Prevalence of and risk factors for dynapenic abdominal obesity in community-dwelling older adults: a cross-sectional study

This article aims to identify the prevalence of and factors associated with dynapenic abdominal obesity (DAO) in older adults in a city in the northern region of Brazil. A cross-sectional study was conducted with 382 community-dwell-ing older adults in Macapa, Amapa, Brazil. Socioeconomic, clinical, and health information were collected using a structured form. DAO was defined as a combination of dynapenia (hand-grip strength of < 26 kgf for men and < 16 kgf for women) and abdominal obesity (abdominal circumference > 102 cm for men and > 88 cm for women). Descriptive and inferential analyses were performed using chi-squared tests, Student’s t-tests, and a Poisson regression. The prevalence of DAO was 10.73%. In the preliminary bivariate analysis, the variables of age range, marital status, number of diseases, functional disability for basic and instrumental activities of daily living, gait speed, and level of physical activity met the established criterion. The final model indicated that only gait speed was a predictor of DAO in older adults. DAO affects nearly 11% of commu-nity-dwelling older adults from northern Brazil; gait speed was a predictor of DAO and could be a useful tool for managing and monitoring this population’s health.


temAs livRes fRee themes introduction
The aging process is marked by several physiological changes that lead to the progressive decline of all organic functions 1 . Despite the generalized decline, specific changes -such as those related to body composition -deserve particular study from clinical and public health professionals. Age-related loss of muscle mass is associated with reductions in muscle strength and physical performance 2 and poor outcomes, such as falls and fractures 3 . However, a reduction in muscle mass alone cannot fully explain losses of muscle strength 4 ; the term "dynapenia" has been recently coined to describe the aging-associated loss of muscle strength as an independent event.
Another salient change in body composition that humans experience during the aging process is the rise in body fat mass. This condition is also associated with a higher risk of comorbidities, disability 5 , and metabolic diseases 6 . However, fat accumulated in the abdominal area is associated with more severe impacts on metabolic disorders than overall obesity [7][8][9] . Both dynapenia and abdominal obesity are important predictors of poor outcomes, but the confluence of both is a downward spiral that may lead to disability 10 and multimorbidity 11 . Indeed, dynapenia and abdominal obesity together may better predict functional disability than each phenomenon individually 10 .
Therefore, this recently recognized condition has been termed "dynapenic abdominal obesity" (DAO) 10,12 . DAO represents a severe public health problem, as it is associated with several health burden indicators and healthcare demands. The condition is not only associated with functional disability 10 , but also with an increased risk of falling 12 , hospitalizations 13 , and death 14 .
Although the consequences of DAO to older adults may be devastating, it has been little investigated. Some studies have examined the association of DAO with specific conditions, such as functional disability 10 , falls 12 , hospitalizations 13 , and death 14 . However, to our knowledge, no previous studies have evaluated multiple risk factors -socioeconomic, clinical, and health -and their relationships with the presence of DAO. Furthermore, Brazil is a country with continental dimensions and important socioeconomic, structural, and cultural differences. Despite that, few studies have looked at the elderly population from northern Brazil 15,16 . To address this gap, to identify the prevalence of and factors associated with DAO in older adults in a city in the northern region of Brazil.

method
This population-based cross-sectional study was carried out in 2017 in the urban area of Macapa, the capital of the state of Amapa, in Brazil. Macapá is a Brazilian city located on the banks of the Amazon river, in northern Brazil. The city (latitude -0° 2'20» N; longitude -51° 3' 59» W) is the only Brazilian state that has its capital cut by the imaginary line of the Equator 17 . With a Human Development Index (HDI) of 0.733 and a life expectancy at birth of 74.2 years 18 , Macapá in the year 2010 had an estimated population of 456,171 inhabitants, of which 5.21% (96% urban and 4% rural) were aged 60 years or older 19 . An appropriate sample size was calculated, assuming a prevalence of 50% of health problems among a finite population of 19,955 older adults living in the urban area of Macapa and setting an accuracy level of 5% and a 95% confidence interval (CI). A minimum of 377 individuals was found to be necessary to be representative of the population. To select older adults' residences, the research team carried out a two-stage cluster sampling process by first sampling government-defined pre-specified census tracts and subsequently identifying residences with older people. Select information from this study has been previously analyzed and is available in previously published studies 17,20 .
Adults aged 60 years or over who did not have cognitive decline measured on the Mini-Mental State Examination 21 and were able to walk with or without walking assistance devices were included. Older adults who were not found after three visits, were institutionalized and/or hospitalized, or affected by neurological diseases that made evaluations impossible were excluded from the study. This study was approved by the Ethics Committee of the Federal University of Amapá (opinion 1,738,671). All the participants agreed to participate in the research by signing an Informed Consent Form. The individuals were recruited and assessed at their respective homes. All interviews were conducted face-to-face by properly trained undergraduate students and monitored by field supervisors (researcher teachers).

exploratory variables
A structured questionnaire was used to collect data on all socioeconomic variables: gender, age, education level, physical health variables; referred morbidities, the number of medications in regular use, falls in the past 12 months (yes/ no), and hospitalizations in the last year (yes/ no). Self-reported morbidities were surveyed using an inventory 22 in which participants reported whether any physician had informed if they had certain health condition. Regularly used medications were verified by checking medical prescriptions and by directly asking participants about the medications they were taking. Physical activity levels were evaluated with the long version of the International Physical Activity Questionnaire (IPAQ) 23 . As suggested by the College of Sports Medicine and the American Heart Association, participants who spent 150 minutes or more per week on physical activities were considered sufficiently active; those who spent less than 150 minutes per week on physical activities were classified as insufficiently active 24 . The Katz Index 25 and the Lawton and Brody Scale 26 were used to assess functional impairment for basic and instrumental activities of daily living, respectively. Finally, a gait speed test, or measuring the time to walk a four-meter distance, was conducted following the Brazilian version of the Short Physical Performance Battery (SPPB) 27 .

Dependent variable
Dynapenic abdominal obesity was measured with both muscle strength and abdominal obesity. Muscle strength was measured using a hydraulic hand dynamometer, SAEHAN Hydraulic Hand Dynamometer, model SH5001. The participants were first instructed to sit with their forearm in a neutral position. The device was placed in the participants' dominant hand, and three maximum strength tests were performed with a one-minute rest between tests. The average of these measurements was recorded for analysis. Values below 26 kilogram-force (kgf) for men and below 16 kgf for women were considered suggestive of reduced muscle strength 28 . The second component of DAO, waist circumference, was measured using a non-elastic tape, placed at the midpoint between the last rib and iliac crest, with the abdomen relaxed at the end of exhalation. Standardized measurements were taken with all participants in a standing position, with their arms relaxed at their sides, and clothing removed from the abdominal region 29 . Cut-off points of abdominal circumference > 102 cm for men and > 88 cm for women were used to classify abdominal obesity 29 . Participants who fulfilled both the loss of muscle strength and abdominal obesity criteria were considered to have DAO.

Data analysis
The data were analyzed in version 21.0 of the Statistical Package for Social Sciences (SPSS) program and using both descriptive and inferential statistics. Differences between the participants characteristics were analyzed using means, standard deviations, chi-square tests, and Student's t-tests. The tests were considered significant when p < 0.20. The variables that met the pre-established inclusion criterion (p < 0.20) were inserted in the multivariate regression model. The factors associated with DAO were identified from the prevalence ratios (PR) using a Poisson regression, considering a significance level of 5% (p < 0.05) and 95% confidence interval (CI).

Results
Three hundred eighty-two older adults participated in this study. The number of recruited, assessed, included, and excluded participants are outlined in Figure 1. The prevalence of abdominal obesity and dynapenia in the sample was 65.7% (n = 251) and 18.6% (n = 71), respectively, and the prevalence of DAO was 10.73% (CI = 08.01-14.24) (n = 41). Older people with DAO had a higher age range, number of diseases, without a partner, functional disability for basic activities of daily living, insufficient level of physical activity, and lower gait speed compared to those without DAO (Table 1).
Age range, marital status, number of diseases, functional disability for basic and instrumental activities of daily living, level of physical activity, and gait speed all met the pre-established criterion in the crude analysis (p < 0.20). The final model indicated that only gait speed remained associated with DAO (PR = 0.17; 95% CI: 0.05-0.54; p = 0.002), suggesting that an increase in one unit of gait speed may decrease the probability of occurrence of this condition by 83% in older adults (Table 2).

Discussion
This study in a representative sample of community-dwelling older adults from northern Brazil identified that DAO is a pervasive condition in this population. To the best of our knowledge, this is the first study to evaluate multiple risk factors associated with the presence of DAO and provides important insights upon which to base future studies on the subject. The findings presented here may also help formulate strategies for screening patients at higher risk of developing DAO and should be taken into account when designing intervention strategies to preventing the worsening of conditions that could lead to DAO in older adults, such as those focused on improving gait speed. The final model indicated that only gait speed remained a significant predictor of this condition.
In line with a previous study that found a 9.3% prevalence of DAO in Brazilian older adults 12 , the prevalence of DAO in our sample was 10.7%. Nevertheless, a wide variation in the prevalence of DAO has been found worldwide, with estimations ranging from 4 to 18.8% 9,10,12,14 . The variation in the prevalence may be due to the differences between the studied samples and the population characteristics. A previous study that evaluated older adults from the Amazon region found that these individuals engage in high levels of physical activity 15 . Further studies may clarify the influence of physical activity levels in older adults from the Amazon region and the prevalence of DAO. Additionally, as the concept of DAO is relatively new 30 , no consensus on the diagnostic procedures has been reached yet. There-fore, a possible explanation for this variability in DAO prevalence is that DAO may vary according to the setting and the different studies' operational definition.
The results of our preliminary analysis are also in accordance with previous studies that have found that DAO is associated with age, the number of comorbidities, functional disability for basic and instrumental activities of daily living, level of physical activity, and gait speed 13 . Nevertheless, in our sample, only gait speed remained inversely associated with DAO in the final model.
The reduced gait performance in older people with DAO compared to those without this condition is supported by previous investigations 31,32 . In a cross-sectional study, dynapenic obesity was associated with an increased risk of slow gait speed 32 . Another study highlights that concomitant obesity and low muscle strength may lead not only to a decline in gait speed but also to an increase in the risk of functional disability 31 . Therefore, gait speed limitations can have a substantial impact on autonomy and endurance during activities of daily living. Furthermore, gait speed represents an independent predictor of other adverse health outcomes such as cognitive impairment, institutionalization, falls, and death 33 .   Of note, reduced leg strength 34 and a high fat mass 35 are modifiable risk factors that, if not early managed, can lead to potentially-disabling conditions and a potentialized burden in older adults' lives. Targeted health care interventions such as exercise and nutrition may help prevent unfavorable health outcomes in this population.
Taken together, our findings reinforce the relevance of the gait speed assessment as a useful tool for managing and monitoring this population's health. Importantly, the cross-sectional design of this study does not allow us to ascertain a causal relationship between DAO and slow gait speed. Further longitudinal studies may help clarify the temporal sequence of those events. Additionally, although the IPAQ was developed for international use and has evidence of reliability and validity 36 , as it is a self-reported instrument some older adults older adults may not fully comprehend the amount of time they may spend active, sitting and/or lying during waking hours, especially those with lower educational level. On the other hand, this study made up of a representative sample of community-dwelling older adults the final model indicated that gait speed is a central element to be considered while evaluating older adults with DAO. It is a simple, non-invasive, low-cost, and easy-to-perform assessment that can help identify older adults at higher risk of adverse health outcomes 33 , help implement effective interventions, and prevent further deterioration in quality of life.

Conclusion
DAO affects nearly 11% of community-dwelling older adults from northern Brazil. The final model revealed that only gait speed remained associated with DAO, indicating that a 1 unit increase in gait speed may decrease the likelihood of occurrence of DAO by 83% in older adults. Gait speed could be a useful tool for managing and monitoring this population's health.

Collaborations
All the authors contributed to the conception and design, analysis and interpretation of the research data, as well as to the writing and critical review of the article. All authors read and approved the final version of the manuscript for publication.