2.1 Inclusion criteria

Inclusion criteria was set based on the PICO question format (population, intervention, comparison or control and outcome). The population under investigation for this systematic review was children (aged 0–18 years) who have an allergy to one or more foods, or family members of children (aged 0–18 years) who have an allergy to one or more foods, or both family members and children. The study had to describe a psychological intervention with reference to psychological theory or describe the underpinning psychological mechanism(s) on which the intervention was based. Interventions were included regardless of geographical location, sample size and publication date. Studies with or without comparison groups were included. Studies needed to report changes in measures of psychological wellbeing in children with one or more food allergies or family members of children with one or more food allergies. Psychological wellbeing was defined as being inclusive of measures of quality of life, depression, anxiety, coping, stress, and worry.

2.2 Exclusion criteria

Abstracts, reviews, discussion papers, non-research letters, editorials, protocols, and reviews of clinical services were excluded.

2.3 Search strategy

The search strategy was formulated using key terms related to the population (children and adolescents with food allergy and family members of children and adolescents with food allergy) and the intervention (psychological services, psychological support or interventions for children or adolescents with food allergy as well as psychological services, psychological support or interventions for family members of children or adolescents with food allergy). A systematic search of the literature was conducted on the following databases: Web of Science, Scopus, MEDLINE, Cochrane Library, WHO trials registry for randomized controlled trials, PsycINFO and EMBASE. Relevant gray literature was also searched using Google Scholar and ProQuest Dissertations and Theses database. Additional references were checked through searching the references cited by the identified studies and systematic reviews, through papers identified in Research Gate and through discussion with experts in the field. Databases were searched from inception to 30th October, 2023. Search terms can be found in Data S1.

2.4 Study selection

After duplicates were removed using EndNote (n = 235), the title and abstracts were manually screened for all papers identified during the search of the literature (n = 1122) by two authors. Full-text copies of identified articles, which met the inclusion criteria, were obtained (n = 40) and reviewed by three members of the research team to assess their inclusion eligibility. Articles were removed if they did not fit the inclusion criteria. Two papers were discussed by the team and were removed as they contained reference to a fictional case study rather than primary data.

2.5 Quality assessment strategy

Quality assessments were independently carried out on each study by two reviewers using the Mixed Methods Appraisal Tool (MMAT).⁹ The MMAT is designed for the appraisal of the methodological quality of quantitative, qualitative or mixed-methods study designs. Each study is rated as to whether it includes quality indicators relevant to the study design. A score is not calculated, but a detailed representation of the ratings is reviewed to assess the quality of the study. No studies were excluded in this review based on poor quality and any discrepancies in ratings were resolved by discussion with a third reviewer.

2.6 Data extraction, analysis, and synthesis

Data were extracted by the research team (RK, CJ, and LH) onto a data extraction form. This included population demographics, study methodology including any comparison groups, mode of delivery and length of the intervention or support offered, the type of psychological service provided, changes in psychological outcomes for children living with food allergy, or their family members. A narrative synthesis of the data was undertaken to summarize the demographic characteristics of children or family members who had been offered support, the support needs identified for the children or family member, the type of psychological support offered, the success of the identified support or interventions regarding improvements in reported measures of psychological well-being. As the intervention types, populations and outcome measures varied across studies, a meta-analysis was not possible.

3.1 Description of studies

A total of 11 papers were included in the final dataset (Figure 1). A summary of study characteristics and findings can be found in Table 1. The majority of studies had small sample sizes; for children, the sample size range was 10–61 with a total of 156 across all studies and for parents the range was 10 to 205 with a total of 682 across all studies. Studies were conducted in the United States (n = 7),^10-16 or the UK (n = 4).^17-20 Intervention types were based on CBT (n = 7)^{10, 11, 16-20} psycho-education (n = 1),¹⁴ peer mentoring (n = 1)¹⁵; self-regulation theory (n = 1)¹²; and coping (n = 1).¹³ Only two interventions were for children only,^{10, 13} three were for children and parents^{11, 14, 16} and six for parents only.^{12, 15, 17-20} No interventions were found for other family members. Only two studies used a clinically anxious sample.^{10, 17} Four studies used a randomized controlled trial (RCT) design,^{11, 12, 18, 19} and one was a non-randomized case control study,¹⁷ with control groups receiving usual care and/or on a waiting list before receiving the intervention. All other studies used a within participant design. Interventions focused on the management of anxiety, stress and/or depression, improvement of coping or problem-solving skills and self-efficacy. One also assisted parents in transitioning FA management responsibility to their child.¹⁶ Outcome measures included QoL, self-efficacy, stress, intolerance of uncertainty, anxiety, worry, depression, and perceived social support. Health outcomes such as adherence or food allergy reactions were not generally measured; however, one intervention assessed food allergy knowledge,¹⁶ one measured medical information such as number of anaphylactic reactions pre and post the intervention¹⁴ and the intervention for oral immunotherapy¹¹ measured dose adherence and IgE/IgG4 levels.

3.2 Quality Assessment

Papers were rated for quality using the MMAT.⁹ Whilst all studies provided clear research questions and study aims, there was variable quality in terms of reporting across all studies and all designs. For the four studies utilizing RCTs, randomization was conducted appropriately in only two of the these, with high levels of missing data reported across most studies and largely inadequate reporting of whether participants adhered to the intervention. For the experimental studies without randomization or control groups, all suffered from lack of diversity and representativeness, with confounders often unaccounted for in analysis, though intervention fidelity (the extent to which an intervention is delivered as planned in order to draw valid conclusions concerning the intervention's effectiveness) was consistently better (Table 2).

3.3 Cognitive Behavioral Therapy Interventions for a clinically anxious food allergy population

Only one study has reported on the effectiveness of a CBT-based intervention for children assessed as clinically anxious¹⁰ and only one study for parents with food allergy assessed as clinically anxious,¹⁷ based on a clinical interview. Both studies were published as proof-of-concept or pilot studies and had small sample sizes but reported significant improvements with moderate to large effect sizes. Both utilized psychologists to deliver the intervention. Dahlsgaard et al.¹⁰ evaluated the feasibility, acceptability and proof-of-concept of the Food Allergy Bravery (FAB) brief manualised CBT-based intervention for children (n = 10). These were group sessions facilitated by a psychologist and delivered either in person or via video. Knibb¹⁷ evaluated the benefits of CBT to improve psychological outcomes for mothers of children with food allergy (n = 5) with a face-to face, one-to-one intervention with a psychologist for approximately 1 h per week for 12 weeks. Both studies reported significant improvements for anxiety and QoL, with moderate-to-large effect sizes (d = 1.48–2.43). Knibb also reported significant improvements for depression and worry, again with large effect sizes, and a trend for improvement in stress.

3.4 Cognitive Behavioral therapy interventions for a general food allergy population

Five studies used CBT-based interventions for a general food allergy population. One of these was for children and parents,¹⁶ and three for parents only.^18-20 One study developed an intervention for children and parents for oral immunotherapy.¹¹ The Food Allergy Mastery (FAM) program¹⁶ was developed for children and parents and consisted of six sessions that could be delivered by a masters-level counselor via telehealth over 3 months. Five of the sessions are held with the child and caregiver, with the sixth run as a peer group session. The intervention focuses on managing social situations, identifying symptoms as an allergic reaction or anxiety and transitioning food allergy management from caregivers to the child. In a pilot study, 10 dyads of children and caregivers rated the program as enjoyable and relevant. Descriptive analysis from this pilot study showed that from baseline to follow-up, food allergy knowledge improved, social support increased and self-efficacy remained stable but with improvements in recognizing and treating an allergic reaction.

For interventions offered to parents for general food allergy management, one was in person¹⁸ and two online.^{19, 20} Boyle et al.¹⁸ reported on a brief single session of CBT with telephone support to reduce maternal state anxiety. This was a face-to-face one-to-one session with mothers (n = 200) in an allergy clinic, provided by healthcare professionals trained to a competent standard in CBT techniques. Vreeken-Ross et al.²⁰ ran two online group sessions facilitated by a trainee clinical psychologist, lasting 2 h each within a week of each other, for 41 mothers of children with food allergy. The intervention consisted of psycho-education, managing early signs of anxiety (e.g., graded exposure to situational avoidance, challenging negative thoughts) and parenting skills (e.g., managing parenting hotspots (such as overprotection and perfectionism)) and emotion coaching. Sugunasingha et al.¹⁹ evaluated an RCT comparing an entirely self-guided psychoeducational website aimed at improving self-efficacy and decreasing anxiety among parents of children with food allergy compared to wait-list control.

Results across these studies were mixed, with greater effects seen for the two interventions that were facilitated by a healthcare professional trained in CBT. The single session offered in clinic by Boyle et al.¹⁸ resulted in no difference in state anxiety between intervention and control groups at 6 weeks, however for the subgroup that started with moderate/high levels of anxiety at baseline there was a significant reduction in anxiety at 6 weeks (with a moderate effect size r = 0.5). The intervention also reduced risk perception and salivary cortisol response (with a small effect). For the online group sessions reported by Vreeken-Ross et al.,²⁰ a significant decrease was seen in parental anxiety and self-efficacy across all follow-up assessments with medium-large effects sustained at 3 months (d = .57 and d = .76, respectively). Parent-rated child outcomes for anxiety and QoL showed no significant results; however, there was a trend for short-lived improvement in QoL for 8-to 12-year-olds and a potential for more sustained decrease in anxiety for 6.5-to 16-year-olds, though authors caution these findings due to small subgroup analysis.

For the self-guided online website,¹⁹ no change was seen in any outcomes at any time point and intervention engagement was low with an average of less than 3 minutes spent on the site. Subgroup analysis revealed that for those scoring within the clinical threshold for depression at baseline, food allergy-related QoL did improve though participant numbers were small and results were not sustained after correcting for multiple comparisons. Analysis of baseline data identified that in addition to self-efficacy, intolerance of uncertainty was found to significantly predict parental food allergy-related QoL and could be a focus for future interventions.

Only one intervention has been published to support children undergoing oral immunotherapy (OIT) and their parents.¹¹ Primary outcomes for this RCT included OIT symptom mindset and anxiety, OIT dosing, symptom occurrence, and dose adherence, staff contact, time to OIT completion, and biomarkers associated with desensitization. Outcomes were assessed via questionnaires developed by the study team. All participants (n = 50 children and their parents) attended 7 monthly group sessions. Families in the intervention group endorsed OIT symptoms as positive signals more than control group families. Throughout the course of OIT, intervention group families reported less anxiety related to symptoms, were more likely to indicate OIT dosing went well, were less likely to contact staff about non-life-threatening symptoms, were less likely to experience non-life-threatening symptoms as OIT doze sizes increased, and were less likely to skip OIT doses due to anxiety. IgG4 levels differed between groups pre- and post OIT, but IgE and IgG4/IgE ratios were not significantly different between groups. The authors concluded that patients' cognitions regarding OIT symptoms may affect treatment experiences and outcomes.

3.5 Non-cognitive behavioral therapy interventions for a general food allergy population

Four studies have evaluated interventions that are not based on CBT and all were to help improve day-to-day management of food allergy in a general food allergy population.^12-15 Two studies were individual interventions for parents,^{12, 15} one was an individual intervention for children,¹³ and one was a group workshop for both parents and children.¹⁴

For parents, interventions have been run based on self-regulation or on peer-mentoring. Baptist et al.¹² ran a pilot RCT to see whether a self-regulation intervention could improve QoL in parents of children with food allergy compared to standard care (food allergy educational information). The intervention involved three 25-minute telephone sessions with a trained nurse to set goals, problem-solve and implement coping behaviors. The authors found that food allergy specific QoL significantly improved across four items on the Food Allergy Quality of Life Parental Burden questionnaire (FAQL-PB) from baseline to three months for the intervention group compared to the control group (helplessness, anxiety related to food allergy, feeling frustrated by others, and feeling frightened) though no change was observed in the self-efficacy items, measured using a non-validated study specific scale.

Ramos et al.¹⁵ conducted a pilot study of a 6-month parent peer mentoring program for parents of young children newly diagnosed with food allergy, designed to improve food allergy-related psychosocial functioning. Eight mentors (parents of older children who were diagnosed with food allergy at a young age) were trained by the research team and matched with 10 mentees whose children were under the age of 5 years and had been diagnosed with food allergy within the last year. Authors noted the intervention was feasible, and mentees reported high acceptability for the intervention. Although there were no significant differences in psychosocial outcomes from baseline to post-intervention due to the small sample size, there were trends toward improvements in QoL, self-efficacy, social support, and food allergy knowledge.

The only non-CBT-based intervention for children aimed to improve coping with food allergy. Cole et al.¹³ extended their prior work with the Cellie Coping Kit, an evidence-based psychological intervention for children with cancer, to children with food allergy. They offered a brief in-person intervention session with 25 children aged 7–13 years and their parents, recruited from oral food challenge clinics. A research assistant presented the Cellie Coping Kit for Food Allergies to the children and their parents, providing guidance about how to use the coping cards and utilize them at home. The cards referenced common allergy-related challenges and provided specific guidance on how to cope with them. Cole et al. reported that the Cellie Coping Kit for Food Allergies was rated as acceptable and feasible by the participants. Just over half of participants (65% of children, 57% of parents) reported at follow up that they learned something about how to cope with food allergy from the intervention including ways to cope with anxiety and ways to discuss food allergies. Although the Cellie Coping Kit is primarily for children, this intervention also shows promise in helping parents to find new ways of coping with their child's food allergy.

The only non-CBT based intervention for both parents and children was delivered by LeBovidge et al.¹⁴ as a half-day group workshop designed to support families in their food allergy management and coping and to enhance children's connections to other children diagnosed with food allergy. During the workshop parents listened to a presentation by a pediatric psychologist about child and family food allergy management and coping and participated in group discussion. Children concurrently participated in a group led by a child life specialist that included games and medical play. Primary outcomes included parents' perceived competence at coping with food allergy, assessed via an author-created measure, and parent food allergy-related burden, assessed via the validated FAQL-PB. Results indicated that the group intervention was rated as acceptable to parents. Parental confidence improved from baseline to post-intervention and again at follow-up 4–8 weeks later. There was also a significant decrease in parental burden from baseline to post-intervention. For children, 88% said enjoyed the workshop and 62% rated the medical play as their most favorite, which helped them learn how to manage their food allergy. The authors concluded that there was preliminary support for the use of group interventions to support coping among children with food allergy and their parents.

4.1 Overall effectiveness across interventions

Despite the variety in the interventions reported, improvements were generally seen for intervention groups compared to control groups or in intervention groups over time, across a number of outcome measures. CBT-based interventions reported significant improvements in food-allergy specific or general anxiety, QoL, self-efficacy, worry, and depression. Of the four non-CBT interventions, two reported significant improvements in outcome measures, which included QoL and self-efficacy. One reported increased learning and ability to cope with food allergy and one reported trends for improvements in QoL, self-efficacy, social support, and food allergy knowledge.

Effectiveness of the interventions did seem to be related to the mode of delivery. Interventions offering face-to-face sessions (either in person or online) facilitated by a healthcare professional appeared to provide the best evidence for effectiveness. This has been shown in other long-term conditions where therapist guided internet-based CBT methods resulted in greater reductions in anxiety and depression than self-guided approaches.²¹ Although study sample sizes were generally small and studies were pilot or proof-of-concept studies, medium to large effect sizes were reported, demonstrating potential clinical effectiveness. The one self-help online study did not find any significant improvements in scores but the engagement with the tool was very low. Ways in which engagement can be improved in future studies need to be investigated before this type of intervention can be properly assessed.

Overall, it appears that psychological interventions for children and parents provide some benefits in terms of psychological outcomes. Only two interventions measured any clinical outcomes and so it is not known whether interventions such as those described in this review could result in better clinical outcomes for food allergy, such as fewer accidental food allergy reactions, reduction in use of an AAI or other medication, or reduction in admissions to hospital. Adherence was also not measured so it is unknown if these interventions can improve frequency of carriage of an AAI. CBT-based interventions have been shown to improve adherence in other conditions such as diabetes²² and so this type of outcome should be measured in future studies.

4.2 Strengths and limitations of studies in this review

There are limitations of the studies in this review. Quality ratings showed a distinct lack of diversity with most participants largely representative of white, higher income households. Less consistently or appropriately reported elements of studies included intervention fidelity, completeness of outcome measures and whether confounders had been accounted for in analysis, which limited the studies overall quality ratings. Sample sizes were typically small reflecting the focus of feasibility and acceptability of many of the studies. There was a lack of information on the content of the intervention for many papers and publication of an intervention protocol would be useful. There was great variability in study design, intervention provided, and outcome measures used, and so it was not possible to run a meta-analysis to assess the overall effectiveness of psychological interventions. Most studies used a general food allergy population rather than one that had been identified as having specific food allergy difficulties in need of psychological support, which could have had an impact on the findings. Strengths of the studies in this review include the use of theory-based interventions and well-validated psychometric scales, but adherence and clinical outcomes were not measured, and it was difficult to ascertain whether findings had clinical importance however a number of studies did report effect sizes.

4.3 Recommendations

The results from this systematic review are promising; however, there are gaps in our knowledge base and further research is needed. The key areas that future research needs to address can be seen in Figure 2. It is also important to ensure that reporting of interventions is detailed enough for researchers and healthcare professionals to make informed decisions about the effectiveness of interventions and so recommendations for reporting of interventions can be found in Table 3. Although further research is needed, it is likely that a stepped care model for interventions for children with food allergies and their families will be most beneficial. Interventions in this review targeting patients and parents who had very high anxiety levels showed the best response and improvement. There is a big unmet need for psychological support for food allergy in the UK⁷ and elsewhere across Europe and North America^{23, 24} and it is unlikely that psychological support will be available at every allergy clinic any time in the near future. It is therefore important that support is provided for those who need it most. Face-to-face online interventions should be considered, owing to the ease and lower costs of this style of intervention compared to in-person interventions. A recommended stepped care model can be found in Figure 3, which also highlights the current evidence available for the type of interventions suggested at each step.