Food allergy-related bullying and associated peer dynamics among Black and White children in the FORWARD study

Abstract

Background:

The experiences of Black children with food allergy (FA) are not well characterized, particularly with respect to bullying victimization and other psychosocial outcomes.

Objective:

To evaluate bullying experiences of Black and White children with FA, including associations with peer relationships, anxiety, and school policies.

Methods:

Surveys were administered to parents of 252 children with physician-diagnosed FA enrolled in the multisite FORWARD cohort. The surveys assessed demographics, atopic disease, bullying victimization, and school FA management practices and policies. Descriptive statistics of bullying by race were compared by χ² tests. Multiple logistic regression analyses adjusting for race, age, parental education, household income, child sex, and multi-FA compared adjusted probabilities of bullying victimization by school policies.

Results:

Nearly 20% of school-aged children were bullied for FA with no substantial racial differences overall, though for children ages 11 years and up, White children reported higher rates of bullying. However, Black children experienced non-FA-related bullying twice as frequently as White children (38.6% vs 17.7%; P = .002). Most of the caregivers (85.7%) who intervened in their child’s bullying reported that it was helpful. Among parents, 17.3% reported that they were teased or bullied owing to their child’s FA. More than half of the respondents (54.8%) reported that some allergens are banned from their child’s school, most typically peanut. In schools banning peanuts, FA-related bullying was less frequently reported by all students who have food allergy.

Conclusion:

Bullying owing to FA is common, and caregivers, medical professionals, and school administrators can help reduce bullying by screening for bullying and supporting and educating school policies.

Introduction

Food allergy (FA) is a growing public health concern and can place substantial psychosocial burden on children and their families.¹ The impact on children with food allergy and their families is extensive and can impair quality of life.² Considering the amount of time children spend engaged in school-related activities, appropriate management of FA is pivotal to their safety and well-being.³ In addition, emerging literature reveals that FA-related bullying victimization is remarkably common and may lead to distress and anxiety.⁴

Recent studies estimate that 8% of children in the United States have FA.⁵ Beyond the physical manifestations of FA, allergen avoidance behaviors essential to FA management may also lead to social isolation, depression, anxiety, and other forms of psychological distress, which bullying may exacerbate.⁶ In a 2013 study, 31.5% of children with food allergy reported bullying related to FA and 24.7% of parents reported that their child was victimized by such bullying.⁷ A 2010 cross-sectional survey of 353 FA conference attending teens and young adults (up to age 25) found that 35.2% had been bullied, teased, or harassed as a result of FA.⁸ Of those, 85.9% reported that the bullying happened more than once.⁸ Although the reasons for negative peer experiences varied, most of the respondents indicated that they were bullied solely owing to FA; other reported reasons were having to carry medication or being placed in distinct groups to prevent exposure to the allergen.⁸

Studies indicate that teachers may not be fully aware of the psychosocial challenges of children with FA, and only half of the teachers in a recent study knew that students with food allergy were at a greater risk of bullying.10, 11, 9 Although the consequences of FA-related bullying are becoming more understood, research on parental responses to FA-related bullying, impact of bullying on child peer relationships and anxiety, and relationships between school FA management policies and bullying warrants further investigation. In addition, to the knowledge of the authors, this is the first study to evaluate and conduct a statistical analysis on the effect of race in regard to FA-related bullying. The present study evaluates the bullying-related experiences of a socioeconomically and geographically diverse sample of Black and White children with FA, in addition to the association between FA-related bullying and school-level FA policies.

Methods

FORWARD is a prospective multicenter cohort study, initiated to evaluate racial differences in the natural history of FA among Black and White children. We also sought to study other effects of FA among these children and their families, given the underrepresentation of Black families in existing FA-related psychosocial research. Children 12 years of age and younger at study enrollment who had allergist-diagnosed immunoglobulin E-mediated FAs and were identified as either non-Hispanic Black or non-Hispanic White and were recruited from allergy clinics affiliated with the Northwestern University, including Ann & Robert H. Lurie Children’s Hospital of Chicago, Rush University Medical Center, Cincinnati Children’s Hospital Medical Center, and Children’s National Hospital in Washington, DC. Upon enrollment, parents or guardians of children with FA provided written consent and completed electronic surveys at intake and quarterly for 24 months after enrollment. Bullying was assessed by parent-reported questionnaire (eFig 1) administered at 9- and 21-month follow-up. For individuals with multiple completed questionnaires, the most recent available responses were used. All study activities were approved by each center’s Institutional Review Board.

Child peer relationships were assessed by the Parent-Proxy Peer Relationships Scale, published by the National Institutes of Health Patient-Reported Outcomes Measurement Information System (PROMIS). Similarly, child anxiety was assessed by the parent-report National Institutes of Health PROMIS Emotional Distress—Anxiety scale.¹² These questionnaires were administered at 12- and 24-month follow-up. Items for each scale are rated on a 5-point Likert scale (1 = never to 5 = always); higher scores indicate better peer relationships and greater anxiety. For individuals with responses at both waves, mean scores were calculated for all items with multiple measurements; mean composite scores were also calculated for all individuals with multiple completed PROMIS surveys. Both scales revealed excellent internal consistency (PROMIS Peer Relationships scale Cronbach’s α = 0.95; PROMIS Emotional Distress-Anxiety scale Cronbach’s α = 0.90). Confirmatory factor analyses of each scale confirmed that unifactorial models provided excellent fit to the data (comparative fit indices ≥0.950; standardized root mean squared residuals <0.05).

Descriptive counts and proportions were calculated to summarize key study variables. The χ² tests assessed the magnitude of unadjusted associations between participant characteristics (eg, race, age), parental education, household income and bullying, and peer relationships and anxiety, with 2-sided P < .05 indicative of statistically significant relationships. Standardized T scores were calculated for each PROMIS instrument using the HealthMeasures scoring service. Multiple logistic regression models estimated covariate adjusted effects of sociodemographic and allergic disease characteristics on bullying victimization. Furthermore, although the presented data focus exclusively on Black and White children owing to the design of the FORWARD cohort, as of Fall 2020, this cohort also began enrolling Latinx children with FA. This recent expansion of the sampling frame of the FORWARD study to include Latinx patients will facilitate more racially and ethnically inclusive data collection and enhance the external validity of the resulting findings.

Results

Demographic Characteristics

Data were collected from 252 parents or guardians of children with FA who met inclusion criteria and completed a bullying questionnaire. Nearly two-thirds (62.3%) of the children were of male sex, and most were identified as non-Hispanic White (73.4%) (Table 1). More than half (54.4%) of the enrolled children were 4 to 10 years old, and 20.4% were 11 years or older at the most recent study assessment period (Winter and Spring 2020). The most common FAs among all children were peanut (68.8%) and tree nut (58.5%). Multiple FAs were reported for 72.2% of the children.

Table 1:

Demographic Characteristics Among Children Surveyed With Food Allergy (Among Those With Completed Bullying Questionnaire)

Chacteristic	All participants		White participants		Black participants		Test statistic
	N	%	N	%	N	%	χ2; P value
Age
<4 years old	63	25.2	53	29	10	14.9	χ2 (2df) = 10.9; P =.004
4–11 years old	136	54.4	101	55.2	35	52.2
>11 years old	51	20.4	29	15.9	22	32.8
Sex
Male	157	62.3	114	61.6	43	64.2	χ2 (1df) = 0.14; P = .71
Female	95	37.7	71	38.4	24	35.8
Race
White	185	73.4	185	100	0	0
Black or African American	67	26.6	0	0	67	100
Type of food allergy
Peanut	174	68.8	127	68.7	47	70.2	χ2 (1df) = 0.05; P = .82
Milk	67	26.5	54	29.2	13	19.4	χ2 (1df) = 2.4; P = .12
Egg	112	44.3	82	44.3	30	44.8	χ2 (1df) = 0.004; P = .95
Fin fish	34	13.4	13	7	21	31.3	χ2 (1df) = 24.9; P < .001
Shell fish	42	16.6	11	6	31	46.3	χ2 (1df) = 57.6; P < .001
Wheat	19	7.5	7	3.8	12	17.9	χ2 (1df) = 14.1; P < .001
Tree nut	148	58.5	102	55.1	46	68.7	χ2 (1df) = 3.7; P = .05
Sesame	48	19	40	21.6	8	11.9	χ2 (1df) = 3.0; P = .08
Soy	18	7.1	10	5.4	8	11.9	χ2 (1df) = 3.2; P = .08
Other	39	15.4	30	16.2	9	13.4	χ2 (1df) = 0.29; P = .59
Highest attained parental education
Some high school, no diploma	5	2.1	3	1.7	2	3.3	χ2 (7df) = 46.5; P < .001
High school graduate, diploma, or equivalent	13	5.5	5	2.9	7	11.5
Some college, no degree	30	12.7	11	6.3	19	31.2
Associate’s degree	10	4.2	5	2.9	5	8.2
Bachelor’s degree	69	29.2	53	30.5	16	26.2
Master’s degree	76	32.2	65	37.4	11	18
Professional degree	14	5.9	14	8.1	0	0
Doctoral degree	19	8	18	10.3	1	1.6
Annual household income
<$50,000	54	22.9	16	9.2	37	60.7	χ2 (6df) = 75.4; P < .001
$50,000-$99,999	22	9.3	15	8.6	7	11.5
$100,000-$149,999	40	17	35	20.1	5	8.2
$150,000-$199,000	23	9.8	20	11.5	3	4.9
$200,000-$299,000	34	14.4	31	17.8	3	4.9
$300,000 or more	45	19.1	43	24.7	2	3.3
Decline to answer	18	7.6	14	8	4	6.6
Number of food allergy
Single	70	27.8	56	30.3	14	20.9	χ2 (1df) = 2.2; P = .14
Multiple	182	72.2	129	69.7	53	79.1
Comorbiditiesa
Physician-diagnosed allergic rhinitis	117	49.8	77	44.5	40	64.5	χ2 (1df) = 7.3; P = .007
Current asthma	94	39.8	58	33.3	36	58.1	χ2 (1df) = 13.3; P = .001
Outgrown asthma	9	3.8	9	5.2	0	0
Physician-diagnosed lifetime eczema	195	83	137	79.2	58	93.6	χ2 (1df) = 6.7; P = .1

^aComorbidity data only available from subset of N = 235 patients.

Significant differences in parental education and annual household income were reported across the two racial groups within our sample (P < .001). More than two-thirds (67.9%) of all White respondents reported having attained either a Bachelor’s or Master’s degree in comparison to 44.2% of Black respondents. In addition, almost three-quarters (74.1%) of the White respondents indicated an annual household income of ≥$100,000 vs most of the Black respondents who reported annual household income of <$50,000 (60.7%).

Bullying

As found in Table 2, 18.7% of all parents of children 4 to 15 years old reported that their child with FA had been bullied owing to their FA and 24.1% reported that the child had been bullied for reasons other than FA. FA-related bullying was most typically reported among older children, affecting 33.3% of children at least 11 years old vs 13.2% of children 4 to 10 years old. No significant racial differences in FA-related bullying prevalence were identified overall (20.0% of White participants vs 15.6% of Black participants; χ² (1_df) = 0.46; P = .50); however, among children ages 11 years and up, the rates were significantly higher among White participants (44.8% vs 18.2% of Black participants; χ² (1_df) = 4.0; P = .046). In contrast, bullying unrelated to FA was higher among Black children (38.6%), lower in White children (17.7%), and significantly different between the 2 groups for all ages; χ² (1_df) = 9.5; P = .002.

Table 2:

Food Allergy-Related Bullying

Variable	Yes overall, N (%)	White participants, N (%)	Black participants, N (%)	χ2; P value
Ever bullied owing to FA (all ages)	35 (18.7)	26 (20.0)	9 (15.6)	χ2 (1df) = 0.46; P = .50
4–11 year olds only	18 (13.2)	13 (12.9)	5 (14.3)	χ2 (1df) = .04; P = .83
>11 year olds only	17 (33.3)	13 (44.8)	4 (18.2)	χ2 (1df) = 4.0; P = .046
Ever bullied for reasons other than FA (all ages)	45 (24.1)	23 (17.7)	22 (38.6)	χ2 (1df) = 9.5; P = .002
4–11 year olds only	29 (21.3)	15 (14.9)	14 (40.0)	χ2 (1df) = 9.8; P = .002
>11 year olds only	16 (31.4)	8 (27.6)	8 (36.4)	χ2 (1df) = 0.45; P = .50
Child disclosed bullying to parent or guardian (all ages)	27 (14.4)	20 (15.4)	7 (12.3)	χ2 (1df) = 0.3; P = .58
4–11 year olds only	13 (9.6)	10 (9.9)	3 (8.6)	χ2 (1df) = 0.05; P = .82
>11 year olds only	14 (27.5)	10 (34.5)	4 (18.2)	χ2 (1df) = 1.7; P = .20
Parent or guardian did something to stop bullying
Spoke with child	20 (12.6)	11 (10.4)	9 (17.0)	χ2 (1df) = 1.4; P = .24
Spoke with offender or offender's parent	11 (7.0)	2 (1.9)	9 (17.3)	χ2 (1df) = 12.8; P < .001
Spoke with a teacher	27 (16.8)	15 (13.8)	12 (23.1)	χ2 (1df) = 2.2; P = .14
Spoke with a principal or administration	25 (15.5)	14 (12.8)	11 (21.2)	χ2 (1df) = 1.9; P = .17
Other
If a parent or guardian did something to stop bullying, was it helpful?	36 (85.7)	22 (81.5)	14 (93.3)	χ2 (1df) = 1.1; P = .29
Parent-reported that they (the parent) were made fun of or teased in a hurtful way because of their concerns about their child's food allergy (all ages)	36 (19.3)	31 (23.9)	5 (8.8)	χ2 (1df) = 5.8; P = .02
4–11 year olds only	27 (19.9)	24 (23.8)	3 (8.6)	χ2 (1df) = 3.8; P = .052
>11 year olds only	9 (17.7)	7 (24.1)	2 (9.1)	χ2 (1df) = 1.9; P = .16
Who was the source of this parental bullying?
Spouse	11 (4.2)	9 (4.7)	2 (2.9)	χ2 (1df) = 0.4; P = .54
Friend	17 (6.7)	15 (8.0)	2 (3.0)	χ2 (1df) = 1.9; P = .16
Parent of another child	24 (9.6)	23 (12.4)	1 (1.5)	χ2 (1df) = 6.7; P = .01
Teacher or principal at the child's school	10 (4.0)	9 (4.9)	1 (1.5)	χ2 (1df) = 1.5; P = .22
My parents (the child’s grandparents)	9 (3.6)	8 (4.4)	1 (1.5)	χ2 (1df) = 1.2; P = .28
My spouse’s parents	15 (6.0)	15 (8.1)	0 (0)	χ2 (1df) = 5.7; P = .02
Another family member	24 (9.5)	20 (10.8)	4 (5.9)	χ2 (1df) = 1.4; P = .24
Other	7 (2.8)	6 (3.3)	1 (1.5)	χ2 (1df) = 0.54; P = .46

Abbreviation: FA, food allergy.

Overall, 14.4% of the respondents reported that the child had disclosed being bullied to their parent or guardian, with similar rates reported for Black (15.4%) and White (12.3%) participants; χ² (1_df) = 0.3; P = .58. Black parents were more likely than White parents to report taking actions to address their child’s bullying by all assessed modalities (Table 2). The greatest racial differences were found with respect to speaking with the offender or offender’s parent, χ² (1_df) = 12.8; P < .001). All parents reported that it was helpful if they did something to stop the bullying, but benefit was more likely to be reported by Black parents (93.3% of Black parents vs 81.5% of White parents); χ² (1_df) = 1.1; P = .29.

In addition, questions on whether parents themselves had experienced bullying owing to their child’s FA were assessed. Overall, 17.3% reported that they had been made fun of or teased, including 20.7% of the White parents and 6.9% of the Black parents, χ² (1_df) = 7.2; P = .004. Within our sample, the most common sources of parental FA-related bullying were the parent of another child (9.6%) or another family member (9.5%).

Child Peer Relationships and Anxiety

For both PROMIS instruments administered in this study, a T score of 50 corresponds to the mean for the US general pediatric population with a population-level SD of 10. The mean T score observed within the entire sample on the peer relationships scale was 50.0 (SD = 9.1); which is good according to the PROMIS score cut points (T score range 40–60). The mean T score value for this sample at baseline on the anxiety measure was 51.4 (SD = 9.3), which fell within normal limits (T score <55).

Parental perceptions of child peer relationships and anxiety were analyzed for children ages 4 to 15 (Table 3 and Fig. 1). The mean composite peer relationship scores for all children with FA-related bullying and for children who were bullied for other reasons revealed no significant differences when compared with children without such bullying (P_{FA-related bullying} = 0.73, P_{Non FA-related bullying} = 0.19). White respondents scored significantly higher on the item, “My child was able to count on his or her friends,” than Black respondents (M_White = 4.33, SD = 0.67, M_Black = 3.98, SD = 0.96; P = .03). Furthermore, there were no significant differences in composite peer relationship scores when comparing White and Black respondents with FA-related bullying and White and Black respondents with non–FA-related bullying (P > 0.05). Moreover, mean scores for both White and Black respondents indicated an overall perception of good child peer relationships.

Table 3.:

PROMIS Peer Relationships and Anxiety Measures

Variable	No report of FA-related bullying Mean (SD)	Report of FA-related bullying Mean (SD)	T test P value	No report of other bullying Mean (SD)	Report of other bullying Mean (SD)	T test P value	Overall White respondents Mean (SD)	Overall Black respondents Mean (SD)	T test P Value
Peer relationships
My child felt accepted by other kids his or her	4.45 (0.66)	4.26 (0.54)	0.21	4.48 (0.61)	4.15 (0.68)	0.01	4.43 (0.60)	4.37 (0.76)	0.65
age
My child was able to count on his or her	4.27 (0.71)	4.17 (0.72)	0.56	4.28 (0.71)	4.17 (0.72)	0.48	4.33 (0.67)	3.98 (0.96)	0.03
friends
My child was good at making friends	4.25 (0.92)	4.26 (0.75)	0.95	4.29 (0.91)	4.09 (0.83)	0.3	4.29 (0.85)	4.07 (1.05)	0.24
My child and his or her friends helped each	4.26 (0.76)	4.17 (0.65)	0.61	4.28 (0.77)	4.13 (0.64)	0.36	4.31 (0.69)	4.04 (0.92)	0.08
other out
Other kids wanted to be my child's friend	4.19 (0.85)	4.17 (0.83)	0.92	4.20 (0.88)	4.15 (0.72)	0.77	4.21 (0.84)	4.12 (0.88)	0.61
Other kids wanted to be with my child	4.22 (0.83)	4.24 (0.77)	0.91	4.27 (0.82)	4.06 (0.79)	0.23	4.28 (0.79)	4.01 (0.93)	0.11
Other kids wanted to talk my child	4.27 (0.81)	4.26 (0.75)	0.95	4.32 (0.80)	4.07 (0.78)	0.15	4.29 (0.78)	4.21 (0.91)	0.63
Composite peer relationship score	4.27 (0.68)	4.22 (0.58)	0.73	4.31 (0.67)	4.12 (0.62)	0.19	4.30 (0.63)	4.11 (0.81)	0.17
Anxiety
My child felt nervous	2.31 (0.86)	2.41 (0.94)	0.59	2.27 (0.80)	2.54 (1.06)	0.15	2.45 (0.84)	2.03 (1.04)	0.02
My child felt scared	2.10 (0.84)	2.30 (1.02)	0.3	2.11 (0.82)	2.22 (1.09)	0.56	2.17 (0.84)	2 (1.04)	0.35
My child felt worried	2.22 (0.85)	2.5 (1.01)	0.17	2.18 (0.80)	2.57 (1.10)	0.04	2.37 (0.80)	1.97 (1.07)	0.02
My child felt like something awful might	1.62 (0.70)	2.09 (1.07)	0.01	1.63 (0.69)	1.96 (1.08)	0.05	1.74 (0.79)	1.66 (0.94)	0.65
happen
My child worried when he or she was at home	1.61 (0.68)	1.96 (0.82)	0.04	1.61 (0.66)	1.89 (0.86)	0.08	1.75 (0.75)	1.5 (0.74)	0.1
My child got scared really easy	1.74 (0.80)	2 (1.04)	0.18	1.72 (0.79)	2.02 (1.02)	0.11	1.82 (0.88)	1.72 (0.90)	0.58
My child worried about what could happen to	1.70 (0.72)	2.36 (1.04)	<0.001	1.75 (0.72)	2.06 (1.11)	0.09	1.85 (0.80)	1.70 (0.98)	0.36
him or her
My child worried when he or she went to bed	1.68 (0.82)	2.04 (1.19)	0.08	1.70 (0.81)	1.91 (1.18)	0.29	1.85 (0.95)	1.49 (0.85)	0.05
at night
Composite anxiety score	1.87 (0.61)	2.25 (0.88)	0.02	1.87 (0.60)	2.18 (0.88)	0.04	2.00 (0.68)	1.77 (0.79)	0.12

Abbreviation: FA, food allergy.

Figure 1.

Child peer relationships and anxiety PROMIS outcomes by FA-related bullying and noneFA-related bullying victimization. Bar graphs visualizing item-specific and composite scores on parent-proxy reported peer relationship and anxiety PROMIS instruments stratified by White and Black race. FA, food allergy

Significantly higher child anxiety scores were observed from parents who reported that their child experienced both FA-related bullying and bullying for other reasons when compared with those without report of either type of bullying (M_{Yes, FA-related bullying} = 2.25, SD = 0.88, M_{No FA-related bullying} = 1.87, SD = 0.61; P = .02; M_{Yes, other bullying} = 2.18, SD = .88, M_{No other bullying} = 1.87, SD = .60; P = .04). Black respondents scored significantly higher on the “My child felt nervous” and “My child felt worried” items than White respondents (P = .02 for both). There were also no significant differences in composite anxiety scores when comparing White and Black respondents with FA-related bullying and White and Black respondents with non–FA-related bullying (P > .05). When analyzing the effects of bullying, the observed peer and anxiety relationships remained consistent after adjusting for age, sex, parental education, income, and presence of non–FA-related bullying.

School-Level FA Management

Among those participants with complete data for school management, bullying, and PROMIS surveys (n = 82), 30.5% of the children were in daycare or preschool, 57.3% were in elementary or middle school, and 1.2% were high school students (Table 4). Of this sample, half of the White students (50.0%) and more than three-fourths of the Black students (76.2%) attended public school. Moreover, 48.1% of the White students and 14.3% of the Black students attended private school.

Table 4.:

School Management Policies (Among Those With Complete School Management, Bullying, and PROMIS Data)

Variable	All participants	White participants	Black participants	χ2; P value
	N (%)	N (%)	N (%)
Daycare and preschool	25 (30.5)	19 (31.7)	6 (27.3)	χ2 (3df) = 4.2; P = .24
Elementary and middle school	47 (57.3)	33 (55.0)	14 (63.6)
High school	1 (1.2)	0 (0)	1 (4.6)
N/A	9 (11.0)	8 (13.3)	1 (4.6)
Which type of school does your child attend?
Public	42 (57.5)	26 (50.0)	16 (76.2)	χ2 (3df) = 9.2; P = .027
Private	28 (38.4)	25 (48.1)	3 (14.3)
Home school	2 (2.7)	1 (1.9)	1 (4.8)
Other	1 (1.4)	0 (0)	1 (4.8)
Did you discuss food allergy management with your child's school?
Yes	71 (97.3)	51 (98.1)	20 (95.2)	χ2 (1df) = 0.45; P = .50
Yes, spoke with principal	31 (43.7)	24 (47.1)	7 (35.0)	χ2 (1df) = 0.85; P = .36
Yes, spoke with teacher	60 (84.5)	44 (86.3)	16 (80.0)	χ2 (1df) = 0.43; P = .51
Yes, spoke with school nurse	46 (64.8)	29 (56.9)	17 (85.0)	χ2 (1df) = 5.0; P = .026
Yes, spoke with caseworker	14 (19.7)	7 (13.7)	7 (35.0)	χ2 (1df) = 4.1; P = .04
Do you have a written food allergy management documents on file at your child’s school?
Yes	70 (98.6)	50 (98.0)	20 (100)	χ2 (2df) = 0.40; P ¼ = .53
No	0	0	0
Unsure	1 (1.4)	1 (2.0)	0 (0)
(If yes to above), which of the following do you have on file at your child’s school?
Food allergy action plan or anaphylaxis emergency care plan	68 (97.1)	49 (98.0)	19 (95.0)	χ2 (1df) = 0.46; P = .50
Individualized education program	12 (17.1)	5 (10.0)	7 (35.0)	χ2 (1df) = 6.3; P = .012
504 plan	15 (21.4)	7 (14.0)	8 (40.0)	χ2 (1df) = 5.7; P = .02
Other plan	3 (4.3)	3 (6.0)	0 (0)	χ2 (1df) 1.3; P = .26
Is stock epinephrine available at your school in case of emergency?
Yes	19 (26.0)	11 (21.2)	8 (38.1)	χ2 (2df) = 2.3; P = .32
No	22 (30.1)	17 (32.7)	5 (23.8)
Unsure	32 (43.8)	24 (46.2)	8 (38.1)
Does your child have their own epinephrine at school?
Yes	67 (91.8)	49 (94.2)	18 (85.7)	χ2 (2df) = 2.9; P = .23
No	5 (6.9)	2 (3.9)	3 (14.3)
Unsure	1 (1.4)	1 (1.9)	0 (0)
(If yes to above) where is your child’s epinephrine kept?
My child self-carries	11 (16.4)	7 (14.3)	4 (22.2)	χ2 (4df) = 5.8; P = .22
Teacher’s desk	14 (20.9)	12 (24.5)	2 (11.1)
Nurse’s office	30 (44.8)	19 (38.8)	11 (61.1)
Principal’s office	5 (7.5)	4 (8.2)	1 (5.6)
Other	7 (10.5)	7 (14.3)	0 (0)
Are some foods not allowed in your child’s school owing to food allergies?
Yes	40 (54.8)	27 (51.9)	13 (61.9)	χ2 (2df) = 13.9; P = .001
No	25 (34.3)	23 (44.2)	2 (9.5)
Unsure	8 (11.0)	2 (3.9)	6 (28.6)
(If yes to above) which foods are not allowed in your child’s school because of food allergies?
Peanuts	39 (97.5)	27 (100)	12 (92.3)	χ2 (1df) = 2.1; P = .14
Tree nuts	30 (75.0)	20 (74.1)	10 (76.9)	χ2 (1df) = 0.04; P = .85
Other	2 (5.0)	1 (3.7)	1 (7.7)	χ2 (1df) = 0.29; P = .59
Is there a classroom policy regarding food in the classroom for snacks, parties, et cetera?
Yes	101 (80.2)	68 (79.1)	33 (82.5)	χ2 (2df) = 1.3; P = .52
No	15 (11.9)	12 (13.9)	3 (7.5)
Unsure	10 (7.9)	6 (7.0)	4 (10.0)
(If yes to above) which of these applies to your child’s classroom?
Only allergen-free food is allowed	49 (48.5)	30 (44.1)	19 (57.6)	χ2 (1df) = 1.6; P = .20
Only food with a clear ingredient label is allowed	35 (34.7)	26 (38.2)	24 (72.7)	χ2 (1df) = 1.2; P = .28
No food is allowed	21 (20.8)	16 (23.5)	5 (15.2)	χ2 (1df) = 0.9; P = .33
I am not sure	7 (6.9)	3 (4.4)	4 (12.1)	χ2 (1df) = 2.0; P = .15
Other	10 (9.9)	9 (13.2)	1 (3.0)	χ2 (1df) = 2.6; P = .11
Are there allergen-free areas for students with food allergy to eat lunch?
Yes	29 (39.7)	25 (48.1)	4 (19.1)	χ2 (2df) = 10.3; P = .006
No	29 (39.7)	21 (40.4)	8 (38.1)
Unsure	15 (20.6)	6 (11.5)	9 (42.9)
(If yes to above) what allergen-free lunch areas are available at your child’s school?
Peanut-free table	26 (89.7)	22 (88.0)	4 (100.0)	χ2 (1df) = 0.54; P = .46
Tree nut-free table	12 (41.4)	11 (44.0)	1 (25.0)	χ2 (1df) = 0.51; P = .47
Dairy-free table	24 (82.8)	20 (80.0)	4 (100.0)	χ2 (1df) = 0.97; P = .33
Other allergen-free areas	4 (13.8)	3 (12.0)	1 (25.0)	χ2 (1df) = 0.49; P = .48

Abbreviation: N/A, not available.

Roughly all respondents reported that they had discussed FA management and had written FA management documents on file at their child’s school (97.3% discussed and 98.6 written documents). Of note, 92% of the respondents reported that their child had their own epinephrine autoinjector at school, yet only 26.0% reported that stock epinephrine was available in case of an emergency, and 43.8% were unsure. In addition, more than half (54.8%) of the respondents reported that some foods were not allowed in their child’s school owing to FA. Of those respondents, 97.5% reported that peanuts were not allowed and 75.0% reported that tree nuts were not allowed. When asked if allergen-free lunch areas were available for students with FA, 48.1% of the White respondents and 19.1% of the Black respondents reported yes, and 40.4% of the White respondents and 38.1% of the Black respondents reported no. Of those respondents who said yes, peanut-free tables and dairy-free tables were the most typically reported allergen-free lunch areas available in schools (89.7% and 82.8%, respectively).

Associations

As indicated in Table 5, we evaluated adjusted associations for any FA-related bullying and any non–FA-related bullying. Children greater than 11 years of age were significantly more likely to report any FA-related bullying (OR: 4.57; CI: 1.86–11.27) and report telling their parents that they were bullied (OR: 4.90; CI: 1.90–12.61) in comparison to children 4 to 11 years old. Children with egg allergy were also significantly more likely to experience any FA-related bullying (OR: 3.22; CI: 1.25–8.32), any non–FA-related bullying (OR: 2.58; CI: 1.13–5.86), and inform their parents that they were bullied (OR: 3.25; CI: 1.17–9.06), in comparison to children without egg allergy. Coefficient plots visualizing these findings are provided in Fig. 2.

Table 5:

Multiple Logistic Regression Analyses Examining Demographic and Clinical Predictors of Bullying Victimization

Patient characteristics associated with any FA-related bullying	Odds ratio	95% CI
Black (vs White)	0.49	0.15	1.58
>11 y (vs 4–11 y)	4.57	1.86	11.27
Parental education	0.73	0.54	1.01
Annual household income	1.13	0.88	1.44
Female (vs male)	1.81	0.77	4.26
Multiple FA (vs single FA)	0.58	0.15	2.19
Peanut allergy	1.48	0.51	4.33
Tree nut allergy	0.57	0.2	1.64
Egg allergy	3.22	1.25	8.32
_cons	0.35	0.06	2.01
Patient characteristics associated with any non-FA-related bullying	Odds ratio	95% CI
Black (vs White)	1.88	0.73	4.81
>11 y (vs 4–11 y)	1.74	0.76	3.95
Parental education	0.74	0.56	0.98
Annual household income	0.99	0.79	1.23
Female (vs male)	1.66	0.77	3.56
Multiple FA (vs single FA)	0.65	0.19	2.2
Peanut allergy	1.25	0.47	3.34
Tree nut allergy	1.29	0.5	3.35
Egg allergy	2.58	1.13	5.86
_cons	0.53	0.11	2.55

Abbreviations: CI, confidence interval; FA, food related.

Figure 2.

Covariate-adjusted predictors of FA-related bullying and non-FA-related bullying victimization. Plots of exponentiated logistic regression coefficients (eg, odds ratios) corresponding to covariate-adjusted predictors of bullying victimization with corresponding 95% confidence intervals. FA, food allergy. NOTE: Parental Education and Annual Household Income were modeled as continuous variables; the remaining variables were modeled as dichotomous.

Discussion

Bullying is widely defined as repeated, unwanted, aggressive behavior that involves a real or perceived power imbalance.¹³ The power imbalance may be physical or psychological, and the aggressive behaviors may be verbal, social, physical, or psychological. Studies have reported bullying rates of 17% to 35% among school-aged children.^14–17 This study extends previous work by evaluating the differences between races associated with FA-related bullying. Our results indicated that 1 in 5 school-aged children in our sample experienced bullying solely owing to their FA. No significant racial differences regarding FA-related bullying were identified overall; though for children ages 11 years and up, White children reported higher rates of bullying. However, significant differences were observed concerning bullying for reasons other than FA, as Black participants experienced bullying at a rate of more than twice that of White participants. Parents who knew that their child was being bullied and subsequently did something to stop the bullying reported that it was helpful in most cases. Black parents were more likely than White parents to address their child’s bullying. When considering peer dynamics among children with FA, no significant racial differences among mean composite peer relationship scores or mean composite anxiety scores were found.

To effectively manage FA, lifestyle changes often include strict food monitoring and preparation, carrying medical equipment (eg, epinephrine autoinjector), and being placed in special groups, or excluded from food-related activities, which may predispose children to bullying.^7,8 Children with FA have experienced a variety of threats in school including verbal threats of allergen exposure, being touched by an allergen, having an allergen thrown or waved at them, as well as intentional cross-contact of their food with an allergen.^7,8 These events could lead to life-threatening allergic reactions and additional emotional distress.

The lived experiences of school-aged children with FA are becoming better understood, and many studies have identified associations between FA and peer dynamics. Among participants who had experienced FA-related bullying, 65.7% expressed sadness or depression and 64.2% indicated feelings of embarrassment or humiliation.⁸ Among the present cohort, parent-proxy report of child peer relationships and anxiety were slightly better among children with no FA-related bullying; however, mean composite scores for child peer relationships and anxiety were generally favorable. It has been previously found that parental knowledge of bullying is linked with less distress and greater quality of life in children with FA.⁷ This was also true in our cohort; 85.7% of parents with a child aged 4 to 15 years, reported that it was helpful if they did something to stop their child from being bullied.

Owing to the amount of time children spend in school-related activities, there is a critical need for managing both the medical and social aspects of FA in schools.¹⁸ A balance must be reached when considering how to safely manage FA in schools without physically isolating children with FA.¹⁹ Children with FA who are separated from their peers may have increased anxiety and lessened social skills, making them more susceptible to bullying and its psychological effects.^17,20 Parents of children with FA also experience vulnerability and anxiety during their child’s transition to school. Their concerns include their child’s safety and other parents’ negative attitudes toward FA.²¹ Currently, there are no well-controlled studies assessing the rates of allergen exposures or reactions between schools with and without school-wide allergen restriction policies, and no policy has been associated with a complete absence of allergic reactions.^9,22

The psychosocial impact of daily FA management and FA-related bullying is far-reaching. Schools have the responsibility to ensure that children with FA feel safe, thereby empowering them to focus on learning rather than risk management. Students should also feel included in social activities, a fundamental experience that supports socioemotional development. Multiple studies have indicated that educational interventions to ensure properly trained school personnel are crucial in FA management.^10,23–26 Schools could consider implementing policies to help children feel safe from allergen exposure and confident that allergic reactions will be treated appropriately. For example, school nurses could instruct all teachers on best practices for FA safety rather than just teachers who have children with FA in their classroom. Policies regarding allergen exclusion from classrooms, parties, and other events could be implemented across the entire school rather than isolated to classrooms with children with FA. With robust antibullying policies, these allergy-related policies would lessen the risk of children with FA being blamed for such policies, while also promoting safety and social inclusion.

Medical professionals who work with patients with FA can also play a role in reducing the psychosocial impact of FA in general, and bullying specifically, by routinely educating families about school FA policies that promote safety and social inclusion, the benefits of having a school FA management plan on file, and the possibility that FA-related bullying could occur. Furthermore, medical professionals can assess the school and peer experiences of their FA patients during appointments and encourage parents to speak with school officials when needed. Clinicians should use open-ended queries about patient experiences that do not require children to specifically label their experiences as bullying. In addition to clinician intervention, parental mediation can also help decrease FA-related bullying. Among our sample, we found that when parents interfered, it was helpful in more than three-fourths of all cases indicating parents can play a greater proactive role in bullying.

Strengths of the present study include its clinically recruited sample, which is likely more generalizable to the broader FA population than previous work that has largely relied upon convenience samples. The FORWARD study’s prospective cohort design also allows repeated assessment of key outcomes of interest, resulting in reduced measurement error. However, the fact that not all constructs evaluated in this study were assessed concomitantly is also a limitation, given the relatively high rates of missing data within the cohort for any given survey assessment. It is important to note, however, that missing data are addressed by ongoing participant outreach to encourage survey completion. As the cohort ages, this will permit further examination of change in FA-related bullying and other peer dynamics over time. Perspectives of high school-aged youth with FA are lacking in the present study, owing to its inclusion criteria. Other limitations of the study include reliance on parent-proxy report of child bullying, which may not be fully indicative of the child’s experiences. Similarly, the description of school-level policies was based upon parent-report and therefore may be inaccurate. Nevertheless, future work should consider building upon the present work by incorporating multimodal assessment of bullying and school-level policies for more accurate and reliable perspectives.

Increasing school policy awareness for parents and greater educational FA awareness for students is critical. Although we found that most parents reported previously discussing FA management with their child’s school, more than 40% were unsure if stock epinephrine was available in case of emergency and 1 in 5 were unsure if there were allergen-free areas for students with FA to eat lunch. Thus, schools need to make sure FA management discussions are intentional and that parents understand school FA-related policies. Moreover, schools, parents, and children with FA should work together to educate the school community about FA. If students (and caregivers) understand the potential severity of FA and its burden on affected students, and ways to support students with FA, this may help reduce FA-related bullying. A previous study indicated that using peer-developed educational videos improved FA knowledge among elementary to high school-aged children.²⁷ After watching the videos, most students reported that they learned something new about FA, could identify common food allergens, and increased their knowledge regarding FA symptoms.²⁷ Awareness of FA leads to accommodation and acceptance, which may reduce the social exclusion that would otherwise be magnified in youth during this psychologically and sociologically formative time in their lives.^28,29

In sum, this study found that both White and Black children experience FA-related bullying at comparable rates, with older children experiencing more FA-related bullying than younger children. Furthermore, for older children, White children reported higher rates of bullying. In addition, Black parents were more likely than White parents to address their child’s bullying. Bullying experiences were relatively unrelated to parent reports of child anxiety or peer experiences, although these were poorer among children who experienced FA-related bullying. Further research is needed that directly assesses children’s perspectives on their experiences of school FA policies and bullying. School policies have the potential to positively impact the experiences of their students with FA if they enact school-wide policies that are socially inclusive.