How to cite this article: Ortiz-Hernández L, Pérez-Salgado D, Tamez-González S. Socioeconomic inequality and health in Mexico. Rev Med Inst Mex Seguro Soc. 2015 May-Jun;53(3):336-47.
Received: June 26th 2014
Accepted: December 29th 2014
Luis Ortiz-Hernández,a Diana Pérez-Salgado,a Silvia Tamez-Gonzáleza
aDepartamento de Atención a la Salud, Universidad Autónoma Metropolitana Xochimilco, Distrito Federal, México
Communication with: Luis Ortiz-Hernández
Telephone: 01 (55) 5483 7573, Fax: 52(55) 5483 7218
Objective: To establish the relationship between socioeconomic inequality and health problems amongst Mexican population reviewing studies with national or regional representation.
Methods: A literature search was performed at national and international databases using the following keywords: health, disease, mental disorders, nutrition, food, social class, social status, unemployment, employment, occupation, income, wage, poverty and socioeconomic status. Reports of national or regional surveys conducted from the nineties were included.
Results: Mostly, diseases events were more common among people from low socioeconomic status: anencephaly, viral infections, anemia, transit accidents by run over, metabolic syndrome, hypertension, affective disorder, anxiety and substances abuse; some malignancies, difficulties to perform activities of daily living, and poor perceived health status. On the opposite, as it goes down in the social scale, are less frequent some protective factors (e.g. fruits or vegetables intake and physical activity) and there is less access to medical aid and preventive interventions (e.g. condom use or diagnosis and treatment for HIV infection, hypertension or obesity).
Conclusions: Socioeconomic status affects all living conditions; therefore, its effects are not confined to certain diseases, but a general precarious state of health. The conceptual and public policy implications related with social inequalities in health are discussed.
Keywords: Socioeconomic status; Social class; Health; Social determinants of health
In recent decades there has been a growing worldwide interest in the study and the elimination of disparities in health. In 2008 the Commission on Social Determinants of Health, World Health Organization1 (WHO) published the report Closing the gap in a generation, in which socioeconomic differences are exemplified between and within different countries selected. The third general recommendation that the commission issued states is that it is necessary to recognize that getting the magnitude of health inequity measured at national and global level is an essential starting point for action. Social inequalities and inequities in the health field have been considered unfair, unjust, and unnecessary (though not inevitable nor irremediable) since they systematically aggravate the conditions of vulnerable populations, given the underlying social structures and political, economic, and legal institutions.2
It is well known that among the characteristic features of Mexican society is the concentration of wealth and that most of its population lives and has lived in poverty. The population in a situation of material poverty (i.e., with insufficient income to meet needs for food, health, education, clothing, housing, and transport) went from 53.1% in 1992 to 69.0% in 1996, down to 42.7% in 2006, and has rebounded to 51.3% in 2010.3 Food poverty (which refers to insufficient income to cover food requirements) was experienced by 18.8% of the population in 2010, and capability poverty (insufficient income to meet food, health and education needs) by 26.7%. This high poverty rate is largely due to the concentration of income: 10% of the wealthiest households in the country account for 34.6% of current monetary income, while 60% of the poorest households have 26.6%.4
In Mexico several academic groups have begun the study of social inequalities starting in the seventies; however, the approaches were methodologically poor because they did not use representative samples.5 Starting in the late eighties our country began doing representative surveys at the national, regional and local levels; lots of data obtained from these have been published in national and international journals. The results of these studies should be systematized because, beyond just being representative, in most of the cases they were performed following appropriate protocols for measuring the variables of interest.
Considering this, the objective of this study was to identify the relationship between socioeconomic inequality and health problems in the Mexican population based on review of nationally representative surveys conducted starting in the nineties.
A bibliographic search on investigations that have analyzed the relationship between socioeconomic stratification and health and disease in Mexico was conducted through the following databases: MEDLINE (http://www.ncbi.nlm.nih.gov/entrez/) Scientific Electronic Library Online (SciELO) (http://www.scielo.org) Periódica y Clase de la Dirección General de Bibliotecas of Universidad Nacional Autónoma de México (UNAM), Índice de Revistas Médicas Latinoamericanas (IMBIOMED) and compilations included in the EBSCO HOST Research Databases (Academic Search Premier, Fuente Académica, Medica Latina, Medline, Psychology and Behavioral Sciences, and Sociological Collection).
Inclusion criteria were studies that had among its objectives to empirically address the relationship between health and socio-economic stratification, and report data representative of a well-defined geographical area (municipality, state, or nation) of Mexico. Ecological studies were excluded because the interest was to document the relationship between social position and health events at an individual level. The search was restricted to documents published between 1990 and 2012.
In each database a search was conducted with the following keywords: salud, enfermedad, trastornos mentales, nutrición, alimentación, clase social, estrato social, desempleo, empleo, ocupación, ingreso, salario, pobreza y socioeconómico (health, disease, mental disorders, nutrition, diet, social class, social stratum, unemployment, employment, occupation, income, wages, poverty, and socio-economic). In addition, article references were reviewed to identify others that had not been found in the databases consulted. Subsequently, studies were classified according to age group studied, forming three groups: children and adolescents (under 18), adults (18-49 years), and older adults (50-60 years or more). The age limits were difficult to establish, because each study used different criteria; therefore, when individuals under 49 were included, the study was categorized as adults (although people aged 60 or older were included), but when it included people over 50, these were located in studies of older adults. Studies that included people 12 years or over, without presenting age subgroups, were located in adult studies.
The following information was extracted from each study: population, measurement of health events, measurement of socioeconomic position or socioeconomic status, and results in terms of differences in prevalence rates or measures of association. Tables showing detailed information of each study were sent as supplementary material for the evaluation of the article and can be requested from the authors.
Due to space constraints, here is a summary of the study findings is presented. Three tables (children and adolescents, adults, and people over 50) were made to summarize the results of 48 studies. The tables of results show the association as negative (i.e., the frequency of the event was lower in people with higher socioeconomic position) or positive (the frequency of the event was greater in people with higher socioeconomic status) when there were differences between at least two groups. Difference or association was considered present when studies reported a value of p < 0.050 (when proportions, rates or averages were compared), or confidence intervals that did not contain units (in the case of measures of association, such as odds ratios or prevalence ratios) or non-overlapping confidence intervals (in the case of measures of central tendency). In the case of the studies that reported measures of trend and reported no statistical significance, it was considered that there was an association when there was a linear relationship between the indicator of socioeconomic stratification and the health event.
Table I presents the review of studies of children and adolescents.6-20 Children and adolescents from low socioeconomic levels (as identified with by one of the following indicators: low-income households, illiterate parents, poor housing, less ownership of goods, or not participating in a cash transfer program) were more likely to have antibodies to the hepatitis A virus, to have lower consumption of fruits and vegetables, and to receive a smaller proportion of caloric and micronutrient requirements such as vitamin A and zinc. The risk of anencephaly is more frequent in lower classes. Preschool and school-age children of low socioeconomic status had a higher risk of susceptibility to poliovirus 1 and anemia, as well as higher levels of cortisol.
In adolescents of low socioeconomic status (those with lower monthly expenses or whose parents have low education and low subjective social status in society), condom use was less common, and herpes infection was more frequent, as was susceptibility to presenting diseases of moderate severity.
By contrast, children and adolescents with better socioeconomic status (attending private school, paid work, higher maternal education) were more likely to be overweight or obese. Teenagers in this group also had higher alcohol use rates, and more frequently reported illnesses perceived to be of mild severity.
Tobacco consumption among adolescents in some cases was associated with indicators of higher socioeconomic position, and in others with those denoting low social status. In adolescents there were no differences between socioeconomic strata in the general perception of health status, or in the use of drugs other than tobacco and alcohol.
|Table I Review of literature on the association of socio-economic stratification with health events in children and up to age 21|
|Population (age group)||Socio-economic factor analyzed||Event||Association *|
|Newly born-SESNTD in Puebla, Guerrero and Estado de México6||Maternal schooling, income, and mother’s occupation||Risk of anencephaly||Negative|
|1 to 9 years-ENSA 20007||Literacy of head of family||Risk of susceptibility to poliovirus 1||Negative|
|1 to 19 years-ENSA 20008||Family income||Antibodies to hepatitis A virus||Negative|
|2 to 6 years in rural areas
(< 2500 inhabitants) belonging to low-income families9
|Enrollment in PDHO government program||Concentrations of cortisol in saliva||Negative|
|Preschoolers, school-age children, and teenagers-ENSANUT 200610||Index of wellbeing in housing conditions||Consumption of fruits and vegetables||Positive for the consumption of fruit in preschoolers and school-age children, and for consumption of vegetables in all three age groups|
|1 to 4 years-ENSANUT 200611||Index of socio-economic level||Inadequate intake of calories and nutrients (fat, vitamins A and C, folate, iron, zinc, and calcium)||Negative for calories, fat, vitamins A and C, zinc, and calcium|
|5 to 11 years-ENSANUT 200612||Index of socio-economic level||Inadequate intake of calories and nutrients (vitamins A and C, folate, iron, zinc, and calcium)||Negative for calories and all nutrients|
|12 to 19 years-ENSANUT 200613||Index of socio-economic level||Inadequate intake of calories and nutrients (protein, vitamins A and C, folate, iron, zinc, and calcium)||Negative in men for vitamins A and C , iron, zinc and calcium, and in women for calories, protein, vitamin A, and zinc|
|Children under 12 -ENSANUT 200614||Index of socio-economic level||Prevalence of anemia: hemoglobin < 110 g/l for children 12 to 71 months and < 120 g/l for children from 6 to 11 years of age||Negative|
|Children under 5 -ENN 198815||Schooling of head of household||Overweight||No association|
|2 to 18 years - ENSANUT 200616||Index of socio-economic level||Prevalence of overweight and obesity according to International Obesity Task Force criteria||Positive in all age groups|
|10 to 21 years in suburban towns in areas of high marginalization- PDHO evaluation survey17||Indicator of poverty, paid work, attending school, and type of school (public, private)||Tobacco use||Positive for paid work and private school, and negative for poverty and school attendance|
|Teenagers from poor households from small urban areas18||Monthly monetary expenditure and schooling by age||Consumption of tobacco and alcohol, condom use, and sexually transmitted infections (Chlamydia, specific antibodies for herpes type 2)||Positive for alcohol consumption and condom use by expenditure and negative for tobacco with schooling by age and with herpes type 2 by expenditure|
|Teenagers from poor urban communities in seven states19||a) OSEP: maternal schooling and monetary expenditure.
b) SSSC and SSSS
|Smoking, alcohol consumption, and drug use||Positive for tobacco and alcohol with OSEP and by SSSC, negative for tobacco and alcohol with SSSS and no association for drug use|
|12 to 19 years from Morelos schools20||Maternal schooling, paternal schooling, mother’s occupation, and father’s occupation||Self-report of disease and disease severity (moderate or slight)||Negative for self-report of moderate disease severity, positive for slight disease severity|
* Based on measures of statistical significance
SESNTD = system of epidemiological surveillance for neural tube defects; ENSA = encuesta nacional de salud; ENSANUT = encuesta nacional de salud y nutrición; EPDHO = Programa de Desarrollo Humano Oportunidades; OSEP = objective socio-economic position; SSSC = subjective social status within community; SSSS = significant social status within the society
Table II presents the review of the studies of Mexican adults.21-46 The population with low socioeconomic status (i.e., less educated, illiterate, low position at work, who perceived having lower subjective social status within the country, experiencing financial insecurity, coming from households of poverty or low income, or living in homes with poor conditions) had a greater risk of infection by V. cholerae and hepatitis A and B viruses, human immunodeficiency virus, and of having anti-Treponema pallidum antibodies present, as well as getting run over in traffic accidents, having metabolic syndrome and hypertension (diagnosed at the time of the survey and at death), perceiving that their health was not good, and experiencing affective symptoms or disorders, anxiety, and substance abuse. Women living in households of low socioeconomic status had a higher risk of cervical intraepithelial neoplasia, as well as depressive symptoms or anemia. People of lower socioeconomic status also less frequently undertook physical activity, less often were diagnosed as obese and received care and treatment; besides tending to consume fewer fruits and vegetables, while at the same time having higher prevalence of inadequate intake of calories, proteins, and nutrients (vitamins A and C, iron, zinc and calcium). Among people with low schooling, condom use and having been tested for HIV infection were less frequent.
|Table II Review of literature on association of socio-economic stratification with health events in adults in Mexico|
|Cases of cholera reported 1991-6 in Mexico21||Poverty index||Incidence of cholera||Positive|
Population over 20 years
|Literacy, position at work (PW) and income||Hepatitis C virus (HCV) Infection||No association|
|Population over 20 years
|Literacy, position at work, and income||Hepatitis B virus (HCV) Infection||Negative with literacy|
|Population over 20 years
|Family income||Total antibodies against hepatitis A virus||Negative|
|Population over 20 years
Population over 20 years
|Schooling and position at work (PW)||Human immunodeficiency virus infection||Negative for PW
No association with schooling
|Women aged 20 to 80 IMSS enrollees in the State of Morelos26||Index of socio-economic level (ISL) and schooling||Cervical intraepithelial neoplasia or cervical cancer||Negative with ISL No association with schooling|
|People of all ages ENSANUT 200627||Income||Traffic accidents (crashes and being run over)||Negative in being run over
Positive in crashes
|Population 12 to 59 years
|Index of wellbeing of housing conditions||Consumption of fruits and vegetables||Positive|
|People from 20 to 59 years
|Index of socio-economic level||Inadequate intake of calories, fiber, protein, fat, carbohydrates, vitamins A and C, folate, iron, zinc, and calcium||Negative for calories, protein, vitamins A and C, iron, zinc and calcium|
|Women 12 to 49 years-
ENN 1999 and ENSANUT 200629
|Index of socio-economic level||Anemia (Hb < 12.0 g/dL for non-pregnant women and < 11.0g/dL for pregnant women)||Negative|
|Women 12 to 49 years
|Adults 20 to 69 years
|Index of socio-economic level (ISL) and schooling||Risk of overweight (BMI of 25 - 29 kg/m2) and obesity (BMI > 30kg/m2)||Positive|
|Adults over 20 years
|Index of socio-economic level (ISL) and schooling||Risk of overweight (BMI of 25 - 29 kg/m2) and obesity (BMI > 30kg/m2)||No association with overweight
Positive for ISL with obesity
|Population 20 to 64 years
|Family and education expenditure||Diagnosis of obesity by health professional (Dx), having followed a treatment for weight control (Tx), having lost weight in the last year (WL), and prevalence of overweight and obesity||Positive for expenditure and schooling with Dx and Tx
Positive for expenditure with obesity
Negative for schooling with obesity
No association for overweight and weight loss
|Population 18 and up Encuesta Nacional de Evaluación del Desempeño33||Household expenditure||Moderate or vigorous physical activity||Negative in women and among inhabitants of urban areas and cities|
|Women 12 to 49
Encuesta Nacional de Nutrición 199934
|Index of socio-economic level and schooling||Athletics and low physical activity||Positive for ISL with athletics|
|Adults 20 and up-ENSANUT 200635||Schooling and income||Metabolic syndrome (SxM) defined by the 3 criteria (AHA/NHLBI, NCEP-ATP and IDF)||Negative with schooling in Sx according to IDF criteria
No association with income
|Adults 20 and up-ENSA 200036||Schooling and income||Dm2 defined by previous medical diagnosis (PD) or finding in the survey (FS, capillary blood glucose fasting ≥ 126 mg/mL or casual capillary blood glucose ≥ 200 mg/dL)||In women: negative with schooling
No association with income
|Population 30 and up from low socio-economic level families in the metropolitan area of Mexico City
|Index of socio-economic level (ISL) and schooling||Dm2 defined by previous medical diagnosis (PD) or finding in the survey (FS, capillary blood glucose fasting ≥ 126 mg/mL or casual capillary blood glucose ≥ 200 mg/dL)||Negative for ISL with FS
Negative for schooling with PD and FS
|Adults 20 and up-200638
|Index of socio-economic level (ISL)||Dm2 (fasting blood glucose ≥ 126 mm/dL or previous medical diagnosis)||Positive|
|Adults 20 and up
|Schooling and income||Early onset of Dm2 (diagnosis before 40 years of age)||Negative for schooling in people older than 40 years
No association with income
|Adults 20 and up
|Schooling and income||High cholesterol (cholesterol ≥ 200 mg/dL), hypertriglyceridemia (trigliceridemia ≥ 150 mg/dL) e hypoalphalipoproteinemia (HDL < 40 mg/dL)||No association|
|Adults 20 and up
|Education and index of socio-economic level (ISL)||Hypertension defined by previous diagnosis (PD) or survey findings (FS, systolic blood pressure ≥ 140 mm Hg and/or diastolic blood pressure ≥ 90 mm Hg)||Negative for schooling with total prevalence, PD and FS
Negative for ISL in total prevalence and FS
Positive for ISL with PD
|Population 18 to 65 years
|Financial security and subjective social status in country and community||Affective mental disorders, anxiety, and substance use||Negative|
|Poor women from 20 to 70 from rural areas of 7 states43||Schooling of women and head of family||Depressive symptoms||Negative for schooling of head of household
No association with schooling of women
|Population 18 and up -
Encuesta Nacional de Evaluación del Desempeño 200244
|Family income and schooling||Depressive symptoms (DS) and perception of good health (PGH)||Negative for income with depressive symptoms
Positive for schooling with condom use and HIV testing
|Population 15 and up45||Schooling||Daily consumption of tobacco||No association|
|Records of mortality associated with hypertension from 2000 to 200846||Schooling||Mortality due to hypertension||Negative|
|SEP = socio-economic position; ENSA = Encuesta Nacional de Salud; ENURBAL = Encuesta Urbana de Alimentación; ENEP = Encuesta Nacional de Epidemiología Psiquiátrica; IMSS = Instituto Mexicano del Seguro Social; ENSANUT = Encuesta Nacional de Salud y Nutrición; HB = hemoglobin; Dm2 = type 2 diabetes mellitus; BMI = body mass index
= Only trends were identified but measures of statistical significance were not reported
= Based on measures of statistical significance
By contrast, people with better socioeconomic status (better housing conditions, less overcrowding, and increased ownership of goods) presented a higher frequency of crashing in traffic accidents and previous diagnosis of hypertension.
The results of studies on the prevalence of type 2 diabetes mellitus (Dm2) and obesity are inconsistent. On the one hand people who lived in homes with more infrastructure and property were more likely to develop type 2 diabetes; on the other, among people with less education and lower socioeconomic index, prevalence of this disease was higher. It has also been found that among people with lower education, a higher percentage become sick with Dm2 at an early age (before age 40). In the case of obesity, it has been positively related to indicators based on household spending and housing conditions (prevalence was higher in those with better position), but negatively with schooling (prevalence was higher in those with lower position).
There were no differences between socioeconomic groups in the prevalence of infection with hepatitis C virus, dyslipidemia, overweight, nor with the frequency of intentional weight loss or tobacco consumption.
Table III presents the review of studies conducted in adults over 50.47-54 Older adults with low socioeconomic status (those who perceived their economic situation as poor, had less education or illiteracy, worse housing conditions in childhood, or lower income or wealth) were more likely to have difficulty performing basic activities, instrumental activities of daily living or undertaking physical activity, as well as rating their health as poor, suffering physical pain or arthritis. Women of low social status were more likely to have respiratory disease, heart attack, stroke, and diabetes mellitus. The men of low status were more likely to have depressive symptoms.
|Table III Review of literature on association of socio-economic stratification with health events in older adults in Mexico|
|People over 65 - ENASEM 200147||Perception of economic situation||Functional dependency in activities of daily living||Negative|
|People over 60 - ENASEM 200148||Schooling and self-perception of socio-economic level||Risk of overweight||Positive|
|Sample selected from database - ENASEM 200149||Childhood housing conditions in (CHC) and schooling of the interviewee||Hypertension (HT), respiratory diseases (RD), heart attack (HA), stroke (S), arthritis (A) and diabetes (D)||Women: Negative for CHC with RD and schooling with RD, HA, S, A and D
Men: Negative for schooling with A. Positive for CHC with HA
Both sexes: positive for schooling with HT
|People over 50 - ENASEM 200150||Schooling and income||Use of preventive health services (UPHS), visits to the doctor (DV) and hospitalizations (H)||Positive for schooling and income with UPHS and DV|
|People over 50 years enrolled in Oportunidades program51||Literacy||Depressive symptoms||Negative in men|
|People over 50 - ENASEM 200152||Schooling||Frequent physical pain||Negative|
|People over 50 - ENASEM 200153||Education, income, and wealth (i.e. net value of individual properties such as homes, businesses, rental properties, capital, vehicles and other assets)||Health self-assessment (HSE, from excellent to poor), difficulties in performing activities of daily living (DADL, e.g. dressing, walking across a room, or bathing), limitations on physical activity (LPA, e.g. climbing some stairs, bending over, extending arms above shoulders or lifting heavy objects), obesity (OBES, BMI > 30), alcoholism (> 3 drinks / day in the last 3 months) and smoking cigarettes (at time of survey)||Urban areas (≥ 100,000 inhabitants):
Negative for schooling with HSE, DADL, LPA and OBES
Negative for income and wealth with HSE, DADL, and LPA
Positive for income with smoking and alcoholism
Less urban areas (≤ 100,000 inhabitants):
Negative for schooling with HSE and alcoholism
Negative for wealth with tobacco consumption
Positive for schooling with obesity
Positive for income with obesity and alcoholism
|People over 65. Analysis of ENASEM 200154||Schooling and employment situation (unemployed)||Current consumption of alcohol and tobacco||Positive for schooling with alcohol consumption
Unemployment was associated with increased tobacco consumption
|ENASEM = Encuesta Nacional de Salud sobre Envejecimiento en México; BMI = body mass index
Based on measures of statistical significance
By contrast, people with better socioeconomic status (more education or income and better living conditions in childhood) were more likely to report having hypertension, to use preventive health services, and to have visited the doctor. In men, better social status was associated with increased likelihood of reporting heart attacks.
There were no consistent patterns for the consumption of tobacco and alcohol, as for rates of obesity and overweight. However, it was found that in older adults in urban areas, the relationship of obesity to schooling was negative, but in the less urban areas the ratio was positive. Moreover, in the total population, better social position was associated with increased consumption of alcohol. In urban areas, better socioeconomic status was positively related to smoking, while in suburban areas the ratio was negative by education and wealth, but by positive income. Finally, unemployment was associated with increased tobacco consumption.
With this review of the literature it became apparent that most, though not all, disease events are more common in people of low socioeconomic status in Mexico. Many of these differences are confirmed in all age groups, which may mean that disparities begin in childhood and persist throughout the life cycle. Similarly, most protective factors (consumption of fruits and vegetables or doing physical activity) are less common in people of low socioeconomic status, while they less often receive medical care or have less access to preventive measures (such as condom use or the diagnosis and treatment of diseases such as hypertension or obesity). The implications of these results are discussed in detail below, distinguishing the conceptual issues related to the study of health inequalities, for the implications in terms of measures undertaken to reduce these disparities.
Conceptualization of inequalities
The fact that people of low status have more risk for most negative health events implies that socioeconomic status is a fact that seems to be widespread for these groups. That is to say, it affects the totality of the living conditions of people and therefore their effects are not confined to certain diseases, but rather these people are conditioned to a precarious state of physical and mental health. This is because the social location of groups determines a structure of risks and opportunities that limit or promote the development of the potentialities inherent in the human.55
Therefore, socioeconomic status cannot be understood as one more "risk factor" with particular effect on certain diseases.
The review also verified the existence of a socioeconomic gradient in various conditions (such as overweight in women of reproductive age and depression). That is, differences are not only observed between poor and non-poor, but within the non-poor differences in health conditions were recorded, i.e. that these differences depend not only on the existence of poverty. The existence of a socioeconomic gradient in health indicates that apart from the material disadvantages that inequality (poverty) might involve, the existence of hierarchies within a relatively economically homogeneous population can have deleterious effects on the health of those individuals or groups who find themselves at the bottom. These negative effects may occur through psychosocial processes and neuroendocrine responses.56,57
In this review it was noted that some negative health events (death by car crash or higher rates of overweight, obesity, and alcohol consumption among children and adolescents) are more frequent in the population with better socioeconomic status. In addition, the severity of some health events also varies between strata: the more severe the condition, the more likely it is to have a socioeconomic gradient adverse to those of low status. For example, obesity and early-onset diabetes is more common in adults of low socioeconomic status, while the opposite occurs with overweight and diabetes in general; similarly, adolescents of low socioeconomic status more often report severe illness; while teenagers with better position more often reported mild disease severity. And in some health events no difference between strata have been proven (such as perception of health status in adolescents or dyslipidemia in adults). These trends imply a need to explore the specific mechanisms by which social inequality has an impact on particular health events.58
Another finding was that as one moves up the social ladder, one is more likely to adopt preventive measures and treatment, including: the consumption of fruits and vegetables, undertaking physical activity, condom use, going to health services, being diagnosed and/or receiving treatment for diseases such as hypertension or obesity, as well as getting tested for HIV infection. These differences may reflect the limited access that people with low socioeconomic status have to health services; they may also show that people with higher socioeconomic position seek more services and are more receptive to health education campaigns. Thus, the sectors with better socioeconomic status benefit more from scientific knowledge and technological developments related to health.59 The end result is that preventative measures and medical care may be contributing to social inequalities.
Actions to reduce socio-economic inequality in health
The evidence derived from studies on social inequalities in health can give directionality to the design of government policies and programs to reduce inequality and therefore poverty. Some suggestions are listed here.
This review has shown that differences associated with social position are common in the Mexican population and, therefore, first it is necessary to create an epidemiological system for surveillance or monitoring of socioeconomic inequalities in health. Currently the official health statistics are recorded and reported regardless of differences between social groups, so it is not possible to know whether the health gains are achieved in equal measure by the entire population, or if the improvement in averages was parallel to increased differences. Therefore, it is necessary for official health statistics to incorporate monitoring of socioeconomic differentials. Also the methodological discussion should be resolved about which is the best indicator of socioeconomic status to include in statistics. This review did not find any one to be superior to any other. For example, neither schooling nor family income (nor expenditures) showed superiority over the other in terms of discerning social differences. Perhaps maternal or family head schooling is the ideal indicator for its simplicity of assessment and relative comparability. However, in Mexico there is little experience in the use of occupational class as indicator of socioeconomic stratification, although it has been shown that these allow more precise identification of health differentials.60 For future studies and epidemiological surveillance systems, the usefulness of different conceptual and methodological proposals of class typologies based on occupation should be explored.60
Secondly, with regard to health care services and prevention, the evidence shows that in Mexico use and effect are clearly stratified by socioeconomic position. These results should alert planners on the need to promote the use and effectiveness of services in areas and populations of low socioeconomic status. In the specific case of health workers and professionals, these results call to reflect on the training of doctors specifically and of all health workers in general. It would be good to include in health professions curricula a review of social inequalities in this area as a way to sensitize professionals to the problems which they will face every day, especially if they are in public services. This would also help them to recognize that a majority of the Mexican population is facing economic and social obstacles, so they are unable to adopt healthy behaviors, and these obstacles cannot be modified by people.
Third, it is necessary in setting the public agenda and government actions, that the discussion focused on poverty be broadened to also include the structural social inequality that characterizes Mexican society. Currently government action is focused on reducing extreme poverty through cash transfer programs (such as the Oportunidades [Opportunities] program). These programs have a limited approach,61 as they emphasize only the physical dimension of the health of children under five and their mothers (i.e., the nutritional status of children and maternal and infant mortality). This ignores aspects such as mental health and other events and behaviors related to health throughout life. It should be stressed that socio-economic inequalities in the Mexican population are general and predispose different health events in different age groups. Although targeted programs can improve survival and aspects of physical health, they are unlikely to help reduce health inequities.
If socioeconomic status affects the whole existence of the subject, which is reflected in health differences, then measures to reduce disparities should respond to this "global" effect. A central strategy is government regulation in key aspects such as the definition of a salary for a healthy life, regulating hiring and working conditions, and regulating the production and marketing of basic goods such as food, housing and health services.62 Undoubtedly, reducing or eliminating inequalities is a complex issue that will not have easy or immediate solutions. These changes involve substantial changes of government institutions, which are achieved through political mobilization.63 Unfortunately, the government agenda does not recognize socioeconomic disparities in health, so it is to be expected that in Mexico this situation will remain an unattended public health problem.
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.