Abstracts

Introduction

Frailty in the elderly is defined as a clinical syndrome characterized by a decrease in energy reserve, strength, and performance, resulting in a cumulative decline in multiple physiological systems, leading to a state of greater vulnerability ¹1 . Morley JE, Vellas B, van Kan GA, Anker SD, Bauer JM, Bernabei R, et al. Frailty consensus: a call to action. J Am Med Dir Assoc 2013; 14:392-7.^,22 . Fried LP, Tangen CM, Walston J, Newman AB, Hirsch C, Gottdiener J, et al. Frailty in older adults: evidence for a phenotype. J Gerontol A Biol Sci Med Sci 2001; 56:M146-56.. This condition causes practical harm to the life of elders and their families, both clinically and psychosocially, since it is associated with greater risk of adverse consequences such as dependency, falls, hospitalization, institutionalization, and death ¹1 . Morley JE, Vellas B, van Kan GA, Anker SD, Bauer JM, Bernabei R, et al. Frailty consensus: a call to action. J Am Med Dir Assoc 2013; 14:392-7.^,22 . Fried LP, Tangen CM, Walston J, Newman AB, Hirsch C, Gottdiener J, et al. Frailty in older adults: evidence for a phenotype. J Gerontol A Biol Sci Med Sci 2001; 56:M146-56.^,33 . Fried LP, Ferrucci L, Darer J, Williamson JD, Anderson G. Untangling the concepts of disability, frailty, and comorbidity: implications for improved targeting and care. J Gerontol A Biol Sci Med Sci 2004; 59:M255-63.^,44 . Ministério da Saúde. Envelhecimento e saúde da pessoa idosa. Brasília: Ministério da Saúde; 2006. (Cadernos de Atenção Básica, Normas e Manuais Técnicos, 19).^,55 . Espinoza SE, Fried LP. Risk factors for frailty in the older adult. Clin Geriatr 2007; 15:37.. Prevalence in Americans is 6.3% ²2 . Fried LP, Tangen CM, Walston J, Newman AB, Hirsch C, Gottdiener J, et al. Frailty in older adults: evidence for a phenotype. J Gerontol A Biol Sci Med Sci 2001; 56:M146-56., and in Brazilians it varies from 10 to 25% in persons above 65 years and 46% above 85 years ⁴4 . Ministério da Saúde. Envelhecimento e saúde da pessoa idosa. Brasília: Ministério da Saúde; 2006. (Cadernos de Atenção Básica, Normas e Manuais Técnicos, 19)..

The syndrome should not be confused with functional dependency or comorbidity. Frailty refers to the fact that a person needs help or requires assistance to perform an activity, or fails to perform it. Individuals are classified as dependent when they need help from someone else or are unable to perform a task ³3 . Fried LP, Ferrucci L, Darer J, Williamson JD, Anderson G. Untangling the concepts of disability, frailty, and comorbidity: implications for improved targeting and care. J Gerontol A Biol Sci Med Sci 2004; 59:M255-63.^,66 . Alves LC, Leimann BCQ, Vasconcelos MEL, Carvalho MS, Vasconcelos AGG, Fonseca TCO, et al. A influência das doenças crônicas na capacidade funcional dos idosos do Município de São Paulo, Brasil. Cad Saúde Pública 2007; 23:1924-30.. Meanwhile, comorbidity is a general concept that encompasses the presence of several diagnosed illnesses ³3 . Fried LP, Ferrucci L, Darer J, Williamson JD, Anderson G. Untangling the concepts of disability, frailty, and comorbidity: implications for improved targeting and care. J Gerontol A Biol Sci Med Sci 2004; 59:M255-63..

Studies have focused on understanding the causes and pathophysiology of frailty, defining and improving criteria to identify elderly at risk and analyzing factors that influence development of the syndrome. There are different definitions for the identification of frailty. The most widely used is that of Fried et al. ²2 . Fried LP, Tangen CM, Walston J, Newman AB, Hirsch C, Gottdiener J, et al. Frailty in older adults: evidence for a phenotype. J Gerontol A Biol Sci Med Sci 2001; 56:M146-56., who define frail elderly as those with three or more of the following indicators: unintentional weight loss, low level of physical activity, reduced grip strength, reduced gait speed, and self-reported fatigue. Another criterion that has been discussed in the scientific literature is that of Rockwood et al. ⁷7 . Rockwood K. Mitnitski A. Frailty in relation to the accumulationof deficits. J Gerontol Med Sci 2007 62:722-7., which adds cognitive and emotional aspects to the diagnostic indicators. There are still other criteria, with no consensus in the academic community on the best approach to diagnosis; however, in a recently published report ¹1 . Morley JE, Vellas B, van Kan GA, Anker SD, Bauer JM, Bernabei R, et al. Frailty consensus: a call to action. J Am Med Dir Assoc 2013; 14:392-7., experts agreed that health professionals should choose a well-validated model among the existing ones.

Since the factors related to the syndrome are not fully known, it is extremely important to understand it in order for targeted measures to be planned and implemented. Many of these health problems can be prevented at the primary care level, as long as healthcare professionals are alert to the determinant factors for the syndrome and aware of the importance of early detection. Studies have shown that various physiological, socio-demographic, psychological, and nutritional factors may be involved in the origin of frailty, in addition to related comorbidities ⁵5 . Espinoza SE, Fried LP. Risk factors for frailty in the older adult. Clin Geriatr 2007; 15:37.^,88 . Levers MJ, Estabrooks CA, Kerr JCR. Factors contributing to frailty: literature review. J Adv Nurs 2006; 56:282-91.^,99 . Kaiser MJ, Bandinelli S, Lunenfeld B. Frailty and the role of nutrition in older people: a review of the current literature. Acta Biomed 2010; 81 Suppl 1:37-45..

The current systematic review aims to identify socio-demographic, psycho-behavioral, health-related, nutritional, and lifestyle factors associated with frailty in the elderly.

Methods

Databases and search strategy

Articles published from 2001 to 2013 were selected from the following databases: MEDLINE via PubMed, Scopus, LILACS, and ISI Web of Knowledge. The descriptors and MesH terms consulted in the search engines were: “age factors” OR “risk factors” OR “socioeconomic factors” OR “demographic factors” OR “clinical factors”, “biological factors” OR “behavior factors” OR “elderly nutrition” OR “nutrition”, “health status” OR “epidemiological factors” OR “elder nutritional physiological phenomena” in the field all words in the literature bases, in combination using the Boolean connector AND with “aging” OR “aged” OR “elderly” OR “senescence”, in the field all words and associated by the Boolean connector AND with “frail elderly” OR “frailty” OR “fragility” OR “elderly frail” OR “frail older adults” in the field Title and/or abstract + key words. Articles in English, Spanish, and Portuguese were selected.

Selection criteria

The review used the following selection criteria: original scientific articles published in Brazilian or international periodicals; publication from 2001 to March 2013; study population 60 years or older; observational study design (cross-sectional, cohort, or case-control); individual selection by probabilistic sample or article showing the sampling design; and identification of factors associated with frailty in the elderly as the principal or secondary objective.

Importantly, there are different diagnostic criteria for frailty, with no consensus in the literature as to the most adequate markers for its identification. However a widely used and well-accepted criterion in the scientific community is that of Fried et al. ²2 . Fried LP, Tangen CM, Walston J, Newman AB, Hirsch C, Gottdiener J, et al. Frailty in older adults: evidence for a phenotype. J Gerontol A Biol Sci Med Sci 2001; 56:M146-56., published in 2001. Based on a study of Americans participating in the Cardiovascular Health Study, the group proposed that the syndrome’s pathophysiology can be identified by a phenotype, using five measurable components:

• Self-reported unintentional weight loss of 4.5kg or 5% of body weight in the previous year;

• Self-reported fatigue assessed by the following: “I felt tired all the time” and “I could not get going”, from the depression scale of the Center for Epidemiological Studies (CES-D) ¹⁰10 . Orme J, Reis J, Herz E. Factorial and discriminate validity of the Center for Epidemiological Studies depression (CES-D) scale. J Clin Psychol 1986; 42:28-33.;

• Decreased grip strength, measured with a dynamometer in the dominant hand, stratified by gender and body mass index (BMI) quartiles;

• Low level of physical activity measured as weekly energy expenditure in kcal, with information obtained from the reduced version of the Minnesota Leisure Time Activity Questionnaire ¹¹11 . Taylor HL, Jacobs Jr. DR, Schucker B, Knudsen J, Leon AS, Debacker G. A questionnaire for the assessment of leisure-time physical activities. J Chronic Dis 1978; 31:745-55., stratified by gender;

• Decreased gait speed in seconds, calculated by recording the time to walk 4.6m at a comfortable pace, stratified by gender and mean height.

The presence of three or more components defines a frail elder. The presence of one to two identifies those at high risk of developing the syndrome (pre-frail). We only selected articles that reported using this criterion to identify frailty, so the article search began in 2001, the year this definition was published.

Data extraction

Article selection and data extraction were performed independently by three reviewers, using a standardized instrument containing: country and study site; sampling number; type of sample; study design; characteristics of sample member; study variables; criterion used to identify frailty; statistical technique; principal results; and limitations.

Assessment of risk of bias

Assessment of the articles included in the analysis used verification of the risk of bias, as suggested by the Cochrane Collaboration ¹²12 . Higgins JPT, Altman DG. Chapter 8: assessing risk of bias in included studies. In: Higgins JPT, Green S, editors. Cochrane Handbook for Systematic Reviews of Interventions, Version 5.0.1. The Cochrane Collaboration; 2008. http://www.cochrane-handbook.org (accessed on 20/May/2013).
http://www.cochrane-handbook.org... . To orient the assessment of this risk, an adapted version of the Newcastle-Ottawa Scale ¹³13 . Wells GA, Shea B, O’Connell D, Peterson J, Welch V, Losos M, Tugwell P. The Newcastle-Ottawa Scale (NOS) for assessing the quality of nonrandomised studies in meta-analyses. http://www.ohri.ca/pro grams/clinical_epidemiology/oxford.asp (accesssed on 20/May/2013).
http://www.ohri.ca/pro grams/clinical_ep... was used (Table 1). The original Newcastle-Ottawa Scale was developed to assess the quality of observational studies and contains eight items that analyze three dimensions: selection, comparability, and outcome (in the case of cohort studies) or exposure (case-control). For each item there is a series of options in which the one that best reflects quality is marked with a star; the more the stars, the higher the study’s quality ¹⁴14 . Stang A. Critical evaluation of the Newcastle-Ottawa scale for the assessment of the quality of nonrandomized studies in meta-analyses. Eur J Epidemiol 2010; 25:603-5.. In the current study, questions were adjusted to investigate exposure and outcome (frailty according to the definition by Fried et al. ²2 . Fried LP, Tangen CM, Walston J, Newman AB, Hirsch C, Gottdiener J, et al. Frailty in older adults: evidence for a phenotype. J Gerontol A Biol Sci Med Sci 2001; 56:M146-56.), and risk of bias was divided into low risk of bias, uncertain risk of bias, and high risk of bias, according to the item being assessed. Each star represents an item for classification of low risk of bias.

Results

The literature search identified 4,183 publications. Of these, we eliminated 629 duplicates that came from two or more databases, and after reading the titles and abstracts, 3,372 were ruled out because frailty was an independent variable, the study was designed as an intervention, review, or validation of a diagnostic criterion, or the publication was a book or journal commentary or editorial. A total of 182 complete articles were selected for analysis. After reading them and applying the selection criteria, 35 eligible articles were left. Figure 1 shows the study selection flow.

Overall study characteristics

In the 35 studies, the data collection period ranged from 1989 to 2011. The year with the most publications was 2012 (n = 10). Most studies were on North American participants (n = 12), followed by Europeans (n = 11), Latin Americans (n = 9), and Asians (n = 3). The number of subjects varied from 77 to 40,657, and most studies had samples greater than 600 individuals. Age of the elderly was greater than 65 years, except for three studies, in which it was greater than 60. The results analyzed in this article are mainly from cross-sectional studies (n = 27).

Table 2 shows details on the main characteristics of the 35 studies, with the design, independent variables, statistical technique, principal results, and limitations listed by authors.

The most widely studied independent variables were demographic (n = 33), diseases and health conditions (n = 30), socioeconomic (n = 30), psycho-behavioral (n = 23), and nutritional (n = 17), and the least studied were lifestyle variables (n = 11). The majority of the studies used logistic regression models (n = 24). All the results presented here were statistically significant.

Demographic and socioeconomic factors and frailty

Of the 35 studies, demographic factors were assessed by 33 and socioeconomic factors by 30. The most frequently assessed demographic variable was age (n = 31) and the most common socioeconomic value was schooling or educational level (n = 27). In general, age, black race/color, and female gender showed a positive association with frailty, while there was an inverse association between frailty and schooling and income.

Diseases, health conditions, and frailty

The principal diseases assessed by the studies were cardiovascular diseases (CVD) (n = 17), diabetes mellitus (n = 17), systemic arterial hypertension (SAH) (n = 14), pulmonary diseases (n = 10), arthritis (n = 11), cancer (n = 8), and stroke (n = 7). Fourteen studies also included comorbidities/diseases as an independent variable. Frailty showed a positive association especially with CVD and number of comorbidities/diseases. As predicted, no disease showed an inverse association with frailty or was considered to have a protective effect.

Sixteen studies investigated functional status, measured mainly by activities of daily living (ADL) and instrumental activities of daily living (IADL), and diagnosis of functional incapacity showed a positive association with frailty in nine.

Eight studies analyzed self-rated health and found a positive association between poor self-rated health and frailty.

Psycho-behavioral factors and frailty

Depressive symptoms were assessed in 17 studies, and cognitive function was tested in 15. The instrument most widely used to assess cognitive function was the Mini Mental State Examination (MMSE) ¹⁵15 . Folstein MF, Folstein SE, McHugh PR. Mini-Mental State: a practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 1975; 12:189-98., and the elderly that tested highest were considered to have the best cognitive function. An inverse association was found between cognitive function and frailty, while depressive symptoms showed a positive association with the syndrome.

Nutritional status, lifestyle and frailty

The most widely assessed nutritional variable was BMI (n = 13). Lifestyle factors were the least analyzed in the selected articles and included smoking (n = 10), alcohol consumption (n = 6), and physical activity (n = 2). A positive association was found between smoking and frailty in two studies, and an inverse association was observed between alcohol and frailty in three studies. Most of the studies found a positive association between frailty and BMI, and two studies showed that underweight elderly according to BMI had a positive association with frailty.

Limitations listed by the authors of the articles analyzed in the final sample

The limitations most frequently cited by the authors of the selected articles were: cross-sectional design (not allowing causal inferences); adaptations of scales suggested by Fried et al. ²2 . Fried LP, Tangen CM, Walston J, Newman AB, Hirsch C, Gottdiener J, et al. Frailty in older adults: evidence for a phenotype. J Gerontol A Biol Sci Med Sci 2001; 56:M146-56.; and self-reporting of data.

Assessment of risk of bias

Table 3 summarizes the assessment of risk of bias in the studies, and Figure 2 shows the graph for each question in the adapted Newcastle-Ottawa Scale ¹³13 . Wells GA, Shea B, O’Connell D, Peterson J, Welch V, Losos M, Tugwell P. The Newcastle-Ottawa Scale (NOS) for assessing the quality of nonrandomised studies in meta-analyses. http://www.ohri.ca/pro grams/clinical_epidemiology/oxford.asp (accesssed on 20/May/2013).
http://www.ohri.ca/pro grams/clinical_ep... .

In relation to analysis of the risk of bias, 34 studies collected information on the independent variables using a structured interview, anthropometric measurements, and clinical tests (low risk of bias), and only one failed to clearly describe the method (uncertain risk of bias).

As for participant selection, all were local community-dwelling, non-institutionalized elders (low risk of bias).

More than half of the studies (n = 19) showed changes in three or more of the components in the criterion adopted by Fried et al. ²2 . Fried LP, Tangen CM, Walston J, Newman AB, Hirsch C, Gottdiener J, et al. Frailty in older adults: evidence for a phenotype. J Gerontol A Biol Sci Med Sci 2001; 56:M146-56. (high risk of bias).

Only one article mentioned that the sample was representative of the local population (low risk of bias).

Only one of the longitudinal studies (n = 7) specified that the sample did not present the syndrome at the beginning of the cohort (low risk of bias).

Discussion

The principal socio-demographic, psycho-behavioral, health-related, nutritional, and lifestyle factors positively associated with frailty were: age, female gender, black race/color, cardiovascular diseases, number of comorbidities/diseases, functional incapacity, poor self-rated health, depressive symptoms, BMI, and smoking. Inversely associated factors were schooling, income, cognitive function, and alcohol use.

Although the selected studies had different designs, sample sizes, and locations, they showed homogeneity in the relations between the demographic and socioeconomic variables and frailty. A longitudinal study of 5,317 North Americans over 65 years of age showed that prevalence of frailty was higher in the oldest old, women, blacks, and low-income individuals ²2 . Fried LP, Tangen CM, Walston J, Newman AB, Hirsch C, Gottdiener J, et al. Frailty in older adults: evidence for a phenotype. J Gerontol A Biol Sci Med Sci 2001; 56:M146-56.. This association was also found in a longitudinal study of French elderly ¹⁶16 . Ávila-Funes JA, Helmer C, Amieva H, Barberger-Gateau P, Le Goff M, Ritchie K, et al. Frailty among community-dwelling elderly people in France: the three-city study. J Gerontol A Biol Sci Med Sci 2008; 63:1089-96. and in cross-sectional studies of both American and Spanish elderly ¹⁷17 . Michelon E, Blaum C, Semba RD, Xue QL, Ricks MO, Fried LP. Vitamin and carotenoid status in older women: associations with the frailty syndrome. J Gerontol A Biol Sci Med Sci 2006; 61:600-7.^,1818 . Alcalá MVC, Puime ÁO, Santos MTS, Barral AG, Montalvo JIG, Zunzunegui MV. Prevalencia de fragilidad en una población urbana de mayores de 65 años y su relación con comorbilidad y discapacidad. Aten Primaria 2010; 42:520-7.. At more advanced ages there is an increase in pre-frail and frail elderly, suggesting that frailty is a progressive condition; the phenomenon occurs more significantly after 80 years of age. One hypothesis for this relationship between increasing age and frailty lies in the cellular oxidative stress that accumulates over the years, modulated by exogenous and endogenous agents that influence the production of reactive oxygen species, leading to DNA damage. Such damage induces alterations at the cellular and systemic levels, with deregulations in the processes of inflammation, apoptosis, necrosis, and proliferation that result in various adverse conditions that increase over the years, such as loss of muscle mass (sarcopenia), diabetes, cancer, and frailty ¹⁹19 . Mulero J, Zafrilla P, Martinez-Cacha A. Oxidative stress, frailty and cognitive decline. J Nutr Health Aging 2011; 15:756-60.^,2020 . Karahalil B, Bohr VA, Wilson DM 3rd. Impact of DNA polymorphisms in key DNA base excision repair proteins on cancer risk. Hum Exp Toxicol 2012; 31:981-1005.^,2121 . Cutler RG. Human longevity and aging: possible role of reactive oxygen species. Ann N Y Acad Sci 1991; 621:1-28..

The higher prevalence of frailty in women can be explained by the greater physiological muscle mass loss in females during aging, in addition to their being more prone to the development of sarcopenia, an intrinsic risk for developing the frailty syndrome ⁵5 . Espinoza SE, Fried LP. Risk factors for frailty in the older adult. Clin Geriatr 2007; 15:37.. Other hypotheses included women’s greater longevity and the fact that they show a higher prevalence of chronic illnesses than men ²²22 . Pinheiro RS, Viacava F, Travassos C, Brito AS. Gênero, morbidade, acesso e utilização de serviços de saúde no Brasil. Ciênc Saúde Coletiva 2002; 7:687-707..

Race is a strong conditioning factor for health status, since blacks are at a disadvantage in relation to whites. Studies have shown that black race/color is an important indicator of low socioeconomic status and is associated with deficient health and high mortality risk ²³23 . Fiorio NM, Flor LS, Padilha M, Castro DS, Molina MCB. Mortalidade por raça/cor: evidências de desigualdades sociais em Vitória (ES), Brasil. Rev Bras Epidemiol 2011; 14:522-30.^,2424 . Batista LE. Masculinidade, raça/cor e saúde. Ciênc Saúde Coletiva 2005; 10:71-80., contributing indirectly and directly to development of the syndrome. Furthermore, some authors believe that race is a marker for genetic polymorphisms that have an influence on the emergence of frailty ²⁵25 . Hirsch C, Anderson ML, Newman A, Kop W, Jackson S, Gottdiener J, et al. The association of race with frailty: The Cardiovascular Health Study. Ann Epidemiol 2006; 16:545-53..

Income and schooling do not act directly in the pathophysiology of frailty, but interfere in the individual’s lifestyle and quality of life and thus in factors that vary with socioeconomic status, including gender and age, which can influence the frailty process ²⁵25 . Hirsch C, Anderson ML, Newman A, Kop W, Jackson S, Gottdiener J, et al. The association of race with frailty: The Cardiovascular Health Study. Ann Epidemiol 2006; 16:545-53..

As for diseases associated with frailty, CVD and the presence of two or more comorbidities are relevant for the occurrence of this syndrome in the elderly. In a cross-sectional study of 1,008 elderly Mexicans, self-reported chronic diseases such as CVD, hypertension, diabetes mellitus, and arthritis were associated with frailty ²⁶26 . Masel MC, Graham JE, Reistetter TA, Markides KS, Ottenbacher KJ. Frailty and health related quality of life in older Mexican Americans. Health Qual Life Outcomes 2009; 7:70.. A cross- sectional and longitudinal study in North Americans, but with diagnosis by clinical examination, showed an equivalent association ²2 . Fried LP, Tangen CM, Walston J, Newman AB, Hirsch C, Gottdiener J, et al. Frailty in older adults: evidence for a phenotype. J Gerontol A Biol Sci Med Sci 2001; 56:M146-56.^,2727 . Newman AB, Gottdiener JS, McBurnie MA, Hirsch CH, Willem JK, Tracy R, et al. Associations of subclinical cardiovascular disease with frailty. J Gerontol A Biol Sci Med Sci 2001; 56:M158-66.. Some researchers contend that CVD and some comorbidities are related to atherosclerosis, a chronic inflammatory state that can result in systemic catabolism and other pathophysiological changes, which can contribute to the clinical manifestations of frailty ²2 . Fried LP, Tangen CM, Walston J, Newman AB, Hirsch C, Gottdiener J, et al. Frailty in older adults: evidence for a phenotype. J Gerontol A Biol Sci Med Sci 2001; 56:M146-56.^,2727 . Newman AB, Gottdiener JS, McBurnie MA, Hirsch CH, Willem JK, Tracy R, et al. Associations of subclinical cardiovascular disease with frailty. J Gerontol A Biol Sci Med Sci 2001; 56:M158-66..

A direct association was also observed between functional incapacity and frailty. Recent cross-sectional studies in both Chinese and Spanish subjects showed that a major portion of the frail elderly show functional incapacity ¹⁸18 . Alcalá MVC, Puime ÁO, Santos MTS, Barral AG, Montalvo JIG, Zunzunegui MV. Prevalencia de fragilidad en una población urbana de mayores de 65 años y su relación con comorbilidad y discapacidad. Aten Primaria 2010; 42:520-7.^,2828 . Chen CY, Wu SC, Chen LJ, Lue BH. The prevalence of subjective frailty and factors associated with frailty in Taiwan. Arch Gerontol Geriatr 2010; 50:S43-7.. A longitudinal study in a robust sample of 5,317 elderly also showed this relationship ²2 . Fried LP, Tangen CM, Walston J, Newman AB, Hirsch C, Gottdiener J, et al. Frailty in older adults: evidence for a phenotype. J Gerontol A Biol Sci Med Sci 2001; 56:M146-56.. The authors contend that frailty can be a precursor of functional incapacity. However, one cannot overlook the possibility of reverse causality between functional capacity and frailty.

In the area of psycho-behavioral variables, decreased cognitive function and the presence of depressive symptoms have been related to frailty. Studies with different samples, (American, Mexican, and French elderly) showed increased prevalence of frailty in elders submitted to different questionnaires with scales for depression or cognitive function and that presented depressive symptoms or cognitive impairment according to the tests ²2 . Fried LP, Tangen CM, Walston J, Newman AB, Hirsch C, Gottdiener J, et al. Frailty in older adults: evidence for a phenotype. J Gerontol A Biol Sci Med Sci 2001; 56:M146-56.^,1616 . Ávila-Funes JA, Helmer C, Amieva H, Barberger-Gateau P, Le Goff M, Ritchie K, et al. Frailty among community-dwelling elderly people in France: the three-city study. J Gerontol A Biol Sci Med Sci 2008; 63:1089-96.^,1818 . Alcalá MVC, Puime ÁO, Santos MTS, Barral AG, Montalvo JIG, Zunzunegui MV. Prevalencia de fragilidad en una población urbana de mayores de 65 años y su relación con comorbilidad y discapacidad. Aten Primaria 2010; 42:520-7.. Elderly with cognitive impairment probably experience greater difficulty in eating, exercising, and walking, which can lead to weight loss and decreased motor function and favor the syndrome’s onset and progression. As for depressive symptoms, the literature shows that the relationship to frailty is biologically plausible, since depressed persons normally present weight loss, limited activity, and isolation, thus predisposing to progressive loss of muscle mass and strength, conditions that accelerate the establishment of the syndrome ⁵5 . Espinoza SE, Fried LP. Risk factors for frailty in the older adult. Clin Geriatr 2007; 15:37..

As for nutritional and lifestyle variables, underweight elderly according to BMI and those with a higher proportion of overweight according to BMI showed a higher prevalence of frailty. In a study of elderly Mexicans, Masel et al. ²⁶26 . Masel MC, Graham JE, Reistetter TA, Markides KS, Ottenbacher KJ. Frailty and health related quality of life in older Mexican Americans. Health Qual Life Outcomes 2009; 7:70. found that underweight was related to frailty. However, although Woods et al. ²⁹29 . Woods NF, LaCroix AZ, Gray SL, Aragaki A, Cochrane BB, Brunner RL, et al. Frailty: emergence and consequences in women aged 65 and older in the Women's Health Initiative Observational Study. J Am Geriatr Soc 2005; 53:1321-30. found the same association, they showed that frailty could also be associated with overweight and obesity. The association between frailty and underweight may be related to the common loss of muscle mass in individuals with unintentional weight loss ³⁰30 . Chemin SS, Mura JDP. Tratado de alimentação, nutrição e dietoterapia. São Paulo: Roca; 2007.. Meanwhile, the relationship between frailty and overweight and obesity may be due to the fact that excess weight can be associated with activation of inflammatory processes, which trigger systemic alterations, which in turn can influence the onset of frailty ⁵5 . Espinoza SE, Fried LP. Risk factors for frailty in the older adult. Clin Geriatr 2007; 15:37.. Still other authors speculate that weight extremes in the elderly are related to loss of muscle mass in arms and legs, and that the phenomenon of “sarcopenic obesity”, referring to weight gain concurrent with loss of muscle mass, leads to difficulty in mobility, reduced strength, and thus physical inactivity, one of the elements in the frailty cycle ³¹31 . Roubenoff R. Sarcopenic obesity. Does muscle loss cause fat gain? Lessons from rheumatoid arthritis and osteoarthritis. Ann N Y Acad Sci 2000; 904:553-7.^,3232 . Jenkins KR. Obesity’s effects on the onset of functional impairment among older adults. Gerontologist 2004; 44:206-16..

As for studies that showed an inverse association between alcohol consumption and frailty ¹⁶16 . Ávila-Funes JA, Helmer C, Amieva H, Barberger-Gateau P, Le Goff M, Ritchie K, et al. Frailty among community-dwelling elderly people in France: the three-city study. J Gerontol A Biol Sci Med Sci 2008; 63:1089-96.^,1717 . Michelon E, Blaum C, Semba RD, Xue QL, Ricks MO, Fried LP. Vitamin and carotenoid status in older women: associations with the frailty syndrome. J Gerontol A Biol Sci Med Sci 2006; 61:600-7.^,2929 . Woods NF, LaCroix AZ, Gray SL, Aragaki A, Cochrane BB, Brunner RL, et al. Frailty: emergence and consequences in women aged 65 and older in the Women's Health Initiative Observational Study. J Am Geriatr Soc 2005; 53:1321-30. and a positive association between smoking and the syndrome ¹⁷17 . Michelon E, Blaum C, Semba RD, Xue QL, Ricks MO, Fried LP. Vitamin and carotenoid status in older women: associations with the frailty syndrome. J Gerontol A Biol Sci Med Sci 2006; 61:600-7.^,2929 . Woods NF, LaCroix AZ, Gray SL, Aragaki A, Cochrane BB, Brunner RL, et al. Frailty: emergence and consequences in women aged 65 and older in the Women's Health Initiative Observational Study. J Am Geriatr Soc 2005; 53:1321-30., the authors do not discuss the possible explanations for such findings. Woods et al. ²⁹29 . Woods NF, LaCroix AZ, Gray SL, Aragaki A, Cochrane BB, Brunner RL, et al. Frailty: emergence and consequences in women aged 65 and older in the Women's Health Initiative Observational Study. J Am Geriatr Soc 2005; 53:1321-30. merely mention that when stratifying the variable in doses per week, elders with moderate alcohol consumption had 13 to 31% lower odds of presenting frailty syndrome, even after adjusting for chronic diseases that have been associated with moderate alcohol use. Caution has been suggested when analyzing such associations, especially those related to alcohol consumption, since not all the studies conducted regression analyses or adjusted for potential confounders when cross-analyzing such variables.

Since 9 of the 35 studies (26%) only included women, a comparison was made been factors associated with frailty in both genders and in studies only with females, showing that there was no difference between the associated factors, suggesting that the elder’s gender does not have a decisive weight in the establishment of the syndrome.

Importantly, the studies varied in both their design and the methods used to measure the independent variables. Most adopted a cross-sectional design, which does not allow establishing a cause-and-effect relationship between the independent variables and the outcome. In addition, 18 studies performed bivariate analyses, and a total of 16 did not adjust for potential confounders. However, in general such limitations appear not to have influenced the associations, considering the consistency between results.

As for assessment of risk of bias, a question that called considerable attention was the adequacy of the diagnostic assessment of frailty. As mentioned, the choice of the criterion adopted by Fried et al. ²2 . Fried LP, Tangen CM, Walston J, Newman AB, Hirsch C, Gottdiener J, et al. Frailty in older adults: evidence for a phenotype. J Gerontol A Biol Sci Med Sci 2001; 56:M146-56. (among various existing criteria) to assess frailty in this review was due to the lack of a consensus or gold standard for identifying the syndrome and to the fact that Fried’s definition is widely used in other Brazilian and international scientific studies. The current review did not aim to critically discuss the instruments for evaluating frailty proposed by the literature, so the analysis of bias in the assessment of frailty merely aimed to verify the extent to which the studies analyzed in the sample deviated from the original proposal by Fried et al. ²2 . Fried LP, Tangen CM, Walston J, Newman AB, Hirsch C, Gottdiener J, et al. Frailty in older adults: evidence for a phenotype. J Gerontol A Biol Sci Med Sci 2001; 56:M146-56..

Focusing on this point, we found that only 36% of the studies assessed frailty comprehensively as Fried et al. ²2 . Fried LP, Tangen CM, Walston J, Newman AB, Hirsch C, Gottdiener J, et al. Frailty in older adults: evidence for a phenotype. J Gerontol A Biol Sci Med Sci 2001; 56:M146-56. proposed, and that 74% performed some modification of the five components. Changes in proposed criteria can lead to erroneous conclusions when comparing the results to those of other studies. Meanwhile, the instruments proposed to assess some component, as for example the questionnaire proposed by Fried et al. ²2 . Fried LP, Tangen CM, Walston J, Newman AB, Hirsch C, Gottdiener J, et al. Frailty in older adults: evidence for a phenotype. J Gerontol A Biol Sci Med Sci 2001; 56:M146-56. to estimate low level of physical activity, may contain items that do not agree with the study’s local reality, which would probably lead the authors to adapt the questionnaire to obtain a more adequate and true response. Furthermore, other authors adopted other criteria and validated such changes in relation to the proposal by Fried et al. ²2 . Fried LP, Tangen CM, Walston J, Newman AB, Hirsch C, Gottdiener J, et al. Frailty in older adults: evidence for a phenotype. J Gerontol A Biol Sci Med Sci 2001; 56:M146-56., while still others did not conduct a validation process or failed to report it. Thus, common sense is recommended when analyzing articles for comparison with data from other authors.

In addition, some studies drawing on the same cohort ¹⁷17 . Michelon E, Blaum C, Semba RD, Xue QL, Ricks MO, Fried LP. Vitamin and carotenoid status in older women: associations with the frailty syndrome. J Gerontol A Biol Sci Med Sci 2006; 61:600-7.^,3333 . Blaum CS, Xue QL, Michelon E, Semba RD, Fried LP. The association between obesity and the frailty syndrome in older women: The Women's Health and Aging Studies. J Am Geriatr Soc 2005; 53:927-34.^,3434 . Semba RD, Bartali B, Zhou J, Blaum C, Ko CW, Fried LP. Low serum micronutrient concentrations predict frailty among older women living in the community. J Gerontol A Biol Sci Med Sci 2006; 61:594-9.^,3535 . Chaves PH, Varadhan R, Lipsitz LA, Stein PK, Windham BG, Tian J, et al. Physiological complexity underlying heart rate dynamics and frailty status in community-dwelling older women. J Am Geriatr Soc 2008; 56:1698-703.^,3636 . Szanton SL, Allen JK, Seplaki CL, Bandeen-Roche K, Fried LP. Allostatic load and frailty in the women's health and aging studies. Biol Res Nurs 2009; 10:248-56.^,3737 . Chang SS, Weiss CO, Xue QL, Fried LP. Patterns of comorbid inflammatory diseases in frail older women: the Women’s Health and Aging Studies I and II. J Gerontol A Biol Sci Med Sci 2010; 65: 407-13.^,3838 . Szanton SL, Seplaki CL, Thorpe RJ, Allen JK, Fried LP. Socioeconomic status is associated with frailty: the Women’s Health and Aging Studies. J Epidemiol Community Health 2010; 64:63-7. used different descriptions of the instruments used to measure the component of the criterion used by Fried et al. ²2 . Fried LP, Tangen CM, Walston J, Newman AB, Hirsch C, Gottdiener J, et al. Frailty in older adults: evidence for a phenotype. J Gerontol A Biol Sci Med Sci 2001; 56:M146-56., which raises doubts in their analysis.

This review presents some limitations. First, by limiting the languages of the publications to English, Portuguese, or Spanish and the databases for the article search, some relevant study may have been left out. The second relates to the limiting the diagnostic criterion for frailty according to Fried et al. ²2 . Fried LP, Tangen CM, Walston J, Newman AB, Hirsch C, Gottdiener J, et al. Frailty in older adults: evidence for a phenotype. J Gerontol A Biol Sci Med Sci 2001; 56:M146-56.. The scientific literature provides different instruments with various markers, which are being tested in international studies. Thus, some outstanding studies may have been lost. Another limitation relates to restricting the presentation of results to those with statistical significance. This decision was due to the fact that one cannot reach conclusions on associations that are not statistically significant, and due to the number and scope of the target variables.

Final remarks

The worldwide increase in prevalence of frailty among the elderly raises challenges for all countries. Knowledge of the factors associated with the syndrome and the complexity of its determinants helps formulate measures for prevention and early intervention, thus fostering aging with better quality of life and greater dignity. Although the studies and their comparison present limitations, this review highlights a series of socio-demographic, psycho-behavioral, health-related, and nutritional factors that assist the identification of more vulnerable groups and that are amenable to intervention.

Importantly, although demographic determinants showed a relationship to frailty in the elderly, some determinants are not subject to changes and interventions. For example, it is impossible to alter age or gender, but they should be considered anyway, since various health conditions increase with age and occur differently between men and women.

Thus, planning of individual and collective health measures for the elderly should consider the factors identified here as related to the frailty syndrome, such as: age, black race/color, female gender, CVD, number of comorbidities/diseases, functional incapacity, poor self-rated health, depressive symptoms, BMI, smoking, schooling, income, cognitive function, and alcohol consumption (the latter with caution). It is also important to investigate other factors not explored in this review, besides conducting meta-analyses aimed at a critical assessment of the evidence and a discussion of the possible heterogeneity of results, in addition to an analysis of the strength of available evidence on the association found between a given variable and frailty in order to better understand how the way of living can interfere in the way of aging and favor the establishment of this syndrome.

Acknowledgments

The authors wish to thank librarian Gizele da Rocha Ribeiro for her outstanding contribution to defining the search strategies and Capes for granting the PhD scholarship to A. C. Mello.

References

Publication Dates

History