Lower Serum Selenium Concentration Associated with Anxiety in Children

Abstract

Purpose:

Few studies have examined the role of selenium in anxiety. This study aimed to evaluate the association between serum selenium concentrations and anxiety disorders and symptoms in children.

Design and Methods:

This study utilized data from 831 children participating in the China Jintan Child Cohort Study (mean age = 12.67 years; 46.1% female). Serum selenium samples were collected and anxiety was assessed using the Chinese version of the Screen for Child Anxiety Related Disorders. Six types of anxiety scores were calculated, including total anxiety, panic/somatic, generalized anxiety, separation anxiety, social anxiety, and school phobia.

Results:

Controlling for covariates, children with lower serum selenium concentrations were more likely to meet clinical cutoffs for total anxiety (OR = 0.992, p < 0.01), panic/somatic disorder (OR = 0.993, p < 0.05), generalized anxiety disorder (OR = 0.990, p < 0.05), social anxiety disorder (OR = 0.991, p < 0.01), and school phobia (OR = 0.989, p < 0.01), but not separation anxiety (OR = 1.000, p > 0.05). Controlling for covariates, lower serum selenium concentrations were also associated with higher continuous total anxiety, generalized anxiety, and school phobia scores (p < 0.05).

Conclusions:

Lower serum selenium concentrations were associated with higher anxiety. To our knowledge, this was the first study to examine the relationship between serum selenium and anxiety disorders in a sample of children. Results indicate an association between children’s micronutrient levels and anxiety disorders.

Practice Implications:

Improving child nutrition may be a promising strategy to help reduce childhood anxiety.

Introduction

Anxiety is among the most common mental disorders affecting males and females across the lifespan. Anxiety disorders—which include social anxiety disorder, separation anxiety, panic disorder, generalized anxiety disorder, as well as other related disorders—share common features of excessive anxiety and related behavioral disturbances (American Psychiatric Association, 2013). Anxiety can have negative effects on various outcomes, including school performance (Carey et al., 2017), relationships (Mughal et al., 1996), health (Kroenke et al., 2013; Meuret et al., 2020), and substance abuse (Bushnell et al., 2019). According to global and national surveys, about 3.6% of the global population has an anxiety disorder (World Health Organization, 2017). Anxiety tends to be relatively common among youth (Liu et al., 2018). Understanding child and adolescent anxiety is important, as this is the primary period of risk for the development of subclinical symptoms of anxiety, as well as anxiety disorders (Beesdo et al., 2009). Studies in children and adolescents have identified risk factors for anxiety across multiple domains, including biology, temperament, adverse life events, parenting, and genetics (Cabral & Patel, 2020). Relatively little research has examined the role of nutrition in the development of anxiety, despite the fact that the association between micronutrient inadequacy and other adverse mental health outcomes has been receiving increasing attention (Dickerman & Liu, 2011; Kaplan et al., 2015; Liu et al., 2014; Wang et al., 2018).

Micronutrients collectively describe the essential vitamins and trace elements needed for proper functioning of the human body. Inadequate micronutrient intake is an important contributor to the global burden of disease (Beal et al., 2017). Common dietary sources of selenium include seafood, bread, grains, meat, poultry, fish, and eggs. Because selenium intake is often in the form of grains, there is substantial regional variation in the selenium content of foods depending on the soil in which it is grown. Recently, studies examining the role of selenium concentrations on health outcomes have implicated its role in several disorders, including thyroid disease (Ventura et al., 2017), adverse reproductive health (Hsieh et al., 2020), and diabetes (Vinceti et al., 2016). Selenium’s protective effect is thought to be linked to the activities of its selenoproteins, which modulate numerous cellular processes through its antioxidant, anti-inflammatory, and regulatory effects (Steinbrenner et al., 2016). The critical role of selenoproteins in fighting oxidative stress and modulating neuronal growth has led to a growing body of research examining the role of selenium in mental health disorders, including depression (Colangelo et al., 2014; Conner et al., 2015; Gao et al., 2012; Ghimire et al, 2019.; Gosney et al., 2008; Li et al., 2018; Wang et al., 2018). Although depression and anxiety are highly comorbid (Cummings et al., 2014; Mobach et al., 2019), the role of selenium in anxiety disorders has been relatively understudied.

Studies investigating the relationship between selenium concentrations and short-term mood states have largely found a beneficial effect of selenium supplementation, particularly when initial selenium concentrations are suboptimal (Rayman, 2000). An intervention study demonstrated an improvement in anxious mood in a British sample of 14–74 year olds after five weeks of selenium supplementation (Benton & Cook, 1991). Similarly, Finley and Penland (1998) reported decreased anxious mood following selenium supplementation in adult men. Lower serum selenium concentrations have also been associated with a greater incidence of depressed, anxious, confused and hostile mood states in adults (Finley & Penland, 1998; Hawkes & Hornbostel, 1996). In sample of older adults living in residential facilities, Gosney et al. (2008) found a trend toward a negative relationship between serum selenium concentration and anxiety, though micronutrient supplementation did not reduce anxiety. These findings are suggestive that selenium may be linked to anxiety.

Despite the increasing volume of literature studying the role of selenium in mental health outcomes, the few studies that have examined the relationship between selenium and anxious mood have largely focused on adults (Benton & Cook, 1990, 1991; Finley & Penland, 1998; Hawkes & Hornbostel, 1996). Given the critical neurodevelopmental changes that occur during childhood, the present study aims to evaluate the association between serum selenium concentrations and anxiety in a community sample of Chinese children.

Methods

Participants and Procedures

The current study was part of the larger China Jintan Child Cohort Study, an on-going prospective longitudinal study with the aim of exploring early health risk factors on children’s neurobehavioral development. Details from this cohort study are described elsewhere (Liu et al., 2015; Liu et al., 2011b; Liu et al., 2010). Participants were drawn from four preschools chosen to represent the entire city’s geographical, social, and economic profiles. In fall 2004, 3- to 5-y-old children attending the four preschools and their parents were invited to participate in this study. In the present study, a total of 846 children from the Wave 2 follow up (n = 1100) completed a self-reported anxiety assessment using the Screen for Child Anxiety Related Disorders (SCARED) questionnaire and completed a blood laboratory assessment involving micronutrient sampling. These children had a mean age of 12.67 years (SD = 0.92; Range: 11–14). Written consents were obtained from both children and parents. Institutional Review Board approval was obtained from the University of Pennsylvania and the ethical committee for research at Jintan Hospital in China.

Measures

Collection and analysis of serum selenium samples.

Children’s serum samples were collected by trained pediatric nurses in Jintan Hospital from 2011–2013 when children were in their final month of sixth grade (last month of elementary school). Approximately 0.5 mL of venous blood was collected in a lead-free ethylenediaminetetraacetic acid (EDTA) tube and kept frozen at −20 degrees Celsius until analysis. All samples were shipped to the Research Center for Environmental Medicine of Children at Shanghai Jiaotong University for analysis. Serum selenium concentration was determined by atomic absorption spectrophotometry (BH model 5.100, Beijing Bohu Innovative Electronic Technology Corporation), with duplicate readings taken with an integration time of 2 (Liu et al., 2011a). One high outlier (selenium > 10 standard deviations from the mean) was excluded from the current analyses.

Anxiety measures.

Children’s anxiety was assessed using the Chinese version of the self-reported Screen for Child Anxiety Related Emotional Disorders (SCARED; Birmaher et al., 1997). This self-reported instrument consists of 41 items. Children responded to statements about their anxiety symptoms covering the last 3 months using a 3-point scale (0 = not true or hardly ever true; 1 = sometimes true; 2 = true or often true). Scores were summed to create total anxiety scores and five subscale anxiety scores, including panic/somatic disorder, generalized anxiety disorder, separation anxiety disorder, social anxiety disorder, and school phobia, with lower scores indicating fewer anxiety problems. The presence or absence of specific kinds of anxiety disorders was determined based on accepted clinical cutoffs (0 = does not meet clinical cutoff; 1 = meets clinical cutoff). This measure has been widely implemented within academic and professional settings, showing well-documented reliability and validity (Birmaher et al., 1997), including the Chinese version (Su et al., 2008). In the current sample, Cronbach’s α for the total anxiety scale was 0.94 and was also acceptable for all subscales (Cronbach’s α = 0.63–0.85).

When calculating anxiety scores, imputed anxiety scores were generated by multiplying the mean of completed values by the number of total items in order to keep as much collected information as possible, rather than eliminating children with fewer than three missing answers. One child with more than three missing answers was not included in the current analysis.

Covariates.

Sex (0 = male, 1 = female), age, location (rural [reference category], suburban, or city), father’s education (middle school or less [reference category], completed high school, or college or higher), and mother’s education (middle school or less [reference category], completed high school, or college or higher) were included as covariates. 13 children were missing parental education data, resulting in a final sample of 831 children with complete data.

Statistical Analysis

All analyses were performed using SPSS Statistics 26. Analyses were performed using the 831 children with complete data. Independent samples t-tests and one-way ANOVA were used to compare children with complete data to those with missing data on key study variables. Descriptive statistics were then calculated. Independent samples t-tests and one-way ANOVA were used to compare serum selenium concentrations across levels of categorical variables. To examine the association between serum selenium concentration and children’s continuous anxiety scores, bivariate Pearson correlation coefficients were computed. OLS regression models were used with serum selenium concentrations predicting continuous anxiety scores, controlling for covariates. Logistic regression models were used to examine the relationship between serum selenium concentrations and dichotomous anxiety scores, controlling for covariates.

Results

Sample Characteristics and Bivariate Analyses

Compared to children with complete data, Wave 2 children with missing data were slightly older (t = 10.12, df = 527.45, p < 0.001), were less likely to have a college-educated mother (χ² = 8.54, df = 2, p < 0.05) and father (χ² = 13.66, df = 2, p = 0.001), and were more likely to live in a rural location (χ² = 17.46 df =2, p < 0.001), but did not differ in terms of sex (χ² = 0.06, df = 1, p > 0.10). Full comparisons of children with complete and missing data can be found in Supplemental Table 1.

Table 1 presents the descriptive characteristics for the sociodemographic variables, serum selenium concentrations, and anxiety scores. The mean (SD) serum selenium concentration was 141.10 (39.45) μg/L. Serum selenium concentrations were lowest among children living in rural areas, F(2) = 12.69, p < 0.001, but did not differ by sex or parental education (p > 0.05).

Table 1.

Descriptive statistics and comparison of serum selenium concentration across levels of categorical variables (n = 831)

	%	Mean (SD)	Mean (SD) Selenium (pg/L)	t(df)/F(df)
Sex				t(829) = −0.16
Male	53.9%		140.90 (39.71)
Female	46.1%		141.34 (39.19)
Residence/School Location				F(2) = 12.69**
Rural	18.2%		127.20 (33.25)
Suburban	40.7%		142.16 (41.51)
City	41.2%		146.19 (38.56)
Father’s Highest Education				F(2) = 0.75
Middle school or below	29.0%		141.00 (39.25)
High School	28.9%		138.74 (36.88)
College or higher	42.1%		142.79 (41.29)
Mother’s Highest Education				F(2) = 1.11
Middle school or below	40.2%		139.03 (37.98)
High School	23.9%		144.27 (43.72)
College or higher	35.9%		141.31 (38.03)
Age (years)		12.67 (0.92)
Selenium (μg/L)		141.10 (39.45)
Total Anxiety		15.23 (12.66)
Panic/Somatic		3.88 (4.20)
Generalized Anxiety		3.48 (3.44)
Separation Anxiety		2.90 (2.77)
Social Anxiety		3.95 (3.24)
School Phobia		1.02 (1.42)

Note. Selenium = serum selenium concentrations. Total anxiety and anxiety sub-scores were measured using the Screen for Child Anxiety Related Emotional Disorders. t-tests and one-way ANOVA were performed to compare mean selenium concentration across levels of categorical variables.

^**p < 0.01.

As shown in Table 2, children who met the clinical cut-offs for total anxiety, panic disorder/somatic symptoms, generalized anxiety, social anxiety, and school phobia had lower mean serum selenium concentrations (p < 0.01). Bivariate correlations are shown in Table 3. Serum selenium concentration was significantly negatively associated with continuous generalized anxiety (r = −0.08, p < 0.05) and school phobia scores (r = −0.08, p < 0.05) and marginally negatively associated with continuous total anxiety (r = −0.07, p < 0.10) and panic/somatic scores (r = −0.06, p < 0.10). Serum selenium concentration was not significantly associated with social phobia (r = −0.03, p > 0.10) or separation anxiety (r = −0.04, p > 0.10).

Table 2.

Descriptive statistics and comparison of serum selenium concentration across anxiety diagnosis groups (n = 831)

	%	Mean (SD) Selenium (μg/L)	t(df)
Total Anxiety Diagnosis			3.06(251.84)**
Yes	18.8%	133.00 (35.93)
No	81.2%	142.98 (40.01)
Panic/Somatic Diagnosis
Yes	19.3%	133.80 (34.38)	2.89(273.91)**
No	80.7%	142.84 (40.40)
Generalized Anxiety Diagnosis			3.67(127.68)***
Yes	10.3%	129.85 (28.62)
No	89.7%	142.40 (40.33)
Separation Anxiety Diagnosis			0.60(829)
Yes	24.5%	139.67 (38.84)
No	75.5%	141.57 (39.67)
Social Anxiety Diagnosis			2.72(829)**
Yes	14.4%	132.06 (36.29)
No	85.6%	142.63 (39.78)
School Phobia Diagnosis			3.38(829)**
Yes	13.7%	129.59 (31.37)
No	86.3%	142.93 (40.30)

Note. Selenium = serum selenium concentrations. Total anxiety and anxiety sub-scores were measured using the Screen for Child Anxiety Related Emotional Disorders. Diagnosis group was based on whether the child met the clinical cut-off for the disorder.

^**p < 0.01.

^***p < 0.001.

Table 3.

Correlations between serum selenium concentration and children’s continuous anxiety scores (n = 831)

	1	2	3	4	5	6	7
1. Selenium	1
2. Total	−0.07†	1
3. Panic	−0.06†	0.91***	1
4. Generalized	−0.08*	0.89***	0.77***	1
5. Separation	−0.04	0.83***	0.70***	0.68**	1
6. Social	−0.03	0.78***	0.58***	0.60**	0.57***	1
7. School	−0.08*	0.67***	0.58***	0.58**	0.48***	0.38***	1

Note. Total = Total anxiety; Panic = Panic/somatic; Generalized = Generalized Anxiety; Separation = separation anxiety; Social = social anxiety; School = school phobia. Selenium = serum selenium concentration (μg/L)

^†p < 0.10.

^*p < 0.05.

^***p < 0.001.

Multivariate Analysis Predicting Continuous Anxiety Scores

Next, we examined separate OLS regressions of the continuous anxiety scores on serum selenium concentration, controlling for covariates. Results are shown in Table 4. Serum selenium concentration significantly predicted total anxiety (β = −0.09, p < 0.05), generalized anxiety (β = −0.11, p < 0.01), and school phobia (β = −0.10, p < 0.05). Serum selenium was also marginally significant as a predictor of panic/somatic symptoms (β = −0.07, p < 0.10). Serum selenium did not significantly predict separation anxiety (β = −0.04, p > 0.05) or social phobia (β = −0.06, p > 0.05).

Table 4.

OLS Regression of Continuous Anxiety Scores on Serum Selenium Concentration (μg/L; n = 831)

	B (SE)	β	p
Total Anxiety	−0.03 (0.01)	−0.09*	0.024
Panic/Somatic	−0.01 (0.00)	−0.07†	0.071
Generalized Anxiety	−0.01 (0.003)	−0.11**	0.004
Separation Anxiety	−0.003 (0.003)	−0.04	0.345
Social Phobia	−0.005 (0.003)	−0.06	0.13
School Phobia	−0.004 (.001)	−0.10*	0.011

Note. Dependent variables are anxiety scores. Regression coefficients are for selenium and are adjusted for sex, residence/school location, father’s education, mother’s education, and age.

^*p < 0.05,

^**p < 0.01,

^†p < 0.10.

Multivariate Analysis Predicting Dichotomous Anxiety Scores

Finally, we examined separate logistic regressions of the dichotomous anxiety scores (based on clinical cutoffs) on serum selenium concentration, controlling for covariates. Results are shown in Table 5. Lower serum selenium concentration was associated with a higher likelihood of a diagnosis of total anxiety (OR = 0.992, p < 0.01), panic/somatic disorder (OR = 0.993, p < 0.05), generalized anxiety disorder (OR = 0.990, p < 0.05), social anxiety disorder (OR = 0.991, p < 0.01), and school phobia (OR = 0.989, p < 0.01), but was not associated with separation anxiety disorder (OR = 1.000, p > 0.10).

Table 5.

Logistic Regressions of Dichotomous Anxiety Scores on Serum Selenium Concentration (μg/L; n = 831)

	Odds Ratio (95% CI)	p
Total Anxiety	0.992 (0.987, 0.998)**	0.007
Panic/Somatic	0.993 (0.988, 0.999)*	0.013
Generalized Anxiety	0.990 (0.983, 0.998)*	0.010
Separation Anxiety	1.000 (0.996, 1.005)	0.881
Social Anxiety	0.991 (0.985, 0.998)**	0.006
School Phobia	0.989 (0.982, 0.995)**	0.001

Note. Dependent variables are dichotomous anxiety scores (0 = does not meet clinical cut-off; 1 = meets clinical cut-off). Odds ratios are for selenium and are adjusted for sex, residence/school location, father’s education, mother’s education, and age.

^*p < 0.05,

^**p < 0.01.

Discussion

In this community sample, we found that children with lower serum selenium concentrations had higher continuous total anxiety, generalized anxiety and school phobia scores. In addition, children with lower selenium concentrations were more likely to meet clinical cutoffs for five of six tested anxiety disorders, including total anxiety, panic disorder, generalized anxiety disorder, social anxiety disorder, and school phobia. These relationships were significant after controlling for sociodemographic covariates. To our knowledge, this study is the first to examine the relationship between serum selenium concentrations and anxiety in children.

Very few studies have examined anxiety in relation to selenium concentration in humans, particularly in children and adolescents. Findings from this community sample of Chinese children suggest that lower serum selenium concentrations may be associated with anxiety status in younger populations. Benton & Cook (1991) documented a decrease in anxious mood following 5 weeks of 100 mcg selenium supplementation in a British sample of 50 males and females when compared to individuals receiving placebo. Similarly, Finley and Penland (1998) reported decreased anxious mood after 105 days of high-selenium diet supplementation in 30 adult men. However, in a sample of 11 adult men living in a metabolic research unit, Hawkes & Hornbostel (1996) did not find significant changes in mood as a result of 120 days of selenium supplementation, although they reported an association between initial selenium concentrations and mood scores. While these studies have offered preliminary evidence of an association between selenium concentration and anxiety, all reported studies have been limited to largely adult samples in either US or British populations. Moreover, these studies were limited by small sample sizes which are prone to sampling error, and as such, the findings cannot be generalized to the global population. They also examined anxious mood, rather than more stable symptoms of anxiety disorders. Our investigation of the relationship between selenium concentration and childhood anxiety disorders provides novel insights into the role of selenium in a developing brain. Taken together with results from animal studies supporting selenium’s role in neurodegeneration prevention and nerve formation (Valentine et al., 2008; Yant et al., 2003), these findings strengthen the assertion that selenium may be a crucial component of mental health in various populations, thus suggesting a possible role of selenium in childhood anxiety.

Our finding that lower selenium was linked to higher levels of anxiety is largely consistent with previous research examining the link between selenium and depression. A study in 1737 rural Chinese older adults found that individuals who met criteria for depression had lower levels of selenium (Gao et al., 2012). A study of 978 young adults in New Zealand ages 17–25 years found that both especially low and high levels of serum selenium were associated with an increased likelihood of depressive symptoms (Conner et al., 2015). More recently, studies in adults have found that participants who met the recommended dietary allowance for selenium, compared to those who did not, had significantly lower odds of depression (Ghimire et al., 2019; Li et al., 2018). Alternatively, a study of 3735 US adults ages 20–32 years found that higher selenium levels were actually associated with increased depressive symptoms (Colangelo et al., 2014). In a randomized controlled trial, selenium supplementation decreased incidence of postpartum depression in Iranian women when compared to those receiving a placebo (Mokhber et al., 2011). Together, these findings suggest a possible relationship between selenium and depression. However, research is needed that investigates this association in children and adolescents (Campisi et al., 2020). Although depression and anxiety are distinct disorders, they are often highly comorbid and their co-occurrence is common in children (Cummings et al., 2014; Garber & Weersing, 2010; Mobach et al., 2019; Seligman & Ollendick, 1998). One proposed explanation for this comorbidity is that anxiety and depression share common etiological factors (Garber & Weersing, 2010; Seligman & Ollendick, 1998). Consistent with the possibility of a shared micronutrient risk profile for anxiety and depression, a study in older adults found that serum selenium was significantly negatively associated with depression scores and marginally (though not significantly) negatively associated with anxiety scores (Gosney et al., 2008). Together with the current findings, this research suggests that in addition to depression, selenium may also contribute to the development of anxiety disorders.

While previous studies suggested a role for selenium in the modulation of mood and anxiety, biochemical mechanisms remain unclear. Oxidative damage has been previously implicated in the etiology of anxiety (Smaga et al., 2015) and a recent study reported a relationship between oxidized glutathione in blood serum and severity of anxiety (Steenkamp et al., 2017). Selenium may act as a modulator through the antioxidant effects of selenoproteins, such as glutathione peroxidases, thioredoxin reductases and selenoprotein P, which protect cells from oxidative damage stemming from lipoperoxidation (Młyniec et al., 2015). Another potential mechanism of selenium is through its role in regulating thyroid function as a component of iodothyronine deiodinases, which control thyroid hormone turnover (Schomburg & Köhrle, 2008). However, studies linking thyroid function to anxiety and panic disorders have mixed results (Engum et al., 2002; Ittermann et al., 2015; Kikuchi et al., 2005). Lastly, low serum selenium concentration has been found to increase dopamine turnover in a rat model (Castaño et al., 1997) and studies have demonstrated a modulatory role for dopamine receptors D1 & D2 in anxiety-like behavior (Zarrindast & Khakpai, 2015).

We should note that the optimal serum selenium concentration for children that avoids toxicity but maximizes health is unknown (Manzanares & Hardy, 2016). Hays, Macey, Nong, and Aylsward (2014) estimated that a plasma selenium concentration of 80 μg/L corresponds to the Institute of Medicine’s Estimated Average Requirement (EAR) for selenium intake in North American adults, while 230 μg/L corresponds to the Tolerable Upper Intake Level (UL) to avoid toxicity (though this UL level is still considered safe). They suggest that these adult biomonitoring equivalents may potentially be adequate for screening use in children, but urge caution. In the current sample, the mean (SD) serum selenium concentration was 141.10 (39.45) μg/L. Only three participants had serum selenium concentration below 80 μg/L and 4.5% of the sample exceeded 230 μg/L. These observed levels may be attributable to the regional soil profile, as the selenium content of food is affected by the soil in which it is grown. Within China, there is substantial variation in the selenium content of soil (Dinh et al., 2018; Li et al., 2014; Long et al., 2018). To our knowledge, the soil selenium concentration in the current study area of Jintan, which is located in Jiangsu Province, has not been examined. However, high selenium concentrations were found in agricultural soils in the Xuzhou District, which is also located in Jiangsu (Huang et al., 2009). This may account for the relatively high serum selenium levels observed in the current sample. Given the uncertainty regarding optimal serum selenium concentrations in children, as well as the relatively high levels observed in the current sample, we urge caution in recommending dietary selenium supplementation as a potential treatment for symptoms of anxiety in children. More research is first needed that identifies optimal concentrations of selenium for physical health in children.

Several potential limitations of the study should be recognized. First, this study measured serum selenium rather than erythrocyte selenium, which has been suggested to better reflect long-term selenium concentration given that selenium is incorporated in a large variety of proteins of varying biological functions. Second, we did not control for family history of anxiety disorders, general signs of malnutrition, or recent major events that could have an effect on children’s mood. Third, children with missing data differed from children with complete data, which may limit generalizability of findings. Fourth, we did not measure symptoms of selenium toxicity. While higher serum selenium appeared beneficial in terms of anxiety, we do not know whether children experienced any negative health effects. Lastly, since this is a cross-sectional study, the relationship between serum selenium and anxiety cannot be interpreted as causal. Furthermore, given cultural and dietary differences, future studies in the same age group in different racial and cultural populations are warranted. Randomized controlled trials in children are also needed to determine whether improving dietary intake of selenium or selenium supplementation may improve child anxiety. Future research is also needed that determines the optimal level of selenium for childhood anxiety while avoiding potential negative health effects of excess selenium, which can lead to toxicity (Lopez et al., 2010; Morris & Crane, 2013; Yang et al., 1983).

Practice Implications

Children’s emotional wellbeing is an important component of growth and development and an essential aspect of their overall health status. Pediatric nurses play an important role in providing primary care and dietary/nutrition resources during well child visits. The current study could have implications for pediatric nurses in helping reducing anxiety in children. A recent study in youth found that administration of a daily multivitamin reduced symptoms of anxiety (Zhang et al., 2013). These findings are suggestive that nutritional supplementation may help to improve symptoms of anxiety in community children, though we urge caution in recommending selenium supplementation due to the potential for toxicity. When treating childhood anxiety, pediatric nurses may consider providing information for parents on promoting diets rich in dietary sources of selenium, including seafood, grains, meat, poultry, fish, and eggs. It may also be beneficial for pediatric nurses to conduct brief dietary assessments or referral to dietitians when treating children with anxiety.

Conclusions

Our findings from a large community sample of children suggest a potential role of selenium in childhood anxiety. Elucidating the role of selenium in human health may be crucial in developing novel ways of preventing and combatting global health burdens. Future research is needed to examine the short-term and long-term effects of selenium on anxiety using larger cohorts, animal models, and randomized clinical trials in various populations to better understand the role of selenium in mental health.

Highlights.

• Children with lower serum selenium concentrations had higher levels of anxiety.

• Prior research into selenium and anxiety has largely focused on adults.

• Improving nutrition may be a promising strategy to help reduce childhood anxiety.