Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Open Access

Peer-reviewed

Research Article

Characterizing the within-person variability of food insecurity in everyday life

  • Caitlin T. Hines ,

    Roles Conceptualization, Data curation, Formal analysis, Funding acquisition, Methodology, Writing – original draft, Writing – review & editing

    cth5379@psu.edu

    Affiliation Department of Biobehavioral Health, The Pennsylvania State University, University Park, Pennsylvania, United States of America

    ⨯
  • Rebecca M. Ryan,

    Roles Supervision, Writing – review & editing

    Affiliation Department of Psychology, Georgetown University, Washington, DC, United States of America

    ⨯
  • Joshua M. Smyth

    Roles Conceptualization, Methodology, Supervision, Writing – review & editing

    Affiliation Department of Psychology, The Ohio State University, Columbus, Ohio, United States of America

    ⨯

Abstract

Food insecurity (FI), the lack of access to adequate food, is linked with negative health and psychological outcomes. FI is typically measured retrospectively over the last year; although this measurement is useful to understand FI prevalence to inform broad policy, it leaves the experience of FI in everyday life poorly understood. Understanding how FI varies across shorter periods of time (days or weeks) can help inform FI prevention and/or intervention. This study characterizes within-person (day-to-day) variance in FI in everyday life. Low-income parents of school-aged children at risk for FI (n = 153) completed daily text message surveys in two-week bursts. Daily FI was measured with 4 yes/no items ranging in severity: worry about food, parent eating less than they should, child eating less than they should, and skipping meals. Items were analyzed as a sum score and individually (to examine FI severity). Among parents who reported FI at least once, FI meaningfully varied day-to-day within individuals (~26% of variation). Different indicators of FI, however, had different proportions of between- and within-person variability: Worry about food, a less severe aspect of FI, had 32% daily variation, whereas the more severe aspect of meal skipping had 45% daily variation. Thus, although substantial between-person differences in FI exist, there is meaningful within-person variability in FI. -person FI variability may be related to the indicator (e.g., severe FI shows greater within-person variability). Considering within-person FI variability, and not just average FI level, may help us understand how FI undermines functioning and how and when best to intervene.

Citation: Hines CT, Ryan RM, Smyth JM (2025) Characterizing the within-person variability of food insecurity in everyday life. PLoS ONE 20(1): e0312543. https://doi.org/10.1371/journal.pone.0312543

Editor: Kamran Baig, Sindh Social Protection Authority, PAKISTAN

Received: February 6, 2024; Accepted: October 9, 2024; Published: January 17, 2025

Copyright: © 2025 Hines et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: Our limited data set is publicly available at the following website: https://repository.library.georgetown.edu/handle/10822/1089089 The data file includes all of the variables necessary to recreate the results, but not the demographic table. We do not have permission to share that information.

Funding: This work was supported by a grant from Georgetown University in addition to a grant from Eunice Kennedy Shriver National Institute of Child Health and Human Development of the National Institutes of Health under award number F32HD110175. There was no additional external funding received for this study.

Competing interests: The authors have declared that no competing interests exist.

Introduction

Food insecurity, or lack of consistent access to enough food to live an active and healthy life, is linked with negative health and psychological outcomes across the lifespan [1, 2]. Food insecurity is a widespread issue, but households with children are particularly at risk. Not only are they more likely to experience food insecurity than households without children [3], but its negative effects carry the potential to undermine development of the child(ren) in every domain [2].

Traditionally, food insecurity is measured by asking about experiences with food over the past 12 months, an approach that has led to a better understanding of food insecurity prevalence and informed food assistance policies. There is, however, little understanding of how food insecurity operates across shorter timelines (e.g., weeks or days); as such, a 12-month measure of food insecurity may not accurately capture the everyday experience of food insecurity. Understanding if and how food insecurity varies across shorter time intervals may provide insight into how food insecurity works to disrupt family functioning and child development, something that, despite decades of research on food insecurity prevalence, is still largely not understood. The first step in studying food insecurity across shorter intervals (e.g. days or weeks) is to identify if food insecurity varies on those time scales; this paper addresses this issue by using a daily measure of food insecurity to characterize food insecurity variability and identify if food insecurity is variable across shorter time intervals.

Food insecurity and children

For children, food insecurity is linked with worse overall health across childhood and increased risk of specific health conditions like anemia and asthma [4, 5]. In addition to undermining physical health, among school-aged children, food insecurity is linked with increased externalizing and internalizing behavior problems [6, 7], and worse math and reading skills [8, 9]. Researchers hypothesize that food insecurity disrupts child development in two primary ways, through inadequate nutrition and stress, both of which are supported throughout the literature. Inadequate nutrition, especially during key developmental windows, may impact functioning thus altering physical and cognitive development [10, 11]. Food insecurity likely works through stress to undermine development by first increasing emotional distress in parents, interfering with their parenting, which in turn, affects their children [2, 12].

Food insecurity measurement

Food insecurity has historically been conceptualized as something occurring over long periods of time. The primary way food insecurity is measured in the United States is using the U.S. Household Food Security Survey Module [13]. Created by the USDA, the HFSSM askes a variety of questions about experiences with food over the past year, including questions about eating patterns and having enough money to purchase food, but also about the worry that food will run out without enough money to buy more. Using this measure, the USDA estimates that more than 13 million children in the United States were exposed to food insecurity in 2022 [3]. This measure has transformed our understanding of food insecurity. It has documented food insecurity prevalence, identified those at risk, and shaped and informed food assistance programs.

The majority of studies examining the effects of food insecurity and its possible mechanisms use variations of the HFSSM and examine food experiences over the past 12 months to draw cross-sectional or longitudinal conclusions. There is, however, a growing interest and awareness that food insecurity may also vary across shorter periods of time. For example, the Census Pulse tracks food insufficiency using a 7-day reference period [14], and a small body of work uses daily measures of food insecurity [15–19]. Food insecurity encompasses things like hunger, skipping meals, and worry about food, which likely vary over shorter periods of time (e.g., some days will have more worry about food than others).

Daily food insecurity

To start, we know that food insecurity is not a static construct. Even when it is examined on a yearly level, food insecurity is dynamic, with families transitioning into and out of it [20, 21]. Further, those transitions and patterns into and out of food insecurity are differentially associated with outcomes for children [6, 7]. Early work using daily measures of food insecurity found that food insecurity can vary throughout the SNAP (Supplemental Nutrition Assistance Program) monthly cycle for adults and adolescents. Specifically, food insecurity decreases after the initial SNAP benefit transfer, and increases as benefits decrease throughout the month [16, 22]. Other emerging work using daily measures finds associations between daily food insecurity and outcomes like parent and child well-being and family routines [15, 17]. This initial work documents that there is day-to-day variability in food insecurity, and it has started to highlight the potential in measuring food insecurity using shorter measurement windows. Much about how food insecurity operates on shorter time intervals, however, is unknown. The aim of this paper is to characterize food insecurity within persons, across short time scales. Best practices for daily measurement, and how daily food insecurity measurement could be used moving forward will be discussed.

The present study

Research using 12-month HFSSM developed a foundational understanding of food insecurity, but we propose that there is value in supplementing this person-level characterization of food insecurity with more (temporally) granular measures that can capture the dynamics of someone’s lived experience. Understanding if and how food insecurity varies across shorter intervals, such as days or weeks, may increase our understanding of this economic stressor in multiple ways. Theoretically, it may provide new insight into the processes through which the long documented negative effects of food insecurity develop. Practically, understanding the timing and patterns of food insecurity across shorter time intervals may inform how and when to best intervene. Variability likely exists on multiple time scales. Research on the SNAP cycles suggest food insecurity varies periodically, within a month [16], but it may also vary day-to-day or week-to-week. For example, if food insecurity varies more over weeks compared to days, it may be that cyclical patterns like paydays or monthly benefit cycles influence much of the variation in food insecurity. Most broadly, it is unlikely that there is one time scale that can adequately characterize all of the nuance in experiences of food insecurity, and exploring if and how food insecurity varies across shorter time scales will provide insight into the lived experience of food insecurity. Using daily data on food insecurity among a low-income sample with school-aged children, this study seeks to characterize food insecurity by partitioning its variance into between- and within-person differences to determine the value in considering food insecurity measurement at shorter time intervals.

Partitioning variance in food insecurity into between- and within-person components will contribute to a working understanding of how food insecurity operates on shorter time intervals and provide insight into how best to use a daily measure of food insecurity. In broad terms, between-person variance represents largely stable differences between people, and within-person variance represents fluctuations in food insecurity experienced by a given person over time. The traditional 12-month measure, the HFSSM, largely captures and reflects those stable differences between people, or between-person variance. Using the HFSSM, each individual gets a value and corresponding food insecurity severity category that represents their experience over the past year. By design, the HFSSM does not capture any within-person variance in food insecurity, or how food insecurity may fluctuate within a person. Therefore, when using a daily measure, any variance attributable to within-person differences is potentially unique information not captured by the 12-month measure. Using a measure of daily food insecurity, comparing proportions of between- and within-person variance will help us understand the degree to which food insecurity itself is dynamic, or stable, across short time periods, as well as provide insight into what more frequent (e.g., daily) food insecurity measures may capture to supplement what is obtained using traditional food insecurity measures.

Additionally, food insecurity encompasses many experiences, ranging from worry about food to skipping meals. Although typically considered in aggregate (which is sensible to determine a person-level risk estimate), it is plausible that these experiences are distinct. For example, they reflect different levels of food insecurity severity and may vary differently over time. Thus, we believe it helpful to examine how distinct elements of food insecurity vary within people over short periods of time. This characterization will provide important insight into how food insecurity, and its encompassing experiences, may change among individuals and families over time, across circumstances and events.

Methods

Data

Data for this study are drawn from a larger evaluation of a school-based food assistance program, Power Packs Project (PPP). PPP is active throughout southeastern Pennsylvania, but data collection for this study was limited to two counties. After IRB approval, parents were recruited to participate in this study when they signed up for PPP between October and December of 2019. All children who received free and reduced priced lunch were eligible to enroll in PPP. A total of 271 families across six schools signed the consent form to participate in the study. One primary caregiver per household signed up to receive daily text message surveys to their phone (90% mothers). The youngest child in elementary school in the household was selected as the focal child for child focused items, resulting in a sample of children aged 4–11 years old (pre-Kindergarten through 5th grade).

Participants responded via daily text-message surveys for two consecutive weeks each month between January 9th and May 30, 2020. The sample was randomly divided in half, such that half of the sample was being surveyed at a time. One group started on January 9th, completed two-weeks of daily surveys, and then the other group completed two weeks of daily surveys. This pattern of two-week measurement bursts followed by two weeks with no surveys continued for five burst cycles, through May 30, 2020. Moving forward we will refer to each two-week period of daily surveys as a “burst”.

Sample

The families participating in the study were from six Title 1 elementary schools in two school districts in Pennsylvania. This low-income, partially- rural sample was majority Latinx and participated in at least one food assistance program (SNAP or WIC). Demographic information for the sample was collected in a one-time survey distributed in June 2020. If a participant did not complete that survey, their race and education status were drawn from administrative data collected by the PPP team. See Table 1 for full demographic information for the analytic sample.

thumbnail
Download:
Table 1. Analytic sample demographics.

https://doi.org/10.1371/journal.pone.0312543.t001

Measures

Food insecurity was measured using four yes/no items adapted from the HFSSM [23] to asses daily food insecurity. Child focused items included the name of one focal child, the youngest child in elementary school in the household.

  1. Today, were you worried that your food would run out before you got money to buy more?
  2. Today, did you eat less than you felt you should because there wasn’t enough money to buy food?
  3. Did [child name] eat less today than you felt he/she should because there wasn’t enough money to buy food?
  4. Did you or [child name] skip a meal today because your family didn’t have enough money for food?

The first item taps the parent psychological experience of food insecurity, whereas the next three items focus on food insufficiency of both parent and child. Following the approach taken in the HFSSM, responses to the food insecurity items were summed. Items were summed each day to create a daily food insecurity composite score, with a 0–4 range. Responding “no” to each questions yields a score of 0, while answering “yes” to each question results in a score of 4.

Analytic strategy

To characterize daily food insecurity, we completed a series of null, or empty, multilevel models in order to partition variance and calculate intraclass correlation coefficients (ICCs). We estimated three level models to account for the nested nature of the data, where daily responses (Level 1), were nested within measurement bursts (Level 2), and between different persons (Level 3).

Total within-person variance is the sum of variance in daily responses (Level 1) and variance at the measurement burst level (Level 2). Day-to-day variance captures variance within individuals within bursts (and error), whereas burst level variance captures variance within individuals across bursts. Level 3, or person level variance, represents between-person variance, or that which is attributable to largely stable differences between individuals. ICCs were calculated as the proportion of variance of interest to the total variance in the model (See Eqs 1–4). The models and ICCs were estimated and calculated separately for the daily food insecurity sum score (0–4), and each individual food insecurity item.

(1)(2)(3)(4)

We estimated linear equations and calculated the variance components following the same procedure for the food insecurity sum score and individual food insecurity items. Although the individual food insecurity items were binary (yes/no response options), this approach is appropriate when proportions are not extreme [24].

Given that we are examining variation in daily food insecurity, we limited the analytic sample to only those who reported food insecurity at some point during data collection (resulting in the loss of 27 cases). Meaning, if a participant answered “No” to all four food insecurity items for the entirety of the study period, they were excluded from analyses. Although the entire sample was at risk for food insecurity, if a parent never endorsed any food insecurity item in the study period, their responses would artificially lower estimates of variance for food insecurity (as they did not report the experience of food insecurity). As our goal is to partition variance in food insecurity, we limited the sample to those who reported food insecurity at some point during the study period. Further, we limited the sample to those participants who completed at least 50% of the daily surveys to ensure those included in analyses had adequate responses for within-person analyses, resulting in an analytic sample of 153 participants. No significant differences in demographics or food insecurity were present between the analytic and full sample. Overall, participants included in the analytic sample completed 89.78% of the daily surveys, and response rates stayed fairly consistent across the measurement period. The percentage of surveys completed in each of the five 14-day measurement bursts ranged from 86.33% in burst 1 to 87.91% in burst 5, with a peak of 91.97% in burst 4.

Results

Fig 1 and Table 2 include results from the variance partitioning analyses for the food insecurity sum score and the individual food insecurity items. Analyses reveal that there is variation in food insecurity both between people and within people (variance attributable to both day and burst). ICCs indicate that 59% of the variation in the food insecurity sum score is at the person level, or between persons, with the remaining 40% due to within-person variability (variance attributable to burst, day, and error). The food insecurity sum score showed slightly lower within-person variability compared to the individual food insecurity items, which had proportions of within-person variance ranging from 47%–57%.

thumbnail
Download:
Fig 1. Daily food insecurity: Variance components.

Total within-person variance = burst-level variance + day-level variance.

https://doi.org/10.1371/journal.pone.0312543.g001

thumbnail
Download:
Table 2. Food insecurity: Intercepts, variances, and variance components.

https://doi.org/10.1371/journal.pone.0312543.t002

A larger proportion of the within-person variability was due to day-to-day variation in responses compared to burst level variability. The proportion of variability attributable to measurement burst was fairly similar across measures, ranging from 9% for children eating less than they should, to 15% for parents worrying food would run out.

Exploratory analyses

We completed additional analyses to examine if variance components changed over time. Notably, the onset of the COVID-19 pandemic occurred about halfway through data collection, in March of 2020. Previous work, in this data set and others, document that levels of food insecurity initially increased with the onset of COVID-19 [18, 25], but there were also supplemental support policies introduced in response to COVID-19. If and how such changes were related to within-person variability in food insecurity is unknown.

In order to examine how variance in daily food insecurity changed across the study period, we used a similar approach to our main analyses, computing variance components by burst. This allowed us to examine if and how variance components changed across the study period. In order to do this, we completed empty two-level models, accounting only for nesting of days (Level 1) between persons (Level 2). Accounting for burst level variance was not necessary in these models, as each burst was analyzed separately.

Results from partitioning variance in food insecurity by burst are in Fig 2; these analyses reveal that person-level ICCs (representing between-person variance) for the food insecurity sum score and the individual items increased over time. This indicates that day-to-day variability in food insecurity (not shown in the figure) decreased across the study period. The food insecurity sum score person-level ICC increased by 30% (0.627 to 0.815) while parents eating less than they should had the largest increase of 71% (0.395 to 0.677). By the end of the study period, within-person, or day-to-day, variation was low. Only 18.5% of the variance in the food insecurity sum score was due to day-to-day variability (person-level ICC in burst 5 of 0.815).

thumbnail
Download:
Fig 2. Person level ICCs across study period.

This figure represents changes in person-level ICCs across the study period, not means. Person-level ICCs represent between-person variance, or the proportion of variance attributable to differences between persons.

https://doi.org/10.1371/journal.pone.0312543.g002

We completed further analyses to explore possible reasons for this decrease in within-person variance. First, we examined if food insecurity means decreased over time, as zero-inflated variables should exhibit less variability. Although food insecurity peaked at the onset of COVID-19, mean levels of endorsement at the end of the study period were very similar to those in the beginning of data collection. Therefore, in these data, low endorsement does not appear to explain the decrease in within-person variation.

As mentioned, the COVID-19 pandemic and associated school closures began about halfway through the study period, during burst 3. Although rates of food insecurity spiked with the onset of the pandemic, the first programmatic and policy changes were enacted in late March 2020 to help counteract increased food insecurity [26]. It is possible that aid from social programs and policies limited day-to-day variability in food insecurity. To explore this hypothesis, we compared variance components for those participating in SNAP and those not participating in SNAP. We did not have adequate sample size to examine differences in SNAP use before and after the onset of COVID-19, so analyses were conducted with data collapsed across bursts. These exploratory analyses (Fig 3) revealed that the group receiving SNAP had less within-person variability, or day-to day variability, than the group who did not receive SNAP. This pattern emerged in the food insecurity sum score, but it was especially pronounced for children eating less than they should and meal skipping. For children eating less than they should, within-person variance accounted for 66% of the variance in the group not receiving SNAP, and only 40% for the group receiving SNAP. For adult or child skipping a meal, within-person variance (burst and day+error) accounted for 64 percent of the variance in the group not receiving SNAP, and only 51 percent of the group receiving SNAP.

thumbnail
Download:
Fig 3. Food insecurity variance components by SNAP status.

SNAP, Supplemental Nutrition Assistance Program; SNAP n = 90; NO SNAP n = 50.

https://doi.org/10.1371/journal.pone.0312543.g003

Discussion

Results from this study indicate that food insecurity varies on a daily basis, but that there is heterogeneity in that variation. Overall, about 40% of the variation in food insecurity reflects within-person differences, and a larger proportion of within-person variation across days relative to across bursts demonstrates that it varies more over shorter time scales (days), than longer ones (bursts). Our results are similar to two previous studies that found about 35% of the variance in food insecurity was due to within-person differences, although those studies used slightly different food insecurity measures and different samples [16, 19]. Our analyses divided within-person variance in food insecurity between days and bursts. In these data, measurement bursts represent 14 days of nightly surveys, therefore burst-level variance in food insecurity can be interpreted as more periodic variation, compared to day-to-day. Although the majority of within-person variance of food insecurity is attributable to days, and not bursts, the presence of burst-level variance suggests that some variability in food insecurity may be cyclical in nature. Burst-level variation reflects variation that occurs over a two-week period, which may be associated with cyclical patterns like payroll or SNAP benefit distribution [16].

These results confirm expectations that substantial, largely stable between person differences in food insecurity exist, but extend our understanding to include the view that there is meaningful within-person variability in food insecurity over weeks and days. Additionally, item level analyses reveal differences in variance components by indicator. Indicators of more severe food insecurity (e.g. eating less than you should and skipping meals) show greater within-person variability. Within-person variability was lowest for parent worry about food, the least severe indicator of food insecurity. This was a somewhat surprising finding, as worry is a psychological component of food insecurity, and it seemed plausible to us that such a psychological construct would vary more day-to-day relative to more severe indicators reflective of deprivation (e.g., skipping meals). It could be that this item was worded in such a way that does not capture worry that changes day-to-day, or, it is possible that worry about food is more of a person-level trait (that is, some people may be prone to worry, whereas others are less so).

We should, however, interpret item-level analyses with caution, as any preliminary conclusions are based on single indicators. Further, each item had more within-person variation than the sum score. This may be because individual items contain more measurement error, or it could be because the food insecurity sum score obscures differences between experiences. For example, a score of 2 on the sum score could reflect various experiences. Understanding and exploring these nuances is an important area for future research to identify and optimize which food insecurity items are most suited to within-person measurement.

Next, our exploratory analyses suggest that within-person variability may change in different contexts or circumstances. Between-person variability increased over the study period. Over time, our participants’ responses became more stable. It may be that changes in environmental circumstances may have resulted in decreases in daily variability. With COVID-19 came economic supports and increases in food assistance, and it is possible that these supports smoothed consumption in a way that decreased instability and variance in food insecurity. Results from the exploratory SNAP analyses are supportive of this theory, as those receiving SNAP had less within-person variation. Future work should further examine the association between food assistance and food insecurity variability, but our exploratory results suggest that those who do not receive SNAP may have increased food insecurity instability.

Daily food insecurity may be a useful tool in exploring food insecurity mechanisms and informing food assistance programs. Research should continue, however, to explore how daily variation in food insecurity varies by food insecurity experience and circumstance in order to identify when it is most useful and appropriate to implement. When there are low levels of within-person variation in any construct, daily measurement, which is costly and time consuming for both participants and researchers, may not be a necessary or appropriate approach [27]. Future research should work to characterize contributors to within-person variability in food insecurity and identify under which circumstances food insecurity variability is highest. Further, we should identify if, and how, that variance, or instability in food insecurity, influences functioning. Typically, instability is considered a negative stressor across domains [28, 29]. Thus, it is possible that it is instability or uncertainty in food insecurity which is especially harmful, not just food insecurity severity or level. Daily measurement is a promising method to explore this.

This work should be interpreted with its limitations in mind. To start, we had a relatively small sample size of parents with young children from a singular geographic location, and results may not generalize to other populations. The fairly homogeneous nature of the sample at this specific time period may have made people appear more similar to one another than during a different time, with a different sample. Further, we limited our analytic sample to those who reported food insecurity, meaning results may not generalize to other low-income samples that do not experience food insecurity. Person-level estimates may be an overestimate, as participants in our sample were regionally close and surveyed at the same point in time, maximizing their similarity. It may be easier to detect daily variation in food insecurity among a larger and more heterogeneous sample in location and circumstance. Additionally, we only used four yes/no food insecurity items, so the granularity of information regarding food insecurity experience is limited (e.g., versus using a continuous measure).

In this study, 40 to 57% of the variance in food insecurity was due to within-person differences, leaving 43–60% of the variance due to differences between people. We suspect the typical 12-month measure of food insecurity likely captures the majority of person-level variance in food insecurity in an effective and efficient way, and it should continue to be used. As we move into attempting to capture experiences and flux in food insecurity on a shorter time scale, however, this work provides preliminary guidance on how those items may be adopted or extended to be sensitive to short-term variation. Researchers should consider integrating daily measurement with larger macro scale measurement, episodically, to increase understanding of specific circumstances. For example, short daily measurement bursts strategically timed within a larger study may provide invaluable insight into specific periods of interests or policy changes.

Although substantial and largely stable between-person differences in food insecurity exist, there is meaningful within-person variability in food insecurity. Moreover, within-person food insecurity variability may be related to the indicator (e.g., indicators of more severe FI show greater within-person variability) and to the context (e.g., receiving government supports to reduce food insecurity or not). Such findings highlight the potential importance of predictability and stability in the food insecurity experience. Considering food insecurity variability, and not just level, may help us understand how food insecurity undermines functioning, and how and when best to intervene. If within-person food insecurity variability itself is related to negative outcomes, efforts to reduce food insecurity instability may be a promising and novel way to combat this widespread public health issue. In sum, continued research using daily measurement of food insecurity is a promising tool to better understanding and working to reduce food insecurity and its harmful effects.

References

  1. 1. Gundersen C, Ziliak JP. Food insecurity and health outcomes. Health Affairs. 2015;34: 1830–1839. pmid:26526240
  2. 2. Hines CT, Markowitz AJ, Johnson AD. Food Insecurity: What Are Its Effects, Why, and What Can Policy Do About It? Policy Insights from the Behavioral and Brain Sciences. 2021;8: 127–135.
  3. 3. Rabbitt MP, Hales LJ, Burke MP, Coleman-Jensen A. Household Food Security in the United States in 2022. 2022.
  4. 4. Kirkpatrick SI, McIntyre L, Potestio ML. Child hunger and long-term adverse consequences for health. Archives of Pediatrics & Adolescent Medicine. 2010;164: 754–762. pmid:20679167
  5. 5. Mangini LD, Hayward MD, Dong YQ, Forman MR. Household food insecurity Is associated with childhood asthma. The Journal of Nutrition. 2015;145: 2756–2764. pmid:26491120
  6. 6. Grineski SE, Morales DX, Collins TW, Rubio R. Transitional dynamics of household food insecurity impact childrenʼs developmental outcomes. Journal of Developmental & Behavioral Pediatrics. 2018;39: 715–725. pmid:29957684
  7. 7. Kimbro RT, Denney JT. Transitions into food insecurity associated with behavioral problems And worse overall health among children. Health Affairs. 2015;34: 1949–1955. pmid:26526254
  8. 8. Hobbs S, King C. The unequal impact of food insecurity on cognitive and behavioral outcomes among 5-year-old urban children. Journal of Nutrition Education and Behavior. 2018;50: 687–694. pmid:29753634
  9. 9. Wang Y, Black M. Child food insecurity and neighborhood food environment in relation to academic performance and behavior problems among a national sample of kindergarten children. Current Developments in Nutrition. 2019;3.
  10. 10. Metallinos-Katsaras E, Colchamiro R, Edelstein S, Siu E. Household food security status Is associated with anemia risk at age 18 months among low-income infants in Massachusetts. Journal of the Academy of Nutrition and Dietetics. 2016;116: 1760–1766. pmid:27451132
  11. 11. Tanner EM, Finn-Stevenson M. Nutrition and brain development: Social policy implications. American Journal of Orthopsychiatry. 2002;72: 182–193. pmid:15792058
  12. 12. Gee KA, Asim M. Parenting while food insecure: Links between adult food insecurity, parenting aggravation, and children’s behaviors. Journal of Family Issues. 2019;40: 1462–1485.
  13. 13. Bickel GW, Hamilton WL, Cook JT, Thompson WW, Buron LF, Frongillo EA, et al. Household Food Security in the United States in 1995 Summary Report of the Food Security Measurement Project. 1997.
  14. 14. Schanzenbach D, Pitts A. Estimates of Food Insecurity During the COVID-19 Crisis: Results from the COVID Impact Survey, Week 1 (April 20–26, 2020). Northwestern Institute for Policy Research; 2020.
  15. 15. Hines CT, Steimle S, Ryan R. Associations between daily food insecurity and parent and child well-being. Developmental Psychology. 2024;60: 809–839. pmid:38451703
  16. 16. Gassman-Pines A, Schenck‐Fontaine A. Daily food insufficiency and worry among economically disadvantaged families with young children. Journal of Marriage and Family. 2019;81: 1269–1284.
  17. 17. Ryan RM, Gassman-Pines A, Steimle S, Baker G, Hines CT, Johnson AD. The role of public and private food assistance in supporting families’ food security and meal routines. Children and Youth Services Review. 2023;150: 106994.
  18. 18. Steimle S, Gassman-Pines A, Johnson AD, Hines CT, Ryan RM. Understanding patterns of food insecurity and family well‐being amid the COVID‐19 pandemic using daily surveys. Child Development. 2021;92. pmid:34435668
  19. 19. Hazzard VM, Loth KA, Crosby RD, Wonderlich SA, Engel SG, Larson N, et al. Relative food abundance predicts greater binge-eating symptoms in subsequent hours among young adults experiencing food insecurity: Support for the “feast-or-famine” cycle hypothesis from an ecological momentary assessment study. Appetite. 2023;180: 106316. pmid:36167172
  20. 20. Huang X, King C. Food insecurity transitions and housing hardships: Are immigrant families more vulnerable? Journal of Urban Affairs. 2018;40: 1146–1160.
  21. 21. Mabli J, Monzella K, Franckle RL, Delgado PL. Food Insecurity Transitions and Changes in Employment and Earnings. American Journal of Preventive Medicine. 2023;64: 368–376. pmid:36464558
  22. 22. Gassman-Pines A, Bellows L, Copeland WE, Hoyle RH, Odgers CL. Day-to-day variation in adolescent food insecurity. Children and Youth Services Review. 2023;149: 106954. pmid:37304155
  23. 23. Coleman-Jensen A, Rabbitt MP, Gregory CA, Singh A. Household Food Security in the United States in 2021. 2022. www.ers.usda.gov
  24. 24. Mbekwe Yepnang AM, Caille A, Eldridge SM, Giraudeau B. Association of intracluster correlation measures with outcome prevalence for binary outcomes in cluster randomised trials. Stat Methods Med Res. 2021;30: 1988–2003. pmid:34218744
  25. 25. Gundersen C, Hake M, Dewey A, Engelhard E. Food Insecurity during COVID-19. Applied Eco Perspectives Pol. 2021;43: 153–161. pmid:33042509
  26. 26. Bitler M, Hoynes H, Schanzenbach DW. The Social Safety Net in the Wake of COVID-19. Cambridge, MA: National Bureau of Economic Research; 2020 Sep p. w27796. Report No.: w27796.
    • 27. Podsakoff NP, Spoelma TM, Chawla N, Gabriel AS. What predicts within-person variance in applied psychology constructs? An empirical examination. Journal of Applied Psychology. 2019;104: 727–754. pmid:30640492
    • 28. Brown ED, Anderson KE, Garnett ML, Hill EM. Economic instability and household chaos relate to cortisol for children in poverty. Journal of Family Psychology. 2019;33: 629–639. pmid:31169392
    • 29. Li Z, Sturge-Apple ML, Martin MJ, Davies PT. Interactive effects of family instability and adolescent stress reactivity on socioemotional functioning. Developmental Psychology. 2019;55: 2193–2202. pmid:31343228