The Difference of Ultra-Processed Food Consumption Based on Individual Characteristics and other Factors among Non-Health Undergraduate Students in Universitas Indonesia in 2023

Ultra-processed food/UPF is a product that undergoes a series of industrial techniques and processes, also has low nutritional value. Overconsumption of UPF can increase the risk of overweight and obesity. The purpose of this study is to determine the differences in the proportion of UPF consumption based on individual characteristics, environmental factors, and lifestyle factors among non-health undergraduate students in Universitas Indonesia in 2023. A cross sectional study design conducted on 149 samples. The data was collected using a self-administered online questionnaire. The data was analyzed using univariate and bivariate (chi-square) analyses. The results showed that 50.3% of the students consumed a high level of UPF. The bivariate analysis showed a significant difference in the proportion of UPF consumption based on peer influence and access to UPF.


Introduction
classified food and beverages into four groups, one of which is ultra-processed food (UPF).UPF is defined as formulations of predominantly industrial processes [1].Products within this group include soft drinks, sweet or savory packaged snacks, frozen foods, dairy products and derivatives, ready-to-drink tea and coffee, and others [1,2].However, excessive consumption of UPF can lead to various health problems, one of which is an increased risk of obesity [3][4][5][6].A 10% increase in the proportion of UPF consumption is associated with an 18% increase in the risk of overweight/obesity, particularly visceral obesity, an increase of 0.38 kg/m2 in body mass index (BMI), and an increase in waist circumference [4,6].
On the other hand, all regions and countries are undergoing a transition towards more processed diets, closely related to food system industrialization, technological changes, and globalization [2].A global systematic review showed that the highest UPF consumption rates were observed in the United States and the United Kingdom, with UPF accounting for more than 50% of total energy intake [7].South Asia and Southeast Asia experienced the highest percentage increase in per capita UPF sales volume from 2002 to 2016, reaching 67.3% and ultra-processed beverages (UPB) at 120.0% [8].Survey of Individual Food Consumption in Indonesia 2014 revealed that the proportion of instant tea, branded powdered coffee, branded powdered beverages, packaged liquid beverages, and sweetened condensed milk consumption ranged from 5.8% to 31.2% [9].Riset Kesehatan Dasar 2018 showed that the proportion of habitual consumption of foods containing seasoning, instant noodles/other instant foods, and carbonated beverages (≥1x/day) ranged from 2.2% to 77.6% [10].Previous studies conducted in this research location indicated that university students had high consumption rates of instant noodles, sugar-sweetened beverages, and fast food, ranging from 48% to 69.2% [11][12][13].
Various international studies have identified factors related to UPF consumption rates.However, research in Indonesia regarding UPF consumption, especially among university students, is limited.Given the increasing trend of UPF consumption in the literature, the researchers are interested in investigating the differences in the proportion of ultraprocessed food consumption based on factors associated with consumption among university students.While they tend to have substantial knowledge about a healthy diet, it may not always be accompanied by the right dietary behavior [20].Studies have shown an increase in the frequency of UPF consumption among students at Universitas Mulawarman in Samarinda, East Kalimantan [21].This finding aligns with research conducted among students at the Universitas Indonesia, which revealed high percentages of UPF consumption, particularly in non-health faculties [11][12][13].

Methods
This study employed a crosssectional design with a quantitative approach.The dependent variable in this study was ultra-processed food consumption.Independent variables included individual characteristics (sex, knowledge about UPF, allowance, and attitude towards UPF), environmental factors (media exposure, peer influence, family influence, and access to UPF), and lifestyle factors (physical activity, food labels, reading habit, and smoking behavior).The research was conducted at the Universitas Indonesia located in Depok, West Java.Data collection took place from April to June 2023.The study was carried out online, distributing an online questionnaire to non-health students at the Universitas Indonesia through various platforms.It was supported by free campus Wi-Fi accessible to all students.
The study population consisted of all non-health undergraduate students of Universitas Indonesia (UI) from the class of 2019 to 2022.The sample for this study comprised non-health undergraduate students from UI in the year 2023 who met the inclusion and exclusion criteria.Inclusion criteria were non-health undergraduate UI students from the class of 2019-2022 in the academic year 2022/2023 and were currently active students.Exclusion criteria included students undergoing specific diets and/or having certain illnesses leading to UPF consumption restrictions.The total sample size for this study was 149 participants.The minimum sample size was obtained from calculations using the hypothesis test formula for the difference between two proportions [22].
Primary data were collected through a self-administered online questionnaire filled out by respondents meeting the criteria.The study's instrument was an online questionnaire that respondents filled out independently.It included questions related to willingness to participate and provide informed consent, personal identification data, and questions based on each variable.Data processing involved four stages: editing, coding, processing, and cleaning.The data were subjected to univariate analysis using frequency distribution and bivariate analysis using chi-square and logistic regression.
UPF consumption was defined as the number of times respondents consume UPF that has food packaging and labels also can be purchased in stores/minimarkets/ supermarkets per day within a one-month period [23].The data was obtained through the completion a Food Frequency Questionnaire (FFQ).The choices for the frequency ranged from none, 3x/month, 2x/month, 1x/month, 1x/week, 2x/week, 3x/week, 4x/week, 5x/week, 6x/week, 1x/day, and >1x/day.The analysis was conducted by standardizing all denominators to per day, then summed and grouped based on median values.It was categorized into "High" (≥ 4.59x/day) and "Low" (< 4.59x/day).
Sex was defined as biological characteristics that define individuals as male and female [24].The data was obtained by completing a questionnaire in the respondent's identity section, either "male" or "female." Knowledge about UPF was defined as the level of understanding of respondents regarding UPF [25].The data was acquired through 10 questions questionnaire.The score of '1' for correct answers and score '0' for incorrect answers.The final score was calculated by dividing the total correct answers by the total number of questions and multiplying by 100%.It was categorized into "Poor" (<60%), "Fair" (60-79%), and "Good" (≥80%).
Allowance was defined as the amount of money used by respondents to purchase food and beverages over a specific period [26].The data was obtained through questions regarding the allowance used for food and beverages per day.It was categorized into "High" (≥ Rp50.000,00) and "Low" (< Rp50.000,00).
Attitude towards UPF was defined as the responses of respondents to their UPF consumption habits [27].The data was assessed through a questionnaire with 9 statements.The response options for statements number 1-2 and 6-9 were range from "strongly agree" (score 4) to "strongly disagree" (score 1).Meanwhile, for statements number 3-5 were "strongly agree" (score 1) to "strongly disagree" (score 4).The scores were totaled.It was categorized into "Positive"/supports UPF consumption (≥score 22.6) and "Negative"/ does not support UPF consumption (< score 22.6).
Media exposure was defined as exposure to UPF advertising in mass media that influences the UPF consumption habits [11,28].The data was obtained by filling out a questionnaire consisting of 4 questions.Response options were "yes" (score 1) and "no" (score 0).The total score was calculated and then categorized based on the median value.It was categorized into "Exposed" (≥ score 3) and "Not exposed" (< score 3).
Peer influence was defined as the influence of peer friends (tend to influence and individual's beliefs and behaviors) on respondents UPF consumption habits [29].Family influence was defined as the influence of family (reside in the same household) on respondents UPF consumption habits [30].Peer influence and family influence were assessed through 3 questions questionnaire each.Response options were "yes" (score 1) and "no" (score 0).The total score was calculated and then categorized based on the median value.It was categorized into "Influenced" (≥ score 3) and "Not influenced" (< score 3) for each variable.
Access to UPF was defined as the distance for home to the store that sells UPF to assess the convenience of obtaining and consuming UPF [31,32].The data was obtained by completing a questionnaire regarding the ease of accessing UPF, which was assessed based on the distance between the respondent's home and the stores that sells UPF.It was categorized into "Near" (< 100 m), "Moderate" (100 -500 m), and "Far" (> 500 m).
Physical activity was defined as the frequency and duration of heavy, moderate, and light physical activities of respondents within one week [33].The data was obtained through the completion on the International Physical Activity Questionnaire -Short Form.The questionnaire consisted of 7 questions regarding low, moderate, and vigorous level of physical activity.It was categorized into "Low", "Moderate", and "High".
Food label reading habit was defined as the respondents' conditions based on several questions related to smoking habits [35].The data was acquired through a questionnaire consisting of 10 questions.If the respondent answered "always", as score of '1' was assigned.If the respondent answered "sometimes/rarely/never", a score of '0' was given.The total score was calculated and then categorized based on the median value.It was categorized into "Poor" (≥ score 3) and "Good" (< score 3).
Smoking behavior was defined as the respondents' conditions based on several questions related to smoking habits [35].The data was obtained by responding to questions about whether the individual smokes or does not smoke.It was categorized into "Smoker" (have ever smoked and still smoke) and "Nonsmoker" (do not smoke).
In Table 2

Discussion
This study indicated that there were 50.3% of students with a high consumption level of ultra-processed food (UPF) (frequency ≥ 4.59x/day).The most consumed UPF products among the respondents in this study were instant noodles, UHT milk, packaged ice cream, and packaged bread.This is in line with the perception that instant noodles are a convenient and affordable food choice, readily available in various places [36], and students often consider them a practical substitute for regular meals [37].The consumption of processed milk is associated with the upper-income group, while bread is a significant contributor to daily calorie intake for men and the second most for women [38].
The study revealed that 99.3% of respondents consumed products categorized as sugary beverages, with UHT milk, packaged tea, and packaged coffee being the top three.The modern era has contributed to the proliferation of conveniently packaged products that are easily accessible and consumed.The appeal of these sugary beverages to adolescents is driven by factors such as practical packaging, sweet taste, and ease of availability [39].
Statistical analysis showed no significant difference in UPF consumption proportions based on sex (p-value = 1.000).However, there is a tendency for male respondents to consume more UPF compared to female respondents.These differences in UPF consumption between sexes are linked to eating habits, nutritional behaviors, and varying health beliefs [40].Better knowledge of processed products and healthier food choices could lead to lower UPF consumption among females [16].Other factors influencing sex-related consumption differences include taste preferences, sex norms affecting food access, and the perception that UPF can lead to weight gain [41].
The statistical analysis also indicated no significant difference in UPF consumption proportions based on knowledge about UPF (p-value = 0.778).Respondents with good knowledge (54.2%) tended to consume more highlevel UPF compared with fair to poor knowledge (46.2%).This aligns with a study on American teenagers that showed knowledge alone doesn't necessarily lead to reduced consumption of such beverages [42].
Results indicated no significant difference in UPF consumption proportions based on allowance (p-value = 0.610).High-level UPF consumption among students with higher allowances was comparable to those with lower allowances.UPF consumption is influenced by factors related to allowances, such as food process, affordability, and accessibility [40].Socioeconomic status also plays a role; low-income individuals tend to consume more instant noodles and processed meats as substitutes for healthier food, while higher-income individuals consume more processed milk and snacks [37].
Statistical tests showed no significant difference in UPF consumption proportions based on attitudes toward UPF (p-value = 0.154).However, there was a tendency for respondents with positive attitudes towards UPF to consume more high-level UPF compared to those with negative attitudes.This is in line with other research that found groups with supportive attitudes towards sugary beverages (included in UPF) tend to consume more compared to those with unsupportive attitudes [11].
Results showed no significant difference in UPF consumption proportions based on exposure to mass media (p-value = 0.214).Nevertheless, respondents exposed to mass media tended to consume more high-level UPF compared to those not exposed.Advertisements for processed foods and beverages are associated with preferences, beliefs, and eating behaviors [43].Prolonged and frequent exposure to food advertisements can influence consumption patterns [18].
Statistical tests revealed a significant difference in UPF consumption proportions based on peer influence (pvalue = 0.013).Respondents influenced by peers tended to consume more UPF compared to those not influenced.This aligns with previous research showing a connection between peer influence and sugary beverage consumption, included in UPF [17,44].Young adults often succumb to the influence of their peers in various habits, including eating behaviors [45].Kotler (2004) stated that peer influence can be related to exposure to information, new behaviors, self-concepts, and social pressures affecting product choices [46].This influence is often found to have a negative impact on increasing the consumption of high-energy and lownutrient foods [47].
Results indicated a tendency for respondents influenced by family to consume more UPF compared to those not influenced, although there was no significant difference (p-value = 0,061).Family consumption patterns and the availability of food at home can affect UPF consumption [48,49].However, changes in dietary habits can occur when students live independently.
Statistical tests showed a significant difference in UPF consumption proportions based on access to UPF (pvalue = 0.005).Respondents with closer access to UPF tended to consume more compared to those with less access.This finding is consistent with research that demonstrates easy access to high-calorie beverages as a determinant factor related to the consumption of such products among students [44].Urban environments facilitate UPF consumption due to greater access and commercial diversity [14].Given that students spend a considerable amount of time on campus, the presence of various easily accessible UPF products in the campus vicinity this supports students in purchasing such products [44].The univariate analysis in this study revealed that the campus cafeteria is the primary location for UPF purchases for over 50% of the respondents.
Results showed no significant difference in UPF consumption proportions based on physical activity level (p-value = 0.817).However, lower physical activity levers were associated with higher proportions of high-level UPF consumption.An unhealthy lifestyle, including low physical activity, is linked to increased UPF consumption [50].Beliefs about the benefits of UPF for optimizing fitness after physical activity can also influence consumption [51].
There was no significant difference in UPF consumption proportions based on reading food label habits (p-value = 0.402).However, respondents with good label reading habits tended to consume more high-level UPF compared to those with poor habits.Knowledge of the purpose of nutritional labels and the perception of their importance can impact reading habits [52].Yet, reading labels alone may not fully change consumption behavior.
Results indicated no significant difference in UPF consumption proportions based on smoking behavior (p-value = 0.818).Although there was a larger proportion of smokers among highlevel UPF consumers, consumption levels were comparable between smokers and non-smokers.Smokers often experience diminished taste sensitivity, potentially leading them to consume more to enjoy the taste [53].

Conclusion
It can be concluded that 50.3% of nonhealth undergraduate students from Universitas Indonesia in the year 2023 consumed ultra-processed food at a high level (≥ 4.59x/day).The most consumed types of UPF among the respondents are instant noodles, UHT milk, and packaged ice cream.Additionally, there are more female respondents with good knowledge about UPF, high allowance, supportive attitudes towards UPF, exposure to mass media, peer influence, family influence, moderate access to UPF, high level of physical activity, good food label reading habits, and non-smokers.Based on the bivariate analysis, significant differences in UPF consumption were observed based on peer influence and access to UPF.
Students are advised to increase awareness about food and beverage choices and become advocates within their peer groups regarding these matters.Information regarding nutrition and health can be disseminated through universityaffiliated social media platforms accessible to all students.The campus could provide health education about UPF and healthy food consumption to students and implement regulations related to vending machines.Policymakers can enhance the delivery of health messages to students, support educational institutions in promoting healthy eating on campus, and support research on students' dietary patterns and factors influencing UPF consumption.Future researchers can explore different populations using varying variables and methods.

Table 1 . Distribution of Characteristics of Non-Health Undergraduate Universitas Indonesia Students
, based on the results of the bivariate analysis using the chi-square test, significant differences in UPF consumption were identified based on peer influence [p-value = 0.013; OR = 2.353 (1.123 -4.566)] and access to UPF [p-value = 0.005; OR = 2.059 (0.969 -4.376) and 4.745(1.796-12.536)].