Consumo de frutas e hortaliças por idosos de baixa renda na cidade de São Paulo Fruit and vegetable intake among low income elderly in the city of São Paulo , Southeastern Brazil

MÉTODOS: Estudo transversal de base populacional com 2.066 idosos (≥60 anos) de baixa renda residentes na cidade de São Paulo, SP, em 2003-2005. Para a avaliação do consumo de frutas e hortaliças foi aplicado questionário de freqüência alimentar. As respostas foram transformadas em consumo diário e comparadas às recomendações da Organização Mundial da Saúde (consumo de cinco ou mais porções diárias). A relação entre consumo recomendado de frutas e hortaliças e variáveis socioeconômicas foi avaliada mediante modelos de regressão logística.


INTRODUCTION
The fast ageing of the Brazilian population indicates the current figure of approximately 15 million elderly individuals.a It is estimated that in 20 years, this figure will more than double, placing Brazil among the five oldest populations in the world. 1,11,12This population ageing will result in an increase in the prevalence of non-communicable chronic diseases, which affect predominantly the elderly. 1 In 2004 the World Health Organization (WHO) b proposed recommendations based on lifestyle changes aiming to prevent and minimize the worldwide a Instituto Brasileiro de Geografia e Estatística.Censo 2000: Brasil.Brasília; 2000 [cited 2007 Jan 01].Available from: http://www.ibge.gov.br/censo2000b World Health Organization.Global Strategy on diet, physical activity and health.Fifty seventy word health assembly.Geneva; 2004[cited em 2007 nov 01].Available from: http://www.who.int/dietphysicalactivity/en/prevalence of non-communicable chronic diseases.One of its main recommendations was the daily intake of five or more servings of fruits and vegetables.The increased intake of these foods would play an important role in reducing the risk for the main chronic diseases, especially due to the greater offer of vitamins, antioxidant minerals, and dietary fibers. 3,20uit and vegetable intake is partly determined by the socioeconomic conditions of the population.In two recent studies, low family income 7 proved to be independently associated with inadequate dietary practices, especially with low daily intake of fruits and vegetables. 9,12Other factors have also been associated with the Brazilian elderly individuals' low fruit and vegetable intake, such as: low schooling, inappetence, difficulties in acquiring and preparing the food and presence of chronic diseases. 2,6w Brazilian studies have investigated the dietary patterns of elderly individuals, and research focusing on fruit and vegetable intake is even scarcer.A populationbased study which used a food frequency questionnaire with 283 elderly people from three regions of the municipality of São Paulo showed that the fruit and vegetable intake increased as the socioeconomic level of the regions increased. 14However, the study did not examine the proportion of elderly individuals who met the recommendations of five daily servings of fruits and vegetables.In a Brazilian inquiry that evaluated a national probabilistic sample of more than 5,000 men and women aged 18 years or older, it was found that among the individuals aged 65 or older, only 20.6% of the women and 14.8% of the men consumed five or more servings of fruits and vegetables per day. 9evertheless, such information was obtained only in relation to the frequency of the intake, and it was not possible to identify which fruits and vegetables were more or less consumed. 8e present study aimed to estimate the socioeconomic and sociodemographic factors associated with the intake of five servings of fruits and vegetables, recommended by WHO, by elderly people living in low income areas, identifying the main fruits and vegetables that compose the diet of this population.

METHODS
The present study is part of the population-based epidemiological cohort "São Paulo Ageing and Health study" (SPAH), [15][16][17] which studied the factors associated with the prevalence and incidence of dementia and other mental disorders in low income elderly of the municipality of São Paulo between 2003-2005.In the first phase of SPAH, a cross-sectional study was conducted a World Health Organization.Global Strategy on diet, physical activity and health.Fifty-seventh world health assembly.Geneva; 2004[cited 2007 Nov 01].Available from: http://www.who.int/dietphysicalactivity/en/ with elderly individuals living in low income areas of the western region of the municipality of São Paulo. 15ll individuals aged 65 years or older who lived in the districts of Butantã, Rio Pequeno and Raposo Tavares were eligible.These districts included the poorest census tracts, with slums and/or areas covered by the Programa Saúde da Família (Family Health Program), representing the regions with the lowest Human Development Index (HDI) of the study's area.The elderly were interviewed and evaluated at their homes, preferably in one single visit. 9Between the years 2003-2005, 2,072 participants were included, and 2,066 answered the food frequency questionnaire about fruit and vegetable intake, constituting the sample of the present study.
6][17] For 89 elderly individuals who presented serious physical or mental disability, close informants answered the interview.Details on the evaluation of physical or mental disability were already published. 15 assess the fruit and vegetable intake, the corresponding section of a Food Frequency Questionnaire (FFQ) was used.This semi-quantitative questionnaire was developed to obtain information on the diet of the general adult population residing in the Metropolitan Region of São Paulo. 7 the FFQ there are nine possible consumption responses for each food item in the list: never or less than once a month, one to three times a month, once a week, two to four times a week, five to six times a week, once a day, one to three times a day, four to five times a day, and six or more times a day.These categories of responses are based on standardized servings for each food item in the list.For example, the standard serving for bananas is one unit; if the participant reports that he consumed an average of three bananas per day, he will be classified in the category "one to three times a day".In the present study, we administered the part of the FFQ that represented the groups of "vegetables", with ten items, and "fruits and natural juices", with 17 food items.
To calculate how many elderly individuals met the WHO recommendations a for fruit and vegetable intake, the responses of each individual that indicated the daily intake of each food item were selected, corresponding to the following frequency categories: one, two to three, four to five and six or more times a day.The sum of all the FFQ responses referring to each food item was calculated and thus, the number of daily servings of fruits, vegetables, and fruits and vegetables (combined) per individual were obtained.Therefore, the total daily fruit and vegetable intake resulted in a continuous variable, equivalent to the total number of daily servings of fruits and vegetables (combined) consumed by the elderly.This final continuous variable was categorized as: "non-daily fruit and vegetable intake", "daily fruit and vegetable intake", and "recommended fruit and vegetable intake" (five or more servings/day).
The statistical analyses were carried out with the software Stata 9.0.First, the socioeconomic and demographic characteristics were analyzed in a descriptive way, using measures of central tendency and percentage distribution.The analyzed characteristics included sex, age group (65-69, 70-74, 75-79, 80 years of age or older), years of residence in the city of São Paulo, schooling (literacy and years at school) and per capita monthly income (in minimum salaries at the time of the study).
The intensity of the associations between socioeconomic and sociodemographic characteristics and adequate daily fruit and vegetable intake was estimated by odds ratios (OR), with their respective 95% confidence intervals using logistic regression models adjusted for age and sex.Statistical significance was evaluated through Wald tests and for ordered categorical variables the test for linear trend was used.Then, multivariate logistic regression models were used to identify the independent associations.Thus, if one or more variables were associated with p> 0.15, the one that had the highest p value was removed from the model.A new model was estimated and the significance of the variables was examined again.This iterative process continued until only the variables with a value of p< 0.15 remained in the model.The effect of potential confounding variables was examined by including each variable sequentially into the multivariate logistic regression models and observing if the estimated ORs varied by more than 10%. 13e study was approved by the Ethics Committee for the Analysis of Research Projects of the Board of Clinical Directors of the Hospital das Clínicas da Faculdade de Medicina da USP (Clinical Hospital of the School of Medicine of USP) (Process no.0361/07).The participants signed a consent document and the consent of the participants with cognitive deficit was provided by the informants.

RESULTS
Table 1 presents the participants' sociodemographic characteristics.Out of 2,066 participants, the majority (1,250; 60.5%) were women, and 370 (45.3%) men and 518 (41.4%) women were in the age group 65 -69.Only 97 (11.9%) men and 99 (7.9%) women had more than four years of schooling.Median per capita monthly income was R$ 346.67 and 8% reported not having any individual source of income.Four hundred and fifty two (55.4%)men and 585 (47.2%) women had been living in São Paulo for at least 40 years.
Table 2 shows the consumption of different fruits and vegetables according to the response categories of the FFQ.
The elderly who reported consuming five or more daily servings of fruits and vegetables were 19.8% (CI 95%: 18.1; 21.5).Another 45.2% presented daily fruit and vegetable intake but they did not reach the recommendations, and 35% did not consume fruits and vegetables on a daily basis.The eight most consumed fruits and vegetables were: tomato, lettuce, carrot, banana, orange/tangerine, apple/pear (Table 2), and they represented 82% of the fruit and vegetable intake of the participants who met the daily recommendations of the WHO. a Table 3 shows the crude and adjusted associations between the socioeconomic and sociodemographic variables and the recommended fruit and vegetable intake.In the multivariate analysis, the recommended fruit and vegetable intake was independently associated with level of schooling and per capita income.The recommended fruit and vegetable intake increased significantly according to years of schooling.The association between the recommended fruit and vegetable

DISCUSSION
In the present study, one fifth of the participants achieved the WHO a recommendations regarding daily servings of fruits and vegetables, while 35% did not consume any type of fruit or vegetable daily.The fruit and vegetable intake increased significantly as the level of schooling and per capita income increased.However, although there is a great offer and diversity of fruits and vegetables in Brazil, we found a limited variability of consumption among the people who reached the WHO recommendations.
Studies conducted in Brazil have shown that sex, purchasing power and schooling/information are determinant for a balanced diet in these populations. 2,6,9,10,14aime & Monteiro 9 (2005) investigated the fruit and vegetable intake in a national probabilistic sample composed of 5,000 individuals, including elderly people.These authors used short questions to obtain information on the intake of these food items and their results were similar to the ones of the present study, with 20.6% of the women meeting the WHO recommendations, but only 14.8% of the men. 9The elderly were the ones who most consumed fruits and vegetables according to the WHO recommendations and a clear positive association was observed between the level of schooling and fruit and vegetable intake. 9A recent study with a sample of adults from the municipality of São Paulo, which examined fruit and vegetable intake using short questions in telephone interviews, showed a significant association between level of schooling and intake of these food items for both sexes. 6 the present study, inadequate fruit and vegetable intake was strongly associated with the elderly individuals' low level of schooling and low income ranges.According to Jaime et al 8 (2007), the estimates are that the fruit and vegetable intake in Brazil corresponds to less than half of the nutritional recommendations, mainly in low income families.A study carried out with data from Pesquisa de Orçamento Familiar (POF -Family Budget Research) of Fundação Instituto de Pesquisas Econômicas of Universidade de São Paulo in 1998/1999 about the acquisitions of food and beverages for family consumption during one month showed that the participation of fruits and vegetables in the overall amount  purchase of vegetables. 10Low level of schooling can also influence consumption and the small variability of fruits and vegetables, due to the adoption of inadequate dietary habits and due to the higher frequency of agerelated problems (locomotion difficulties, chewing and swallowing problems, diseases such as depression and dementia). 2,4,14,16The effect of low level of schooling could explain, at least in part, the small variability in fruit intake observed here.We found that among the least consumed fruits were those whose cost is high in the state of São Paulo, b like strawberry, fig, peach and of calories increases as the family income increases and the price of the fruits and vegetables decreases. 4e analysis of data about the acquisition of food items obtained from POF 2002/2003 a showed that the families with monthly income up to R$ 400.00 allocated only 3.6% of the expenditures on food for the acquisition of fruits and 3.3% for the acquisition of vegetables. 10On the other hand, the families with monthly income higher than R$ 4,000.00allocated 23.5% of the expenditures on food for the acquisition of fruits and 12.7% for the grapes.However, some fruits with lower cost, such as lime, avocado, watermelon and melon, were little consumed by the elderly.In the case of vegetables, we observed a similar pattern, with some vegetables whose cost is low and which can be easily purchased, like cabbage, chayote, cucumber and chard, being little consumed by the study's participants.
9][20] Our study used FFQ to assess the fruit and vegetable intake and the results were consistent with those obtained by Brazilian investigations that used short questions. 6,9herefore, we believe that the use of the FFQ in the present work did not influence the assessment of the level of fruit and vegetable intake and allowed a more accurate identification of the consumed food items.A positive aspect of the present study is the fact that, for the 89 participants who could not answer the questionnaires due to serious mental disability, the information on fruit and vegetable intake was provided by their informants, increasing the sample's representativeness regarding the participants' health condition.
The results of the present work should be generalized with some caution to other populations of elderly people in Brazil, as there are areas where the access to this kind of food is more restricted and the per capita income is lower.Nevertheless, we believe that elderly individuals in other low income areas of other urban centers in Brazil do not present a better situation than the one observed here in relation to daily fruit and vegetable intake.
To conclude, the fruit and vegetable intake by low income elderly individuals was well below the current WHO recommendations.a This inadequacy was associated with unfavorable socioeconomic conditions, low level of schooling and low purchasing power, even in a relatively homogeneous population group.The most consumed fruits and vegetables on a daily basis are in accordance with the most consumed food items by the

Table 2 .
Frequency of fruit and vegetable intake of the sample.Municipality of São Paulo, Southeastern Brazil, 2003-2005.
a Low content of carbohydrates b Medium content of carbohydrates

Table 3 .
Crude and adjusted analyses of the associations between recommended fruit and vegetable intake (≥ 5 daily servings) and socioeconomic and sociodemographic variables.Municipality of São Paulo, Southeastern Brazil, 2003-2005.(N=2.066)Wald test population of the state of São Paulo, but the variety in the consumption of these food items is limited.Public policies and health promotion programs must incorporate forms of increasing the consumption of these food items by the elderly population in Brazil.
a Test for linear trend b