Food consumption by NOVA food classification, metabolic outcomes, and barriers to healthy food consumption among university students

Abstract The NOVA food classification system is a simple tool that can be used to assess the consumption levels of different categories of foods based on their level of processing. The degree to which food is processed has a significant impact on health outcomes. In Ghana, no study exists on the consumption of the different NOVA food groups among tertiary students and how it relates to their metabolic outcomes. This study assessed the frequency of food intake according to the NOVA classification and how they relate to body mass index, waist circumference, and blood pressure. The barriers to the consumption of healthy foods among students were also assessed. This was a cross‐sectional study conducted among 352 students of the Takoradi Technical University. Questionnaire was used to obtain sociodemographic information as well as data on perceived barriers to healthy food consumption. Food frequency questionnaire was used to obtain data on dietary intake. The weight, height, waist circumference, waist‐to‐hip ratio, and blood pressure of all participants were measured. Chi‐square was used to compare categorical variables between males and females and to determine the association between the frequency of food intake according to the NOVA classification and metabolic indicators. The prevalence of overweight and obesity was 23.8%. More than half (51.1%) of the students had elevated blood pressure. The majority of study participants (54.2%) had a high frequency of consumption of both unprocessed and ultra‐processed foods. Male students who frequently consumed ultra‐processed foods (1–6 times/day) had significantly high blood pressure. High consumption of both ultra‐processed and unprocessed foods was also associated with elevated blood pressure among male students. Limited time to prepare healthy meals and the high cost of unprocessed foods were among barriers to which most students strongly agreed to. Establishment of canteens that provide affordable healthy foods, teaching students time management, and nutrition education can mitigate barriers to healthy food consumption.


| INTRODUC TI ON
Dietary guidelines based on food groups and nutrient profiles provide nutritional information and recommendations aimed at assisting the global population to make healthy food choices (FAO, 2016).
These recommendations are geared toward the prevention of deficiency diseases, obesity, and other diet-related noncommunicable diseases (Menegassi et al., 2019).Classifying foods based on the nutrients they contain can be misleading considering that the level of processing of various foods can also impact a person's health.Chronic non-communicable diseases are currently the world's leading cause of morbidity and mortality and account for about 17.9 million deaths each year (Roth et al., 2018).They are mainly attributed to unhealthy diets, especially those that are ultra-processed (Chen et al., 2020;Pagliai et al., 2021).
The NOVA food classification system was developed by the Centre for Epidemiological Studies in Health and Nutrition, School of Public Health, University of Sao Paulo, Brazil (Monteiro et al., 2018).This classification system groups foods based on their level of processing rather than by their nutrient profile.The NOVA food classification is based on the physical, biological, and chemical processes that foods are taken through after they are separated from nature and before they are consumed or used in the preparation of dishes and meals.There are four classes within the NOVA food system: unprocessed or minimally processed foods, culinary ingredients, processed foods, and ultra-processed foods.
Unprocessed foods comprise those that have undergone cleaning and removal of inedible or unwanted parts without the addition of oils, fats, sugar, and salt.Culinary ingredients include oils, butter, sugar, and salt.Processed foods have been preserved or improved by adding culinary ingredients.The fourth category includes industrial formulations made mostly or entirely from substances derived from or extracted from food constituents or synthesized in laboratories from food substrates or other organic sources.Normally, natural foods make up a small percentage of ultra-processed products or are entirely absent.
Adolescents tend to increase their consumption of ultraprocessed foods and beverages when they enter university and other tertiary institutions.This increase in the consumption of ultraprocessed foods and beverages is often a result of independence from their families, convenience, and inadequate time to cook owing to the demands of academic work.The aforementioned factors culminate in opting for fast foods, and low-nutrient dense cuisines, which puts them at risk of adverse health outcomes (Sprake et al., 2018).
The adoption of such a dietary pattern is of public health concern since it has a long-term impact on their health and well-being.Being overweight or obese during adolescence and young adulthood is also a strong predictor of obesity and associated chronic diseases later in life (Abdelhafez et al., 2020;Isa & Masuri, 2011;Momanyi & Akinyi, 2014).
Currently, in Ghana, there are no data or research on the level of consumption of the different NOVA food groups among university students and how they are associated with health outcomes such as blood pressure and obesity.There is also a dearth of information with regard to the barriers these university students face in the acquisition of healthy foods within the university campus.
Therefore, this study sought to assess the frequency of food intake according to the NOVA classification and their association with overweight, obesity, and blood pressure as well as barriers to healthy food consumption among students of the Takoradi Technical University in Ghana.

| ME THODS
This was a cross-sectional study that was undertaken among students of the Takoradi Technical University in Ghana.Trained research assistants assisted in the collection of data.Questionnaires were used to collect information on sociodemographics, meal consumption patterns, and barriers to the consumption of healthy foods.The questionnaire was pretested among nursing and midwifery students.Feedback from pretesting was used to modify the questionnaire prior to actual data collection.Weight, height, waist circumference, and body composition measurements of all participants were taken.Blood pressure measurements were also taken for all participants.

| Study area
The study was conducted in the Takoradi Technical University which is within the Western Region of Ghana.Data collection was done within the various Faculties of the university.

| Sample and sampling technique
Krejcie and Morgan sample size estimation formula was used to estimate the sample size for the study.The total population of students at the Takoradi Technical University was obtained from the students' records as 15,106.The total student population of each Faculty was also obtained.The sample size determined from the Krejcie and Morgan sample size table for the given population was between 306 and 310 (Chuan & Penyelidikan, 2006).Two percent of each faculty's population was calculated and all added to form the total sample of 302.The actual number of students who partook in this study was 352.Data collection took place within the faculties and within each faculty, study participants were randomly selected.

| NOVA food consumption assessment
A detailed food frequency questionnaire was used to collect data on food consumption.The Prospective Urban Rural Epidemiology (PURE) study food frequency questionnaire which is validated was adapted for the study (Baker et al., 2020).The food item list consisted of unprocessed and processed foods.The NOVA food classification system was used to categorize foods as unprocessed, processed,culinary ingredients and ultra-processed foods.The foods included in the food frequency questionnaire were typical local and exotic foods consumed in Ghana.The foods were grouped under the four NOVA food classes.For each food item, participants were to indicate the frequency at which they consumed it, whether daily (1-3 times daily), weekly (1-3 times per week), monthly (1-2 times per month), occasionally (once in several months), or never.These five responses were coded from 1 to 5. To assess the average frequency of consumption of each NOVA class, the response codes for foods within the same NOVA category were summed up and divided by the total number of foods within the category.The average responses were grouped into four; daily, weekly, monthly, and occasionally/never.These four responses were subsequently categorized into two for the unprocessed and ultraprocessed foods categories.Daily to weekly consumption were categorized as high frequency of consumption and coded as 1 while monthly and occasionally/never were categorized as low frequency of consumption and coded as 2. The codes for the ultra-processed and unprocessed foods were subsequently computed by addition.A resulting code of 2 was interpreted as high frequency of consumption of both ultra-processed foods and unprocessed foods, 3 as low and high consumption of either and 4 as low frequency of consumption of both.

| Barriers to healthy food consumption
Perceived barriers to the consumption of healthy foods were assessed using a Likert scale questionnaire.The items in the questionnaire were based on common barriers to healthy food consumption reported among adolescents and young adults, specifically in Africa (Abdelhafez et al., 2020;Lima et al., 2021;Musaiger et al., 2013).
Students were to indicate their level of agreement with a set of barriers provided within the questionnaire.The responses ranged from strongly agree to strongly disagree.Additionally, participants were also asked about the period within the day they had adequate time to eat, what constituted their favorite meals, if they planned their meals in advance, and if they consumed breakfast before leaving their hostels for lectures.

| Ethical clearance
Ethical clearance for the study was granted by the Cape Coast Teaching Hospital Institutional Review Board (CCTHERC/ EC/2022/106) in Ghana.Written permission was sought and obtained from the university authorities prior to data collection.
Participants were made to sign an informed consent form to indicate voluntary participation.

| RE SULTS
A total of 352 students participated in the study.The mean age of study participants was 22.3 ± 2.4 years and the majority resided in hostels (49.7%).The majority of participants were single (99.1%) and Christians (92.6%).For the anthropometric characteristics, the mean weight and BMI of participants were 64.3 ± 10.6 kg and 21.8 ± 3.4 kg/m 2 , respectively.About 10% of study participants were underweight, 65.9% were of normal weight, and about a quarter (23.8%) were overweight or obese.The prevalence of overweight and obesity was significantly higher in females (37.2%) compared to males (13.6%; p > .001).The mean systolic and diastolic blood pressure were 122.8 ± 13.9 and 71.2 ± 10.4 mm Hg and systolic blood pressure was significantly higher among males (122.8 ± 13.9 mm Hg) compared to females (115.0 ± 11.4 mm Hg; p < .001).Table 1 shows the descriptive characteristics including sociodemographic characteristics and anthropometric measures of study participants.
Table 2 shows the consumption pattern of the NOVA food categories by gender.For unprocessed foods, most participants consumed them 1-6 times per day (41.2%) and 1-3 times per week (38.9%).This pattern of consumption was similar for culinary ingredients.For ultra-processed foods, most students consumed it monthly (31.8%) and 1-6 times daily (29.5%).The consumption frequency of ultra-processed foods was significantly higher among females compared to males.
Among male students, the frequency of consumption of unprocessed foods, ultra-processed and high consumption of both unprocessed and ultra-processed foods was significantly associated with blood pressure but this was not the case for female students.
Among males with elevated blood pressure, 77.8% consumed unprocessed or minimally processed foods less frequently (1-3 times per month) and 76.9% consumed ultra-processed foods daily.Among female students, the consumption of culinary ingredients was significantly associated with blood pressure.Among females who had normal blood pressure, 67.8% consumed culinary ingredients daily while 35.7% consumed them 1-3 times per week.For females with elevated blood pressure, 32.2% consumed culinary ingredients daily while 64.3% consumed culinary ingredients 1-3 times per month.
NOVA food category consumption was not significantly associated with waist circumference and BMI status.Tables 3-5 show the association between NOVA food consumption frequency and metabolic outcomes.
Table 6 shows the meal consumption patterns and perceived barriers to healthy food consumption among students.The majority of study participants did not consume breakfast before leaving their hostel or home nor did they plan their meals in advance.These findings were similar among males and females.Also, the majority of the students indicated that the period within the day they had adequate time to eat was in the afternoon (54.3%) and evening (38.9%) compared to the morning (6.8%).More than half (52.7%) of the study participants indicated that fast foods constituted their favorite foods and this was followed by local/traditional foods (29.3%).
For the barriers to healthy food consumption, a high proportion of participants strongly agreed with the statements healthy foods are more expensive (29%), cooking healthy foods is more difficult compared to cooking processed/ultra-processed foods (28.1%), and limited time to prepare healthy foods (26.4%) and nonperiodical follow-up on nutritional status (27.3%).The majority of students strongly disagreed with the statement succumbing to peer pressure (27.0%) and the statement university food environment is not suitable for consuming a healthy diet (31.3%).

| DISCUSS ION
This study assessed the level of consumption of the four categories of NOVA food among students of the Takoradi Technical University located in the Western Region of Ghana and how they relate to weight and blood pressure as well as barriers to healthy food consumption faced by students.Findings from the study reveal that the prevalence of elevated pressure is high among the students while the prevalence of overweight and obesity is lower than what is reported among the general adult population (Asosega et al., 2023;Atuahene et al., 2017;Ofori-Asenso et al., 2016;Owiredu et al., 2019;Solomon et al., 2017;Tuoyire et al., 2016).The mean BMI of study participants was within the normal range but blood pressure was slightly above normal with males recording a significantly higher value.Though the blood pressure measurement of participants in this study was taken once, elevated blood pressure observed among them is of grave concern.Elevated blood pressure among the students can be linked to a reliance on out-of-home cooked meals which tend to be high in salt and fat and or skipping breakfast which is linked to snacking and other inappropriate habits later in the day.
Additionally, male students are more likely to depend on out-ofhome cooked foods compared to females and have lower intakes of vegetables which may make them more prone to elevated blood pressure (Mello Rodrigues et al., 2019).Elevated blood pressure in adolescence is linked to high pulse, thickened carotid intimamedia, hypertrophy of left ventricular, and cardiovascular diseaserelated mortality in adulthood (Yang et al., 2020).In a similar study among college students in Ethiopia, Tadesse and Alemu (2014) reported that the male gender was associated with a higher risk of hypertension.Similar findings have also been reported among university students in Congo where the prevalence of hypertension was 26.4% (Lebbie et al., 2017).In Ghana, hypertension awareness tends to be lower among younger persons compared to older adults probably resulting from their perceived lower risk of acquiring the condition (Sanuade et al., 2020).This calls for public health education and routine screening among adolescents and young adults to conscientize them on their blood pressure levels and the need to adopt healthful lifestyle.This is especially essential as majority of students strongly agreed that non-periodical follow-up on their nutritional status was a barrier to healthy food consumption.Hypertension is a risk factor for several chronic diseases including stroke, kidney diseases, and heart diseases, and one of the leading causes of death in Ghana (Owusu et al., 2021).
The prevalence of overweight and obesity was significantly higher among females compared to males.Several studies conducted among adolescents and adults within and outside Ghana have reported a higher prevalence of overweight and obesity among the female gender compared to the male gender (Lartey et al., 2020;Ofori-Asenso et al., 2016).
Furthermore, findings on consumption revealed that students had a higher frequency of consumption of unprocessed foods as well as ultra-processed foods.This implies that participants consume protective foods such as vegetables, fruits, whole grains, and other whole foods but at the same time consume lots of unhealthy options such as sugar-sweetened drinks and biscuits among others.
The high consumption of ultra-processed foods has the potential to negate the positive health impact of unprocessed food consumption and thus puts students at risk of adverse health outcomes.
Ultra-processed foods are high in fat, salt, and sugar and these increase the risk of hypertension, obesity, and other chronic diseases.A study conducted by Cordova et al. (2021) revealed that the consumption of ultra-processed foods is strongly associated with a 5-year increase in weight in a dose-response manner.In this study, the beneficial impact of unprocessed food consumption on health was not clearly evident.In fact, among males, the majority of those with a consumption frequency of 1-6 times per day for unprocessed foods had elevated blood pressure.Unprocessed foods are most effective in promoting cardiometabolic health when they are   consumed frequently as part of an overall healthy dietary pattern and lifestyle such as regular physical activity (Smiljanec et al., 2020).
High consumption of ultra-processed food is linked to excessive caloric intake and inadequate consumption of essential nutrients (Cattafesta et al., 2020;Rossato et al., 2023).Additionally, among female participants, the frequency of consumption of culinary ingredients was higher among those with normal blood pressure.
Culinary ingredients such as sugar may be added to whole-grain porridges like oats, wheat, or corn.Vegetable oil and animal fat are often used in the preparation of gravies, most of which are vegetable based.Their health impacts are also dependent on the quantities consumed.It is, therefore, possible that the female participants with normal blood pressure and perhaps study participants in general did not consume high quantities of these foods.The majority of students did not consume breakfast nor did they plan their meals; these practices predisposed them to high consumption of ultraprocessed foods as most of them also indicated that fast foods constituted their favorite meal.Individuals who plan their meals TA B L E 7 Binary logistic regression of the predictors of blood pressure.
Weight and height of all participants were measured following standard procedures.All participants were made to empty their pockets and remove cardigans and other heavy clothing.The Salter digital weight scale, model 9102 , was used to measure the weight of all study participants.The scale was put on a hard leveled flat surface and set to 0. Participants were made to stand motionless on the weight scale with arms by the side and weight was measured to the nearest 0.1 kg.The Seca 213 portable stadiometer was used to assess the height of all study participants.Participants were made to stand upright on the stadiometer with their buttocks and upper back against the rod of the stadiometer, and heels against the base plate.Participants were also asked to look straight ahead to achieve a Frankfort plane, with shoulders relaxed and arms by the side.The head plate of the stadiometre was then pulled down to touch the highest point of the skull and height was taken to the nearest 0.1 cm.The height and weight of participants were used to calculate their Body Mass Index (BMI) by dividing their weight in kilograms by their height in meters squared (m 2 ).Participants were subsequently categorized as underweight (<18.5 kg/m 2 ), normal (18.5-24.9kg/m 2 ), overweight (25-29.9kg/m 2 ), and obese (≥30 kg/m 2 ) as per the World Health Organization's standard (World Health Organization, 2010).

2. 5 |
Blood pressure measurement Blood pressure measurements were taken following standardized procedures.The blood pressure measurement of all participants was taken with the Omron M7 Intellisense digital sphygmomanometer .Participants were made to sit and rest for about 5 minutes after they had gone through the anthropometric measurements.Blood pressure was taken once and recorded for all participants.Blood pressure below 120 and 80 mm Hg for systolic and diastolic was classified as normal.Blood pressure measurement ≥120 mm Hg systolic and/or ≥80 mm Hg diastolic was classified as elevated blood pressure (World Health Organization, 2023).

Statistical
Package for Social Sciences (IBM SPSS) version 23 was used for data analysis.Pearson's chi-square and Fisher's exact tests were used to compare categorical variables between males and females.Independent sample t-test was used to determine the differences among males and females for continuous variables such as weight and height.A p-value of <.05 was used for statistical significance.Binary logistic regression was used to model the factors associated with blood pressure.Variables included in the model were age, gender, body mass index, and waist-to-hip ratio.The consumption levels of unprocessed foods and ultra-processed foods were also included in the model.

Table 7
shows the binary logistic regression of the predictors of blood pressure.Body mass index, gender, and consumption of unpro- Descriptive statistics.Consumption pattern of NOVA food category by gender.
TA B L E 1 Note: Bold values indicate p values ≤ 0.05.TA B L E 2 Note: Bold values indicate p values ≤ 0.05.TA B L E 3 Association between consumption of NOVA food class and blood pressure.Note: Bold values indicate p values ≤ 0.05.| 1989 SEKYI et al.
Association between NOVA food class and waist circumference.NOVA food consumption and weight status of study participants.
TA B L E 4 Meal consumption and perceived barriers to healthy food consumption.
Note: Variable set as reference for blood pressure is elevated blood pressure.
Note: Bold values indicate p values ≤ 0.05.TA B L E 6 (Continued) a Variable set as reference.