDA definition was based solely on Hb and MCV values (Hb < 10.5gr/dl, 60fl < MCV < 75fl). Due to a lack of information in our database, ferritin and iron values were not considered. However, congenital anemia was ruled out, indicating that the main reason for anemia was indeed IDA. Second, the timing of IDA is an important factor in neurocognitive development [31]. In our study, IDA was evaluated at age 9–18 months, and we have no information from earlier life to determine the initiation of IDA (i.e., mother's diet iron content during pregnancy or whether the child had significant iron deficiency from birth to the examination of IDA). A third limitation is the definition of ADHD, which was based on medical prescription. This results in a non-differential misclassification of the outcome for the two groups. Accordingly, some children who meet the criteria for ADHD diagnosis may choose not to receive pharmaceutical treatment. Nonetheless, other studies have classified people as having ADHD according to medical treatment [19, 20, 32].
This study did not find an association between IDA at age 9–18 months and ADHD development during childhood. Further studies are warranted, to examine the effects of ID in the early developmental stages of the brain (during pregnancy and in the first weeks after birth) and other environmental factors, on the development of ADHD in childhood.