Introduction

According to the World Obesity Atlas, the adult obese rate was 15 % in 2020, with a higher prevalence among women than men (17 % vs. 13 %). By 2030, this rate is expected to increase to 18 % (1). Specific biological and social determinants influence the onset of this pathology. In biological terms, for example, these factors include excessive energy consumption relative to energy expenditure (2), physical inactivity, genetic effects, and gene-environment interactions. Among the social aspects, environment, socioeconomic situation, residential segregation, access to health services, transport, and social support all play a role in this malnutrition by excess (3). Obesity is also associated with metabolic conditions, including insulin resistance (4) and the onset of arterial hypertension (5).

In Latin American countries such as Colombia, the prevalence of adult obesity continues to increase. In 2005, 13.7 % of adults were classified as obese (6), but by 2015, the figure had risen to 18.7 %. Notably, obesity was more prevalent among women (22.4 %) than men (14.4 %). Additionally, significant disparities in obesity rates were observed across different socioeconomic groups, with the highest prevalence found among individuals with medium (20.5 %) and low (19.4 %) wealth index, compared to those with high (18.6 %) or very low (16.8 %) wealth index, respectively. Furthermore, when considering race/ethnicity, the Afro-Colombian group had the highest prevalence at 22.9 %, significantly surpassing the figures for indigenous people (14.9 %) and others (predominantly mestizo) (18.5 %) (7). Meanwhile, a prognostic study foresees that, by around 2030, shifts from being overweight to being obese will sharply increase among adults of the lowest socioeconomic status (8).

Based on this statistical background, this scoping review seeks to answer five questions related to the literature on adult obesity in Colombia from 2000 to 2021: 1) What has been the reported prevalence in the studies conducted? 2) What study designs, methods and measures have been used? 3) What demographic and socioeconomic variables have been included in the studies? 4) What were the practices of public health interest linked to the evidence? 5) Which pathologies have been included in the studies? Addressing the above questions aims to identify and describe the sociodemographic and public health dimensions of adult obesity in Colombia. This information will highlight gaps in the literature that need to be closed to support the development of public health policies. Such policies are essential for promoting healthy lifestyles and preventing non-communicable diseases.

Methodology

This study, in the form of a scoping review, helps to identify the types of evidence available, clarify key concepts and characteristics related to an idea and to recognise and analyse knowledge gaps (9) as in the case with adult obesity in Colombia. It is important to note that some nutritional events, such as food choices and food environments, have been studied using this methodology (10,11). We followed the guidelines established in the PRISMA statement to ensure a replicable process (12).

The search, which accepted works in English and Spanish, was conducted on April 27 2021, in PubMed and on May 4 2021, in Scielo and Lilacs. It included the following search terms: “Obes* AND Colombia”; “BMI AND Colombia NOT Obes”; “IMC AND Colombia NOT Obes”; Adiposity AND Colombia NOT Obes*” and was restricted to papers published between 2000 and 2021 because we considered it an appropriate cut-off year based on the availability of nutritional status data in the country. Mendeley was the software used for managing the references, removing duplicates, and obtaining complete documents for review.

Regarding inclusion criteria, original articles were included if they described: 1. Any study of obesity in adults over 18 in Colombia; 2. Any transversal, longitudinal, cohort, quasi-cohort, ecological, or panel study of Colombia. The following criteria were used to exclude articles from the review: 1. Studies concerning children and young people under 18 years of age; 2. Clinical studies due to difficulties in generalising findings resulting from selection or rigidity in intervention strategies and case-control research because the time frame between exposure and pathology was not always easy to establish; 3. Studies about pregnant women; 4. Studies of the student population (i.e., school and university); 5. Studies conducted outside of Colombia, as the aim of this review is to understand the phenomenon of obesity based on the conditions of this country.

Concerning variables of interest, the following were included:

• Nutritional status variables: Body Mass Index (BMI), obesity, abdominal obesity, waist circumference, and body adiposity index.

• Health variables: Physical activity practice and associated comorbidities.

• Demographic variables: Sex, age, and racial/ethnic group.

• Socioeconomic variables: Occupation, educational level, and Socioeconomic Status (SES). In Colombia, SES is a variable that is not classified at the individual level but by dwelling strata, denominated as 1. Low-low; 2. Low; 3. Medium-low; 4. Medium; 5. Medium-high; and 6. High. Of these, strata 1, 2, and 3 correspond to lower levels encompassing small, lower-quality dwellings with limited or no access to municipal services such as sewerage system, running water, and electricity. In contrast, the high strata 5 and 6 contain higher quality dwellings in neighbourhoods where these services are provided, and property owners are obliged to pay taxes for them (13).

The variables of interest were used in thematised narrative form to present the results. The outcome (obesity) is shown through its prevalence in the population. For evaluating the quality of the studies, researchers independently used the Newcastle-Ottawa scale, designed for assessing cross-sectional studies. This scale employs a star system, allocating a maximum of 10 stars to evaluate bias risk of in three areas: selection (five stars), comparability (two stars), and outcome (three stars). Scores of three and below are considered low quality, four to six are deemed average, and seven and above represent high quality. The full scale is ten stars, with more indicating better quality (14).

Three independent researchers obtained information from the studies using a data extraction format. To resolve a few discrepancies, we followed the strategies proposed by PRISMA. We reviewed the inclusion and exclusion criteria and variables defined in the protocol. Meetings were held to reach a consensus with one of the senior researchers acting as the mediator.

Results

Synthesis of the Search

The search strategy initially yielded 1,766 articles. After duplicates (n=49) were eliminated, and another 1,677 were excluded by title and abstract since they did not identify or describe the demographic or socioeconomic dimensions of obesity and their links with adult public health, the list linked 40 citations for assessment by eligibility criteria. Twenty-one articles were identified for full-text review (Figures 1 and 2). In terms of the Newcastle-Ottawa scale, 76.2 % (n=16) of the studies were high quality, and 23.8 % (n=5) were average quality (Table 1). The main characteristics of the included articles are presented in Tables 2 and 3.

What has been the reported prevalence in the studies conducted?

Prevalence of obesity was reported in 16 studies, presenting values ranging from 7.0 % to 28.2 % (Table 3) (15-30). Prevalence of abdominal obesity appeared in eleven studies, ranging from 9.0% to 82.9% (16,17,20,22,25,26,29-33). Moreover, one study included an analysis of average adiposity (0.68 ≤ r ≤0.82, p<0.001) (34). Although two studies covered the race/ethnicity variable, only one offered a race/ethnic analysis of obesity in Colombia. This study showed lower levels of abdominal obesity in the Emberá population (an indigenous ethnic group) of Jardín-Antioquía than those of mixed race living in the provinces of Antioquía, Bolívar and Nariño, with prevalences of 42.6% and 48.7%, respectively (n=159 individuals) (17).

What study designs, methods, and measures have been used?

Twenty articles (95.2 %) used a cross-sectional methodology, and one applied a cohort analysis (35). Most reports included BMI (n=20) (15-33,35), one article focused on the adiposity average (24,27), two studies calculated the level of food security (19,30), and two included the wealth quintile index (19,30).

What demographic and SES variables have been included in the studies?

Regarding demographic variables, four studies only included women (16,21,22,34), and two focused on men (31,32). The articles, focusing on both women and men, were conducted in the country's three main cities: Bogotá, Medellín and Cali. The fifteen remaining studies included men and women (15,17-20,23-30,33,35). All twenty-one studies considered age and covered the population over 18 years in all cases. As noted above, two studies included the race/ethnicity variable (17,31).

In addition, the review found several studies that related a range of socioeconomic aspects to obesity. The identified issues focus on SES, occupation, and means of transport. Firstly, regarding SES-related matters, one study calculated the relationship between general adiposity and SES using factor analysis in a group of women in Cali. Findings showed that the adiposity factor increased in all SES categories, especially in the middle range (34). In the same city, another study analysed trends of SES, height, and obesity of women and found that the low SES showed higher obesity rates than those of medium and high SES (16). An additional study established the social and economic factors of obesity in Medellín and found that obesity is more prevalent in low SES, low educational levels, and people whose income is below 1,400,000 pesos (COP) (23). In addition to low academic levels, being female meant a greater probability of suffering from obesity (27).

The above results are consistent with a study that established a clear association of obesity with other variables. Gender, age, educational level, occupation, family income, and SES were related to the risk of obesity, which is higher in women than in men. Obesity increases with age, it is higher in lower educational levels, it increases in people doing any work (domestic work, in formal and informal employers and employees), and those belonging to middle and low SES groups (28).

Likewise, another study estimated the effects of a tax on sugary drinks on obesity and found that after its implementation, obesity in the lowest SES could be reduced by between 1.1 to 2.4 percentage points (15). In turn, another paper which examined the correlation between overweight, obesity and the perception of fitness among women in Bogotá identified that the number of household goods and the number of births are factors associated with obesity (22).

Our scoping review also identified a significant correlation with the wealth index. This index approximates the relative economic status of households using a measurement that aggregates several economic indicators grouped and classified into quintiles (36). One study showed that belonging to a higher wealth quintile, being female, food insecurity, living in urban areas, and the Pacific region were positively associated with obesity (24). When the index was applied on the national scale, it was also found that higher levels of wealth were associated with the appearance of obesity. However, after calculating social inequalities using the Lorenz curve, it was observed that obesity occurs at every income level and is not limited to any specific economic condition (27).

Secondly, the review identified that workers in different occupational environments exhibited diverse behaviours related to becoming obese. For health professionals, there was no significant association between shift work and obesity, indicating the need for more studies. Conversely, in a company in the metal-mechanical sector, it was found that those with more significant abdominal obesity presented more medical disorders (31).

Finally, studies have shown that owning cars or motorbikes increases the risk of obesity in men (25). The literature even suggests that the more time spent riding a motorbike, the greater the risk of obesity (30).

What were the practices of public health interest linked to the evidence?

One study found that a history of alcohol consumption was related to excess malnutrition. The habit of smoking tobacco is also associated with the development of obesity in overweight individuals, especially in occasional and heavy smokers, the latter being people who smoke more than 40 packets per year (26), according to the authors.

Which pathologies have been included in the studies?

Several studies inquired into associated pathologies and health variables related to obesity. The central themes evidenced were insulin resistance, dependence in older adults, and associated pathologies.

One study showed that the Emberá exhibit lower levels of central obesity and better function of pancreatic β cells than the mixed-race population (17). Concerning dependence in older adults, obesity has a detrimental effect on everyday activities, creates dependency and is also associated with pathologies such as diabetes, chronic obstructive pulmonary disease, coronary heart disease, heart failure, cancer, hearing and vision problems, and arterial hypertension (19).

Other obesity-related diseases are breast cancer, rheumatoid arthritis, and hypertension. In the first case, a study showed that in overweight premenopausal patients, there was a positive association with tumours typical of breast cancer (21). In the second case, after evaluating a programme for patients with rheumatoid arthritis, it was found that obese adults might have fewer possibilities for improvement in their condition than overweight patients (35). Finally, in hypertension, one study showed that prehypertension and stages one and two of arterial hypertension are associated with obesity (33).

On the other hand, abdominal obesity is related to the appearance of diseases such as diabetes, arterial hypertension, and dyslipidaemia (20). One study found that a higher value of visceral adiposity was associated with more elevated LDL cholesterol, glucose levels, and a high lipid-metabolic index by comparison with individuals presenting with less visceral adiposity (32).

Discussion

In general, this scoping review highlights several important points. In terms of demographics and socioeconomic status, obesity in Colombia is more prevalent among women and those of low SES (as defined according to the household economic resources in Colombia). However, it has also recently been identified in medium and high socioeconomic levels. Likewise, it is more prevalent in lower educational levels and low-income households. Furthermore, owning or using a motorbike or car as a means of transport increases the risk of obesity.

Regarding health, the results suggest that the consumption of alcohol and tobacco are risk factors for obesity. Moreover, obesity is associated with pathologies like diabetes, arterial hypertension, dyslipidaemia, hearing and eyesight problems, breast cancer, and rheumatoid arthritis.

Regarding the prevalence of obesity at the local level, some studies reported higher levels in certain cities compared to the national average, such as in Medellín with 28.2 % (21) and Soledad with 24.6 % (29). However, this local obesity prevalence is still lower when compared to those registered in other Latin American countries, such as Argentina and Mexico, which have values of 33.9 % (37) and 36.1 % (38), respectively. Nevertheless, despite the geographic variations, studies in several countries have established that adult populations in social vulnerability are at a greater risk of obesity (39,40).

Despite the scoping review reporting two articles about race/ethnicity, only one analysed the abdominal obesity between Emberá indigenous from Jardin Antioquia and the mixed-race population from Antioquia, Bolivar and Nariño. This finding underscores the significance of understanding obesity within ethnic populations, given that Afro-Colombian and indigenous communities constitute a substantial portion of Colombia's population, accounting for approximately 13 % (41). These groups exhibit distinct dietary patterns (42,43), and their geographic locations often overlap with regions affected by armed conflicts (44). Moreover, many have experienced forced displacement, a primary driver of food insecurity (45,46). Therefore, future studies should incorporate a race/ethnicity-based analysis, offering valuable insights for policymakers. This approach is analogous to developing dietary guidelines for pregnant and breastfeeding mothers, infants, children under two years of age, and the general population, with adaptations tailored to Colombia's diverse food cultures (47,48).

Social vulnerability included variables from the socioeconomic component. Education is one example since -as the review also reports- studies affirm that illiterate people and/or those with primary education or less are at greater risk of obesity (49). When analysing the phenomenon by years of education, the evidence suggests that older adults with six years or less of education are at greater risk of obesity (50). The latter situation is confirmed by studies indicating that more educated individuals present a lower BMI than less educated individuals (51). Even though the causal links between education and obesity are not yet fully understood, prior studies have suggested that a) having greater access to health-related information and enhanced ability to manage such information; b) having a clearer understanding of the risks associated with lifestyle choices; and c) having better self-control and a more stable set of preferences over time may all play a role (52,53).

Low education is often associated with earning less income, and evidence shows that low income and low SES are related to higher rates of obesity and diabetes (54). Several studies indicate that the relationship between obesity and poverty in Latin America shows a rising trend in the population of lower individual-level SES. In poor individuals, obesity is linked with periods of malnutrition during infancy, which produces an adaptive response to low energy intake and, at older ages, increased consumption and hence the onset of obesity (55). As a result, it is essential to incorporate food and nutrition education into various focal points, such as schools, communities, and workplaces, while considering the diversity of food cultures and education levels.

In the study by Gomez et al., (2016), no significant association between workers with fixed shift work hours, e.g., health professionals, and obesity was found (18). Conversely, a systematic review and meta-analysis of nurses found that the risk is higher among professionals doing shift work than among those who do not, especially in the case of night shifts, which turned out to be statistically significant (56).

Another aspect reported as a risk factor for obesity is the ownership and use of motorbikes and cars as means of transport. The Latin American Study of Nutrition and Health (ELANS) reports that active mobility, e.g., walking or riding a bicycle, is associated with lower BMI levels (57). While some Colombian cities, like Bogotá, have reported active bike mobility among their citizens, socioenvironmental factors, such as fatal collisions and high-risk areas prone to catastrophic events, continue to limit its use. (58). The above observation suggests an additional consideration for policymakers regarding city infrastructure.

Finally, the studies relate obesity and/or abdominal obesity with the appearance or coexistence of chronic diseases, as suggested by the literature worldwide. For instance, reviews at the global level suggest connections between obesity and the metabolic complication of diabetes type 2 (59), cardiovascular disease (60), hypothyroidism, Cushing's syndrome, polycystic ovary syndrome, dyslipidaemia, arterial hypertension, coronary heart disease, heart failure, sleep apnoea, and some kinds of cancer (61).

As for country-specific examples in diverse locations worldwide, we found similarities. A cross-sectional study of workers in a region from Spain reported a greater prevalence of diabetes, arterial hypertension, dyslipidaemia, and metabolic syndrome in overweight people (62). In Brazil, a national study found that obesity increases the risk of having diabetes (prevalence ratio: 2.9) by comparison with the risk associated with being overweight (prevalence ratio: 1.8) (63). Likewise, a representative study of citizens using healthcare services in South Korea reported that central obesity and BMI are significantly associated with arterial hypertension, dyslipidaemia, and diabetes (64). Notably, health associations like the European Association for the Study of Obesity (EASO) emphasise the link between cardiovascular disease and obesity (65). While the relationship between obesity and the onset of non-communicable and cardiovascular diseases has been studied, further research is needed in Colombia, particularly considering the country's racial/ethnic, socioeconomic, and geographic diversity.

In conclusion, this scoping review has identified a clear relation in previous studies conducted in Colombia between sociodemographic determinants and obesity, including socioeconomic status, educational levels, environmental factors, and relevant public health diseases. Addressing these determinants comprehensively through integrated public policies is essential, rather than approaching them individually. Additionally, it was somewhat unexpected that the scoping review did not detect any longitudinal, quasi-cohort, or synthetic cohort studies (66) that analyse obesity trends in adults, allowing for the identification of generational changes and the progression of this chronic, non-infectious condition as individuals age, this represents a knowledge gap and a significant opportunity for future research.

Strengths and Weaknesses

The main strength of this study lies in its rigorous search methodology following the guidelines of the PRISMA statement (12). Additionally, the review involves three well-established search engines and employs uniform and reproducible data extraction methods, using a preestablished checklist to minimise errors and bias (14). These three electronic databases focus on studies related to Colombia and Latin America. Nevertheless, a significant limitation was identified. Focusing exclusively on peer-reviewed articles may have resulted in the exclusion of grey literature, which can contribute to understand the phenomenon of adult obesity in Colombia through other dimensions.

Funding

The study supported with funding from the DEMOS_2021 contract through the R&D project “Salud de las personas de edad avanzada el análisis de la comorbilidad las múltiples causas de muerte y las desigualdades de género y socioeconómicas en la salud” (COMORHEALTHSES PID2020-113934RB-I00x|) financed by the Spanish Ministry of Science and Innovation (PI Jeroen Spijker).

References

Notes

* Review article.

Author notes

^a Autora de correspondencia. Correo electrónico: pcastro@ced.uab.es

Additional information

Cómo citar: Castro, P. A., Spijker, J. y Recaño Valverde, J. (2024). Adult Obesity in Colombia from the Sociodemographic and Public Health Perspective: A Scoping Review. Revista Gerencia y Políticas de Salud, 23. https://doi.org/10.11144/Javeriana.rgps23.aocs

Descargar