Video Abstract

Video Abstract

Close modal
OBJECTIVE

We sought to evaluate the use of behavioral economics approaches to promote the carrying of epinephrine auto-injectors (EAIs) among adolescents with food allergies. We hypothesized that adolescents who receive frequent text message nudges (Intervention 1) or frequent text message nudges plus modest financial incentives (Intervention 2) would be more likely to carry their epinephrine than members of the usual care control group.

METHODS

We recruited 131 adolescents ages 15 to 19 with a food allergy and a current prescription for epinephrine to participate in a cohort multiple randomized controlled trial. Participants were randomly assigned to participate in Intervention 1, Intervention 2, or to receive usual care. The primary outcome was consistency of epinephrine-carrying, measured as the proportion of checkpoints at which a participant could successfully demonstrate they were carrying their EAI, with photo-documentation of the device.

RESULTS

During Intervention 1, participants who received the intervention carried their EAI 28% of the time versus 38% for control group participants (P = .06). During Intervention 2, participations who received the intervention carried their EAI 45% of the time versus 23% for control group participants (P = .002).

CONCLUSIONS

Text message nudges alone were unsuccessful at promoting EAI-carrying but text message nudges combined with modest financial incentives almost doubled EAI-carriage rates among those who received the intervention compared with the control group. However, even with the intervention, adolescents with food allergies carried their EAI <50% of the time. Alternative strategies for making EAIs accessible to adolescents at all times should be implemented.

What’s Known on This Subject:

Adolescence is the period of highest risk of death from anaphylaxis for children with food allergies. Yet, there is a lack of adherence to preventive strategies, including the carrying of potentially life-saving epinephrine auto-injectors (EAIs), and few evidence-based interventions exist.

What This Study Adds:

Although text message nudges alone did not promote EAI-carrying, text message nudges combined with modest financial incentives almost doubled EAI carriage rates compared with usual care. However, at any given point in time, only 54% of adolescents carried their EAI.

In the United States, an estimated 4% to 8% of children have immunoglobulin E-mediated food allergies.13  Allergic reactions can be mild or severe, including anaphylaxis, a potentially life-threatening condition that is treatable with the prompt administration of epinephrine.4  Inpatient hospital stays and emergency department visits make up ∼60% of the total direct medical costs of food allergies in children.5  No universal cure exists for food allergies, leaving those affected by them to practice careful self-management, which includes preventive measures (ie, avoidance of potentially allergenic foods) and emergency response (ie, administration of epinephrine during anaphylaxis).6 

Adolescence is the period with the highest risk of death from anaphylaxis.79  This higher risk results in part from a lack of adherence to prevention strategies, such as the consistent carrying of epinephrine auto-injectors (EAIs), because of increased independence from parents and greater risk-taking behaviors.1012  One study found that, among college students with food allergies with a history of anaphylactic symptoms, only approximately one-third reported ever carrying an EAI; only one-half of them carried it with them at all times.13  Multiple studies have identified barriers and facilitators to carrying and using epinephrine, including characteristics of EAI devices (eg, bulkiness, ease of use) and the social environment (eg, ability to carry epinephrine in a backpack at school but difficulty when socializing outside of school).12,14,15  Yet few interventions exist to improve epinephrine carriage rates; most interventions that have been implemented are purely educational.16,17 

The use of behavioral economics approaches to promote disease management offers promising results across multiple chronic conditions.1821  These same concepts can be applied to promote safe food allergy management practices.22  Behavioral economics acknowledges that individuals typically have bounded rationality, meaning that they often do not behave perfectly rationally but do so in predictable ways on the basis of mental shortcuts or heuristics.23  Based on a pilot intervention with promising results for nudge approaches to improve epinephrine-carrying among young adults,22  we experimentally evaluated 2 behavioral economics interventions to promote epinephrine-carrying among adolescents. Using a cohort multiple randomized controlled trial design,24  we hypothesized that:

  1. Adolescents with food allergies who receive frequent text message nudges (Intervention 1) would be more likely to carry their epinephrine than members of the usual care control group.

  2. Adolescents with food allergies who receive frequent text message nudges plus financial incentives (Intervention 2) would be more likely to carry their epinephrine than members of the usual care control group.

Informed by our pilot study that employed text message nudges and modest financial incentives,22  we developed 2 sequential interventions on the basis of 5 principles of behavioral economics to promote consistent epinephrine-carrying among adolescents with food allergies.

Principle 1: Availability Heuristic

The availability heuristic refers to the ease with which examples about a certain topic come to mind.25  It offers a way to overcome patients’ limited bandwidth to attend to healthy behaviors, such as consistent epinephrine-carrying, by nudging patients. In this study, nudges were text message prompts to increase the salience of food allergy management among adolescents. Text message-based interventions have been successfully employed among youth, including to manage diabetes and asthma.26  Working with a convenience sample of adolescents and young adults with food allergies who were not part of this study, we piloted and cognitively tested a bank of text message nudges in cognitive interviews.27  Messages provided information, facts, or tips about food allergies or requested feedback from participants (eg, “What is the most important advice you would give a friend with food allergies?”).

Principle 2: Feedback

Providing feedback about a certain behavior allows individuals to learn.25  We provided automated feedback to participants when they succeeded or failed to document they were carrying their EAIs.

Principle 3: Incentives

Incentives are a cornerstone of behavioral economics.25  In our pilot study, we found that modest financial incentives (approximately $100) were a successful motivator for carrying epinephrine among young adults.22 

Principle 4: Loss Aversion

Loss aversion refers to individuals’ tendency to be more effectively motivated by the threat of losing something of value than by the gain of something of equivalent value.28 

Principle 5: Present Bias

Present bias refers to individuals’ tendency to focus on the present rather than the future.28,29  In the case of food allergy management, the benefits of consistently carrying epinephrine may not be immediately apparent given (hopefully) rare use in emergency situations.

Intervention 1 tested the use of nudges (Principle 1) and feedback (Principle 2) to promote epinephrine-carrying by sending intervention group participants intermittent, randomly timed (∼30 over a 10-week period, primarily outside of regular school hours) text messages. Intervention 2 used all 5 principles, including the nudges and feedback components from Intervention 1 with the addition of financial incentives (Principle 3), which were designed by leveraging the loss aversion principle (Principle 4) and with information about the incentive for carrying epinephrine provided immediately (Principle 5). At the beginning of the second intervention period, intervention group participants were notified that $100 was placed in a virtual account for them and that they would lose $10 every time they failed to successfully demonstrate they were carrying their EAI at each of the 10 random checkpoints.

This study used a cohort multiple randomized controlled trial design to test the 2 interventions to promote epinephrine-carrying and characterize food allergy management behaviors.24  The base cohort was followed for a period of 2 years during which they were asked to fill out 3 surveys (at baseline, midway, and at the end of the study period) about their food allergy knowledge, attitudes, and behaviors. During recurring text-based assessments to monitor adverse events, participants were also asked about the incidence of food allergic reactions. At enrollment, participants self-reported their age, sex, and race/ethnicity to assess sample representativeness and generalizability of findings. All participants were part of the base cohort. Within the base cohort, participants were randomly assigned to participate in Intervention 1 (text message nudges), Intervention 2 (text message nudges and financial incentives), or to receive usual care (Fig 1). During Intervention 1, the control group consisted of participants who were (1) randomly assigned to receive usual care and (2) randomly assigned to later participate in Intervention 2. During Intervention 2, the control group consisted of only participants randomly assigned to receive usual care throughout the entire study period. Participants randomly assigned to participate in Intervention 1 were followed during Intervention 2 to assess the long-term effects of the intervention on epinephrine-carrying but were excluded from the primary analysis for Intervention 2 to avoid confounding the comparison with any potential carryover effects of Intervention 1.

FIGURE 1

Study intervention flow.

FIGURE 1

Study intervention flow.

Close modal

We worked with the Children’s Hospital of Philadelphia (CHOP) Recruitment Enhancement Core to send out emails or letters to >2500 patients on the basis of the following search criteria: (1) ages 15 to 19, (2) epinephrine prescription, (3) seen by an allergy specialist at any CHOP location within the last 36 months, and (4) had 1 of the following International Classification of Diseases codes for food allergy (T78.0–T78.2, T78.40, T78.49, Z91.018). We also recruited patients by phone, by posting physical and online flyers, and by publishing the trial on Penn Medicine’s trial portal. Recruitment began in June 2018 and ended in February 2019.

To participate in the study, participants had to meet the following inclusion criteria: (1) have a food allergy, (2) have a current prescription for an EAI, (3) be between 15 and 19 years of age, (4) be fluent in English, and (5) have a cellphone that could send and receive photos. The study coordinator recruited 131 participants for the base cohort. Participants were then randomly assigned by the study biostatistician to receive Intervention 1 (n = 26), Intervention 2 (n = 54), or usual care (n = 51; Supplemental Fig 1). Given the relatively small sample size, we used permutated block randomization and stratified by sex to ensure a balance between groups.30  With 25 participants per group, we estimated 80% power to detect a 23% difference in the proportion of successful checkpoints (our primary outcome) between each intervention group and the control group, a smaller difference than we found in our pilot study.22 

We obtained informed consent from all participants aged 18 and older and parental consent and participant assent for those <18 years of age. Consent was obtained during a phone call with the study coordinator, and participants were asked to sign the consent form electronically. Participants received $20 for completing each of the 3 surveys and $20 for participating in a debrief interview after the study was over.

The administration of the interventions was fully automated by using the Way to Health platform.31  After participants enrolled in the study, they registered their cellphone numbers with the platform to receive all study surveys and interventions by text message.

The primary endpoint was consistency of epinephrine-carrying, measured as the proportion of checkpoints at which a participant could successfully demonstrate they were carrying their epinephrine. Checkpoints occurred for all participants (intervention and control) once per week at a random time during both intervention periods and included a text message with a request to share a photograph of their EAI next to a code word of the day to ensure they were taking the photo in real time. For example:

Are you carrying your epinephrine now? If not, text “no” in response. If yes, text a photo of your epinephrine auto-injector next to the written code word “computer”.

Participants were classified as having successfully demonstrated they were carrying their EAI if they replied to the checkpoint message with a photo of their EAI and the code word of the day within 30 minutes of receiving the message (Fig 2). This assessment method was successfully employed in the pilot study.22 

FIGURE 2

Example of a successful checkpoint photograph.

FIGURE 2

Example of a successful checkpoint photograph.

Close modal

We initially planned to measure the primary outcome as the proportion of 10 checkpoints during which participants were carrying their EAIs. Early technical glitches with the study platform during Intervention 1 led to a failure to reach multiple study participants; 12 checkpoint attempts were made from which 9 yielded usable data. Thus, for Intervention 1, the primary outcome was the proportion of 9 checkpoints during which participants carried their EAIs. During Intervention 2, all 10 checkpoints worked as planned.

We also collected process data, using a mix of open- and close-ended questions, during the midway and exit surveys to evaluate whether participants found the text message nudges useful and informative, assess the frequency at which text message nudges were sent, and understand how motivating the financial incentives were for participants who received them.

All data were analyzed in RStudio version 1.3.959. To assess the effectiveness of the interventions as measured by the primary endpoint and test hypotheses 1 and 2, we used a 2-sided Wilcoxon rank test (at a 0.05 significance level) to determine between-group differences in the proportion of successful checkpoints for each intervention.32 

Analyses were conducted by using complete case analysis with missing data treated as a failure to document epinephrine-carrying, following the intention-to-treat principle.33  We conducted a sensitivity analysis that counted successful responses to checkpoints even if they were late (ie, outside the 30-minute window, per our predefined criteria).

This study was approved by the Institutional Review Boards of the University of Pennsylvania and the CHOP.

Cohort participants identified primarily as non-Hispanic white, with a mean age of 16 years. The most common food allergies were to tree nuts and peanuts, and more than two-thirds of participants were allergic to >1 food. All participants self-reported consistently carrying their EAIs; only one-half had ever been treated with an injection of epinephrine. Participants’ characteristics were similar across all 3 study groups (Table 1).

TABLE 1

Participant Demographic Characteristics, Overall and by Intervention Group

CharacteristicCohort (n = 131)Intervention 1Intervention 2
Intervention Group
1: Text Messages
Only (n = 26)
Control Group 1 (n = 105)Intervention Group
2: Text Messages and Financial Incentives (n = 54)
Control Group 2 (n = 51)
Sex, n (%)a      
 Female 65 (50) 13 (50) 52 (50) 27 (50) 25 (49) 
 Male 65 (50) 13 (50) 52 (50) 27 (50) 25 (49) 
 Decline to answer 1 (1) 0 (0) 1 (1) 0 (0) 1 (2) 
Age, mean (SD) 16 (1) 16 (2) 17 (1) 17 (1) 16 (1) 
Race, n (%)a      
 American Indian or Alaskan Native 1 (1) 0 (0) 1 (1) 0 (0) 1 (2) 
 Asian 5 (4) 2 (8) 3 (3) 2 (4) 1 (2) 
 Black 15 (12) 5 (19) 10 (10) 4 (7) 6 (12) 
 White 104 (79) 19 (73) 85 (81) 45 (83) 40 (78) 
 Other 5 (4) 0 (0) 5 (5) 3 (6) 2 (4) 
 Decline 1 (1) 0 (0) 1 (1) 0 (0) 1 (2) 
Ethnicity, n (%)      
 Hispanic 3 (2) 0 (0) 3 (3) 1 (2) 2 (4) 
 Decline to answer 1 (1) 0 (0) 1 (1) 0 (0) 1 (2) 
Grade, n (%)a      
 9 9 (7) 0 (0) 9 (9) 3 (6) 6 (12) 
 10 36 (28) 11 (42) 25 (24) 10 (19) 15 (30) 
 11 30 (23) 4 (15) 26 (25) 14 (26) 12 (24) 
 12 22 (17) 3 (12) 19 (18) 10 (19) 9 (18) 
 College Freshman 28 (21) 6 (23) 22 (21) 14 (26) 8 (16) 
 College Sophomore 6 (5) 2 (8) 4 (4) 3 (6) 1 (2) 
Allergic to, n (%)b      
 Peanut 79 (70) 15 (71) 64 (70) 30 (65) 34 (74) 
 Tree nuts 84 (74) 14 (67) 70 (76) 39 (85) 31 (67) 
 Soy 8 (7) 1 (5) 7 (8) 4 (9) 3 (7) 
 Wheat 3 (3) 1 (5) 2 (2) 0 (0) 2 (4) 
 Milk 18 (16) 5 (24) 13 (14) 6 (13) 7 (15) 
 Egg 18 (16) 3 (14) 15 (16) 8 (17) 7 (15) 
 Fish 11 (10) 2 (10) 9 (10) 6 (13) 3 (7) 
 Shellfish 20 (18) 6 (29) 14 (15) 6 (13) 8 (17) 
 Sesame 26 (23) 3 (14) 23 (25) 11 (24) 12 (26) 
 Other 21 (19) 7 (33) 14 (15) 10 (22) 4 (9) 
Multiple allergies, n (%) 82 (73) 14 (67) 68 (74) 35 (76) 33 (72) 
Age at diagnosis, mean (SD)c 4 (4) 4 (3) 3 (4) 4 (4) 3 (3) 
Ever used EAI, n (%)c 59 (52) 9 (43) 50 (54) 25 (54) 25 (54) 
Ever had an allergic reaction, n (%)c 81 (73) 13 (62) 68 (74) 36 (78) 32 (70) 
Consistency of EAI carrying (1–10), mean (SD)c 8 (3) 8 (2) 8 (3) 8 (3) 8 (3) 
CharacteristicCohort (n = 131)Intervention 1Intervention 2
Intervention Group
1: Text Messages
Only (n = 26)
Control Group 1 (n = 105)Intervention Group
2: Text Messages and Financial Incentives (n = 54)
Control Group 2 (n = 51)
Sex, n (%)a      
 Female 65 (50) 13 (50) 52 (50) 27 (50) 25 (49) 
 Male 65 (50) 13 (50) 52 (50) 27 (50) 25 (49) 
 Decline to answer 1 (1) 0 (0) 1 (1) 0 (0) 1 (2) 
Age, mean (SD) 16 (1) 16 (2) 17 (1) 17 (1) 16 (1) 
Race, n (%)a      
 American Indian or Alaskan Native 1 (1) 0 (0) 1 (1) 0 (0) 1 (2) 
 Asian 5 (4) 2 (8) 3 (3) 2 (4) 1 (2) 
 Black 15 (12) 5 (19) 10 (10) 4 (7) 6 (12) 
 White 104 (79) 19 (73) 85 (81) 45 (83) 40 (78) 
 Other 5 (4) 0 (0) 5 (5) 3 (6) 2 (4) 
 Decline 1 (1) 0 (0) 1 (1) 0 (0) 1 (2) 
Ethnicity, n (%)      
 Hispanic 3 (2) 0 (0) 3 (3) 1 (2) 2 (4) 
 Decline to answer 1 (1) 0 (0) 1 (1) 0 (0) 1 (2) 
Grade, n (%)a      
 9 9 (7) 0 (0) 9 (9) 3 (6) 6 (12) 
 10 36 (28) 11 (42) 25 (24) 10 (19) 15 (30) 
 11 30 (23) 4 (15) 26 (25) 14 (26) 12 (24) 
 12 22 (17) 3 (12) 19 (18) 10 (19) 9 (18) 
 College Freshman 28 (21) 6 (23) 22 (21) 14 (26) 8 (16) 
 College Sophomore 6 (5) 2 (8) 4 (4) 3 (6) 1 (2) 
Allergic to, n (%)b      
 Peanut 79 (70) 15 (71) 64 (70) 30 (65) 34 (74) 
 Tree nuts 84 (74) 14 (67) 70 (76) 39 (85) 31 (67) 
 Soy 8 (7) 1 (5) 7 (8) 4 (9) 3 (7) 
 Wheat 3 (3) 1 (5) 2 (2) 0 (0) 2 (4) 
 Milk 18 (16) 5 (24) 13 (14) 6 (13) 7 (15) 
 Egg 18 (16) 3 (14) 15 (16) 8 (17) 7 (15) 
 Fish 11 (10) 2 (10) 9 (10) 6 (13) 3 (7) 
 Shellfish 20 (18) 6 (29) 14 (15) 6 (13) 8 (17) 
 Sesame 26 (23) 3 (14) 23 (25) 11 (24) 12 (26) 
 Other 21 (19) 7 (33) 14 (15) 10 (22) 4 (9) 
Multiple allergies, n (%) 82 (73) 14 (67) 68 (74) 35 (76) 33 (72) 
Age at diagnosis, mean (SD)c 4 (4) 4 (3) 3 (4) 4 (4) 3 (3) 
Ever used EAI, n (%)c 59 (52) 9 (43) 50 (54) 25 (54) 25 (54) 
Ever had an allergic reaction, n (%)c 81 (73) 13 (62) 68 (74) 36 (78) 32 (70) 
Consistency of EAI carrying (1–10), mean (SD)c 8 (3) 8 (2) 8 (3) 8 (3) 8 (3) 

SD, standard deviation.

a

Percentages may not add up to 100% because of rounding.

b

Numbers do not sum to group totals and percentages do not add up to 100% because participants could report multiple allergies.

c

Cohort: n = 113, intervention group 1: n = 21, control group 1: n = 92, intervention group 2: n = 46, control group 2: n = 46.

The proportion of participants who successfully documented they were carrying their EAI at each individual checkpoint is presented in Figure 3. During Intervention 1 (text message nudges), participants who received the intervention carried their EAI 28% of the time versus 38% for control group participants (P = .06). During Intervention 2 (text message nudges and financial incentives), participants who received the intervention carried their EAIs 45% of the time versus 23% for control group participants (P = .002). This difference was even larger in the sensitivity analysis that allowed for late responses (52% vs 32%) (Table 2).

FIGURE 3

Proportion of participants who carried their epinephrine auto-injector at each individual checkpoint.

FIGURE 3

Proportion of participants who carried their epinephrine auto-injector at each individual checkpoint.

Close modal
TABLE 2

Proportion of Checkpoints at Which Participants Successfully Demonstrated They Were Carrying Their EAIs

Study GroupResponse TimeCarrying Epinephrine, %P
Intervention 1: text message nudges only     
 Intervention group 1 (n = 26) On time (primary outcome) 28 .06  
 Control group 1 (n = 105) 38  
 Intervention group 1 (n = 26) On time and late (sensitivity analysis) 38 .015 
 Control group 1 (n = 105) 53  
Intervention 2: text message nudges and financial incentives    
 Intervention group 2 (n = 54) On time (primary outcome) 45 .002  
 Control group 2 (n = 51) 23  
 Intervention group 2 (n = 54) On time and late (sensitivity analysis) 52 .003 
 Control group 2 (n = 51) 32  
Study GroupResponse TimeCarrying Epinephrine, %P
Intervention 1: text message nudges only     
 Intervention group 1 (n = 26) On time (primary outcome) 28 .06  
 Control group 1 (n = 105) 38  
 Intervention group 1 (n = 26) On time and late (sensitivity analysis) 38 .015 
 Control group 1 (n = 105) 53  
Intervention 2: text message nudges and financial incentives    
 Intervention group 2 (n = 54) On time (primary outcome) 45 .002  
 Control group 2 (n = 51) 23  
 Intervention group 2 (n = 54) On time and late (sensitivity analysis) 52 .003 
 Control group 2 (n = 51) 32  

Participants in both intervention groups had, on average, neutral ratings for how informative the text messages were (5.67 and 5.43, respectively, on a scale from 1 to 10) and reported that the frequency at which text message nudges were sent was “just right” (72.2% and 76.3% of respondents, respectively). In contrast to their perceptions of the usefulness of the text messages, Intervention 2 participants reported, on average, that the financial incentives were highly motivating (7.71 on a scale from 1 to 10).

This study evaluated the effectiveness of 2 behavioral economics-informed interventions to promote epinephrine-carrying among adolescents with food allergies. Our results indicate that frequent text message nudges were unsuccessful in promoting EAI-carrying, but nudges combined with modest financial incentives increased EAI-carrying. Data for Intervention 2 are consistent with those of our pilot study among young adults with food allergies. In our pilot study, we found that participants who received text messages combined with financial incentives carried their EAIs twice as often as control group participants who received text messages only (54% vs 27% of the time, P = .023).22  The results from these 2 studies provide important evidence that interventions, especially those using financial incentives, can be effective in promoting a potentially life-saving behavior. The use of text message nudges alone, however, may simply not be enough to promote epinephrine-carrying. In fact, intervention participants were neutral about how informative the text message nudges were, suggesting that these were not perceived as beneficial.

Our findings suggest that individuals with food allergies may overestimate how consistently they carry their EAIs. Participants rated the statement “I carry my epinephrine auto-injector everywhere that I go” as an 8, on average, on a scale from 1 (completely disagree) to 10 (completely agree). Yet when we documented actual EAI carriage rates with photographic evidence, we found they carried their EAIs, at most, 53% of the time. Despite the success of the text message nudge plus financial incentives intervention, even during that intervention, adolescents and young adults with food allergies still carried their EAIs ~50% of the time.22  Other work has revealed that the primary reason for not using an EAI after an allergic reaction is that the device was not available.34 

Given the unpredictability and potential severity of food allergy reactions, epinephrine, the first line of defense for anaphylaxis, should be accessible at all times and promptly administered when an allergic reaction begins. In fact, time to administration is a contributing factor to death from anaphylaxis.7,8  Given that these and others’ results suggest that consistent EAI carrying may be difficult for most adolescents, a multilevel approach to safe food allergy management is needed. This may include more individually targeted interventions, especially when combined with improvements to EAI devices.35  Moving beyond the individual, EAIs should be universally accessible in public places, schools, recreation centers, and restaurants. Epinephrine is a safe, low-risk medicine that could be made available in the same way that automated external defibrillators are.36  Some EAIs are designed to provide verbal instructions during use, like automated external defibrillators, and are associated with lower rates of error during administration.37  A comparison of different models for supplying epinephrine in schools found that having undesignated stock epinephrine, without requiring individuals to also carry their own EAIs, was the most cost-effective.38 

Our study provides important information on a unique way to promote EAI-carrying. Most previous interventions among adolescents with food allergies are information-based interventions,39,40  with an emphasis on training individuals with food allergies and their families how to use EAIs.16  One such study among adolescents found that a mobile health decision support tool led to better symptom recognition and treatment selection for anaphylaxis, although it did not address access to epinephrine in cases when it would be the treatment of choice.41  We found only 1 other behavioral intervention to promote safer food allergy management that goes beyond information provision.40  This study assessed whether periodic checkpoints with school nurses influenced students’ epinephrine-carrying behavior.42  Checkpoints were not associated with more consistent epinephrine-carrying but were associated with carrying unexpired epinephrine.

The interventions evaluated in this study are simple and scalable. Although a text message-only intervention is inexpensive, our data currently do not support its implementation. The provision of modest financial incentives was successful and could be a beneficial strategy at an individual level. Future studies should explore whether smaller incentive amounts, nonfinancial incentives, or different incentive schemes (eg, lottery) are similarly effective. These could offer a lower-cost way of supporting adolescents in managing food allergies to organizations invested in this type of work, including insurance companies and advocacy organizations. In this study, we also could not determine if financial incentives alone, without text message nudges, would be successful as a stand-alone intervention to encourage EAI-carrying.

This study has several limitations. We had a small sample size that decreased over time. The study population was fairly homogenous and primarily included adolescents followed by allergy specialists, who are more likely to be more engaged in the management of their food allergies. Therefore, our results may not be generalizable to all adolescents with food allergies. It is possible that those adolescents not followed by an allergy specialist would more greatly benefit from interventions that increase the salience of food allergies.

The use of text message nudges combined with modest financial incentives was associated with higher rates of epinephrine-carrying among adolescents with food allergies in this randomized controlled trial. Despite this success and the replication of the results from our pilot study, adolescents still carried their life-saving medications ~50% of the time. Alternative strategies that emphasize making EAIs available in a variety of settings, rather than relying on each individual to remember, may be necessary to increase the safety of adolescents with food allergies.

Kate Kearns and Caleb Epstein contributed to the study but not by authorship.

This trial has been registered at www.clinicaltrials.gov (identifier NCT03284372).

Ms Dupuis contributed to data collection, conducted data analyses, interpreted the data, and wrote the first draft of the manuscript; Ms Feuerstein-Simon and Ms Luna Marti contributed to data collection, conducted data analyses, interpreted the data, and provided substantive input on the manuscript; Drs Cannuscio, Brown-Whitehorn, Mollen, and Spergel contributed to the study design and analysis plan, secured funding for the study, contributed to data collection, interpreted the data, and provided substantive input on the manuscript; Drs Meisel, Troxel, and Volpp contributed to the study design and analysis plan, secured funding for the study, interpreted the data, and provided substantive input on the manuscript; Drs Block, Gortmaker, and Kenney interpreted the data and provided substantive input on the manuscript; and all authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.

The funder had no role in design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

FUNDING: Research reported in this publication was supported by the National Institute of Child Health & Human Development under award number R21HD088941. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health (NIH).

CONFLICT OF INTEREST DISCLOSURES: The authors have indicated they have no potential conflicts of interest related to this article to disclose.

EAI

epinephrine auto-injector

1
Branum
AM
,
Lukacs
SL
.
Food allergy among children in the United States
.
Pediatrics
.
2009
;
124
(
6
):
1549
1555
2
Gupta
RS
,
Springston
EE
,
Warrier
MR
, et al
.
The prevalence, severity, and distribution of childhood food allergy in the United States
.
Pediatrics
.
2011
;
128
(
1
):
e9
e17
3
National Academies of Sciences, Engineering, and Medicine, Health and Medicine Division, Food and Nutrition Board; Committee on Food and Allergies: Global Burden, Causes, Treatment, Prevention, and Public Policy
.
Finding a Path to Safety in Food Allergy: Assessment of the Global Burden, Causes, Prevention, Management, and Public Policy
.
Oria
MP
,
Stallings
VA
, eds.
Washington, DC
:
National Academies Press
;
2016
4
Shaker
MS
,
Wallace
DV
,
Golden
DBK
, et al;
Collaborators
;
Chief Editors
;
Workgroup Contributors
;
Joint Task Force on Practice Parameters Reviewers
.
Anaphylaxis-a 2020 practice parameter update, systematic review, and Grading of Recommendations, Assessment, Development and Evaluation (GRADE) analysis
.
J Allergy Clin Immunol
.
2020
;
145
(
4
):
1082
1123
5
Gupta
R
,
Holdford
D
,
Bilaver
L
,
Dyer
A
,
Holl
JL
,
Meltzer
D
.
The economic impact of childhood food allergy in the United States
.
JAMA Pediatr
.
2013
;
167
(
11
):
1026
1031
6
Sicherer
SH
,
Sampson
HA
.
Food allergy: a review and update on epidemiology, pathogenesis, diagnosis, prevention, and management
.
J Allergy Clin Immunol
.
2018
;
141
(
1
):
41
58
7
Bock
SA
,
Muñoz-Furlong
A
,
Sampson
HA
.
Fatalities due to anaphylactic reactions to foods
.
J Allergy Clin Immunol
.
2001
;
107
(
1
):
191
193
8
Bock
SA
,
Muñoz-Furlong
A
,
Sampson
HA
.
Further fatalities caused by anaphylactic reactions to food, 2001-2006
.
J Allergy Clin Immunol
.
2007
;
119
(
4
):
1016
1018
9
Ramsey
NB
,
Guffey
D
,
Anagnostou
K
, %
Coleman
NE
,
Davis
CM
.
Epidemiology of anaphylaxis in critically ill children in the United States and Canada
.
J Allergy Clin Immunol Pract
.
2019
;
7
(
7
):
2241
2249
10
Anagnostou
A
,
Sharma
V
,
Herbert
L
,
Turner
PJ
.
Fatal food anaphylaxis: distinguishing fact from fiction
.
J Allergy Clin Immunol Pract
.
2022
;
10
(
1
):
11
17
11
Gupta
RS
.
Anaphylaxis in the young adult population
.
Am J Med
.
2014
;
127
(
1 Suppl
):
S17
S24
12
Sampson
MA
,
Muñoz-Furlong
A
,
Sicherer
SH
.
Risk-taking and coping strategies of adolescents and young adults with food allergy
.
J Allergy Clin Immunol
.
2006
;
117
(
6
):
1440
1445
13
Karam
M
,
Scherzer
R
,
Ogbogu
PU
,
Green
TD
,
Greenhawt
M
.
Food allergy prevalence, knowledge, and behavioral trends among college students - A 6-year comparison
.
J Allergy Clin Immunol Pract
.
2017
;
5
(
2
):
504
506.e5
14
Leach
L
,
Smith
HE
,
Brown
C
,
Davies
M
,
Jones
CJ
.
Young people’s views on the design of adrenaline auto-injectors: a qualitative study
.
J Allergy Ther
.
2018
;
9
:
1
15
Macadam
C
,
Barnett
J
,
Roberts
G
, et al
.
What factors affect the carriage of epinephrine auto-injectors by teenagers?
Clin Transl Allergy
.
2012
;
2
(
1
):
3
16
Muraro
A
,
Mendoza Hernandez
DA
.
Managing food allergy and anaphylaxis: a new model for an integrated approach
.
Allergol Int
.
2020
;
69
(
1
):
19
27
17
Lieberman
JA
,
Wang
J
.
Epinephrine in anaphylaxis: too little, too late
.
Curr Opin Allergy Clin Immunol
.
2020
;
20
(
5
):
452
458
18
Haisley
E
,
Volpp
KG
,
Pellathy
T
,
Loewenstein
G
.
The impact of alternative incentive schemes on completion of health risk assessments
.
Am J Health Promot
.
2012
;
26
(
3
):
184
188
19
Loewenstein
G
,
Price
J
,
Volpp
K
.
Habit formation in children: evidence from incentives for healthy eating
.
J Health Econ
.
2016
;
45
:
47
54
20
Volpp
KG
,
John
LK
,
Troxel
AB
,
Norton
L
,
Fassbender
J
,
Loewenstein
G
.
Financial incentive-based approaches for weight loss: a randomized trial
.
JAMA
.
2008
;
300
(
22
):
2631
2637
21
Volpp
KG
,
Troxel
AB
,
Pauly
MV
, et al
.
A randomized, controlled trial of financial incentives for smoking cessation
.
N Engl J Med
.
2009
;
360
(
7
):
699
709
22
Cannuscio
CC
,
Dupuis
R
,
Graves
A
, et al
.
A behavioral economics intervention to encourage epinephrine-carrying among food-allergic adults: a randomized controlled trial
.
Ann Allergy Asthma Immunol
.
2015
;
115
(
3
):
234
240.e1
23
Thorgeirsson
T
,
Kawachi
I
.
Behavioral economics: merging psychology and economics for lifestyle interventions
.
Am J Prev Med
.
2013
;
44
(
2
):
185
189
24
Relton
C
,
Torgerson
D
,
O’Cathain
A
, %
Nicholl
J
.
Rethinking pragmatic randomised controlled trials: introducing the “cohort multiple randomised controlled trial” design
.
BMJ
.
2010
;
340
:
c1066
25
Thaler
RH
,
Sunstein
CR
.
Nudge: Improving Decisions About Health, Wealth, and Happiness
.
London, UK
:
Penguin Books
;
2009
26
Fedele
DA
,
Cushing
CC
,
Fritz
A
,
Amaro
CM
,
Ortega
A
.
Mobile health interventions for improving health outcomes in youth: a meta-analysis
.
JAMA Pediatr
.
2017
;
171
(
5
):
461
469
27
Carbone
ET
,
Campbell
MK
,
Honess-Morreale
L
.
Use of cognitive interview techniques in the development of nutrition surveys and interactive nutrition messages for low-income populations
.
J Am Diet Assoc
.
2002
;
102
(
5
):
690
696
28
Strohacker
K
,
Galarraga
O
,
Williams
DM
.
The impact of incentives on exercise behavior: a systematic review of randomized controlled trials
.
Ann Behav Med
.
2014
;
48
(
1
):
92
99
29
Mitchell
MS
,
Goodman
JM
,
Alter
DA
, et al
.
Financial incentives for exercise adherence in adults: systematic review and meta-analysis
.
Am J Prev Med
.
2013
;
45
(
5
):
658
667
30
Broglio
K
.
Randomization in clinical trials: permuted blocks and stratification
.
JAMA
.
2018
;
319
(
21
):
2223
2224
31
Asch
DA
,
Volpp
KG
.
On the way to health
.
LDI Issue Brief
.
2012
;
17
(
9
):
1
4
32
Lachin
JM
.
Nonparametric statistical analysis
.
JAMA
.
2020
;
323
(
20
):
2080
2081
33
Detry
MA
,
Lewis
RJ
.
The intention-to-treat principle: how to assess the true effect of choosing a medical treatment
.
JAMA
.
2014
;
312
(
1
):
85
86
34
Warren
CM
,
Zaslavsky
JM
,
Kan
K
,
Spergel
JM
,
Gupta
RS
.
Epinephrine auto-injector carriage and use practices among US children, adolescents, and adults
.
Ann Allergy Asthma Immunol
.
2018
;
121
(
4
):
479
489.e2
35
Gallagher
M
,
Worth
A
,
Cunningham-Burley
S
,
Sheikh
A
.
Epinephrine auto-injector use in adolescents at risk of anaphylaxis: a qualitative study in Scotland, UK
.
Clin Exp Allergy
.
2011
;
41
(
6
):
869
877
36
Wood
JP
,
Traub
SJ
,
Lipinski
C
.
Safety of epinephrine for anaphylaxis in the emergency setting
.
World J Emerg Med
.
2013
;
4
(
4
):
245
251
37
Umasunthar
T
,
Procktor
A
,
Hodes
M
, et al
.
Patients’ ability to treat anaphylaxis using adrenaline autoinjectors: a randomized controlled trial
.
Allergy
.
2015
;
70
(
7
):
855
863
38
Shaker
MS
,
Greenhawt
MJ
.
Analysis of value-based costs of undesignated school stock epinephrine policies for peanut anaphylaxis
.
JAMA Pediatr
.
2019
;
173
(
2
):
169
175
39
Cutler
DM
.
Behavioral Health Interventions: What Works and Why?
In:
Bulatao
RA
,
Anderson
NB
, eds.
Understanding Racial and Ethnic Differences in Health in Late Life: A Research Agenda
.
Washington, DC
:
National Academies Press
;
2004
.
40
de Silva
D
,
Singh
C
,
Muraro
A
, et al;
European Academy of Allergy and Clinical Immunology Food Allergy and Anaphylaxis Guidelines Group
.
Diagnosing, managing and preventing anaphylaxis: systematic review
.
Allergy
.
2021
;
76
(
5
):
1493
1506
41
Gallagher
JL
,
Rivera
RD
,
Van Shepard
K
, et al
.
Life-threatening allergies: using a patient-engaged approach
.
Telemed J E Health
.
2019
;
25
(
4
):
319
325
42
Spina
JL
,
McIntyre
CL
,
Pulcini
JA
.
An intervention to increase high school students’ compliance with carrying auto-injectable epinephrine: a MASNRN study
.
J Sch Nurs
.
2012
;
28
(
3
):
230
237

Supplementary data