Research | Open Access | Volume 9 (1): Article 45 | Published: 12 March 2026
Menu, Tables and Figures
Boyo Constant Paré1,&.*, Désiré Lucien Dahourou2,*, Solo Traoré3,4,*, Ad Bafa Ibrahima Ouattara3,5, Ter Tiero Elias Dah3,6, Oumar Guira1,7
1Training and Research Unit in Health Sciences, Department of Public Health, Joseph Ki-Zerbo University, Ouagadougou, Burkina Faso, 2Department of Biomedical and Public Health, Research Institute of Health Sciences, Ouagadougou, Burkina Faso, 3Training and Research Unit in Health Sciences (UFR/SS), Lédéa Bernard OUÉDRAOGO University, Ouahigouya, Burkina Faso, 4Department of Medicine and Medical Specialties, Ouahigouya Regional Teaching Hospital, Ouahigouya, Burkina Faso, 5Department of Paediatrics, Ouahigouya Regional Teaching Hospital, Ouahigouya, Burkina Faso, 6Department of Public Health, Ouahigouya Regional Teaching Hospital, Ouahigouya, Burkina Faso, 7Department of Internal Medicine, Yalgado Ouédraogo Teaching Hospital, Ouagadougou, Burkina Faso, *These authors contributed equally to this work and are joint first authors
&Corresponding author: Boyo Constant Paré, Training and Research Unit in Health Sciences, Department of Public Health, Joseph Ki-Zerbo University, Ouagadougou, Burkina Faso. Email: boyoconstantp@gmail.com ORCID: https://orcid.org/0000-0002-6842-127X
Received: 27 May 2025, Accepted: 11 Mar 2026, Published: 12 Mar 2026
Domain: Non-communicable Disease Epidemiology
Keywords: Impaired fasting glucose, women, prevalence, associated factors, urban environment
©Boyo Constant Paré et al. Journal of Interventional Epidemiology and Public Health (ISSN: 2664-2824). This is an Open Access article distributed under the terms of the Creative Commons Attribution International 4.0 License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Cite this article: Boyo Constant Paré et al., Prevalence and factors associated with impaired fasting glucose among urban women in Burkina Faso. Journal of Interventional Epidemiology and Public Health. 2026; 9(1):45. https://doi.org/10.37432/jieph-d-25-00130
Introduction: Diabetes mellitus increases the risk of heart disease by fourfold in women and twofold in men. It is a growing problem in developing countries, including Burkina Faso. Impaired Fasting Glucose (IFG) can predict increased risk for developing diabetes. We determined the proportion of IFG among adult women and identified their risk factors.
Method: This was a cross-sectional study of female volunteers over the age of 18 years, recruited from four screening sites that targeted occupational profiles whose lifestyle would be factors influencing IFG, which can predict an increased risk for developing diabetes: university, market, police school, and nursing school. Data were collected in November 2020, in the urban setting of Ouagadougou, Burkina Faso. We performed a logistic regression to identify risk factors associated with IFG (blood glucose ≥6.1 mmol/L).
Results: 356 women participated in the study at four collection sites. Of these women, 10.1% (36/356; 95% CI: 7.3-13.7) were classified as IFG. The mean age was 28 ±9.3 years, with extremes ranging from 18 to 71 years. IFG represented 46% among women aged 45 to 55 years, 16% among women with abdominal obesity, 19% among women with obesity, 28% among freelance women. In the multivariate analysis, abdominal obesity (aOR=2.6; 95%CI: 1.1-6.2), recruitment at the market site (aOR=10.8; 95%CI: 1.3-86.9) were significantly associated with a higher risk of IFG in women.
Conclusion: We observed a higher prevalence and risk of IFG among women recruited from the market setting, where women predominantly engage in activities related to trade or entrepreneurship. Further studies among women working in such settings with this type of socio-professional lifestyle could help to improve the performance of diabetes prevention and control programs in Burkina Faso.
Type 2 Diabetes Mellitus (T2DM) seriously affects women’s lives. It quadruples, for instance, the risk of heart disease in women and doubles that in men [1]. Worldwide, the prevalence of diabetes among women aged 20–79 years was estimated at 10.2% in 2021 [2]. In Africa, 19 million cases were reported in 2019 [3]. A proportion of 60% of the diabetes cases were estimated to not be diagnosed in this region [3]. According to the World Health Organization (WHO), the largest rise in prevalence of chronic diseases like diabetes is expected in the developing world by 2030 [4]. Diabetes is becoming a major problem in developing countries, including Burkina Faso [5]. In Burkina Faso (BF), the type 2 diabetes mellitus (T2DM) prevalence was slightly increased between 2013 and 2021 among women, with prevalence of 4.7% and 6.8% [6,7].
The risk factors for diabetes comprise overweight or obesity, being aged above 45 years, having a parent or brother/sister with type 2 diabetes mellitus (T2DM), physically activity less than 3 times a week, presence of non-alcoholic fatty liver disease, having ever experienced gestational diabetes, or having given birth to a baby who weighed above 4082 grams [8,9]. Some of these factors are modifiable while others are not. Job-related characteristics may also influence the risk factors for T2DM [10–12]. Work-related stress, shiftwork, long working hours, and sedentary work conditions might increase the risk of T2DM and generalised obesity [10–12]. Studies conducted in the United States revealed that the prevalence of diabetes varied according to occupation [10,12,13]. Nonetheless, occupation has rarely been examined as an important factor in diabetes-related studies in Low and Middle Income Countries (LMICS) [14]. Moreover, some studies stress a difference in cardio-metabolic risk factors distribution between women and men [15,16]. There are differences between men and women regarding type 2 diabetes and other cardiovascular risk factors with respect to comorbidities, the manifestation of complications, and the initiation of and adherence to therapy [17,18]. The majority of diabetics discover their disease through the cardinal signs of diabetes mellitus (27%) or the onset of complications (61.53%) [19,20]. In Burkina Faso, 89% of participants in the 2021 STEPS survey in urban areas had never measured their blood glucose levels [7]. Diabetes mellitus testing should serve as a gateway to the prevention and control of Non-Communicable Diseases (NCDs) [21]. Impaired Fasting Glucose (IFG) can predict increased risk for developing diabetes. Diabetes prevention or delay is effective when targeting patients with IFG [22]. Greater knowledge of the groups that are less likely to be aware of their IFG could result in a more targeted approach to screening, increased potential for detection, and therefore more effective management of diabetes. We aimed to estimate the prevalence of IFG among adult women recruited from diverse socio-professional settings and investigate how individual lifestyle and physiological factors contribute to IFG risk within these specific urban contexts
Study setting
The data were collected in November 2020, in the city of Ouagadougou, a large urban area of Burkina Faso. Ouagadougou is the capital city of the country and the largest one [23]. The urban commune of Ouagadougou is located in the province of Kadiogo, in the Central region. Ouagadougou has 12 arrondissements [23–25]. It has 2,415,266 people and 502,938 households [23–25]. Half (50%) of the workforce were government workers, while the rest were in the informal sector (45%) and the unemployed (5%) [23–25]. In addition, with a young population (24 years on average), Ouagadougou is the commune with the country’s highest population density, with over 4300 inhabitants per km². More than 61% of the city’s residents above 15 years are literate.
Type and study population
This study is a cross-sectional study among the adult population in Ouagadougou [25]. We extracted data from adult female participants (≥18 years) for a secondary analysis.
Data collection
The survey was conducted during a mass campaign of diabetes diagnosis for the 2020 international diabetes day. We intentionally selected four different screening sites—a university, a market, a police academy, and a nursing school—to gather data from diverse socio-professional groups. The assumption was that the lifestyles associated with these environments might influence impaired fasting glucose. A quick survey has been conducted during this mass campaign [25]. Our focus was on women, so we extracted all the women’s data from the initial database. We considered all the women who were at least 18 years of age in the database. We exclude pregnant women and those who reported that they were taking medication for diabetes [25].
An anonymous questionnaire was administered to collect data, and capillary blood glucose was measured using the SD CodeFreeTM glucose analyzer. A capillary whole blood was taken by finger prick and immediately analyzed. A control solution test was performed each time a new bottle of strips was opened. All other procedures described by the manufacturer were followed [25].
Weight was taken with the subjects wearing light clothing and no shoes, using a digital floor scale to the closest 0.1 kg. The BMI was calculated as the weight (kg)/ height (m2). Waist circumference (marked using a tape) was measured while standing, at the halfway level between the lowest rib and iliac crest. The value was recorded to the nearest centimeter (cm) at the navel level at the end of regular expiration [25].
Study variables
Outcome variable: The outcome variable in our study was impaired fasting glucose (IFG). IFG was defined as fasting (after at least 8 h of fasting) capillary glucose ≥ 6.1 mmol/L and < 7.0 mmol/L [7,22].
Independent variables: independent variables was age, family history of diabetes mellitus (classified into no family history, first-degree family history and second-degree family history); body mass index (in kg/m2 ); waist circumference; physical activity (categorized as ≥ 30 minutes of daily physical activity or not); daily fruit and vegetable intake, history of hyperglycemia; the use of anti-hypertensive treatment [25].
Statistical analysis
The study sample was characterized using frequency tables with percentages for qualitative variables. Comparisons between groups were done by the Chi square test or Fisher test. The quantitative variables were presented through mean and standard deviation. We estimated the prevalence of women with IFG with 95% confidence interval. We used logistic regression to identify the factors associated with IFG. We included in a multivariable model non-colinear independent variables associated in univariable analysis at the 20% threshold. We used a stepwise manual step-down procedure to get the final model with STATA 15 software.
Ethical considerations
This study was approved by the Health Science Ethics Committee (No. 2020-8-146) [25]. We obtained the health authorities’ authorization to conduct the study. This study was conducted during a mass campaign. This was the minimal-risk study involving volunteer participants in the context where literacy is very low. In addition, the participant’s decision to participate in this study was based on a deliberate and affirmative action. Before they gave oral consent to participate, we informed participants that participation in this survey was free and voluntary [25]. No data that could identify participants was collected. Participants with capillary glucose levels > 13.75 mmol/L (indicative of severe hyperglycemia) were systematically referred for urgent evaluation at the Internal Medicine Department of the Yalgado Ouédraogo University Hospital (CHU-YO). Participants with results below this threshold but within the IFG or diabetic range were provided with a formal referral for an outpatient consultation at the same specialized center for diagnostic confirmation and management.
Overall, 377 women were screened. Among these women, 356 women (94%) were included in this analysis. The mean age was 28 (standard deviation: 9.4 years), with extremes ranging from 18 to 71 years. Most of the participants were under 35 years (81.4%). The prevalence of IFG was 10.1% (36/356; 95% CI: 7.3-13.7). The prevalence of IFG was 46% among women aged 45 to 55 years, 16% among women with abdominal obesity, 19% among women with generalised obesity, 28% among freelance women. Generalised obesity was identified among 17.6% of participants, and 84.4% of women declared that they eat at work, not at home. Abdominal obesity was identified in 49.7% of women (Table 1).
In the univariate analysis the following variables were associated with IFG among women: being screened in the market i.e. freelance women (OR=15.81; 95% CI: 2.01-124.35), age group of [35-45[ (OR=5.69; 95% CI: 2.44-13.25) & [45-55[ (OR=13.76; 95% CI: 4.16-45.49), presence of abdominal obesity (OR=4.01; 95% CI: 1.77-9.08), being married (OR=2.25; 95% CI: 1.10–4.60, being obese (OR=3.33; 95% CI: 1.43-7.74). In the multivariate analysis, abdominal obesity (aOR=2.59; 95% CI:1.08-6.20), being screened in the market, i.e. freelance women (trade, business, entrepreneurship) (aOR=10.77; 95% CI:1.33-86.92 significantly increased the odds of IFG in women (Table 2).
We observed a high proportion of women with IFG, exceeding the national prevalence reported in the 2021 STEPS survey. This suggests that urban-specific metabolic risks may be intensifying faster than national figures reflect. Our findings align with similar studies in Ouagadougou, which reported among men and women [26] and a high proportion of diabetes occurrence in women compared to men [5]. Significant disparities were observed in IFG proportion across data collection sites, reflecting the diverse socio-demographic profiles within the city. The lower rates at the Police academy , the private university and the nursing school likely represent baseline risks among the younger, more physically active and health literate groups. In contrast, the proportion of IFG at the central market is likely driven by the fact that it was the site with the older participants and the higher rates of obesity. These results suggest that current national strategies, which largely rely on passive hospital-based detection, are suboptimal for reaching high-risk urban sub-populations.
A high prevalence of IFG was observed among women aged 45 to 55 years, compared to those aged less than 35 years. Although getting old increases the risk of IFG [27], and it was likely participants who felt a certain risk of an unhealthy state who came for the screening, these figures are worrying. Indeed, in Burkina Faso, the latest 2021 STEPS survey showed 89.0% of participants had never had their blood glucose level measured. Community screening can therefore be used in such settings to identify people at risk and people living with impaired fasting glucose who are unaware of their condition.
We acknowledge that using the data collection site as a proxy for socio-professional risk profiles is a limitation of this study. This approach was chosen due to significant missing data in the self-reported occupation variables. While individuals tested at the market may include transient visitors from other professional backgrounds, the site-specific prevalence remains a robust indicator of the ecological risk environment. The IFG proportion disparity, ranging from 2.4% at the police academy to 28.3% at the central market—highlights that certain urban hubs in Ouagadougou act as ‘hotspots’ for impaired fasting glucose. These findings suggest that screening strategies should be tailored to specific urban settings where high-risk populations congregate, even if individual occupational status remains heterogeneous. Typical lifestyle due to socio-professional environment might be a factor influencing glycemia [13,28].
Indeed, several studies have shown that health outcomes may be impacted by types of food and restaurants available in our environment and the food choices we make due to our daily activities [29,30]. In addition, socio-economic inequalities expose women to the main risk factors for diabetes, such as poor diet, physical inactivity, smoking, and harmful alcohol consumption. This poor lifestyle may be responsible for the high prevalence of obesity among women, one of the main risk factors for diabetes [17,18,31].
Participants from the university, mostly students, have a youthful lifestyle, eat fast food and occasionally do sports. Thus, IFG represented 3.9% of this group. It represented 19% among women with generalised obesity, 28% among women from the market’s site. Participants from the market’s site were mainly self-employed women working at the market, making sales for a living. This market is located in the heart of downtown and is the biggest. Most of them go there in the morning and return home only in the evening, due to the distance. As a result, their breakfasts and lunches are bought at the market (restaurants or food vendors nearby or have them delivered by restaurants around). Indeed, 298 (84%) declared to eat outside. Market participants were very much inclined to eat out (88%), which means no real control over their carbohydrate consumption. Due to their professional activities of selling in small shops, they likely do less sport. In contrast, participants from the police academy were more likely than the other groups to take part in sports on a regular basis, and they had the lowest proportion of IFG in their group.
The proportion of participants from nursing school who presented IFG was moderate. This contrasts with the fact that nurses and healthcare professionals, because of their medical knowledge and educational training, would be less likely to be at high risk of diabetes [32]. However, a study among health professionals in Ouagadougou showed a paradox with the low consumption level of fruits (12% daily) and vegetables (22% daily) among health professionals, coupled with a low level of knowledge of the functions of fruits and vegetables [33]. Moreover, this category of students is mainly selected by an examination from the civil service office, and they are students who are beginning to have a more stable monthly income. In addition, the design with data collection through a mass campaign may explain some of these results.
The risk for developing diabetes is often associated with overweight, generalised obesity, and abdominal obesity [34]. In this survey, IFG represented 16% among women with abdominal obesity and 12% and 19%, respectively, among overweight and obese women. Abdominal obesity significantly increased the risk of diabetes among women. It underscores that central adiposity might be a primary driver of metabolic dysregulation among women in Ouagadougou. Waist circumference is a factor frequently associated with diabetes in many settings [26,35–38]. The national STEPS data (2013 – 2021) indicate a broad increase in abdominal obesity prevalence in women; our findings pinpoint specific urban vulnerabilities[6,7].
The increase between 2013 and 2021 is probably due to the habits of communities that may have changed with urbanisation and modernisation. The high prevalence of abdominal obesity compared to generalized obesity suggests a thin-fat phenotype often observed in populations undergoing rapid nutrition transitions . This is likely exacerbated by the fact that 84.4% of our participants consume meals at their place of employment, where access to nutrient-dense food may often be limited in favour of high glycemic, energy-dense street foods. This suggests that health education models should focus on workplace-based environmental dietary interventions. It should also be noted that in our context, the high proportions of overweight and obese women could be explained by socio-cultural motivations. In many sub-Saharan countries, being overweight is considered a sign of beauty and well-being for women [39]. The socio-professional environment could therefore contribute to a progressive leverage effect on diabetes risk factors. Being able to tackle habits among women, taking into account their socio-professional environment, may contribute to controlling the diabetes epidemic at the population level [40].
Limitations of the study
The findings of this study should be interpreted in light of several limitations. First, it was a cross-sectional design, which precludes any causal inference between the socio-professional environment and the onset of the IFG. Second, the fact that data were collected at fixed sites during a mass screening campaign introduces a potential for self-selection bias; individuals who perceive themselves to be at higher risk may have been more likely to volunteer for screening than those who perceived their health as normal. Consequently, the observed IFG prevalence of 10.1% may be an overestimation of the true prevalence among the general female population in Ouagadougou. Furthermore, blood glucose was measured once and could not be repeated, as required for the diagnosis of diabetes mellitus in the standards of excellence [41,42], and the capillary blood glucose was measured instead of plasma glucose. The measurement of most exposures was based on participant declarations. We also acknowledge that some well-known metabolic risk factors for diabetes mellitus were not included in the study, such as lipids, history of fetal macrosomia, smoking, and alcohol intake. This could lead to information bias, even if these risk factors are rare among women in our setting. Despite these constraints, the study provides critical insights into the metabolic health of women within specific urban professional clusters that are often overlooked in national surveys.
A market setting where women mostly have activities related to trade business or entrepreneurship is associated with a higher prevalence of impaired fasting glucose in women. These findings highlight the importance of targeting such urban hubs for screening and prevention programs. Further studies among women working in such settings with this type of socio-professional lifestyle could help to improve the performance of IFG prevention and control programs.
What is already known about the topic
What this study adds
B.C.P., D.L.D. and S.T., contributed equally to this work and share the first authorship. B.C.P, D.L.D. and S.T., designed the study and analyzed the data. B.C.P., D.L.D., S.T., A.B.I.O., T.T.E.D. and O.G. interpreted the data. B.C.P., S.T. and D.L.D. wrote the first draft. B.C.P., D.L.D., S.T., A.B.I.O., T.T.E.D. and O.G. critically reviewed the manuscript. All the authors have read and approved the final version of the article.
| Characteristics | Non IFG | IFG | P value | Total | ||
|---|---|---|---|---|---|---|
| N=320 | % | N=36 | % | |||
| Data collection sites | <0.001 | |||||
| Police academy | 40 | 97.6 | 1 | 2.4 | 41 | |
| Nursing school | 188 | 92.2 | 16 | 7.8 | 204 | |
| Biggest Market of the city | 43 | 71.7 | 17 | 28.3 | 60 | |
| Private university | 49 | 96.1 | 2 | 3.9 | 51 | |
| Age (years) | <0.001 | |||||
| <35 | 273 | 94.1 | 17 | 5.9 | 290 | |
| 35-45 | 31 | 73.8 | 11 | 26.2 | 42 | |
| 45-55 | 7 | 53.8 | 6 | 46.2 | 13 | |
| >55 | 9 | 81.8 | 2 | 18.2 | 11 | |
| Family history of diabetesa | 0.28 | |||||
| No previous history | 226 | 91.5 | 21 | 8.5 | 247 | |
| 1st degree family historyb | 45 | 84.9 | 8 | 15.1 | 53 | |
| 2nd degree family historyc | 49 | 87.5 | 7 | 12.5 | 56 | |
| 30 minutes of daily physical activity | 0.65 | |||||
| Yes | 192 | 90.8 | 13 | 9.2 | 205 | |
| No | 128 | 89.3 | 23 | 10.7 | 151 | |
| Daily consumption of fruit and vegetables | 0.20 | |||||
| Not every day | 267 | 90.8 | 27 | 9.2 | 294 | |
| Every day | 53 | 85.5 | 9 | 14.5 | 62 | |
| Prescription of anti-hypertensive | ||||||
| Yes | 10 | 76.9 | 3 | 23.1 | 0.11 | 13 |
| No | 310 | 90.4 | 33 | 9.6 | 343 | |
| History of hyperglycemia | 0.22 | |||||
| Yes | 7 | 77.8 | 2 | 22.2 | 9 | |
| No | 313 | 90.2 | 34 | 9.8 | 347 | |
| Waist circumference | <0.001 | |||||
| No abdominal obesityd | 171 | 95.5 | 8 | 4.5 | 179 | |
| Abdominal obesitye | 149 | 84.2 | 28 | 15.8 | 177 | |
| BMI in kg/m2 | 0.01 | |||||
| Normalf | 184 | 93.4 | 13 | 6.6 | 197 | |
| Overweightg | 85 | 88.5 | 11 | 11.5 | 96 | |
| Generalized Obesityh | 51 | 80.9 | 12 | 19.1 | 63 | |
| Marital Status | 0.04 | |||||
| Married | 125 | 85.6 | 21 | 14.4 | 146 | |
| Single | 188 | 93.1 | 14 | 6.9 | 202 | |
| Widowed | 3 | 75.0 | 1 | 25.0 | 4 | |
| Divorced | 0 | 0.0 | 0 | 0.0 | 0 | |
| Means of transportation | 0.89* | |||||
| Motorcycle | 257 | 89.9 | 29 | 10.1 | 286 | |
| Bicycle | 19 | 90.5 | 2 | 9.5 | 21 | |
| Pedestrian | 26 | 86.7 | 4 | 13.3 | 30 | |
| Car | 15 | 93.7 | 1 | 6.3 | 16 | |
| Type of physical activity | 0.50 | |||||
| Walk | 119 | 88.8 | 15 | 11.2 | 134 | |
| Work | 131 | 89.7 | 15 | 10.3 | 146 | |
| Leisure | 32 | 97.0 | 1 | 3.0 | 33 | |
| None | 33 | 86.8 | 5 | 13.2 | 38 | |
| Lunch location | 0.24 | |||||
| At home | 47 | 85.5 | 8 | 14.5 | 55 | |
| Out of the house | 270 | 90.6 | 28 | 9.4 | 298 | |
| Contraceptive | 0.90 | |||||
| Yes | 90 | 90.0 | 10 | 10.0 | 100 | |
| No | 215 | 89.6 | 25 | 10.4 | 240 | |
| Univariate | Multivariate | |||||
|---|---|---|---|---|---|---|
| cOR | 95% CI | p | aOR | 95% CI | p | |
| Data collection sites | <0.001 | |||||
| Police academy | 1 | 1 | ||||
| Nursing school | 3.40 | 0.43-26.41 | 0.24 | 3.21 | 0.41-25.12 | 0.26 |
| Biggest market of the city | 15.81 | 2.01-124.35 | <0.01 | 10.77 | 1.33-86.92 | 0.02 |
| Private university | 1.61 | 0.14-18.66 | 0.69 | 1.53 | 0.13-17.71 | 0.77 |
| Age (years) | <0.001 | |||||
| <35 | 1 | |||||
| 35-45 | 5.69 | 2.44-13.25 | <0.001 | |||
| 45-55 | 13.76 | 4.16-45.49 | <0.001 | |||
| >55 | 3.56 | 0.71-17.82 | 0.12 | |||
| Family history of diabetesa | 0.29 | |||||
| No previous history | 1 | |||||
| 1st degree family historyb | 1.91 | 0.79-4.58 | 0.14 | |||
| 2nd degree family historyc | 1.53 | 0.61-3.81 | 0.35 | |||
| 30 minutes of daily physical activity | 0.65 | |||||
| Yes | 0.84 | 0.41-1.73 | ||||
| No | 1 | |||||
| Daily consumption of fruit and vegetables | ||||||
| Not every day | 1 | |||||
| Every day | 1.67 | 0.74-3.77 | 0.21 | |||
| Prescription of anti-HTA | ||||||
| Yes | 2.81 | 0.73-10.75 | 0.12 | |||
| No | 1 | |||||
| History of hyperglycemia | ||||||
| Yes | 2.63 | 0.52-13.16 | 0.23 | |||
| No | 1 | |||||
| Waist circumference | ||||||
| No abdominal obesityd | 1 | |||||
| Presence of abdominal obesitye | 4.01 | 1.77-9.08 | <0.01 | 2.59 | 1.08-6.20 | 0.03 |
| BMI (kg/m2) | 0.01 | |||||
| Normalf | 1 | |||||
| Overweightg | 1.83 | 0.78-4.25 | 0.15 | |||
| Generalized Obesityh | 3.33 | 1.43-7.74 | <0.01 | |||
| Marital status | 0.05 | |||||
| Married | 2.25 | 1.10-4.60 | 0.02 | |||
| Single | 1 | |||||
| Widowed | 4.47 | 0.43-45.88 | 0.20 | |||
| Divorced | – | – | – | |||
| Means of transportation | 0.89 | |||||
| Motorcycle | 1.69 | 0.21-13.28 | 0.47 | |||
| Bicycle | 1.57 | 0.13-19.12 | 0.72 | |||
| Pedestrian | 2.30 | 0.23-22.59 | 0.61 | |||
| Car | 1 | |||||
| Type of physical activity | 0.56 | |||||
| Walk | 0.83 | 0.28-2.45 | 0.73 | |||
| Work | 0.75 | 0.25-2.22 | 0.61 | |||
| Leisure | 0.20 | 0.02-1.86 | 0.16 | |||
| None | 1 | |||||
| Lunch location | ||||||
| At home | 1 | |||||
| Out of the house | 0.60 | 0.26-1.41 | 0.25 | |||
| Contraceptive | ||||||
| Yes | 0.95 | 0.44-2.07 | 0.90 | |||
| No | 1 | |||||