Healthy aging profile in octogenarians in Brazil

ABSTRACT Objective: to identify the healthy aging profile in octogenarians in Brazil. Method: this population-based epidemiological study was conducted using household interviews of 335 octogenarians in a Brazilian municipality. The decision-tree model was used to assess the healthy aging profile in relation to the socioeconomic characteristics evaluated at baseline. All of the tests used a p-value < 0.05. Results: the majority of the 335 participating older adults were women (62.1%), were aged between 80 and 84 years (50.4%), were widowed (53.4%), were illiterate (59.1%), had a monthly income of less than one minimum wage (59.1%), were retired (85.7%), lived with their spouse (63.8%), did not have a caregiver (60.3%), had two or more children (82.7%), and had two or more grandchildren (78.8%). The results indicate three age groups with a healthier aging profile: older adults aged 80 to 84 years (55.6%), older adults aged 85 years and older who are married (64.9%), and older adults aged 85 and older who do not have a partner or a caregiver (54.2%). Conclusion: the healthy aging profile of octogenarians can be explained by age group, marital status, and the presence of a caregiver.


Introduction
The aging of the world population is occurring heterogeneously. In 1999, 10% of the population was aged 60 years and older, ranging from 19% in developed countries to 5% in developing countries. The United Nations estimates that this percentage will double by 2050 (1) .
The last Brazilian census confirms that this process is occurring faster in Brazil than it occurred in Europe at the beginning of the demographic transition. In 2000, the population aged over 60 years corresponded to 8.6% of the total population, whereas this percentage increased to 10.8% in the 2010 census. Adults older than 80 years constitute the age group with the highest percentage of growth in recent years, representing 14.3% of older adults in Brazil and 1.5% of the total population of Brazil in 2010 (2) .
Camarano (3) reported that the elderly population is also aging, i.e., the proportion of those aged ≥ 80 years is also increasing, thereby changing the age composition of the group. The analysis of the different trajectories of life of older people reveals that they fall into different social and economic statuses in Brazil.
However, few studies in this area have investigated the lifestyle and health status of older adults of this new age stratum. This context challenges governments and society to pursue actions and health promotion policies considering a broader perspective of health and wellbeing in old age (4) . The concept of healthy aging can help explain these demographic and epidemiological changes.
From a broader perspective, healthy aging is a continuous process of learning and personal development aimed at achieving autonomy and independence for elderly people (4) . This process also involves the balance of the interactions between the various dimensions of life of this age group: physical and mental health, independence and autonomy in activities of daily living, social involvement and support, family interaction and support, and economic independence (5) .
A survey conducted in a subsample of the Healthy Aging Processes (Processos do Envelhecimento Saudável-PENSA) study sought to investigate how older people perceive the multiple dimensions of the concept of healthy aging. Regarding the factors considered necessary to achieve healthy aging, the categories most frequently mentioned by older people were physical health (53%), social health (46%), emotional health (37%), concern about nutrition and exercise (36%), and the prevention of risk factors (19%) (6) .
In practice, the lack of parameters related to functional, cognitive, physical, and social aspects, as well as the lack of physiological parameters, has limited the identification of healthy older adults (7) , particularly among octogenarians.
Therefore, the aim of this study was to assess the healthy aging profile of octogenarians in Brazil.

Methods
The Aging, Gender, and Quality of Life (AGEQOL) study is a population-based cohort study conducted in Sete Lagoas, state of Minas Gerais, Brazil, in 2012, with a sample representative of the older population of the municipality (8) .
Sampling was performed in two stages: in the first stage, census sectors were selected; in the second stage, households were selected, and their number was proportional to the size of each sector. In each household, all residents aged 60 years and older were interviewed, regardless of their marital status or degree of kinship. Data were collected by trained researchers via interviews with the older adults in their homes.   The use of medications was evaluated by the number of drugs taken at the time of the study. A dichotomous question was used to assess the presence or absence of chronic or acute diseases known at the time of the study. The frequency of alcohol intake was evaluated for the three months before the beginning of the study using the following question: "On average, how many days a week do you consume alcoholic beverages such as beer, wine, liquor, and sugarcane rum?" The response categories were none, less than one day per week, one day per week, two to three days per week, four to six days per week, and every day. With regard to smoking, the participants responded whether they 1) never smoked, 2) smoked but quit, 3) smoked occasionally, i.e., less than one cigarette per day, or 4) currently smoked, i.e., at least one cigarette per day (7) .
To assess cognitive status, the Mini-Mental State Examination (MMSE) validated in Brazil (9) was used with The outcome variable of the study sample was healthy aging. This variable was evaluated using a structured questionnaire applied in the homes of the participants between January and March 2014. This variable was constructed with adaptations from the study conducted by Carrasco (7) . The protocol was developed by a team of geriatric medicine experts to identify healthy individuals from the community considering the principles of healthy aging together with a low comorbidity burden. In this study, the following criteria were used to identify a healthy older adult: positive selfperceived health (good and very good); functionally independent for all daily and instrumental activities; not suffering from cognitive impairment; capable of walking at least three blocks without assistance; no acute or chronic diseases; taking fewer than three medications; not smoking (never having smoked or having smoked but quit); and rarely consuming alcohol (none or less than one day per week). Initially, the data were analyzed descriptively.
The associations between the variables of interest were evaluated using the Chi-square test at a level of significance of 5%.
The decision-tree model was used to analyze the healthy aging profile relative to the other predictors. This method consists of decision rules used to successively subdivide the dataset to make it increasingly homogeneous for the outcome variable. The decision tree is presented in the form of a graph and starts with a root node that includes all the characteristics of the study sample. The nodes produced in sequence represent subdivisions of the data in groups that are increasingly homogeneous (11) .
The method used was the Chi-squared Automatic Interaction Detector (CHAID) algorithm, which can detect and record the non-linear effects on the response variable and the interactions between the predictors. The interpretation of the tree focuses primarily on analyzing the group with the largest number of individuals formed by the last node of the tree, which represents the final result of the model (12) .

Results
The age of the study sample in 2014 ranged between 80 and 108 years, with a mean age of 85.2 ± 4.6 years (85.3 ± 4.8 for women and 85.1 ± 4.1 for men).
The analysis between the genders (Table 2) indicated that most women did not consume alcohol       Figure 2).
lifestyle and habits (not smoking or drinking alcohol and taking few medications), adequate mobility, and lack of morbidities.
Previous studies have shown that healthy behaviors, including not smoking, management of weight and blood pressure, and regular exercise, are associated with healthy aging and improved quality of life in older people (23)(24)(25) . For this reason, it is necessary to invest in individual and public health interventions to guide these subjects and to develop new strategies to ensure a longer and healthier life for future generations (26) .
Notably, we expected to find gender differences for healthy aging among octogenarians in the tree model, based on the literature on aging (8,(27)(28) . of married older adults with a better psychological wellbeing (16) .
The stratum of unhealthy subjects older than 85 years was associated with being single, widowed, or divorced (p = 0.011) and requiring the assistance of a caregiver (p = 0.020). These results indicate that this age group comprised older adults with more morbidities and functional limitations, who required more care, and did not have a partner to share their life with and help in daily activities.
The caregiver is often a close relative and someone of the female gender (spouses, daughters, and granddaughters) who resides in the home of the older adults and becomes responsible for all aspects of their life (30) . Intervention actions in this age group are unique and should enable family members and formal caregivers to better address the difficulties related to functional, physical, cognitive, and psychological limitations of older individuals.

Discussion
In general, the profile of the population studied was similar to that of previous studies conducted in Brazil (4,13) and other countries (14)(15)(16) . The majority of the participants were women, widows, had a low education level, were retired, had an income of less than one minimum wage, and were living with family members or caregivers.
In the last Brazilian population census, the illiteracy rate among older people was 26.2% (17) , and this rate is similar to that found in the AGEQOL study (28.2%) (8) . These values were even higher (59.1%) in the ≥ 80 year age group, as reported in other studies (4,13) . and Aging (Saúde, Bem-estar e Envelhecimento-SABE) study (18) and in other Brazilian studies involving octogenarians (19)(20) .
The condition of living alone was a cause for concern and was more frequent among women, possibly because of the greater likelihood of remarrying observed among men (18) , which does not occur among women.
These data demonstrate the importance of adjusting to a new family arrangement in this age group (20) .
In addition, substantial differences were observed between the genders for smoking and alcohol  (21) . The studies on smoking and alcohol consumption and their consequences among older people suggest correlations with gender, ethnicity, and socioeconomic status (22) .
In this study, the concept of healthy aging included an adequate perception of health, independence to perform ADLs, absence of cognitive impairment, healthy Rev. Latino-Am. Enfermagem 2016;24:e2724 Living alone is a risk factor for depressive symptoms and decline in psychological well-being among older adults (16) . Therefore, older people with a stable and reliable family support system can build strong family relationships and better overcome possible losses during the aging process (21) .
The limitations of this study include the lack of genetic information and the lack of corroboration with baseline data. The lack of previous data on the variables that compose the concept of healthy aging in this study prevented the separate calculation of survival for the healthy and unhealthy groups. The healthy aging profile was investigated using subjective self-reported information, which could lead to recall bias.
Nonetheless, this is one of the few studies that has evaluated octogenarians using baseline data from a random sample with a high response rate to make inferences using complex statistical tools such as the decision tree.
The incorporation of other variables and the geoprocessing of data can help broaden the discussion and establish a temporal relationship between healthy aging and marital status among octogenarians in Brazil.
Considering that marital status was an important determinant of the healthy aging profile in this sample, other data should be evaluated, including the subtypes of family arrangements (with two or three generations), the reasons for living at home, and whether these older adults were the head of the household (20) .
Further research should better assess the different requirements of and formulate public policies for this age group, considering the heterogeneity of this population regarding age and socioeconomic status (3) .

Conclusion
Healthier older adults had a positive self-perceived health; the absence of functional impairment, cognitive impairment, and other morbidities; adequate mobility; and healthy habits. In the final model, this profile was determined by age, marital status, and the need for a caregiver.
The concept of healthy aging adopted here can be considered a reliable and practical model in to the continued care of older adults whose health has been impaired by morbidities or the natural weakening process because of increasing age and should provide strategies to improve the relationship between older adults and family caregivers.