How to cite this article: García de Alba-García JE, Salcedo-Rocha AL, Hayes Bautista D, Milke Najar ME. Cultural domain of the causes of diabetes in three generations of popular stratum in Guadalajara, Mexico, México. Rev Med Inst Mex Seguro Soc. 2015 May-Jun;53(3):308-15.
ORIGINAL CONTRIBUTIONS
Received: June 26th 2014
Accepted: January 26th 2015
Javier E. García de Alba-García,a Ana L. Salcedo-Rocha,a David Hayes Bautista,b María Eugenia Milke Najarc
aUnidad de Investigación Social, Epidemiológica y de Servicios de Salud, Instituto Mexicano del Seguro Social/Centro Universitario de Ciencias de la Salud, Universidad de Guadalajara
bCentro de Estudios de Salud y Cultura Latinas, Universidad de California, Los Ángeles
cCentro Universitario de Ciencias de la Salud, Universidad de Guadalajara
Guadalajara, México
Communication with: Ana L Salcedo-Rocha
Telephone: (55) 3617 0060, 3668 3000, extensions 31818 y 31887
Email: analeticia_salcedorocha@yahoo.com.mx
Background: The growing prevalence of diabetes must be confronted in several ways. Establishing the generational transmission of cultural knowledge offers some guidelines to prevent and control the disease. Once we identify and compare the semantic structures of shared knowledge we lay the foundations of a culturally comprehensive care. The objective was to characterize the main elements about cultural domain of the causes of diabetes in a population of grandparents, parents and children belonging to popular strata in Guadalajara, Jalisco, Mexico.
Methods: A cognitive anthropological study performed in 104 subjects selected randomly in Guadalajara. We applied the free listing technique in order to obtain the semantic model and the average of cultural knowledge on the causes of the disease through a consensus analysis.
Results: The studied groups were divided by generation: grandparents, parents and children. The data evidences intergenerational transmission, in form of a basic semantic structure, and a significant consensus around a single model.
Conclusions: The semantic structure on the causes of the diabetes includes: a) the emotions, as traditional dimension; b) certain behaviours related with the lifestyle, as everyday dimension; c) some biomedical concepts, like an emergent dimension.
Keywords: Type 2 diabetes mellitus; Cultural domain; Cultural consensus
Type 2 diabetes mellitus (DM2) is a global public health problem that has surpassed generational barriers. It currently affects 140 million people, and in two decades 300 million will suffer from it.1 In Mexico, the prevalence of DM2, 6.7% in 1993, increased to 8.2% in 2000,2 and in 2012 it was 14.6%.3
This increase has affected the response capacity of health institutions and threatens the financing of services,4 plus it raises other challenges, such as glycemic control in patients with DM2.5 It is estimated that this control is exercised in 35.8% of patients in the United States6 and in Mexico it ranges between 36 and 28%.2,7
This implies a challenge to the current model of DM2 care, which aims to promote the prevention and effective treatment of this disease.8
The spatial and temporal extent and complexity of the DM2 problem requires considering its cultural importance, which we believe has as much influence as transgenerational genetic and epigenetic effects.9,10 An example is some associated risk factors, such as obesity.11
The process of cultural agreement is a complex consensual phenomenon that involves aspects of social reproduction and identity, linked in turn with customs, knowledge, practices, images, worldview, etc. This transmission has been evidenced as semantic similarity, which permeates health-related behaviors (in members of a family), an interaction in which knowledge and beliefs underlie these behaviors. Here we define the need to investigate these to integrate a broader conceptual framework.12
The importance of cultural consensus on chronic diseases has been highlighted in several studies,13 which have shown how the urban Latin American population continues to believe in explanatory models of disease that are different from the biomedical. This has obvious implications for care.8,14
Medical anthropology considers cultural knowledge as a process that takes place based in the patient, i.e., from their culture, where people give value, meaning, and significance to their ailments and health, as the bases of their quality and style of life.
Understanding and considering this knowledge from a transgenerational perspective can help obtain evidence for care with better human quality, mainly before the present and future challenges posed by DM2.15 Therefore the theoretical orientation of this work is the cognitive aspect of anthropology, which, through the theory of consensus,16 gives us evidence in specific groups of people and cultural models of a particular aspect of reality or semantic domain, in our case that of the cause of DM2 in three generational groups of the lower classes.
Although there are various definitions of culture, for this work we understand it as shared knowledge of systems of meanings and significance, communicated by natural language and other symbol systems with representational, directive, and affective functions that are capable of creating other cultural entities and giving a particular meaning to reality.17
Note that, like all social processes, culture affects and is affected by other systems of personality and of material and immaterial life. As a structuring and structured structure, this shared knowledge ensures intergroup and intragroup variability of constructed cultural models, either by sharing the culturally agreed knowledge or by social or economic restrictions in the context of the actors,18 which influences the behavior, attitudes, beliefs, norms, and values of everyday life of these people.
The method for determining existence according to the construction of cultural knowledge is called cultural consensus analysis16,18,19 and it requires three conditions for it to be achieved:
Our objective was to characterize the main elements related with the cultural domain of the causes of diabetes in a population of grandparents, parents, and children belonging to the lower classes in Guadalajara, Jalisco, Mexico.
Cognitive anthropological study, with random representative sample of 104 people over 20 years old, belonging to upper-low and middle-low strata (according to geo-economic classification of the Instituto Nacional de Estadística, Geografía e Informática [INEGI]) and who are residents of the sector Libertad in Guadalajara. These people were selected by block sampling, previously numbered. The minimum size was calculated by the formula for cultural consensus studies, which is based on a proportion of 50% with 95% confidence and 5% error, and 90% potency.18,20
The participants, who gave written informed consent, were members of three generational groups: grandparents (defined as having grandchildren over 18 years), parents (defined as having children over 18 years old) and children (adults aged 18 years and up), who live together in the same housing unit.
The study included adults of both sexes with or without evidence of DM2.
Data collection for the participating researchers was conducted during the second half of 2010, by using a form to gather patient demographic data (age, sex, marital status, education, and occupation), and the use of the technique of free lists, which consisted of asking each participant to express "all known causes that produce diabetes."
The free lists technique was applied to each respondent, by the same trained investigator, independently, in a quiet and private environment in each home. The researcher avoided verbally suggesting answers, encouraging the patient with monosyllables, to express the reasons they considered appropriate.18
The descriptive statistical analysis of the data consisted of obtaining absolute and relative frequencies, means, and standard deviations, according to the measurement scale. With inferential analysis we looked for significant differences (p < 0.05) between the distributions studied; we applied Snedecor F-test for numeric data and chi-squared for categorical data, using the program Epi Info, version 6.4.
Cultural knowledge about the causes of diabetes was obtained with the help of the program Anthropac,19 to obtain:
The protocol was approved by the IMSS Committee for Scientific Research and Ethics, with registration number 2010-785-001.
General features
Table I shows how grandparents, parents, and children, as generational groups, showed statistically significant differences in some characteristic variables, such as average age 69, 49 and 28 (F = 304, p < 0.001); marital status: mostly married (chi-squared = 73.6, p < 0.000); schooling inversely proportional to average group age (F = 251, p < 0.000); the percentage who have social security, mostly grandparents (chi-squared = 35.64, p < 0.000); the percentage of productive employment, inversely proportional to average group age (chi-squared = 17.86, p = 0.001); the proportion of people with parental history of diabetes, which was inversely proportional to average group age (chi-squared = 13.28, p = 0.01); and the percentage of subjects with history of diabetes, which was directly proportional to average group age (chi-squared = 33.26, p = 0.000).
Table I General characteristics of the three generational groups (N = 104) | ||||||
Generational group | ||||||
Grandparents | Parents | Children | ||||
n= 23 | n= 37 | n= 44 | ||||
Age (in years) | 69.95 ± 9.23 | 48.78 5.13 ± | 28.72 ± 6.04 | |||
Education (in years) | 5.17 ± 4.54 | 8.29 ± 3.27 | 10.63 ± 3.05 | |||
Mon thly income (in pesos) | 1.56 ± 1.53 | 2.72 ± 2.24 | 3.18 ± 2.04 | |||
Weight (in kg) | 71.33 ± 12.07 | 79.66 ± 13.03 | 75.97 ± 24.45 | |||
Height (in m) | 1.56 ± 0.12 | 1.59 ± 0.09 | 1.64 ± 0.100 | |||
BMI (in kg/m2) | 29.25 ± 5.91 | 31.12 ± 4.77 | 27.74 ± 6.67 | |||
Waist (in cm) | 96.65 ± 24.39 | 96.36 ± 25.75 | 88.96 ± 24.84 | |||
Glucose (in mg/dL | 150.78 ± 112.33 | 116.55 ± 37.89 | 104.18 ± 20.15 | |||
SBP * (in mm HG) | 127.39 ± 48.78 | 131.1 ± 1635 | 118.52 ± 15.30 | |||
DBP * (in mm HG) | 72.08 ± 24.97 | 83.91 ± 1019 | 75.5 ± 10.85 | |||
Pulse (in beats per minute) | 69.65 ± 26.77 | 74.67 ± 21.79 | 80.43 ± 12.37 | |||
% | % | % | ||||
Sex | Men | 43.47 | Men | 32.43 | Men | 43.18 |
Women | 56.52 | Women | 67,56 | Women | 56.81 | |
Marital status | Married | 47.82 | Married | 81.08 | Married | 70.45 |
Single | 8.69 | Single | 16.21 | Single | 25 | |
Widowed | 43.47 | Divorced | 2.70 | Widowed | 2.27 | |
Divorced | 2.27 | |||||
Residential status | Citizen | 100 | Citizen | 97.29 | Citizen | 100 |
Other | 2.70 | |||||
Nationality | Mexican | 100 | Mexican | 97.29 | Mexican | 100 |
Other | 2.70 | |||||
Social security | No | 21.73 | No | 62.16 | No | 50 |
Yes | 78.26 | Yes | 37.83 | Yes | 50 | |
Occupation | In the home | 34.78 | Employee | 35.13 | In the home | 27.27 |
Retired | 17.39 | In the home | 40.54 | Employee | 40.90 | |
Employee | 13.04 | Unemployed | 5.40 | Unemployed | 9.09 | |
Unemployed | 21.73 | Laborer | 2.70 | Student | 2.27 | |
Laborer | 4.34 | Business-owner | 16.21 | Business-owner | 13.63 | |
Business-owner | 8.69 | Laborer | 6.81 | |||
Fa ther's disease history | High blood pressure | 13.04 | Diabetes mellitus | 27.02 | Diabetes mellitus | 34.09 |
Diabetes mellitus | 17.39 | High blood pressure | 16.21 | High blood pressure | 13.63 | |
Obesity | 8.69 | Hypercholesterolemia | 2.70 | Hyperuricemia | 4.54 | |
Other | 4.34 | CVD | 5.40 | Hypercholesterolemia | 4.54 | |
None | 65.21 | Other | 13.51 | Obesity | 9.09 | |
None | 51.35 | None | 47.27 | |||
CVD | 9.09 | |||||
Mother's disease history* | High blood pressure | 17.39 | Hypercholesterolemia | 5.40 | High blood pressure | 15.90 |
Diabetes mellitus | 21.73 | High blood pressure | 21.62 | Diabetes mellitus | 9.09 | |
Obesity | 8.69 | Obesity | 10.81 | CVD | 9.09 | |
Hypercholesterolemia | 4.34 | Diabetes mellitus | 29.72 | Obesity | 4.54 | |
Other | 4.34 | CVD | 16.21 | Other | 2.27 | |
None | 52.17 | Other | 8.10 | None | 72,72 | |
None | 37.83 | |||||
Personal disease history | Hypercholesterolemia | 4.34 | High blood pressure | 13.51 | High blood pressure | 4.54 |
High blood pressure | 39.13 | Diabetes mellitus | 18.91 | Diabetes mellitus | 2.27 | |
Obesity | 13.04 | Hypercholesterolemia | 8.10 | Uric acid | 2.27 | |
Diabetes mellitus | 52.17 | Hyperuricemia | 2.70 | Cholesterol | 6.81 | |
CVD | 4.34 | Obesity | 10.81 | Obesity | 6.81 | |
Other | 17.39 | Other | 21.62 | None | 70.45 | |
None | 30,43 | None | 48.64 | Other | 9.09 | |
Bad habits | Television | 95.65 | Television | 89.18 | Television | 93.18 |
Soft drinks | 69.56 | Soft drinks | 72.97 | Soft drink | 81.81 | |
Tobacco | 17.39 | Beer | 16.21 | Beer | 22.72 | |
Liquor | 4.34 | Tobacco | 29.72 | Tobacco | 36.36 | |
Beer | 13.04 | Table wine | 2.70 | Liquor | 6.81 | |
None | 2.70 | Table wine | 2.27 | |||
None | 2.27 | |||||
BP = blood pressure; CVD = cerebrovascular disease * Did not present statistical differences |
In somatometry, height was greater in the younger generations (F = 5.29, p = 0.006); body mass index (BMI) was of overweight-obesity in all cases (F = 3.31, p = 0.04); waist was at-risk for grandparents and parents (F = 1.14, p = 0.02); also, in some metabolic parameters such as glucose, the average was definitely high in grandparents (F = 4.82, p = 0.009). The average diastolic blood pressure was higher in parents (F = 5.31, p = 0.006).
In addition, participants did not show statistical differences in sex ratio (chi-squared = 3.42, p = 0.18), maternal history (chi-squared = 5.89, p = 0.20), weight (F = 1.43, p = 0.24), systolic blood pressure (F = 2.35, p = 0.10) and pulse (F = 2.38, p = 0.09) (Table I).
Cultural consensus model
Table II shows the semantic models of grandparents, parents and children about the causes of diabetes
Table II Semantic models of causes reported by each generational group | |||||||
Cause | Grandparents n= 23 |
Parents n= 37 |
Children n= 44 |
Observations | |||
Place | % | Place | % | Place | % | ||
Shock | 1st | 48 | 3rd | 24 | 2 nd | 34 | Hierarchical structure of shared cause items: Chi-squared = 10.18 Degree of freedom = 10 p= 0.42 |
Anger | 2 nd | 26 | 6 th | 14 | 5 th | 18 | |
Obesity | 2 nd | 26 | 2 nd | 38 | 5 th | 18 | |
Poor diet | 5 th | 17 | 1st | 46 | 1st | 39 | |
Foods wi th sugar | 5 th | 17 | 4. | 22 | 7 th | 14 | |
Joys | 7 th | 13 | 8. | 8 | 10 th | 7 | |
Mortification | 4. | 22 | |||||
Lack of exercise | 4. | 22 | 7 th | 14 | |||
Genetics | 7th | 14 | 3rd | 32 | |||
Overweight | 4. | 20 | |||||
Fats | 9 th | 11 | |||||
Stress | 11 th | 3 | |||||
Ratio of variances between the first and the other estimated models | 20.22 accounting for 92% of total variability | 22.61 accounting for 93% of total variability | 23.10 accounting for 92% of total variability | High level of consensus and cultural knowledge for a single semantic model of causes | |||
Group knowledge Average Standard deviation |
0.836 ± 0.06 |
0.866 ± 0.05 |
0.848 ± 0.05 |
The sample is representative of the upper-low and middle-low strata of the Libertad sector of the city of Guadalajara, and although it has the causal constraints of a quantitative cross-sectional study, it should be noted that its internal validity and cultural approach encourages the need to deepen the theme studied, given the demographic importance of the lower classes in urban areas of the country, in order to understand their perspective on the genesis of diabetes and improve health promotion.
Moreover, in the results reported, the differences are mainly explained by the socio-biological characteristics of the three groups studied.
In contrast, semantic models found in the three groups express generational agreement as similarities in their semantic elements (p = 0.42), which we believe are mediated by the culture that the informants share and control.11
We think that the items or elements that make up the semantic model of each generation are the cultural result of generational dynamics, related to the process of socialization and the experience of disease.21
In the structures of the generations we found that three major cognitive dimensions coexist:
• The traditional, which is related to emotional elements: shock, anger, mortification, and joy, which show the affect based in ancestral folk knowledge.
• The everyday, which is related to components of the lifestyle of lower-class groups: poor diet, foods with sugar and fat, lack of exercise,
• The emerging, which is built based on modern biomedicine and is gradually shared by the generations studied: genetics, obesity, overweight, and stress.
This three-dimensional model, in the three generations studied, presents high rates of shared knowledge (> 0.80), results that are consistent with other studies,6,22 but conducted in populations that do not differentiate generational groups, so our results show not only the existence and permanence of a lay cultural model, but also suggest the movement of shared cultural knowledge, in this case three generation groups from the lower classes, urban area inhabitants, for the causes of diabetes.
This leads us to suggest that the model found lies in a process of generational cultural transmission that works to, first, maintain the tradition that survives in Mexico to attribute loss of health to "strong" emotions (such as shock),23 causes that come before biomedical disease etiology and embody illness in the experience of people, to reconfigure their identity and to make sense of current stress and psychological distress,24 especially if we consider that the lower classes historically suffer structural violence, whether social, gender, or family. Hence their mentioning causes such as shock, anger, rage, and even redefining paradoxical emotions such as joy.
Secondly, this process of cultural transmission serves to express priorities within the daily life of the lower classes, in which the satisfaction of the daily requirement of feeding oneself7 makes the food component occupy a prominent place, ranging from 43, 38 and 45% (p = 0.58) for generations of grandparents, parents and children, and thus the health and functionality of the individual is maintained.
Thirdly, this process serves to incorporate and adapt to its existence the impact of contemporary biomedicine, as it adds new items from this discipline: 25% in the generation of parents and 45% in the children.
According to Segall and Goldstein,25 similarities and differences found between and within the semantic models described, include and express, at first, the way people link health and disease patterns to wealth and poverty. For example, until the twentieth century, obesity was a marker of wealth, and malnutrition a marker of poverty. However, the development of a consumer economy paradoxically fostered obesity in poverty,11 which produced a high prevalence of overweight-obesity (> 70%)3 in the lower classes.
In a second stage, these similarities and differences explain for Segall et al. the incorporation of critical attitudes of dissatisfaction or recognition of the limits of biomedicine, as well as some traditions related to care practices, such as: the main causes have an external locus of control,26 such as those of emotion and food, which are issues that involve tensions with health professionals (as they favor the culture of reprimand, prohibition, and enforcement), a situation that requires understanding to facilitate self-care and care; otherwise, educational activities might seem like coercion by the health team.
In a third stage, these similarities and differences explain for Segall and Goldstein lay or specialized knowledge on the effects of lifestyle on health and the desire to exercise or not their personal responsibility for health. For example, it is said that the idea of chronic disease is not well digested by the community, which sees many of its causes as reversible, causing treatment failures in the lay population, which believes that simply by no longer eating sweets, resting, or drastically reducing daily intake, their health will improve; however, by proposing this in a communication context of care, though formalistic, it may be defective, because of failure to use indirect communication for bad news, highly respected in lower classes, for whom conversational customs involve ritual presentation, demonstrations of respect, tranquility in the topic (not hurried) and relative proximity (a situation that is sometimes left out when communicating the diagnosis of diabetes).
The intergenerational transmission of knowledge and health behaviors is given by mechanisms such as values, socialization, assigned roles (whose extent depends on the presence of parents), quality of the relationship, similarity of gender, inherited status, exposure to similar structures of opportunity and, in the case of young adults, the propensity to use cultural stereotypes disseminated by the media, and religion.27 For example: socialization in the family makes its members learn what disease is and how it is treated; that is, as a mechanism for selection, it also homogenizes the cultural capital of its members, the experience of the disease. Especially if it is chronic.28
Thus the cultural dominance of the causes of diabetes evidences a complex process in which the dimensions are dynamically constructed and reconstructed: traditional, which maintains folk significance and meanings; everyday, product of domestic praxis for meeting basic needs; and the emerging, harmonizing this knowledge with modern biomedical care.
For this last reason it is necessary, among other things, to take into account the generational aspect to provide culturally comprehensive care when implementing specific adjustments to programs for prevention and control of diabetes.
Cultural studies formalize what appears obvious, as in the case studies of generational knowledge of the causes of diabetes mellitus, which can help create alternative programs29,30 that take into account the three dimensions of lay causality for diabetes mellitus, adapting these programs to group subculture, whether they are applied individually or in groups, both for healthy individuals and patients with type 2 diabetes mellitus, during institutional opportunities, being attended by the health team, situations in which it is possible to influence not only by increasing their knowledge, as is traditionally done, but also to develop skills and practices that generate self-confidence in the patient and consciousness in the health team.
A representative sample of grandparents, parents, and children belonging to the upper-low and middle-low strata in the sector Libertad in Guadalajara presented a model of lay causes of diabetes mellitus, similar in shared semantic elements, for three dimensions: traditional, everyday, and emerging. This structure can be a useful means of establishing culturally comprehensive educational programs for both patients and the health team.
The project received INN03D support from the Programa de Investigación en Migración y Salud (PIMSA), the Initiative for the Americas, and California Program on Access to Care to Berkeley School of Public Health.
We thank PIMSA for their support (INN03D) for the development of this work.
Conflict of interest statement: The authors have completed and submitted the form translated into Spanish for the declaration of potential conflicts of interest of the International Committee of Medical Journal Editors, and none were reported in relation to this article.