STRENGTH AND LIMITATIONS OF THIS STUDY

One of the strengths of this study was the use of a validated and reliable measurement tool, the health belief model (HBM).

This study clarifies the impact of all HBM components on obesity management practices among college students.

The study had a relatively large sample size, and by including students from both medical and non-medical colleges, we enhanced the generalisability of our findings to a broader student population.

Its cross-sectional design limits its ability to establish causal relationships.

Introduction

Body mass index (BMI) is a statistical measure of body fat based on height and weight in males and females of any age.1 The National Institute of Health (NIH) and WHO categorise a person as underweight, normal weight, overweight or obese based on their BMI.2 The global prevalence of overweight and obesity has reached epidemic proportions, affecting both developed and developing nations.3–5 According to the WHO, approximately 1.9 billion adults aged ≥18 years were overweight in 2016, with at least 650 million classified as obese.6 A critical period of risk for the development of obesity is the university phase of life, where changes in physical activity and eating patterns can lead to significant weight gain.7 8 In particular, Saudi Arabia is witnessing an alarming increase in overweight and obesity among young adults, posing a significant threat to public health.9 10 Maintaining healthy weight has become a vital health priority for university students because of these lifestyle changes. Several preventive and treatment strategies have been applied for weight control.11 However, adherence to treatment strategies varies among individuals for various reasons.11 12 Therefore, there is an urgent need to develop effective weight management strategies explicitly tailored to university students. To create effective weight management strategies for university students, it is essential to understand the factors that influence appropriate weight loss behaviour. Previous studies have demonstrated that psychosocial variables such as self-efficacy and beliefs about health and obesity play essential roles in the success of weight treatment strategies.13–15 The health belief model (HBM) is a psychosocial model that can be used to guide health promotion and disease prevention interventions.16 17 The six constructs of the HBM address an individual’s specific perceptions of susceptibility, benefits, barriers, cues and self-efficacy, which predict individual health-related behaviours.

HBM has been used in many previous studies as an important predictor of definite intention to take influenza and COVID-19 vaccines.18 19 The relationship between perceptions of obesity and their impact on obesity management remains relatively unexplained in the existing literature in Saudi Arabia. University students in Saudi Arabia exhibited a high perception of the severity of obesity and the benefits of physical activity towards obesity prevention in a previous study. However, only 50% of them believe they have a higher risk of obesity.20 The researchers used only four constructs of the HBM: perceived seriousness of obesity, perceived susceptibility to obesity, the perceived benefit of physical activity in preventing obesity and perceived barriers to physical activity in preventing obesity. The perceived benefits of a healthy diet and cues to action towards adopting a healthy diet were not assessed. Therefore, this study aimed to explore the significant HBM constructs that determine university students’ perceptions of obesity, to analyse their association with the BMI categories of the students and to determine their association with the success of behavioural interventions for obesity management. This study provides critical information to develop effective obesity management interventions for university students.

MethodStudy setting

This cross-sectional study was conducted with university students recruited from Jazan University between January and April 2023. Jazan University is one of the largest institutions of higher education in Saudi Arabia. The main central campus of Jazan University is located on the southwest coast of Saudi Arabia close to the Red Sea. Additional campuses were present in Abu Arish, Sabya, Samtah, Farasan, Ad-darb, Al-Daer and Al-Ardah. The university has 23 colleges.

Sample size and sampling technique

The StatCalc function of the Epi Info formula was used to estimate the sample size required for this study. A sample size of 526 was the minimum estimated sample size based on the following assumptions: population size of 60 000 students, 5% acceptable margin of error, 95% CI and expected prevalence of 50% of students with a high risk of obesity (as there are no available studies to estimate the prevalence among university students). With the addition of a 10% non-response rate, the final sample size was 579 participants. A multistage sampling technique was used to ensure participant recruitment. In the first stage of sampling, we divided the colleges into three strata (health sciences, sciences and humanities colleges). We then randomly selected six colleges to ensure the equal representation of health sciences, sciences and humanities colleges. Two colleges were selected for each group. Second, we ensured that the identified sample of participants represented different stages of students’ university routes. In the final sampling unit, we distributed the questionnaire to the selected classes while ensuring an equal representation of male and female students; see online supplemental appendix 1.

Participant selection and data collection

Data were collected using a self-administered questionnaire to measure the demographics and HBM components; see online supplemental appendix 2. The collected demographic details were related to the participants’ age, gender, college, year of study and social status. Students registered at Jazan University at the time of recruitment were included in this study. Those who were not Jazan University students were excluded from this study. Well-trained and qualified students distributed and handled the questionnaires. Informed written consent was obtained from all participants before the start of the study, and the questionnaires were completed while the students were present in their classes.

Measurement tool

The first part of the questionnaire focused on the baseline characteristics of the participants, such as age, gender, college, class, marital status, height (cm) and weight (kg), whether any members of the student’s family were obese, experience with diet therapy, experience with weight loss behaviour, experience with exercise therapy and experience with medical treatment. The second part of the questionnaire was for the key constructs of the HBM, including 89 statements.16 17 Each statement was rated on a 5-point Likert scale, ranging from 1 (strongly disagree) to 5 (strongly agree). Thirteen questions evaluated perceived severity on three subscales (emotional/mental, health, physical health/fitness and social professional). Seven questions evaluated perceived susceptibility and consisted of two subscales (lifestyle and environmental). Fourteen questions dealt with perceived barriers, comprising three subscales (practical concerns, emotional/mental health and awareness). Thirteen questions on perceived benefits included three subscales (emotional/mental health, physical health/fitness and social/professional). The 12 questions on cues to action consisted of two subscales (internal and external cues to action). Eighteen questions assessed self-efficacy in dieting and included two subscales (habits, preferences and emotional/mental health). The seven questions on self-efficacy in exercise and five questions on the behavioural intention of weight management consisted of two subscales (dieting and exercising). All statements were rated on a 5-point Likert scale, ranging from 1 (strongly disagree) to 5 (strongly agree).

Reliability and validity

A subgroup of 30 randomly selected students was asked to repeat the survey after a 2-week interval to assess the test–retest reliability of the questionnaire. The necessary changes were made: some questions were rephrased, others were cancelled, and the tools were reconstructed and ready for use. The intraclass correlation coefficient (ICC) was computed to evaluate the stability over time. The ICC indicated excellent agreement (ICC=0.86). The piloted samples were excluded from the study. Reliability was calculated using internal consistency (Cronbach’s alpha). Alpha coefficients equal to or higher than 0.70 were considered satisfactory.21 The overall reliability of the instrument based on Cronbach’s alpha was 0.92.

Statistical analysis

Data were revised, coded and analysed using the Statistical Package for Social Sciences, V.26. The characteristics and beliefs of the participants were described using means (±SD) and frequencies (percentages), wherever appropriate. Weight groups were divided into three categories: underweight (BMI <18.5 kg/m2), normal weight (18.5 ≤BMI <25 kg/m2), overweight (BMI 25–29.9 kg/m2) and obese (BMI ≥30 kg/m2) according to the NIH and the WHO classification for white, Hispanic, and black individuals.2 Cross-tabulation of BMI categories was performed using the paired t-test and the χ2 test. Based on the results of the Kolmogorov-Smirnov and Shapiro-Wilk tests, the BMI variable did not follow a normal distribution. This information is relevant for understanding the distributional properties of BMI data and may have implications for subsequent statistical analyses or modelling approaches. Therefore, we opted to use the χ2 test because the BMI variable did not follow a normal distribution. Multivariate logistic regression analysis was performed. Gender, age, college year, marital status, income, perceived severity, cue to action, perceived susceptibility, perceived barriers, perceived benefits, perceived self-efficacy in exercise and perceived self-efficacy in diet were used to explain the variability in the behavioural intention of weight management (dependent variable). Each variable was included individually to obtain the crude OR using logistic regression models. Statistical significance was set at p <0.05.

Patient and public involvement statement

Students were involved in the pilot study to enhance the reliability of the questionnaire, and actively participated in the training session before data collection. The results of this study will be disseminated to students as a reflection of their perception of overweight and obesity status and will be involved in the implementation of future obesity management interventions.

ResultsDemographic and baseline information

A total of 579 undergraduate students participated in this study and completed the entire online survey. Based on self-reported weight and height data, the mean BMI was 24.21 (±6.52; range, 12.70–65.67) kg/m2. The baseline characteristics of the respondents based on the BMI categories are presented in table 1. The percentages of individuals who were underweight, normal-weight, overweight and obese were 16.4%, 47.9%, 19.9% and 15.9%, respectively. More than half the participants were from health colleges. Most of the students (85.5%) were single. They belonged to the age group of 21–23 years (58.7%). A highly significant relationship (p <0.001) was found between the participants’ BMI category and the gender. Compared with the percentage of female students in the same categories, the percentage of male students who were overweight or obese was much higher. There was a significant relationship between a self-reported family history of obesity and BMI category (p <0.001). Age, marital status, income and undergraduate institution did not show significant associations (p >0.05) with BMI categories. Half of the participants (59.2%) reported being engaged in physical activities. Diet therapy is a term used to describe the adoption of a diet to simply promote optimum weight. In this study, there was a statistically significant difference (p <0.001) between the participants’ experiences with diet therapy and their weight categories. Similar results were observed between BMI categories and experience with weight loss behaviour and exercise therapy (p <0.001), as shown in table 1. Most of the students, regardless of their BMI categories, agreed that their motivation to maintain healthy weight was to improve their health and appearance.

Table 1

The association between the BMI categories and the demographic and baseline characteristics of the participants

HBM constructs and intention scale among BMI categories

Reliability (internal consistency) was calculated using Cronbach’s alpha. Table 2 shows the alpha coefficients for all HBM constructs and intention scales. All scales had values higher than 0.67. Based on Cronbach’s alpha, the overall reliability of the HBM is 0.97. The mean scores of the total perceived severity scale of the whole group were 3.38 ±0.03 for all respondents, which showed significant differences among the three groups (p <0.001). Students in the underweight category showed the highest mean score for perceived severity (3.62 ±0.08). The emotional and mental health subscale beliefs received higher ratings than the other severity subscales (3.55 ±0.04). Underweight and normal-weight students rated the emotional/mental health subscale higher than the overweight group (p <0.01). Moreover, the mean physical health/fitness scores and social/professional subscales showed significant differences among the three groups (p <0.05). Regarding the perceived susceptibility to obesity risk, the total number of respondents yielded a mean score of 3.47 ±0.04. The students who were considered underweight had the highest average score (3.61 ±0.10); nevertheless, there were substantial disparities between each of the four categories. The four categories of the environmental subscale for perceived susceptibility showed no significant difference (p =0.111). The mean score for the perceived barriers scale was 3.42 ±0.04, with statistically significant variations between the BMI categories. The underweight category showed the highest levels of perceived barrier (3.58 ±0.09), followed by the normal weight category (3.46 ±0.05) and the underweight category (3.31 ±0.08). The mean score for the perceived benefits of adopting healthy food and exercise habits was 3.39 ±0.05. Significant differences were found between the four subgroups on the social/professional subscale (3.61 ±0.13). The mean score for perceived cues to action for obesity management was 3.50 ±0.04. The normal-weight category achieved the highest score (3.63 ±0.10), but there were no notable differences between the four groups. The study groups did not differ in their mean ratings of internal cues to action. The student groups’ mean ratings of external cues to action differed significantly (p <0.05). The mean rating of self-efficacy in dieting was 3.47 ±0.04, which showed no significant differences among the four groups (p =0.194), as students in the underweight category showed the strongest belief about their self-efficacy in dieting (3.55 ±0.11), followed by students in the normal-weight (3.45 ±0.06), obese (3.32 ±0.11) and overweight categories (3.31 ±0.09). Habits/preferences and emotional/mental health subscales showed no significant differences among BMI categories. Moreover, perceived self-efficacy on the exercise scale showed no significant differences. The overall mean score for behavioural intention of obesity management was 3.38 ±0.05. According to the findings, students planned to manage their weight by dieting rather than engaging in physical activity. The four groups did not differ significantly from one another in terms of the overall mean score of the behavioural intention of the obesity management scale, nor did they differ in terms of either of the two subscales.

Table 2

HBM constructs and intention scale among BMI categories

Multivariate logistic regression

Table 3 shows the variability in the behavioural intentions of the management of obesity. Gender, age, college year, marital status, income, perceived severity, cue to action, perceived susceptibility, perceived barriers, perceived benefits, perceived self-efficacy in exercise and perceived self-efficacy in diet were used to explain the variability in the behavioural intentions of obesity management (dependent variable). Each variable was included individually to obtain the crude OR using logistic regression models. Except for age, college, year, marital status and income, all variables showed significant associations with behavioural intention of obesity management. However, in multivariate logistic regression analysis, some variables showed a significant association. Perceived self-efficacy in exercise and diet was significantly associated with BMI categories, with adjusted ORs of 2.82 (2.10 to 3.79) and 1.51 (1.09 to 2.09), respectively. The data used met the assumptions of the logistic regression. The Hosmer-Lemeshow test showed a good fit of the data to the logistic regression (χ2 =9.98, p =0.267).

Table 3

Multivariate logistic regression analysis of variables that influence behavioural intentions of obesity management

Discussion

Body weight loss and, in particular, maintaining an ideal body weight, are challenging because of the difficulty of permanently altering lifestyle choices instead of temporarily. It is essential to predict and manage obesity in its early stages to prevent the development of chronic diseases. Increasing physical activity and maintaining a balanced diet are two popular behavioural approaches to overcome obesity.22 The primary goal of this study was to apply the HBM as a framework to determine the predictors of behavioural intentions around obesity management among students at Jazan University. Successful adoption of lifestyle choices for the sake of obesity management requires comprehensive strategies, individual actions and integration between stakeholders at local, national and international levels. Perceptions of responsibility are likely to play a role at various levels, including society, governmental policies, healthcare planning and social work, prevention organisations and food and marketing industries.

Our study revealed a significant association between gender and self-reported of existence of at least one obese family member and the BMI categories of students. These findings are consistent with previous research, underscoring the complex interplay between genetic and environmental factors in obesity.20 23 24 However, the complexity of the relationships and confounders of other socioeconomic indicators could account for this heterogeneity. Extensive quantitative studies and adjusted models are required to reflect the specific attributes of socioeconomic status.

Nearly one-third of the students in this study reported attempting to lose weight. This experience differed significantly between the BMI categories. Almost half of the students reported having previously attempted to control their diet and exercise to lose weight, while a few reported having tried medical therapy. This result is consistent with other studies that examined weight loss strategies among youth and university students.23 25–27 This consistency across different populations underscores the global nature of the obesity challenge and the widespread need for effective, evidence-based obesity management strategies.

The motivating factor for weight control among the students in this study was a combination of health and appearance. This finding diverges from previous studies that found health or appearance to be the primary motivator.23 26 28 This discrepancy may reflect cultural differences or shifts in societal attitudes towards health and body image. Health promotion strategies should consider the diverse motivations for weight control and aim to foster healthy attitudes towards body image.

Perceptions about the severity of obesity varied among participants, with students in underweight and normal-weight categories rating the emotional/mental health subscale higher than students in overweight category did. This is consistent with prior research indicating that individuals with obesity often have a lower perceived severity of their weight status.23 However, this contrasts with other studies, in which most patients with obesity seeking help recognised obesity as a disease and were aware of the risks associated with obesity.29–31 This discrepancy emphasises the need for education and awareness campaigns to enhance understanding of the health risks associated with obesity.

In terms of self-efficacy, our study found that underweight students had the strongest beliefs in their self-efficacy regarding dieting and exercise, followed by students who are normal, obese and overweight. However, no significant differences were observed between the BMI categories. This aligns with previous research that found that self-efficacy plays a crucial role in adopting new health activities.23 32 33 This suggests that interventions to improve obesity management should consider strategies to boost self-efficacy, particularly among students who are overweight and obese.

Perceived barriers to healthy eating and regular physical activity significantly influenced the behavioural intentions of obesity management. This is consistent with previous research, indicating that practical concerns often hinder adherence to healthy behaviours.34 35 These findings highlight the need for interventions that address practical barriers to healthy eating and physical activity, such as time constraints, lack of access to healthy foods or exercise facilities, and lack of knowledge about healthy behaviours.

Finally, our multivariate logistic regression analysis revealed that perceived severity, cues to action, barriers and self-efficacy in dieting and exercise were significant predictors of behavioural intentions for obesity management. Self-efficacy in dieting and exercise emerged as the most significant predictor among the independent variables. This finding underscores the importance of self-efficacy in adopting and maintaining healthy behaviours, aligning with the theoretical underpinnings of the HBM and previous empirical research.23 36

There was no substantial difference across the BMI categories regarding the perceived benefits of adopting healthy dietary and exercise habits. Emotional and mental health benefits were rated higher than other subscale constructs. This finding aligns with a previous study that found stronger beliefs about these benefits in underweight and normal-weight categories.23 This emphasises the positive psychological and physical health outcomes of nutrient-rich diets and regular physical activity. Perceived cues to action were linked to obesity management in our study. Normal-weight students reported high ratings for these cues, although the differences across BMI categories were not statistically significant. However, external cues were rated significantly higher than internal cues, a departure from a previous study that found that both cues mediate the relationship between cues to action and obesity management intentions.23 This discrepancy suggests the need to further explore how internal and external cues, which relate to self-esteem and the facilitation of task-specific rules, respectively, contribute to achieving personal and professional goals. Multivariate logistic regression identified perceived severity, cues to action, barriers and self-efficacy in dieting and exercise as significant predictors of obesity management intentions. Among these, self-efficacy in dieting and exercise emerged as the most influential factor.

Conclusions

This study applied the HBM to understand behavioural intentions around obesity management in Jazan University students. The results highlight the complex interplay between gender, self-reported family history and self-efficacy in diet and exercise. Health promotion strategies should focus on enhancing the awareness of obesity risks, addressing perceived barriers and boosting self-efficacy. Discrepancies with previous research suggest cultural and societal influences on health and body image attitudes, indicating the need for tailored intervention strategies. Future research should consider longitudinal studies and explore additional factors influencing obesity management behaviours. Based on these findings, this study provides recommendations to various stakeholders. University administrators are encouraged to implement educational programmes and initiatives that raise awareness about the health risks associated with overweight and obesity. Health science colleges can integrate obesity prevention and management courses into their curricula and provide training on the application of the HBM for obesity management interventions. Science and humanities colleges should promote interdisciplinary approaches and collaborations to address obesity management issues comprehensively. Student support services can offer counselling and support groups to address psychological and emotional aspects related to body image and self-esteem while providing resources and guidance on nutrition and exercise.

Data availability statement

Data are available upon reasonable request. ‘Not applicable’.

Ethics statementsPatient consent for publication

Consent obtained directly from patient(s)

Ethics approval

This study involves human participants and was approved by Standing Committee for Scientific Research—Jazan University (HAPO-10-Z-001), Saudi Arabia (Reference No.: REC-44/06/463). Participants gave informed consent to participate in the study before taking part.

Acknowledgments

The authors are grateful to the participants of this study. The authors extend their appreciation to the Deputyship for Research and Innovation, Ministry of Education in Saudi Arabia for funding this research work through the project number ISP23-47.