The nutritional status of the Chilean population has changed rapidly from a high prevalence of malnutrition in the 1970s to almost its total eradication at the end of the 1980s. Malnutrition measured by the Monthly Consolidated Registry, which includes 1.2 million pre-school children attending national health-care centres, decreased from 15.5% to 5.0% between 1975 and 1993. Low birth weight (<2500 g) declined from 11.0% to 4.9% in the same period. In pregnant women and children, a decrease in the proportion of wasting along with a significant increase in the prevalence of overweight and obesity has been observedReference Vio and Albala1. According to the theory of the nutrition transition, these shifts are due to modifications in the diet which in turn have been associated with demographic, economic, social and epidemiological changes. During the 1990s, Chile increased its per capita income twofold, from $US 2600 in 1987 to $US 5000 in 1997; however, an important proportion of this increase has been spent on processed foods, rich in fat, sugar and salt, television sets, appliances and cars. This has led to an increase in energy intake as well as a decrease in physical activity, impacting on obesity ratesReference Vio and Albala2.
During the 1970s, malnutrition was a major problem in Chile, decreasing progressively until the 1980s, when an economic setback increased the nutritional deficit. However, at the same time, the prevalence of obesity in adults, particularly in low-income women, was high. This coexistence of obesity and malnutrition has been one of the underlying factors in transitional periods and occurred in Chile during the 1980sReference Albala, Vio, Kain and Uauy3. In the following years, this situation shifted rapidly to the eradication of malnutrition and a rise in obesity rates. Diets have become higher in fat (especially derived from animal products), refined sugar and salt, and lower in fibre. Presently, Chile is in a classical post-nutritional transition stage.
The current paper describes the changes in the nutritional status of the population in the nutrition transition process, analysing the trends in the prevalence of obesity in pregnant women, children, adults and the elderly until the year 2005, and comparing them with the goals established by the Ministry of Health for the period 2000–2010.
Methods
Design and setting
The trends in the prevalence of obesity presented here are based on secondary analyses of individual datasets from population-based studies and aggregate datasets collected routinely by Chilean national public institutions. The individual datasets analysed here originate from a large study of schoolchildren carried out by Muzzo et al.Reference Muzzo, Burrows, Cordero and Ramirez4, the 2003 National Health Survey5 and the SABE (Health, Well-Being, and Aging) StudyReference Albala, Lebrao, Leon Diaz, Ham-Chande, Hennis and Palloni6.
The school-based studyReference Muzzo, Burrows, Cordero and Ramirez4 is a cross-sectional serial survey conducted in Santiago (where 40% of the Chilean population lives), two regions from the South and two regions from the North, representing the country’s diversity. The first study was carried out in 1987 and included 8536 schoolchildren from primary education; the second study was carried out in 1991 and included 3280 schoolchildren; the third one in 1994 included 244 schoolchildren; and the fourth one in 1998 included 1521 schoolchildren.
The 2003 National Health Survey5 is a cross-sectional study aimed at estimating the prevalence of chronic diseases and nutrition-related risk factors among people 17 years and older. Data were collected between May and December 2003 from a national representative sample of 3428 subjects (1559 men and 1869 women).
The SABE StudyReference Albala, Lebrao, Leon Diaz, Ham-Chande, Hennis and Palloni6 is a multi-centre survey whose goal is to investigate the health and well-being of older people in seven capital cities of Latin America and the Caribbean. In Santiago, Chile, the study included 1301 subjects aged 60 years and older (819 women and 411 men), selected from a random representative sample of older people, of whom 1230 had complete measurements.
Aggregate datasets emanate from the Ministry of Health, the National Board for Child Day-Care Centres (JUNJI) and the National Board for School Assistance and Scholarships (JUNAEB). The Ministry of Health collects aggregate data for all the population attending the national network of health services that covers approximately 70% of the national population aged 0–6 years. JUNJI collects data from approximately 70 000 children of low and middle–low socio-economic status attending day-care centres and JUNAEB collects data on approximately 220 000 children entering first grade (around 70% of the national population in this grade).
Variables
For children under 6 years of age, data for over 1 million children attending national primary health-care centres and 70 000 children attending day-care centres were analysed. At the health centres, weight and height were measured by nurses, twice a year, according to the national programme of children’s health controls. At JUNJI, the measurements were done by pre-school teachers, twice a year (May and November).
The data on weight and height obtained from JUNAEB for about 220 000 children entering first grade (6–8 years) are obtained by classroom teachers in March (beginning of the school year). The nutritional status of pre-school children and schoolchildren entering first grade was analysed utilising the weight-for-height indicator and a cut-off point of >2 standard deviations to define obesity using the National Center for Health Statistics (NCHS) reference7. Prepubertal and pubertal boys and girls were classified as obese when their body mass index (BMI) was ≥95th percentile according to the Centers for Disease Control and Prevention/NCHS 2000 growth charts8. For pregnant women, aggregate data from the Ministry of Health for all Chilean pregnant women attending national primary health-care centres for routine pregnancy health control were collected. The nutritional status was determined using the Rosso-Mardones referenceReference Rosso9, which is based on weight/height. Although this criterion was changed in 2005 to one based on BMIReference Atalah, Castillo, Castro and Aldea10, the data presented here correspond to the previous period (1987–2004).
The nutritional status of adults (≥18 years) and elders was determined using the World Health Organization (WHO) criteria11 for BMI categories: overweight, BMI=25.0–29.9 kg m−2; obese, BMI=30.0–39.9 kg m−2; morbid obese, BMI≥40.0 kg m−2.
Results
Prevalence of childhood obesity
It is well known that obesity is the result of interactions between genetic and environmental factors. Among genetic factors, it is considered that there exists a hereditary predisposition to develop this disease. Several studies in identical twins have shown that despite living separately from each other, their BMI and percentage of body fat run in parallelReference Lobstein, Baur and Uauy12. When both parents are obese, a child has 80% likelihood to be obese, with this risk decreasing to 40% when only one parent is obese. It has been determined that interaction between inherited and environmental factors has led to increased obesity. Since no relevant shifts have been produced in the genetic constitution of the Chilean population in the last decades, it is clear that factors related to lifestyle have been responsible for the obesity phenomenon, particularly an increase in both energy consumption and sedentary behaviour.
Not only there has been a significant rise in childhood obesity, but it is also being observed at earlier ages. As a consequence, complications are affecting a growing number of subjects. Clinical studies have shown that obese children are at higher risk of suffering important physical and psychological problems. Among the first, a greater prevalence of asthma, insulin resistance, type 2 diabetes and hypertension have been observed. Among the psychological problems, which are probably more prevalent, we find poor self-esteem, stigmatisation and isolation, derived from social stereotypes and unfavourable attitudes towards obesity attributing to these children negative characteristics such as laziness, dirtiness and irresponsibilityReference Barlow and Dietz13.
We show here the evolution of obesity prevalence in children based on available epidemiological data. The figures clearly demonstrate a remarkable increase in its prevalence in the last two decades.
Table 1 summarises the evolution in overweight and obesity from 1985 until 2005 in children aged between 0 and 5 years (0–2 years and 2–5 years) attending the national primary health-care centres. In the younger age group, the proportion with overweight has remained relatively stable, between 16% and 17%, while obesity has risen by 52% between 1985 and 1995 without further modifications until 2004, while a small decrease was observed in 2005. In the 2–5 years age group, the situation is similar with respect to overweight; however obesity has risen sharply from 5.0% in 1985 to 8.2% in 2005. We determined (using the test of proportions) if there was a significant difference between the prevalence of obesity in 1985 and 1995 and between 1995 and 2005 in both age categories. In both age groups, there was a significant increase between the obesity prevalence in 1985 and 1995, while between 1995 and 2005 it was significantly higher only in the 2–5-year-olds.
Evolution of overweight and obesity prevalence in the pre-school population under health control (1985–2005)
NA – not available.
* Significant increase in obesity prevalence between 1985 and 1995: P < 0.01.
† Significant increase in obesity prevalence between 1995 and 2005: P < 0.01.
Another important source of data on the nutritional status of pre-school children (2–5-year-olds) comes from the National Board for Day-Care Centres (JUNJI). Anthropometric data for children attending these centres are collected by this institution. Figure 1 shows the evolution in the prevalence of overweight and obesity between 1985 and 2005; overweight has remained stable at about 22% throughout the years, whereas the proportion of obesity has risen from 8.6% to 10.6% between 1995 and 2000, fluctuating between 10.1% and 10.6% in later years. We compared the percentage of obesity using the test of proportions relative to the 1995 value; these results indicate that there was a significant rise in the years 1997 and 2000 compared with the rate in 1995 (P < 0.05).
Prevalence of overweight and obesity in 2–5-year-old pre-school children attending day-care centres (National Board for Day-Care Centres (JUNJI), 1995–2005). *Significant increase in obesity prevalence relative to the 1995 value: P < 0.05
Figure 2 shows how the obesity prevalence has varied according to age in latter years. Two-year-old children have the lowest obesity prevalence, approximately 6%; in 3-year-olds it increases almost twofold while continuing to increase in the 4-year-old group, with a prevalence of 14%. These data confirm the picture shown in Fig. 1 that the percentage of obesity has remained relatively stable since the year 2000; however, a slight increase was observed in 2005 in the two youngest age groups. We tested if there were differences in the obesity prevalence within the same year according to age, using the test of proportions. The results showed that statistical differences exist between the obesity prevalence of 2- and 3-year-olds and also between 3- and 4-year-olds in each of the years (P < 0.01 and P < 0.05, respectively).
Evolution of obesity prevalence according to age in 2–5-year-old pre-school children attending day-care centres (National Board for Day-Care Centres (JUNJI), 1995–2005). *Statistically significant difference between the obesity prevalence of 2- and 3-year-olds: P < 0.01. †Statistically significant difference between the obesity prevalence of 3- and 4-year-olds: P < 0.05
In relation to schoolchildren, since 1987, weight and height data have been collected yearly by the National Board for School Assistance and Scholarships (JUNAEB) for approximately 70% of the schoolchildren entering first grade in the country; this represents approximately 220 000 children. This database has been essential to know the nutritional status of this age group in the last 15 years.
Figure 3 shows the prevalence of obesity in schoolchildren entering first grade between 1987 and 2005. It can clearly be observed that a marked rise in obesity occurred between 1987 and 1997, a period in which the proportion of obesity doubled. From 1997, obesity has risen slowly until 2001–2002, remaining stable thereafter with a small increase observed in 2004 and 2005 (17.3% and 18.5%, respectively).
Prevalence of obesity in schoolchildren attending first grade in public schools (National Board for School Assistance and Scholarships (JUNAEB), 1987–2005)
With respect to obesity prevalence in adolescents, published results from four cross-sectional descriptive studies carried out in various cities in the countryReference Muzzo, Burrows, Cordero and Ramirez4 are summarised in Table 2. Statistical analysis of the trend of variations in obesity was calculated using the χ 2 test for tendency and odds ratios. Obesity prevalence increased significantly for both sexes, from 1.6% to 14.6% (P < 0.0001) for male students and from 2.3% to 17.6% for female students (P < 0.0001).
Obesity prevalence (%) in adolescentsFootnote * from various Chilean cities (1986–1998)
* Number of subjects – 2885 males aged 12–16 years and 3586 females aged 10–16 years in 1986; 497 males aged 12–16 years and 916 females aged 10–16 years in 1991; 478 males aged 12–16 years and 709 females aged 10–16 years in 1994; 343 males aged 12–16 years and 465 females aged 10–16 years in 1998.
† Significant increase in obesity prevalence between 1986 and 1998: P < 0.0001.
Obesity prevalence in pregnant women
In relation to pregnant women, the Ministry of Health issued a statistical series in 1987 when the Rosso-Mardones referenceReference Rosso9, which considers weight and height by gestational age, was first adopted. This reference categorises women attending prenatal check-ups in health-care centres into four groups: low weight, normal, overweight and obese. In this epidemiological series, a significant decrease in low-weight pregnant women was observed along with an important increase in obesity (from 12% in 1987 to 34% in 2002) with a slight decrease, to 32.6%, in the year 2004 (Fig. 4). In 2005, the Ministry of Health changed to a reference based on BMI by gestational ageReference Atalah, Castillo, Castro and Aldea10. With this new reference, 20.3% of pregnant women were classified as obese in 2005.
Obesity prevalence in pregnant women (Ministry of Health, Chile, 1987–2004). In 2005, the Ministry of Health reference based on weight/height was changed to one based on body mass index; this resulted in an obesity prevalence of 20.3%
Obesity prevalence in adults
For adults, no follow-up system to assess nutritional status is available, as is the case for children and pregnant women. There are only cross-sectional studies including representative samples from Santiago in 1987 and 1992Reference Berríos, Jadue, Zenteno, Ross and Rodríguez14–Reference Berríos16, where obesity prevalence increased from 6% in 1987 to 11% in 1992 in men and from 16% in 1987 to 24% in 1992 in women. A study carried out in Valparaiso in 1997 found a prevalence of 14% for men and 23% for womenReference Jadue, Vega, Escobar, Delgado, Garrido and Lastra17. Only recently, with the 2003 National Health Survey5, are representative data available for the country by region and by urban–rural categories, which evidence a prevalence of 22% for obesity (BMI>30 kg m−2), 38% for overweight and 1.3% for morbid obesity (BMI > 40 kg m−2). The highest prevalences of obesity and morbid obesity are observed in women aged 45 to 64 years, as depicted in Table 3. Obesity prevalence varies according to socio-economic level, measured in this case by education as a proxy indicator, showing that subjects with low educational level have a twofold higher prevalence than those with a high educational level (Fig. 5).
Prevalence of overweight, obesity and morbid obesityFootnote * in Chilean adults by gender
Values are expressed as % (95% confidence interval).
* Overweight – body mass index (BMI)=25.0–29.9 kg m−2; obesity – BMI=30.0–39.9 kg m−2; morbid obesity – BMI ≥ 40.0 kg m−2.
Obesity prevalence in adults by educational level (National Health Survey 2003)
Obesity prevalence in older people
The SABE Study has provided data on BMI in the elderly from Santiago and other five cities in Latin America and the CaribbeanReference Albala, Lebrao, Leon Diaz, Ham-Chande, Hennis and Palloni6. In Santiago, a high prevalence of obesity in both men and women was found, higher in women and in the younger group. The highest prevalence of obesity was found in the group aged 60–64 years for both men (35.6%) and women (44.1%), decreasing to 18.3% and 25.7%, respectively, in the group aged 75 years and older. On the other hand, a very low prevalence of underweight was observed, increasing with age (Table 4).
Nutritional status of elderly Chileans by gender and age: SABE Study 2002
WHO – World Health Organization; BMI – body mass index.
Values are expressed as % (95% confidence interval).
Discussion
On 21 November 2000, the National Board for Health Promotion (VIDA Chile) presented a six-year goal strategy (2000–2006) to the Chilean President18. These goals mandated a reduction in obesity prevalence among 2–5-year-old pre-school children attending day-care centres, from 10% to 7%; in 6–8-year-old schoolchildren admitted in first grade, from 16% to 12%; and in pregnant women from 32% to 28%. Thereafter, during the discussion of the national health reform, these goals for obesity were established as the Sanitary Goals for the decade (2000–2010)19.
Five years later, at the end of 2005, obesity rates have not changed; in pre-school children this figure has remained stable at about 10.6%; in schoolchildren in first grade, it has increased to 18.5%; and in pregnant women the figure in 2004 was 32.6%. In children, the trend in obesity is similar to that in the UK, where the Government set a goal to stabilise childhood obesity prevalence at the level of 10%, but instead it increased from 9.9% in 1995 to 13.7% in 2003Reference Cole20.
In the last five years VIDA Chile has continued to take actions and set clear regulations at central level in order to tackle obesity throughout the life cycleReference Salinas and Vio21. Educational materials on food and nutrition have been developed in order to be included in primary-school programmesReference Olivares, Moron, Kain, Zacarias, Andrade and Lera22; however they have not been implemented yet by the Ministry of Education. Guidelines for Active Living23 have been developed with clear messages to promote physical activity for all population groupsReference Salinas and Vio24; but they have had little dissemination and application. Food education for the population has continued through media messages and delivery of booklets given to customers at supermarketsReference Domper, Zacarías, Olivares, González and Vio25. Food and Dietary Guidelines issued in 1997Reference Castillo, Uauy and Atalah26 have been reviewed to include counselling on physical activity and tobacco, constituting Guidelines for Healthy Living27. Nevertheless, these efforts have been insufficient for decreasing the obesity prevalence in the country. In general, the population knows both what healthy eating means (as shown in different national surveys) and the benefits of engaging in physical activity, but there is persistence in eating foods high in fat, sugar and salt and also in leading inactive lives. Behaviour changes are difficult to attain and require strong political decision-making in order to set a national health priority involving the academic sector, the media, and public and private sectors.
Policy options
The question that arises is: Why have the goals set for obesity not been accomplished?
There are at least three answers. The first one is that the policies implemented have not been intense enough considering the magnitude of the problem, and that the application of a more efficient State Policy is required. The second possibility involves analysing the situation that would have occurred if any policy were implemented: in this case, the probable answer is that obesity would have continued to increase. The third answer is epidemiological: after a rapid rise in obesity in pre-school children (Table 1), school-age children (Fig. 3), adolescents (Table 2) and pregnant women (Fig. 4) between 1987 and 2000, a stabilisation period or a ‘plateau’ was observed. In that situation, very effective interventions are required to reduce obesity, because it is extremely difficult to reverse the trend.
Under this perspective, there have been effective policy options. For example, some interventions implemented in Chile have been effective in decreasing obesity rates in schoolchildren. A three-year intervention in public schools in a small city, Casablanca – which included educational activities in food and nutrition with materials developed by the Institute of Nutrition and Food Technology (INTA) and the Food and Agriculture Organization of the United NationsReference Olivares, Moron, Kain, Zacarias, Andrade and Lera22, and increased physical education time as well as improved quality of utilisation of this extra time – proved to be very successful. With this intervention, obesity decreased by 50% at the end of the second year, and remained stable thereafterReference Kain, Vio, Leyton, Cerda, Olivares and Uauy28. This intervention, which was funded by a private company, is an example of a partnership between a university, the private sector and government to tackle the obesity epidemic. Presently, it is being replicated in one of Santiago’s counties and, in addition, a new initiative based on the Casablanca experience and funded by private companies has been launched nationwide, based on the social responsibility of some enterprises. In addition, the Ministry of Health is implementing the Global Strategy against Obesity (EGO-Chile)29, 30, based on the strategy recommended by the WHO and Pan American Health Organization31.
State policy in health promotion to prevent obesity
To accomplish change in the nutritional situation of the country, a global priority policy in health promotion to prevent obesity is needed. Presently, the priorities of government ministries are focused neither in health promotion nor in obesity prevention. The Ministry of Health is facing the urgent task to carry out health reform that includes the General Guarantees in Health (GGH) law, also called ‘Plan AUGE’ (Universal Access with Explicit Guarantees), based on a curative plan that initially tackles 40 specific diseasesReference Letelier and Bedregal32. At the Ministry of Education, the most important goal is to improve the equity and the quality of education, measured by periodic tests that evaluate mathematics and linguistics skills. In the other ministries and governmental institutions, the situation is similar, with no integrated policy to cope with obesity and health promotion in order to improve the quality of life of the Chilean people.
A comprehensive State Policy, that includes the public and private sectors related with obesity, is needed. This policy, which requires special funding, should include: regulations for food advertisement and food marketing of unhealthy foods, an increase in physical education classes from two to at least four hours per week, and the provision of accessible places to engage in physical activity and recreation in the cities. A strong marketing campaign through the mass media (particularly television) is needed, informing parents of the risks involved in childhood obesity, on the foods that are unhealthy and of the benefits of physical activity. An example of this type of marketing is the ‘5-a-day’ campaign aimed to increase the consumption of fruits and vegetables in the population. Political decisions introducing initiatives to curb the sale of unhealthy foods should impact on changing long-established habits in the population. The healthy choices represent new challenges and also new opportunities and rules for the private sector. A successful association between public and private sectors requires not only sharing of goals but also sharing of ethics.
Conclusion
In conclusion, no effective actions to tackle the obesity epidemic will be possible without a clear priority strategy to prevent obesity at the national level, establishing synergy and alliances between the different actors: the scientific community, governmental and non-governmental organisations, legislators, health professionals, food industry and the media. Moreover, effective actions require local compromise at the pre-school and school levels in the counties; national commitment of the Government; compromises from the private sector; and Congress to enact laws and regulations against food marketing of unhealthy foods, while promoting physical activity.
Government and non-governmental organisations should play a role in promoting and protecting an environment that supports the growth and development of infants and children, monitoring food marketing, and facilitating community-based initiativesReference Caballero33. The successful Casablanca project, promoting healthy eating and physical activity in children and their familiesReference Kain, Vio, Leyton, Cerda, Olivares and Uauy28 that is being replicated currently in seven public primary schools in Santiago, should be an example to be followed in the near future by other public schools in the country.
Acknowledgements
Sources of funding: The work was funded by INTA, University of Chile.
Conflict of interest declaration: None.
Authorship responsibilities: F.V. was involved in the conception and design of the work, analysis and interpretation of the data, writing the article, and giving final approval of the version to be published. C.A. was involved in the conception and design of the work, analysis and interpretation of the data (in particular the data related to ageing), revising the article critically for important intellectual content, and giving final approval of the version to be published. J.K. was involved in the analysis and interpretation of the data (in particular the data related to infancy and children, including adolescence), and revising the article critically for the final approval of the version to be published.
