Identifying a Fine Line between Food Insecurity and Food Acquisition Stress: A Mixed Methods Exploration

Abstract

This study introduces the concept of Food Acquisition Stress (FAS), stress associated with food acquisition among those who do not necessarily screen positive for food insecurity.. This study used an exploratory sequential mixed methods approach among a sample of predominantly early childhood educators to develop a 7-item tool for measuring current and retrospective FAS. Using this tool, we identified that 61% of individuals who had FAS did not meet criteria for food insecurity. Capturing FAS, even among those categorized as food secure, has the potential to identify individuals who may need supportive interventions. Future research can explore how FAS is related to health behaviors.

Introduction

An estimated 12.7% of households in the United States (US) are affected by poverty with greater estimates among households with children under age 18 (17.6%) and age 6 (19.5%), households led by females (14%), and ethnic minority households (Black 22%, Hispanic 19.4%).¹ Current estimates of food insecurity for adults in the US range between 14% ² and 19.3%.³ Similarly, approximately 16% of households with children in the US experience food insecurity.⁴ This rate corresponds to over 50 million adults and over 6 million households with children that do not have access to adequate food for active, healthy living due to a lack of financial resources. ^2,5 Food insecurity negatively affects physical health, emotional well-being, and quality of life.^6–8

In the US, food insecurity is assessed using the Household Food Security Survey Module (HFSSM) from the United States Department of Agriculture (USDA).¹ The full version is 18 items, and short forms of 6 and 10 items are also available from the USDA. Questions focus on assessing if there has been inadequate money to get needed food in the last 12 months in a household (e.g., “food ran out before we got money…”) as well as assessing if hunger resulted in the home (e.g., “cut meals or skipped meals…”). Scoring conventions group respondents into levels of food insecurity based on the number of affirmative responses (i.e., sometimes or often) to the questions.

Outside the United states, the 8-item Food Insecurity Experiences Scale (FIES) from the Food and Agricultural Organization of the United Nations is based on the HFSSM and aims to capture the full range of severity of food insecurity.^9,10 Some questions are similar to the HFSSM (e.g., “your household ran out of money”) while some aimed to measure “uncertainty and anxiety regarding food access” (e.g., “ate only a few kinds of food”, p 10⁹). To date, the FIES has commonly been used to compare food security levels among nations.¹⁰ The Household Food Insecurity Access Scale (HFIAS) was designed in 2006 also for cross-cultural use to measure different domains of food insecurity (i.e., anxiety, food quality, food insufficiency).^11,12 The measure allows for determination of whether food insecurity occurred, and also the frequency of food insecurity experiences. According to instrument developers, a strength of this tool is to provide detailed population-level information to inform geographically targeted food insecurity interventions.¹¹ Thus, the various tools provide a number of options for food insecurity assessment.

Despite multiple assessment options and documentation of food insecurity rates across countries and salient sub-groups, assessment tools of less severe experiences have been developed and validated to a lesser extent. That is, there are limited data and measurement options available to understand nuances in stressful economic experiences related to food that may not reach the threshold for identification of food insecurity status. We hypothesize that stress associated with food acquisition among those who do not necessarily screen positive for food insecurity has the potential to impact individuals in salient ways.

An example of an important gap between an accepted measurement threshold and missing a point at which intervention may show impact is illustrated by work in the area of maternal depression screening. In studies of mothers with low depression symptoms (sub-threshold for positive screen) with preschool-age children, children were more likely to exhibit internalizing and externalizing behavior problems.^13,14 Mothers with low-level depression were also less likely to extend support for learning to their children.¹⁵ These studies suggested long-held discrete cut-points for identifying persons at-risk resulted in the under-identification of families that may benefit from interventions.

An assessment of stress associated with food acquisition may prove to be a salient indicator of economic instability with potentially important relationships to health outcomes like excess weight and diabetes. Further, retrospective assessment of food insecurity in all forms (i.e., Food insecure with/without food associated stress; Food secure with/without food associated stress) could prove useful in understanding a history of economic instability and food-related stress, potentially providing insights into current health outcomes (e.g., obesity), eating habits, and attitudes toward food. We posit that reporting on childhood food insecurity experiences, much like Adverse Childhood Experiences (ACEs),^16–18 may provide critical insights into salient experiences with food that continue to impact the health of individuals in adulthood.

Several segments of the US workforce may be more subject to food acquisition stress (FAS). For example, early childhood educators are a workforce of over 550, 000 who are paid $22,290 per year on average.¹⁹ Understanding the impacts and nuances of food insecurity experiences from the perspectives of early childhood educators is important given their daily interactions with children around food and the potential of their behaviors to influence children.^20–22 Further, while unique from other workforces in some ways (e.g., primarily female, education requirements), the early childhood education workforce has similarities to others fields such as health care support (e.g., nursing assistant, home care aids), service industry workers (e.g., servers), and facilities maintenance. All these occupations are also marked by low median income, high turnover, and above average health risks.^23–25 Thus, understanding of FAS among early educators has the potential to increase understanding in other workforces as well.

This study used an exploratory sequential mixed methods approach²⁶ to address the need for a nuanced and realistic understanding of FAS that does not reach threshold for food insecurity designation. In brief, qualitative interviews focused on experiences of stress related to food acquisition. These interviews were used to derive language for survey items, which were tested subsequently in a larger sample. Specifically, the purpose of the current study was twofold: (1) to capture the lived experiences of individuals related to current and past memories of FAS (qual) and (2) to leverage interview data to develop and provide preliminary data on properties of a tool to assess FAS, both at present and in childhood (QUANT). This mixed methods approach aimed to develop a measure that is relevant for understanding FAS and for studying its associated correlates and outcomes in the future.

Methods

Approach

We employed an exploratory sequential mixed methods design (qual → QUANT)²⁶ in which we gathered qualitative interviews (N = 28), generated and refined survey items, and conducted surveys in a Southern state. Our goal was to determine the generalizability of themes uncovered through in-depth interviews. After thematic analysis of narratives,²⁷ an extensive survey item bank was generated. This bank of items was refined and narrowed through expert and stakeholder review and cognitive interviewing. The final survey was collected from 1203 individuals. The Institutional review board at the University of Arkansas for Medical Sciences reviewed and approved this study.

Data Collection

Qualitative Data Collection to Inform Survey Development

Procedures for the qualitative portion of this study are described in more extensive detail elsewhere.²⁸ In brief, a trained research assistant²⁹ conducted interviews between February 2015 and May 2015 with early care and education teachers (ECETs) Research staff stratified early care and education centers by obesity prevalence in the community in which the center is located, randomly selected using a random number generator, and sent materials to recruit educators upon approval from administration. One educator per site was included in the study; 90% of invited sites participated. The research team invited interested teachers who did not match selection criteria to join a wait list for future research projects. All teachers who completed a screening interview received free 3-month access to an online continuing education platform, which supports achievement of state requirements for continuing education credits. Teachers completing the interview received a $50 cash compensation.

Quantitative Data Collection to Inform Survey Development

Subsequent to item development (for more detail, see below), an invitation for the quantitative survey collection was extended to 2413 individuals attending regional professional development trainings offered through the Cooperative Extension Service between January 2017 and May 2017. The trainer provided a cover sheet to potential participants detailing the study and informing educators that participation was voluntary and not related to the training they would receive. Participants completed surveys with paper and pencil. A final page was detached and submitted separately from the survey if participants elected to enter a drawing for $75. A total of 50% (N = 1203) elected to participate.

Measures

Qualitative Measure

Interview Guide

Belsky’s Determinants of Parenting Model³⁰ informed design of the interview guide which sought to elicit personal backgrounds (childhood up to present day) related to food and meals in their (participants’) homes. The final version reflected input from community stakeholders and a qualitative expert as well as modifications made after three pilot interviews. Additional detail is provided elsewhere.²⁸ All participants answered all questions although the exact order differed to follow each conversation.

Quantitative Measure

Survey

The PI (TS) generated quantitative survey items based on qualitative themes, striving to use educator language when possible. The item bank was reviewed and discussed with an expert in the field (SJ). Selected items were then pre-tested and revised through two cognitive interviews in the target population.³¹ Additionally, 65 Cooperative Extension agents (who train and interact regularly with community audiences) reviewed the draft surveys for feedback. Items from the USDA’s 10-item HFSSM were added for validation purposes. The short form was selected over the full 18-item version to minimize participant burden, and because the 10-item short form has shown criterion validity, construct validity, structural validity, and internal consistency.³² Participants provided answers to demographic questions (16 items) at the end of the survey.

Analysis

Qualitative Toward Development of Survey

Qualitative interviews were transcribed verbatim and imported into QSR NVivo 10 for analysis. ³³ In our prior work, we employed a directed content analysis with a start list informed by the Belsky model to code qualitative interviews.²⁸ Once team members came to agreement about themes arising outside the start list (i.e., inductive coding), team members discussed their recurrence, defined parameters of the theme, incorporated them into the codebook, and coded all interviews with the adjusted codebook. Thus, the analysis reflects a hybrid approach of combining deductive and inductive approaches.³⁴

Initial Examination of Survey Items and Scales

Statistical analyst used quantitative statistics to explore the properties of survey items and proposed scales. Analysts started from screening for and excluding careless responders (CR) using the following criteria: 1) Mahalanobis distance³⁵ that shows each respondent’s distance from the average response pattern were calculated; then, indicators were generated if the Mahalanobis distance (distributed as χ²) exceeded a critical value corresponding to p = 0.01³⁶; 2) Indicators for Even-Odd consistency³⁶, that assesses the extent to which participants choose equivalent response options to measuring similar constructs, were generated using the value of −0.1 as the cutoff based on the distribution of Even-Odd correlation; and 3) Indicators for response pattern, that using Longstring values³⁷ to identify participants repeatedly choose the same response, were generated using 10 as the cutoff value based on the design of the survey. Cases meeting two or more of the criteria were excluded from analyses (N =32). After this exclusion, 1,173 cases were retained. Descriptive statistics for items (frequencies, means) assessed for ceiling and floor effects and provided information of the distribution across response options. Responses to the 10-item HFSSM measure were summed and categorized into high, marginal, low, and very low food security in accordance with scoring directions.³⁸ For items conceived to represent a single construct (based on relationship to qualitative sub-themes), Cronbach’s alpha was calculated to provide an index of internal consistency.

Results

Qualitative Results Toward Development of Survey

Sample

ECETs in the qualitative portion of the study (n = 28) included 75.0% Caucasian, 14.3% African American, 7.1% Hispanic, and 3.5% Other ethnicities. More than half of the sample were lead teachers in their classrooms (67.9%). A majority held college degrees (60.7%), followed by those with some college education (32.1%), and 7.1% with a high school diploma or General Education Development (GED) certificate. The average age of participants was 40.7 years with 10.8 years of experience in the field. Educators in the qualitative component of our study reported FI at a rate of 38% in the present and 34% in their childhood based on the adapted HFSSM measure.

Memories of Childhood Food Insecurity

Table 1 provides a summary of themes with exemplar quotes as well as sample quantitative items that resulted from that theme. Several participants, irrespective of current food security status, described a pervasive experience of food insecurity in childhood. Hallmarks included an awareness that their parents were skipping meals to protect them and an uncertainty about the source of the next meal. In some cases, this experience was related to parental absence from the home as described by one educator who said, “Mom was at work, dad was wherever. You know, sometimes we’d just, or we’d go fix a peanut butter and jelly sandwich, or um, mainly we lived on cereal a lot when I was a kid (15).” The detailed descriptions of childhood food insecurity by participants reflects the poignancy of these experiences as illustrated by a participant who described uncertainty about timing and frequency of meals.

Table 1.

Qualitative Themes, Exemplar Quotes, and Related Item Examples

Theme	Exemplar Quotes	Quantitative Item Example(s)
Memories of Child Food Insecurity	“But we were in hard times though, we were kind of poor and food was kind of.... But... we had a big garden and that’s how we ate. We rarely went to a grocery store.” “Because he [my dad] was also a musician and sometimes, I remember that there were some evenings that we did not eat. We were waiting for him to come to see how much he earned to bring. But my mom, however, she would sit outside some evenings and she would give us whatever we had to eat, but she would go outside. You know, that is when it started, I think. Then sometimes, when my dad would arrive he would bring chicken, you know, something big, a big meal and they would wake all of us up to eat.”	- Adapted, retrospective HFSSM items  - The food that your family bought just didn’t last, and you didn’t have money to get more.
Memories of Food-Acquisition Stress	“She (mom) used to go to this church and get what they used to call a bread box or a bread basket. Every other Thursday they would give out of it... if you need a little extra food.” [Mom would say,] ‘Oh, I haven’t been to the grocery store yet,’ and I think that might have been code for ‘We have to wait until payday’ but there was always something, there was always food. It might not have been what we wanted [laughs]...Of course, my aunt and my uncle, they didn’t have kids until later, later in life. And I can remember them bringing... You know, I didn’t think much of it then when I was a kid, but I can remember my aunt making soup or spaghetti or something... now that I’m an adult and I think back on it, and I’m thinking mom ran out of money, you know.” “was a time in my life, my dad, he was in an accident and he had a motorcycle wreck so he didn’t get to work. And at that time, it seemed like we did not receive any food stamps or anything like that. So we did eat a lot of beans and lots of tortillas and gardening food, but there was times that we did not get to have steak. We did have something like beans and chicken, something that was cheaper, and I do remember that. And there was time, there was a time that we did fix the chicken, we did get one piece, during that time.” “Something we did not have before. We can do it now [afford meals]. We don’t waste food either or our money.“	- We changed the way we ate when money was tight. - There was plenty of food in my home (reverse coded).
Current Food-Acquisition Stress	“How can I buy a week’s worth of groceries on $40?” “When my, my boys were little -- they were little so we I worked part time, which didn’t even equal 20 hours a week. And we had to make it on what I was making to pay everything. So we had lots of beans and rice and macaroni and cheese and things that would go a long ways, but didn’t cost a whole lot...”	- I made food choices based on what we can afford. -We could afford all the foods we enjoy (reverse coded).

Memories of Food Acquisition Stress

Narratives provided rich detail about experiences with varying degrees of food insecurity. At the least extreme, participants reported strong household efforts to avoid food waste, changing their diets at the end of the month, and eating often with family members when they were children. Quotes supporting development of this theme are found in Table 1. Other participants discussed a common sub-theme which was inability to afford desired foods and a necessity to eat foods perceived to be of lower quality (e.g., “macaroni and cheese (17)).”

Current Food Acquisition Stress

Many educators continue to experience various levels of food insecurity and FAS as adults. One educator described how she “fills up on water” the last two weeks of the month because the “food that’s in the house is for the kids (4).” Additional quotes are found in Table 1.

Quantitative Results toward Survey Development

Survey Items

Survey items in Table 2 reflect the language used by educators to describe their experiences. We also adapted items from the HFSSM to be retrospective reports of food insecurity; these items mirrored questions about current experiences to as about childhood experiences. The items we developed are compared to USDA published items in Table 3.

Table 2.

Items Means and Standard Deviations For Scales and Items

Memories of Child Food Insecurity
As a child, how often were the following true: (Never = 0, Always = 4)	Mean	Standard Deviation
The food that your family bought just didn’t last, and you didn’t have money to get more.	0.71	0.93
People in your household cut the size of meals or skipped meals because there wasn’t enough money for food.	0.51	0.84
Memories of Food-Related Stress	1.49	0.88
We changed the way we ate when money was tight.	1.60	1.17
Wasting food was a big deal with my family.	2.10	1.46
There was plenty of food in my home. (reverse coded)	0.93	1.13
There were times we had to go without foods that we enjoyed because we could not afford them.	1.23	1.13
We could afford steak when we wanted. (reverse coded)	2.21	1.23
Current Food Related Stress	1.57	0.86
In the last 12 months, how often were the following true: (Never = 0, Always = 4)
We change our diet at the end of the month when money was tight.	0.92	1.05
We made sure we do not waste food.	2.67	1.21
We could afford all the foods we enjoy. (reverse coded)	1.46	1.09
I made food choices based on what we can afford.	2.34	1.23

^*Cronbach’s alpha values reported for scale after removal of food waste items, which were not internally consistent with other items for either scale.

Table 3.

Comparison of Items to Screen for Food Insecurity and Food Acquisition Stress

Items from USDA Household Food Security Survey Module (Short Form)*	Items from Current Food Acquisition Stress *
1. “The food that (I/we) bought just didn’t last, and (I/we) didn’t have money to get more.” Was that often, sometimes, or never true for (you/your household) in the last 12 months?	1. We change our diet at the end of the month when money was tight.
2. “(I/we) couldn’t afford to eat balanced meals.” Was that often, sometimes, or never true for (you/your household) in the last 12 months?	2. We could afford all the foods we enjoy. (reverse coded)
3. Did (you/you or other adults in your household) ever cut the size of your meals or skip meals because there wasn’t enough money for food? (Yes, No, Don’t Know)	3. I made food choices based on what we can afford.
4. [IF YES ABOVE, ASK] How often did this happen—almost every month, some months but not every month, or in only 1 or 2 months? (Almost every month, Some months but not every month, Only 1 or 2 months, Don’t Know)	Items from Memories of Food Acquisition Stress*
4. Did you ever eat less than you felt you should because there wasn’t enough money for food? (Yes, No, Don’t Know)	1. We changed the way we ate when money was tight.
5. Were you every hungry but didn’t eat because there wasn’t enough money for food?	2. There was plenty of food in my home. (reverse coded)
	3. There were times we had to go without foods that we enjoyed because we could not afford them.
	4. We could afford steak when we wanted. (reverse coded)

^*Both measures ask respondents to think about the last 12 months. The USDA measure uses the response options as provided above. The Food Acquisition Stress Measure uses a 5 point scale of with Never = 0, Always = 4.

Sample

Participants in the surveys were largely educators (N = 76.0%) as well other adults attending the training (e.g., foster parents, volunteers, directors, site owners, N = 24.0%) and represented a sample that was 96.2% female. The majority of the respondents were White (64.1%) or Black (26.3%). The mean age was 40.7 (SD =15.6) with 10.9 years of working experience (SD = 10.0). Annual household income for participants was largely at or below $34,999 (59.7%); 24% of participants reported an annual household income between $35,000 and $64,999; and 16.4% reported a household income greater than $65,000. Approximately 12% had a Bachelor’s degree or higher. About a third (34.1%) had at least some college; 12.5% had an Associate’s degree; and 37.3% reported high school or GED as their highest level of education. Participants were from 68 of the 75 counties in a rural state.

Food Security Status

For the items related to current food security, the overall raw score to the 10 items ranged between 0 and 10 with a mean of 1.5 and a standard deviation of 2.3. Based on responses to the 10 items, 56.4% reported high food security; 21.3% reported marginal food security; 13.7% reported low food security; and 8.5% reported very low food security. The two-item retrospective assessment of childhood food security identified 22.2% of the sample with food insecurity in childhood.

Properties of the Food Acquisition Stress Tool

Initially, the Memories of Food Acquisition Stress scale exhibited a Cronbach’s alpha of 0.71 using all items in Table 2; this value increased to 0.74 if the item about food waste was deleted from the scale. The Cronbach’s alpha for Current Food Acquisition Stress scale was 0.60 with all items appearing in Table 2; this increased to 0.65 after deleting the item about food waste. Internal consistency remained acceptable without inclusion of the item about affording steak (.67), and we acknowledge that research in diverse cultures may choose not to include this item reflective of cultural, religious, or dietary preference. Average scores for both scales exhibited an observed minimum of 0 and an observed maximum of 4. Scale means and standard deviations are included with item means in Table 2.

Binary indicators of FAS were created that mirror the logic for the 2-item food insecurity screen developed by Hager et al.³⁹ That is, if a subscale had a mean with a value of 2 or greater (indicating sometimes or more frequently, on average), then FAS was said to be present, and a value of 1 was indicated. This was completed for both memories of and current FAS. By this classification, 33.1% of participants indicated Memories of Food Acquisition Stress, and 38.3% indicated Current Food Acquisition Stress.

Relationship Between Food Acquisition Stress, Food Insecurity, and Food Waste

Box plots (Figures 1 and 2) present the stair-step relationship between food security status groups and FAS. Specifically, there is a stepwise relationship between the HFSSM categories of food security and both Memories of Food-Related Stress and Current-Food Related Stress which appears steeper for Current-Food Related Stress. In addition, chi-square statistics tested for the association between Food Security Status and FAS Status at both time points and for the relationship between past and current FAS.

Figure 1.

Current Food Security Status and Memories of Food-Acquisition Stress

Figure 2.

Current Food Security Status and Current Food-Acquisition Stress

Memories of Food Acquisition Stress was significantly related to Childhood Food Security Status with those who reported food insecurity in childhood much more likely to report FAS in childhood (χ² = 340.10, p<.001). In addition to this expected overlap, 19.4% of participants who were food secure in childhood reported Memories of Food-Related Stress. For Current Food-Related Stress, a similar pattern was noted. Current Food Insecurity was related to Current Food-Related Stress (χ² =287.77, p<.001); yet a relatively large proportion of participants who had Current Food Acquisition Stress met the designation of Food Secure (51.7%). Memories of Food Acquisition Stress were also related to Current Food Acquisition Stress (χ² = 127.62, p<.001).

Across all Food Security Status levels, item means for wasting food were consistent and relatively high. For memories of food waste, item means were above 2 for all groups and approximately 3 for the very low food security status group. For current avoidance of food waste, item means were approximately 3 for all groups. The corrected item-total correlation for past food waste item (r=.31) and current food waste item (r=.23) were respectively the lowest on each scale. Although these items were related to food stress, it seems that the value of not wasting food was also endorsed by those who are/were food secure. Thus, food wastes items were not useful for distinguishing among participants.

Discussion

The current mixed methods study presents development and preliminary data on a new tool to measure stress related to food acquisition. Qualitative interviews provided a rich data source from which to develop items likely to be salient to persons with similar experiences. These lived experiences were leveraged to develop the resulting 7-item Food Acquisition Stress Tool with 2 scales (Table 2). The Food Acquisition Stress Tool demonstrated relatively strong internal consistency, no indications of floor or ceiling effects, and minimal concern with careless responding (2.7%). The stair-step relationship between FAS and food insecurity lends validity to the tool, which has the ability to capture the nuance of FAS not embedded in current assessments of food insecurity. The Food Acquisition Stress Tool has potential use to complement existing tools and to identify individuals that may not reach the threshold for being identified as food insecure.

Using this tool, most food insecure individuals also indicated food acquisition stress (85%); a high proportion, as would be expected. However, we also identified that most individuals who had food acquisition stress did not meet criteria for food insecurity (61%). In the short-term, stress has been associated with different eating patterns including increased fat intake and eating in response to stress.⁴⁰ Persistent stress can contribute to increased cortisol, higher blood pressure, glucose intolerance, and insulin resistance.^41,42 Thus, capturing stress about food acquisition, even among individuals who are currently categorized as food secure, has the potential to aid in identification of individuals who may be at greater risk and in need of supportive interventions.

Our study also provides information about memories of childhood experiences with food insecurity and stress about food acquisition. Childhood food insecurity (adapted retrospective HFSSM) was significantly related to memories of food acquisition stress as well as current food acquisition stress. Several participants (18.8%) did not screen as food insecure in their childhood but reported stress about food acquisition in childhood. Further, memories of food acquisition stress were strongly related to current food acquisition stress, which suggests early life experiences may contribute to food-related stress patterns into adulthood. Regardless, stress in childhood has life-long effects. For example, studies investigating ACEs and chronic early life stress have consistently illustrated the relationships between childhood experiences and adult behaviors and health outcomes.^43,44 At present, memories of food insecurity are included in the ACEs questionnaire (i.e., You did not have enough to eat) as an indicator of neglect along with other possible indications of neglect (e.g., You had to wear dirty clothes; There was someone to take you to the doctor if you needed it). Future studies could explore if experiences around food insecurity and food stress in childhood have unique impacts on development compared to other adverse experiences. Interventions to elucidate and address memories of food acquisition stress could contribute to formation of healthier habits and reduced stress in adulthood.

An additional interesting finding is the relative high endorsement of items about food waste across the sample, regardless of food insecurity and food acquisition stress. This finding may reflect societal values related to food waste. That is, items about food waste could elicit participant responses about real familial rules and norms of prevention of food waste or induce social desirability in responses to be consistent with current societal trends toward emphasizing more sustainable living.⁴⁵ This is consistent with prior research which showed a primary motivation for minimizing food waste was the prevention of negative emotions (e.g., guilt, regret, embarrassment).⁴⁶ Thus, although participants in our qualitative sample reflected on the power of avoiding food waste in their own lives, related quantitative survey items were not useful for the purpose of distinctions about experiences with FAS.

Our study included a high portion of early childhood educators. Previous investigation of food security among early childcare educators identified that between 34 % and 46% are food insecure, depending on their role (lead vs assistant teacher) and place of residence.^47–49 This study illustrates that even greater numbers experience FAS. This is concerning given that recent studies have shown that the experiences of educators can translate into classroom feeding and nutrition promotion practices.⁵⁰ There are also policy implications for early education based on this work. Current estimates suggest a $1 investment in early childhood education is estimated to yield a $4 - $12 return including better academic outcomes for children in K-12 education, reduction in department of corrections cases throughout the lifespan, higher graduation rates, and higher earnings for those who participate in high quality early childhood education.⁵¹ One implication of the current study is the opportunity to invest in the workforce of early childhood education. Options such as higher wages to workers and feeding providers on the job could help to alleviate FAS and influence practices in their own families and classrooms.

Although this study was conducted in a sample largely comprised of early childhood educators, several segments of the US workforce may be more subject to FAS. For example, early childhood educators are a workforce of over 550, 000 who are paid $22,290 per year on average.¹⁹ Further, while unique in some ways from other workforces (e.g., primarily female, limited education requirements), the early childhood education workforce has similarities to others fields such as health care support (e.g., nursing assistant, home care aids), service industry workers (e.g., servers), and facilities maintenance whose occupations are also marked by low median income, high turnover, and above average health risks.^23–25 Thus, translation of research described for this current project to other audiences has the potential to increase understanding in other workforces as well.

The measurement of FAS may be useful in clinical settings as well. Previous studies have advocated food-insecurity screening for all patients.^52,53 Pairing food security status with information on FAS could provide a rich picture of patients’ dietary environment. Given that even low-levels of food insecurity are associated with chronic health conditions,⁵⁴ identification of FAS may allow clinicians to provide referrals and community resources to patients to prevent development into further food insecurity and hunger. For example, community interventions have shown improved health behaviors and outcomes for patients with food insecurity,^55,56 and health care organizations have been successful in improving patient’s Hba1c levels, blood sugar levels, cholesterol and triglycerides by proving free, healthy food and diabetes-education classes to those who are food insecure.⁵⁷ Individuals with FAS may benefit in similar ways.

This study has both limitations and strengths. Although the lack of gender diversity in our sample may limit generalizability to other populations, it is representative of early childhood educators in the US (between 91% and 98% are women). In order to make comparisons to other, more diverse workforces, additional data should be gathered. This study also did not collect any health outcome data to understand potential associations between FAS and BMI, diabetes, or other chronic disease, which would be of interest for future work. Future studies could also explore differences in FAS among different demographic groups and those participating in the Supplemental Nutrition Assistance Program. A strength of the study is the mixed methods approach, which allowed for the examination of generalizability of themes found in qualitative work. Further, we collected a relatively large number of participants which exceeds recommendations for the number of cases per new item of survey development.⁵⁸

To date, the full range of food stress experiences and adjacent stressors are largely unexplored, particularly through approaches that reflect lived experiences relative to food and stress. Further, under-identification of households affected by FAS due to “discrete cut points” in previous research may have missed a group of individuals who are at risk for economic stress/hardship that affects them in important ways across the lifespan. Gaining knowledge related to the experience of food insecurity through use of existing gold-standard tools, while beneficial, may not go far enough in identifying those individuals and families who are at high risk for experiencing poverty-related health outcomes and stressors. Widening the considerations to include those who may have a history of stress related to feeding themselves or their families, (either from childhood or other previous experience) can contribute to the growing body of knowledge surrounding outcomes of a strained income. This information may also offer opportunities for educational intervention as well as contextual understanding regarding potential difficulties experienced by educators when interacting with children in their care regarding food.