SciELO - Scientific Electronic Library Online

vol.21 suppl.2Oral health profile among community-dwellingelderly and its association with self-rated oral healthDepressive symptoms among the elderly in São Paulo city, Brazil: prevalence and associated factors (SABE Study) author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand




Related links


Revista Brasileira de Epidemiologia

Print version ISSN 1415-790XOn-line version ISSN 1980-5497

Rev. bras. epidemiol. vol.21  supl.2 São Paulo  2018  Epub Feb 04, 2019 


Nutrition status of elderly smokers and former smokers of São Paulo City, Brazil

Isis Bonfitto GonçalvesI 

Maria Lúcia LebrãoI  *

Yeda Aparecida de Oliveira DuarteI 

Gabriela Arantes WagnerII 

Dirce Maria Trevisan ZanettaI 

IDepartment of Epidemiology, School of Public Health, Universidade de São Paulo - São Paulo (SP), Brazil.

IIDepartment of Preventive Medicine, Universidade Federal de São Paulo - São Paulo (SP), Brazil.



The concomitance of smoking and poor dietary habits represents a worsened prognosis of health and quality of life for elderly. The aim of this study was to characterize the nutritional status of elderly who were smokers and former smokers and residents of São Paulo city.


A cross-sectional study was conducted in 2010 with a representative sample of 1,345 individuals aged 60 years and over, who were part of the elderly cohort monitored at the SABE Study. Sociodemographic, health, and nutritional aspects of the elderly were described, according to their tobacco use in life.


The proportion of smokers and former smokers was 12.9 and 54.7%, 11.0 and 25.2%, and 11.8 and 37.2% for male, female, and total population, respectively. For both genders, increasing age decreased the proportion of smokers. The proportion of proper fruit intake was smaller for female smokers. Poorer nutritional status was observed in smokers, who had fewer meals per day and greater frequency of underweight compared with elderly nonsmokers.


Considering the impact of inappropriate eating habits and smoking on health, elderly smokers deserve special attention on their nutritional status.

Keywords: Nutritional status; Feeding behavior; Aged; Tobacco; Brazil; Epidemiology


Population aging is an established worldwide phenomenon, which raises concerns related to health, as it is accompanied by an increase in the occurrence of chronic and disabling diseases1.

Some aspects of the current lifestyle, such as smoking, excessive alcohol consumption, sedentary lifestyle, high intake of foods rich in sugars and fats, and low consumption of fruits and vegetables, increase the risks to health2,3. Smoking stands out among the habits that offer higher risks to health because it is the leading cause of preventable deaths worldwide, especially those related to cardiovascular diseases and cancer4.The smoking effects are not studied much in the elderly population5; however, it is assumed that this population presents higher risks due to the longer exposure6.

Literature points to a higher prevalence of smoking among the underweight elderly, which raises concerns because the concurrence of two or more risk behaviors may be more harmful than the sum of their separate effects6.7. This study examined smoking and its association with nutritional status of elderly residents in the city of São Paulo.



This is a cross-sectional study with data from the SABE Study (Health, Well-being, and Aging), collected in 2010. SABE study is composed of multiple cohorts and was initiated in 2000 with a random sample of elderly patients (aged ≥ 60 years) living in São Paulo, SP. Itemployed a stratified sampling in two stages, based on census tracts in the city. Details on the sample design of the initial study are described in a previous publication8.In 2006, the second interview was conducted with those elderly individuals, and a new sample of elderly individuals who were aged between 60 and 64 years was obtained by means of the same procedure of the first data collection point, as this age group was no longer represented. The third follow-up visit was conducted in 2010, when a third sample of elderly individuals who were aged between 60 and 64 years was added. The total sample of 1,345 elderly individuals was evaluated that year (aged ≥ 60 years), representing a population of 1,338,138 elderly individuals living in São Paulo, as the sample weights were recalculated based on the 2010 Census to maintain a representative sample of the elderly population living in the city.


Data were collected by means of home interviews conducted by trained interviewers. Sociodemographic variables evaluated were gender (male, female); age (60 - 64; 65 - 69; 70- 74; ≥ 75 years); schooling in complete years of study (< 1 - 3 years, 4 - 7 years, ≥ 8 years of schooling); income in minimum wages, which were categorized into quartiles; ethnicity (white, others); and family arrangement (living alone, living together). The characteristics of the health were evaluated with the following variables: number of self-reported chronic diseases (none, 1, ≥ 2); frequency of alcohol consumption in the last 3 months (< 1 day/week, 1 - 3 days/week; ≥ 4 days/week), and self-rated health status (good, not so good), and tobacco use in life, which was obtained through self-report to the following question: “Doyou currently smoke or have you ever smoked?”(never smoked, ex-smokers, and smokers).

The assessment of nutritional status was determined by the body mass index (BMI), and it was classified according to the recommendation from the Pan American Health Organization9 as underweight (BMI ≤ 23 kg/m2), normal weight (23 kg/m2< BMI <28kg/m2), and overweight and obese (BMI ≥ 28 kg/m2) elderly.

Nutritional variables were number of full meals per day (< 3 meals; ≥ 3 meals), consumption of dairy products at least once a day (yes, no); consumption of eggs and beans at least once a week (yes, no); meat consumption at least three times a week (yes, no); consumption of fruits at least twice a day (yes, no); and consumption of vegetables at least twice a day (yes, no). Other features related to nutritional status were weight loss in the last 12months (none, 1 to 3 kg; > 3 kg), waist-hip ratio (< 0.9, ≥ 0.9 for men and < 0.85; ≥0.85 for women), and nutritional self-perception (good, not so good).


For the descriptive analyses, ratios for the categorical variables were calculated. Allanalyzes incorporated weights to correct differences in the selection probability of the participants and the results were presented by the weighted values. Differences between the groups were estimated by the Rao-Scott test for the categorical variables, which considered sampling weights to the population estimates with population weights. Survey package of the STATA software was used, which provides procedures for analyzing complex sample surveys and allows incorporating different weights of the observations that influence the point estimates of the total population parameters.

The study was approved by the Human Research Ethics Committee of the School of Public Health of the Universidade de São Paulo. Participation was voluntary and informed consent forms signed by all participants were obtained. The authors report no conflict of interest in relation to this study.


The behavior in relation to smoking differed among genders (Table 1), showing a higher proportion of women who had never smoked (63.8%). Although approximately 70% of men had been smokers, 54.7% reported having stopped smoking. The proportion of current smokers was similar in both the genders. The highest proportion of smoking occurred among those aged between 60 and 64 years in both the genders, with a significant decrease among those aged above 70 years. The frequency of smokers did not differ between white men and other ethnicities, but there was a higher proportion of nonsmokers among white women in relation to the other ethnicities. Higher proportion of smokers was seen among women living alone than those living together. There was no difference in the use of tobacco among the different levels of schooling evaluated. With regard to the income, there was a predominance of male smokers in the lower strata, whereas those with higher incomes showed the highest proportion of nonsmokers (Table 2).

Table 1. Tobacco use in life by gender in 2010; SABE study (Health, Well-being, and Aging). 

Men Women Total
% (95%CI) % (95%CI) % (95%CI)
Nonsmoker 32.4 (27.8 - 37.5) 63.8 (59.5 - 67.9) 51.1 (47.5 - 54.6)
Ex-smoker 54.7 (49.6 - 59.6) 25.2 (22.1 - 28.6) 37.2 (34.4 - 40.0)
Smoker 12.9 (9.7 - 16.9) 11.0 (8.7 - 13.8) 11.8 (9.7 - 14.3)

χ2 with Rao-Scott adjustment: p < 0.001.

Table 2. Distribution of sociodemographic characteristics according to tobacco use in life by gender in 2010; SABE study (Health, Well-being, and Aging).  

Men Women Total
¥ NS % ES % S % p-value* NS % ES % S % p-value NS % ES % S % p-value
Tobacco 32.4 54.7 12.9 < 0.001 63.8 25.2 11.0 < 0.001 51.1 37.2 11.8 < 0.001
Age (years)
60 - 64 29.3 51.0 19.7 0.027 53.5 31.5 15.0 < 0.001 42.9 40.1 17.1 < 0.001
65 - 69 41.8 48.5 9.8 56.0 29.4 14.6 50.2 37.2 12.6
70 - 74 36.1 53.7 10.3 76.2 17.2 6.7 60.0 31.9 8.1
≥ 75 26.3 66.8 6.8 73.7 19.9 6.4 56.6 36.9 6.6
White 34.1 52.6 13.3 0.622 68.5 21.1 10.4 0.009 54.7 33.7 11.6 0.013
Others 30.2 57.5 12.4 57.3 31.0 11.8 45.9 42.1 12.0
Schooling (grade)
Up to 3rd 26.5 61.2 12.3 0.295 62.6 26.7 10.8 0.531 50.0 38.7 11.3 0.750
4th - 7th 29.3 56.8 13.9 65.6 25.2 9.2 50.6 38.2 11.2
> 8th 40.8 46.8 12.5 63.1 22.9 13.9 52.7 34.1 13.3
Family arrangement
Living alone 27.7 55.1 17.2 0.711 59.2 23.6 17.3 0.032 50.6 32.1 17.3 0.074
Living together 33.0 54.6 12.4 64.9 25.6 9.6 51.1 38.1 10.8
Monthly income (quartile)
1st 15.7 61.6 22.7 0.005 68.4 22.2 9.4 0.583 56.0 31.5 12.5 0.195
2nd 29.7 52.8 17.6 58.3 28.8 12.9 48.3 37.2 14.6
3rd 26.7 67.1 6.2 65.7 23.5 10.9 47.2 44.2 8.7
4th 41.5 48.4 10.2 63.1 23.3 13.6 49.9 38.6 11.5

Data were weighted to be representative of the elderly population in 2010 in São Paulo, Brazil; *χ2 with Rao-Scott adjustment; ¥NS: nonsmokers, ES: ex-smokers, S: current smoker.

Regarding nutritional status, 47.0% of men presented normal weight, 13.3% were underweight, and 39.7% were overweight or obese whereas among women, the proportions were 34.4, 12.4, and 53.4%, respectively. The prevalence of overweight and obesity was higher among women than among men (p < 0.001).

Male smokers were predominantly underweight, whereas nonsmokers were overweight or obese (Table 2). The same pattern was observed among the women smokers. Amongthe male smokers, there was higher proportion of individuals who presented high alcohol consumption and reported having had less than three meals a day, with a significant difference seen among women. Moreover, a good self-perception of health predominated among these women. Individuals who reported two or more chronic diseases presented lower proportion of smokers, which was significant among women (Table 3).

Table 3. Distribution of nutritional characteristics and health according to tobacco use in life by gender in 2010; SABE study (Health, Well-being, and Aging).  

Men Women Total
¥ NS % ES % S % p-value* NS % ES % S % p-value* NS % ES % S % p-value*
Classification by BMI&
Underweight 28.9 49.2 21.9 0.360 50.1 22.3 27.6 < 0.001 41.3 33.5 25.2 < 0.001
Normal weight 32.3 52.9 14.8 60.1 26.4 13.5 46.8 39.0 14.1
Overweight/obese 32.7 57.7 9.7 68.8 25.6 5.6 56.8 36.2 7.0
Weight loss in the past year
None 17.4 56.7 25.9 0.076 64.9 22.4 12.7 0.880 51.3 32.2 16.5 0.213
1 - 3 kg 27.9 69.6 2.5 63.5 29.0 7.5 50.4 44.0 5.7
≥ 3 kg 33.8 53.7 12.5 63.7 25.3 11.0 51.1 37.3 11.6
Number of full meals per day
< 3 20.5 60.1 19.3 0.161 46.8 29.0 24.2 0.002 33.8 44.4 21.8 0.002
≥ 3 34.2 53.8 12.0 65.5 24.8 9.7 53.1 36.3 10.6
Self-perception of nutrition
Good 32.9 54.1 13.0 0.083 65.0 24.5 10.5 0.083 51.7 36.8 11.5 0.295
Not so good 23.4 65.6 11.1 50.2 33.0 16.7 42.3 42.6 15.1
Self-perception of health
Good 27.7 55.1 17.2 0.711 59.2 23.6 17.3 0.032 50.6 32.1 17.3 0.074
Not so good 33.0 54.6 12.4 64.9 25.6 9.6 51.1 38.1 10.8
Waist-hip ratio
Above 38.9 46.7 14.4 0.534 60.6 26.2 13.3 0.412 55.3 31.1 13.6 0.099
Below 30.6 55.9 13.5 64.9 25.1 10.0 49.5 39.0 11.6
Consumption of alcoholic beverages (days/week)
< 1 32.3 56.4 11.4 0.427 64.4 25.7 10.0 0.059 53.9 35.7 10.4 0.022
1 - 3 27.4 56.8 15.9 58.4 23.5 18.0 38.8 44.6 16.7
> 4 39.5 45.2 15.4 59.7 11.8 28.5 42.9 39.5 17.6
Chronic diseases
None 36.4 48.9 14.8 0.285 68.3 19.8 12.0 0.039 52.3 34.3 13.4 0.021
1 33.4 52.5 14.1 64.4 22.6 13.0 53.0 33.6 13.4
≥ 2 28.2 61.6 10.2 60.9 30.7 8.4 48.3 42.6 9.1

Data were weighted to be representative of the elderly population in 2010 in São Paulo, Brazil; *χ2 with Rao-Scott adjustment; ¥NS: nonsmokers, ES: ex-smokers, S: current smoker.

There was no significant difference in the consumption of dairy products (p = 0.737 for men and p = 0.188 for women), meat (p = 0.321 and p = 0.341), eggs and beans (p = 0.320 and p = 0.493), and vegetables (0.328 and 0.870), considering the use of tobacco in life as well as in fruit consumption among men (p = 0.664). The proportion of inadequate consumption of fruits (p = 0.005) was higher among female smokers (Figure 1).

NS: nonsmoker, ES: ex-smoker, S: current smoker.

Figure 1. Proportion of food intake according to gender and tobacco use in life, in 2010; SABE study (Health, Well-being, and Aging). 


The frequency of smoking in this study was 12%. Although approximately 70% of the elderly men have smoked at least once in life, and this proportion was much lower among the women with approximately 35%, current smokers had similar proportion in both genders, as the proportion of ex-smokers was considerable among men. Historically, the prevalence of tobacco use is higher among them, but smoking among the female elderly has risen because of the increase in the smoking habit among younger women 10. Stopping smoking should be encouraged among the elderly, as there are benefits observed in all age groups, including functional improvement and reduction in morbidity and mortality11,12.

Similar to other studies, smoking habit decreased with advancing age. Literature shows association of the smoking habit with the following characteristics of the elderly: living alone, lower education and lower income6,13. In this study, there was no significant difference with regard to schooling; living alone was associated with smoking among women and low income was associated with smoking among men.

It was observed that a high frequency of female smokers presented with being underweight, which is similar to the findings of Zaitune etal.6 in the population of the southwest region of the Greater São Paulo, Campinas, and Botucatu, and to the findings of Berto etal.3 in the population of Rio Grande do Sul. Weight gain as a consequence of smoking cessation may represent an obstacle to the treatment of smoking, especially among women, in whom overweight is seen more frequently14. However, in this study, there was an increase in the overweight frequency among the nonsmoking women, but the same did not occur among ex-smokers. Changes in the dietary pattern with aging may be originated from different physiological or psychosocial causes, but smoking is a factor that increases the potential of the negative effects on the eating habits of the elderly, as the smoking habit causes reduction in taste and smell3,15. Furthermore, the anorectic effects of nicotine act on the appetite; thus, the body weight tends to be lower among the smokers6.The determinants for low weight were not described in this work, but it was observed that female smokers had fewer meals a day. In this study, with regard to the quality of food intake, the results followed some patterns observed in the elderly population in Brazil, presenting higher proportion of adequacy in the consumption of food that are protein sources, with the exception of milk and dairy products. Smoking did not significantly harm the consumption of these foods. There was only a reduction in the proportion of adequate consumption of fruits among female current smokers; however, the frequency of adequacy remained above 80%. In contrast, Subar etal.2 observed that the consumption of fruits and vegetables was lower with higher intensity of smoking, increasing the risk of cancer. These authors also observed that smokers are those who skip breakfast more often. Underweight may lead to functional and physical impairment, decreased quality of life, and increased hospitalization and mortality, due to low caloric and nutrients intake16.

The concurrence of smoking with other unhealthy habits, observed in this study, has been identified in several studies, such as excessive alcohol consumption and poor diet3,6 .

The frequency of smokers decreased and ex-smokers increased with the presence of two or more chronic diseases. This distribution was also observed among women, when analyzed with respect to gender. Further analysis is needed to justify these data; however, Zaitune etal.6 pointed out two phenomena that may explain this fact, which are as follows: smoking cessation as a consequence of the onset of diseases and the increased risk of early mortality among the smokers. Moreover, the highest proportion of female smokers with good self-perception of health and smokers in the age group of 60 to 64 years support this assumption.

This study has limitations of not being able to determine whether the nutritional status observed in smokers is due to smoking, as the cross-sectional design of the study does not establish the temporality of events. Although the nutritional assessment by the BMI has its limitations, such as the identification of body composition and central adiposity, it is still considered a good indicator in epidemiological studies, according to Silveira etal.17, because it presents low cost, easy application, association with noncomunnicable chronic diseases, among others.


Smoking habit among the elderly individuals was associated with the worst nutritional status, with fewer meals per day compared with nonsmoker elderly, contributing to the higher frequency of underweights observed in this study. In view of the observed results and considering the impact on health, elderly smokers deserve special attention in relation to their nutritional status.


1. Veras, R. Envelhecimento populacional contemporâneo: demandas, desafios e inovações / Population aging today: demands, challenges and innovations. Rev Saúde Pública 2009; 43(3): 548-54. [ Links ]

2. Subar AF, Harlan LC, Mattson ME. Food and nutrient intake differences between smokers and non-smokers in the US. Am J Public Health 1990; 80(11): 1323-9. [ Links ]

3. Berto SJ, Carvalhaes MA, Moura EC. Tabagismo associado a outros fatores comportamentais de risco de doenças e agravos crônicos e não transmissíveis / Smoking associated with other behavioral risk factors for chronic non-communicable diseases. Cad Saúde Pública 2010; 26(8): 1573-82. [ Links ]

4. Mathers CD, Loncar D. Projections of global mortality and burden of disease from 2002 to 2030. PLoS Med 2006; 3(11): e442. DOI: 10.1371/journal.pmed.0030442. [ Links ]

5. Bezerra FC, Almeida MI de, Nóbrega-Therrien SM. Estudos sobre envelhecimento no Brasil: revisão bibliográfica / Studies on aging in Brazil: literature review. Rev Bras Geriatr Gerontol 2012; 15(1): 155-67. DOI: 10.1590/S1809-98232012000100017. [ Links ]

6. Zaitune MP, Barros MB, Lima MG, César CL, Carandina L, Goldbaun M, et al. Fatores associados ao tabagismo em idosos: Inquérito de Saúde no Estado de São Paulo (ISA-SP) / Factors associated with smoking in the elderly: a health survey in São Paulo (ISA-SP). Cad Saúde Pública 2012; 28(3): 583-96. [ Links ]

7. Pires SL, Gagliardi RJ, Gorzoni ML. Estudo das frequências dos principais fatores de risco para acidente vascular cerebral isquêmico em idosos / Study of the main risk factors frequencies for ischemic cerebrovascular disease in eldery patients. Arq Neuropsiquiatr2004; 62(3B): 844-51. [ Links ]

8. Lebrão ML, Laurenti R. Saúde, bem-estar e envelhecimento: o estudo SABE no Município de São Paulo / Health, well-being and aging: the SABE study in São Paulo, Brazil. Rev Bras Epidemiol 2005; 8(2): 127-41. [ Links ]

9. Organização Pan-Americana da Saúde. Saúde, bem estar e envelhecimento. O Projeto SABE de São Paulo: uma abordagem inicial. Brasília: OPAS; 2003. [ Links ]

10. Global Youth Tobacco Survey Collaborating Group. Differences in worldwide tobacco use by gender: findings from the Global Youth Tobacco Survey. J Sch Health 2003; 73(6): 207-15. [ Links ]

11. Allen SC. Tobacco smoking in older people. Reviews in Clinical Gerontology 2009; 19(3): 185-92. DOI: 10.1017/S0959259809990256. [ Links ]

12. Doll R, Peto R, Boreham J, Sutherland I. Mortality in relation to smoking: 50 years' observations on male British doctors. BMJ 2004; 328(7455): 1519. [ Links ]

13. Peixoto SV, Firmo JO, Lima-Costa MF. Fatores associados ao hábito de fumar entre idosos (Projeto Bambuí) / Factors associated to smoking habit among older adults (The Bambuí Health and Aging Study). Rev Saúde Pública 2005; 39(5): 746-53. [ Links ]

14. Nascimento C de M, Ribeiro AQ, Cotta RM, Acurcio F de A, Peixoto SV, Priore SE etal. Estado nutricional e fatores associados em idosos do Município de Viçosa, Minas Gerais, Brasil / Nutritional status and associated factors among the elderly in Viçosa, Minas Gerais State, Brazil. Cad Saúde Pública 2011; 27(12): 2409-18. [ Links ]

15. Campos MTFS, Monteiro JBR, Ornelas APR de C. Fatores que afetam o consumo alimentar e a nutrição do idoso / Factors that affect the aged people food intake and nutrition. Rev Nutr 2000; 13(3): 157-65. DOI: 10.1590/S1415-52732000000300002. [ Links ]

16. Schilp J, Wijnhoven HA, Deeq DJ, Visser M. Early determinants for the development of undernutrition in an older general population: Longitudinal Aging Study Amsterdam. Br J Nutr 2011; 106(5): 708-17. DOI: 10.1017/S0007114511000717. [ Links ]

17. Silveira EA, Lopes ACS, Caialfa WT. Avaliação do estado nutricional de idosos. In: Kac G, Sichieri R, Gigante DP, organizadores. Epidemiologia nutricional. Rio de Janeiro: Atheneu/Fundação Oswaldo Cruz; 2007. [ Links ]

Financial support: Fundação de Amparo à Pesquisas do Estado de São Paulo (FAPESP), process 53778-3/2009; DMTZ partially funded by the National Scientific and Technological Development Council (CNPq), process 71702/2011-0.

Received: March 12, 2015; Accepted: March 31, 2015

Corresponding author: Dirce Maria Trevisan Zanetta. Faculdade de Saúde Pública da Universidade de São Paulo. Departamento de Epidemiologia. Avenida Dr. Arnaldo, 715, CEP: 01246-904, São Paulo, SP, Brasil. E-mail:

*in memoriam

Conflict of interests: nothing to declare

Creative Commons License Este é um artigo publicado em acesso aberto sob uma licença Creative Commons