Psychosocial and dietary influences on eating behaviors and body mass index among geographically diverse Indian university students

Abstract

In the context of India’s growing youth population, the interplay of dietary habits, social influences, and psychological factors critically shapes nutritional health outcomes. This cross-sectional study examined the socio-demographic, dietary, psychological, and social determinants of eating behaviour and their association with body mass index (BMI) among 3,150 undergraduate students from agricultural universities across six Indian regions. Using a validated self-administered questionnaire and anthropometric assessments, the study applied Principal Component Analysis (PCA) on 14 binary behavioural variables to identify latent dimensions influencing dietary patterns. The key behavioral components associated with BMI were: External Cues & Emotional Reactivity (PC2) (β = +2.04, r = + 0.77), characterized by emotional eating and responsiveness to food advertisements; Dietary Practices & Cultural Disengagement (PC1) (β = −1.54, r = − 0.58), reflecting structured dietary habits independent of cultural food norms; and Positive Affect & Peer Independence (PC6) (β = −0.84, r = − 0.31), indicating emotionally positive, peer-independent eating. Dietary assessments revealed that although cereal consumption was adequate (100%), notable deficiencies were observed in the intake of green leafy vegetables, milk, and fruits, with consumption levels falling approximately 31–34%, 13–18%, and 10–20% below recommended dietary guidelines, respectively. The findings highlighted that beyond nutritional deficits, psychosocial factors including emotional reactivity and social pressures significantly influence students’ weight status. Emotional vulnerability and environmental cues linked with higher BMI, whereas structured eating and emotional well-being are protective. This research underscores the need for gender and socially sensitive nutrition programs in Indian universities to prevent both obesity and undernutrition.

Introduction

India is home to one of the world’s largest populations of young adults, with a substantial proportion of individuals in the 18–25-year age group transitioning into higher education. This demographic group is increasingly exposed to rapid dietary and lifestyle changes associated with urbanization, academic pressures, and evolving food environments. The proportion of India’s working age population has increased, with United Nations predictions¹ indicating it was approximately 58% in 2000 and may rise to 64% by 2035. Moreover, demographic estimates indicate that India’s median age was merely 29 in 2020, in contrast to over 40 in other industrialized countries such as the USA, Europe, and Japan². India is currently experiencing a complex nutritional landscape characterized by the coexistence of undernutrition, overweight, and obesity. National evidence (NFHS-5) indicates that while undernutrition remains a concern, particularly micronutrient deficiencies, the prevalence of overweight and obesity is increasing among university students and young adults (increased to 24% in women and 23%) in men due to rapid dietary and lifestyle transitions³. This positions India as one of the world’s ‘largest youthful nations.’ When this demographic advantage is combined with conducive policy conditions that promote health, education, skill development, and viable employment opportunities, it can accelerate India’s economic growth^4,5.

Previous research has shown that healthier dietary patterns, including greater consumption of plant-based foods, are associated with improved health outcomes as Springmann et al.⁶ estimated that shifts toward healthier diets could substantially reduce diet-related health risks at the population level. While India has made notable progress in food availability and overall health indicators, the country continues to face a dual burden of undernutrition and overweight, reflecting shifts in dietary practices and lifestyle behaviors. These changes are particularly relevant among university students, for whom emerging adulthood represents a critical period in the formation of long-term eating behaviors that can influence body mass index (BMI). Shifting to healthy diets that include sustainability considerations can contribute to reducing health and climate change costs by 2030 because the hidden costs of these diets are lower compared with those of current consumption patterns⁷. FAO⁸ Estimated that $12.7 trillion/year in hidden costs are embedded in global food systems $8 trillion from unhealthy diets and undernutrition, and $3.1 trillion from environmental damage. In the past decade, India has achieved significant gains in health and nutrition. Although India has achieved notable progress in reducing infant mortality and improving food production capacity, these advances have not uniformly translated into optimal nutritional outcomes across all age groups^9,10.

Young adulthood, particularly the university years, represents a critical period during which eating behaviours are shaped by increased independence, academic pressures, and changing food environments. The persistence of unhealthy eating practices during this stage may contribute to both undernutrition and excess weight, highlighting the need to better understand eating behaviours and their psychosocial correlates among university students. Despite substantial improvements in food availability¹¹ and the implementation of multiple national nutrition initiatives, India continues to face a dual burden of undernutrition and diet-related non-communicable diseases. Several researches have shown that prevalence of overweight, obesity and associated risk factors including dietary habits among Indian university students have resulted high levels of malnutrition highlighting the growing public health burden in young populations in India^12,13. Recent national and global assessments indicate limited progress in addressing nutrition-related health risks, including overweight, obesity, and diabetes among adults. These challenges coexist with persistent dietary inadequacies and unhealthy eating patterns, reflecting broader shifts in food consumption and lifestyle behaviors. Such trends are increasingly evident among young people, highlighting the importance of understanding eating behaviors and nutritional health during early adulthood. Among young adults, irregular eating patterns, meal skipping, unhealthy food choices and body image concerns were found to have key influences on BMI; both underweight and overweight prevalence in Indian context¹⁴. Springmann et al.¹⁵ highlighted that diet costs across 150 countries, showing that healthy diets aligned with sustainability goals are more affordable than current consumption patterns in many regions, particularly high-income countries. Beyond dietary intake alone, psychological and social factors exert a substantial influence on eating behaviours among young adults. Emotional states such as stress, anxiety, boredom, and loneliness have been linked to maladaptive eating patterns, including emotional eating and external cue–driven consumption. Social environments, peer influence, campus food availability, and exposure to food marketing further shape food choices, contributing to variability in dietary behaviours and BMI among university students. Inadequate nutrition and obesity represent significant health challenges for contemporary society, impacting both health outcomes and healthcare costs¹⁶. Approximately 40% of India’s youth population is at significant risk of cerebrovascular accidents, primarily due to detrimental lifestyle choices, including poor management of hypertension, diabetes, hyperlipidaemia, insufficient physical activity, unhealthy dietary practices, alcohol consumption, and the use of tobacco products or smoking. Suboptimal dietary practices constitute a significant public health issue among young adults transitioning to university life¹⁷, during which they encounter stress and time constraints¹⁸. Prevalent unhealthy eating behaviour among young adults included meal omission, dining out, snacking, and fast-food intake¹⁹. This situation renders employment physically arduous, necessitating considerable physiological exertion and the preservation of proper posture. It is exceedingly exhausting, time-intensive, and stressful²⁰.

University students possess insufficient understanding regarding appropriate food choices, which may adversely impact their eating habits and nutritional status²¹. The proliferation of shopping malls, convenience stores, vending machines, and fast-food establishments has engendered a concerning environment for young adults to adopt poor dietary practices²². Eating habits evolve through a multifaceted process shaped by various elements; societal influence intensifies with age, but parental and familial effect is predominantly formed during childhood and, to a lesser degree, adolescence²³. Emerging adulthood (ages 18 to 25) is a pivotal period in which individuals attain independence and develop enduring health behavior patterns, concurrently marked by poor lifestyle traits and an elevated risk of obesity and chronic diseases¹⁶. One of the key recommendations of Global Nutrition Report 2022 is to screening of diet quality when assessing food security interventions, in order to equitably address hunger and diet-related diseases²⁴. Limited research has been undertaken in India to examine the food habits of students and the factors influencing these habits. With this background, the present study aims to evaluate eating habits and their associated aspects, emphasizing social and psychological influences among Indian students by examining the following researchable issues:

• What is the socio-demographic and physiological characteristics of Indian university students?

• What are the dietary habits and consumption patterns of students in relation to a balanced diet?

• Is there an association between dietary habits scores and socio-demographic factors?

• What psychological and social factors influence the eating behaviors of students?

The primary objective of this study was to examine eating behaviors and their associations with body mass index (BMI) among Indian university students. Specifically, the study aimed to assess dietary practices, psychological factors, and social influences related to eating behavior, and to explore how these behavioral dimensions are associated with BMI in a large, multi-zonal sample of students enrolled in state agricultural universities across India.

Hypotheses

Based on existing literature, the study tested the following hypotheses:

H₁

Unhealthy dietary practices and lower dietary diversity are associated with higher BMI among university students.

H₂

Psychological factors such as emotional eating and reactivity to external cues are positively associated with BMI.

H₃

Social influences, including peer pressure and environmental cues, are associated with variations in eating behavior and BMI.

H₄

Distinct behavioral patterns derived from dietary, psychological, and social variables show differential associations with BMI.

Materials and methods

Context of the study and sampling

India hosts 66 State Agricultural Universities, 4 Deemed Universities, 3 Central Agricultural Universities (CAUs), and 4 Central Universities (CUs) with Agricultural faculties. Consequently, there are 77 universities offering agriculture as a discipline. The current 74 universities have been spread over the North, South, East, West, Central, and North-East zones. The present cross-sectional investigation involved 3,150 students from multiple state agriculture universities representing all zones. The study focused on students from state agricultural universities because these institutions are distributed across all major regions of India and enrol students from diverse rural and urban backgrounds. While offering a relatively uniform academic setting, they reflect substantial regional, cultural, and dietary diversity. Moreover, shared campus food environments enable standardized data collection and facilitate meaningful comparison of dietary, psychological, and social influences on eating behaviour. The data has been gathered to accurately reflect all regions, thereby representing the diversity of Indian eating customs and food consumption patterns. Table 1 presents the details of respondents from the agriculture university.

Table 1.

Respondents from agricultural university representing all zones.

Zone	Name of university/institute	State	Number of students selected
North zone	Ch. Sarwan Kumar Himachal Pradesh Krishi Viswavidyalaya, Palampur	Himachal Pradesh	150
	Sher-e-Kashmir University of Agricultural Science & Technology, Jammu	Jammu and Kashmir	150
	Punjab Agricultural University, Ludhiana	Punjab	150
	Sardar Vallabhbhai Patel University of Agriculture & Technology, Meerut	Uttar Pradesh	150
	Chandra Shekhar Azad University of Agricultural & Technology, Kanpur	Uttar Pradesh	150
South zone	University of Agricultural Sciences, Dharwad	Karnataka	150
	Tamil Nadu Agricultural University, Coimbatore, Coimbatore	Tamil Nadu	150
	Acharya NG Ranga Agricultural University, Guntur	Andhra Pradesh	150
	University of Agricultural Sciences, Bangalore	Karnataka	150
East zone	Birsa Agricultural University, Ranchi	Jharkhand	150
	Odisha University of Agricultural & Technology, Bhubaneswar	Odisha	150
	Uttar Banga Krishi Viswavidhyalaya, Cooch Behar	West Bengal	150
West zone	Anand Agricultural University, Anand	Gujarat	150
	Agriculture University, Kota	Rajasthan	150
	Maharana Pratap University of Agriculture & Technology, Udaipur	Rajasthan	150
	SKN Agriculture University, Jobner	Rajasthan	150
Central zone	Nanaji Deshmukh Pashu Chikitsa VisvaVidyalaya, Jabalpur	Madhya Pradesh	150
	Jawaharlal Nehru Krishi Viswa Vidyalaya, Jabalpur	Madhya Pradesh	150
	Indira Gandhi Krishi Viswa Vidhyalaya, Raipur	Chhattisgarh	150
North-East zone	Central Agricultural University, Imphal, Manipur	Manipur	150
	Assam Agricultural University, Jorhat	Assam	150
Total			3150

The study focused specifically on students from agricultural universities; therefore, the findings reflect eating behaviors within this academic context.

This cross-sectional study was conducted among undergraduate students enrolled in state agricultural universities across six agro-climatic zones of India. The sample size was determined based on feasibility and the objective of achieving broad regional representation rather than formal power calculation, given the exploratory nature of the study. Students aged 18–25 years who were enrolled as full time undergraduates and provided informed consent were included in the study. Students with self-reported chronic medical conditions affecting diet or body weight and those with incomplete questionnaire or anthropometric data were excluded from the final analysis. Responses with missing or inconsistent information were removed prior to analysis. The study assessed five categories of variables: sociodemographic characteristics, eating habits, psychological factors influencing eating behavior, sociological influences on food choice, and anthropometric measures including Body Mass Index (BMI).

Sociodemographic characteristics

Sociodemographic refers to the characteristics of a population that can influence behaviors and preferences. For the present study age, gender, marital status, educational background, and lifestyle factors such as smoking, alcohol use and physical activity were included under sociodemographic characteristics, based on extensive review of literature. Association between eating habits score and socio-demographic factors was analyzed to identify those sociodemographic characteristics which significantly affects the dietary pattern of the respondents.

Eating habits

Eating habits were assessed based on the dietary habits of the respondents through 10 items (regular meals, daily breakfast, frequency of snacks, cereals consumption, pulses consumption, vegetables consumption, fruits consumption, milk consumption, consumption of fried food and water intake). For data collection, a self-administered questionnaire on eating behavior was developed, based on the referring from Yahia et al.²⁵ and review of existing literature. The questionnaire was also reviewed for contextual relevance and minor linguistic adaptation to ensure clarity and cultural appropriateness for Indian university students. Average consumption of various food items by respondents were compared to the recommended daily intake based on ICMR guidelines to outline the nutritional gaps among respondents. Dietary intake data were verified through a 24-hour dietary recall to enhance the reliability of the food frequency questionnaire responses.

Psychological and social factors affecting eating behavior

Psychological and social factors influencing eating behavior were studied with the items adapted from the validated Compulsive Eating Scale²⁶. Psychological factors were studied based on the data captured through items covering emotional eating triggers such as loneliness, boredom, happiness, and anxiety. Social factors were studied based on the data captured through items covering sociological influences such as parental and peer pressure, cultural beliefs, and the influence of role models. Each behavioral question was scored as 1 (healthy response) or 0 (unhealthy response), and scores were summed to derive an overall eating behavior with higher scores indicating healthier habits.

Anthropometric measures

Anthropometric measurements were obtained using standardized procedures. BMI was calculated as weight in kilograms divided by height in meters squared (kg/m²), and classified using WHO (2004) Asian criteria²⁷. In addition to BMI, Ponderal Index (PI) was computed as weight divided by height cubed (kg/m³). Basal Metabolic Rate (BMR) was estimated using the Harris-Benedict equations, and Resting Energy Expenditure (REE) was estimated using the Mifflin equations, adjusted for gender and age.

Validation and internal consistency of the study

Before the main survey, a pilot study was conducted to validate the questionnaire with all its segments (sociodemographic characteristics, eating habits, psychological and social factors affecting eating behavior and anthropometric measures) aiming to ensure the accuracy and consistency of each measurement item, following Hair et al.²⁸ recommendations. Prior to the main survey, the adapted questionnaire was pilot-tested among a small group of Indian university students who were not included in the final sample. The pilot testing aimed to assess comprehension, clarity of items, and feasibility of administration. The pilot study involved 210 students from non-sampled areas representing all selected universities and included additional open-ended questions. Based on the pilot feedback, no significant changes were made to the core structure of the questionnaire, as the reliability scores (Cronbach’s alpha values ranging from 0.725 to 0.812 (predefined threshold of 0.7), indicating that the questionnaire was reliable and suitable for use in the main survey) were satisfactory. In this process, we identified and addressed potential response biases. So, minor revisions were made to clarify a few measurement items that could have been misinterpreted by respondents. In addition to this, content validity expert judgment was done by qualified subject matter experts to systematically evaluate the items so that the items accurately represent the domain of knowledge or construct being measured. The Content Validity Ratio (CVR) was above 0.78, which indicates that subject matter experts agreed the item were essential. Based on Experts suggestions the items were reviewed and refined for relevance and consistency across all respondents. The questionnaire was also reviewed for contextual relevance and minor linguistic adaptation to ensure clarity and cultural appropriateness for Indian university students. All procedures involving human participants were conducted in accordance with the ethical principles outlined in the Declaration of Helsinki. Participation in the study was entirely voluntary, and written informed consent was obtained from all participants prior to data collection. Participants were informed of their right to withdraw from the study at any stage without any consequences. Confidentiality and anonymity were ensured by using coded data and restricting access to research records. No invasive procedures were involved, and all efforts were made to minimize potential harm or discomfort.

Measurement of reliability and validity

In line with the methodology detailed by Hair et al.²⁸, we calculated the composite reliability (CR) using the formula:

where, λ is the standardized factor loading for each item in the construct.

To assess divergent validity (DV) and ensure that theoretically unrelated constructs do not exhibit significant correlations, we organized the squared correlations of latent variables in a matrix. We then compared these correlations with their average variance extracted (AVE). Following Hair et al.²⁸, the AVE was computed using the formula:

where, n denotes the total number of items in a factor.

Formulae for calculations:

1. Body Mass Index (BMI)

• b)Ponderal Index

• c)Mifflin equations for REE (calories):

• d)Harris-Benedict equations for BMR:

Data analysis included descriptive statistics (means, standard deviations, and frequencies), Student’s t-tests, and ANOVA to examine eating pattern differences across socio-demographic groups.

Principal component analysis (PCA)

Principal Component Analysis (PCA) was performed on 14 binary behavioural variables to extract latent factors based on eigen values > 1, and factor loadings ≥ ± 0.30. Multiple linear regression models were applied to assess the associative relationships of PCA-derived behavioural components on BMI, while hierarchical regression identified factors associated with the eating behaviour score. PCA was applied as an exploratory method to identify underlying patterns among dietary, psychological, and social behaviour variables. Although several variables were binary in nature, PCA was used to summarize correlated behaviours and reduce dimensionality in a large sample. The analysis was conducted on the correlation matrix, and components were interpreted descriptively. Pearson’s correlation coefficients examined the relationships between BMI, physiological parameters, and behavioural components. Multi-collinearity was assessed using the Variance Inflation Factor (VIF). PCA was conducted on 14 dietaries, psychological, and social variables among 3,150 Indian university students revealed six distinct components, cumulatively explaining 46.01% of the total variance in eating behaviours indicating a moderate level of explanatory power. These components are interpreted as exploratory behavioral patterns rather than definitive latent constructs.

Six behavioural factors were included under PCA analysis. PC1: Dietary Practices & Cultural Disengagement; involves the erosion of traditional, often healthier, eating patterns in favor of diets high in processed foods, fats, and sugars. PC2: External Cues & Emotional Reactivity; such as environmental, social, or sensory stimuli (e.g., sights, sounds, smells), directly influence emotional reactivity by triggering immediate psychological and physiological responses. These signals can induce stress, joy, or craving, often overriding internal states and prompting impulsive behaviors like emotional eating or addiction-related relapse. PC3: Socialized Eating Patterns; refer to how social contexts, relationships, and cultural norms influence what, when, and how much people eat, often leading to increased consumption in groups. These behaviors, driven by social facilitation and modeling, promote community but can also override hunger cues, causing people to eat more or mirror the choices of companions. PC4: Structured Eating vs. Loneliness; represent two contrasting, yet often interrelated, aspects of daily life and mental health. Structured eating involves purposeful, organized, and often planned nutritional intake (set meal times, balanced meals). Loneliness is the subjective, distressing feeling of social isolation or a perceived deficit in meaningful connections. PC5: Boredom-Driven Eating; form of emotional eating where food is consumed to combat feelings of emptiness, lack of stimulation, or boredom rather than physical hunger. It serves as a distraction or a way to seek novelty, often leading to consumption of high-calorie, sugary, or salty snacks. PC6: Positive Affect & Peer Independence; which are closely linked, with prosocial peer connections fostering increased positive emotions, confidence, and social-emotional skills in adolescents. Positive, supportive friendships act as a buffer against anxiety and depression, while enabling independence, allowing teenagers to develop their own identities.

Multiple linear regression analysis was conducted to examine the associations between six behavioral components derived from Principal Component Analysis (PCA) and Body Mass Index (BMI) among university students. Pearson’s correlation coefficients were also computed to examine the direct associations between each component and BMI.

Results

Socio-demographic characteristics

It is evident from Table 5 that most respondents (74%) belonged to the age bracket of 18–21 years, comprising 67.3% male and 80.6% females. Concerning the educational attainment of mothers, 19.7% possessed non-formal education, 51.33% completed high school or below, and 29% achieved intermediate education or higher. Most of the mothers were not employed (78.4%). Regarding paternal education, the majority possessed an intermediate level or more (56.3%), 24.3% had completed high school or less, and 19.4% had received non-formal education. A majority resides in nuclear households (67.6%) and has an average monthly household income of 20,000–30,000 rupees (49.7%), whereas 64.4% live with their relatives. Smoking (20.6%) and alcohol consumption (6%) have been recorded exclusively in males. Only 32.6% of respondents engaged in regular exercise.

Table 5.

Association between eating habits score and socio-demographic factors (n = 3150).

Characteristic	Category	Males (n = 2000) %	Female (n = 1150) %	Mean (SD)	Test Value (df)	p-value
Age	18–21	67.30	80.60	5.77 (± 1.88)	t(3148) = -2.31	0.02 *
	≥ 22	32.60	19.30	6.12 (± 1.93)
Mother’s education level	Non-formal Education	21.30	18.00	5.95 (± 1.76)	F(2, 3147) = 0.26	0.77
	High school or less	47.30	55.30	5.11 (± 1.99)
	Intermediate and above	31.30	26.60	5.85 (± 1.67)
Father’s education level	Non-formal Education	14.00	24.60	5.41 (± 1.20)	F(2, 3147) = 0.22	0.80
	High school or less	28.00	20.60	5.22 (± 1.11)
	Intermediate and above	58.00	54.60	5.78 (± 1.90)
Monthly income (INR)	5000–10,000	07.33	10.60	5.44 (± 1.03)	F(3, 3146) = 1.32	0.27
	10,000–20,000	22.60	32.60	6.21 (± 1.17)
	20,000–30,000	60.00	39.30	6.40 (± 2.01)
	> 30,000	10.00	17.30	6.11 (± 2.17)
Family type	Joint	36.00	28.60	6.26 (± 1.97)	t(3148) = 1.25	0.21
	Nuclear	64.00	71.30	6.01 (± 1.88)
Mother’s working status	Yes	25.30	18.00	5.93 (± 1.77)	t(3148) = -1.23	0.22
	No	74.60	82.00	6.31 (± 1.66)
Smoking status	Smoker	20.60	0	5.01 (± 1.04)	t(3148) = -2.58	0.01 *
Alcohol consumption	Yes	06.00	0	4.95 (± 1.32)	t(3148) = -2.12	0.03 *
Regular exercise	Yes	27.30	38.00	6.55 (± 1.59)	t(3148) = 2.16	0.03 *

* Significant at 5% level.

Physiological characteristics of respondents

The physiological attributes of male and female students were documented in an observational sheet, indicating individual age, gender, height, and relative values of BMI (kg/m²) and Ponderal Index, as presented in Table 2. The mean age of the subjects was 21.58 ± 7.26 years for males and 21.29 ± 2.47 years for females, with respective standard deviations. The corresponding physiological parameters for male and female recorded as height 164.7 ± 7.21 (cm) and 157.8 ± 6.34 (cm); BMI, 21.83 ± 6.52 and 21.09 ± 7.05 (kg/m²); Ponderal index 13.27 ± 3.51 and 13.41 ± 3.47 (kg/ (m)³; Resting Energy Expenditure 1511.3 and 1241.5 calories; and Basal Metabolic Rate (BMR) 1,514.5 and 1242.5 Calories/day, respectively (Table 2).

Table 2.

Physiological characteristics and prevalence of malnutrition among students based on BMI by gender of respondents.

Physiological characteristics	Male (n = 2000) (Mean ± SD)	Female (n = 1150) (Mean ± SD)
Age, Years	21.58 ± 7.26	21.29 ± 2.47
Weight, Kg	58.53 ± 5.84	51.89 ± 3.20
Height, cm	164.7 ± 7.21	157.8 ± 6.34
BMI, kg/m2	21.83 ± 6.52	21.09 ± 7.05
Ponderal Index (kg/ (m)³	13.27 ± 3.51	13.41 ± 3.47
Resting Energy Expenditure (calories)	1511.3	1241.5
Basal Metabolic Rate (BMR) Calories/day	1514.5	1242.5

Body mass index*	Male (n = 2000)	Female (n = 1150)	Total
Underweight (< 18.5)	14.0	17.3	15.6
Normal (18.5–22.9)	50.0	46.0	48.0
Pre-Obese (23.0-27.4)	27.4	26.0	26.7
Obese class (27.5–34.9)	8.6	10.7	9.6

*BMI is calculated on WHO criteria for Asian Population.

Dietary practices and average intake of diverse food items in relation to the ICMR’s dietary guidelines for Indians

Table 3 shows dietary practices among respondents and Table 4 presents data on the average daily consumption (g/day) of various food products relative to a balanced diet necessary to fulfil daily nutrient requirements by age group.

Table 3.

Dietary practices among respondents.

Characteristic	Male (%) (n = 2000)	Female (%) (n = 1150)	Total (%)	Chi-square (χ²)	p-value
Regular meals (3 times/day)
Yes	68.6	64.6	66.6	5.07	0.024*
No	31.4	35.4	33.4
Daily breakfast
Yes	72.6	78.0	75.3	10.69	0.001*
No	27.4	22.0	24.7
Frequency of snacks (per week)
<3 times	14.0	21.3	17.6	19.79	< 0.001**
≥3 times	86.0	78.7	82.4
Cereals consumption
Daily	100.0	100.0	100.0	–	–
Pulses consumption
Daily	82.6	74.0	78.4	21.76	< 0.001**
3–4 times per week	14.0	22.0	18.0
1–2 times per week	3.4	4.0	3.6
Vegetables consumption
Daily	67.4	64.7	66.0	5.63	0.018*
3–4 times per week	16.6	20.6	18.6
1–2 times per week	16.0	14.7	15.4
Fruits consumption
Daily	7.3	6.0	6.6	22.24	< 0.001**
3–4 times per week	37.3	47.4	42.3
1–2 times per week	55.4	46.6	51.0
Milk consumption
Daily	26.0	14.0	20.0	39.65	< 0.001**
3–4 times per week	31.4	35.3	33.4
1–2 times per week	36.6	47.3	42.0
Rarely	6.0	3.4	4.6
Consumption of fried food
3–4 times per week	30.0	39.3	34.6	17.14	< 0.001**
1–2 times per week	62.6	48.0	55.4
Rarely	7.4	12.7	10.0
Water intake (Liters/day)
<2	78.0	84.6	81.3	16.47	< 0.001**
≥2	22.0	15.3	18.7

Table 4.

Average consumption of various food items (g/day) in comparison to balance diet to meet daily nutrient requirement.

Food Items	Recommended requirement (Male)	Average consumption Mean ± SD (Male, N = 2000)	Percent change	‘t’ value	Standard error of difference	p-value	Recommended requirement (Female)
Cereals and Millets (g)	370	376.89 ± 13.13	1.86	23.46*	0.294	< 0.0001	270
Pulses and beans (g)	120	122.58 ± 18.24	2.15	6.32*	0.408	< 0.0001	90
Green leafy vegetables (g)	100	68.56 ± 15.89	-31.44	88.48*	0.355	< 0.0001	100
Other vegetables (g)	200	198.62 ± 17.23	-0.69	3.58**	0.385	= 0.0003	200
Roots and tubers (g)	100	95.50 ± 15.56	-4.5	12.93**	0.348	< 0.0001	100
Milk and milk products (g)	300	258.59 ± 18.89	-13.80	98.03**	0.422	< 0.0001	300
Fruits (g)	100	89.56 ± 17.54	-10.44	26.61**	0.392	< 0.0001	100
Fats and oils (g)	40	42.78 ± 12.85	6.95	9.67*	0.287	< 0.0001	30

Food Items	Average consumption Mean ± SD (Female, N = 1150)	Percent change	‘t’ value	Standard error of difference	p-value	F-value	p-value
Cereals and Millets (g)	272.55 ± 20.78	0.94	4.16**	0.613	< 0.0001	11381.0	< 0.0001
Pulses and beans (g)	85.79 ± 18.89	-4.67	7.55*	0.557	< 0.0001	8713.0	< 0.0001
Green leafy vegetables (g)	65.89 ± 19.59	-34.11	59.04**	0.578	< 0.0001	14.6	0.0001
Other vegetables (g)	185.86 ± 18.52	-7.07	25.89	0.546	< 0.0001	766.3	< 0.0001
Roots and tubers (g)	93.29 ± 18.26	-6.71	12.46**	0.538	0.0003	16.7	< 0.0001
Milk and milk products (g)	245.89 ± 20.55	-18.04	89.29**	0.606	< 0.0001	184.3	< 0.0001
Fruits (g)	79.59 ± 16.57	-20.41	41.77*	0.489	< 0.0001	462.5	< 0.0001
Fats and oils (g)	29.59 ± 12.89	-1.36	1.07NS	0.380	0.2809	1236.4	< 0.0001

* Significant at 5% level, ** Significant at 1% level, NS-Non-Significant Note= Recommended requirements for various food items (g/day) as per age group and lifestyle with reference point of Nutrient requirement for Indians 2020; One-Sample t-test (Each Gender and their Recommended Intake); ANOVA (Comparison Between Male and Female Intake).

Association between eating habits score and socio-demographic factors

The mean and standard deviation of the total score of eating habits were compared across category factors in the study. The mean for individuals aged 18–21 years was 5.77 (SD ± 1.88), while for those aged ≥ 22 years it was 6.12 (SD ± 1.93), with a significant difference (p = 0.02). A notable disparity in eating habits scores was observed between smokers, 5.01 (SD ± 1.04), and non-smokers, 6.41 (SD ± 1.93), (p = 0.01); between individuals who consume alcohol, 4.95 (SD ± 1.32), and those who abstain, 5.72 (SD ± 1.66), (p = 0.03); and between individuals who engage in regular exercise. 6.55 (SD ± 1.59) for those who did, and 5.78 (SD ± 1.44) for those who did not. No substantial correlation was identified between dietary practices and other socio-demographic variables (Table 5).

Psychological and social factors affecting eating behavior and their association with eating habits

Table 6 indicates that most females consumed food due to feelings of loneliness (72.6%) in contrast to males (52%). Among males, 21.4% reported feeling entirely out of control regarding eating, 14% consumed food till experiencing discomfort, 22% ate due to feelings of distress or anxiety, and 35.4% ate out of boredom. Among females, 25.3% reported feeling entirely out of control regarding food, 12.6% consumed food till experiencing discomfort, 38% ate due to feelings of distress or anxiety, and 27.3% ate out of boredom. A significant majority of guys (77.4%) consumed food due to feelings of happiness, in contrast to females (35.3%). Table 6 also indicates that 34% of males ate due to maternal pressure, while 28% did so due to paternal pressure. A significant proportion of females (71.3%) consumed food due to self-consciousness, in contrast to males (50.6%), although males (68%) engaged in eating more frequently in the company of friends or family compared to females (39.3%). Television and radio food commercials had a greater influence on the eating behaviours of males (41.3%) than on females (27.3%), but cultural beliefs significantly impact the eating behaviours of females (52%) more than those of males (22.6%). Peer pressure influences eating patterns in boys and females at rates of 79.4% and 74.6%, respectively.

Table 6.

Association between eating habits and social and psychological factors (n = 3150).

Factors affecting dietary habits	Male (n = 2000)		Female(n = 1150)		Mean (SD)		p Value
	Yes (%)	No (%)	Yes (%)	No (%)	Yes	No
Psychological factors
Eat due to feelings of loneliness	52.00	48.00	72.60	7.30	4.91(1.23)	5.32 (1.27)	0.02
Experience a loss of control while eating	21.40	78.60	25.30	74.70	5.12 (1.31)	5.56 (1.36)	0.51
Overeating to the point of discomfort	14.00	86.00	12.60	87.40	6.17 (1.40)	6.57 (1.51)	0.07
Eat in response to feelings of upset or anxiety	22.00	78.00	38.00	62.00	5.91 (1.32)	6.21(1.34)	0.01
Eat out of boredom	35.40	64.60	27.30	72.70	6.11 (1.53)	6.35 (1.61)	0.03
Eat as a result of feeling happy	77.40	22.60	35.30	64.70	6.94 (1.37)	7.13(1.40)	0.04
Social factors
Eat because of mother’s pressure	34.00	66.00	44.60	55.40	5.15 (1.76)	5.71 (1.78)	0.08
Eat because of father’s pressure	28.00	72.00	31.30	68.70	5.01(1.65)	5.31 (1.64)	0.07
Eat because of self-consciousness	50.60	49.40	71.30	28.70	6.43 (1.66)	6.87 (1.72)	0.02
Eat more within group of friends/ family	68.00	32.00	39.30	60.70	6.77 (1.01)	6.90 (1.86)	0.04
Eating behaviour is influenced by food advertisements in TV/Radio	41.30	58.70	27.30	72.60	5.18 (1.92)	5.49 (1.90)	0.01
Eating behaviour is influenced by cultural belief	22.60	77.40	52.00	48.00	5.29 (1.76)	5.10 (1.53)	0.02
Eating habits are influenced by peer pressure	79.40	20.60	74.60	25.40	4.91 (1.37)	6.09 (1.33)	0.03

Principal component analysis of dietary, psychological, and social factors

It is evident from Table 7 that the first component (PC1: Dietary Practices & Cultural Disengagement) explained 8.10% of the variance, primarily characterized by high positive loadings for daily milk (+ 0.45) and fruit (+ 0.35) consumption, as well as parental pressure (+ 0.39), reflecting structured dietary habits. Conversely, it showed a strong negative loading for cultural influence (− 0.54), suggesting that these dietary practices were pursued independently of cultural food norms. The second component (PC2: External Cues & Emotional Reactivity) accounted for 7.99% of the variance and reflected susceptibility to external and emotional triggers. This component was strongly defined by influences from food advertisements (+ 0.39) and eating when anxious (+ 0.39), while showing a negative association with happy eating (− 0.38) and eating in groups (− 0.35), indicating solitary, emotionally reactive, and advertisement-influenced eating patterns. The third component (PC3: Socialized Eating Patterns) explained 7.65% of the variance, capturing the influence of group dynamics and fruit intake, with key loadings on daily fruits (+ 0.41) and eating in groups (+ 0.32). It inversely associated with regular meal patterns (− 0.34) and daily vegetable intake (− 0.45), indicating that social eating may sometimes disrupt structured dietary behaviours. The fourth component (PC4: Structured Eating vs. Loneliness) contributed 7.54% of the variance and combined positive breakfast consumption (+ 0.52) and advertisement influence (+ 0.36), contrasting with a strong negative loading for eating when lonely (− 0.49). This component suggests that students with structured meal patterns (e.g., breakfast) may experience lower emotional eating linked to loneliness. The fifth component (PC5: Boredom-Driven Eating) explained 7.41% of the variance and was defined by a strong positive loading for eating when bored (+ 0.67) and a negative association with eating in groups (− 0.48), highlighting solitary eating driven by boredom rather than social settings. Finally, the sixth component (PC6: Positive Affect & Peer Independence) accounted for 7.33% of the variance. It showed a positive loading for eating when happy (+ 0.37) and advertisement influence (+ 0.29), with a pronounced negative loading for peer pressure eating (− 0.64), reflecting individuals who eat independently of peer influence, driven more by internal positive emotions and media cues. BMI, although widely used, does not distinguish between lean and fat mass or capture central adiposity. Future work should incorporate body composition measures as waist‑to‑hip ratio to provide a more nuanced assessment of nutritional status.

Table 7.

Principal component analysis of dietary, psychological, and social factors (n = 3150).

Variable	PC1: dietary practices & cultural disengagement	PC2: external cues & emotional reactivity	PC3: socialized eating patterns	PC4: structured eating vs. loneliness	PC5: boredom-driven eating	PC6: positive affect & peer independence
Regular meals	+ 0.29	+ 0.15	−0.34	−0.34	−0.28	+ 0.27
Daily breakfast	−0.17	−0.17	−0.06	+ 0.52	+ 0.10	−0.11
Daily vegetables	0.00	−0.26	−0.45	+ 0.21	−0.21	0.01
Daily fruits	+ 0.35	+ 0.30	+ 0.41	+ 0.19	−0.00	−0.31
Daily milk	+ 0.45	+ 0.07	−0.22	−0.06	+ 0.30	+ 0.17
Eat when anxious	0.02	+ 0.39	−0.06	+ 0.20	+ 0.10	−0.22
Eat when lonely	−0.27	0.02	+ 0.41	−0.49	+ 0.06	+ 0.08
Eat when bored	+ 0.10	−0.31	+ 0.06	−0.08	+ 0.67	−0.13
Eat when happy	−0.05	−0.38	+ 0.06	+ 0.14	+ 0.23	+ 0.37
Peer pressure eating	+ 0.05	−0.01	−0.27	−0.23	+ 0.07	−0.64
Influenced by ads	+ 0.07	+ 0.39	−0.04	+ 0.36	+ 0.17	+ 0.29
Eat in group	+ 0.13	−0.35	+ 0.32	+ 0.20	−0.48	−0.13
Parental pressure	+ 0.39	−0.32	−0.10	−0.07	−0.01	−0.19
Cultural influence	−0.54	+ 0.11	−0.32	−0.02	+ 0.04	−0.17
Eigenvalue	1.13	1.12	1.07	1.06	1.04	1.03
Variance explained (%)	8.10	7.99	7.65	7.54	7.41	7.33
Cumulative variance explained (%)	8.10	16.09	23.73	31.27	38.68	46.01

Component labels were assigned to facilitate interpretation and do not imply formal theoretical constructs.

Discussion

Among university students, dietary patterns are often shaped by irregular meal timing, frequent snacking, increased consumption of processed and convenience foods, and inadequate intake of fruits, vegetables, and dairy products. The transition to university life is frequently associated with academic stress, time constraints, altered sleep patterns, and reduced physical activity, all of which may contribute to unhealthy eating practices and weight-related outcomes.

This study’s findings reveal that most students were classified as having a normal weight according to BMI standards (Table 2). The prevalence of malnutrition among students, categorized by BMI and gender, revealed that less than half (48%) exhibited a normal BMI, while 15.6% were underweight, 26.7% were pre-obesity, and 9.6% were classified as obese Class I (Table 2). Normal weight was more common among males (46%) than females (50%), but obesity and underweight were more prevalent in females (10.7%, 17.3%) compared to males (8.6%, 14%), respectively. In Indian context, these findings are in line with National Family Health Survey (NFHS-4) data (2015–16) which reported that the prevalence of both underweight and overweight/obesity was higher among women compared to men^29–31. This study found that most of both males and females consumed meals regularly (66.6%) and breakfast (75.3%). These figures are lower than those reported in a Chinese study, where 83.6% of university students had regular meals, with 79% eating at least three times daily³², and higher than a Lebanese study that indicated 61.4% of students had regular daily meals²⁵. The regular consumption of breakfast by students is crucial for adequate energy intake to combat fatigue resulting from a demanding daily academic schedule³³. This study reveals that a considerable proportion of respondents do not consume meals three times daily (33.4%) and frequently omit breakfast (24.7%), indicating an unhealthy behaviour (Table 3). This finding is in line with past Indian studies highlighting that irregular meal consumption and frequent breakfast skipping are common among young populations in India, with breakfast being the most frequently omitted main meal³⁴. The regular intake of snacks and light meals is a notable characteristic of adolescent dietary habits³⁵. The current study revealed that most respondents (82.4%) consumed snacks three or more times weekly. All individuals consume cereals daily (100%). The majority consumes vegetables (66%) and pulses (78.4%) daily, surpassing the consumption rates in China (47.9%) and Bahrain (26.3%)^32,35.

Over half of the respondents (51%) consume fruits once or twice weekly, while a mere 6.6% consume them daily. Nonetheless, it has been shown that insufficient consumption of fruits and vegetables is linked to many chronic diseases in adulthood³⁶. The Indian Council of Medical Research (ICMR) advocates for daily milk consumption; nevertheless, the current study indicates that just 20% of individuals consume milk daily, with the majority (42%) consuming it once or twice a week. Prior research indicates that students frequently choose fast food because of its taste, accessibility, and convenience³⁷. A significant proportion of respondents, specifically 33.4%, consume fried food twice weekly or more. Furthermore, over 50% of respondents consume fried food once or twice weekly (55.4%). Most individuals used less than two liters of water everyday (81.3%). The daily intake of pulses, vegetables, fruits, and milk is higher among males (82.60%, 67.40%, 7.30%, and 26%) compared to females (74.0%, 64.7%, 6.0%, and 14.0%) correspondingly.

The analysis of dietary habits among male and female students revealed several statistically significant gender-based differences (Table 3). Regarding regular meal consumption, 68.6% of males and 64.6% of females reported having three regular meals daily (χ² = 5.07, p = 0.024), showing a modest but significant difference. Daily breakfast consumption was reported by 72.6% of males and 78.0% of females, reflecting a significant gender variation (χ² = 10.69, p = 0.001), with females displaying better breakfast habits. Snack consumption frequency showed a highly significant difference (χ² = 19.79, p < 0.001), where 21.3% of females reported consuming snacks less than three times per week, compared to only 14.0% of males, indicating that females had comparatively healthier snack habits. Cereal consumption was reported as 100% daily across both genders, with no difference observed. However, daily pulses consumption was higher among males (82.6%) compared to females (74.0%), a difference that was highly significant (χ² = 21.76, p < 0.001). Vegetable consumption patterns also varied significantly (χ² = 5.63, p = 0.018), though the difference was smaller, with 67.4% of males and 64.7% of females reporting daily vegetable intake. A sharper contrast appeared in fruit consumption, where only 7.3% of males and 6.0% of females consumed fruits daily, but females were more likely to consume fruits 3–4 times per week (47.4% females vs. 37.3% males), resulting in a significant difference (χ² = 22.24, p < 0.001).

Milk consumption patterns showed a strong gender difference (χ² = 39.65, p < 0.001), with 26.0% of males consuming milk daily compared to 14.0% of females. Conversely, females more frequently reported lower milk intake. Fried food consumption also differed significantly (χ² = 17.14, p < 0.001); 39.3% of females consumed fried foods 3–4 times per week compared to 30.0% of males, suggesting higher frequency of unhealthy food choices among females in this category. Water intake patterns showed that a larger proportion of females (84.6%) consumed less than 2 L of water daily compared to males (78.0%), with this difference also being highly significant (χ² = 16.47, p < 0.001). Overall, these results indicate that while some dietary practices like cereal consumption are consistent across genders, significant differences exist in other areas such as breakfast habits, milk and fruit intake, snack consumption, and fried food intake. This finding is in line with past Indian studies highlighting staples food like cereals are widely consumed in India, but other dietary behaviours such as milk/dairy intake, fruit and vegetable consumption, and snack choices differ substantially among groups, including by gender and behavioural patterns^38,39. These findings highlight the importance of gender-sensitive nutritional awareness and intervention programs that address these distinct patterns among university students.

The dietary consumption was evaluated against the recommended daily allowances for different food items (g/day) according to age group and lifestyle, referencing the nutrient requirements for Indians 2020. The daily energy requirement (Kcal) for boys over 18 years with a moderately active lifestyle is approximately 2680 Kcal, which should include a balanced diet comprising cereals and millets (370 g), pulses and beans (120 g), green leafy vegetables (100 g), other vegetables (200 g), roots and tubers (100 g), fruits (100 g), nuts (45 g), milk and milk products (300 ml), and fats and oils (40 g). The statistics indicated that the average food consumption among the boys was far below the required nutritional requirements (Table 4). Boys’ consumption of green leafy vegetables (31.44%), root and tubers (4.5%), milk and dairy products (13.80%), and fruits (10.44%) fell below the Recommended Dietary Intake. The daily energy requirement (Kcal) for girls over 18 years with a moderately active lifestyle is approximately 2125 Kcal, which may include a balanced diet comprising cereals and millets (270 g), pulses and beans (90 g), green leafy vegetables (100 g), other vegetables (200 g), roots and tubers (100 g), fruits (100 g), nuts (40 g), and milk and milk products (300 ml), along with fats and oils (30 g). For females, the consumption of pulses and beans (4.67%), green leafy vegetables (34.11%), other vegetables (7.07%), root and tubers (6.71%), milk and dairy products (18.04%), and fruits (20.41%) fell below the recommended dietary intake (Table 4).

Among males, consumption of cereals and millets, pulses, and fats and oils slightly exceeded the recommended intake, with statistically significant differences, suggesting a tendency toward over-consumption in certain categories. However, substantial under-consumption was noted for green leafy vegetables (− 31.44%), milk and milk products (− 13.80%), and fruits (− 10.44%), all of which showed highly significant differences (p < 0.0001), indicating critical nutritional gaps. Similarly, females also exhibited major deficiencies, particularly in the intake of green leafy vegetables (− 34.11%), pulses (− 4.67%), milk and milk products (− 18.04%), and fruits (− 20.41%). These results reflect poor dietary diversity and inadequate micronutrient intake, especially among females. The data exhibited statistical significance at the 1% and 5% levels (Table 4). Further, ANOVA results comparing actual intake between males and females revealed statistically significant differences across all food categories. Males consistently reported higher intake of cereals, pulses, vegetables, milk, fruits, and fats and oils than females, underscoring a gender disparity in dietary access and consumption. This difference may stem from socio-cultural practices, intra-household food distribution, or differences in occupational energy requirements. This finding are consistent with past Indian studies highlighting found statistically significant gender differences in actual intake levels of major food and nutrient categories, including energy, carbohydrates, proteins, and fats; with males typically consuming higher total energy and macronutrients compared to females^40,41. Collectively, the findings highlight the need for targeted nutritional interventions and gender-sensitive dietary policies to ensure adequate and equitable food intake, with a focus on improving the consumption of green leafy vegetables, fruits, and dairy products, especially among female.

Two psychological components, specifically ‘experience a loss of control while eating’ and ‘overeating to the point of discomfort’, along with two social factors, ‘eating due to maternal pressure’ and ‘eating due to paternal pressure,’ exhibited no significant correlation with the eating score (p > 0.05) (Table 6). The current study found a substantial association between smoking, alcohol intake, and the dietary habits of university students. Multiple Indian studies on similar lines also highlighted substance use among college students, including tobacco and alcohol consumption, and identifies associations with broader lifestyle behaviours. These studies found prevalence and correlates of substance use, and highlights how these behaviours cluster with other risk factors like stress, peer pressure, academic pressures, and lifestyle changes, which often include changes in dietary patterns and unhealthy eating behaviours among students in Indian university settings^42,43. Comparable results were also observed among Chinese university students³²; however, only male participants reported smoking (20.6%) and alcohol consumption (6%), as alcoholism and smoking among females are socially unacceptable in Indian society. Only 32.6% of respondents engaged in regular exercise. The eating behaviours score in this study was markedly lower among individuals who affirmed the following statements: “eat due to feelings of loneliness,” “eat in response to feelings of upset or anxiety,” and “eat out of boredom.” This finding aligns with the results of a study conducted on Malaysian university students³⁷. Prior research indicated that the behavioural ramifications of stress may influence dietary patterns⁴⁴. Individuals residing in a high-stress society often increase their food intake as a coping mechanism for stress. The current study reveals that a greater proportion of females (72.6%) consumed food due to feelings of loneliness compared to males (52%), whereas a higher percentage of males (77.4%) ate in response to feelings of happiness, in contrast to females (35.3%). A significant majority of females (71.3%) consumed food due to self-consciousness, in contrast to males (50.6%). Conversely, a higher percentage of males (68%) ate in the company of friends or family compared to females (39.3%). Television and radio food commercials had a greater influence on the eating behaviour of males (41.3%) than on females (27.3%), but cultural beliefs significantly impact the eating behaviours of females (52%) more than those of males (22.6%). Peer pressure influences eating patterns in boys and females at rates of 79.4% and 74.6%, respectively.

Under PCA Analysis, The Dietary Practices & Cultural Disengagement factor (PC1) demonstrated a significant negative association with BMI, with a regression coefficient of β = −1.54 (p < 0.001) and a bivariate correlation of r = − 0.51 (Table 8). This finding indicates that students adhering to healthier dietary practices such as regular consumption of milk and fruits and showing reduced cultural dietary constraints, tend to have lower BMI values. These dietary behaviors likely offer protection against overweight and obesity. However, it is important to interpret this association in the context of maintaining BMI within a healthy range; excessively low BMI may reflect potential risks of undernutrition, emphasizing the need for balanced eating patterns rather than dietary restriction alone. Although several associations reached statistical significance, the observed effect sizes were generally modest, and statistical significance should not be interpreted as evidence of strong practical or clinical importance.

Table 8.

Relationship of PCA-derived behavioural factors with BMI.

Association component	β coefficient	Standard error	t-value	p-value	Pearson’s r with BMI
PC1: Dietary practices & cultural disengagement	−1.54	0.03	−62.18	< 0.001	−0.51
PC2: External cues & emotional reactivity	+ 2.04	0.03	+ 81.64	< 0.001	+ 0.67
PC3: Socialized eating patterns	+ 0.01	0.03	+ 0.25	0.805	0.00
PC4: Structured eating vs. Loneliness	−0.02	0.03	−0.81	0.418	−0.01
PC5: Boredom-driven eating	+ 0.01	0.03	+ 0.22	0.827	0.00
PC6: positive affect & peer independence	−0.84	0.03	−32.12	< 0.001	−0.26
Constant	22.04	0.03	835.65	< 0.001	—

The External Cues & Emotional Reactivity component (PC2) exhibited a strong positive association with BMI, reflected by a regression coefficient of β = +2.04 (p < 0.001) and a bivariate correlation of r = + 0.67 (Table 8). This suggests that students whose eating behaviors are influenced by emotional triggers (e.g., anxiety, happiness) and external stimuli such as food advertisements are more likely to exhibit higher BMI, potentially due to increased consumption of calorie-dense, comfort foods. The Positive Affect & Peer Independence component (PC6) also showed a significant negative association with BMI, with β = −0.84 (p < 0.001) and r = − 0.26, indicating that students who eat independently of peer pressure and in response to positive emotions tend to maintain lower BMI levels (Table 8). This may reflect a more autonomous and emotionally balanced approach to eating. In contrast, PC4 (Structured Eating vs. Loneliness) and PC5 (Boredom-Driven Eating) were not significantly associated with BMI (p > 0.40 for both components) and exhibited negligible correlations (r ranging from 0.00 to − 0.01), suggesting limited observable relationships between these behavioral domains and BMI in this student population. These findings highlight the multifactorial nature of eating behavior and its impact on BMI. These findings are in line with earlier studies in Indian context, which shows that complex, multifactorial dietary behaviours (not single foods) correlate with BMI and related risk outcomes, highlighting nutrition’s interplay with lifestyle, socioeconomic and demographic influences^45,46. Healthy dietary routines and emotional self-regulation are protective against elevated BMI, while external environmental cues and emotional vulnerabilities are risk factors. Importantly, interventions aiming to promote healthy weight in university students should encourage balanced dietary habits that support optimal BMI, while preventing both excessive weight gain and undernutrition.

The scatterplot matrix (Fig. 1) displayed the bivariate associations between selected principal components and Body Mass Index (BMI) for 3,150 university students. Each component score and BMI followed an approximately normal distribution. The analysis highlighted that PC2 (External Cues & Emotional Reactivity) had the strongest positive correlation with BMI (r = + 0.77), indicating that students who tend to eat in response to emotional states and external environmental cues exhibited a significantly higher BMI. PC1 (Dietary Practices & Cultural Disengagement) showed a moderately strong negative correlation with BMI (r = − 0.58), suggesting that healthier eating patterns, such as daily milk and fruit consumption, and disengagement from restrictive cultural dietary practices, were associated with lower BMI values. Similarly, PC6 (Positive Affect & Peer Independence) demonstrated a weaker but still notable negative correlation (r = − 0.31) with BMI, indicating that students characterized by positive emotional states and lower peer pressure in eating habits maintained healthier weight profiles.

Fig. 1.

Scatterplot matrix depicting the associations between principal components of eating behavior and body mass index (BMI).

However, the other principal components—PC3 (Socialized Eating Patterns), PC4 (Structured Eating vs. Loneliness), and PC5 (Boredom-Driven Eating)—had negligible correlations with BMI (|r| < 0.05), suggesting limited direct influence on body weight. These findings visually support the earlier regression results, emphasizing the multidimensional behavioral determinants of nutritional status among university students.

Overall, the findings of the present study are broadly consistent with evidence from previous studies conducted among Indian university students, which have reported the coexistence of undernutrition and overweight, irregular meal patterns, low fruit and milk consumption, and gender differences in dietary practices. Similar studies from Indian higher education settings have documented frequent breakfast skipping, inadequate intake of fruits and vegetables, and increased reliance on energy dense foods, particularly during the transition to university life. The prevalence of emotional and socially influenced eating observed in this study also aligns with Indian research highlighting the role of stress, peer influence, and campus food environments in shaping students’ eating behaviours and nutritional status.

Conclusion

India is experiencing a rapid nutrition transition, with the adult obesity rate increasing more than threefold in recent decades and the annual rise in childhood obesity among the highest globally (World Obesity Federation, 2024). NFHS-5 (2019–21) reports that obesity among men aged 18–69 years increased to 22.9% from 18.9% in NFHS-4 (2015–16), and prevalence remains substantially higher in urban than rural areas (29.8% vs. 19.3% in males; 33.2% vs. 19.7% in females). Against this backdrop, the present cross-sectional study assessed nutritional status and behavioral determinants among 3,150 students from agricultural universities across India. Most participants were 18–21 years old and from nuclear, middle-income families. A dual burden of malnutrition was evident: 15.6% were underweight, while 36.3% were pre-obese or obese (WHO Asian BMI criteria). Diet quality showed important gaps. Although cereal intake generally met recommendations, consumption of nutrient-dense foods was low—only 20% consumed milk daily and 6.6% consumed fruit daily—significantly below Indian dietary guidelines (p < 0.001).

Psychosocial factors strongly shaped eating behavior. Principal Component Analysis identified six behavioral dimensions spanning dietary practices, emotional reactivity, socialized eating, structured meal patterns, boredom-driven eating, and peer independence, explaining 46.01% of variance. Healthier dietary practices were negatively associated with BMI, whereas emotional and advertisement-driven eating showed a strong positive association. These findings highlight the multifactorial nature of student nutrition, where diet quality, emotions, and social environments interact. Interventions should be holistic, combining nutrition education, mental well-being support, physical activity promotion, and reduction of alcohol and smoking. Gender-sensitive and context-specific strategies are warranted. Given the cross-sectional design, longitudinal and implementation research across diverse universities is needed before broad policy translation.

Implications of the study

Amidst the global transition towards industrialization and cultural transformation, the awareness and practice of healthy diets have declined in both emerging and developed economies. University students, who represent one of the most vulnerable demographic groups, are particularly affected by these shifts. Limited availability, affordability, and accessibility of nutritious food options on university campuses and in surrounding food environments have contributed to the adoption of unhealthy eating habits. This study provides a unique and comprehensive evaluation of the socio-demographic and physiological characteristics, dietary practices, and obesity prevalence among university students in India, along with an assessment of the social and psychological influences on eating behavior. Unlike many previous studies limited to specific regions, this research utilized primary data from the North, South, East, West, Central, and North-East zones of India, offering a pan-Indian perspective. Although the study relied on self-reported dietary intake data, the use of validated food frequency questionnaires and retrospective 24-hour recall methods enhanced the reliability of the dietary assessments. Nevertheless, future research could benefit from incorporating clinical nutritional assessments and biochemical markers, to provide more objective measurements of nutritional status.

The use of validated psychological and dietary behavior scales has ensured that the findings are based on credible behavioral constructs. However, further studies could deepen this understanding by exploring nutrient-specific intakes (e.g., calcium, iron, vitamins) and evaluating regional variability in food access and dietary patterns. Overall, the findings unequivocally demonstrate the critical role of social and psychological factors such as emotional triggers, peer influence, and environmental cues in shaping students’ eating behaviors and, consequently, their nutritional status. These insights carry practical implications for agricultural universities and potentially for higher educational institutions across India. University administrators, student affairs bodies, and campus health services can utilize these findings to design capacity-building and nutrition training programs, aligned with recommended dietary guidelines. Additionally, the study’s outcomes are valuable for policymakers and public health planners who aim to integrate psychosocial factors into national nutrition and health intervention frameworks for young adults. In the broader context of national development, these findings highlight the importance of investing in the nutritional health of India’s youth population. A healthy student population is essential for realizing the potential of India’s demographic dividend. Consequently, this study supports the development and implementation of evidence-based nutrition education initiatives that promote healthy, balanced eating habits and empower students to make informed dietary choices.

Limitations and future research directions

The study population was restricted to students enrolled in agricultural universities, which may differ from students in general, technical, or private universities in terms of academic environment, socioeconomic background, and food exposure. Therefore, the results are context specific and may not be generalizable to the broader Indian university student population. Future studies should include students from diverse academic disciplines and institutional settings to enhance generalizability. Binary response options were used to enhance clarity and consistency across study. The use of binary scoring for dietary, psychological, and social behavior items represents an important limitation of the study. While this approach facilitated standardized data collection across a large multi-regional sample, it may oversimplify complex behaviors and reduce variability, potentially attenuating or distorting underlying relationships. The use of PCA on binary variables represents a methodological limitation of the study. While PCA facilitated the identification of exploratory behavioral patterns, it relies on assumptions of linearity that may not be fully met with binary data. Consequently, the extracted components should be interpreted cautiously. Also, the PCA solution in present study explains a moderate proportion of the total variance and therefore the PCA findings are exploratory and descriptive rather than definitive. Future studies can explore alternative methods, such as tetrachoric correlation based factor analysis and should explore gender-specific challenges and strategies for more inclusive policy development. By addressing these limitations, future research can provide a more holistic understanding of the social and psychological factors influencing dietary patterns and help in development of targeted interventions for diverse communities.

Author contributions

GSM and PJ have done the conceptualization and design of the study. SB, PJ and GSM led the primary data collection from the respondents. SB, SS, Su.B led the data analysis and interpretation of data. S., GSM, PJ led the original draft preparation. RRB, S., GSM, PJ and SB led the editing and proofreading. All authors contributed to the article and approved the submitted version.

Funding

This research received no external funding.

Data availability

The data were derived from a primary dataset generated by the authors through field-based survey research. The datasets generated and analyzed during the current study are available from the corresponding author upon reasonable request. However, the data are not publicly available due to ethical considerations, including the need to safeguard participant privacy and confidentiality. Data sharing will be considered in accordance with applicable ethical standards and institutional guidelines.

Declarations

Competing interests

The authors declare no competing interests.

Ethical approval

All methods used in this study were conducted in accordance with the relevant guidelines and regulations. The study received approval from the Research Ethics Committee of the Division of Agricultural Extension, ICAR-Indian Agricultural Research Institute, New Delhi, India (Approval No. Ag Extn/2025/579(A), Date: 01 January, 2025). Informed consent was obtained from each participant, and their confidentiality was ensured throughout the study.