Changes in anthropometric indicators and gait speed in older adults: cohort study

Abstract

Objective

To characterize changes in anthropometric indicators in older adults and investigate whether being overweight was associated with lower gait speed (GS), based on measurements taken at an interval of nine years.

Methods

Cohort study with older adults (≥65 years), conducted in 2008-2009 (baseline) and 2016-2017 (follow-up) in the city of Campinas/SP and in Ermelino Matarazzo/SP, Brazil. Body weight, height, waist circumference (WC) and hip (HC) measurements were taken and used to determine the following indicators: body mass index (BMI), waist-to-height ratio (WHtR), waist-to-hip ratio (WHR) and conicity index (C index). The T and Wilcoxon tests for paired samples were used to estimate the differences.

Results

Information from 537 older adults (70.0% women) with a mean age of 72.2 years at baseline and 80.7 years at follow-up were analyzed. After nine years, the men showed significant decreases in weight, height and BMI, and an increase in the C index. In women, decreases in weight, height and BMI, and increases in WC, HC, WHtR, WHR and C index were observed. The percentage variations observed were: -3.89% (weight), -0.36% (height), -4.18% (BMI) and +2.27% (C index) among men; -2.95% (weight), -0.65% (height), -0.73% (BMI), +3.33% (WC), +1.59% (HC), +3.45% (WHtR), +2.27% (WHR) and +4.76% (C-Index) among women. Being overweight was associated with greater odds ratio of stability and new cases of lower GS at follow-up.

Conclusion

Changes were identified in weight, height, BMI, and indicators of abdominal obesity, especially in women, together with an association between being overweight and lower GS.

Keywords
Aged; Anthropometry; Body Composition; Gait; Obesity, Abdominal; Longitudinal Studies

Resumo

Objetivo

Caracterizar mudanças em indicadores antropométricos em idosos e investigar se o excesso de peso associou-se com menor velocidade da marcha (VM), com base em medidas realizadas a um intervalo de nove anos.

Métodos

Estudo de coorte com idosos (≥65 anos), realizado em 2008-2009 (linha de base) e 2016-2017 (seguimento) em Campinas/SP e Ermelino Matarazzo/SP, Brasil. Foram aferidas medidas de peso corporal, estatura, circunferência da cintura (CC) e do quadril (CQ), usadas para obter os indicadores: índice de massa corporal (IMC), razão cintura-estatura (RCE), razão cintura-quadril (RCQ) e índice de conicidade (Índice C). Os testes T e de Wilcoxon para amostras pareadas foram usados para estimar as diferenças.

Resultados

Foram analisadas informações de 537 idosos (70,0% mulheres) com idade média de 72,2 anos na linha de base e 80,7 anos no seguimento. Após nove anos, os homens apresentaram reduções significativas do peso corporal, estatura e IMC, e aumento do Índice C. Nas mulheres, observou-se declínio do peso, estatura e IMC, e elevação da CC, CQ, RCE, RCQ e Índice C. Observaram-se variações percentuais de: -3,89% (peso), -0,36% (estatura), -4,18% (IMC) e +2,27% (Índice C) nos homens; -2,95% (peso), -0,65% (estatura), -0,73% (IMC), +3,33% (CC), +1,59% (CQ), +3,45% (RCE), +2,27% (RCQ) e +4,76% (Índice C) nas mulheres. O excesso de peso associou-se com maiores chances de estabilidade e de novos casos de menor VM no seguimento.

Conclusão

Foram identificadas mudanças no peso, estatura, IMC, nos indicadores de obesidade abdominal, especialmente nas mulheres, e associação entre excesso de peso e menor VM.

Palavras-Chave:
Idoso; Antropometria; Composição Corporal; Marcha; Obesidade Abdominal; Estudos Longitudinais

Introduction

The aging process, or senescence, is associated with changes in body composition that include a reduction in muscle and bone tissue, and an increase in and redistribution of adipose tissue¹,². The loss of muscle tissue causes a decrease in the basal metabolic rate, predisposing older adults to weight gain¹,², together with higher incidence of chronic non-communicable diseases (CNCDs), regardless of age, sex and body composition³.

Loss of muscle mass loss and increased fat mass heighten the risk of mortality⁴-⁶ and produce negative effects on health and quality of life, including a decline in gait speed⁷,⁸ and functional capacity⁶,⁹-¹¹, higher occurrence of falls⁶,¹¹, frailty¹¹-¹³ and CNCDs⁶. A follow-up study involving North American older adults showed a higher incidence of mobility limitation (difficulty walking or climbing stairs) among overweight or obese men and women at 25, 50 and 70 to 79 years of age, compared with those who maintained a healthy weight⁹. A meta-analysis with data from two cohorts conducted on older adults in Spain detected a higher risk of frailty among obese individuals, higher scores in the fatigue criteria, low levels of physical activity and low handgrip strength¹³.

Excess visceral adipose tissue and ectopic fat deposits (liver, pancreas, heart, musculoskeletal system, and bone marrow) increase the production of inflammatory cytokines and reduce the production of adiponectin, a protein that has an anti-inflammatory, antidiabetic and antiatherogenic role¹⁴. In old age, the activation of the innate immune system triggers a low-grade chronic inflammatory process called inflammaging, which accelerates the development of chronic diseases and loss of muscle mass¹⁴,¹⁵.

There are several anthropometric indicators considered practical, inexpensive and that show good reliability in the assessment of body composition, such as body mass index (BMI), waist circumference (WC) and waist-to-hip ratio (WHR) that are widely used, in addition to others like the waist-to-height ratio (WHtR) and the conicity index (C index), which are rarely used in clinical practice and in population studies. Since it adjusts for height, WHtR is better than WC at detecting cardiovascular diseases, diabetes, arterial hypertension and dyslipidemia in men and women¹⁶. The C index comes from measurements of weight, height and WC, and is based on changes in body design – from the shape of a cylinder to a double cone (two cones with a common base) – due to the concentration of fat in the abdomen¹⁷. With aging, the redistribution of adipose tissue and its accumulation in the abdominal region affect the ability of these indicators to classify older adults with excess adiposity⁶,¹⁵. BMI does not assess the distribution of body fat, especially that deposited in the visceral region, which makes it less accurate for detecting increased cardiometabolic risk than the other indicators mentioned¹⁸,¹⁹.

The Frailty Profile of Elderly Brazilians (FIBRA Study) is a multicenter, population-based survey that was developed in 2008-2009 in 17 cities located in all five geographic regions of Brazil, selected by criteria of convenience. It aimed to characterize frailty profiles in adults aged 65 years old and over, considering a profusion of instruments and variables. One of the consequences of this research was a follow-up study, in 2016-2017, involving older adults from the initial study who were still alive and residing in Campinas/SP and Ermelino Matarazzo/SP. In the follow-up survey, the sociodemographic, anthropometric, frailty phenotype and mental status variables collected in the initial survey were repeated.

The literature provides accumulated evidence on the nutritional status of older adult populations and associated factors. In contrast, there are few national studies that analyze changes in body composition and associations with adverse health outcomes, particularly in a sample with a considerable portion of adults aged 80 years old and over.

The aim of this study was to characterize changes in anthropometric indicators in older adults and to investigate whether being overweight is associated with lower gait speed, based on measurements taken at an interval of nine years.

Methods

This is a multicenter, populational cohort study conducted using data from the FIBRA Study. Data collection originally took place in 2008-2009, in cities chosen for convenience in the five Brazilian geographic regions, which were gathered in poles coordinated by four public universities, including the State University of Campinas and its survey of seven cities. In each one, a representative sample of the urban population of older adults aged 65 years and over²⁰ was selected. In 2016-2017, Campinas/SP and Ermelino Matarazzo, a district of the city of São Paulo, conducted a cohort study involving older adult who had participated in the initial study and who still resided there, and the data obtained were analyzed in this research.

In 2008-2009 (baseline), 90 urban census sectors were randomly selected in Campinas and 62 in Ermelino Matarazzo. All households in the selected sectors were visited to identify the presence of older adults who met the inclusion criteria: 65 years of age or older, agreeing to participate in the research, residing in the household, and presenting sufficient independence and autonomy, and sensory, psychomotor, language, and comprehension abilities. The study excluded older adults who were bedridden, those with terminal disease or neoplasia (except for the skin), severe sensory or cognitive problems, aphasia or neurological diseases with signs of aggravation²⁰.

Recruited from households and flow points, the older adults were invited to attend public places, in easily accessible areas, for a data collection session. Recruitment at flow points, places of confluence for older adults located in the selected census sectors, was the except and was used when households were difficult to access. Recruitment was carried out until the quotas of men and women by age group (65 to 69, 70 to 74, 75 to 79 and ≥80 years old) were completed in proportions compatible with the census distribution of the same in the selected sectors, having anticipated possible losses or refusals²⁰.

In 2016-2017, a follow-up study was conducted involving the older adult participants at baseline. The addresses registered in the Campinas and Ermelino Matarazzo databases served as a basis for locating these older adults. Recruitment and data collection were carried out at home by graduate students in gerontology and undergraduate students in medicine, organized in pairs. Up to three attempts were made to find each older adult.

For both time points of the study, body weight, height, and waist (WC) and hip (HC) circumference were measured. Weight was measured with a portable electronic scale, with the older adult standing erect on the equipment platform, facing the scale, with their eyes fixed forward, feet parallel and barefoot, while wearing light-weight clothes. For height, a portable stadiometer was used and the older adults stood upright, with their backs to the scale, barefoot and feet together, with their heads positioned in the Frankfurt Plane. WC was verified at the midpoint between the lower edge of the last rib and the iliac crest, with the individual standing and the waist region naked. HC was measured in the area with the greatest volume of the buttocks, with the older adult standing and wearing clothes below the buttocks²¹,²².

The following anthropometric indicators were calculated:

- Body Mass Index (BMI): [weight (kg)/height (m²)].

- Waist-to-height ratio (WHtR): [waist circumference (cm)/height (cm)].

- Waist-to-hip ratio (WHR): [waist circumference (cm)/hip circumference (cm)].

- Conicity Index (C Index):

$$\frac{\text{waist circumference (m)}}{0.109\sqrt{\frac{\text{body weight (kg)}}{\text{height (m)}}}}$$

The anthropometric variables and indicators were presented according to sex and age group at baseline (65-69, 70-74 and 75 years old or over) and at follow-up (72-79, 80-84 and 85 years old or over).

The usual gait speed (GS) was evaluated by the time in seconds it took the older adult to walk a distance of 4 meters on a flat floor. Three attempts were made, allowing the use of a walking stick or walker. The average travel time was calculated. The cut-off point ≤0.8 m/s was used to identify older adults who presented slow gait²³. Next, a dichotomous variable was created that reflects stability or change in GS from baseline to follow-up, composed of: older adults with higher GS (>0.8m/s) at baseline and follow-up or who began to present higher GS at follow-up; lower GS (≤0.8m/s) at the two time periods or who began to present lower GS at follow-up.

Being overweight was identified from the anthropometric variables and respective cut-off points:

- WC: ≥96.0 cm for men and ≥88.7 cm for women²⁴.

- WHtR: ≥0.58 for both sexes²⁴.

- BMI: ≥27 kg/m^{2 21}.

- WHR: >1.0 for men and >0.85 for women²⁵.

- C index: ≥1.25 for men and ≥1.18 for women²⁶.

The cut-off points used for WC, WHtR and BMI were defined for older adults, while cut-off points for WHR were defined for adults and for the C index were defined for adults aged 30 to 74 years.

Data analysis used descriptive statistics (mean, standard deviation, median and interquartile distance) for the variables considered at baseline and at follow-up, according to sex. To assess the differences between the measurements studied during the period, the normality of the distribution of variables was initially verified using the Shapiro-Wilk statistical test. Thus, the appropriate statistical tests were used – Student’s t test for paired samples, and the non-parametric Wilcoxon test – considering a significance level lower than 5%. The percentage changes in measurements and anthropometric indicators in older adults were also calculated between baseline and follow-up for both sexes.

Next, the incidences (%) of lower GS according to being overweight at baseline were estimated, and the associations were verified using Pearson’s chi-square test (p<0.05). Logistic regression adjusted for sex and age was used to obtain the odds ratios (OR) and respective 95% confidence intervals (95%CI) of slow gait, and associations with being overweight were determined by the Wald test, p<0.05.

The FIBRA Study projects were approved by the Ethics Committees of the Campinas State University (report 1,332,651, CAAE 49987615.3.0000.5404) and the University of São Paulo (report 2,952,507, CAAE 92684517.5.3001.5390). All participants signed a term of free, informed consent.

Results

At baseline, 1,284 older adults composed the sample, 900 in Campinas and 384 in Ermelino Matarazzo. At follow-up, only 549 older adult participants remained, 192 were deceased and 543 could not be located. Regarding the baseline samples, the losses (older adults not located or who refused to participate) represented 41.9% in Campinas and 43.2% in Ermelino Matarazzo.

Among the 549 older adults interviewed at baseline and at follow-up, 12 were excluded due to the lack of complete data on anthropometric measurements in both periods. Thus, data from 537 older adults were analyzed in this study (Figure 1).

Figure 1
Flowchart of the sample used in this research. FIBRA study, Older adults, Campinas and Ermelino Matarazzo, SP, Brazil

Data were analyzed from 537 older adults whose weight, height, WC and HC were measured in 2008-2009 and 2016-2017. Women represented 70.0% of the sample evaluated in both survey time periods, and the mean age was 72.2 years (±5.2) at baseline and 80.7 years (±4.8) at follow-up. Mean GS was 0.43 m/s (±0.49) for the group of older adults at baseline, and 0.81 m/s (±0.39) at follow-up.

For men, weight and height were normally distributed (p>0.05). All other variables for both men and women did not show normal distribution. Between baseline and follow-up, among men, decreases in the average weight and height, in the median BMI, and an increase in the median C index were observed (Table 1).

Table 1
Means and medians of anthropometric variables in older adult males, according to age (n=161). FIBRA Study, Older adults, Campinas and Ermelino Matarazzo, SP, Brazil, 2008-2009 and 2016-2017.

Variables by age group	n (%)	Mean (standard deviation)	Median (interquartile distance)
Weight (kg) – baseline
65-69	48 (29.8)	77.8 ± 14.2	76.7 (16.4)
70-74	68 (42.3)	73.9 ± 11.2	74.2 (13.5)
≥ 75	45 (27.9)	72.1 ± 10.2	72.7 (11.6)
Total	161	74.6 ± 12.1	75.0 (13.5)
Weight (kg) – follow-up
72-79	57 (35.4)	74.0 ± 15.2	75.8 (20.1)
80-84	70 (43.5)	72.5 ± 13.2	72.3 (18.9)
≥ 85	34 (21.1)	66.0 ± 8.8	65.1 (10.4)
Total	161	71.7 ± 13.4	71.1 (18.4)
p value (difference: baseline – follow-up)			<0.001 a
Height (cm) – baseline
65-69	48	168.1 ± 6.9	168.0 (11.5)
70-74	68	167.3 ± 6.0	167.5 (9.5)
≥ 75	45	165.9 ± 5.2	166.0 (7.0)
Total	161	167.1 ± 6.1	167.0 (9.0)
Height (cm) – follow-up
72-79	57	167.8 ± 7.1	168.0 (10.0)
80-84	70	165.5 ± 6.3	165.7 (9.0)
≥ 85	34	166.2 ± 6.0	166.5 (7.0)
Total	161	166.5 ± 6.6	166.0 (8.0)
p value (difference: baseline – follow-up)			0.038 a
Waist circumference (cm) – baseline
65-69	48	96.6 ± 12.8	96.0 (16.7)
70-74	68	95.9 ± 9.9	95.5 (11.7)
≥ 75	45	93.7 ± 11.3	94.0 (15.0)
Total	161	95.5 ± 11.2	95.0 (16.0)
Waist circumference (cm) – follow-up
72-79	57	97.3 ± 12.3	97.0 (14.0)
80-84	70	96.0 ± 13.3	94.5 (19.0)
≥ 85	34	92.3 ± 8.0	94.0 (11.0)
Total	161	95.7 ± 12.1	96.0 (15.5)
p value (difference: baseline – follow-up)			0.402b
Hip circumference (cm) – baseline
65-69	48	98.1 ± 9.5	98.5 (12.2)
70-74	68	98.5 ± 7.6	97.5 (9.7)
≥ 75	45	97.4 ± 8.0	98.0 (11.0)
Total	161	98.1 ± 8.3	98.0 (10.5)
Hip circumference (cm) – follow-up
72-79	57	99.4 ± 10.0	99.0 (13.0)
80-84	70	99.3 ± 7.9	98.0 (9.0)
≥ 85	34	97.3 ± 5.7	98.5 (8.0)
Total	161	98.9 ± 8.3	98.0 (9.0)
p value (difference: baseline to follow-up)			0.121b
Body mass index (kg/m2) – baseline
65-69	48	27.5 ± 4.9	27.2 (5.8)
70-74	68	26.4 ± 3.7	26.2 (5.4)
≥ 75	45	26.2 ± 3.7	26.0 (4.4)
Total	161	26.7 ± 4.1	26.3 (5.4)
Body mass index (kg/m2) – follow-up
72-79	57	26.2 ± 5.2	25.8 (6.2)
80-84	70	26.4 ± 4.4	25.9 (6.7)
≥ 85	34	23.9 ± 3.4	23.6 (4.3)
Total	161	25.8 ± 4.6	25.2 (6.2)
p value (difference: baseline – follow-up)			<0.001 b
Waist-to-height ratio – baseline
65-69	48	0.57 ± 0.08	0.57 (0.10)
70-74	68	0.57 ± 0.06	0.57 (0.09)
≥ 75	45	0.56 ± 0.07	0.55 (0.10)
Total	161	0.57 ± 0.07	0.57 (0.09)
Waist-to-height ratio – follow-up
72-79	57	0.58 ± 0.08	0.57 (0.09)
80-84	70	0.58 ± 0.08	0.57 (0.10)
≥ 85	34	0.56 ± 0.05	0.55 (0.07)
Total	161	0.57 ± 0.07	0.57 (0.09)
p value (difference: baseline – follow-up)			0.180b
Waist-to-hip ratio – baseline
65-69	48	0.98 ± 0.07	1.00 (0.11)
70-74	68	0.97 ± 0.06	0.97 (0.09)
≥ 75	45	0.96 ± 0.08	0.96 (0.11)
Total	161	0.97 ± 0.07	0.98 (0.10)
Waist-to-hip ratio – follow-up
72-79	57	0.98 ± 0.06	0.98 (0.07)
80-84	70	0.96 ± 0.08	0.97 (0.09)
≥ 85	34	0.95 ± 0.06	0.95 (0.07)
Total	161	0.96 ± 0.07	0.97 (0.08)
p value (difference: baseline – follow-up)			0.349b
Conicity index – baseline
65-69	48	1.30 ± 0.07	1.31 (0.10)
70-74	68	1.33 ± 0.06	1.32 (0.07)
≥ 75	45	1.30 ± 0.10	1.30 (0.13)
Total	161	1.31 ± 0.08	1.32 (0.10)
Conicity index – follow-up
72-79	57	1.35 ± 0.09	1.35 (0.09)
80-84	70	1.33 ± 0.10	1.34 (0.11)
≥ 85	34	1.35 ± 0.07	1.35 (0.08)
Total	161	1.34 ± 0.09	1.35 (0.10)
p value (difference: baseline – follow-up)			<0.001 b

• ^a p-value, paired T test; ^b p-value, paired Wilcoxon test.

For women, decreases in the median weight, height and BMI, and increases in the median WC, HC, WHtR, WHR and C index were observed (Table 2).

Table 2
Means and medians of anthropometric variables in older adult females, according to age (n=376). FIBRA Study, Older adults, Campinas and Ermelino Matarazzo, SP, Brazil, 2008-2009 and 2016-2017.

Variables by age group	n (%)	Mean (standard deviation)	Median (interquartile distance)
Weight (kg) – baseline
65-69	143 (38.0)	67.6 ± 11.0	66.1 (13.4)
70-74	117 (31.1)	66.4 ± 12.2	65.1 (15.1)
≥ 75	116 (30.9)	63.5 ± 12.2	62.1 (15.7)
Total	376	65.9 ± 11.9	64.4 (14.9)
Weight (kg) – follow-up
72-79	164 (43.6)	66.7 ± 11.0	66.9 (12.9)
80-84	135 (35.9)	64.0 ± 14.1	61.7 (17.3)
≥ 85	77 (20.5)	58.5 ± 10.0	57.5 (13.6)
Total	376	64.1 ± 12.4	62.5 (15.4)
p value (difference: baseline – follow-up)			<0.001 a
Height (cm) – baseline
65-69	143	155.1 ± 6.8	155.0 (10.0)
70-74	117	154.0 ± 6.1	154.0 (8.0)
≥ 75	116	152.6 ± 6.7	153.0 (7.7)
Total	376	154.0 ± 6.6	154.0 (8.0)
Height (cm) – follow-up
72-79	164	153.8 ± 6.5	154.0 (9.0)
80-84	135	152.5 ± 6.5	153.0 (9.0)
≥ 85	77	149.7 ± 7.7	151.0 (8.0)
Total	376	152.5 ± 6.9	153.0 (9.0)
p value (difference: baseline – follow-up)			<0.001 a
Waist circumference (cm) – baseline
65-69	143	90.4 ± 11.1	90.0 (13.5)
70-74	117	89.9 ± 11.3	89.0 14.5)
≥ 75	116	87.9 ± 12.9	88.5 (16.2)
Total	376	89.5 ± 11.7	90.0 (14.5)
Waist circumference (cm) – follow-up
72-79	164	94.7 ± 11.4	93.2 (14.0)
80-84	135	93.7 ± 15.0	93.0 (19.0)
≥ 85	77	88.5 ± 11.4	90.0 (18.0)
Total	376	93.1 ± 13.0	93.0 (16.7)
p value (difference: baseline – follow-up)			<0.001 a
Hip circumference (cm) – baseline
65-69	143	101.9 ± 9.3	100.0 (12.0)
70-74	117	102.0 ± 9.7	101.0 (11.5)
≥ 75	116	101.1 ± 9.7	100.7 (11.4)
Total	376	101.7 ± 9.5	100.4 (11.0)
Hip circumference (cm) – follow-up
72-79	164	103.9 ± 11.7	103.0 (13.5)
80-84	135	103.4 ± 12.3	101.5 (15.0)
≥ 85	77	98.1 ± 9.2	98.0 (13.0)
Total	376	102.5 ± 11.7	102.0 (13.7)
p value (difference: baseline to follow-up)			0.036 a
Body mass index (kg/m2) – baseline
65-69	143	28.1 ± 4.5	27.2 (6.0)
70-74	117	27.9 ± 4.5	27.8 (5.5)
≥ 75	116	27.2 ± 4.8	27.1 (6.8)
Total	376	27.8 ± 4.6	27.3 (5.9)
Body mass index (kg/m2) – follow-up
72-79	164	28.3 ± 4.7	27.6 (5.9)
80-84	135	27.5 ± 5.5	27.2 (6.9)
≥ 85	77	26.3 ± 5.3	25.6 (6.4)
Total	376	27.6 ± 5.2	27.1 (6.5)
p value (difference: baseline – follow-up)			0.041a
Waist-to-height ratio – baseline
65-69	143	0.58 ± 0.08	0.58 (0.08)
70-74	117	0.58 ± 0.07	0.58 (0.10)
≥ 75	116	0.58 ± 0.09	0.58 (0.11)
Total	376	0.58 ± 0.08	0.58 (0.10)
Waist-to-height ratio – follow-up
72-79	164	0.62 ± 0.08	0.60 (0.10)
80-84	135	0.61 ± 0.10	0.62 (0.12)
≥ 85	77	0.59 ± 0.09	0.59 (0.12)
Total	376	0.61 ± 0.09	0.60 (0.11)
p value (difference: baseline – follow-up)			<0.001 a
Waist-to-hip ratio – baseline
65-69	143	0.89 ± 0.07	0.89 (0.10)
70-74	117	0.88 ± 0.08	0.87 (0.10)
≥ 75	116	0.87 ± 0.08	0.87 (0.10)
Total	376	0.88 ± 0.08	0.88 (0.10)
Waist-to-hip ratio – follow-up
72-79	164	0.91 ± 0.10	0.91 (0.11)
80-84	135	0.90 ± 0.10	0.90 (0.11)
≥ 85	77	0.90 ± 0.08	0.90 (0.11)
Total	376	0.91 ± 0.10	0.90 (0.11)
p value (difference: baseline – follow-up)			<0.001 a
Conicity index – baseline
65-69	143	1.26 ± 0.09	1.26 (0.11)
70-74	117	1.26 ± 0.10	1.27 (0.13)
≥ 75	116	1.25 ± 0.11	1.26 (0.13)
Total	376	1.26 ± 0.10	1.26 (0.12)
Conicity index – follow-up
72-79	164	1.32 ± 0.09	1.32 (0.12)
80-84	135	1.33 ± 0.13	1.32 (0.13)
≥ 85	77	1.30 ± 0.11	1.30 (0.16)
Total	376	1.32 ± 0.11	1.32 (0.14)
p value (difference: baseline – follow-up)			<0.001a

• ^a p-value, paired Wilcoxon test.

Figure 2 shows the percentage change in anthropometric measurements after nine years, according to sex. Among men, decreases in weight (-3.89%), height (-0.36%) and BMI (-4.18%) were observed. Only the C index showed a positive change (+2.27%). Among women, decreases in weight (-2.95%), height (-0.65%) and BMI (-0.73%) were observed, while the remaining measurements and indicators showed increases: WC (+3.33%), HC (+1.59%), WHtR (+3.45%), WHR (+2.27%) and C index (+4.76%).

Figure 2
Percentage variation in measurements and anthropometric indicators in older adults, between baseline and follow-up. FIBRA study, Older adults, Campinas and Ermelino Matarazzo, SP, Brazil

There were no significant differences between the sexes in the incidence of gait stability or occurrence of slower gait between baseline and follow-up. In contrast, among adults aged 75 years old and over, the incidence of slower gait was 2.6 times higher compared with those 60 to 69 years old. As determined by the anthropometric measurements WC, BMI, WHtR and WHR, being overweight increased the chances of older adults presenting gait stability or a slower gait after nine years (Table 3).

Table 3
Incidence of lower gait speed in older adults, according to sex, age and overweight. FIBRA study, Older adults, Campinas e Ermelino Matarazzo, SP, Brazil, 2008-2009 and 2016-2017.

Variable	Incidence %	Adjusted OR b (IC95%)	p value c
Sex	p= 0.139 a
Male	77.2	1.00
Female	82.7	1.51 (0.95 - 2.41)	0.082
Age (in years)	p= 0.004
60 to 69	77.0	1.00
70 to 74	77.8	1.09 (0.66 - 1.79)	0.733
≥ 75	89.7	2.66 (1.43 - 4.92)	0.002
Waist circumference	p= 0.017
Not overweight	76.8	1.00
Overweight	85.0	1.80 (1.14 - 2.83)	0.011
Body mass index	p= 0.009
Underweight	81.2	1.38 (0.62 - 3.08)	0.428
Eutrophy	75.0	1.00
Overweight	86.1	2.11 (1.31 - 3.39)	0.002
Waist-to-height ratio	p= 0.001
Not overweight	75.6	1.00
Overweight	86.9	2.08 (1.31 - 3.31)	0.002
Waist-to-hip ratio	p= 0.009
Not overweight	75.8	1.00
Overweight	84.9	1.87 (1.17 - 2.99)	0.009
Conicity index	p= 0.128
Not overweight	76.1	1.00
Overweight	82.4	1.60 (0.96 - 2.68)	0.069

• ^a p value, Pearson’s chi-square test; ^b OR: adjusted odds ratio: sex adjusted for age, age adjusted for sex, and overweight adjusted for sex and age; 95%CI: 95% confidence interval; ^c p value, Wald test.

Discussion

This research assessed changes in the anthropometric profile of the older adults recruited in households and at flow points, during the period between the baseline (2008-2009) and follow-up (2016-2017) surveys of the FIBRA Study. Among the eight measures and indicators selected, men showed alterations in four: decreases in weight, height and BMI, and an increase in C index; while women showed alterations in all of them: decreases in weight, height and BMI, and increases in measurements and indicators of central adiposity – WC, HC, WHtR, WHR and C index. Significant associations were observed between being overweight and gait stability or new cases of slower gait. Being overweight/obese impacts the health and quality of life of older adults, resulting from the increased risks of morbidity and mortality, complications and disabilities, while also impacting health care systems through the increase in costs and demand for health services⁴-⁷,¹⁰.

Other studies report the same findings regarding the reduction in weight⁵,²⁷,²⁸ and height²⁷-²⁹ observed in this research. Santanasto et al.⁵ analyzed data from the Health, Aging and Body Composition (Health ABC) cohort of Pittsburgh, PA, and Memphis, TN, in the United States, and observed a reduction in body weight in men (81.6 kg at baseline; -1.5 kg/-1.7%) and women (70.1 kg at baseline; -1.4 kg/-1.8%) after five years. In Norway, a follow-up study, detected decreases in the height of older adults aged between 60 and 69, 70 and 79 and ≥ 80 years at baseline: -1.3 cm, -1.9 cm, -2.4 cm in males and -1.9 cm, -2.3 cm, -2.3 cm in females, 11 years later. Regarding body weight, reductions were observed from the age of 70: -1.3 kg and -2.4 kg in men and -2.4 kg and -5.6 kg in women²⁷. In contrast to the results of this study, Almeida et al.³⁰ observed no significant changes in the weight and height of older adults (≥60 years) included in the SABE Study (Health, Well-being and Aging), between 2000 and 2006, probably due to the younger sample and shorter follow-up time.

Based on data from a cohort of Swedish older adults, Gavriilidou et al.²⁸ observed decreases in height of around 6 cm for men and 8 cm for women, between the ages of 60 to 64 and 85 years or older. The authors also investigated anthropometric classification errors caused by the imprecision of measured height in older adults. To achieve this, they calculated the BMI using the measured height and that estimated by knee height. The results revealed that the use of measured height to calculate the BMI underestimated the prevalence of low weight and overestimated the prevalence of obesity, in both sexes and more intensely in older adults aged ≥80 years, in relation to the estimated measure²⁸. A study conducted in an outpatient clinic identified that frail older adults presented greater differences between the measured and estimated height compared with their robust peers, and recommended the use of the estimated measure, particularly for frail older adults³¹.

The trajectory of human aging involves changes in body composition that include a decrease in height, loss of muscle and bone tissue, and an increase in and redistribution of adipose tissue¹,²,⁶. The progressive decrease in height results from compression of the intervertebral discs, flattening of the vertebrae, changes in body posture, decreased bone mineral density (osteopenia/osteoporosis) and flattening of the plantar arch¹,²,³². Body weight reduction is observed from the age of 70 and over⁶,²⁷ and results from the loss of muscle mass, body water and bone mass¹,²,³². BMI decreases with advancing age due to loss of muscle mass²⁷,³³. Results from the English Longitudinal Study of Ageing (ELSA) show an increase in BMI in the early years of old age, followed by a significant decline from the age of 71 onwards³³.

In this study, women showed an increase in the medians of anthropometric measurements that evaluated the distribution of body fat (WC, HC, WHtR and WHR). Over the course of eight years, the ELSA data revealed an increase in waist circumference up to the age of 80 (0.18 cm/year) and a downward trend from that age onwards, for both sexes³³. In the United States, a five-year prospective study involving older adults aged 70 to 79 years at baseline, identified a reduction in subcutaneous and visceral abdominal fat in women using computed tomography⁵. Adipose tissue increases with advancing age and tends to accumulate in the abdominal region, increasing chronic low-grade inflammation and the risk of cardiometabolic diseases²,⁶,¹⁵. After menopause, with the decline in estrogen levels, the fat deposited in the gluteofemoral region is redistributed to the visceral deposit¹⁵. The findings of this research show that the process of fat mass redistribution in women continued during follow-up, different from that observed in men.

Although there were increases in the medians of the C index in both sexes, it was more intense among women. In a study conducted with older adults assisted by the Estratégia Saúde da Família [Family Health Strategy] in Viçosa, MG, the mean value of the C index for women was also higher than for men (p<0.01)³⁴. In older adults (≥60 years) from Salvador, BA, the accuracy of the C index for classifying visceral obesity was 0.97 and 0.66, respectively for males and females³⁵. Proposed by Valdez¹⁷ in the 1990s, as an indicator of central obesity, the C index is considered a good predictor of diabetes mellitus, hypertension and cardiovascular diseases¹⁹.

A cross-sectional study involving Portuguese older adults (≥65 years) observed greater chances of slower gait (≤0.8 m/s) among overweight (OR= 2.42, 95%CI: 1.13-5.18) and obese women (3.97, 95%CI: 1.63-9.67) than among eutrophic women. Among men, these associations were similar, but the OR estimates fell by half (p value for trend = 0.001)⁷. Data from the SABE Colombia survey showed an inverse association between BMI and gait speed in women and older adults in general⁸. The mechanical overload that being overweight/obese exerts on body joints, such as the knees and hips, and the low-grade inflammation triggered by excess adipose mass are indicated in the literature as causes of slower gait⁷,⁸, highlighting the importance of strategies to prevent weight gain.

Appraisal of the results of this research should consider certain limitations. At baseline, the FIBRA Study selected a sample of older adults with no apparent cognitive deficit and with adequate physical and health status to attend the data collection sites, which may have introduced some bias in the selection of individuals who presented better anthropometric and nutritional profiles. In turn, the survival bias that may have influenced the data could be due to the lower risk of premature death among non-obese older adults with a greater reserve of lean mass. Despite these potential limitations, the possibility of evaluating changes in the anthropometric indicators of older adults nine years later is a major strength of this study.

Conclusion

The results of this study revealed changes in the anthropometric profile resulting from aging. In both sexes, we observed decreases in body weight, height and BMI. Women showed increases in all the indicators of abdominal obesity, while men only showed an increase in the C index. Being overweight was associated with a greater chance of gait stability and new cases of slower gait, nine years after the first survey of measurements.

This study provides information from a cohort of older adults, a considerable portion of whom were aged 80 years old or over, on changes in various anthropometric indicators and in gait speed. Clinical or public health professionals dedicated to the care of older adults and research will benefit from the results, in order to identify, for example, the most sensitive indicators for discerning excess weight during aging, in order to develop interventions that promote healthy aging.

• Funding: CAPES/PROCAD, Nº do processo 2972/2014-01, FAPESP Nº 2016/00084-8, CNPq Nº 424789/2016-7. Post-doctoral scholarship for D. A. PNPD/CAPES, Nº do processo: 88887.320898/2019-00.

REFERÊNCIAS

• ¹
  JafariNasabian P, Inglis JE, Reilly W, Kelly OJ, Ilich JZ. Aging human body: changes in bone, muscle and body fat with consequent changes in nutrient intake. Journal of Endocrinology 2017;234,R37-R51. doi:10.1530/JOE-16-0603.
• ²
  Amarya S, Singh K, Sabharwal M. Changes during aging and their association with malnutrition. J Clin Gerontol Geriatr 2015;6:78-84. http://dx.doi.org/10.1016/j.jcgg.2015.05.003
• ³
  Zampino M, AlGhatrif M, Kuo P-L, Simonsick EM, Ferrucci L. Longitudinal Changes in Resting Metabolic Rates with Aging Are Accelerated by Diseases. Nutrients 2020;12:3061. doi:10.3390/nu12103061.
• ⁴
  Sedlmeier AM, Baumeister SE, Weber A, Fischer B, Thorand B, Ittermann T, et al. Relation of body fat mass and fat-free mass to total mortality: results from 7 prospective cohort studies. Am J Clin Nutr 2021;113:639-46. doi: https://doi.org/10.1093/ajcn/nqaa339.
• ⁵
  Santanasto AJ, Goodpaster BH, Kritchevsky SB, Miljkovic I, Satterfield S, Schwartz AV, et al. Body Composition Remodeling and Mortality: The Health Aging and Body Composition Study. J Gerontol A Biol Sci Med Sci 2017; 72(4):513-19. doi:10.1093/gerona/glw163.
• ⁶
  Ponti F, Santoro A, Mercatelli D, Gasperini C, Conte M, Martucci M, et al. Aging and Imaging Assessment of Body Composition: From Fat to Facts. Front Endocrinol 2020;10:861. doi: 10.3389/fendo.2019.00861.
• ⁷
  Mendes J, Borges N, Santos A, Padrão P, Moreira P, Afonso C, et al. Nutritional status and gait speed in a nationwide population-based sample of older adults. Sci Rep 2018;8:4227. doi:10.1038/s41598-018-22584-3.
• ⁸
  Ramírez-Vélez R, Pérez-Sousa MA, Venegas-Sanabria LC, Cano-Gutierrez CA, Hernández-Quiñonez PA, Rincón-Pabón D, et al. Normative Values for the Short Physical Performance Battery (SPPB) and Their Association With Anthropometric Variables in Older Colombian Adults. The SABE Study, 2015. Front Med 2020;7:52. doi: 10.3389/fmed.2020.00052.
• ⁹
  Houston DK, Ding J, Nicklas BJ, Harris TB, Lee JS, Nevitt MC, et al. Overweight and Obesity Over the Adult Life Course and Incident Mobility Limitation in Older Adults. Am J Epidemiol 2009;169:927-36. doi: 10.1093/aje/kwp007.
• ¹⁰
  Kim S, Leng XI, Kritchevsky SB. Body Composition and Physical Function in Older Adults with Various Comorbidities. Innovation in Aging 2017; 00(00):1-9. doi:10.1093/geroni/igx008.
• ¹¹
  Feng Z, Lugtenberg M, Franse C, Fang X, Hu S, Jin C, et al. Risk factors and protective factors associated with incident or increase of frailty among community-dwelling older adults: A systematic review of longitudinal studies. PLoS ONE 2017;12(6): e0178383. https://doi.org/10.1371/journal.pone.0178383.
• ¹²
  Stenholm S, Strandberg TE, Pitkälä K, Sainio P, Heliövaara M, Koskinen S. Midlife Obesity and Risk of Frailty in Old Age During a 22-Year Follow-up in Men and Women: The Mini-Finland Follow-up Survey. J Gerontol A Biol Sci Med Sci 2014;69(1):73-8. doi:10.1093/gerona/glt052.
• ¹³
  García-Esquinas E, García-García FJ, León-Muñoz LM, Carnicero JA, Guallar-Castillón P, Harmand MG-C, et al. Obesity, Fat Distribution, and Risk of Frailty in Two Population-Based Cohorts of Older Adults in Spain. Obesity 2015; 23:847-55. doi: 10.1002/oby.21013.
• ¹⁴
  Neeland IJ, Ross R, Després J-P, Matsuzawa Y, Yamashita S, Shai I, et al. Visceral and ectopic fat, atherosclerosis, and cardiometabolic disease: a position statement. Lancet Diabetes Endocrinol 2019;7:715-25. http://dx.doi.org/10.1016/S2213-8587(19)30084-1.
• ¹⁵
  Mancuso P, Bouchard B. The Impact of Aging on Adipose Function and Adipokine Synthesis. Front Endocrinol 2019;10:137. https://doi.org/10.3389/fendo.2019.00137.
• ¹⁶
  Ashwell M, Gunn P, Gibson S. Waist-to-height ratio is a better screening tool than waist circumference and BMI for adult cardiometabolic risk factors: systematic review and meta-analysis. Obes Rev 2011;13:275-86. doi: 10.1111/j.1467-789X.2011.00952.x.
• ¹⁷
  Valdez R. A simple model-based index of abdominal adiposity. J Clin Epidemiol 1991;44:955-6. doi: 10.1016/0895-4356(91)90059-i.
• ¹⁸
  Ashwell M, Gibson S. Waist-to-height ratio as an indicator of ‘early health risk’: simpler and more predictive than using a ‘matrix’ based on BMI and waist circumference. BMJ Open 2016;6:e010159. doi: 10.1136/bmjopen-2015-010159.
• ¹⁹
  Piqueras P, Ballester A, Durá-Gil JV, Martinez-Hervas S, Redón J, Real JT. Anthropometric Indicators as a Tool for Diagnosis of Obesity and Other Health Risk Factors: A Literature Review. Front. Psychol 2021;12:631179. doi:10.3389/fpsyg.2021.631179.
• ²⁰
  Neri AL, Yassuda MS, Araújo LF, Eulálio MC, Cabral BE, Siqueira MEC, et al. Metodologia e perfil sociodemográfico, cognitivo e de fragilidade de idosos comunitários de sete cidades brasileiras: Estudo FIBRA. Cad Saúde Pública 2013;29(4):778-92. https://doi.org/10.1590/S0102-311X2013000400015.
• ²¹
  Brasil. Ministério da Saúde. Orientações para coleta e análise de dados antropométricos em serviços de saúde: norma técnica do sistema de Vigilância Alimentar e Nutricional - SISVAN [Internet]. Brasília: Ministério da Saúde; 2011 [acessado em x/x/2021]. 76 p. Disponível em: https://bvsms.saude.gov.br/bvs/publicacoes/orientacoes_coleta_analise_dados_antropometricos.pdf
  » https://bvsms.saude.gov.br/bvs/publicacoes/orientacoes_coleta_analise_dados_antropometricos.pdf
• ²²
  Lohman TG, Roche AF, Martorell R. Anthropometric standardization reference manual. 1 ed. Champaign: Human Kinetics Books; 1988.
• ²³
  Cruz-Jentoft AJ, Bahat G, Bauer J, Boirie Y, Bruyère O, Cederholm T, et al. Sarcopenia: revised European consensus on definition and diagnosis. Age Ageing 2019;48:16-31. doi: 10.1093/ageing/afy169.
• ²⁴
  Assumpção D, Ferraz RO, Borim FSA, Neri AL, Francisco PMSB. Pontos de corte da circunferência da cintura e da razão cintura/estatura para excesso de peso: estudo transversal com idosos de sete cidades brasileiras, 2008-2009. Epidemiol Serv Saude 2020;29(4):e2019502. doi: 10.5123/S1679-49742020000400027.
• ²⁵
  Brasil. Ministério da Saúde. Vigilância alimentar e nutricional - Sisvan: orientações básicas para a coleta, processamento, análise de dados e informação em serviços de saúde. Brasília: Ministério da Saúde, 2004. 120 p.
• ²⁶
  Pitanga FJG, Lessa I. Sensibilidade e especificidade do índice de conicidade como discriminador do risco coronariano de adultos em Salvador, Brasil. Rev Bras Epidemiol 2004;7(3):259-269.
• ²⁷
  Drøyvold WB, Nilsen TIL, Krüger Ø, Holmen TL, Krokstad S, Midthjell K, et al. Change in height, weight and body mass index: Longitudinal data from the HUNT Study in Norway. Int J Obes 2006; 30:935-39. doi: 10.1038/sj.ijo.0803178.
• ²⁸
  Gavriilidou NN, Pihlsgård M, Elmståhl S. High degree of BMI misclassification of malnutrition among Swedish elderly population: Age-adjusted height estimation using knee height and demispan. Eur J Clin Nutr 2015; 69:565-71. doi:10.1038/ejcn.2014.183.
• ²⁹
  Pelclová J, Štefelová N, Olds T, Dumuid D, Hron K, Chastin S, et al. A study on prospective associations between adiposity and 7-year changes in movement behaviors among older women based on compositional data analysis. BMC Geriatrics 2021;21:203. https://doi.org/10.1186/s12877-021-02148-3.
• ³⁰
  Almeida MF, Marucci MFN, Gobbo LA, Ferreira LS, Dourado DAQS, Duarte YAO. Anthropometric Changes in the Brazilian Cohort of Older Adults: SABE Survey (Health, Well-Being, and Aging). J Obes 2013; http://dx.doi.org/10.1155/2013/695496
  » http://dx.doi.org/10.1155/2013/695496
• ³¹
  Jansen AK, Santos DAG, Ramiro DO, Santos RR. Comparação da estatura aferida e estimada em idosos com diferentes classificações funcionais. Mundo da Saúde 2020; 44:445-53. doi:10.15343/0104-7809.202044445453.
• ³²
  Santos ACO, Machado MMO, Leite EM. Envelhecimento e alterações do estado nutricional. Geriatria & Gerontologia 2010; 4(3):168-75.
• ³³
  Zaninotto P, Lassale C. Socioeconomic trajectories of body mass index and waist circumference: results from the English Longitudinal Study of Ageing. BMJ Open 2019; 9:e025309. doi:10.1136/bmjopen-2018-025309.
• ³⁴
  Milagres LC, Martinho KO, Milagres DC, Franco FS, Ribeiro AQ, Novaes JF. Relação cintura/estatura e índice de conicidade estão associados a fatores de risco cardiometabólico em idosos. Cien Saude Colet 2019; 24(4):1451-61. https://doi.org/10.1590/1413-81232018244.12632017.
• ³⁵
  Roriz AKC, Passos LCS, de Oliveira CC, Eickemberg M, Moreira PdA, Sampaio LR. Evaluation of the Accuracy of Anthropometric Clinical Indicators of Visceral Fat in Adults and Elderly. PLoS ONE 2014;9(7):e103499. doi:10.1371/journal.pone.0103499.

Edited by

Edited by: Isac Davidson S. F. Pimenta

Publication Dates

• Publication in this collection
  06 June 2022
• Date of issue
  2022

History

• Received
  28 Nov 2021
• Accepted
  22 Mar 2022