Abstract
Introduction
The burden of diseases related to overweight/obesity is rapidly increasing, especially in low-and-middle-income countries. University students, especially females, are at high risk of being overweight or obese, predisposing them to non-communicable diseases. This study primarily assessed knowledge and prevalence of overweight and obesity; and secondarily factors associated with Body Mass Index (BMI) categories among female undergraduate students of Makerere University Kampala.
Methods
A cross-sectional study was conducted among 380 female undergraduate students, proportionately sampled from all academic units. Data were collected using self-administered questionnaires followed by weight and height measurements to compute Body Mass Index. Knowledge was scored on (1) food characteristics, (2) predisposing practices and environments, and (3) consequences of overweight and obesity; and categorised to adequate (≥70%) and inadequate (<70%). Descriptive statistics and ordered logistic regression were used for analysis.
Results
The response rate for the study was 93.3%. The prevalence of overweight was 24.7% (95% CI 20.5% to 29.4%) while that for obesity was 6.8% (95% CI 4.5% to 9.9%). About 74% (95% CI 69.5% to 78.5%) were knowledgeable about foods, 18.4% (95% CI 14.8% to 22.7%) were knowledgeable about practices and environments that promote obesity or overweight, and 19.7% (95% CI 16.0% to 24.1%) were knowledgeable about consequences of being overweight or obese. The factors associated with BMI categories were being married (adjusted OR (aOR)=1.94, 95% CI 1.05 to 3.57, p=0.034), having adequate knowledge of foods that promote obesity and overweight (aOR=0.58, 95% CI 0.36 to 0.93, p=0.023) and being in higher years of study compared with first years (third year aOR=2.82, 95% CI 1.46 to 5.47, p value=0.002; fourth year aOR=2.99, 95% CI 1.36 to 6.55, p value=0.006).
Conclusions
The prevalence of overweight and obesity in female undergraduate students is high. This calls for designing strategies to control the problem especially targeting higher risk groups like the married and students in higher years of study. Additionally, the proportions of students knowledgeable about practices that promote obesity and overweight as well as its consequences are very low, suggesting a need for strengthening.
Keywords: Public Health, Prevalence, Body Mass Index
WHAT IS ALREADY KNOWN ON THIS TOPIC
There is increasing prevalence of overweight/obesity and related diseases globally, especially among females. University students have been reported in some studies to have considerable magnitudes of overweight/obesity and yet also have poor knowledge with misinformation limiting the control of the problem. Overweight and obesity are likely to vary in different contexts owing to differences in diets, genetic predisposition, knowledge and perceptions about overweight/obesity, and the practices that persons engage in, hence, the need for this study among female students in a low-resource setting.
WHAT THIS STUDY ADDS
This study re-enforces findings of high magnitude of being overweight or obese in this generally young group of people. There is also low knowledge about its promoters and consequences. The study further strengthens findings of relationships between being in higher categories of Body Mass Index (BMI) and being married, in higher years of study, and having low knowledge of foods that predispose to being overweight or obese.
HOW THIS STUDY MIGHT AFFECT RESEARCH, POLICY OR PRACTICE
The findings of this study call for university leadership at various levels to conduct awareness programmes to address the low levels of knowledge. University policies to support the dissemination of information on overweight/obesity risks, consequences and its control ought to be designed. Further research is needed to identify the most feasible and accepted strategies to address overweight or obesity.
Background
The global prevalence of overweight and obesity among adults increased from 25% in 1990 to over 43% in 2022.1 Among adult females, the global prevalence of overweight and obesity increased to 44% in 2022,1 compared with 40% in 2014.2 In Uganda, the prevalence of high Body Mass Index (BMI) and obesity among adults in 2025 were 26% and 9% respectively. Adult females had double the prevalence of high BMI compared with males (34% vs 17%).3 In 2021, overweight and obesity accounted for 15% of deaths from known non-communicable diseases (NCDs), and 27% of ill-health from known NCDs.3 Low-income countries will face the brunt of overweight and obesity-related diseases amidst the high burden of infectious and undernutrition diseases.4
According to the WHO, overweight and obesity are responsible for 40% of type 2 diabetes, 23% of ischaemic heart disease and 41% of cancer cases.5 In addition, they are also associated with hypertension, osteoarthritis, sleep apnoea, low self-esteem, decreased life expectancy and quality of life.6 Furthermore, overweight and obesity have considerable economic impact in several countries.7 8
Several risk factors have been associated with overweight and obesity, including diet, genetics, physical activity, physiological and behavioural factors.9 If no interventions are made, it is estimated that 50% of the world’s adult population will have high BMI with obesity at 22% among females and 17% among males by 2030.3
A study among college students at Sohag University in Egypt found that 17.2% of study group were obese. A majority of participants thought that obesity is a hormonal disease, or psychological or an infectious disease. The study reported that less than 50% of obese persons agreed to have marriage with another obese person.10 This not only indicates a high prevalence of obesity among young adults, but also misinformation and poor attitudes about the condition that undermine efforts to manage and/or prevent it.
University students may be at increased risk of overweight and obesity due to unhealthy food environments that result in unhealthy diets like excessive calorie intake and low fruit and vegetable consumption.11 The students are more likely to depend on food prepared in restaurants or other food outlets due to busy schedules that prevent food preparation, limited availability of cooking facilities in their places of residence, and the easy access to food prepared for purchase. Such food is likely to be dense as students attempt to minimise costs.11 In addition, students may spend more of their leisure time on video games and watching television.12 Female students are at an even higher risk of having high BMI compared with males in tandem with population trends of higher overweight and obesity among females.13 Ignorance among university students on causative factors of overweight and obesity and their health consequences may predispose them to the risk of becoming overweight and obese while at school or in the future after school. Overweight and obesity put females at increased risk of diabetes, cardiovascular diseases, polycystic ovarian syndrome and cancers.14 15 In addition, they may experience the social consequences like stigma.16 There was need to provide information that can be used to address the problem of overweight and obesity in the context of rapidly changing lifestyles in low-resource settings. University students form a special population that nations are looking to for future development but if they are unhealthy, they present a future burden to already constrained health systems. This study therefore primarily assessed knowledge and prevalence of overweight and obesity among female undergraduate students of Makerere University. In addition, we secondarily assessed the factors associated with BMI categories among these students.
Methods
Study design and setting
A cross-sectional study was conducted among female undergraduate students from all academic units at the main campus of Makerere University. Makerere University is the oldest tertiary institution in Uganda and remains one of the leading higher institutions of learning in Africa. The university has a population of about 14 424 female undergraduate students at the main campus. The main campus is located 3 km from Kampala city centre along Bombo Road. It houses most of the academic units of the university and draws students from all regions of Uganda and internationally.
The main campus has 10 academic units (nine colleges and one school), that is, College of Agriculture and Environmental Sciences (CAES); College of Business and Management Sciences (CoBAMS); College of Computing and Information Sciences (CoCIS); College of Education and External Studies (CEES); College of Engineering, Design, Art and Technology (CEDAT); College of Health Sciences (CHS); College of Natural Sciences (CoNAS); College of Veterinary Medicine, Animal Resources and Biosecurity (CoVAB); College of Humanities and Social Sciences (CHUSS); and the School of Law. Some of the students are housed in halls of residence belonging to the university and are thus provided with meals while others are housed in hostels mainly located in the areas surrounding the university. Some students commute from their homes.
Study participants and sampling
We enrolled female undergraduate students from each of the 10 academic units at the main campus of Makerere University from March to April 2019. The participants gave informed consent to participate in the study. Quota sampling was used to get participants from the different academic units. The number of participants that were sampled from each academic unit was based on the total number of female undergraduate students in each unit. The number of female undergraduate students in each academic unit at the time of the study was 921 in CAES, 2151 in CoBAMS, 1461 in CoCIS, 3046 in CEES, 1376 in CEDAT, 704 in CHS, 482 in CoNAS, 494 in CoVAB, 554 in the School of Law and 3235 in CHUSS. Thus, 26 students were sampled from CAES; 61 from CoBAMS; 41 from CoCIS; 85 from CEES; 39 from CEDAT; 20 from CHS; 14 from CoNAS; 14 from CoVAB; 16 from the School of Law; and 91 from CHUSS, making a total of 407 students. Convenience sampling was used to select participants from each academic unit. The students would be accessed in the mornings in the lecture rooms for different classes before the lectures commenced and those who agreed to participate were enrolled.
Sample size estimation
The sample size was estimated using the formula for single proportions.17 The assumptions included 95% level of confidence, 5% tolerable sampling error, and prevalence of overweight and obesity of 50% to give maximum possible sample size.18 This gave sample size of 385 which was adjusted for missing data of about 5%. However, of the 407 students that were selected, the dataset had complete data (questionnaires plus weight and height measurements) from 380 participants.
Variables
The study had two dependent variables: (1) knowledge about overweight and obesity; (2) presence of obesity or overweight for the assessment of prevalence of overweight and obesity. In the regression analysis for associated factors, the primary outcome was BMI category (ordered logistic model). Knowledge domains were modelled as independent variables alongside covariates. Prevalence of BMI categories and knowledge levels were described as summary outcomes. Knowledge was assessed on food characteristics, practices and environment that promote overweight and obesity. Knowledge of food characteristics was assessed using four multiple choice questions. Knowledge regarding practices and environment that promote obesity and overweight; and knowledge of consequences of obesity and overweight were each assessed using 10 multiple choice questions. Each correct answer was given a score of one and a percentage score was computed. A score of 70% and above for each of the three knowledge sections was considered adequate knowledge. This cut-off has been used in previous studies where it is referred to as modified Bloom cut-off.19
Overweight and obesity were assessed from calculation of BMI. Participants whose BMI was <18.5 were considered as underweight; 18.5–24.9 were considered as normal weight; 25–29.9 were considered as overweight; and >=30 were considered as obese.20 In the analysis for associated factors, the categories of underweight and normal BMI were combined. Thus, the outcome variable was a three-level ordinal variable with underweight/normal, overweight and obesity categories.
The independent variables were marital status, year of study, knowledge of foods, practices and environments that promote overweight/obesity, knowledge of consequences of overweight and obesity, and amount of money spent on meals. The amount of money spent on meals per day was categorised based on average costs per meal in restaurants and food vending places around the university. The categories created are indicative of whether one is likely to be having meals from low-cost or high-cost food sources and possibly the number of meals they are likely to be having.
Data collection
Two research assistants were trained on the purpose of the study, the data collection process and ethical considerations before starting the data collection process. Data were collected using a structured questionnaire with knowledge questions formulated from several studies.21,26 It was a 24-item tool (online supplemental file 1). The separate tools from which the questions were derived had been validated in those studies but the tool had not been validated as a combined tool. Participants were to select one of three alternative answers, that is, true, false or I don’t know for each question, of which one was correct. Each right answer was scored one point, a wrong answer was scored zero points. The scores ranged from 0 to 24 points. A percentage score was computed separately for each of the three knowledge areas assessed. The questionnaire also captured demographic information. Height and weight measurements were taken after filling the questionnaire by one data collector to ensure consistency of measurement. Participants’ contacts to which the correct answers for this questionnaire were to be sent were recorded. The answers were sent after conclusion of the data collection process to raise awareness about causative factors and consequences related to overweight and obesity and to clear up any related misconceptions.
This questionnaire was pretested among 20 male undergraduate students from the same university, 2 weeks before starting the data collection process. Analysis of the results from pretesting was carried out and appropriate modifications were done to improve clarity of questions in the final questionnaire meant for data collection. The pilot testing was done among males to avoid having persons that had participated in the pilot test recruited in the main study. Since the questionnaires used did not have any dimensions that would likely differ between males and females, we anticipated that doing the pilot test among males would be adequate.
This questionnaire was delivered to each of the selected students in the respective academic units by researchers and research assistants early in the morning before the start of lectures to minimise errors which could arise due to fatigue. The questionnaire was written in English since it is the official language at Makerere University. The respondents were required to return the questionnaires immediately when they finished answering all the questions. Anthropometric measurements which included weight and height were taken for all participants in light clothing and no shoes. Weight was measured using digital weighing scales. These were calibrated at the start of the study and each time they were moved from one place to another. Height was taken using measuring tape attached to the wall with participants standing without shoes, looking straight with heels together. BMI was calculated as the ratio of weight in kilograms to the square of height in metres.
Data analysis
Data were entered using Epidata. Two data entrants were employed and the two sets of data entered were compared for reproducibility and any discrepancies were rectified. The cleaned dataset was exported to Stata version 15.0 (StataCorp, College Station, TX, USA). The performance of the tool was assessed using the Cronbach’s alpha for reliability; the Delta Index for item discrimination; and the Item Difficulty Index. Descriptive statistics were used to describe participants, determine knowledge and prevalence of overweight and obesity. Categorical data were presented as percentages. Prevalence of overweight and obesity, and knowledge were presented with 95% CIs. Bivariable analysis was done to assess association of each independent variable with BMI categories. Factors that had p values of less than 0.2 at bivariable analysis were considered for multivariable analysis. Interaction was assessed using likelihood ratio tests comparing reduced and full models considering two-way interaction terms formed between variables that were considered for multivariable analysis. The analysis showed no significant statistical interaction. Confounding was assessed by comparing unadjusted and adjusted ORs with a difference of 10% or more considered confounding. Bivariable and multivariable analyses were done using ordered logistic regression. The data were checked for fulfilment of the proportional odds assumption using both the ‘omodel’ and brant command, and they both showed the assumption was fulfilled. The level of significance was 5%. Sensitivity analysis was done for the knowledge variables to assess if it was preferable to analyse them as continuous variables, or as categorised variables using a cut-off of 60% for adequate knowledge, or as categorised variables using a cut-off of 80% for adequate knowledge. These alternate ways of handling the knowledge variables did not alter the conclusions; hence, we chose to proceed with the 70% cut-off that had been prespecified.
Patient or public involvement in research
Patients or the public were not involved in the design, conduct or dissemination plans for our research.
Results
Participant flow
All 407 undergraduate students who were approached consented to participate in the study. However, 380 filled the questionnaires representing 93.3% response rate. All 380 students who returned the questionnaires agreed to have their weight and height measured and were therefore all included in the analysis.
Socio-demographic characteristics of the participants
About a third of the participants (33.2%, 126/380) were in year two, and majority (90.0%, 342/380) were single. Almost half (42.5%, 161/380) of the participants spend between 5000 (equivalent to US$1.5) and 7000 (equivalent to US$2) per day on meals. Sociodemographic characteristics of the participants are summarised in table 1.
Table 1. Social demographics of 380 female undergraduate students from Makerere University.
| Characteristic | Frequency (n=380) | Percentage (%) |
|---|---|---|
| Year of study | ||
| 1 | 80 | 21.1 |
| 2 | 126 | 33.2 |
| 3 | 117 | 30.8 |
| 4 | 57 | 15.0 |
| Marital status | ||
| Married | 38 | 10.0 |
| Single | 342 | 90.0 |
| Money spent on meals* | ||
| 0.000–5000 | 122 | 32.2 |
| 5001–7000 | 161 | 42.5 |
| 7001–10 000 | 74 | 19.5 |
| 10 001–15 000 | 22 | 5.8 |
Approximately 3500 Uganda shillings=US$1.
Prevalence of overweight and obesity
The prevalence of overweight was 24.7% (95% CI 20.5% to 29.4%) while that for obesity was 6.8% (95% CI 4.5% to 9.9%) as summarised in table 2.
Table 2. Body Mass Index levels of 380 female undergraduate students from Makerere University.
| BMI level | Frequency (%) n=380 |
95% CI |
|---|---|---|
| <18.5 (underweight) | 16 (4.2) | 2.4% to 6.7% |
| 18.5–24.9 (normal weight) | 244 (64.2) | 59.2% to 69.0% |
| 25.0–29.9 (overweight) | 94 (24.7) | 20.5% to 29.4% |
| >=30 (obese) | 26 (6.8) | 4.5% to 9.9% |
BMI, Body Mass Index.
Level of knowledge among female undergraduate students
About 74% (95% CI 69.6% to 78.4%) had adequate knowledge regarding the foods that promote overweight and obesity. However, only 18.4% (95% CI 14.8% to 22.7%) had adequate knowledge of practices that promote overweight and obesity and 19.7% (95% CI 16.0% to 24.1%) had adequate knowledge of the consequences (table 3).
Table 3. Knowledge of overweight and obesity among 380 female undergraduate students from Makerere University.
| Characteristic | Frequency (%) n=380 |
95% CI |
|---|---|---|
| Knowledge of foods that promote obesity/overweight | ||
| Adequate | 282 (74.2) | 69.6 to 78.4 |
| Inadequate | 98 (25.8) | |
| Knowledge of practices that promote obesity/overweight | ||
| Adequate | 70 (18.4) | 14.8 to 22.7 |
| Inadequate | 310 (81.6) | |
| Knowledge of consequences of obesity/overweight | ||
| Adequate | 75 (19.7) | 16.0 to 24.1 |
| Inadequate | 305 (80.3) | |
Factors associated with BMI categories
Marital status (adjusted OR (aOR)=1.94 (95% CI 1.05 to 3.57, p=0.034)), having adequate knowledge of foods that promote obesity and overweight (aOR=0.58, CI 0.36 to 0.93, p=0.023), and year of study (third year compared with first year, aOR=2.82, CI 1.46 to 5.47, p value=0.002; fourth year aOR=2.99, 95% CI 1.36 to 6.55, p value=0.006) were significantly associated with BMI categories at multivariable analysis (table 4).
Table 4. Factors associated with BMI categories based on ordered logistic regression among 380 female undergraduate students in Makerere University.
| Characteristic | BMI categories | Unadjusted OR (CI) | P value | Adjusted OR (CI) | P value | ||
|---|---|---|---|---|---|---|---|
| Normal* | Overweight | Obese | |||||
| Marital status | 0.034 | ||||||
| Single | 20 (52.6) | 15 (39.5) | 3 (7.9) | 1.00 | 1.000 | ||
| Married | 240 (70.2) | 79 (23.1) | 23 (6.7) | 1.95 (1.05 to 3.61) | 0.034 | 1.94 (1.05 to 3.57) | |
| Average amount spent on food daily (Uganda shillings)† | |||||||
| 0.000–5000 | 83 (68.0) | 31 (25.4) | 8 (6.6) | 1.00 | |||
| 5001–7000 | 111 (68.9) | 40 (24.9) | 10 (6.2) | 0.96 (0.58 to 1.58) | 0.865 | ||
| 7001–10 000 | 52 (10.3) | 16 (21.6) | 6 (8.1) | 0.93 (0.49 to 1.74) | 0.818 | ||
| 10 001–15 000 | 13 (59.1) | 7 (31.8) | 2 (9.1) | 1.46 (0.59 to 3.60) | 0.408 | ||
| Year of study | |||||||
| I | 64 (80.0) | 11 (13.8) | 5 (6.2) | 1.00 | 1.00 | ||
| II | 98 (77.8) | 25 (19.8) | 3 (2.4) | 1.08 (0.54 to 2.18) | 0.826 | 1.01 (0.50 to 2.04) | 0.973 |
| III | 65 (55.6) | 43 (36.7) | 9 (7.7) | 2.95 (1.52 to 5.74) | 0.001 | 2.82 (1.46 to 5.47) | 0.002 |
| IV | 33 (57.9) | 15 (26.3) | 9 (15.8) | 3.12 (1.40 to 6.96) | 0.005 | 2.99 (1.36 to 6.55) | 0.006 |
| Knowledge of foods that promote overweight/obesity | |||||||
| Inadequate | 58 (59.2) | 33 (33.7) | 7 (7.1) | 1.00 | 1.00 | ||
| Adequate | 202 (71.6) | 61 (21.6) | 19 (6.8) | 0.61 (0.38 to 0.96) | 0.033 | 0.58 (0.36 to 0.93) | 0.023 |
| Knowledge of practices that promote overweight/obesity | |||||||
| Inadequate | 211 (68.1) | 78 (25.2) | 21 (6.8) | 1.00 | |||
| Adequate | 49 (70.0) | 16 (22.9) | 5 (7.1) | 0.93 (0.53 to 1.63) | 0.788 | ||
| Knowledge of consequences of overweight/obesity | |||||||
| Inadequate | 207 (67.9) | 77 (25.2) | 21 (6.9) | 1.00 | |||
| Adequate | 53 (70.7) | 17 (22.6) | 5 (6.7) | 0.88 (0.51 to 1.53) | 0.661 | ||
Interactions that were assessed with p values for likelihood ratio tests included: marital status × knowledge of foods that promote overweight/obesity (p=0.586), marital status × year of study (p=0.443), and knowledge of foods that promote overweight/obesity × year of study (p=0.264).
Comprised of normal BMI combined with underweight.
Approximately 3500 Uganda shillings=US$1.
BMI, Body Mass Index.
Discussion
Summary of key findings
The prevalence of overweight and obesity in the study population was 24.7% (95% CI 20.5% to 29.4%) and 6.8% (95% CI 4.5% to 9.9%) respectively. About 74% (95% CI 69.6% to 78.4%) of all participants were knowledgeable about food characteristics that increase the risk to overweight and obesity. Only 18.4% (95% CI 14.8% to 22.7%) were knowledgeable about practices and environments that promote overweight and obesity, and 19.7% (95% CI 16.0% to 24.1%) were knowledgeable of the social, economic and medical consequences of being overweight and obese. Marital status, knowledge of foods that promote overweight/obesity, and year of study were significantly associated with BMI categories.
Prevalence of overweight and obesity
The findings on overweight (24.7%) and obesity (6.8%) are higher than in a study carried out in 2010 among 683 young adults aged between 18 and 30 years in urban (Kampala city) and rural (Kamuli district) parts of Uganda which found that 17.4% and 2.9% of young adult females from Kampala were overweight and obese respectively.27 This difference could be attributed to the trends of increasing sedentary lifestyles especially in urban areas in addition to the increased availability of fast foods.28 In a cross-sectional study done in Kampala city, street food was found to contribute 49.1% of the daily fat intake of habitual urban consumers.29 In addition, findings are higher than the 9% overweight and 1% obesity found among female students in Bangladesh.30 However, the findings are in the range of findings from another study conducted in 2016 that reported results among adult females aged 20–49 years,31 and in a study in Ghana among university students.32 Further, the findings in our study are lower than in a study in western Balkans among medical students,33 and another study in Jordan among female adolescent students.34 The difference could be due to differences in study populations; our study included all students while the Balkans study included medical students who are likely to be more health conscious, hence less obese/overweight. However, a study in Saudi Arabia among health sciences students found prevalences of obesity and overweight of 23.7% and 11% respectively.35
The nutrition status of an individual is the result of the balance between calories consumed and calories expended. This balance is mainly affected by dietary and physical activity patterns. The observed dietary and physical activity patterns in any population/society and any changes therein are influenced by the environmental changes associated with development and policies that allow individuals access to a healthy lifestyle.1 This explains the differences in nutrition status between similar populations in different countries. Policies can be implemented that support individuals in active living and making healthier dietary choices through making these foods the most accessible, affordable and available.1 It is surprising that female university students are no different from females in the general population in having high BMI levels. This might call for interventions that go beyond the university possibly addressing cultural norms and perceptions.
Knowledge about obesity and overweight
The knowledge about food characteristics that increase the risk to overweight and obesity (74.2%) was higher than the 0.8%–38.8% knowledge levels found among adults in Tanzania,36 likely due to the differences in study populations (university students vs the general population). It is also higher than the knowledge of food risk factors for obesity and overweight (range 20% to 74%) found in rural Cameroon but within the range (66.9% to 91.8%) of that found in urban Cameroon.37
The proportion of participants that were knowledgeable about practices and environments that promote overweight and obesity was low (18.4%). This was in the range of 3.9% to 34.4% knowledge about practices and environments that promote overweight and obesity among secondary school students in Nigeria.38 However, our finding was lower than the finding in the cross-sectional study carried out in 2006 among 301 urban Senegalese women aged between 20 and 50 years.39 Herein, it was found out that those participants had good knowledge of practices and environments that promote obesity and overweight (58.8% to 81.1%). The explanation for this difference may be due to differences in the study population (urban women aged 20 to 50 years vs undergraduate university students).
The study also found out that only 19.7% of the participants were knowledgeable of the social, economic and medical consequences of being overweight and obese. This finding is lower than the 65.4% knowledge of health consequences among Senegalese women aged 20 to 50 years.39 However, it was higher than the 5% level of good knowledge (score of 8–10 of 10 questions) of health risks associated with obesity among female adolescents in Jordan.34 The cut-off for good knowledge used in the Jordan study (80%) was higher than the cut-off used in our study (70%) likely explaining the differences in proportions of level of good knowledge observed. The low knowledge of practices and environments that promote obesity and overweight, and the low knowledge of its consequences imply that the students cannot be intentional in avoiding these practices and environments. In addition, the motivation to avoid overweight and obesity may be minimal if the students have limited knowledge of the consequences. The knowledge findings, however, should be treated cautiously owing to the low performance of the knowledge assessment tool.
Factors associated with BMI categories
The study found that students who were married had odds of being in higher categories of BMI that were about 94% higher than that of students who were single. In addition, students who had adequate knowledge of the foods that promote overweight/obesity had 42% less odds of being in higher categories of BMI compared with those with inadequate knowledge. Students who were in the third year had 2.8 times the odds of being in higher categories of BMI compared with those in the first year, while fourth year students had almost three times the odds of being in higher BMI categories compared with first years. There was no difference between first and second years.
The findings on the association between marital status and BMI categories are in agreement with a systematic review that explored this association.40 A retrospective study conducted in China revealed that the odds of overweight and obesity are higher for married persons; OR: 2.18 (95% CI 1.90 to 2.51) and 1.95 (95% CI 1.57 to 2.43) respectively.41 Another study conducted in Greece also revealed a higher risk for obesity among the married women (OR: 2.31; 95% CI 1.73 to 3.10) than in the unmarried ones.42 Another study in Uganda in the general population among 20–49-year-olds reported higher odds of over-nourishment among married women compared with non-married.31 Various explanations have been forwarded for the observed association between marriage and BMI. The association has been attributed to changes in lifestyle, behaviours and social support when people get married. The protection hypothesis supposes that married adults have better social support and reduced engagement in risky behaviour.43 The marriage market hypothesis put forward that adults in the marriage market (unmarried individuals) tend to be more mindful of their BMI as a leaner person may have higher prospects of finding a partner.44 45 Maintaining a healthy BMI is, however, costly and people in a stable relationship may no longer find it imperative.43 The social obligation hypothesis supposes marriage obliges people to eat more regular meals and/or richer and denser foods.43
The observed association between BMI categories and knowledge of risk foods for obesity is similar to a study done in China among children and adolescents and their mothers.46 There is a likelihood that students who have adequate knowledge about obesogenic foods modify their diets to exclude them, hence minimising overweight/obesity. However, this was a cross-sectional study; hence, the causal relationship could not be ascertained. However, knowledge of practices and environments that promote obesity/overweight, and knowledge of its consequences were not associated with BMI categories in our study. In another study among sports science students, there was no association between knowledge of risk of obesity and obesity rate.47 The findings on associations between knowledge variables and BMI categories should be interpreted with caution because the tools used for knowledge assessment had low reliability with Cronbach’s alpha of 0.554.
Our findings of higher odds of being in higher BMI categories among higher years of study are similar to findings from a study in Ghana.32 A possible explanation for this finding is a longer period of exposure to foods that promote overweight and obesity with increased years of study because the students have spent longer at the university. Another possible explanation is the greater academic stress from financial hardships, academic overload and social expectations among first-year students that eases as they progress through the years due to adaptation.48 Further, the association between year of study and BMI categories might be a proxy of association with age since the students’ age is expected to increase in higher years of study. We did not collect data on students’ age. However, both studies had disproportionate samples of students in the different academic year groups (for the study in Ghana; first year n=122, second year n=65, third year n=58, fourth year n=55). A larger, proportionate sample size would be required to test the association between year of study and BMI. The findings on year of study call for university-level interventions that ensure that students do not develop health risks like obesity while at the university.
The lack of association between amount of money spent on food and BMI categories is supported by findings from a study among medical students from Jazan University where students’ monthly income was not associated with obesity.49 A possible explanation may be that some students were resident at the university and therefore took meals served at their halls of residence; hence, what they spent was not reflective of all the amount of food available to them. The other possible explanation could be that students may spend money on cheap fast foods or snacks that are commonly sold around urban areas of low socioeconomic status.
Study limitations
The convenience sampling strategy for the participants was likely to introduce selection bias. As earlier stated, students found in the lecture rooms before the start of the day’s classes were selected from each academic unit. These students available in the lecture rooms at that time are likely to be different from those who do not come early. The students who come early to class, in comparison to late comers or non-attenders, are likely to be highly engaged students who may be more knowledgeable or even more health conscious, hence affecting BMI. The sampling strategy was also likely to omit students who, by their academic programmes, did not have lectures in the morning. Of note, fifth-year medical students were not included because they were more likely to be in wards. In addition, 27 of the selected students did not complete study procedures, resulting in a response rate of 93.3%. However, the impact of non-response is likely to be minimal since it is quite low (6.7%). Further, the reliability of the knowledge assessment aspects of the questionnaire was quite low, with Cronbach’s alpha of 0.554, below the recommended minimum value of 0.7. This was likely to result in misclassification of the participants’ knowledge based on the tool. However, the Item Difficulty Index was 0.524 (0.775 for knowledge of foods that promote obesity or overweight, 0.484 for knowledge of practices and environments, and 0.463 for knowledge of consequences of obesity or overweight), which is within the recommended range of 0.2–0.8. In addition, the Delta Index showing the discrimination capacity overall was 0.945, which indicates good scale differentiation (0.845 for knowledge of foods, 0.935 for knowledge of practices and environments, and 0.926 for knowledge of consequences of obesity or overweight). We also did sensitivity analysis using the knowledge variables as continuous variables or categorical variables with 60% or 80% cut-offs for adequate knowledge. The conclusions did not differ from the 70% cut-off for adequate knowledge that is presented. Another limitation arose from doing the pretest of the study procedures among male undergraduate students, which could have led to missing some sex-specific differences in the comprehension of questions. However, we anticipate that these differences are likely to be minimal since these are all undergraduate university students. In addition, the measurement of height using a tape attached on the wall, that is, wall height, instead of a stadiometer which is the gold standard, could have led to measurement bias arising from observer positioning, wall irregularities or participant posture. This could have led to misclassifications in the BMI. However, wall height has been reported to be a fairly accurate estimate of stadiometer height.50 And as stated earlier, the measurements of weight and height were done by one person to minimise inter-observer variability. Our study did not include several important variables that could have been risk factors for obesity and overweight or potential confounders, particularly age, behavioural factors like physical activity, dietary practices and stress levels. The cross-sectional nature of the study also limits the establishment of causal relationships between independent variables and BMI categories. The data were also collected in 2019 which is pre-COVID-19 era. There could be post-COVID-19 shifts in student behaviours and in the food environments around the campus. The COVID-19 period was characterised by changes in behaviours where people adopted healthy living behaviours like increased consumption of vegetables, but there were also changes in financial circumstances that affected dietary practices.51 These could have lingered on in this post-COVID-19 era. The findings of this study should be considered in light of the above limitations.
Conclusion
The prevalence of overweight and obesity among female undergraduate students at Makerere University was high, with about one in every four being overweight and almost 7 in every 100 being obese. Although the knowledge of foods that promote overweight/obesity was high at 74%, the knowledge of practices and environments that promote overweight and obesity was low, with only about 18 in 100 patients having adequate knowledge. In addition, about one in every five students had adequate knowledge about the consequences of overweight and obesity. Married students, students in higher years of study, and those who had less knowledge about foods that promote overweight and obesity were more likely to be in higher categories of BMI. The findings suggest a need to strengthen awareness of overweight/obesity among university students, especially the predisposing practices. University systems like the Office of Dean of Students or the University Academic Registrar, together with the University Hospital, could be championing the awareness programmes, along with the student leadership. Programmes to address issues on overweight or obesity could be incorporated in health programmes conducted during university orientation weeks and at other times of general student sensitisation. In addition, students who are at high risk of overweight and obesity should be targeted with interventions for addressing the problem. Further research could identify preferred methods for addressing the problem of overweight and obesity among students. Additionally, other potentially important variables reported in the literature, like physical activity, dietary practices and body perceptions, could be assessed for association with BMI categories.
Supplementary material
Acknowledgements
We thank all study participants for taking time to participate in the study. We also thank members of the Department of Pharmacy of Makerere University for their invaluable inputs.
Footnotes
Funding: The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Provenance and peer review: Not commissioned; externally peer reviewed.
Patient consent for publication: Not applicable.
Ethics approval: The research was approved by the School of Health Sciences Research and Ethics Committee (SHSREC REF: 2019-004). Permission was sought from registrars of different academic units to allow us to carry out research on undergraduate female students therein. Written informed consent was sought from every study participant before including them in the study. The student’s decision to participate in the study did not have any impact on their grades or anything else related to their studies. Access to the data was restricted to the investigators to ensure confidentiality and identification numbers were used instead of participants’ names. All research procedures complied with Uganda National Council of Science and Technology guidelines. Participants gave informed consent to participate in the study before taking part.
Patient and public involvement: Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.
Data availability statement
Data are available upon reasonable request.
References
- 1.World Health Organization Obesity and overweight. 2024. [03-Oct-2024]. https://www.who.int/news-room/fact-sheets/detail/obesity-and-overweight Available. Accessed.
- 2.Africa Health Organization Obesity and overweight fact sheet. [03-Oct-2024]. https://aho.org/fact-sheets/obesity-and-overweight-fact-sheet/ Available. Accessed.
- 3.World Obesity Federation World obesity atlas 2025. 2025:274.
- 4.Ford ND, Patel SA, Narayan KMV. Obesity in Low- and Middle-Income Countries: Burden, Drivers, and Emerging Challenges. Annu Rev Public Health. 2017;38:145–64. doi: 10.1146/annurev-publhealth-031816-044604. [DOI] [PubMed] [Google Scholar]
- 5.World Health Organization Obesity and overweight. Fact sheet no.311. 2012
- 6.Xu S, Xue Y. Pediatric obesity: Causes, symptoms, prevention and treatment. Exp Ther Med. 2016;11:15–20. doi: 10.3892/etm.2015.2853. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Okunogbe A, Nugent R, Spencer G, et al. Economic impacts of overweight and obesity: current and future estimates for eight countries. BMJ Glob Health. 2021;6:e006351. doi: 10.1136/bmjgh-2021-006351. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Tremmel M, Gerdtham U-G, Nilsson PM, et al. Economic Burden of Obesity: A Systematic Literature Review. Int J Environ Res Public Health. 2017;14:435. doi: 10.3390/ijerph14040435. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.WHO Fact Sheet Obesity and overweight. 2015. https://www.who.int/news-room/fact-sheets/detail/obesity-and-overweight Available.
- 10.Mohamed Ali M, Elhamed Zaki NA, Departments of Community Medicine and Internal Medicine, Sohag Faculty of Medicine, Sohag University, Sohag, Egypt. et al. Assessment of prevalence, knowledge, attitude and practices of obesity among college students in Sohag Governorate. Int J Curr Res Med Sci. 2017;3:87–96. doi: 10.22192/ijcrms.2017.03.04.014. [DOI] [Google Scholar]
- 11.Almoraie NM, Alothmani NM, Alomari WD, et al. Addressing nutritional issues and eating behaviours among university students: a narrative review. Nutr Res Rev. 2025;38:53–68. doi: 10.1017/S0954422424000088. [DOI] [PubMed] [Google Scholar]
- 12.Lowry R, Wechsler H, Galuska DA, et al. Television viewing and its associations with overweight, sedentary lifestyle, and insufficient consumption of fruits and vegetables among US high school students: differences by race, ethnicity, and gender. J Sch Health. 2002;72:413–21. doi: 10.1111/j.1746-1561.2002.tb03551.x. [DOI] [PubMed] [Google Scholar]
- 13.Kanter R, Caballero B. Global Gender Disparities in Obesity: A Review. Adv Nutr. 2012;3:491–8. doi: 10.3945/an.112.002063. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Bray GA. Medical Consequences of Obesity. J Clin Endocrinol Metab. 2004;89:2583–9. doi: 10.1210/jc.2004-0535. [DOI] [PubMed] [Google Scholar]
- 15.Hu FB. Overweight and obesity in women: health risks and consequences. J Womens Health (Larchmt) 2003;12:163–72. doi: 10.1089/154099903321576565. [DOI] [PubMed] [Google Scholar]
- 16.Puhl RM, Heuer CA. The stigma of obesity: a review and update. Obesity (Silver Spring) 2009;17:941–64. doi: 10.1038/oby.2008.636. [DOI] [PubMed] [Google Scholar]
- 17.Goodman R, Kish L. Controlled Selection--A Technique in Probability Sampling. J Am Stat Assoc. 1950;45:350. doi: 10.2307/2280293. [DOI] [Google Scholar]
- 18.Egbuchulem KI. THE BASICS OF SAMPLE SIZE ESTIMATION: AN EDITOR’S VIEW. Ann Ib Postgrad Med. 2023;21:5–10. [PMC free article] [PubMed] [Google Scholar]
- 19.Ge H, Hong K, Fan C, et al. Knowledge, attitude, and practice of healthcare providers on chronic refractory cough: A cross-sectional study. Heliyon. 2024;10:e27564. doi: 10.1016/j.heliyon.2024.e27564. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.World Health Organization Malnutrition in women. [20-Sep-2024]. https://www.who.int/data/nutrition/nlis/info/malnutrition-in-women Available. Accessed.
- 21.Swift JA, Glazebrook C, Macdonald I. Validation of a brief, reliable scale to measure knowledge about the health risks associated with obesity. Int J Obes. 2006;30:661–8. doi: 10.1038/sj.ijo.0803165. [DOI] [PubMed] [Google Scholar]
- 22.Styk W, Wojtowicz E, Zmorzynski S. Reliable Knowledge about Obesity Risk, Rather Than Personality, Is Associated with Positive Beliefs towards Obese People: Investigating Attitudes and Beliefs about Obesity, and Validating the Polish Versions of ATOP, BAOP and ORK-10 Scales. Int J Environ Res Public Health. 2022;19:14977. doi: 10.3390/ijerph192214977. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Ayala GX, Rogers M, Arredondo EM, et al. Away-from-home food intake and risk for obesity: examining the influence of context. Obesity (Silver Spring) 2008;16:1002–8. doi: 10.1038/oby.2008.34. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Sonntag D, Schneider S, Mdege N, et al. Beyond Food Promotion: A Systematic Review on the Influence of the Food Industry on Obesity-Related Dietary Behaviour among Children. Nutrients. 2015;7:8565–76. doi: 10.3390/nu7105414. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Vilchis-Gil J, Galván-Portillo M, Klünder-Klünder M, et al. Food habits, physical activities and sedentary lifestyles of eutrophic and obese school children: a case-control study. BMC Public Health. 2015;15:124. doi: 10.1186/s12889-015-1491-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Dhana K, Haines J, Liu G, et al. Association between maternal adherence to healthy lifestyle practices and risk of obesity in offspring: results from two prospective cohort studies of mother-child pairs in the United States. BMJ. 2018;362:k2486. doi: 10.1136/bmj.k2486. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Baalwa J, Byarugaba BB, Kabagambe EK, et al. Prevalence of overweight and obesity in young adults in Uganda. Afr Health Sci. 2010;10:367–73. [PMC free article] [PubMed] [Google Scholar]
- 28.Uganda AS, Bonabana-Wabbi J, Sserunkuuma D, et al. Determinants of fast food consumption in Kampala, Uganda. African Journal of Food, Agriculture, Nutrition and Development. 2012 [Google Scholar]
- 29.Sseguya W, Matovu N, Swann J, et al. Contribution of street food to dietary intake of habitual urban consumers: A cross-sectional study in Kampala city, Uganda. Nutr Health. 2020;26:187–95. doi: 10.1177/0260106020919629. [DOI] [PubMed] [Google Scholar]
- 30.P K, N J. Food Habits, Obesity and Nutritional Knowledge among the University Students in Noakhali Region of Bangladesh: A Cross Sectional Study. J Food Nutr Disor . 2016;5 doi: 10.4172/2324-9323.1000201. [DOI] [Google Scholar]
- 31.Sserwanja Q, Mukunya D, Kawuki J, et al. Over-nutrition and associated factors among 20 to 49-year-old women in Uganda: evidence from the 2016 Uganda demographic health survey. Pan Afr Med J. 2021;39:261. doi: 10.11604/pamj.2021.39.261.26730. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Obirikorang C, Adu EA, Anto EO, et al. Prevalence and risk factors of obesity among undergraduate student population in Ghana: an evaluation study of body composition indices. BMC Public Health. 2024;24:877. doi: 10.1186/s12889-023-17175-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Ilić M, Pang H, Vlaški T, et al. Prevalence and associated factors of overweight and obesity among medical students from the Western Balkans (South-East Europe Region) BMC Public Health. 2024;24:29. doi: 10.1186/s12889-023-17389-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Thaher LM, Alasad J, Maharmeh M, et al. Prevalence of Obesity and Knowledge of Health Risk Associated with Obesity among Female Adolescents in Jordan. OJN. 2018;08:60–8. doi: 10.4236/ojn.2018.81005. [DOI] [Google Scholar]
- 35.Makkawy E, Alrakha AM, Al-Mubarak AF, et al. Prevalence of overweight and obesity and their associated factors among health sciences college students, Saudi Arabia. J Family Med Prim Care. 2021;10:961–7. doi: 10.4103/jfmpc.jfmpc_1749_20. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Msollo SS, Shausi GL, Mwanri AW. Prevalence, knowledge and practices on prevention and management of overweight and obesity among adults in Dodoma City, Tanzania. PLoS ONE. 2024;19:e0297665. doi: 10.1371/journal.pone.0297665. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Dongmo FFD, Asongni WD, Mba ARF, et al. Knowledge, Attitude, and Practices regarding Obesity among Population of Urban (Douala) and Rural (Manjo) Areas in Cameroon. Int J Chronic Dis. 2023;2023:5616856. doi: 10.1155/2023/5616856. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Oyewande AA, Ademola A, Okuneye TA, et al. Knowledge, attitude and Perception Regarding Risk Factors of Overweight and Obesity Among Secondary School Students in Ikeja Local Government Area, Nigeria. J Family Med Prim Care. 2019;8:1391–5. doi: 10.4103/jfmpc.jfmpc_160_19. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Holdsworth M, Delpeuch F, Landais E, et al. Knowledge of dietary and behaviour-related determinants of non-communicable disease in urban Senegalese women. Public Health Nutr. 2006;9:975–81. doi: 10.1017/s1368980006009797. [DOI] [PubMed] [Google Scholar]
- 40.Nikolic Turnic T, Jakovljevic V, Strizhkova Z, et al. The Association between Marital Status and Obesity: A Systematic Review and Meta-Analysis. Diseases. 2024;12:146. doi: 10.3390/diseases12070146. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Liu J, Garstka MA, Chai Z, et al. Marriage contributes to higher obesity risk in China: findings from the China Health and Nutrition Survey. Ann Transl Med. 2021;9:564. doi: 10.21037/atm-20-4550. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Tzotzas T, Vlahavas G, Papadopoulou SK, et al. Marital status and educational level associated to obesity in Greek adults: data from the National Epidemiological Survey. BMC Public Health. 2010;10:732. doi: 10.1186/1471-2458-10-732. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Averett SL, Sikora A, Argys LM. For better or worse: relationship status and body mass index. Econ Hum Biol. 2008;6:330–49. doi: 10.1016/j.ehb.2008.07.003. [DOI] [PubMed] [Google Scholar]
- 44.van Woerden I, Brewis A, Hruschka D, et al. Young adults’ BMI and changes in romantic relationship status during the first semester of college. PLoS One. 2020;15:e0230806. doi: 10.1371/journal.pone.0230806. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Wilson SE. Marriage, gender and obesity in later life. Econ Hum Biol. 2012;10:431–53. doi: 10.1016/j.ehb.2012.04.012. [DOI] [PubMed] [Google Scholar]
- 46.Xu Z, Zhao Y, Sun J, et al. Association between dietary knowledge and overweight and obesity in Chinese children and adolescents: Evidence from the China Health and Nutrition Survey in 2004-2015. PLoS One. 2022;17:e0278945. doi: 10.1371/journal.pone.0278945. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Hassan MHA, Che Muhamed AM, Safii NS, et al. Association between obesity risk knowledge and body mass index among sport students. Proceedings of the 8th International Conference on Movement, Health and Exercise; Springer Nature Singapore. 2023. [Google Scholar]
- 48.Nakalema G, Ssenyonga J. Academic stress: Its causes and results at a Ugandan university. AJOTE. 2014;3 doi: 10.21083/ajote.v3i3.2762. [DOI] [Google Scholar]
- 49.Alqassimi S, Elmakki E, Areeshi AS, et al. Overweight, Obesity, and Associated Risk Factors among Students at the Faculty of Medicine, Jazan University. Medicina (Kaunas) 2024;60:940. doi: 10.3390/medicina60060940. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 50.Gordon SA, Fredman L, Orwig DL, et al. Comparison of methods to measure height in older adults. J Am Geriatr Soc. 2013;61:2244–6. doi: 10.1111/jgs.12572. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 51.United Nations Children’s Fund . Nairobi Kenya: UNICEF; 2022. Impact of the COVID-19 pandemic on diets, nutrition practices and nutrition services in Uganda; p. 10. [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Data are available upon reasonable request.