Introduction

Food insecurity (FI) is the lack of consistent access to safe and healthy food [1]. In 2020, the prevalence of FI among households in the US was estimated to be 10.5%, which was down from 11.1% in 2019 [2]. Comprehensive national data since the start of the COVID-19 pandemic are not yet available, but local data indicate that the prevalence of FI has increased, particularly among the newly food insecure as a result of COVID-19 [3]. Among the US population, pre-COVID-19 rates of FI among college students range from 10 to 75% [4].

While the general college population has been widely studied, less is known about the prevalence of FI among collegiate athletes. One non-peer-reviewed source on this topic is the RealCollege survey of students at 4-year and 2-year institutions in the US conducted by the Hope Center for College, Community, and Justice [5]. The most recent survey in fall 2019 aggregated data for student-athletes for the first time. “Student-athlete” was self-reported as participating in their institution’s athletic department. Of the almost 167,000 total students who completed the survey, 1866 were student-athletes at 4-year institutions and 1640 attended 2-year institutions. Prevalence of FI among athletes at 4-year institutions and 2-year institutions was 23% and 39%, respectively. For non-athletes, these rates were 32% and 43%, respectively. This report, as well as the ripple effects from the pandemic, have spurred increased interest on the topic of FI among college students in general and student-athletes specifically. The purpose of this review is to examine the prevalence of FI among the general college student population and student-athletes, also taking into consideration the impact of the COVID-19 pandemic on FI in college students.

Search Strategy and Selection Criteria

After conducting preliminary searches to identify key words, a systematic search of PubMed, SportDiscus, and Medline (using EBSCOhost Research Platform) was conducted, and all were screened for original, primary research articles published in the last 5 years since 2016. The search strategy was based on sensitivity and did not include terms related to comparators or outcomes in order to allow for the potential retrieval of a larger number of articles. The following search terms were used to search the databases with filters for human subject research and English language applied without restriction on study design: “food insecurity” and “food security” in combination with “college” or “athlete.”

A defined set of eligibility criteria was constructed with the following inclusion criteria: a population of college students in the US, use of a validated tool for food security (FS), results that included a reported overall prevalence of FI, clear methodology that could confirm all inclusion criteria, and the article written in English. Exclusion criteria included not meeting the previous inclusion criteria and/or ambiguous methods that could not confirm all inclusion criteria.

Data were extracted independently by three researchers and included study objective, sample size, and population characteristics (gender, race/ethnicity, 4-year or 2-year institution, non-athlete or athlete status, undergraduate or graduate status, first-generation college student, assessment tool utilized, time frame of assessment, study design, and results and outcomes). The results and outcomes data from each of the studies included an overall % FI, % low and very low FS when available, and factors associated with FI. Full data can be found in Tables 1 and 2.

Table 1 Summary of research studies on food insecurity in college students
Full size table
Table 2 Food insecurity prevalence levels
Full size table

Results and Discussion

Searches of the databases resulted in a total of 801 studies. After screening of the titles and abstracts and excluding review articles and duplicates, 62 articles remained. After a more careful screening of the full texts, 16 were excluded for not meeting the inclusion criteria. One additional publication became available during the preparation of this manuscript and was subsequently included in the data and discussion. Therefore, a total of 47 articles that met the inclusion criteria were analyzed and evaluated [6–9, 10••, 11–17, 18•, 19, 20•, 21•, 22•, 23–28, 29•, 30–37, 38••, 39–52] (see Table 1).

Study Characteristics

Of the 47 studies, 44 were cross-sectional research [6–9, 10••, 11, 15–17, 18•, 19, 20•, 21•, 22•, 23–28, 29•, 30–37, 38••, 39–52] and three were cohort design [12, 13, 44]. Two of the studies were performed at 2-year institutions [25, 46] with the other 45 at 4-year institutions [6–9, 10••, 11–17, 18•, 19, 20•, 21•, 22•, 23, 24, 26–28, 29•, 30–37, 38••, 45, 47–52]. All but two of the studies were focused on the general college student population (non-athletes) while the other two were focused on student-athletes [10••, 38••]. For the administration of the assessment tools, in 41 studies an online version was utilized [6, 8, 10••, 11–17, 18•, 19, 20•, 21•, 22•, 23–28, 29•, 30–37, 39–45, 49–52], a paper version in two studies [9, 38••], and both in one study [7]. There were three studies where the mode of administration was not stated [46–48]. In all but three studies [13, 17, 25] some version of the USDA Economic Research Service’s survey tool was used to assess FS [53]. The USDA has three validated surveys that categorize FS levels: the 18-item Household Food Security Survey Module (HFSSM), the 10-item Adult Food Security Survey Module (AFSSM), and the 6-item AFSSM Short Form. Within the HFSSM, there are specific questions about FS status of children in the household, while these questions are omitted from the other two surveys. In the majority of the assessments, participants were queried about FS in the past 12 months [6–9, 10••, 12, 14, 15, 17, 18•, 19, 20•, 21•, 22•, 23, 24, 26–28, 29•, 30–34, 37, 39–43, 45, 46, 52]. In one, FS status was assessed over the past 9 months [35], and in the remaining studies, FS status was assessed over the past 30 days [11, 16, 36, 47–51]. No time frame was specified in four studies [13, 25, 38••, 44]. FS status was reported on a continuum from high FS to very low FS. A person is considered to have FI if they have “low” or “very low” FS.

Food Insecurity Among the General College Student Population

The overall prevalence of FI from all subjects combined in this current review reveals a much higher prevalence of FI in college students at 32.2 ± 12.9% (range 9.9–72.9%) compared to the national household average in the USA in 2019 (10.5%) [2]. The overall prevalence was obtained by the sum of all published prevalences from each of the studies included in the review, divided by the total of studies included in the review, and then multiplied by 100 to yield a percent. Breaking down the results from the current review, the prevalence of FI at 4-year institutions and 2-year institutions was 32.7 ± 12.9% (range 9.9–72.9%) and 37.8 ± 20.1% (23.36–52%), respectively. These are in line with the results from the Hope Center survey (32% from 4-year institutions and 43% from 2-year institutions) [5]. When comparing the results based on the assessment tool, it appears that the 6-item (35.3 ± 11.7%; range 9.9–48.0%), 10-item (30.3 ± 10.1%; range 9.9–48.0%), and 18-item (33.8 ± 18.5%; range 15.0–52.0%) surveys all yielded similar results while the 2-item (43.7 ± 25.9%; range 23.6–72.9%) and “other” (52.8%) yielded results that were much higher.

Factors Associated with Food Insecurity in College Students

Demographic, Sociocultural, and Socioeconomic Factors

Several factors that were investigated in the 47 studies were correlated with the level of FS reported. Race and ethnicity are factors that are associated with a higher risk of FI. When analyzing the data from the current review for studies with overrepresentation of Hispanics and Black/African Americans, the combined % FI is 35.0 ± 14.8% (range 14.0–72.9%). The highest prevalence of FI (72.9%) was in the study by Duke et al. performed in a population of students from a historically Black college or university [17]. Only El Zein et al. [18] specifically reported on FI prevalence among international students and noted that it was higher (37.6%) than the entire student population (32.0%) at one university, though this was still significantly lower than the rate for Black students (61.7%) [19].

Other demographic factors identified as being associated with increased FI were the female sex/gender, being a single parent, and having dependent children. In an analysis by the Institute for Women’s Policy Research [54], 22% of all undergraduate students are parents. Among students raising children while in college, 70% are mothers, and of these, 62% are single parents. In the US, single mothers with children in the household have the highest rates of FI (27.7%) of any household composition [55]. Considering this data and the fact that a larger proportion of the college student population is female (60%) [2] may provide at least some explanation for the higher rates of FI among college students compared to the general population.

Educationally related factors associated with FI include being a first-generation college student, parental education of high school level or below, and having undergraduate status. First-generation college students are more likely to come from low-income, poor, or working class backgrounds (57%) compared to continuing generation students (12%) [56]. No direct cause is known for why graduate students have lower rates of FI than undergraduate students–are more resources available to graduate students than undergraduates, or are they just more likely to come from more financially stable backgrounds? Socioeconomically, students receiving financial assistance (specifically Pell Grants) had higher rates of FI. Considering access to food, it is not surprising that having a limited meal plan or no meal plan was associated with increased FI. Lack of accessibility to food may also potentially explain why students living off campus and those who do not have access to a car reported increased levels of FI. Finally, food security status may vary for students depending on the time of the semester. The prevalence of FI was higher at the end of the fall and spring semesters compared to the beginning when financial assistance is greater and campus meal plans have not been depleted.

Food Insecurity and Outcomes

The authors of the reviewed literature identified many relationships between FI and specific outcomes among college students. To begin, the outcome most notably unique among this group compared to the general population is academic performance. Higher rates of FI among college students were consistently associated with lower grade point average. There is evidence that dietary behaviors (e.g., eating breakfast) and quality (e.g., increased fruit and vegetable consumption) are related to better academic outcomes [57]. In this review, FI was related to dietary behaviors and quality such as increased frequency of skipping meals; lower fruit, vegetable, and fiber intake; and increased intake of added sugars, processed foods, fast food, dairy, and calcium. Conversely lower FI was associated with higher rates of nutrition literacy, which has been shown to predict adherence to healthier dietary patterns [58]. In regard to FI and other health outcomes, college students experiencing FI were more likely to report poor overall health than students not experiencing FI. These health metrics included less frequent medical visits/exams, poor mental health, increased psychological stress, increased depression, poor sleep quality, disordered eating/eating disorders, increased obesity, less physical activity, increased tobacco and marijuana use, and higher rates of binge drinking. College students experiencing FI employed various coping strategies to meet their food needs such as more frequent attendance at events that offered free snacks and meals, increased use of food pantries, greater frequency of skipping meals, decreasing portion sizes, stretching the available food over a longer period of time, and purchasing low-cost, highly processed foods. While it remains out of the scope of this paper to offer specific solutions to address FI on college campuses, strategies currently being employed include on-campus food pantries and food recovery from campus events directed to students.

Food Insecurity Among Collegiate Student-Athletes

There are only two published studies in peer-reviewed journals that have assessed the prevalence of FI among collegiate student-athletes [10••, 38••]. The average prevalence of FI for student-athletes (see Table 2) was 12.3 ± 3.4% (range 9.9–14.7%) much lower than the general college student population and more in line with the national average. However, this value is much lower than the results of the Hope Center survey for student-athletes (23% from 4-year institutions and 39% from 2-year institutions) [5]. One explanation for this difference could be that there was a wider variety of schools from across the country in the Hope Center survey compared to the two studies where the samples were from one Division I program and Division III athletes from schools nationwide. Poll et al. studied the relationship between FI and disordered eating in a population of male NCAA Division I student-athletes (mostly football players) at a single institution in the southeastern USA and found a 9.9% prevalence of FI [38••]. They also queried the participants about their FS in high school, and 13.5% reported being FI at that time. Collegiate FI status was significantly associated with preoccupation with food and keeping food or hiding food but not with binge eating. As observed in the general college student population, preoccupation with food was significantly correlated with having experienced FI before coming to college (i.e., in high school). No correlations between FI and race/ethnicity were reported in this study.

Brown et al. reported a 14.7% prevalence of FI over the previous 12 months in a nationwide sample of NCAA Division III student-athletes [10••]. The prevalence of FI among subjects who self-identified among the following racial/ethnic groups was 13.3% for white, 18.3% for Hispanic, 31.0% for Black/African American, and 8.5% for Asian. The level of FI was higher for those without a meal plan, those receiving a Pell grant, first-generation college students, and those who have previously experienced FI. Subjects that experienced FI self-reported a negative impact on academic and athletic performance. Contributing factors to FI reported by student-athletes were dining hall hours conflicting with practice times (45.4% of subjects), games during dining hours (22%), and living off-campus with limited money (14.6%).

Since there are so few peer-reviewed, published articles on FI and student-athletes, it is worth discussing a master’s thesis [59] and dissertation [60] on the topic that, although not published, have undergone a rigorous review process. In a master’s thesis by Anziano [59], 18 NCAA Division II student-athletes were interviewed about their level of FS and 61.6% (n = 11) reported having some level of FI which is much higher than in the published work by Poll et al. [38••] and Brown et al. [10••]. It is important to note the small sample size of this qualitative thesis study and consider possible self-selection bias of participants. Contributing factors to FI that were identified by Anziano [59] were limited access to transportation, limited cooking facilities, lack of time to cook or prepare meals, and lack of food options in the dining hall [59]. Outcomes for those experiencing FI were similar to those found in Brown et al. [10••] including self-report of a negative impact on athletic performance. Identified coping strategies were similar to those found in this review mentioned previously.

In a dissertation by Misener [60], 424 Division III student-athletes on 22 teams from a primarily white, liberal arts college in Western Pennsylvania completed a survey to assess the prevalence of FI. Using the 6-item AFSSM Short Form, 129 (29%) were found to experience some level of FI, which was higher than Brown et al. [10••], who also surveyed students at DIII schools. The student-athletes in Misener’s research specifically reported cutting the size of their meals or skipping meals because they ran out of money. Twenty-six percent reported eating less than they felt they should have in the last year because they did not have enough money or food. The teams with the highest percentage of FI were women’s cross country (80%) and men’s field athletes (67%). The reason 23.8% reported not eating a meal is because they ran out of meal swipes. About one-third (29.7%) reported being unable to afford balanced meals.

Impact of COVID-19 on Food Insecurity in College Students

The impact of COVID-19 on FS of college students was only evaluated in three studies [13, 16, 44]. Davitt et al. assessed food security over 30 days during the spring of 2020 and determined that 17.0% of respondents were classified as FI [16]. Soldavani et al. looked at specific changes in FS status and found that 12% had improved status, 68% unchanged, and 20% became worse [44]. In contrast, Christensen et al. reported no differences in prevalence of FI pre-COVID-19 compared to during COVID-19 [13]. In the two longitudinal studies, the sampled populations were predominantly white and female students at single institutions (Soldavani et al. [44] in the southeastern USA and Christensen et al. [13] in the Midwest) prior to the start of the COVID-19 pandemic and the spring of 2020. The timing of the surveys between these two studies may help explain this discrepancy in that Soldavani et al. [44] surveyed students in June and July of 2020, while Christensen et al. [13] surveyed students earlier in April 2020. In addition, different FS assessment tools were used for these studies. Soldavani et al. [44] reported that moving in with family and receiving financial support from family appeared to have a positive impact on students’ FS status.

Limitations

The lack of published peer-reviewed studies on collegiate student-athletes prevents conclusions from being made on this population. Student-athlete resources vary widely across athletic division and location leading to variability in the limited data available. In the general college student studies, the validity of the assessment tools available has been questioned. Ames et al. used the 10-item AFSSM in their assessment of 478 students [7]. They concluded that this survey may lead to inaccurate results in college students due to the questions relating to household FS which may not be appropriate for the specific living situations of the college population. This survey tool was used in 24 of the reviewed studies (66% of total subjects). Self-selection bias could also influence the results as students who are FI may be more likely to participate in a study on this topic. Lastly, our convenience sample was primarily white and female and may not be indicative of other demographics.

Conclusion

This review of the literature further confirmed that FI exists among the general college student population. It also confirmed that there is a lack of peer-reviewed, published literature that addresses FI and collegiate student-athletes. The information gleaned from studies summarized in this article address not only the potential impact of FI on academic performance in the classroom, but also the potential impact outside the classroom. This can include the impact on athletic performance and health and wellness of student-athletes, even if a lack of scientific evidence is not yet available to support it. We found parallels between the peer-reviewed, published, and unpublished studies on both the general college student population and collegiate student-athletes implying that student-athletes are a small subset of college students that have the same challenges, if not more, around FI. The long-term implications of COVID on the prevalence of FI on college campuses are yet to be explored. Future studies will need to assess the effectiveness of initiatives at decreasing the prevalence of FI on college campuses and factors that may contribute to FI among specific segments of the college population such as student-athletes.