- Email: [email protected]
Differences in Caregiver Food Allergy Quality of Life Between Tertiary Care, Specialty Clinic, and Caregiver-Reported Food Allergic Populations
Original Article
Differences in Caregiver Food Allergy Quality of Life Between a Tertiary Care, Specialty Clinic, and a Caregiver Reported Food Allergic Populations Claire Ward, MDa, and Matthew Greenhawt, MD, MBA, MSca,b Ann Arbor, Mich
What is already known about this topic? Food allergy is associated with reduced caregiver quality of life (QoL). However, little is known about how heterogeneous any deficit in QoL may be across a diverse population. What does this article add to our knowledge? Large, significant differences in food allergy QoL exist between caregivers with children followed at referral centers versus self-selected caregivers reporting food allergic children in an online community, despite a common diagnosis. How does this study impact current management guidelines? The food allergic population is heterogeneous, and certain segments experience a disproportionate burden of disease. Such groups may have different needs and risk profiles despite a common diagnosis, which may predispose to different long-term outcomes. BACKGROUND: Food allergy is associated with diminished caregiver quality of life (QoL), but the heterogeneity of this effect is unknown. OBJECTIVE: The objective of this study was to explore potential differences in caregiver QoL between self-selected caregivers reporting a child with food allergy (SS) and caregivers with children followed at a food allergy referral center clinic (RC). METHODS: The Food Allergy Quality of Life Parental Burden (FAQL-PB) index and screening questions regarding the child’s most severe food reaction were administered to caregivers of milk, egg, peanut, or tree nut allergic children. SS were recruited via the email and/or social media networks of 2 large national food allergy advocacy groups, and RC from a tertiary referral center specialty clinic.
a
Division of Allergy and Clinical Immunology, Department of Internal Medicine, University of Michigan Medical School, Ann Arbor, Mich b Department of Pediatrics, Child Health Evaluation and Research Unit, University of Michigan Medical School, Ann Arbor, Mich This study was supported in part by a gift from an anonymous foundation, and by National Center for Advancing Translational Sciences Grant 2UL1TR000433. M. Greenhawt also received support from National Center for Advancing Translational Sciences grant #2KL2TR000434. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. C. Ward reports no funding. Conflicts of interest: The authors declare that they have no relevant conflicts. Received for publication December 4, 2014; revised July 13, 2015; accepted for publication July 30, 2015. Available online -Corresponding author: Matthew Greenhawt, MD, MBA, MSc, Division of Allergy and Clinical Immunology, Department of Pediatrics, Child Health Evaluation and Research Unit, University of Michigan Medical School, 24 Frank Lloyd Wright Dr., Lobby H-2100, Box 442, Ann Arbor, MI 48106. E-mail: mgreenha@med. umich.edu. 2213-2198 Ó 2015 American Academy of Allergy, Asthma & Immunology http://dx.doi.org/10.1016/j.jaip.2015.07.023
RESULTS: Among 2003 SS and 305 RC, the mean total FAQLPB QoL score was 2.67. Compared with SS, RC had a lower (better) mean total QoL score (1.84 vs 2.81, P < .001), individual FAQL-PB domain scores (mean difference range 0.511.93; all P < .001), and lower QoL scores for all allergens (mean difference range 0.89-1.32; peanut P < .001, tree nut P < .001, milk P [ .006, egg P [ .001). In an adjusted multiple linear regression model, RC were associated with a lower QoL score (L1.6 [95% CI, L1.91 to L1.29], P < .001). Factor analysis of the index revealed 2 dimensions. A minimal clinically important difference of 0.3 was calculated for the FAQL-PB using the standard error of measurement method. CONCLUSIONS: Caregiver food allergy QoL is heterogeneous, and worse among SS versus RC. Clinically and statistically significant differences were noted in the total, domain-specific, and allergen-specific QoL scores, which indicated that the food allergic population may be segmented and have different risk profiles and/or burdens of illness, despite a common diagnosis. Ó 2015 American Academy of Allergy, Asthma & Immunology (J Allergy Clin Immunol Pract 2015;-:---) Key words: Food allergy; Quality of life; Food Allergy Quality of Life-Parental Burden; Anaphylaxis; Epinephrine; Self-efficacy; Prevalence; Self-reported food allergy
Food allergy is a growing public health problem that may affect up to 8% of children and 15 million Americans, though estimates may differ based on the methodology used.1-4 Estimates of food allergy prevalence in the United States are based on a proxy report from a caregiver, serologic testing for specific IgE (sIgE), or other indirect methods of assessment such as diagnosis discharge coding, and may vary when compared with methods that use an oral food challenge to confirm diagnosis.1-9 In particular, food allergy may be self-reported in as many as 4to 5-fold more individuals than those in whom it can be verified.10 1
2
WARD AND GREENHAWT
J ALLERGY CLIN IMMUNOL PRACT MONTH 2015
TABLE I. Demographic trends within the populations Abbreviations used FAI- Food allergic individual FAQL-PB- Food Allergy Quality of Life-Parental Burden FASEQ- Food Allergy Self-Efficacy Questionnaire MCID- Minimal clinically important difference QoL- Quality of life RC- Children followed at a food allergy referral center clinic SEM- Standard error of measurement sIgE- Specific IgE SS- Self-selected caregivers reporting a child with food allergy
Food allergy is associated with reduced patient and caregiver health-related quality of life (QoL).11 This has been noted with both generic pediatric QoL indices and through food allergyespecific QoL indices (in both US and international populations).12-24 The degree of heterogeneity within the US food allergic population is poorly understood, but there are 2 general populations of food allergic individuals that have been studied— self-selected caregivers reporting a child with food allergy (SS), and caregivers who have children followed at referral centers (RC). Both populations strongly identify themselves as food allergic, live a food allergic lifestyle, and presumably experience the same problems and QoL reductions.3,16,17,21,25 There has been no prior exploration of QoL differences between SS and RC, and there is limited prior literature exploring other potential differences in specific attributes between these 2 groups that may better explain the heterogeneity within the food allergic population. Thus, it is unknown if the problems, needs, or risk factors for poor QoL are different between these groups. Any potential difference may have significant bearing on future outcomes such as natural history of disease, utilization of health care services, or effects of treatments that may soon become available. Therefore, the purpose of this study was to better understand differences in caregiver QoL within a large food allergic population comprising distinct cohorts (SS and RC), as well as to provide further psychometric validation of the Food Allergy Quality of Life Parental Burden (FAQL-PB) index16 within this large, diverse food allergic population.
METHODS This was a prospective, cross-sectional study of caregiver food allergy QoL, comparing 2 distinct populations. SS were recruited nationally for participation during the summer and fall of 2012 by email contact, social media feeds, and the websites of Kids with Food Allergies and the Food Allergy and Anaphylaxis Network (FAAN, since renamed FARE, Food Allergy Research and Education), 2 large national food allergy advocacy groups. Eligible participants included caregivers who identified themselves as both over the age of 18 and having at least one food allergic individual (FAI) with a physiciandiagnosed food allergy. Caregivers with more than one FAI were asked to retake the survey based on their experience with each FAI independently. Analysis was restricted to those identifying a child with milk, egg, peanut, or tree nut allergy, the 4 most commonly reported allergens in the cohort. RC patients evaluated and managed for milk, egg, peanut, or tree nut allergy at the University of Michigan Division of Allergy and Clinical Immunology clinics between 2009 and 2011 were identified by the review of a divisional food allergy patient database, and were recruited for participation in this study in clinic by mail or by phone
Demographic trends
Current age (y) Age of initial reaction (y) Peanut Tree nut Milk Egg Reported history of anaphylaxis White race Income >$100,000 College graduate Allergist made initial diagnosis Peanut/tree nut allergy
Self-selected caregiver report (n [ 2003)
Clinic (n [ 305)
P
6.9 2.5 2.2 3.6 4.4 2.7 61.9%
7.5 1.9 1.75 3.3 1.4 0.8 37.5%
.03 <.001 NS NS <.001 <.001 <.001
89.3% 54.6% 83.3% 76.9%
79.5% 59.6% 83.3% 86.9%
<.001 NS NS <.001
93.1%
52.8%
<.001
NS, Not significant.
TABLE II. Comparison of symptoms in the populations Symptoms reported
Systemic fives Eczema Other rash Oropharyngeal angioedema Itchy throat Throat tightness Cough Shortness of breath Wheezing Vomiting Abdominal pain Hypotension Syncope
Self-selected caregiver report % (n [ 2003)
Clinic % (n [ 305)
P
49 23.1 10.9 57.4 40.8 38.1 34.4 28.4 26 41.1 21.7 8.1 7.5
31.5 5.6 7.9 28.7 9.2 6.3 11.6 11.6 10.2 33.7 5.6 0 5.3
<.001 <.001 NS <.001 <.001 <.001 <.001 <.001 <.001 .01 <.001 <.001 .17
NS, Not significant.
between November 2011 and August 2012. These 4 allergens represented the most common food allergens seen in the practice. Caregivers were included if their child had a visit in the clinic between 2009 and 2011 and a chart-verified diagnosis of food allergy, defined by a documented, convincing clinical history of an IgEmediated food-induced allergic reaction in the setting of confirmed IgE-mediated sensitization to the reported food (positive skin test and/or serum sIgE), or sIgE and/or a prick skin test wheal >95% positive predictive value for milk, egg, or peanut in a child with atopic dermatitis.10 A total of 1116 families with a unique visit in this 2-year time span were contacted over a year as part of a larger study to create an institutional database, with a total of 572 (51.3%) responding, of whom 92% identified an allergy to either milk, egg, peanut, or tree nut. The selection of those analyzed for chart review was blinded and at random. The data collection was halted at approximately the 300th survey because of financial constraints.23 Caregivers in both cohorts completed separate questionnaires containing identical sections regarding symptoms of the FAI most severe reaction to the allergen, reaction treatment, perception of reaction severity, follow-up care of their food allergy, allergic and
J ALLERGY CLIN IMMUNOL PRACT VOLUME -, NUMBER -
WARD AND GREENHAWT
3
FIGURE 1. Comparison of mean total and individual domain-specific Food Allergy Quality of Life Parental Burden (FAQL-PB) questionnaire scores between the self-selected (caregiver-reported) and clinic groups. The difference was statistically significant (P < .05) in all quality of life (QoL) domains. Error bars represent standard error.
nonallergic comorbidities, anaphylaxis management training, sociodemographic information, the institutionally developed and validated caregiver Food Allergy Self-Efficacy Questionnaire (FASEQ) (Cronbach a ¼ 0.88, intraclass correlation [ICC] ¼ 0.75, Supplemental Text, available in this article’s Online Repository at www.jaciinpractice.org/),26 and the FAQL-PB questionnaire assessing caregiver QoL. RC caregiver-reported questionnaire data were verified through the chart review of the patient’s medical record, to account for potential recall bias. National Institutes of Allergy and Infections Diseases/FAAN anaphylaxis criteria were applied to both the parentreported symptoms and those detailed through the chart review.27 The FAQL-PB (Figure E1, available in this article’s Online Repository at www.jaci-inpractice.org/) is a validated, 17-item, selfadministered questionnaire that measures the effect of pediatric food allergy on caregiver QoL, developed by Cohen et al.16 It has an ICC ¼ 0.93 and Cronbach a ¼ 0.95, indicating both excellent validity and reliability for the cross-sectional assessment of QoL. Caregiver QoL was not assessed at a uniform time from diagnosis during the formulation and validation of this scale. The 17 questions are each 7-point Likert items and the index is scored as a summated rating scale, with a higher FAQL-PB score indicating a worse QoL.16 Dimensionality of the index was not investigated during its validation. The clinical significance of changes in QoL measures is assessed through evaluating a minimal clinically important difference (MCID), the smallest difference in score that patients perceive as beneficial, and would mandate, in the absence of troublesome side effects and effective cost, a change in the patient’s management.28 For 7-point Likert scales analyzing QoL, prior studies have used an approximate measure of 0.5 as an MCID, based on the work by Jaeschke et al in cardiac and pulmonary disease.29 However, 0.5 is not an absolute number adaptable to any 7-point scale, and MCID is considered to be both specific to the particular QoL index and population being studied.30,31 Therefore, no true MCID is known for the FAQL-PB, and 0.5 is speculative.16 Index MCID can be calculated by the assessment of clinical impact through global rating questions, number needed to treat analysis, or through quantifying the standard error of measurement (SEM).32
Statistical analysis Data were analyzed using frequency analysis to report descriptive statistics, and c2/Fisher’s exact tests and Student’s t-test for bivariate
comparison. Adjusted multiple linear regression was performed to explore the relationships between a caregiver’s QoL score and the variables assessed in the questionnaire and our previous studies. Analysis was prespecified based on ongoing QoL studies by our research group.23,24 Multicollinearity of independent variables was assessed using the inspection of the variance inflation factor, and model specificity through the stata linktest (model specification link assessment) and ovtest (Ramsey reset test for omitted variables) commands. The stata margins command was used to assess posterior regression predictive probabilities. Factor analysis was performed to determine index dimensionality using the iterative principal components method with rotation. Scree plots were used to determine the number of relevant factors to retain. MCID for the index was calculated using the SEM method within the individual and total populations, by the following formula: SEM ¼ Standard deviation * square root [(1 reliability)].32 Data were analyzed with Stata SE, Version 13 (Stata Corp., College Station, Tex). Based on a level of significance of 0.05, the study had 80% power to detect a 0.5 difference in QoL score between groups with 65 patients per arm. This study was approved by the University of Michigan Medical School institutional review board.
RESULTS Differences in baseline demographics and reaction attributes A total of n ¼ 302 RC and n ¼ 2003 SS were included in the analysis. Table I details the comparative population demographics. Table II details the comparative presenting symptoms of the most severe allergic reaction. There were numerous significant differences between the groups, in particular in the type and frequency of reported symptoms, and significant predominance of reported peanut and tree nut allergy among the SS. RC had an earlier onset of allergy, but were slightly older at the time their caregiver was surveyed. Both populations were predominantly white and the caregivers college educated. The majority of SS reported that their child was under the care of an allergist. Differences in food allergy QoL Differences in RC and SS QoL were highly apparent. The mean total QoL score and all 17 individual domain-specific QoL scores were significantly different between groups, with higher
4
WARD AND GREENHAWT
J ALLERGY CLIN IMMUNOL PRACT MONTH 2015
FIGURE 2. Comparison of mean total Food Allergy Quality of Life Parental Burden (FAQL-PB) scores, stratified by specific food allergen, between the self-selected (caregiver-reported) and clinic groups. The difference was statistically significant (P < .05) in all quality of life (QoL) domains. Error bars represent standard error.
FIGURE 3. Comparison of mean total Food Allergy Quality of Life Parental Burden (FAQL-PB) scores between the self-selected (caregiverreported) and clinic groups, stratified by history/absence of reported anaphylaxis (A), epinephrine use (B), milk/egg allergy (C), and peanut/tree nut allergy (D). The difference was statistically significant (P < .05) in all quality of life (QoL) domains except where striped. Error bars represent standard error.
mean scores (worse QoL) in SS (mean total QoL score 2.81 [95% CI, 2.74-2.86] vs 1.84 [95% CI, 1.68-1.99], difference 0.96, P < .001). The average domain-specific difference was 0.99 between the groups (domain range 0.51-1.93, domain difference standard error ¼ 0.38) (Figure 1). When individually
explored by milk, egg, peanut, or tree nut allergy, mean QoL scores were significantly higher in SS for all respective allergens (Figure 2). When stratified by just caregivers of FAI with prior reported epinephrine use, mean total and 9/17 domain-specific QoL scores were significantly higher in SS than RC (2.98
WARD AND GREENHAWT
J ALLERGY CLIN IMMUNOL PRACT VOLUME -, NUMBER -
5
TABLE III. Adjusted linear regression models predicting the quality of life score Quality of life score
Coefficient
P
95% CI
Peanut or tree nut allergy
Milk or egg allergy
Referral clinic group 1.60 <.0001 1.91 to 1.29 1.69 (2.03 to 1.35) 0.91 (1.7 to 0.12) Reported anaphylaxis 0.27 .005 0.08 to 0.45 0.22 (0.02 to 0.41) NS Required epinephrine 0.59 .005 0.21 to 1.03 0.58 (0.15 to 1) NS Anaphylaxis/required epinephrine 0.54 .02 0.98 to 0.09 0.51 (0.98 to 0.05) NS interaction Self-efficacy (FASEQ score) 0.44 <.0001 0.35 to 0.54 0.5 (0.4 to 0.6) NS Peanut or tree nut allergy 0.48 <.0001 0.72 to 0.24 e e Other child with FA 0.24 .006 0.07 to 0.42 0.29 (0.1 to 0.47) NS Intercept 2.03 <.0001 1.5 to 2.6 1.37 (0.85 to 1.9) 2.6 (1 to 4.21) Adjusted for age of reaction, allergist diagnosis, income, age and age2, education, race, total number of food allergies, and geographic region. Allergen and geographic region were excluded in sensitivity analyses. FA, Food allergy; FASEQ, Food Allergy Self-Efficacy Questionnaire; NS, not significant.
[95% CI, 2.87-3.09] vs 2.23 [95% CI, 1.89-2.57], difference 0.75, P < .001). Similarly, when stratified by caregivers of FAI with prior reported anaphylaxis, mean total and 16/17 domainspecific QoL scores were significantly higher in SS than RC (mean total score 2.9 [95% CI, 2.81-2.98] vs 2.09 [95% CI, 1.82-2.35], difference 0.81, P < .001) (Figure 3, A-D). Supplemental Table E1 (available in this article’s Online Repository at www.jaci-inpractice.org) details the item specific mean differences in score by population for these aforementioned trends. A multivariable linear regression model was created to better understand the association of participant attributes with the QoL score (Table III). In the adjusted analysis, RC were significantly associated with a lower QoL score (better QoL), confirming what was observed at the bivariate level. Additionally, having a peanut or tree nut allergic child (compared with milk or egg) was also significantly associated with a lower QoL score. Covariates significantly associated with a higher (worse) QoL score included anaphylaxis, epinephrine use, increasing (worsening) FASEQ score, and having multiple FAI. An interaction was noted between reported epinephrine use in individuals with a reported history of anaphylaxis—a lower QoL score was noted in those with both reported anaphylaxis and epinephrine use, but independently (eg, if only one of these attributes was reported but not the other) these were both associated with a higher (worse) QoL score. A separate sensitivity analysis of this regression model restricted to only those with peanut or tree nut allergic FAI, and just milk or egg allergic FAI both showed persistence of the significant association between RC and lower QoL score (Table III).
Psychometric performance of the FAQL-PB within clinical and self-report populations The psychometric performance of the FAQL-PB was assessed within each individual population and the total population. Cronbach a was nearly identical within each population (total group 0.95, SS 0.94, and RC 0.96), which indicated consistent reliability. An index-specific MCID was calculated using the SEM method, which was 0.31 in the RC population, 0.34 in the SS, and 0.32 in the combined population. The FAQL-PB lacks factor analysis and elucidation of its dimensionality within US populations, though this was recently described in a UK population.33 Thus, we performed a factor analysis within each individual population, and for the total combined population to assess underlying dimensionality and item-theme correlations. Using the iterative principal factor
method, 2 factors each were identified in all 3 groupings (SS, RC, and the combined group). Orthogonal and oblique rotations were performed, with the oblique rotation retained to maximize correlations between factors. This identified 12 items in the combined population, 13 items in RC, and 14 items in SS that loaded onto factor 1. Three items consistently loaded onto factor 2 in all 3 populations: vacation planning, restaurant choice in dining out, and limitations on social activities. Table IV details the factor loadings for the groups. Factor 1 appears to have a theme of general health concerns and anxiety related to the FAI. Factor 2 strongly identifies a theme of social interaction with others.
DISCUSSION In this study, we aimed to explore if any differences in QoL were detectable based on being selected from an academic center cohort or through a food allergy advocacy group, and to explore specific performance of the index within these different groups to further validate it as a robust outcome measure. Herein we show that caregiver food allergy QoL is heterogeneous, and dependent on multiple factors, including a large effect related to the “population” through which we surveyed a caregiver. This study offers some comparison of these 2 population segments and shows a range of possible reported symptoms, management, and outcomes of how a similar disease is managed. Limited previous data exist that compare and describe such differences, though more recently the validity of self-reported populations has been challenged, given evidence that there may be overestimation and poor validation, despite the fact that the majority of the data describing food allergy in the United States are from selfreport.1,2,4,5,7,10,24,33-35 What we demonstrate is that among individuals living a “food allergic lifestyle,” some have a much more prescribed deficit in terms of poor food allergy QoL, though the reasons for this are less clear, and were beyond the scope of this present study but may be related to many possible issues including caregiver motivation, individualized effect of diagnosis, or relationship with their provider, for example. Follow-up work to better understand these patterns and influence on the observed segmentation is underway. Our data highlight that a more individualized approach to understanding the daily lives of our food allergic families may be needed, given evidence that different populations have different problems, needs, and risk factors contributing to their poor QoL. Such differences may significantly impact future treatment options or utilization of care if these are not better understood.
6
WARD AND GREENHAWT
J ALLERGY CLIN IMMUNOL PRACT MONTH 2015
TABLE IV. Factor analysis of the Food Allergy Quality of Life Parental Burden in the 3 populations Oblique rotation loadings Domain
Combined population Troubled with the thought your child will have a reaction Troubled by possibility concerns for your child’s health Troubled that child may not overcome their food allergy Troubled you would be unable to help treat a reaction Troubled by the sadness of the burden your child carries Troubled by possibility of leaving child in care of others Troubled by anxiety Troubled your child will not have a normal upbringing Troubled by fear of child being near others while eating Troubled by your child attending group activities Troubled that others lack appreciation of severity Troubled by need to take precaution before leaving home Troubled by concerns for your child’s nutrition Limitation in choice of vacation plans Limitation in choice of restaurant Limitation in participation in social activities Troubled by need to spend extra time preparing meals Factor correlation Clinic population Troubled with the thought your child will have a reaction Troubled that others lack appreciation of severity Troubled that child may not overcome their food allergy Troubled by your child attending group activities Troubled by possibility of leaving child in care of others Troubled by the sadness of the burden your child carries Troubled by possibility concerns for your child’s health Troubled you would be unable to help treat a reaction Troubled by fear of child being near others while eating Troubled by anxiety Troubled your child will not have a normal upbringing Troubled by concerns for your child’s nutrition Limitation in choice of restaurant Limitation in choice of vacation plans Limitation in participation in social activities Troubled by need to spend extra time preparing meals Troubled by need to take precaution before leaving home Factor correlation Self-selected caregiver-report population Troubled with the thought your child will have a reaction Troubled by possibility concerns for your child’s health Troubled that child may not overcome their food allergy Troubled you would be unable to help treat a reaction Troubled by the sadness of the burden your child carries Troubled by anxiety Troubled your child will not have a normal upbringing Troubled by possibility of leaving child in care of others Troubled by fear of child being near others while eating Troubled by your child attending group activities Troubled that others lack appreciation of severity Troubled by need to take precaution before leaving home Troubled by concerns for your child’s nutrition Troubled by need to spend extra time preparing meals
Factor 1
Factor 2
Uniqueness
0.926 0.859 0.791 0.829 0.768 0.741 0.715 0.706 0.708 0.699 0.679 0.406 0.471 0.0875 0.0799 0.1054 0.3689 0.6375
0.1063 0.034 0.0023 0.1332 0.0447 0.089 0.1376 0.1362 0.108 0.1141 0.0765 0.3938 0.1193 0.884 0.881 0.713 0.382
0.256 0.298 0.372 0.435 0.364 0.359 0.344 0.36 0.389 0.396 0.467 0.476 0.692 0.308 0.306 0.384 0.538
0.9447 0.8108 0.8054 0.79 0.7952 0.795 0.777 0.7457 0.7055 0.6565 0.5046 0.3503 0.0645 0.0375 0.0838 0.1871 0.3277 0.6832
0.086 0.0056 0.0678 0.0864 0.1082 0.05 0.1039 0.0471 0.1369 0.2149 0.3449 0.2649 0.9122 0.8693 0.769 0.627 0.5326
0.211 0.349 0.272 0.275 0.238 0.311 0.275 0.49 0.352 0.33 0.389 0.68 0.244 0.287 0.314 0.411 0.371
0.9181 0.8599 0.7798 0.8322 0.7798 0.721 0.7265 0.7324 0.7095 0.6888 0.6655 0.4183 0.4756 0.3872
0.112 0.0486 0.0058 0.1425 0.0058 0.1235 0.1056 0.08 0.0912 0.1078 0.0674 0.3664 0.1117 0.3527
0.27 0.309 0.386 0.433 0.386 0.355 0.366 0.385 0.408 0.422 0.497 0.502 0.696 0.557 (continued)
WARD AND GREENHAWT
J ALLERGY CLIN IMMUNOL PRACT VOLUME -, NUMBER -
7
TABLE IV. (Continued) Oblique rotation loadings Domain
Factor 1
Factor 2
Limitation in choice of vacation plans Limitation in choice of restaurant Limitation in participation in social activities Factor correlation
0.098 0.091 0.102 0.6157
0.8724 0.861 0.696
Uniqueness
0.335 0.3347 0.417
Bolded values show which items load to which of the 2 factors.
In showing that food allergy QoL is significantly different between RC and SS, we suggest that these may be distinct, separate subgroups within the food allergy population likely with differing needs and risk profiles. Caregiver QoL was highly differentiable based on the specific population, with scores nearly a full 1-point worse at a bivariate level and 1.6-point worse in a multiple regression model between RC and SS. These differences exceed both the calculated MCID of 0.32 and the estimated MCID of 0.5 by several fold. We further demonstrate QoL differences between populations based on type of food allergen, prior anaphylaxis, prior use of epinephrine, and presence of multiple FAI in the family. Each of these findings emphasizes how the same illness is not perceived and experienced in the same manner across a diverse population—an important and novel realization that suggests the burden of the same disease is disproportionate. More research is needed to understand why the burden is experienced disproportionately, and pinpoint more specific details that moderate these differences, to enhance understanding of why these exist and what their long-term implications may be. For example, there is need to further elucidate the significance of, and reasons why peanut and/or tree nut allergy is associated with a better adjusted QoL than milk and/or egg allergy—certainly a tailored approach by allergen, within a population segment, may be a tool to lead to better outcomes. We also show data supporting a relationship between selfefficacy and QoL score (increasing [worsening] self-efficacy score is associated with increasing [worsening] QoL score), which seems somewhat intuitive. Self-efficacy is an intriguing predictor of QoL, and may serve as a point of future intervention to help improve QoL, or an independent stand-alone outcome to follow. This is an emerging area of study, and a manuscript more formally describing the FASEQ index and the relationship between self-efficacy and QoL is forthcoming by our group. We also present novel data regarding the psychometric properties of the FAQL-PB in US populations. The FAQL-PB has good performance with high reliability and minimally differing dimensionality in both RC and SS.16,33 This is the first assessment of the dimensionality of this index in a US population. Knibb and Stalker recently validated and performed the factor analysis of the FAQL-PB in a British population, but noted different dimensionality (3 dimensions) than we present.33 This should not be surprising given well-established data that QoL is culturally specific.12,31,36,37 Most importantly, we provide an index-specific (and food allergy population specific) estimate of clinical significance, which is novel. This index-specific MCID can now be used to further enhance the clinical applications of the FAQL-PB as a cross-sectional patient-reported outcome measure. There are several limitations of this study. Foremost this utilized prospectively collected QoL data, whereas all associated covariates used for comparison and adjustment of findings were
based on retrospective data. All such attributes of food allergy were recorded entirely by caregiver report, and the data are subject to both recall bias and possible challenges to validity. This information was verified by chart review in the clinic sample. Participants from the self-selected caregiver-reported cohort were recruited through social media tied to national advocacy groups, with more than 50% of caregivers reporting membership in at least one such group. This may infer some degree of selection bias or bias resulting from another unmeasured attribute tied to the propensity for advocacy group membership, though group membership was adjusted for and was not significant in the regression analysis. Earlier US QoL studies have dealt with similar potential limitations.16 However, there are presently few other feasible and practical methods besides caregiver report to study food allergy at a national level. We were not able to ascertain the degree to which the clinic cohort utilizes similar online resources or participates in such groups. In conclusion, we show that caregiver food allergy QoL is heterogeneous, and dependent on multiple factors, including the segment of the “population” to which one belongs. These data provide evidence of heterogeneity of disease, inferring that distinct segmentation within the general food allergic population may exist, likely with differing needs, risk profiles, and patientreported outcomes. Identifying that certain self-selected caregiver-reported groups may experience a disproportionate burden of the same illness (with different patient-reported outcomes) is an important finding, which will allow further study to develop more tailored ways to help these individuals. We also further validate the FAQL-PB as an instrument for tracking crosssectional caregiver QoL by deriving an MCID to denote clinical change, and the underlying dimensional structure of the index. These properties should enhance the clinical applicability of the FAQL-PB. REFERENCES 1. Branum AM, Lukacs SL. Food allergy among U.S. children: trends in prevalence and hospitalizations. NCHS Data Brief 2008;(10):1-8. 2. Branum AM, Lukacs SL. Food allergy among children in the United States. Pediatrics 2009;124:1549-55. 3. Gupta RS, Springston EE, Warrier MR, Smith B, Kumar R, Pongracic J, et al. The prevalence, severity, and distribution of childhood food allergy in the United States. Pediatrics 2011;128:e9-17. 4. Osborne NJ, Koplin JJ, Martin PE, Gurrin LC, Lowe AJ, Matheson MC, et al. Prevalence of challenge-proven IgE-mediated food allergy using populationbased sampling and predetermined challenge criteria in infants. J Allergy Clin Immunol 2011;127:668-76. e1-2. 5. Sicherer SH, Munoz-Furlong A, Sampson HA. Prevalence of seafood allergy in the United States determined by a random telephone survey. J Allergy Clin Immunol 2004;114:159-65. 6. Sicherer SH, Munoz-Furlong A, Sampson HA. Prevalence of peanut and tree nut allergy in the United States determined by means of a random digit dial telephone survey: a 5-year follow-up study. J Allergy Clin Immunol 2003;112: 1203-7.
8
WARD AND GREENHAWT
7. Sicherer SH, Munoz-Furlong A, Godbold JH, Sampson HA. US prevalence of self-reported peanut, tree nut, and sesame allergy: 11-year follow-up. J Allergy Clin Immunol 2010;125:1322-6. 8. Liu AH, Jaramillo R, Sicherer SH, Wood RA, Bock SA, Burks AW, et al. National prevalence and risk factors for food allergy and relationship to asthma: results from the National Health and Nutrition Examination Survey 2005-2006. J Allergy Clin Immunol 2010;126:798-806.e13. 9. Greenhawt M, Weiss C, Conte ML, Doucet M, Engler A, Camargo CA Jr. Racial and ethnic disparity in food allergy in the United States: a systematic review. J Allergy Clin Immunol Pract 2013;1:378-86. 10. Boyce JA, Assa’ad A, Burks AW, Jones SM, Sampson HA, Wood RA, et al. Guidelines for the diagnosis and management of food allergy in the United States: report of the NIAID-sponsored expert panel. J Allergy Clin Immunol 2010;126:S1-58. 11. van der Velde JL, Dubois AE, Flokstra-de Blok BM. Food allergy and quality of life: what have we learned? Curr Allergy Asthma Rep 2013;13:651-61. 12. Salvilla SA, Dubois AE, Flokstra-de Blok BM, Panesar SS, Worth A, Patel S, et al. Disease-specific health-related quality of life instruments for IgE-mediated food allergy. Allergy 2014;69:834-44. 13. Primeau MN, Kagan R, Joseph L, Lim H, Dufresne C, Duffy C, et al. The psychological burden of peanut allergy as perceived by adults with peanut allergy and the parents of peanut-allergic children. Clin Exp Allergy 2000;30: 1135-43. 14. Avery NJ, King RM, Knight S, Hourihane JO. Assessment of quality of life in children with peanut allergy. Pediatr Allergy Immunol 2003;14:378-82. 15. Akeson N, Worth A, Sheikh A. The psychosocial impact of anaphylaxis on young people and their parents. Clin Exp Allergy 2007;37:1213-20. 16. Cohen BL, Noone S, Munoz-Furlong A, Sicherer SH. Development of a questionnaire to measure quality of life in families with a child with food allergy. J Allergy Clin Immunol 2004;114:1159-63. 17. Cummings AJ, Knibb RC, King RM, Lucas JS. The psychosocial impact of food allergy and food hypersensitivity in children, adolescents and their families: a review. Allergy 2010;65:933-45. 18. DunnGalvin A, de BlokFlokstra BM, Burks AW, Dubois AE, Hourihane JO. Food allergy QoL questionnaire for children aged 0-12 years: content, construct, and cross-cultural validity. Clin Exp Allergy 2008;38:977-86. 19. Flokstra-de Blok BM, DunnGalvin A, Vlieg-Boerstra BJ, Oude Elberink JN, Duiverman EJ, Hourihane JO, et al. Development and validation of the selfadministered Food Allergy Quality of Life Questionnaire for adolescents. J Allergy Clin Immunol 2008;122:139-44. 144 e1-2. 20. Flokstra-de Blok BM, van der Velde JL, Vlieg-Boerstra BJ, Oude Elberink JN, DunnGalvin A, Hourihane JO, et al. Health-related quality of life of food allergic patients measured with generic and disease-specific questionnaires. Allergy 2010;65:1031-8. 21. Springston EE, Smith B, Shulruff J, Pongracic J, Holl J, Gupta RS. Variations in quality of life among caregivers of food allergic children. Ann Allergy Asthma Immunol 2010;105:287-294.e3.
J ALLERGY CLIN IMMUNOL PRACT MONTH 2015
22. van der Velde JL, Flokstra-de Blok BM, de Groot H, Oude-Elberink JN, Kerkhof M, Duiverman EJ, et al. Food allergy-related quality of life after double-blind, placebo-controlled food challenges in adults, adolescents, and children. J Allergy Clin Immunol 2012;130:1136-1143 e2. 23. Howe L, Franxman T, Teich E, Greenhawt M. What affects quality of life among caregivers of food allergic children? Ann Allergy Asthma Immunol 2014;113:69-74. 24. Franxman T, Howe L, Teich E, Greenhawt M. Oral food challenge and food allergy quality of life in caregivers of food allergic children. J Allergy Clin Immunol Pract 2015;3:50e56. 25. Resnick ES, Pieretti MM, Maloney J, Noone S, Munoz-Furlong A, Sicherer SH. Development of a questionnaire to measure quality of life in adolescents with food allergy: the FAQL-teen. Ann Allergy Asthma Immunol 2010;105:364-8. 26. Baptist AP, Dever SI, Greenhawt MJ, Polmear-Swendris N, McMorris MS, Clark NM. A self-regulation intervention can improve quality of life for families with food allergy. J Allergy Clin Immunol 2012;130:263-265.e6. 27. Sampson HA, Munoz-Furlong A, Campbell RL, Adkinson NF Jr, Bock SA, Branum A, et al. Second symposium on the definition and management of anaphylaxis: summary report—Second National Institute of Allergy and Infectious Disease/Food Allergy and Anaphylaxis Network symposium. J Allergy Clin Immunol 2006;117:391-7. 28. Patrick DL, Bergner M. Measurement of health status in the 1990s. Annu Rev Public Health 1990;11:165-83. 29. Jaeschke R, Singer J, Guyatt GH. Measurement of health status. Ascertaining the minimal clinically important difference. Control Clin Trials 1989;10:407-15. 30. Wright A, Hannon J, Hegedus EJ, Kavchak AE. Clinimetrics corner: a closer look at the minimal clinically important difference (MCID). J Man Manip Ther 2012;20:160-6. 31. Cook CE. Clinimetrics corner: the minimal clinically important change score (MCID): a necessary pretense. J Man Manip Ther 2008;16:E82-3. 32. Guyatt GH, Osoba D, Wu AW, Wyrwich KW, Norman GR. Methods to explain the clinical significance of health status measures. Mayo Clin Proc 2002;77: 371-83. 33. Knibb RC, Stalker C. Validation of the Food Allergy Quality of Life-Parental Burden Questionnaire in the UK. Qual Life Res 2013;22:1841-9. 34. Chafen JJ, Newberry SJ, Riedl MA, Bravata DM, Maglione M, Suttorp MJ, et al. Diagnosing and managing common food allergies: a systematic review. JAMA 2010;303:1848-56. 35. Jackson KD, Howie LD, Akinbami LJ. Trends in allergic conditions among children: United States, 1997e2011. NCHS Data Brief 2013;(121):1-8. 36. Muraro A, Dubois AE, Dunngalvin A, Hourihane JO, de Jong NW, Meyer R, et al. EAACI Food Allergy and Anaphylaxis Guidelines. Food allergy healthrelated quality of life measures. Allergy 2014;69:845-53. 37. Goossens NJ, Flokstra-de Blok BM, van der Meulen GN, Arnlind MH, Asero R, Barreales L, et al. Health-related quality of life in food-allergic adults from eight European countries. Ann Allergy Asthma Immunol 2014;113:63-68.e1.
J ALLERGY CLIN IMMUNOL PRACT VOLUME -, NUMBER -
Supplemental Text: Background Detail of the FASEQ The Food Allergy Self-efficacy Questionnaire (FASEQ) is an 8 Likert-item index, designed to assess a caregiver’s confidence in managing various aspects of raising a food allergic child. This was inspired by a similar self-efficacy measure used in asthma, with wording of themes changed to reflect terminology for food allergy. This was assessed for face validity internally within the University of Michigan Division of Allergy and Clinical Immunology, in conjunction with a team from the University of Michigan School of Public Health’s Center for Managing Chronic Disease that has expertise in self-regulatory behavior
WARD AND GREENHAWT
8.e1
and self-efficacy. The index was then piloted as part of a secondary outcome measure in a small randomized controlled trial of a self-regulation intervention for anaphylaxis measurement.26 Items in the FASEQ are scored similarly to the Food Allergy Quality of Life Parental Burden (FAQL-PB), as a summated rating scale with lower score reflective of better self-efficacy, though a 5-point scale as opposed to a 7-point scale (FAQLPB) is used. The index has undergone formal validation (as described in the Methods section), as well as additional psychometric validation. A separate publication is forthcoming that details the specific performance of the FASEQ, including its validation.
8.e2
WARD AND GREENHAWT
J ALLERGY CLIN IMMUNOL PRACT MONTH 2015
FIGURE E1. The FAQL-PB quality of life index.
WARD AND GREENHAWT
J ALLERGY CLIN IMMUNOL PRACT VOLUME -, NUMBER -
TABLE E1. Mean differences in QoL score by individual analysis between the clinic and self-selected caregiver-reported populations Domain
Total Vacation plans Restaurant choice Social activities Meal preparation Public travel Anxiety Not overcome Other caregiver Others appreciate severity Child’s burden Group activity Child’s health Unable to help Normal upbringing Nutrition Near others Risk of reaction
Anaphylaxis
Epinephrine Peanut/tree Milk/egg use nut allergy allergy Overall
0.81 1.45 1.49 1.20 0.74 0.76 0.53 0.16 0.44 0.62
0.75 1.48 1.52 1.53 0.86 0.83 0.48 0.11 0.29 0.44
1.22 1.82 2.11 1.46 0.98 1.08 1.10 0.86 0.98 1.26
1.08 1.27 1.69 1.18 1.09 1.25 0.88 0.53 0.66 0.81
0.96 1.37 1.66 1.21 0.62 1.93 0.83 0.58 0.76 1.03
1.11 0.51 0.60 0.37 1.23
1.20 0.31 0.47 0.34 1.19
1.55 1.05 1.12 0.78 1.34
1.21 0.92 0.74 0.40 1.22
1.24 0.83 0.79 0.61 1.11
0.39 0.91 0.46
0.33 0.67 0.54
0.88 1.23 1.04
1.34 0.96 0.60
0.51 1.05 0.83
Bold values represent those not exceeding the proposed minimal clinical important difference. Italicized values represent differences that were not statistically significant.
8.e3
Differences in Caregiver Food Allergy Quality of Life Between a Tertiary Care, Specialty Clinic, and a Caregiver Reported Food Allergic Populations Claire Ward, MDa, and Matthew Greenhawt, MD, MBA, MSca,b Ann Arbor, Mich
What is already known about this topic? Food allergy is associated with reduced caregiver quality of life (QoL). However, little is known about how heterogeneous any deficit in QoL may be across a diverse population. What does this article add to our knowledge? Large, significant differences in food allergy QoL exist between caregivers with children followed at referral centers versus self-selected caregivers reporting food allergic children in an online community, despite a common diagnosis. How does this study impact current management guidelines? The food allergic population is heterogeneous, and certain segments experience a disproportionate burden of disease. Such groups may have different needs and risk profiles despite a common diagnosis, which may predispose to different long-term outcomes. BACKGROUND: Food allergy is associated with diminished caregiver quality of life (QoL), but the heterogeneity of this effect is unknown. OBJECTIVE: The objective of this study was to explore potential differences in caregiver QoL between self-selected caregivers reporting a child with food allergy (SS) and caregivers with children followed at a food allergy referral center clinic (RC). METHODS: The Food Allergy Quality of Life Parental Burden (FAQL-PB) index and screening questions regarding the child’s most severe food reaction were administered to caregivers of milk, egg, peanut, or tree nut allergic children. SS were recruited via the email and/or social media networks of 2 large national food allergy advocacy groups, and RC from a tertiary referral center specialty clinic.
a
Division of Allergy and Clinical Immunology, Department of Internal Medicine, University of Michigan Medical School, Ann Arbor, Mich b Department of Pediatrics, Child Health Evaluation and Research Unit, University of Michigan Medical School, Ann Arbor, Mich This study was supported in part by a gift from an anonymous foundation, and by National Center for Advancing Translational Sciences Grant 2UL1TR000433. M. Greenhawt also received support from National Center for Advancing Translational Sciences grant #2KL2TR000434. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. C. Ward reports no funding. Conflicts of interest: The authors declare that they have no relevant conflicts. Received for publication December 4, 2014; revised July 13, 2015; accepted for publication July 30, 2015. Available online -Corresponding author: Matthew Greenhawt, MD, MBA, MSc, Division of Allergy and Clinical Immunology, Department of Pediatrics, Child Health Evaluation and Research Unit, University of Michigan Medical School, 24 Frank Lloyd Wright Dr., Lobby H-2100, Box 442, Ann Arbor, MI 48106. E-mail: mgreenha@med. umich.edu. 2213-2198 Ó 2015 American Academy of Allergy, Asthma & Immunology http://dx.doi.org/10.1016/j.jaip.2015.07.023
RESULTS: Among 2003 SS and 305 RC, the mean total FAQLPB QoL score was 2.67. Compared with SS, RC had a lower (better) mean total QoL score (1.84 vs 2.81, P < .001), individual FAQL-PB domain scores (mean difference range 0.511.93; all P < .001), and lower QoL scores for all allergens (mean difference range 0.89-1.32; peanut P < .001, tree nut P < .001, milk P [ .006, egg P [ .001). In an adjusted multiple linear regression model, RC were associated with a lower QoL score (L1.6 [95% CI, L1.91 to L1.29], P < .001). Factor analysis of the index revealed 2 dimensions. A minimal clinically important difference of 0.3 was calculated for the FAQL-PB using the standard error of measurement method. CONCLUSIONS: Caregiver food allergy QoL is heterogeneous, and worse among SS versus RC. Clinically and statistically significant differences were noted in the total, domain-specific, and allergen-specific QoL scores, which indicated that the food allergic population may be segmented and have different risk profiles and/or burdens of illness, despite a common diagnosis. Ó 2015 American Academy of Allergy, Asthma & Immunology (J Allergy Clin Immunol Pract 2015;-:---) Key words: Food allergy; Quality of life; Food Allergy Quality of Life-Parental Burden; Anaphylaxis; Epinephrine; Self-efficacy; Prevalence; Self-reported food allergy
Food allergy is a growing public health problem that may affect up to 8% of children and 15 million Americans, though estimates may differ based on the methodology used.1-4 Estimates of food allergy prevalence in the United States are based on a proxy report from a caregiver, serologic testing for specific IgE (sIgE), or other indirect methods of assessment such as diagnosis discharge coding, and may vary when compared with methods that use an oral food challenge to confirm diagnosis.1-9 In particular, food allergy may be self-reported in as many as 4to 5-fold more individuals than those in whom it can be verified.10 1
2
WARD AND GREENHAWT
J ALLERGY CLIN IMMUNOL PRACT MONTH 2015
TABLE I. Demographic trends within the populations Abbreviations used FAI- Food allergic individual FAQL-PB- Food Allergy Quality of Life-Parental Burden FASEQ- Food Allergy Self-Efficacy Questionnaire MCID- Minimal clinically important difference QoL- Quality of life RC- Children followed at a food allergy referral center clinic SEM- Standard error of measurement sIgE- Specific IgE SS- Self-selected caregivers reporting a child with food allergy
Food allergy is associated with reduced patient and caregiver health-related quality of life (QoL).11 This has been noted with both generic pediatric QoL indices and through food allergyespecific QoL indices (in both US and international populations).12-24 The degree of heterogeneity within the US food allergic population is poorly understood, but there are 2 general populations of food allergic individuals that have been studied— self-selected caregivers reporting a child with food allergy (SS), and caregivers who have children followed at referral centers (RC). Both populations strongly identify themselves as food allergic, live a food allergic lifestyle, and presumably experience the same problems and QoL reductions.3,16,17,21,25 There has been no prior exploration of QoL differences between SS and RC, and there is limited prior literature exploring other potential differences in specific attributes between these 2 groups that may better explain the heterogeneity within the food allergic population. Thus, it is unknown if the problems, needs, or risk factors for poor QoL are different between these groups. Any potential difference may have significant bearing on future outcomes such as natural history of disease, utilization of health care services, or effects of treatments that may soon become available. Therefore, the purpose of this study was to better understand differences in caregiver QoL within a large food allergic population comprising distinct cohorts (SS and RC), as well as to provide further psychometric validation of the Food Allergy Quality of Life Parental Burden (FAQL-PB) index16 within this large, diverse food allergic population.
METHODS This was a prospective, cross-sectional study of caregiver food allergy QoL, comparing 2 distinct populations. SS were recruited nationally for participation during the summer and fall of 2012 by email contact, social media feeds, and the websites of Kids with Food Allergies and the Food Allergy and Anaphylaxis Network (FAAN, since renamed FARE, Food Allergy Research and Education), 2 large national food allergy advocacy groups. Eligible participants included caregivers who identified themselves as both over the age of 18 and having at least one food allergic individual (FAI) with a physiciandiagnosed food allergy. Caregivers with more than one FAI were asked to retake the survey based on their experience with each FAI independently. Analysis was restricted to those identifying a child with milk, egg, peanut, or tree nut allergy, the 4 most commonly reported allergens in the cohort. RC patients evaluated and managed for milk, egg, peanut, or tree nut allergy at the University of Michigan Division of Allergy and Clinical Immunology clinics between 2009 and 2011 were identified by the review of a divisional food allergy patient database, and were recruited for participation in this study in clinic by mail or by phone
Demographic trends
Current age (y) Age of initial reaction (y) Peanut Tree nut Milk Egg Reported history of anaphylaxis White race Income >$100,000 College graduate Allergist made initial diagnosis Peanut/tree nut allergy
Self-selected caregiver report (n [ 2003)
Clinic (n [ 305)
P
6.9 2.5 2.2 3.6 4.4 2.7 61.9%
7.5 1.9 1.75 3.3 1.4 0.8 37.5%
.03 <.001 NS NS <.001 <.001 <.001
89.3% 54.6% 83.3% 76.9%
79.5% 59.6% 83.3% 86.9%
<.001 NS NS <.001
93.1%
52.8%
<.001
NS, Not significant.
TABLE II. Comparison of symptoms in the populations Symptoms reported
Systemic fives Eczema Other rash Oropharyngeal angioedema Itchy throat Throat tightness Cough Shortness of breath Wheezing Vomiting Abdominal pain Hypotension Syncope
Self-selected caregiver report % (n [ 2003)
Clinic % (n [ 305)
P
49 23.1 10.9 57.4 40.8 38.1 34.4 28.4 26 41.1 21.7 8.1 7.5
31.5 5.6 7.9 28.7 9.2 6.3 11.6 11.6 10.2 33.7 5.6 0 5.3
<.001 <.001 NS <.001 <.001 <.001 <.001 <.001 <.001 .01 <.001 <.001 .17
NS, Not significant.
between November 2011 and August 2012. These 4 allergens represented the most common food allergens seen in the practice. Caregivers were included if their child had a visit in the clinic between 2009 and 2011 and a chart-verified diagnosis of food allergy, defined by a documented, convincing clinical history of an IgEmediated food-induced allergic reaction in the setting of confirmed IgE-mediated sensitization to the reported food (positive skin test and/or serum sIgE), or sIgE and/or a prick skin test wheal >95% positive predictive value for milk, egg, or peanut in a child with atopic dermatitis.10 A total of 1116 families with a unique visit in this 2-year time span were contacted over a year as part of a larger study to create an institutional database, with a total of 572 (51.3%) responding, of whom 92% identified an allergy to either milk, egg, peanut, or tree nut. The selection of those analyzed for chart review was blinded and at random. The data collection was halted at approximately the 300th survey because of financial constraints.23 Caregivers in both cohorts completed separate questionnaires containing identical sections regarding symptoms of the FAI most severe reaction to the allergen, reaction treatment, perception of reaction severity, follow-up care of their food allergy, allergic and
J ALLERGY CLIN IMMUNOL PRACT VOLUME -, NUMBER -
WARD AND GREENHAWT
3
FIGURE 1. Comparison of mean total and individual domain-specific Food Allergy Quality of Life Parental Burden (FAQL-PB) questionnaire scores between the self-selected (caregiver-reported) and clinic groups. The difference was statistically significant (P < .05) in all quality of life (QoL) domains. Error bars represent standard error.
nonallergic comorbidities, anaphylaxis management training, sociodemographic information, the institutionally developed and validated caregiver Food Allergy Self-Efficacy Questionnaire (FASEQ) (Cronbach a ¼ 0.88, intraclass correlation [ICC] ¼ 0.75, Supplemental Text, available in this article’s Online Repository at www.jaciinpractice.org/),26 and the FAQL-PB questionnaire assessing caregiver QoL. RC caregiver-reported questionnaire data were verified through the chart review of the patient’s medical record, to account for potential recall bias. National Institutes of Allergy and Infections Diseases/FAAN anaphylaxis criteria were applied to both the parentreported symptoms and those detailed through the chart review.27 The FAQL-PB (Figure E1, available in this article’s Online Repository at www.jaci-inpractice.org/) is a validated, 17-item, selfadministered questionnaire that measures the effect of pediatric food allergy on caregiver QoL, developed by Cohen et al.16 It has an ICC ¼ 0.93 and Cronbach a ¼ 0.95, indicating both excellent validity and reliability for the cross-sectional assessment of QoL. Caregiver QoL was not assessed at a uniform time from diagnosis during the formulation and validation of this scale. The 17 questions are each 7-point Likert items and the index is scored as a summated rating scale, with a higher FAQL-PB score indicating a worse QoL.16 Dimensionality of the index was not investigated during its validation. The clinical significance of changes in QoL measures is assessed through evaluating a minimal clinically important difference (MCID), the smallest difference in score that patients perceive as beneficial, and would mandate, in the absence of troublesome side effects and effective cost, a change in the patient’s management.28 For 7-point Likert scales analyzing QoL, prior studies have used an approximate measure of 0.5 as an MCID, based on the work by Jaeschke et al in cardiac and pulmonary disease.29 However, 0.5 is not an absolute number adaptable to any 7-point scale, and MCID is considered to be both specific to the particular QoL index and population being studied.30,31 Therefore, no true MCID is known for the FAQL-PB, and 0.5 is speculative.16 Index MCID can be calculated by the assessment of clinical impact through global rating questions, number needed to treat analysis, or through quantifying the standard error of measurement (SEM).32
Statistical analysis Data were analyzed using frequency analysis to report descriptive statistics, and c2/Fisher’s exact tests and Student’s t-test for bivariate
comparison. Adjusted multiple linear regression was performed to explore the relationships between a caregiver’s QoL score and the variables assessed in the questionnaire and our previous studies. Analysis was prespecified based on ongoing QoL studies by our research group.23,24 Multicollinearity of independent variables was assessed using the inspection of the variance inflation factor, and model specificity through the stata linktest (model specification link assessment) and ovtest (Ramsey reset test for omitted variables) commands. The stata margins command was used to assess posterior regression predictive probabilities. Factor analysis was performed to determine index dimensionality using the iterative principal components method with rotation. Scree plots were used to determine the number of relevant factors to retain. MCID for the index was calculated using the SEM method within the individual and total populations, by the following formula: SEM ¼ Standard deviation * square root [(1 reliability)].32 Data were analyzed with Stata SE, Version 13 (Stata Corp., College Station, Tex). Based on a level of significance of 0.05, the study had 80% power to detect a 0.5 difference in QoL score between groups with 65 patients per arm. This study was approved by the University of Michigan Medical School institutional review board.
RESULTS Differences in baseline demographics and reaction attributes A total of n ¼ 302 RC and n ¼ 2003 SS were included in the analysis. Table I details the comparative population demographics. Table II details the comparative presenting symptoms of the most severe allergic reaction. There were numerous significant differences between the groups, in particular in the type and frequency of reported symptoms, and significant predominance of reported peanut and tree nut allergy among the SS. RC had an earlier onset of allergy, but were slightly older at the time their caregiver was surveyed. Both populations were predominantly white and the caregivers college educated. The majority of SS reported that their child was under the care of an allergist. Differences in food allergy QoL Differences in RC and SS QoL were highly apparent. The mean total QoL score and all 17 individual domain-specific QoL scores were significantly different between groups, with higher
4
WARD AND GREENHAWT
J ALLERGY CLIN IMMUNOL PRACT MONTH 2015
FIGURE 2. Comparison of mean total Food Allergy Quality of Life Parental Burden (FAQL-PB) scores, stratified by specific food allergen, between the self-selected (caregiver-reported) and clinic groups. The difference was statistically significant (P < .05) in all quality of life (QoL) domains. Error bars represent standard error.
FIGURE 3. Comparison of mean total Food Allergy Quality of Life Parental Burden (FAQL-PB) scores between the self-selected (caregiverreported) and clinic groups, stratified by history/absence of reported anaphylaxis (A), epinephrine use (B), milk/egg allergy (C), and peanut/tree nut allergy (D). The difference was statistically significant (P < .05) in all quality of life (QoL) domains except where striped. Error bars represent standard error.
mean scores (worse QoL) in SS (mean total QoL score 2.81 [95% CI, 2.74-2.86] vs 1.84 [95% CI, 1.68-1.99], difference 0.96, P < .001). The average domain-specific difference was 0.99 between the groups (domain range 0.51-1.93, domain difference standard error ¼ 0.38) (Figure 1). When individually
explored by milk, egg, peanut, or tree nut allergy, mean QoL scores were significantly higher in SS for all respective allergens (Figure 2). When stratified by just caregivers of FAI with prior reported epinephrine use, mean total and 9/17 domain-specific QoL scores were significantly higher in SS than RC (2.98
WARD AND GREENHAWT
J ALLERGY CLIN IMMUNOL PRACT VOLUME -, NUMBER -
5
TABLE III. Adjusted linear regression models predicting the quality of life score Quality of life score
Coefficient
P
95% CI
Peanut or tree nut allergy
Milk or egg allergy
Referral clinic group 1.60 <.0001 1.91 to 1.29 1.69 (2.03 to 1.35) 0.91 (1.7 to 0.12) Reported anaphylaxis 0.27 .005 0.08 to 0.45 0.22 (0.02 to 0.41) NS Required epinephrine 0.59 .005 0.21 to 1.03 0.58 (0.15 to 1) NS Anaphylaxis/required epinephrine 0.54 .02 0.98 to 0.09 0.51 (0.98 to 0.05) NS interaction Self-efficacy (FASEQ score) 0.44 <.0001 0.35 to 0.54 0.5 (0.4 to 0.6) NS Peanut or tree nut allergy 0.48 <.0001 0.72 to 0.24 e e Other child with FA 0.24 .006 0.07 to 0.42 0.29 (0.1 to 0.47) NS Intercept 2.03 <.0001 1.5 to 2.6 1.37 (0.85 to 1.9) 2.6 (1 to 4.21) Adjusted for age of reaction, allergist diagnosis, income, age and age2, education, race, total number of food allergies, and geographic region. Allergen and geographic region were excluded in sensitivity analyses. FA, Food allergy; FASEQ, Food Allergy Self-Efficacy Questionnaire; NS, not significant.
[95% CI, 2.87-3.09] vs 2.23 [95% CI, 1.89-2.57], difference 0.75, P < .001). Similarly, when stratified by caregivers of FAI with prior reported anaphylaxis, mean total and 16/17 domainspecific QoL scores were significantly higher in SS than RC (mean total score 2.9 [95% CI, 2.81-2.98] vs 2.09 [95% CI, 1.82-2.35], difference 0.81, P < .001) (Figure 3, A-D). Supplemental Table E1 (available in this article’s Online Repository at www.jaci-inpractice.org) details the item specific mean differences in score by population for these aforementioned trends. A multivariable linear regression model was created to better understand the association of participant attributes with the QoL score (Table III). In the adjusted analysis, RC were significantly associated with a lower QoL score (better QoL), confirming what was observed at the bivariate level. Additionally, having a peanut or tree nut allergic child (compared with milk or egg) was also significantly associated with a lower QoL score. Covariates significantly associated with a higher (worse) QoL score included anaphylaxis, epinephrine use, increasing (worsening) FASEQ score, and having multiple FAI. An interaction was noted between reported epinephrine use in individuals with a reported history of anaphylaxis—a lower QoL score was noted in those with both reported anaphylaxis and epinephrine use, but independently (eg, if only one of these attributes was reported but not the other) these were both associated with a higher (worse) QoL score. A separate sensitivity analysis of this regression model restricted to only those with peanut or tree nut allergic FAI, and just milk or egg allergic FAI both showed persistence of the significant association between RC and lower QoL score (Table III).
Psychometric performance of the FAQL-PB within clinical and self-report populations The psychometric performance of the FAQL-PB was assessed within each individual population and the total population. Cronbach a was nearly identical within each population (total group 0.95, SS 0.94, and RC 0.96), which indicated consistent reliability. An index-specific MCID was calculated using the SEM method, which was 0.31 in the RC population, 0.34 in the SS, and 0.32 in the combined population. The FAQL-PB lacks factor analysis and elucidation of its dimensionality within US populations, though this was recently described in a UK population.33 Thus, we performed a factor analysis within each individual population, and for the total combined population to assess underlying dimensionality and item-theme correlations. Using the iterative principal factor
method, 2 factors each were identified in all 3 groupings (SS, RC, and the combined group). Orthogonal and oblique rotations were performed, with the oblique rotation retained to maximize correlations between factors. This identified 12 items in the combined population, 13 items in RC, and 14 items in SS that loaded onto factor 1. Three items consistently loaded onto factor 2 in all 3 populations: vacation planning, restaurant choice in dining out, and limitations on social activities. Table IV details the factor loadings for the groups. Factor 1 appears to have a theme of general health concerns and anxiety related to the FAI. Factor 2 strongly identifies a theme of social interaction with others.
DISCUSSION In this study, we aimed to explore if any differences in QoL were detectable based on being selected from an academic center cohort or through a food allergy advocacy group, and to explore specific performance of the index within these different groups to further validate it as a robust outcome measure. Herein we show that caregiver food allergy QoL is heterogeneous, and dependent on multiple factors, including a large effect related to the “population” through which we surveyed a caregiver. This study offers some comparison of these 2 population segments and shows a range of possible reported symptoms, management, and outcomes of how a similar disease is managed. Limited previous data exist that compare and describe such differences, though more recently the validity of self-reported populations has been challenged, given evidence that there may be overestimation and poor validation, despite the fact that the majority of the data describing food allergy in the United States are from selfreport.1,2,4,5,7,10,24,33-35 What we demonstrate is that among individuals living a “food allergic lifestyle,” some have a much more prescribed deficit in terms of poor food allergy QoL, though the reasons for this are less clear, and were beyond the scope of this present study but may be related to many possible issues including caregiver motivation, individualized effect of diagnosis, or relationship with their provider, for example. Follow-up work to better understand these patterns and influence on the observed segmentation is underway. Our data highlight that a more individualized approach to understanding the daily lives of our food allergic families may be needed, given evidence that different populations have different problems, needs, and risk factors contributing to their poor QoL. Such differences may significantly impact future treatment options or utilization of care if these are not better understood.
6
WARD AND GREENHAWT
J ALLERGY CLIN IMMUNOL PRACT MONTH 2015
TABLE IV. Factor analysis of the Food Allergy Quality of Life Parental Burden in the 3 populations Oblique rotation loadings Domain
Combined population Troubled with the thought your child will have a reaction Troubled by possibility concerns for your child’s health Troubled that child may not overcome their food allergy Troubled you would be unable to help treat a reaction Troubled by the sadness of the burden your child carries Troubled by possibility of leaving child in care of others Troubled by anxiety Troubled your child will not have a normal upbringing Troubled by fear of child being near others while eating Troubled by your child attending group activities Troubled that others lack appreciation of severity Troubled by need to take precaution before leaving home Troubled by concerns for your child’s nutrition Limitation in choice of vacation plans Limitation in choice of restaurant Limitation in participation in social activities Troubled by need to spend extra time preparing meals Factor correlation Clinic population Troubled with the thought your child will have a reaction Troubled that others lack appreciation of severity Troubled that child may not overcome their food allergy Troubled by your child attending group activities Troubled by possibility of leaving child in care of others Troubled by the sadness of the burden your child carries Troubled by possibility concerns for your child’s health Troubled you would be unable to help treat a reaction Troubled by fear of child being near others while eating Troubled by anxiety Troubled your child will not have a normal upbringing Troubled by concerns for your child’s nutrition Limitation in choice of restaurant Limitation in choice of vacation plans Limitation in participation in social activities Troubled by need to spend extra time preparing meals Troubled by need to take precaution before leaving home Factor correlation Self-selected caregiver-report population Troubled with the thought your child will have a reaction Troubled by possibility concerns for your child’s health Troubled that child may not overcome their food allergy Troubled you would be unable to help treat a reaction Troubled by the sadness of the burden your child carries Troubled by anxiety Troubled your child will not have a normal upbringing Troubled by possibility of leaving child in care of others Troubled by fear of child being near others while eating Troubled by your child attending group activities Troubled that others lack appreciation of severity Troubled by need to take precaution before leaving home Troubled by concerns for your child’s nutrition Troubled by need to spend extra time preparing meals
Factor 1
Factor 2
Uniqueness
0.926 0.859 0.791 0.829 0.768 0.741 0.715 0.706 0.708 0.699 0.679 0.406 0.471 0.0875 0.0799 0.1054 0.3689 0.6375
0.1063 0.034 0.0023 0.1332 0.0447 0.089 0.1376 0.1362 0.108 0.1141 0.0765 0.3938 0.1193 0.884 0.881 0.713 0.382
0.256 0.298 0.372 0.435 0.364 0.359 0.344 0.36 0.389 0.396 0.467 0.476 0.692 0.308 0.306 0.384 0.538
0.9447 0.8108 0.8054 0.79 0.7952 0.795 0.777 0.7457 0.7055 0.6565 0.5046 0.3503 0.0645 0.0375 0.0838 0.1871 0.3277 0.6832
0.086 0.0056 0.0678 0.0864 0.1082 0.05 0.1039 0.0471 0.1369 0.2149 0.3449 0.2649 0.9122 0.8693 0.769 0.627 0.5326
0.211 0.349 0.272 0.275 0.238 0.311 0.275 0.49 0.352 0.33 0.389 0.68 0.244 0.287 0.314 0.411 0.371
0.9181 0.8599 0.7798 0.8322 0.7798 0.721 0.7265 0.7324 0.7095 0.6888 0.6655 0.4183 0.4756 0.3872
0.112 0.0486 0.0058 0.1425 0.0058 0.1235 0.1056 0.08 0.0912 0.1078 0.0674 0.3664 0.1117 0.3527
0.27 0.309 0.386 0.433 0.386 0.355 0.366 0.385 0.408 0.422 0.497 0.502 0.696 0.557 (continued)
WARD AND GREENHAWT
J ALLERGY CLIN IMMUNOL PRACT VOLUME -, NUMBER -
7
TABLE IV. (Continued) Oblique rotation loadings Domain
Factor 1
Factor 2
Limitation in choice of vacation plans Limitation in choice of restaurant Limitation in participation in social activities Factor correlation
0.098 0.091 0.102 0.6157
0.8724 0.861 0.696
Uniqueness
0.335 0.3347 0.417
Bolded values show which items load to which of the 2 factors.
In showing that food allergy QoL is significantly different between RC and SS, we suggest that these may be distinct, separate subgroups within the food allergy population likely with differing needs and risk profiles. Caregiver QoL was highly differentiable based on the specific population, with scores nearly a full 1-point worse at a bivariate level and 1.6-point worse in a multiple regression model between RC and SS. These differences exceed both the calculated MCID of 0.32 and the estimated MCID of 0.5 by several fold. We further demonstrate QoL differences between populations based on type of food allergen, prior anaphylaxis, prior use of epinephrine, and presence of multiple FAI in the family. Each of these findings emphasizes how the same illness is not perceived and experienced in the same manner across a diverse population—an important and novel realization that suggests the burden of the same disease is disproportionate. More research is needed to understand why the burden is experienced disproportionately, and pinpoint more specific details that moderate these differences, to enhance understanding of why these exist and what their long-term implications may be. For example, there is need to further elucidate the significance of, and reasons why peanut and/or tree nut allergy is associated with a better adjusted QoL than milk and/or egg allergy—certainly a tailored approach by allergen, within a population segment, may be a tool to lead to better outcomes. We also show data supporting a relationship between selfefficacy and QoL score (increasing [worsening] self-efficacy score is associated with increasing [worsening] QoL score), which seems somewhat intuitive. Self-efficacy is an intriguing predictor of QoL, and may serve as a point of future intervention to help improve QoL, or an independent stand-alone outcome to follow. This is an emerging area of study, and a manuscript more formally describing the FASEQ index and the relationship between self-efficacy and QoL is forthcoming by our group. We also present novel data regarding the psychometric properties of the FAQL-PB in US populations. The FAQL-PB has good performance with high reliability and minimally differing dimensionality in both RC and SS.16,33 This is the first assessment of the dimensionality of this index in a US population. Knibb and Stalker recently validated and performed the factor analysis of the FAQL-PB in a British population, but noted different dimensionality (3 dimensions) than we present.33 This should not be surprising given well-established data that QoL is culturally specific.12,31,36,37 Most importantly, we provide an index-specific (and food allergy population specific) estimate of clinical significance, which is novel. This index-specific MCID can now be used to further enhance the clinical applications of the FAQL-PB as a cross-sectional patient-reported outcome measure. There are several limitations of this study. Foremost this utilized prospectively collected QoL data, whereas all associated covariates used for comparison and adjustment of findings were
based on retrospective data. All such attributes of food allergy were recorded entirely by caregiver report, and the data are subject to both recall bias and possible challenges to validity. This information was verified by chart review in the clinic sample. Participants from the self-selected caregiver-reported cohort were recruited through social media tied to national advocacy groups, with more than 50% of caregivers reporting membership in at least one such group. This may infer some degree of selection bias or bias resulting from another unmeasured attribute tied to the propensity for advocacy group membership, though group membership was adjusted for and was not significant in the regression analysis. Earlier US QoL studies have dealt with similar potential limitations.16 However, there are presently few other feasible and practical methods besides caregiver report to study food allergy at a national level. We were not able to ascertain the degree to which the clinic cohort utilizes similar online resources or participates in such groups. In conclusion, we show that caregiver food allergy QoL is heterogeneous, and dependent on multiple factors, including the segment of the “population” to which one belongs. These data provide evidence of heterogeneity of disease, inferring that distinct segmentation within the general food allergic population may exist, likely with differing needs, risk profiles, and patientreported outcomes. Identifying that certain self-selected caregiver-reported groups may experience a disproportionate burden of the same illness (with different patient-reported outcomes) is an important finding, which will allow further study to develop more tailored ways to help these individuals. We also further validate the FAQL-PB as an instrument for tracking crosssectional caregiver QoL by deriving an MCID to denote clinical change, and the underlying dimensional structure of the index. These properties should enhance the clinical applicability of the FAQL-PB. REFERENCES 1. Branum AM, Lukacs SL. Food allergy among U.S. children: trends in prevalence and hospitalizations. NCHS Data Brief 2008;(10):1-8. 2. Branum AM, Lukacs SL. Food allergy among children in the United States. Pediatrics 2009;124:1549-55. 3. Gupta RS, Springston EE, Warrier MR, Smith B, Kumar R, Pongracic J, et al. The prevalence, severity, and distribution of childhood food allergy in the United States. Pediatrics 2011;128:e9-17. 4. Osborne NJ, Koplin JJ, Martin PE, Gurrin LC, Lowe AJ, Matheson MC, et al. Prevalence of challenge-proven IgE-mediated food allergy using populationbased sampling and predetermined challenge criteria in infants. J Allergy Clin Immunol 2011;127:668-76. e1-2. 5. Sicherer SH, Munoz-Furlong A, Sampson HA. Prevalence of seafood allergy in the United States determined by a random telephone survey. J Allergy Clin Immunol 2004;114:159-65. 6. Sicherer SH, Munoz-Furlong A, Sampson HA. Prevalence of peanut and tree nut allergy in the United States determined by means of a random digit dial telephone survey: a 5-year follow-up study. J Allergy Clin Immunol 2003;112: 1203-7.
8
WARD AND GREENHAWT
7. Sicherer SH, Munoz-Furlong A, Godbold JH, Sampson HA. US prevalence of self-reported peanut, tree nut, and sesame allergy: 11-year follow-up. J Allergy Clin Immunol 2010;125:1322-6. 8. Liu AH, Jaramillo R, Sicherer SH, Wood RA, Bock SA, Burks AW, et al. National prevalence and risk factors for food allergy and relationship to asthma: results from the National Health and Nutrition Examination Survey 2005-2006. J Allergy Clin Immunol 2010;126:798-806.e13. 9. Greenhawt M, Weiss C, Conte ML, Doucet M, Engler A, Camargo CA Jr. Racial and ethnic disparity in food allergy in the United States: a systematic review. J Allergy Clin Immunol Pract 2013;1:378-86. 10. Boyce JA, Assa’ad A, Burks AW, Jones SM, Sampson HA, Wood RA, et al. Guidelines for the diagnosis and management of food allergy in the United States: report of the NIAID-sponsored expert panel. J Allergy Clin Immunol 2010;126:S1-58. 11. van der Velde JL, Dubois AE, Flokstra-de Blok BM. Food allergy and quality of life: what have we learned? Curr Allergy Asthma Rep 2013;13:651-61. 12. Salvilla SA, Dubois AE, Flokstra-de Blok BM, Panesar SS, Worth A, Patel S, et al. Disease-specific health-related quality of life instruments for IgE-mediated food allergy. Allergy 2014;69:834-44. 13. Primeau MN, Kagan R, Joseph L, Lim H, Dufresne C, Duffy C, et al. The psychological burden of peanut allergy as perceived by adults with peanut allergy and the parents of peanut-allergic children. Clin Exp Allergy 2000;30: 1135-43. 14. Avery NJ, King RM, Knight S, Hourihane JO. Assessment of quality of life in children with peanut allergy. Pediatr Allergy Immunol 2003;14:378-82. 15. Akeson N, Worth A, Sheikh A. The psychosocial impact of anaphylaxis on young people and their parents. Clin Exp Allergy 2007;37:1213-20. 16. Cohen BL, Noone S, Munoz-Furlong A, Sicherer SH. Development of a questionnaire to measure quality of life in families with a child with food allergy. J Allergy Clin Immunol 2004;114:1159-63. 17. Cummings AJ, Knibb RC, King RM, Lucas JS. The psychosocial impact of food allergy and food hypersensitivity in children, adolescents and their families: a review. Allergy 2010;65:933-45. 18. DunnGalvin A, de BlokFlokstra BM, Burks AW, Dubois AE, Hourihane JO. Food allergy QoL questionnaire for children aged 0-12 years: content, construct, and cross-cultural validity. Clin Exp Allergy 2008;38:977-86. 19. Flokstra-de Blok BM, DunnGalvin A, Vlieg-Boerstra BJ, Oude Elberink JN, Duiverman EJ, Hourihane JO, et al. Development and validation of the selfadministered Food Allergy Quality of Life Questionnaire for adolescents. J Allergy Clin Immunol 2008;122:139-44. 144 e1-2. 20. Flokstra-de Blok BM, van der Velde JL, Vlieg-Boerstra BJ, Oude Elberink JN, DunnGalvin A, Hourihane JO, et al. Health-related quality of life of food allergic patients measured with generic and disease-specific questionnaires. Allergy 2010;65:1031-8. 21. Springston EE, Smith B, Shulruff J, Pongracic J, Holl J, Gupta RS. Variations in quality of life among caregivers of food allergic children. Ann Allergy Asthma Immunol 2010;105:287-294.e3.
J ALLERGY CLIN IMMUNOL PRACT MONTH 2015
22. van der Velde JL, Flokstra-de Blok BM, de Groot H, Oude-Elberink JN, Kerkhof M, Duiverman EJ, et al. Food allergy-related quality of life after double-blind, placebo-controlled food challenges in adults, adolescents, and children. J Allergy Clin Immunol 2012;130:1136-1143 e2. 23. Howe L, Franxman T, Teich E, Greenhawt M. What affects quality of life among caregivers of food allergic children? Ann Allergy Asthma Immunol 2014;113:69-74. 24. Franxman T, Howe L, Teich E, Greenhawt M. Oral food challenge and food allergy quality of life in caregivers of food allergic children. J Allergy Clin Immunol Pract 2015;3:50e56. 25. Resnick ES, Pieretti MM, Maloney J, Noone S, Munoz-Furlong A, Sicherer SH. Development of a questionnaire to measure quality of life in adolescents with food allergy: the FAQL-teen. Ann Allergy Asthma Immunol 2010;105:364-8. 26. Baptist AP, Dever SI, Greenhawt MJ, Polmear-Swendris N, McMorris MS, Clark NM. A self-regulation intervention can improve quality of life for families with food allergy. J Allergy Clin Immunol 2012;130:263-265.e6. 27. Sampson HA, Munoz-Furlong A, Campbell RL, Adkinson NF Jr, Bock SA, Branum A, et al. Second symposium on the definition and management of anaphylaxis: summary report—Second National Institute of Allergy and Infectious Disease/Food Allergy and Anaphylaxis Network symposium. J Allergy Clin Immunol 2006;117:391-7. 28. Patrick DL, Bergner M. Measurement of health status in the 1990s. Annu Rev Public Health 1990;11:165-83. 29. Jaeschke R, Singer J, Guyatt GH. Measurement of health status. Ascertaining the minimal clinically important difference. Control Clin Trials 1989;10:407-15. 30. Wright A, Hannon J, Hegedus EJ, Kavchak AE. Clinimetrics corner: a closer look at the minimal clinically important difference (MCID). J Man Manip Ther 2012;20:160-6. 31. Cook CE. Clinimetrics corner: the minimal clinically important change score (MCID): a necessary pretense. J Man Manip Ther 2008;16:E82-3. 32. Guyatt GH, Osoba D, Wu AW, Wyrwich KW, Norman GR. Methods to explain the clinical significance of health status measures. Mayo Clin Proc 2002;77: 371-83. 33. Knibb RC, Stalker C. Validation of the Food Allergy Quality of Life-Parental Burden Questionnaire in the UK. Qual Life Res 2013;22:1841-9. 34. Chafen JJ, Newberry SJ, Riedl MA, Bravata DM, Maglione M, Suttorp MJ, et al. Diagnosing and managing common food allergies: a systematic review. JAMA 2010;303:1848-56. 35. Jackson KD, Howie LD, Akinbami LJ. Trends in allergic conditions among children: United States, 1997e2011. NCHS Data Brief 2013;(121):1-8. 36. Muraro A, Dubois AE, Dunngalvin A, Hourihane JO, de Jong NW, Meyer R, et al. EAACI Food Allergy and Anaphylaxis Guidelines. Food allergy healthrelated quality of life measures. Allergy 2014;69:845-53. 37. Goossens NJ, Flokstra-de Blok BM, van der Meulen GN, Arnlind MH, Asero R, Barreales L, et al. Health-related quality of life in food-allergic adults from eight European countries. Ann Allergy Asthma Immunol 2014;113:63-68.e1.
J ALLERGY CLIN IMMUNOL PRACT VOLUME -, NUMBER -
Supplemental Text: Background Detail of the FASEQ The Food Allergy Self-efficacy Questionnaire (FASEQ) is an 8 Likert-item index, designed to assess a caregiver’s confidence in managing various aspects of raising a food allergic child. This was inspired by a similar self-efficacy measure used in asthma, with wording of themes changed to reflect terminology for food allergy. This was assessed for face validity internally within the University of Michigan Division of Allergy and Clinical Immunology, in conjunction with a team from the University of Michigan School of Public Health’s Center for Managing Chronic Disease that has expertise in self-regulatory behavior
WARD AND GREENHAWT
8.e1
and self-efficacy. The index was then piloted as part of a secondary outcome measure in a small randomized controlled trial of a self-regulation intervention for anaphylaxis measurement.26 Items in the FASEQ are scored similarly to the Food Allergy Quality of Life Parental Burden (FAQL-PB), as a summated rating scale with lower score reflective of better self-efficacy, though a 5-point scale as opposed to a 7-point scale (FAQLPB) is used. The index has undergone formal validation (as described in the Methods section), as well as additional psychometric validation. A separate publication is forthcoming that details the specific performance of the FASEQ, including its validation.
8.e2
WARD AND GREENHAWT
J ALLERGY CLIN IMMUNOL PRACT MONTH 2015
FIGURE E1. The FAQL-PB quality of life index.
WARD AND GREENHAWT
J ALLERGY CLIN IMMUNOL PRACT VOLUME -, NUMBER -
TABLE E1. Mean differences in QoL score by individual analysis between the clinic and self-selected caregiver-reported populations Domain
Total Vacation plans Restaurant choice Social activities Meal preparation Public travel Anxiety Not overcome Other caregiver Others appreciate severity Child’s burden Group activity Child’s health Unable to help Normal upbringing Nutrition Near others Risk of reaction
Anaphylaxis
Epinephrine Peanut/tree Milk/egg use nut allergy allergy Overall
0.81 1.45 1.49 1.20 0.74 0.76 0.53 0.16 0.44 0.62
0.75 1.48 1.52 1.53 0.86 0.83 0.48 0.11 0.29 0.44
1.22 1.82 2.11 1.46 0.98 1.08 1.10 0.86 0.98 1.26
1.08 1.27 1.69 1.18 1.09 1.25 0.88 0.53 0.66 0.81
0.96 1.37 1.66 1.21 0.62 1.93 0.83 0.58 0.76 1.03
1.11 0.51 0.60 0.37 1.23
1.20 0.31 0.47 0.34 1.19
1.55 1.05 1.12 0.78 1.34
1.21 0.92 0.74 0.40 1.22
1.24 0.83 0.79 0.61 1.11
0.39 0.91 0.46
0.33 0.67 0.54
0.88 1.23 1.04
1.34 0.96 0.60
0.51 1.05 0.83
Bold values represent those not exceeding the proposed minimal clinical important difference. Italicized values represent differences that were not statistically significant.
8.e3