Exploring reasons for high levels of food insecurity and low fruit and vegetable consumption among university students post-COVID-19

High rates of food insecurity and low consumption of fruit and vegetables among university students have been observed prior to the COVID-19 pandemic and intensified during the pandemic. This study aimed to investigate food insecurity among university students and its associations with sociodemographic factors, fruit and vegetable consumption behaviours, and preferred campus programs to address these issues. A convenience sample of 237 Australian university students completed a cross-sectional online survey from October to December 2022. Food insecurity was assessed using the 10-item US Adult Food Security Module, applying the Canadian classification scheme. Sociodemographic variables, fruit and vegetable consumption behaviours, and perceptions of fruit and vegetable access and their affordability were included in the survey. Students were also asked to select the most suitable program(s) and provide reasons for their choice using open-ended questions. Approximately half of respondents (46.4%) were identified as food insecure. The proportion of students meeting the recommended intake of vegetables as specified in the Australian Dietary Guidelines was very low (5.1%) compared with fruit (46.2%). Low fruit consumption was significantly associated with food insecurity (OR = 1.81; 95%CI 1.03, 3.18, p = 0.038). Factors such as the perceived lower accessibility and higher price of fruit and vegetables were significantly associated with higher odds of food insecurity. In terms of potential programs, a free fruit and vegetable campaign was the most popular program, with affordability and physical access being the most frequently cited reasons. These findings suggest that food insecurity is associated with low fruit and vegetable consumption in university students. Therefore, transforming campus food environments and developing food policies at the university level must be considered to address food and nutrition security in university students.


Introduction
The prevalence of food insecurity internationally has substantially increased since 2019-20.It was estimated 30% of the global population experienced food insecurity in 2022, which is 391 million higher than in 2019 (Food and Agriculture Organization, International Fund for Agricultural Development, UNICEF, 2023).According to Food and Agriculture Organization of the United Nations, food insecurity refers to the lack of regular access to adequate safe and nutritious food to maintain an active and healthy life (Food and Agriculture Organization, International Fund for Agricultural Development, UNICEF, 2023).Food insecurity is often viewed as a continuum that begins with uncertainty about obtaining food and reaches its more severe level when there is no food for a day or more.These conditions can be determined by four hierarchical dimensions, namely food availability, food accessibility, food utilisation, and stability (Gross et al., 2000).
Food availability refers to the physical existence of food with sufficient nutritional quality (Simelane & Worth, 2020).However, food availability does not necessarily determine the ability to access food.Food accessibility consists of two fundamental aspects, physical and financial accessibility, to achieve and maintain optimal nutritional status (Simelane & Worth, 2020).Food accessibility can be attained when an individual or household has the ability to physically obtain food and pay for the food to meet dietary guidance.Food utilisation is commonly referred to as the ability to use, prepare, and transform food into safe and nutritious food to meet physiological, sensory, and cultural needs (Gross et al., 2000;Simelane & Worth, 2020).These needs are often determined by various elements, including knowledge, skills, food preparation and storage, cooking facilities, and time availability (Gross et al., 2000;Simelane & Worth, 2020).Without the last dimension of 'stability', the first three dimensions can be challenging to achieve and maintain over time.Stability refers to the temporal determinant of food security, which is categorised into chronic (long-term) and transitory (short-term and temporary) food insecurity (Gross et al., 2000).Chronic food insecurity occurs when individuals or households are unable to fulfill their minimum food needs consistently over an extended period, whereas transitory food insecurity occurs when there is a sudden decline in the capacity to produce or obtain sufficient food to maintain optimal nutritional status (Food and Agriculture Organization, 2008).All dimensions must be fulfilled to ensure food security and, ultimately, nutrition security.
A report from the Australian Institute of Family Studies revealed that the estimated prevalence of food insecurity in the Australian population was 4% in 2011-13 and then increased to 13% in 2016-18 (Bowden, 2020).Despite this, there is a lack of regular monitoring of food security at the population level in Australia, which may overlook high prevalences in vulnerable groups.Indeed, there is a growing body of literature that shows that the prevalence of food insecurity is higher among university students than in the general population (Bruening et al., 2017;Ellison et al., 2021;Loofbourrow & Scherr, 2023).Prior to the COVID-19 pandemic, food insecurity and its prevalence among university students was a concern (Bruening et al., 2017;Gallegos et al., 2014;Nazmi et al., 2019), however it emerged as a critical issue during the pandemic because university students were strongly affected due to job losses and financial hardship (Hastings et al., 2021;Jessup et al., 2022;Morley et al., 2023).
Studies have reported the prevalence of food insecurity across Australian university students during the pandemic ranged between 32% and 48% (Bennett et al., 2022;Dana et al., 2023;Kent et al., 2022;Mihrshahi et al., 2022).It was estimated that up to 50% of food insecure students reported experiencing very low food security (Dana et al., 2023).International university students were reported to have higher estimated prevalence of food insecurity than domestic students with prevalence worsening due to lack of government support, such as pandemic-related financial assistance payments, JobSeeker, and Job-Keeper, during the first year of the pandemic (Bennett et al., 2022;Dana et al., 2023;Kent et al., 2022;Mihrshahi et al., 2022).A large cross-sectional survey at an Australian university showed that the risk of food insecurity was twice as high for international students compared to domestic students (Kent et al., 2022).Our previous study also found that international students are not only more vulnerable to food insecurity but also reported large disruptions to their studies and a worsening of their mental wellbeing compared to domestic students (Mihrshahi et al., 2022).The ongoing rise in food insecurity is expected to persist due to cost-of-living crisis and this phenomenon has become somewhat normalised within this population (Bennett et al., 2022).The Foodbank Hunger Report 2023 highlights that cost of living emerged as the leading cause of food insecurity, driven by rising costs in food, groceries, energy and housing (Foodbank Australia, 2023).
A systematic review on food insecurity and dietary outcomes suggests that food insecure university students have been linked with poorer diet quality, including lower consumption of fruit and vegetables, compared with food secure students (Shi et al., 2021).Previous studies showed that diet quality can be highly influenced by food environments, which is an important factor to determine food choice (Li et al., 2022a;Sawyer et al., 2021).A population study in Australia showed generally low consumption of fruit and vegetables among young adults, with the lowest consumption among males 18-24 years (Nour et al., 2017).Furthermore, a scoping review on vegetable consumption among university students across 30 countries highlighted that the majority of students did not consume the recommended serves of vegetables by the World Health Organization (Rodrigues et al., 2019).
The campus food environment encompasses food outlets, cafés, vending machines, and food sources accessible to students who are physically attending university.As a university student, exposure to the food environment may influence diet behaviours.Low availability of healthy foods (e.g., fruit and vegetables) and higher costs compared to unhealthy options (e.g., sweet snacks and sugar sweetened drinks) are commonly observed in university food environments (Li et al., 2022a).Several studies highlight the need to increase affordable fruit and vegetable options on university campuses due to concerns about student fruit and vegetable consumption (van den Bogerd et al., 2019;Roy et al., 2019).These findings imply that the limited availability and affordability of healthy food options on university campuses could contribute to concerns about food insecurity among students.Therefore, it is important to understand the role of the campus food environment in addressing food insecurity and improving healthy eating among university students.
Limited initiatives have been implemented to address food insecurity long-term in higher education settings, with a lack of student input in planning and implementation of programs or interventions.Food pantry-based interventions were the most commonly identified approach for improving students' diet-related outcomes (An et al., 2019).However, a systematic review on the nutritional quality of food pantries suggested that the majority of the studied food pantries were unable to provide adequate amounts and types of food to sustain a balanced diet (Simmet et al., 2017).Furthermore, negative perceptions towards food pantry users could lead to reluctance in accessing the food pantry (Idehai et al., 2024).Consequently, some existing programs may be poorly designed to meet the needs of students experiencing food insecurity (Hagedorn-Hatfield et al., 2022).This study aimed to investigate the prevalence of food insecurity and its associations with fruit and vegetable consumption behaviours in university students one year after the conclusion of COVID-19 restrictions in 2021 (post-COVID-19).We also explored students' opinions about the most suitable programs to address food insecurity and improve healthy eating on campus, with the aim of providing valuable insights into interventions, strategies and policies to address food insecurity and healthy eating in Australia, and higher education settings in particular.

Design and participants
The study utilised a cross-sectional design for data collection using an online platform, Research Electronic Data Capture (REDCap) (Harris et al., 2009).All students aged at least 18 years and who were enrolled at Macquarie University were eligible to participate in the survey.Macquarie University had more than 40,000 students from 114 countries in 2022, encompassing culturally diverse backgrounds and socioeconomic status.The recruitment process took a place between October 2022 and December 2022.Participants were recruited using social media, flyers on campus, and university newsletter.A survey link or a quick response (QR) code were placed on the advertisements and when used, participants were directed to the Digital Participant Information and Consent Form (PICF).Consent was obtained prior to commencing the survey.
Additionally, students were approached on campus to participate in intercept surveys.Depending on their availability during the intercept surveys, students could opt to scan the QR code and complete the survey at their convenience or complete it immediately.The online questionnaires and intercept survey were both identical.Participation in the research was voluntary and anonymous.To increase the response rate during the exam week (started on November 7, 2022), participants who completed the survey were entered in a lucky draw to win one of four P.N.A. Dharmayani et al. gift vouchers valued at AUD50.The study was approved by Macquarie University Human Research Ethics Committee (Project ID: 11,860; Approval number 520221186039979).
The target sample size was 160 domestic and 69 international students using a power calculation for proportions (https://sample-size.ne t/sample-size-proportions/).The sample size was calculated to achieve 80% power at α = 0.05 based on the assumption that the prevalence of food insecurity between domestic and international students would be differ by 20% and the proportion of international students in the total sample would be 30%.Of 320 respondents who were interested in the study and signed the consent form, a total of 237 respondents (74.4%) were included in the analysis.Eighty-three respondents were excluded for the following reasons: not answering the survey questionnaire (n = 50), having missing data on food insecurity status (n = 32), and being under the age of 18 years (n = 1).

Food security
Student food security status was measured using the United States Department of Agriculture (USDA) 10-item of the US Adult Food Security Survey Module (FSSM) (U. S. Department of Agriculture, 2012).This standard module is considered to be the most accurate tool for gauging food security status in higher education students (Nikolaus et al., 2019).The items were modified to focus on individual rather than the household in this survey.An example item was "I worried whether my food would run out before I got money to buy more" (see Supplementary Materials).The total scores ranged from 0 to 10, with one point for each affirmative response.Internal consistency was excellent within the current sample (McDonald's ω = 0.89).Applying the Canadian classification scheme, respondents were categorised as food secure (score 0), marginally food insecure (score 1), moderately food insecure (score 2-5), and severely food insecure (score 6-10) (Men & Tarasuk, 2022).The current study applied the Canadian classification scheme to determine food security status, which categorised students with total score of two into 'moderately food insecure' instead of 'marginally food insecure'.Growing evidence supports that the 'marginally food insecure' group using the US classification should be recognised as food insecure due to similar characteristics to food insecure populations (Brescia & Cuite, 2022;Men & Tarasuk, 2022).For the purpose of analyses, data was dichotomised to 'food secure' (students with a score of <2) and 'food insecure' (students with a score of 2).

Behaviours and perceptions of fruit and vegetable consumption
Validated questions from the National Health Survey 2017-2018 (Australian Bureau of Statistics, 2018b) on fruit and vegetable consumption were used to obtain a measure of serves of fruit and vegetables consumed per day (see Supplementary Materials).The data were recoded to (<2, ≥2) for fruit intake followed by the recommended vegetable intake according to the Australian Dietary Guidelines (National Health and Medical Research Council, 2013).Due to a very small proportion (5.1%) of respondents following the recommended vegetable intake by the Australian Dietary Guidelines (National Health and Medical Research Council, 2013), the data were recoded (<3, ≥3) based on median scores, which were in line with international recommendation for vegetables (Kalmpourtzidou et al., 2020).Behaviours and perceptions of fruit and vegetables questions drew on existing questions in the NSW Population Health Survey.These included perceived adequacy of fruit and vegetable consumption, barriers to increase fruit and vegetable consumption, and perceived access to fruit and vegetables (Chapman et al., 2016;Wellard-Cole et al., 2023).For reasons to stop eating more fruit/vegetables, participants were able to only select one response.

Sociodemographic data
Sociodemographic data refers to characteristics of a population or group that are related to social and demographic factors.Sociodemographic characteristics including gender, age, indigenous status, country of birth, student status (domestic/international), employment status (full time, part time, casual, unemployed), household size, and location of residence (on-campus/residential postcode) were reported by participants.Additionally, students were asked to disclose their current degree and faculty.

Program ideas
Prior to this survey, a qualitative study was conducted to explore potential food programs on campus to improve healthy eating and food insecurity among students where 13 programs ideas were proposed (Keat et al., 2024).These included ideas such as a free fruit and vegetable campaign, regular farmer's market, food cooperative and an increased number of healthy food outlets.The program ideas were listed following a question "which program(s) do you think will help you to eat more healthily on campus?[select all that apply]" in the current study's survey.A follow up question was asked to prompt students to provide their reasons for choosing the program(s).

Data analysis
The present analyses included students who completed the survey on food security (N = 237).All statistical analyses were performed using STATA/MP version 18 (StataCorp, LP, College Station, Texas, U.S.).Sociodemographic variables were presented as frequencies (n) and percentages (%) for categorical variables, and mean and standard deviation (SD) for continuous variables.Descriptive analyses were performed to compare the differences in sociodemographic variables between food secure and food insecure students using chi-square or Fisher's exact test for categorical variables and independent t-test (parametric) or Mann Whitney U test (non-parametric) for continuous variables.
Univariate logistic regression modelling was performed to examine the association between each behaviour and perception related to fruit and vegetable consumption (e.g., daily consumption of fruit and vegetables consumption, perceived fruit and vegetable accessibility) and food security status.Multivariable logistic regression modelling was further performed to investigate associations for all factors that were found to have a p < 0.1 in the univariate regression model and adjusted for covariate variables.In the univariate analysis, p < 0.1 is often used to identify potential explanatory variables that might be associated with the outcome variables for inclusion in the final model (Bursac et al., 2008;Lang, 2007).The covariates were determined based on significant sociodemographic variables identified in the univariate analyses and previous literature (Soldavini & Berner, 2020) (age, gender, student status, and employment status).The significance level for multivariable logistic regression was set at p < 0.05.
The program ideas were treated as dummy variables.Cross tabulations with Chi-square tests examined the association between the selected program and food security status.The open-ended question regarding the reasons for choosing the program was analysed using inductive content analysis (Elo & Kyngäs, 2008;O'Cathain & Thomas, 2004) by one author (PNAD) using QSR NVivo 20.Inductive content analysis was used to create categories from the raw data without a theory-based categorisation matrix (Elo et al., 2014).The major themes were developed by formulating categories through interpretation using content-characteristic words (Elo & Kyngäs, 2008).This process included open coding, creating categories, and abstraction (Elo & Kyngäs, 2008).The other author (SM) vetted all themes developed.Any discrepancies were discussed until both authors reached agreement.

Sociodemographic characteristics
The average age of the 237 respondents was 24.1 ± 7.0 years.Onehundred sixty-two domestic students (68.4%) and 75 international students (31.7%) participated.The respondents were predominantly female (72.9%), undergraduates (61.3%), and living off campus (i.e., not in university owned accommodation) (87.4%).Sociodemographic characteristics of participants and comparisons with food insecurity status (food secure vs food insecure) are reported in Table 1.Food insecure students were significantly more likely to live in larger household size than food secure students (p = 0.011).There was no difference between domestic and international students in terms of food insecurity status (p = 0.372).

Food insecurity status
Table 2 shows the proportion of food insecurity among international and domestic students.Food insecurity among students was relatively high with 29.5% and 16.9% of students identified to be moderately food insecure and severely food insecure, respectively.Therefore, almost half (46.4%) could be identified as 'food insecure'.

Behaviours and perceptions of fruit and vegetable consumption
Applying the Australian Dietary Guidelines revealed that 46.2% of students consumed at least two servings of fruit, whereas only 5.1% consumed five servings of vegetables or more.More than half students (56.5%) perceived that their fruit consumption was 'too little' or 'don't know', whereas half of them (50.2%)perceived that their vegetable consumption was 'about right'.The main barriers to consuming more fruit were 'habit' (28.1%) and that fruit was 'too expensive' (26.4%).The main barriers to vegetable consumption were also because of 'habit' (20.1%) and students indicating that they 'already have enough in my diet' (18.4%).Behaviours and perceptions related to fruit and vegetable consumption are reported in Table 3.

Association between fruit and vegetable consumption behaviours and food insecurity status
Table 4 shows the results of univariate and multivariable logistic regression between behaviours and perceptions related to fruit and vegetable consumption and food insecurity status.Lower consumption of fruit (<2 serves) showed consistently significant associations with higher odds of food insecurity status in university students after adjusting for covariates (univariate OR = 1.97; 95% CI 1.17, 3.32, p = 0.011; multivariable OR = 1.81; 95% CI 1.03, 3.18, p = 0.038).Although a lower consumption of vegetables (<3 serves) indicated higher odds of food insecurity status, the association was not significant and further attenuated after covariates were included in the model.Students who reported that they 'strongly disagree/disagree' that they have easy access to fruit and vegetables had a significantly higher odds of food insecurity.A similar pattern was also observed for perceived adequacy of fruit consumption, but the significance was diminished in the multivariable model.Students who indicated that fruit and vegetables were 'too expensive' were at higher odds of being food insecure (OR = 4.31; 95% CI 1.91, 9.71, p < 0.001 and OR = 16.70 95% CI 5.03, 55.44, p < 0.001, respectively).

Reasons for choosing program ideas
Furthermore, a total of 152 unique responses were provided by students explaining their reasons for choosing the program ideas.Affordability and physical access to food were the most frequently reported reasons students chose particular programs across all responses (Table 5).These categories exemplify how food costs and the ability to physically obtain food play a critical role in shaping their diet and hinder students from choosing healthier food options.Other categories that emerged were food availability, knowledge and skills, and sustainability.Regarding food availability, students expressed concerns about a lack of healthy food options and very limited choices that meet specific dietary requirements (e.g., halal, vegan options) on campus.Twentyeight students explicitly showed interests in improving their knowledge and skills to cook healthier meals, including seeking consultation with a nutritionist.The sustainability category is mainly related to food co-operative and farmer market program, which can support local producers and create a supportive ecosystem.In addition, four major categories emerged related to healthy eating behaviour and habits, namely (1) nudge to eat healthier food, (2) intention to buy healthier food, (3) fruit and vegetable intake, and (4) healthy diet habit, which can be established or improved by addressing the key identified components in food insecurity (see Supplementary Materials Table S2).

Discussion
This study aimed to investigate the current prevalence of food insecurity and its association with sociodemographic and fruit and vegetable consumption behaviours among university students at an Australian university, post-COVID-19.Additionally, this study explored program ideas and sought students' opinions on which programs would best address food insecurity and improve healthy eating on campus.Approximately half of students (46.4%) reported being food insecure (moderately and severely food insecure), with 16.9% of them being severely food insecure.Food insecurity was significantly associated with low fruit consumption, perceived lack of access to fruit and vegetables, and less affordability of fruit and vegetables.In terms of program ideas, a free fruit and vegetable campaign was the most selected program by students, particularly among food insecure students.These findings highlight the fact that food insecurity remains a cause for concern post-COVID-19.

Food insecurity in domestic and international students
During the COVID-19 pandemic, several studies highlighted a significantly higher rate of food insecurity in international students compared with domestic students (Bennett et al., 2022;Dana et al., 2023;Mihrshahi et al., 2022), including our previous study in 2020 (Mihrshahi et al., 2022).However, our results show that there is no difference in the prevalence of food insecurity between international and domestic students in 2022, whereas our previous study in 2020 found a significant difference (Mihrshahi et al., 2022).Another study conducted at an Australian university observed that there was no difference in the prevalence of food insecurity between international and domestic students from late 2021 to mid-2022 (Shi & Allman-Farinelli, 2023).This may indicate that the food insecurity situation has worsened for domestic students.This may have resulted due to the cessation of the COVID-19 financial support programs by the Australian government early in 2022.Two studies indicated that these programs directly decreased financial constraints (Botha et al., 2022) and observed an increase in the living standards for most households in 2020 (Li et al., 2022b).Domestic students were eligible for these programs but international students were mostly ineligible during the COVID-19 pandemic (Klapdor & Lotric, 2022).Another possible explanation is the cost-of-living crisis in Australia resulting from the aftermath of the COVID-19 pandemic, which is driven by inflation and higher interest rates (Tsiaplias & Wang, 2023).
In our study, there was an association between the number of people in a household and food insecurity, with a greater proportion of individuals living with six or more people experiencing food insecurity.One documented coping strategy was when students share accommodation and live in over-crowded conditions to reduce rental expenses (Buckle et al., 2023;Morris et al., 2023).Although the literature that  a Following the Australian Dietary Guidelines.
b Distribution collapsed into too little/don't know and about right/too much.c Distribution of strongly disagree was too small (n < 5).
focuses on both food insecurity and housing insecurity among university students is rare, a large survey at a public university in the United States found that food insecure students were at higher risk of housing insecurity than food secure students (Robbins et al., 2022).Furthermore, a qualitative study highlights that housing and food are intertwined, posing challenge for students in prioritising between rent payments, food expenses, and other basic needs (Martinez et al., 2021).This indicates that university students are also at risk of housing insecurity as these two social determinants of health are linked and result in poorer mental and physical health.The current study applied the Canadian classification scheme to determine food security status, which categorised students with total score of two into 'moderately food insecure' instead of 'marginally food insecure'.Marginally food insecure should be treated as a distinct group separately from those with high food security (i.e., no affirmative on the food insecurity scales), considering it is detrimental to health outcomes and academic performance (Brescia & Cuite, 2022;Loofbourrow & Scherr, 2023).Growing evidence supports that the 'marginally food insecure' group using the US classification should be recognised as food insecure due to similar characteristics to food insecure populations (Brescia & Cuite, 2022;Men & Tarasuk, 2022).The classification differences may have a bearing on shaping policy and intervention programs to address food insecurity (Men & Tarasuk, 2022).

Fruit and vegetable intake among university students
In addition, this study highlights that fruit and vegetable consumption among university students remains low, and low fruit consumption is significantly associated with food insecurity.Regarding maintaining healthy eating habits, the qualitative findings identified accessibility and affordability factors as crucial among university students.However, the cost of healthy foods and drinks increased by +17.9% from 2019 to 2022, with fruit and vegetables being the major contributor to the marked increase cost (Lewis et al., 2023).In contrast, the cost of unhealthy foods and drinks increased less than healthy foods and drinks (+9% between 2019 and 2022) (Lewis et al., 2023).Our findings reflect this rise in the cost of fruit and vegetables as participants considered them as 'too expensive', and unhealthy food options were more commonly available, accessible and affordable on campus.Furthermore, students expressed the need for programs that address affordability and physical food access on campus.These findings further emphasise the importance of the four dimensions of food security, such as food availability, food accessibility, food utilisation, and stability.
Another salient finding in our research is that there is a discrepancy about the daily serves of vegetables necessary to meet the Australian Dietary Guidelines among university students.Although half of students perceived their consumption of vegetables as adequate, only 5% reported consuming vegetables as recommended (i.e., five servings or more per day for women and six servings or more per day for men).Several national studies (Chapman et al., 2016;Nour et al., 2017;Wellard-Cole et al., 2023) and international studies (Matthews et al., 2016;Teschl et al., 2018;van den Bogerd et al., 2019) also identified this discrepancy between the actual and perceived adequacy of vegetable consumption among young adults.Possible explanations are a tendency to underestimate the serves of vegetables required (Chapman et al., 2016), and social approval bias (Miller et al., 2008).In Australia, young adults remain the poorest consumers of vegetables over the past decade compared to other age groups (Australian Bureau of Statistics, 2015; Australian Bureau of Statistics, 2018a).This may not only indicate a need for improvement in knowledge, but also education to develop cooking and food preparation skills in this population (Bernardo et al., 2021;Larson et al., 2006;Mills et al., 2017).As confirmed by our qualitative findings, many students expressed their intention to enhance their knowledge of healthy eating and develop their cooking skills.

Potential campus-based food security programs
Prior to our current study, a qualitative study was conducted to explore potential campus-based food security programs.A total of 13 programs ideas were proposed in addressing food insecurity and improve healthy eating by food insecure students (Keat et al., 2024).Of these programs, a free fruit and vegetable campaign was the most popular choice, particularly among food insecure students, followed by discounted meal deals and increased healthy food outlets.Furthermore, affordability and physical food access were most frequently mentioned factors in selecting a program aimed at addressing food insecurity and improving healthy eating.The shift of food provisioning in most Australian universities to a more commercial, privatised system, coupled with the scarcity of subsidised food, may have contributed to these inequalities and has further impacted food insecurity among university students (Jeffrey et al., 2022).A systematic review on food environment interventions in higher education settings found that implemented interventions were often a single intervention in short-term period (Roy et al., 2015).This review suggested that employing multiple strategies (e.g., price incentives and increased availability of healthy foods, along with nutrition information) in higher education settings may lead to optimal nutrition-related outcomes.Another review on effectiveness of subsidising healthier foods suggested significant increases in the purchase and consumption of healthier foods, which may modify dietary behaviour, yet little is known about the impact on total diet (An, 2013).Although some studies showed promising results, further studies are required to evaluate the long-term effectiveness of environmental and combinations thereof in improving dietary behaviours.

Proposed framework and implications
The current study compliments the evidence indicating that university students are a high-risk population for experiencing food insecurity and poor diet (Kent et al., 2023;Shi et al., 2021Shi et al., , 2022)).Our results provide a compelling argument for higher education institutions to urgently address deficiencies in the campus food environment.Our findings showed that students' reasons to choose program ideas reflect the four dimensions of food security, namely affordability, accessibility, utilisation and stability.As suggested by Roy et al. (2015) and considering the complexity of food insecurity issues, potential strategies with multiple components should be developed and supported by specific food policies at the university-level to mandate the actions to achieve the provision of variety and affordable healthier food on campus.Fig. 1 depicts a proposed framework of campus food and nutrition security.
Recognising the need to integrate the key dimensions of food security into strategies promoting both food security and healthy eating, these strategies align with the key dimensions and are further supported by university food policy.For example, food availability refers to the physical existence of food with sufficient nutritional quality, encompassing not only quantity but also the variety and quality of food.Therefore, strategies including culturally diverse foods and options catering to different dietary needs, are included.As stability dimension refers to the consistency and sustainability of the first three dimensions, it is present in all strategies.The ultimate outcome is that university students are able to maintain healthy eating habits, thereby achieving food and nutrition security.
Both our quantitative and qualitative analyses demonstrate the need to ensure affordability and increase physical food access in addressing food insecurity and improving healthy eating on campus.This study also attempts to explore and understand students' needs and perspectives, which are pertinent for developing and implementing initiatives to reshape the campus food environment to improve healthy eating.Potential campus-based programs to alleviate financial strain and increase healthier food options should be prioritised, which is reiterated here.Multi-component interventions using a settings based-approach based on the socio-ecological determinants of health framework should be considered to address this complex issue (de Villiers & Faber, 2019).Additionally, the presence of healthy and sustainable food policies at the university-level, coupled with regular evaluations, are needed to ensure food and nutrition security in university students.

Strengths and limitations
This study has several strengths and limitations that should be noted.The strengths of this study include its inclusion of both domestic and international students from different degrees and faculties across campus.Additionally, the use of 10-item USDA FSSM, a validated survey module, was used to determine food security status and the Canadian classification was applied for categorisation.Due to discordance between the US and Canada classification in food security, Men and Tarasuk (Men & Tarasuk, 2022) suggested recognising marginal food insecure individuals/household as "food insecure" rather than "food secure" by using this classification.Furthermore, by collecting both quantitative and qualitative data (i.e., open ended responses), the current study triangulated and corroborated the findings and obtained consensus on barriers to eating healthily in general and specifically on campus, particularly among food insecure students.Further evidence using mixed methods from other universities across Australia is needed to gain a comprehensive understanding of the food insecurity situation "I particularly like the idea of a farmers' market, as it would be great to support Australian agriculture at the same time as being able to eat more healthily.""Programs that supply affordable and sustainable fresh foods are always a great way to get university students to eat better.""Create a supportive ecosystem." a Some responses may overlap with other categories.
of university students post-COVID-19.However, it is important to note the cross-sectional survey was conducted in a single university using convenience sampling and had a small sample size, which may limit the generalisability to the wider university population.Although the information about the number of people in a household was collected, further details about their living situation, such as whether students live with their parents/family was not captured, which may be an important predictor.Future studies should also explore the association between food insecurity and housing security as both are basic needs.In addition, this survey relied on selfreported data including students' behaviours and perceptions related to fruit and vegetable consumption, which may subject to social desirability and recall bias.

Conclusion
University students continue to be at high risk of food insecurity and have low consumption of fruit and vegetables.Lack of affordability and low accessibility of fruit and vegetables have been identified as significant factors associated with food insecurity in this population.The current study suggests the importance of students being able to contribute ideas for addressing food insecurity and healthy eating concerns in order to develop effective interventions.Also, there is a need for healthy and sustainable food policies at the university-level which should be prioritised to ensure food and nutrition security in university students.Food insecurity is a complex and societal problem requiring multidimensional elements and multisectoral action.

Table 1
Participant sociodemographic characteristics of 237 university students completing survey.

Table 2
Food insecurity status between domestic and international students.

Table 3
Behaviours and perceptions related to fruit and vegetable consumption.

Table 4
Univariate and multivariable logistic regression assessing factors associated with food insecurity among university students (N = 237).
a Adjusted for age, gender, student status, employment, household size.b Distribution collapsed into too little/don't know and about right/too much.c Indicates significant (p < 0.05).

Table 5
Qualitative categoriesreasons for choosing the program ideas (n = 152) a ."Iwould be more inclined to purchase healthy food if it was affordable.""The availability of discounted or free fruit and veg will be very helpful for me and will also help me achieve my fitness goals.""The current food places don't allow for healthy eating at a reasonably priced bracket."Physical food access (n = 67, 44.1%) "It will make healthy food more accessible and reliable for vulnerable students just like me." "Greater access.Currently, I don't as healthy as I'd like to because I eat what is convenientthe food on campus, and the food near me, which tends to be unhealthy.""I spend a lot of time on campus and think it would increase my access to nutritious meals."Food availability (n = 34, 22.4%) "Because It would increase the availability of healthy food in campus.""I also really want halal options because many times I want to try new foods instead of chips but I am held back because of the items not being halal.""I barely buy food on campus at Macquarie because there are very limited vegan/vegetarian options plus I don't want to spend that much money."Knowledge and skills (n = 28, 18.4%) "Because I don't really aware of the importance of food so nutritional consultation might help me gain knowledge on the importance of consuming healthy food.""To know more ingredients, I can use and how to cook them.""Definitely especially with the nutritionist, because they be able to personalise the care each student receives, especially taking into account dietary restrictions that people like me have."Sustainability (n = 5, 3.3%)