Symposium 17: Food Insecurity and Disordered Eating: Leveraging Diverse Research Methods to Improve Intervention and Inform Policy

Food insecurity, which refers to a lack of consistent access to enough and the kinds of food to support a healthy lifestyle, is a key social determinant of health that affects approximately 12% of U.S. households. Food insecurity often involves alternating periods of food access and food scarcity, which leads to dietary restriction followed by overconsumption. Although externally imposed, this “feast and famine” cycle mimics disordered eating cycles of restraint and binge eating. Additionally, individuals with weight and shape concerns may use compensatory behaviors in an attempt to mitigate the effects of binge eating on weight gain. Food insecurity has consistently been associated with increased rates of binge eating and compensatory behaviors, which increase risk for negative health outcomes, including cardiovascular disease and depression. Public policy is the first line of defense against food insecurity and provides important context for research and clinical intervention. Policies such as the Supplemental Nutrition Assistance Program (SNAP) and the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC) have been instrumental in increasing population-wide food access in the U.S. However, further research is needed to improve implementation of these policies and understand potential unintended consequences. For example, SNAP participants report worse diet quality and disordered eating compared to income-eligible non-participants. One hypothesis explaining this association is that the monthly frequency of SNAP disbursements contributes to the “feast and famine” cycle that contributes to increased binge eating. Thus, it is imperative that clinical approaches to reduce disordered eating behaviors incorporate the context of food insecurity, developmental stage, and policy.

This symposium brings together presenters from various professional disciplines (e.g., clinical psychology, public health, and implementation science) with expertise in food insecurity, disordered eating, and digital intervention implementation to highlight research on the role of food insecurity in the development and treatment of disordered eating. The presentations draw from diverse methodological approaches (i.e., ecological momentary assessment, longitudinal analysis, and qualitative analysis). The first presentation will present findings using ecological momentary assessment which demonstrated associations between the timing of earnings and benefits and binge eating, highlighting the role of the “feast and famine” cycle in driving disordered eating. By understanding how timing of benefits impacts disordered eating, we may be able to make important updates to policy to increase population-level benefit. Similarly, policy recognizes the importance of maternal and child health and seeks to ameliorate food insecurity in this population through programs like WIC. Understanding how food insecurity interacts with disordered eating at various stages of pregnancy and postpartum has implications for key maternal health outcomes, including gestational diabetes and preeclampsia. It may also inform improvements to WIC and other existing policy approaches. The second presentation will present findings from a longitudinal study of food insecurity during pregnancy and postpartum, which found a positive, but nonsignificant relationship between food insecurity and binge-type behavior during this period. Additionally, data related to how to best to reach individuals with food insecurity and disordered eating for targeted individual interventions may have important implications for broader public health campaigns to improve eating behavior in this population. The third presentation will discuss key considerations for adapting interventions to address binge eating among populations experiencing food insecurity, based on qualitative data collected from participants in a trial of a digital intervention for binge eating among adults with type 2 diabetes mellitus. The discussant, who is an expert in food insecurity and eating behavior, will synthesize findings from the presentations and consider how they may be incorporated into improved treatment of disordered eating among food-insecure populations.


 
1: Patterns of Binge-Eating Symptoms Relative to Timing of Earnings/Benefits Receipt Among Young Adults Experiencing Food Insecurity

Authors: Vivienne M. Hazzard, PhD, MPH, RD1, Ross D. Crosby, PhD2, Katie A. Loth, PhD, MPH, RD3, Scott G. Engel, PhD2, Stephen A. Wonderlich, PhD2, Dianne Neumark-Sztainer, PhD, MPH, RD4

Affiliations: 1Charles Stewart Mott Department of Public Health, Michigan State University College of Human Medicine; 2Sanford Center for Biobehavioral Research; 3Department of Family Medicine and Community Health, University of Minnesota Medical School; 4Division of Epidemiology & Community Health, University of Minnesota School of Public Health

Introduction: The objective of the present study was to characterize patterns of binge-eating symptoms relative to timing of earnings/benefits receipt among young adults experiencing food insecurity (FI).
Method: The sample included 43 young adults from the EAT 2010-2018 study with FI who reported receipt of earnings and/or benefits during a 2-week ecological momentary assessment (EMA) protocol (Mage=25.2±1.9 years; 72.1% women; 67.4% Black, Indigenous, or a Person of Color). Participants indicated the day in the protocol on which they received such earnings/benefits and provided data up to 4 times/day across the protocol on binge-eating symptoms after an eating episode. Multilevel interrupted time-series analyses were conducted to examine trajectories of binge-eating symptoms aggregated at the daily level before and after receipt of earnings/benefits, adjusting for age, gender, and weekday/weekend day.
Results: Results indicated that levels of binge-eating symptoms significantly decreased in days before earnings/benefits receipt, significantly increased immediately following receipt of earnings/benefits, and significantly decreased in the days following earnings/benefits receipt.
Conclusion: These findings highlight the need for stable provision of food access to reduce binge eating.


 
2: Food insecurity during the perinatal period: Characteristics and associations with dysregulated eating

Authors: Riley J. Jouppi, MS1, Christine C. Call, PhD2, Rachel P. Kolko Conlon, PhD2, & Michele D. Levine, PhD1,2

Affiliations: 1Department of Psychology, University of Pittsburgh; 2Department of Psychiatry, University of Pittsburgh Medical Center

Introduction: Little is known about food insecurity (FI) and its relation to dysregulated eating in the perinatal period (i.e., pregnancy and postpartum) when biopsychosocial changes may alter the experience of FI. This study explored the prevalence and persistence of FI and associations with dysregulated eating across the perinatal period.

Method: This is a secondary analysis of a perinatal lifestyle intervention trial that enrolled participants with pre-pregnancy body mass index (BMI)≥25. The sample includes participants who completed the U.S. Department of Agriculture FI measure at early pregnancy (T0), late pregnancy (T1), and/or 6 months postpartum (T6; N=125; Mage=31.7±4.9y; MBMI=32.1±6.6kg/m2). Participants completed the Eating Disorder Examination-Questionnaire 6.0 (EDE-Q) at all timepoints. Linear and logistic regression models explored longitudinal associations between FI and EDE-Q global/subscale scores and odds of experiencing loss of control eating (LOC; 0=never, 1=any), respectively, covarying for previous timepoints and randomization status as appropriate.

Results: FI was endorsed by 28% of participants at T0, 24% at T1, and 20% at T6. Relationships between FI and EDE-Q scores across the perinatal period did not reach statistical significance (ps≥.2) or follow a reliable pattern. Although rates of LOC at T1 and T6 were higher in those who, at previous timepoints (T0 and T1, respectively) reported FI (TI: 31%; T6: 50%) vs. no FI (TI: 28%; T6: 34%), differences were not statistically significant (OR=1.1, 95% CI [-1.5,1.7]; OR=1.4, 95% CI [-1.5,2.3]). Among 53 participants with FI data in pregnancy+postpartum, 17% endorsed FI in both periods, 21% in either pregnancy or postpartum, and 62% never. Though statistically nonsignificant (ps≥.3), participants with FI in pregnancy or postpartum consistently scored highest on the EDE-Q at T6 (Bs=0.2-0.6). Rates of LOC at T6 were higher in those with FI in pregnancy+postpartum (56%) and pregnancy or postpartum (45%) vs. those who never endorsed FI (39%), though not statistically significant (OR=1.3, 95% CI [-1.2,1.7]; OR=2.0, 95% CI [-0.9,2.3]).

Conclusion: FI was prevalent in the perinatal period. Relations between FI and dysregulated eating emerged but were weak and may be better detected in larger samples. Future research should clarify if adapted measures are needed and if these relationships operate differently in the perinatal period, as well as eating impacts of addressing perinatal FI and role of food access-related policies.

3: Improving the dissemination and implementation of a digital intervention for binge eating among adults with food insecurity

Authors: Rebecca L. Flynn1,2, Isabel R. Rooper1,2, Chidiebere Azubuike, MS1,2, Graham Miller1,2, Adrian Ortega, PhD1,3, Leah M. Parsons1,2, Jennifer E. Wildes, PhD4 & Andrea K. Graham, PhD1,2

Affiliations: 1Center for Behavioral Intervention Technologies, Northwestern University Feinberg School of Medicine; 2Department of Medical Social Sciences, Northwestern University Feinberg School of Medicine; 3Department of Preventive Medicine, Northwestern University Feinberg School of Medicine; 4Department of Psychiatry & Behavioral Neuroscience, The University of Chicago

Background: We are conducting the first pilot trial of a 16-week digital intervention for binge eating (“FoodSteps”) adapted for individuals with food insecurity. Additionally, to learn how to effectively engage individuals with food insecurity in the intervention, we are assessing their preferred methods of intervention dissemination (e.g., channels, modes, materials) and ideas for future implementation (e.g., additional intervention adaptations) at post-intervention.

Methods: Adults (N=31; 74% self-identified as female; 16% Hispanic; 29% non-Hispanic Black, 42% non-Hispanic White, 13% non-Hispanic other) with recurrent binge eating, FI, and type 2 diabetes mellitus joined an ongoing pilot trial, ending November 2024. Data here are from participants that have completed the trial (n=16); the full sample will be presented at the meeting.

Results: All participants indicated that disseminating the intervention via trusted sources (e.g., doctor, family/friends) and community events would be useful methods for spreading the word about FoodSteps, with slightly less interest in therapists and fitness centers (94%). Participants were comfortable (81%) or somewhat comfortable (19%) referring a friend or family member to FoodSteps (an indicator of intervention acceptability). All participants thought it would be useful for future users to learn about FoodSteps via websites, emails, and phone/video calls. When deciding to join the program, all participants said dissemination materials should highlight evidence that the program reduces binge eating and is free. Participants were only slightly less interested (94%) in dissemination materials showcasing prior users’ feedback. For future implementation, 50% were interested in having the intervention share their progress with their primary care doctor, while only 19% wanted to share progress with their therapist. Regarding additional intervention adaptations, participants recommended FoodSteps add a budget planner for food spending, recipes for healthy meals, a daily journal, fitness videos, and a physical activity log.

Discussion: Results give insight into how clinical researchers can improve outreach methods to engage individuals with food insecurity into binge eating interventions. Findings particularly highlight that trusted sources have an important role for dissemination and future implementation of digital intervention, including by learning about and supporting users’ intervention progress.

Chair - Lindsey Parnarouskis PhD  Student
Drexel University

Presenter - Vivienne Hazzard PhD, MPH, RD  Student
Michigan State University College of Human Medicine

Co-Author - Ross Crosby PhD  Student
Sanford Research

Co-Author - Katie Loth PhD MPH RD  Student
University of Minnesota

Co-Author - Scott Engel PhD  Student
Sanford Center for Biobehavioral Research

Co-Author - Stephen Wonderlich PhD  Student
Sanford Health

Co-Author - Diane Neumark-Sztainer PhD, MPH, RD  Student
University of Minnesota School of Public Health

Presenter - Riley Jouppi MS  Student
University of Pittsburgh

Co-Author - Christine Call PhD  Student
University of Pittsburgh

Co-Author - Rachel Kolko Conlon PhD  Student
University of Pittsburgh School of Medicine

Co-Author - Michele Levine PhD, FSBM  Student
University of Pittsburgh Medical Center

Presenter - Rebecca Flynn   Student
Northwestern University Feinberg School of Medicine

Co-Author - Isabel Rooper   Student
Northwestern University Feinberg School of Medicine

Co-Author - Chidiebere Azubuike MS  Student
Northwestern University Feinberg School of Medicine

Co-Author - Graham Miller   Student
Northwestern University Feinberg School of Medicine

Co-Author - Adrian Ortega PhD  Student
Northwestern University Feinberg School of Medicine

Co-Author - Leah Parsons   Student
Northwestern University Feinberg School of Medicine

Co-Author - Jennifer Wildes PhD  Student
The University of Chicago

Co-Author - Andrea Graham PhD  Student
Northwestern University Feinberg School of Medicine

Discussant - Lindsey Parnarouskis PhD  Student
Drexel University

Food insecurity, which refers to a lack of consistent access to enough and the kinds of food to support a healthy lifestyle, is a key social determinant of health that affects approximately 12% of U.S. households. Food insecurity often involves alternating periods of food access and food scarcity, which leads to dietary restriction followed by overconsumption. Although externally imposed, this “feast and famine” cycle mimics disordered eating cycles of restraint and binge eating. Additionally, individuals with weight and shape concerns may use compensatory behaviors in an attempt to mitigate the effects of binge eating on weight gain. Food insecurity has consistently been associated with increased rates of binge eating and compensatory behaviors, which increase risk for negative health outcomes, including cardiovascular disease and depression. Public policy is the first line of defense against food insecurity and provides important context for research and clinical intervention. Policies such as the Supplemental Nutrition Assistance Program (SNAP) and the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC) have been instrumental in increasing population-wide food access in the U.S. However, further research is needed to improve implementation of these policies and understand potential unintended consequences. For example, SNAP participants report worse diet quality and disordered eating compared to income-eligible non-participants. One hypothesis explaining this association is that the monthly frequency of SNAP disbursements contributes to the “feast and famine” cycle that contributes to increased binge eating. Thus, it is imperative that clinical approaches to reduce disordered eating behaviors incorporate the context of food insecurity, developmental stage, and policy.

This symposium brings together presenters from various professional disciplines (e.g., clinical psychology, public health, and implementation science) with expertise in food insecurity, disordered eating, and digital intervention implementation to highlight research on the role of food insecurity in the development and treatment of disordered eating. The presentations draw from diverse methodological approaches (i.e., ecological momentary assessment, longitudinal analysis, and qualitative analysis). The first presentation will present findings using ecological momentary assessment which demonstrated associations between the timing of earnings and benefits and binge eating, highlighting the role of the “feast and famine” cycle in driving disordered eating. By understanding how timing of benefits impacts disordered eating, we may be able to make important updates to policy to increase population-level benefit. Similarly, policy recognizes the importance of maternal and child health and seeks to ameliorate food insecurity in this population through programs like WIC. Understanding how food insecurity interacts with disordered eating at various stages of pregnancy and postpartum has implications for key maternal health outcomes, including gestational diabetes and preeclampsia. It may also inform improvements to WIC and other existing policy approaches. The second presentation will present findings from a longitudinal study of food insecurity during pregnancy and postpartum, which found a positive, but nonsignificant relationship between food insecurity and binge-type behavior during this period. Additionally, data related to how to best to reach individuals with food insecurity and disordered eating for targeted individual interventions may have important implications for broader public health campaigns to improve eating behavior in this population. The third presentation will discuss key considerations for adapting interventions to address binge eating among populations experiencing food insecurity, based on qualitative data collected from participants in a trial of a digital intervention for binge eating among adults with type 2 diabetes mellitus. The discussant, who is an expert in food insecurity and eating behavior, will synthesize findings from the presentations and consider how they may be incorporated into improved treatment of disordered eating among food-insecure populations.


 
1: Patterns of Binge-Eating Symptoms Relative to Timing of Earnings/Benefits Receipt Among Young Adults Experiencing Food Insecurity

Authors: Vivienne M. Hazzard, PhD, MPH, RD1, Ross D. Crosby, PhD2, Katie A. Loth, PhD, MPH, RD3, Scott G. Engel, PhD2, Stephen A. Wonderlich, PhD2, Dianne Neumark-Sztainer, PhD, MPH, RD4

Affiliations: 1Charles Stewart Mott Department of Public Health, Michigan State University College of Human Medicine; 2Sanford Center for Biobehavioral Research; 3Department of Family Medicine and Community Health, University of Minnesota Medical School; 4Division of Epidemiology & Community Health, University of Minnesota School of Public Health

Introduction: The objective of the present study was to characterize patterns of binge-eating symptoms relative to timing of earnings/benefits receipt among young adults experiencing food insecurity (FI).
Method: The sample included 43 young adults from the EAT 2010-2018 study with FI who reported receipt of earnings and/or benefits during a 2-week ecological momentary assessment (EMA) protocol (Mage=25.2±1.9 years; 72.1% women; 67.4% Black, Indigenous, or a Person of Color). Participants indicated the day in the protocol on which they received such earnings/benefits and provided data up to 4 times/day across the protocol on binge-eating symptoms after an eating episode. Multilevel interrupted time-series analyses were conducted to examine trajectories of binge-eating symptoms aggregated at the daily level before and after receipt of earnings/benefits, adjusting for age, gender, and weekday/weekend day.
Results: Results indicated that levels of binge-eating symptoms significantly decreased in days before earnings/benefits receipt, significantly increased immediately following receipt of earnings/benefits, and significantly decreased in the days following earnings/benefits receipt.
Conclusion: These findings highlight the need for stable provision of food access to reduce binge eating.


 
2: Food insecurity during the perinatal period: Characteristics and associations with dysregulated eating

Authors: Riley J. Jouppi, MS1, Christine C. Call, PhD2, Rachel P. Kolko Conlon, PhD2, & Michele D. Levine, PhD1,2

Affiliations: 1Department of Psychology, University of Pittsburgh; 2Department of Psychiatry, University of Pittsburgh Medical Center

Introduction: Little is known about food insecurity (FI) and its relation to dysregulated eating in the perinatal period (i.e., pregnancy and postpartum) when biopsychosocial changes may alter the experience of FI. This study explored the prevalence and persistence of FI and associations with dysregulated eating across the perinatal period.

Method: This is a secondary analysis of a perinatal lifestyle intervention trial that enrolled participants with pre-pregnancy body mass index (BMI)≥25. The sample includes participants who completed the U.S. Department of Agriculture FI measure at early pregnancy (T0), late pregnancy (T1), and/or 6 months postpartum (T6; N=125; Mage=31.7±4.9y; MBMI=32.1±6.6kg/m2). Participants completed the Eating Disorder Examination-Questionnaire 6.0 (EDE-Q) at all timepoints. Linear and logistic regression models explored longitudinal associations between FI and EDE-Q global/subscale scores and odds of experiencing loss of control eating (LOC; 0=never, 1=any), respectively, covarying for previous timepoints and randomization status as appropriate.

Results: FI was endorsed by 28% of participants at T0, 24% at T1, and 20% at T6. Relationships between FI and EDE-Q scores across the perinatal period did not reach statistical significance (ps≥.2) or follow a reliable pattern. Although rates of LOC at T1 and T6 were higher in those who, at previous timepoints (T0 and T1, respectively) reported FI (TI: 31%; T6: 50%) vs. no FI (TI: 28%; T6: 34%), differences were not statistically significant (OR=1.1, 95% CI [-1.5,1.7]; OR=1.4, 95% CI [-1.5,2.3]). Among 53 participants with FI data in pregnancy+postpartum, 17% endorsed FI in both periods, 21% in either pregnancy or postpartum, and 62% never. Though statistically nonsignificant (ps≥.3), participants with FI in pregnancy or postpartum consistently scored highest on the EDE-Q at T6 (Bs=0.2-0.6). Rates of LOC at T6 were higher in those with FI in pregnancy+postpartum (56%) and pregnancy or postpartum (45%) vs. those who never endorsed FI (39%), though not statistically significant (OR=1.3, 95% CI [-1.2,1.7]; OR=2.0, 95% CI [-0.9,2.3]).

Conclusion: FI was prevalent in the perinatal period. Relations between FI and dysregulated eating emerged but were weak and may be better detected in larger samples. Future research should clarify if adapted measures are needed and if these relationships operate differently in the perinatal period, as well as eating impacts of addressing perinatal FI and role of food access-related policies.

3: Improving the dissemination and implementation of a digital intervention for binge eating among adults with food insecurity

Authors: Rebecca L. Flynn1,2, Isabel R. Rooper1,2, Chidiebere Azubuike, MS1,2, Graham Miller1,2, Adrian Ortega, PhD1,3, Leah M. Parsons1,2, Jennifer E. Wildes, PhD4 & Andrea K. Graham, PhD1,2

Affiliations: 1Center for Behavioral Intervention Technologies, Northwestern University Feinberg School of Medicine; 2Department of Medical Social Sciences, Northwestern University Feinberg School of Medicine; 3Department of Preventive Medicine, Northwestern University Feinberg School of Medicine; 4Department of Psychiatry & Behavioral Neuroscience, The University of Chicago

Background: We are conducting the first pilot trial of a 16-week digital intervention for binge eating (“FoodSteps”) adapted for individuals with food insecurity. Additionally, to learn how to effectively engage individuals with food insecurity in the intervention, we are assessing their preferred methods of intervention dissemination (e.g., channels, modes, materials) and ideas for future implementation (e.g., additional intervention adaptations) at post-intervention.

Methods: Adults (N=31; 74% self-identified as female; 16% Hispanic; 29% non-Hispanic Black, 42% non-Hispanic White, 13% non-Hispanic other) with recurrent binge eating, FI, and type 2 diabetes mellitus joined an ongoing pilot trial, ending November 2024. Data here are from participants that have completed the trial (n=16); the full sample will be presented at the meeting.

Results: All participants indicated that disseminating the intervention via trusted sources (e.g., doctor, family/friends) and community events would be useful methods for spreading the word about FoodSteps, with slightly less interest in therapists and fitness centers (94%). Participants were comfortable (81%) or somewhat comfortable (19%) referring a friend or family member to FoodSteps (an indicator of intervention acceptability). All participants thought it would be useful for future users to learn about FoodSteps via websites, emails, and phone/video calls. When deciding to join the program, all participants said dissemination materials should highlight evidence that the program reduces binge eating and is free. Participants were only slightly less interested (94%) in dissemination materials showcasing prior users’ feedback. For future implementation, 50% were interested in having the intervention share their progress with their primary care doctor, while only 19% wanted to share progress with their therapist. Regarding additional intervention adaptations, participants recommended FoodSteps add a budget planner for food spending, recipes for healthy meals, a daily journal, fitness videos, and a physical activity log.

Discussion: Results give insight into how clinical researchers can improve outreach methods to engage individuals with food insecurity into binge eating interventions. Findings particularly highlight that trusted sources have an important role for dissemination and future implementation of digital intervention, including by learning about and supporting users’ intervention progress.

Chair - Lindsey Parnarouskis, PhD,
Drexel University

Presenter - Vivienne M. Hazzard, PhD, MPH, RD,
Michigan State University College of Human Medicine

Co-Author - Ross D. Crosby, PhD,
Sanford Research

Co-Author - Katie A. Loth, PhD MPH RD,
University of Minnesota

Co-Author - Scott G. Engel, PhD,
Sanford Center for Biobehavioral Research

Co-Author - Stephen A. Wonderlich, PhD, PhD,
Sanford Health

Co-Author - Dianne Neumark-Sztainer, PhD, MPH, RD,
University of Minnesota School of Public Health

Presenter - Riley J. Jouppi, MS,
University of Pittsburgh

Co-Author - Christine C. Call, PhD,
University of Pittsburgh

Co-Author - Rachel P. Kolko Conlon, PhD,
University of Pittsburgh School of Medicine

Co-Author - Michele D. Levine, PhD, FSBM,
University of Pittsburgh Medical Center

Presenter - Rebecca L. Flynn,
Northwestern University Feinberg School of Medicine

Co-Author - Isabel R. Rooper,
Northwestern University Feinberg School of Medicine

Co-Author - Chidiebere Azubuike, MS,
Northwestern University Feinberg School of Medicine

Co-Author - Graham Miller,
Northwestern University Feinberg School of Medicine

Co-Author - Adrian Ortega, PhD,
Northwestern University Feinberg School of Medicine

Co-Author - Leah M. Parsons,
Northwestern University Feinberg School of Medicine

Co-Author - Jennifer E. Wildes, PhD,
The University of Chicago

Co-Author - Andrea K. Graham, PhD,
Northwestern University Feinberg School of Medicine

Discussant - Lindsey Parnarouskis, PhD,
Drexel University