Vegetable Consumption and Factors Associated with Increased Intake among College Students: A Scoping Review of the Last 10 Years

Vegetable consumption is a predictor for improved health outcomes, such as reduced obesity and likelihood of food-related noncommunicable diseases. Young adults are a key population, being in a transitional stage-of-life: Habits gained here are taken through the lifespan. This review establishes insight into the consumption of vegetables among young adults during their college/university years, and factors associated with increased consumption. Seventy-one papers were extracted, published between January 2009 and October 2018. Search terms related to consumption; vegetables; and college/university setting and sample. A diverse range of definitions, guidelines, and study approaches were observed. Findings identify that the majority of students do not consume World Health Organization recommendations. Being female was the most frequent predictor of higher intake of vegetables, and no consumption patterns were identified by countries. Living at family home; body mass index; happiness and stress level; perceived importance of healthy eating; socioeconomic level; breakfast consumption; stage of study; openness to new experiences; sleep pattern; nutrition knowledge; activity level; alcohol usage; and energy intake were identified as influential factors. Public policies and new strategies to encourage vegetable consumption among college students are indispensable, especially targeting subgroups with even lower intakes, such as males and those living outside family home.


Introduction
Young adulthood is a particularly important time for the promotion of healthy eating, because several behaviours are developed and established during this period [1]. However, as characterised as a transitional life stage which may include many significant changes, such as leaving the family home, commencing college, entering the workforce, partnering, or becoming a parent, many people lack interest in following a healthy and balanced diet, or struggle to prioritise this [2,3].
Young adults include people from different backgrounds in a relatively large age range, and the great majority of college students are part of this group. They are beginning to take responsibility for their own dietary habits as they undergo a critical period in the consolidation of eating habits The terms used in the search comprised four categories that were combined using the Boolean operator "AND" as follows: (a) consumption (food consumption OR food intake OR eating), (b) type of food (vegetable), (c) setting (college OR university OR "higher education" OR faculty) and (d) sample (student OR freshman OR sophomore OR young adult OR millennial OR late adolesc* OR emerging adult* OR "new adult"). The combinations were adapted to use more general or more specific terms based on the limitations of each database. For the Scopus and Scielo databases, the search was performed considering the title, abstract and keywords, while in Medline/Pubmed, the terms were searched in the full text due to the low number of references (n = 20) when searching just title and abstract. Preliminary searches were helpful for adjusting search terms and their combination in order to find the largest possible number of articles related to the topic.
Studies published in Portuguese, Spanish and English were included. This was possible due to the international team of authors which enabled full assessment of studies published in these three languages. It was felt that inclusion of the widest possible range of studies enhances the value of the review, representing findings from a wide diversity of cultures and settings. Exclusion criteria comprised qualitative studies; studies focusing on eating disorders (e.g., bulimia) or specific groups (e.g., athletes, pregnant women), biomarkers and supplementation; studies with patients (e.g., menopause women, anaemics, people with coeliac disease); specific minerals association with vegetable consumption; validation and reliability of questionnaires; hypothetical scenarios and case-control studies. Intervention studies were not included to ensure that this review focused on establishing a base line account and to avoid repetition of the recent systematic review by Appleton et al. [16].
Duplicates were removed, followed by irrelevant titles. The abstracts of the remaining papers were reviewed, and potential studies were considered based on the inclusion/exclusion criteria. The studies which analysed the intake of food or food groups, or the dietary patterns of college students were read and judiciously analysed in full text. Studies were not considered if they presented data on aggregated fruit and vegetable consumption; if they presented data of consumption in scores where it was impossible to estimate the consumption; if they were developed in the college setting with a different population (e.g., lecturers); and if they were not developed with college students. Figure 1 shows how the database search and article selection process resulted in 71 articles being included in this review.
The data of selected studies were extracted to a Microsoft Excel spreadsheet for analysis, including study details (i.e., authors, location, year of publication, and design), study population, sample and participant demographics, food intake assessment instruments, definition of vegetables, data on vegetable consumption and associated factors with increased vegetable intake. The information extracted from each study is presented in the summary tables. The percentage of male and female participants and mean age across all studies was calculated. The common results were grouped and presented separately according to the type of data provided (i.e., frequency of overall intake, frequency of intake according to portions/servings, average intake of portions/servings per week, consumption in grams/day and comparison of consumption with relevant guidelines). Mean daily vegetable intake was calculated across studies presenting the consumption of vegetables in frequencies of intake. A few studies are presented in more than one table. Details from all studies were tabulated by one review author (VR) and checked by AF or GB. Tables are provided in the Results section. Tables outline vegetable consumption organised by frequency, quantity and comparison against relevant guidelines.

Design and Participants
This study reviewed data from 71 articles regarding the vegetable consumption of 65,971 college students from more than 155 different colleges located in 30 countries from Africa (2), Asia (8), Europe (13), North (2) and South America (4) and Oceania (1). The majority of participants were female (69.8%), and the mean age of the students was 21.6 years old. Almost 95% of the studies (n = 67) were designed as cross-sectional. The other designs were mixed (cross-sectional and longitudinal) [17], microlongitudinal (21 days) [18], time series analysis [19], and retrospective survey [20].

Design and Participants
This study reviewed data from 71 articles regarding the vegetable consumption of 65,971 college students from more than 155 different colleges located in 30 countries from Africa (2), Asia (8), Europe (13), North (2) and South America (4) and Oceania (1). The majority of participants were female (69.8%), and the mean age of the students was 21.6 years old. Almost 95% of the studies (n = 67) were designed as cross-sectional. The other designs were mixed (cross-sectional and longitudinal) [17], microlongitudinal (21 days) [18], time series analysis [19], and retrospective survey [20].

Vegetable Definition
Most studies (60.6%) did not define what was being considered as vegetables in their investigations. The term was only presented in the tables or text referring to the group, without specifying whether participants were told what to consider as a vegetable, or whether different types of vegetables consumed were grouped in this category. From the studies which mentioned what was considered in the analysis, eight divided vegetables into raw or fresh vegetables (including salads), and cooked vegetables [37,43,46,47,49,53,56,72]. Another six studies [19,34,36,38,44,58] declared that the intake of both raw and cooked vegetables was considered to calculate "vegetable consumption". One study [45] only considered salad and raw vegetables in their analysis. Four studies divided vegetables into different categories: green, yellow, other vegetables, and salads [60], sautéed leafy greens, leafy greens, nonleafy cooked vegetables and nonleafy raw vegetables [21]; fresh, frozen, canned and stewed [22]; and fresh, tinned, legumes and potatoes [29]. Finally, eight studies included different forms of vegetables to create a single variable in their analyses: green-, red-or yellow-coloured vegetables [64]; vegetables without tubers, roots and bananas [48], fresh, canned or juice [25]; vegetables and juices [85]; raw, cooked, canned or frozen [77]; coloured and other types of vegetables, mushrooms and sea vegetables [63]; fresh, cooked or frozen, as well as green salad and did not count potatoes [31]; and vegetable side dishes or salads [32]. It is clear that the complexity of defining vegetables either by botanical or culinary descriptors makes it difficult to provide an aggregated analysis.
No study mentioned the degree of food processing related to the vegetables consumed i.e., whether the vegetables were fresh, minimally processed (e.g., washed, sliced, peeled), juiced, or preserved in brine or sugar. Additionally, there was no discussion regarding the type of production of these vegetables (organic or conventional), if they were originated from genetic modified crops, or the type of commercialization, for instance, whether they were part of fair trades, locally produced, or imported from other countries.

Vegetable Consumption
Vegetable consumption is summarised in five tables, according to the type of outcome measure provided in the studies. In the first table, data from 30 studies are presented with the frequency of vegetable intake (Table 1). Mean frequency of daily vegetable intake was 40.2%, varying from 11.2% to 72.4%. The highest frequency of daily intake of vegetables was observed in Finland, where 72.4% of females and 57.3% of males eat vegetables daily [45]. In a study conducted only with female participants in Poland, 65.0% ate salad and raw vegetables every day [75]. In Cyprus [38], 56.5% of the participants ate fresh or cooked vegetables daily and, from these, 29.5% more than once a day. Other studies also demonstrated a high frequency of intake, such as in Lithuania [49], where 60.0% eat vegetables 4-7 times a week, and in Italy [82], where 42.1% of participants ate vegetables at least once a day, and 16% in a frequency of 5-6 times a week.
On the other hand, some studies demonstrated frequencies of daily intake as low as 11.2% in Saudi Arabia [64], 12.4% in South Africa [59] and 14.3% in Zimbabwe [55]. Brazilian studies showed the lowest frequencies of vegetable intake. In Perez et al.'s [56] study, 28.4% and 25.5% of college students answered never eating raw vegetables/salads and cooked vegetables, respectively. In Cansian et al.'s [21] study, 25.2% of participants answered never or rarely eating sautéed leafy greens. Associated factors with increased intake were being female [36,49], regular health self-rate [79], lower BMI and lower blood pressure (both genders) [84], in later years of study [72], not being a quota student (an affirmative action approved by Law which reserves 50% of spots in Brazil's federal universities for students coming from public schools, low-income families and who are of African or indigenous descent) [56], the importance given for eating healthy [47], and living at family home [44]. Table 2 presents frequency of vegetable consumption according to the portions/servings consumed. The most common frequency of intake was 1 portion/day. This level of consumption was achieved by 51.6% of Iranian students [25], 44.0% of Indian students [65] and 35.8% of Chilean [66]. The form of presenting the results was not uniform, and sometimes, only a percentage related to the consumption of a determined amount was presented, without specifying the distribution of the remaining percentage of consumption among students. For instance, the study of Duran-Aguero et al. in 2014 [67] presented that 21.8% of the investigated students consumed 2 portions of vegetables a day, while their study of 2016 [69] compared the percentage of normal and overweight/obese students who consumed 2 portions of vegetables a day, which was 32.4% and 43.9%, respectively. Moreover, a Saudi Arabian study [23] demonstrated that almost two thirds of nutrition department students (64.3%) consumed ≥3 servings of vegetables/day, while the non-nutrition department students consumed a lower percentage (45.5%). The studies showed that consuming two or more servings of vegetables was a protective factor for overweight/obesity [66] and that the measure of happiness was positively associated with the amount of vegetables consumed [25]. A higher frequency of vegetable intake was associated with both a higher frequency of eating episodes and a regular breakfast habit, and this association with breakfast habits is stronger for males than for females, while the association with the number of eating episodes was similar between sexes. A higher socioeconomic status and the intention to lose weight represented independent factors associated with more favourable vegetable consumption [28]. Students who lived in the family home consumed more helpings of vegetables each day, compared with young adults who lived independently [78].
In Table 3, the studies present the average portions/servings of vegetables consumed overall or by groups. The highest average intake of vegetables was identified in New Zealand and Canada. In New Zealand, Conner et al. [22] found frequencies of 2.5 servings/day in sample 1 and 2.8 servings/day in sample 2, while the microlongitudinal study developed by White et al. [18] found an average intake of 2.5 servings/day. In Canada, the regular intakes were 2.5 servings/day in 2010 [77] and 2.7 servings/day in 2013 [40]. Italian students living in the family home consumed higher quantities of cooked vegetables, whilst those living away from home were characterised by higher consumption of raw vegetables [37]. Italian women and students living at family home were positively associated with a greater consumption of vegetables [53]. In New Zealand, openness to new experience was the most consistent significant predictor of higher vegetable consumption. Young adults higher in openness ate more daily servings of vegetables than young adults lower in openness across both samples tested [22]. In Croatia, nutrition knowledge was significantly positively correlated with intake of vegetables [52], and women, senior students and those who prepare food for themselves demonstrated higher nutrition knowledge scores. In Italy, intention significantly affects vegetable eating behaviour in participants with low habits, while perceived behavioural control is the main predictor of the behaviour in the high habits group. This indicates that vegetable consumption may be intentional as well as habitual, depending on the level of habit strength [33]. Finally, in Spain, overweight people consumed significantly fewer vegetables than the normal weight ones. Females ate more raw or cooked vegetables than men [54].
A point to be considered in results presented in Tables 2 and 3 is that not all the studies mentioned the equivalence in grams for the portions evaluated or whether participants were told what to consider as a portion.
In Table 4, studies are summarised according to vegetable consumption in grams/day. The higher average intake in grams per day was found in Iran [42] (263 g/day), followed by Japan (217.5 g/days) [35]. The lowest intake was observed in first-year students in Croatia (80 g/days) [86]. In Iran, compared with those in the lowest tertile, women in the top tertile of dietary energy density had the lowest diversity score for vegetables [41], and breakfast consumers had a larger intake of vegetables and higher scores for the dietary diversity score for vegetables (1.6 versus 1.2) [42]. In Japan, late midpoint of sleep was significantly negatively associated with the energy-adjusted intake of vegetables [63]. Additionally, in a Dutch study, vegetable intake was lower among students who were non-Dutch, living in the family home, not adhering to physical activity guidelines and moderate and heavy alcohol drinkers [32]. Table 5 presents studies comparing the vegetable consumption with relevant guidelines or recommendations that varied from daily [47] to five portions of vegetables per day [48]. The median frequency of participants who achieved relevant recommended vegetable intake was 35.4% but varied widely. The lowest frequency of compliance with recommendations was found in South Africa, where only 2.5% of participants met the recommendation of 3 portions/day [59], followed by two studies in the USA, in which 7.0% [85] and 12.4% [71] of the participants met the recommended 2.5 cups/day. The highest frequency (74.0%) was found among fourth-year students from the Netherlands (150 g/day) [17], followed by 68.4% for salad/raw vegetable intake (daily or several times a day) in a study in Finland [47]. Associated factors with meeting the relevant recommendations of vegetable intake were being female [46,47,54,79], importance given for eating healthy [47], normal weight [33,68] and less stress [46].

Discussion
This study presents worldwide data regarding vegetable consumption from almost 70 thousand college students. The findings demonstrate that the majority of young adults do not consume vegetables as frequently as recommended by the WHO, nor in sufficient quantities to satisfy other relevant guidelines. No consumption patterns according to country or region were apparent. Being female was the most frequent predictor associated with higher intake of vegetables [36,46,47,49,53,54,80]. The variation between genders was highlighted and might explain the large disparity in consumption among studies. For example, the highest consumption percentages were found in studies only with women or in studies with more than 70% of female respondents, such as Finland [45], Poland [75] and Spain [54]. This finding is also consistent with previous research showing that females eat healthier than males, as, for instance, male young adults eat more frequently at fast-food restaurants than female young adults [88], and male college students consume fewer servings of fruit and vegetables daily than female (4.3 vs 4.8; p < 0.05) [89]. In addition to this main predictor, the following factors were associated with higher intake of vegetables: normal weight [33,54,66,68]; living in the family home [37,44,53,78]; greater perception of happiness and less pressure and stress [18,25,43,46]; importance given for healthy eating [33,47,79]; higher socioeconomic level [28,56]; having breakfast [28,42]; lower BMI and lower blood pressure [84]; later stage of study [72]; more openness to new experiences [22]; early mid-point of sleep [63]; nutrition knowledge [52]; being more active and drinking less alcohol [32]; and lower energy diet density [41].
Studies on vegetable consumption often indicate health benefits from high consumption, such as reduced risk of cardiovascular disease, diabetes type 2, various cancers, stroke, dementia and cognitive decline [8]. The majority of studies, however, do not investigate vegetable consumption independent of fruit consumption or other aspects of the diet. While fruits and vegetables are frequently consumed together, associations may reflect not just the relationship with vegetables but with product consumption in general, or with a healthier diet/lifestyle [16]. Despite the study focus being the intake of vegetables in particular, the majority of data came from studies evaluating the overall diet intake of students. This might have hampered the analysis, considering that often only partial information regarding the consumption was available. Additionally, comparison between studies is problematic due to differing study approaches (e.g., grams per day; frequency of intake). In order to minimise this limitation, the results were grouped by type of data available.
There was a lack of definition for vegetables in many studies, and many times, it was not possible to identify if any definition was given to participants at the point of research. In a few studies, the authors only used the term 'vegetables' to refer to the category. Other studies divided vegetables into 'cooked' and 'raw/salads', and a few were specific, dividing them, for instance, in 'green', 'yellow', 'salad' and 'other vegetables'. Moreover, potatoes were sometimes included in the category and sometimes not considered at all. The authors believe that this lack of definition is a limitation in the study of vegetable consumption.
Many issues were identified as challenges in the design, conduct, measurement, evaluation and comparison between studies on vegetable intake. Previous studies have highlighted that the term 'vegetables' covers a heterogeneous group of foods, especially across cultures and geographic locales. For instance, legumes (dried beans and peas), which are not by botanical definition vegetables, are often included for calculating vegetable intake, but not always in the same manner. Botanically speaking, foods that develop from the flower of a plant are defined as a fruit, while those from other parts are vegetables. This definition is not consistent with the culinary parlance which classifies plant-based ingredients on the basis of taste. Due to these differing approaches, inconsistencies appear between studies adding complexity to the field. Potatoes are also sometimes included in studies, but sometimes excluded. While a traditional staple for many Caucasian groups, they are not for many Asian/Pacific populations [90]. Accurate determination of vegetable consumption is essential to determine current intake patterns and for evaluating interventions developed to increase consumption. Additionally, it is not clear for the majority of the studies whether participants were asked to consider intake of vegetables per se or as part of a composite dish such as in a casserole, which may lead to inconsistencies between study findings.
Another issue is related to the lack of investigation regarding the degree of processing of the vegetables consumed. Vegetables eaten fresh or minimally processed (i.e., unprocessed foods altered by processes to make their preparation easier or more diverse, such as removal of inedible or unwanted parts, drying, fractioning and refrigeration or placing in containers) tend to preserve their main characteristics and nutrient profile. Canned and bottled vegetables (i.e., processed foods), despite having increased durability and enhanced sensory characteristics, frequently have the addition of salt, sugar or fat as preservers, which may have a negative impact on the nutrient profile of the original food. Finally, if substances such as colours, flavours, emulsifiers and other additives are added to increase palatability and attractiveness, which is frequently done, it can be named as an ultra-processed food, whose formulation, presentation and marketing often promote overconsumption [91,92]. Differently from what is expected with fresh or minimally processed food consumption, the intake of processed, and mainly of ultra-processed, versions has been increasingly associated with unhealthy dietary nutrient profiles and several diet-related noncommunicable diseases [92,93]. These products are also troublesome from social, cultural, economic, political and environmental points of view [94]. Therefore, processed, and mainly, ultra-processed vegetables should be accounted for and analysed separately by the studies.
Furthermore, the type of production and commercialisation of the vegetables can also have an impact on consumption. For instance, organically produced foods have higher concentrations of antioxidants, lower concentrations of the toxic metal Cd and a lower incidence of pesticide residues than non-organic comparators across regions and production seasons [95]. Additionally, they are more sustainable, with gains to species diversity in organically farmed fields in comparison to conventional farms [96]. However, organic foods are often more expensive than conventional alternatives, and this could moderate consumption [97].
It is evident that studies investigating vegetable consumption focused much more on the amount consumed than on the nutritional quality of these vegetables and the impact of their production-consumption chain. Nevertheless, such factors need to be considered, mainly when recommendations are made, as they have direct influence on human and environment health.
Studies which evaluated the frequency of vegetable intake among college students indicated a higher percentage of daily intakes in Finland, Poland, Cyprus, Lithuania and Italy. The lowest rates for daily intake were found in Saudi Arabia, South Africa and Zimbabwe. In Brazil, the frequencies of students who answered not eating vegetables at all were the highest among the studies which evaluated consumption by frequency. It would be expected that populations living within the culture of the Mediterranean diet, such as Spain, Italy, Cyprus, Croatia and Greece, at least report a higher consumption of vegetables than in other similar, non-Mediterranean, populations [98]. However, results from these countries support a shift away from the Mediterranean diet toward less healthy eating patterns [36][37][38][39][40]. This phenomenon appeared more evident among students living outside the family home, who also are more inclined towards lower consumption of homecooked meals and more frequent use of fast food [37]. Gaining primary responsibility for food shopping and preparation can lead to unhealthy dietary habits among college students living out of the family home. By contrast, students living at the family home might receive more support for healthier food habits.
The studies which presented consumption by frequency of portions/servings consumed also indicated a more frequent consumption of 1 portion/day, which is lower than the WHO recommendation of eating at least 5 portions per day (considering vegetables and fruits together). Considering the average intake by portions, the only studies where the average reached 2 or more servings per day were in New Zealand and Canada. Additionally, the Canadian studies demonstrated high standard deviation, highlighting strong variability within these results. The terms 'portion size' and 'serving size' are sometimes used interchangeably, and therefore, both terms were included in the results. However, it is widely accepted that these terms have different meanings. Portion size refers to 'the amount of food intended to be consumed by an individual in a single eating occasion', whereas serving size refers to 'the quantity recommended to be consumed in a single eating occasion' [99]. Further, considering that not all the studies mentioned the equivalence in grams for the portions evaluated or whether participants were told what to consider as a portion, the results might not reflect exactly the same basis for assessing consumption.
The majority of studies did not mention the setting in which vegetables are consumed, which is deemed a significant gap in knowledge, especially considering that it has been previous identified that setting may have an impact on consumption, as, for instance, more frequent use of fast-food restaurants has been associated with lower intake of key nutrients and healthful foods, and conversely, more frequent use of full-service restaurants was related to higher intake of vegetables [88].
In the studies summarised according to vegetable consumption in grams/day, higher average intakes were found in Iran [35] (263 g/day) and Japan (217.5 g/days) [33]. In Iran [35], breakfast consumers had higher scores of dietary heathy eating index and dietary diversity for fruits, vegetables and whole grains compared with nonconsumers. In Japan [33], 17.5% of males and 23.5% of females had a daily vegetable intake of 350 g or more, and the results of analyses conducted separately for males and females showed that the significant relationship between breakfast skipping and poor vegetable intake was found only in males. However, it was not clear whether those who habitually eat breakfast consume vegetables at breakfast or whether they ate vegetables at other meals but not breakfast, because they did not examine dietary intake of breakfast separately.
It is frequently stated by international agencies, national governments and nongovernmental organisations that regular breakfast consumption is associated with higher intakes of micronutrients, a better diet that includes fruits and vegetables and less frequent use of soft drinks. Further, breakfast eaters tend to achieve the recommended dietary allowance for vitamins and minerals more often compared to breakfast skippers and to have higher scores for healthy eating indexes [100]. Additionally, there is an association between skipping breakfast and low nutritional adequacy of adult diets. Moreover, comprehensive dietary counselling that supports daily breakfast consumption may be helpful in promoting healthy dietary habits throughout the day [101].
Finally, for the studies which compared vegetable consumption with relevant guidelines or recommendations, the median frequency of participants who achieved recommended vegetable intake was 35.4% but varied widely. The lowest frequency of compliance with guidelines was found in South Africa [55] and in the USA [69,84], as opposed to the highest found in Netherlands [15] and in Finland [40]. It was possible to notice, however, a wide variation in recommendations. The global guidelines from WHO do not disaggregate fruits from vegetables, and this might make it difficult for people to understand the quantities of each type of food that should be consumed. This combined recommendation hinders the standardization of national guidelines towards vegetable consumption. Irrespective of the guidelines considered, young adults are struggling to achieve the recommendation, demonstrating that this is a crucial point to be addressed.
Innovative public policy initiatives designed to stimulate vegetable consumption are required. For instance, implementing healthy food prescriptions within large government healthcare programs to promote healthier eating could generate substantial health gains and be highly cost-effective. A study using US nationally representative data and a validated microsimulation model to evaluate policy scenarios for adults found that, over a lifetime, a 30% subsidy on fruits and vegetables would prevent 1.93 million cardiovascular disease events and 0.35 million deaths and save $40 billion in healthcare costs. This subsidiary program would be highly cost-effective from a healthcare and societal perspective, and theoretical results were consistent across subgroups within each insurance group, including by age, race/ethnicity, education and income [102]. Other public health initiatives focused on influential factors identified in this study could also be discussed and implemented to increase healthy eating, such as enhancing sleeping patterns [103,104], overall wellbeing [105,106], and nutritional knowledge [34], and related to accessibility and affordability of fresh, organic and locally produced vegetables, focusing on more sustainable options [107,108] and stimulating homemade preparation [109,110].

Limitations and Further Studies
Limitations are accepted on any study regarding vegetable consumption. Firstly, many studies do not disaggregate fruit from vegetables, and this is also true with regard to guidelines. Secondly, the definition of what is described as a vegetable varies considerably between studies, policy and even consumer parlance. Thirdly, measurement of vegetable intake varies from grams/day, to portions and serving sizes. Lastly, no account is taken of where vegetables are consumed or if they are part of a composite dish. The authors have ensured robust analysis of the available papers and collated key issues of debate.
Taken together, the limitations highlight the need for greater consistency in future studies in this area. Further studies would usefully include greater consideration of the cooking method employed and the impact of this on both the nutrient contribution and consumer acceptance. Additional research into the most effective methods to encourage greater consumption levels would be valuable to both the field and this population.
It is apparent that studies on vegetable consumption are focused on assessing consumption and the achievement of nutritional guidelines by individuals. However, it is important to think that, beyond achieving quantities, it is necessary to discuss the quality of food consumed from a health and sustainability perspective. The type of production (e.g., organic, genetically modified) and commercialisation (e.g., fair trade, local) of the vegetables are key elements to be considered, as both have impact on the sustainability of the system. The contribution of vegetables is important not just from a dietary perspective but given their potential to reflect UN sustainable development goals and deliver economic gain.

Conclusions
There is a paucity of data on the factors influencing vegetable consumption. While there have been studies of perceptions of freshness, psychosocial, environmental and life course factors influencing fruit consumption, there are very little data on vegetables, and they constitute an under-researched area. Hence, this review makes a vital and timely contribution. It is well known that the majority of students do not consume recommended levels of vegetables; however, existing efforts to understand the drivers of vegetable intake are fragmented, do not consider the population of young adults sufficiently and often aggregate fruit and vegetable consumption together, introducing bias. This review, for the first time, provides a comprehensive assessment of vegetable consumption and finds that being female is the strongest predictor associated with higher intake. Country or region differences were not observed; however, a number of other contributory factors have been identified, such as having a higher socioeconomic level; living in the family home; later stage of study; having a greater perception of happiness and less pressure and stress; and being more open to new experiences. Considering modifiable factors, overall elements related to healthy quality behaviour were the most prominent in the studies.
Public policies and new strategies to encourage vegetable consumption among college students are indispensable, especially targeting subgroups with even lower intakes, such as males and those living outside the family home. Positive eating habits are established in young adulthood and involve adoption of healthy food practices, including adequate vegetable consumption. Additionally, recommendations broadly agree that a sustainable diet should be based on vegetable consumption and reduced meat consumption and should prioritize the consumption of locally produced, seasonal and organic foods, strengthening short food supply chains and decreasing environmental impact, climate change, soil degradation, gas emissions, water contamination and loss of biodiversity. Therefore, not only are positive eating patterns important to promote positive health outcomes such as the reduction of obesity and noncommunicable diseases, but also for global sustainability.