This site is intended for healthcare professionals only

Journal of
Diabetes Nursing


Share this article

Predictive factors of impaired glycaemic state in community-dwelling older adults in Brazil

Claudineia Matos de Araujo
, Marcos Fernandes
, José Ailton Oliveira Carneiro
, Raildo da Silva Coqueiro
, Rafael Pereira

This study aimed to identify and analyse predictive factors for impaired glycaemic state (IGS) in community-dwelling old adults, in order to create a predictive model for this population. Data from 316 community-dwelling older adults (≥60 years) were analysed, with IGS as the dependent variable and age and anthropometric and motor performance indicators as predictor variables. In the resulting model, only waist circumference for men and BMI for women were significant predictors of IGS.

The ageing process is associated with an increased prevalence of impaired glycaemic state (IGS), a collective term that includes impaired fasting glucose (i.e. fasting blood glucose levels of 5.6–6.9 mmol/L) and diabetes (Danaei et al, 2011; American Diabetes Association, 2014). Alarmingly, the global prevalence of diabetes and/or IGS is increasing, with high impact in developing countries (Danaei et al, 2011; Schmidt et al, 2015). Recent surveys suggest that approximately 9 million Brazilians have diabetes, and around 3.5 million of these are aged 65 years or over (Iser et al, 2015).

Diabetes is one of the most common causes of mortality and morbidity in the world, increasing the risk of death from cardiovascular and kidney disease. The development of diabetes among older adults is associated with accelerated functional decline (Schaap et al, 2013; de Rekeneire and Volpato, 2015); changes in body composition, especially reductions in muscle mass and increases in adipose tissue (Koster and Schaap, 2015); and cardiovascular complications, such as hypertension (Nayak et al, 2014), all of which are typical features of the ageing process. The close relationship of these changes with the pathogenesis of diabetes could help the identification of older people who have IGS.

From this perspective, the identification of predictors of IGS may be an affordable alternative to facilitate screening, with a view to early diagnosis of IGS in the older population. It may also enable the adoption of preventive strategies and early intervention by health services in order to improve the health of older people, especially in communities with low indicators of health and quality of life, where the cost of conducting biochemical tests for the entire population is prohibitive. Thus, this study was performed to identify and analyse predictive factors for IGS in community-dwelling older adults, in order to create a predictive model of IGS for this population.

This was a descriptive study with a cross-sectional design, conducted by the Center for Research on Epidemiology of Aging at the State University of Southwest Bahia, Brazil. The study was conducted in the municipality of Lafaiete Coutinho. During the data collection period, the population consisted of 3901 inhabitants registered in the Primary Health Care Strategy (PHCS). Of these, 2104 individuals were living in the urban area and 598 (15%) were aged ≥60 years.

The study population comprised all individuals aged ≥60 years who were not institutionalised and were living in the urban area of Lafaiete Coutinho. Of the 355 people identified, 316 (89.0%) participated in the study; 17 refusals were registered (4.8%) and 22 subjects (6.2%) were not contactable after three home visits on different days.

The research protocol was approved by the local Ethics Committee (No. 064/2010). Participation was voluntary, and individuals signed an informed consent form, according to the ethical standards required by the Helsinki Declaration.

Data collection
Data were collected in January 2011 by the interviewers, with the support of healthcare professionals from each area of the PHCS. Data from sociodemographic questions, blood tests, blood pressure, heart rate, anthropometric measurements and motor performance tests were recorded.

Glycaemic state (dependent variable)
The population was stratified as impaired (IGS) or normal glycaemic state (NGS) based on two criteria that did not exclude one another: fasting blood glucose ≥5.6 mmol/L and/or continuous use of insulin and/or oral hypoglycaemic drugs. The cut-off point of ≥5.6 mmol/L was adopted based on American Diabetes Association (2014) criteria.

Anthropometric indicators (predictor variables)
The anthropometric indicators evaluated were BMI, as an indicator of general obesity; waist circumference, hip circumference and waist-to-hip ratio, as indicators of central (i.e. visceral) obesity; and arm circumference and calf circumference, as indicators of muscle mass.

Motor performance indicators (predictor variables)
The “chair stand test” was used to assess the strength and endurance of the lower limbs, as proposed by Tinetti and Ginter (1988) and Guralnick et al (1995). The time required to perform five repetitions of rising from a chair and sitting down was recorded, and the test was considered successful when done in ≤60 seconds. Hand grip strength was recorded using the SH5001 hydraulic dynamometer (Saehan Corporation, Republic of Korea). Walking speed was tested over a path of 2.44 m, and the participants were requested to walk along the path at their usual speed, with assistive devices if required.

Details of the statistical analyses performed are presented in Box 1.

Of the 316 older adults included in this study, 309 had all the variables of interest recorded; of these, 170 (55%) were women and 139 (45%) men. The prevalence of IGS was 22.9% among women and 18.0% among men.

Mean age, anthropometric indicators and motor performance indicators of the studied population are presented in Table 1. Among men, BMI, waist circumference, arm circumference and chair stand test times met the inclusion criteria for the multiple logistic regression, while BMI, waist circumference and arm circumference met the inclusion criteria in women.

Table 2 shows the final predictive model for IGS in older adults according to gender. In the multiple logistic regression model for men, only waist circumference remained, indicating that this variable alone was significantly associated with IGS. In the model for women, only BMI remained.

The discriminatory capacity of the models to identify older adults with IGS was evaluated by receiver operating characteristic (ROC) curve parameters, and the results showed that both models had moderate discriminatory capacity. For men, the area under the curve (AUC) for waist circumference was 0.65 (95% confidence interval [CI], 0.57–0.73), the sensivity was 45.8 (95% CI, 25.6–67.2) and the specificity was 87.6 (95% CI, 80.1–93.1). The cut-off point was >101 cm.

For women, the (AUC) for BMI was 0.62 (95% CI, 0.54–0.70), the sensivity was 38.9 (95% CI, 23.2–56.5) and the specificity was 85.0 (95% CI, 77.6–90.7). The cut-off point was >29 kg/m2.

This study aimed to investigate whether anthropometric and motor performance indicators together could identify and predict IGS in community-dwelling older adults. The results obtained in the multiple logistic regression model showed that, for older men, waist circumference alone had discriminatory and predictive potential for IGS, while only BMI showed this potential for older women.

It is interesting that two different anthropometric indicators were significant in the final model for men and women. This divergence could be explained by differences in body fat distribution between the two genders, as men are usually characterised by android-type fat distribution, with fat accumulation predominantly in the abdominal region (Ley et al, 1992; Blouin et al, 2008), and the amount of fat located inside the abdominal cavity (intra-abdominal or visceral adipose tissue) tends to be twice as high in men as in women (Blouin et al, 2008).

Furthermore, it is known that the relationship between BMI and percentage body fat in men and women diverges during the ageing process, with BMI being a better reflection of body fat composition in older women than in older men (Meeuwsen et al, 2010). This might explain the absence of BMI in the predictive model generated for men.

From a biological point of view, these results could be explained by the fact that free fatty acids deriving from lipolysis may induce insulin resistance (Montague and O’Rahilly, 2000; Ribeiro Filho et al, 2006). Another mechanism could be the secretion of numerous adipokines from adipocytes, which directly influence tissue sensitivity to insulin (Lebovitz and Banerji, 2005; Ribeiro Filho et al, 2006).

It is noteworthy that the final model obtained for both genders presented cut-off points (waist circumference >101 cm for older men and BMI >29 kg/m2 for older women) with high specificities of around 85%. This makes these factors a good potential screening tool, especially given that they are cheap and easy to measure. The cut-off points obtained here should be considered in future studies as an option to aid in screening for IGS in clinical practice and/or in epidemiological studies involving older adults. These results should only be applied to populations with similar characteristics (e.g. age group and ethnicity), but further studies to determine the generalisability of these findings to other populations are encouraged.

From these results, it can be concluded that waist circumference is predictive of IGS in older men (aged ≥60 years), while BMI is the better predictor in older women. These two anthropometric measures can be used to screen for this outcome in the community, which could contribute to early identification of IGS risk and enable the adoption of preventive actions and control strategies.

This study was supported by a grant from the State of Bahia Research Foundation (FAPESB).


American Diabetes Association (2014) Diagnosis and classification of diabetes mellitus. Diabetes Care 37(Suppl 1): 81–90
Blouin K, Boivin A, Tchernof A (2008) Androgens and body fat distribution. J Steroid Biochem Mol Biol 108: 272–80
Buatois S, Perret-Guillaume C, Gueguen R et al (2010) A simple clinical scale to stratify risk of recurrent falls in community-dwelling adults aged 65 years and older. Phys Ther 90: 550–60
Danaei G, Finucane MM, Lu Y et al (2011) National, regional, and global trends in fasting plasma glucose and diabetes prevalence since 1980: systematic analysis of health examination surveys and epidemiological studies with 370 country-years and 2.7 million participants. Lancet 378: 31–40
Dawson B, Trapp RG (2004) Basic and Clinical Biostatistics (4th edition). McGraw-Hill, New York, NY, USA
de Rekeneire N, Volpato S (2015) Physical function and disability in older adults with diabetes. Clin Geriatr Med 31: 51–65
Guralnik JM, Ferrucci L, Simonsick EM et al (1995) Lower-extremity function in persons over the age of 70 years as a predictor of subsequent disability. N Engl J Med 332: 556–61
Iser BP, Stopa SR, Chueiri PS et al (2015) Self-reported diabetes prevalence in Brazil: results from National Health Survey 2013. Epidemiologia e Serviços de Saúde 24: 305–14 (in Portuguese)
Koster A, Schaap LA (2015) The effect of type 2 diabetes on body composition of older adults. Clin Geriatr Med 31: 41–9
Lebovitz HE, Banerji MA (2005) Point: visceral adiposity is causally related to insulin resistance. Diabetes Care 28: 2322–5
Ley CJ, Lees B, Stevenson JC (1992) Sex- and menopause-associated changes in body-fat distribution. Am J Clin Nutr 55: 950–4
Meeuwsen S, Horgan GW, Elia M (2010) The relationship between BMI and percent body fat, measured by bioelectrical impedance, in a large adult sample is curvilinear and influenced by age and sex. Clin Nutr 29: 560–6
Montague CT, O’Rahilly S (2000) The perils of portliness: causes and consequences of visceral adiposity. Diabetes 49: 883–8
Nayak BS, Sobrian A, Latiff K et al (2014) The association of age, gender, ethnicity, family history, obesity and hypertension with type 2 diabetes mellitus in Trinidad. Diabetes Metab Syndr 8: 91–5
Ribeiro Filho FF, Mariosa LS, Ferreira SR, Zanella MT (2006) Visceral fat and metabolic syndrome: more than a simple association. Arq Bras Endocrinol Metabol 50: 230–8 (in Portuguese)
Schaap LA, Koster A, Visser M (2013) Adiposity, muscle mass, and muscle strength in relation to functional decline in older persons. Epidemiol Rev 35: 51–65
Schmidt MI, Duncan BB, Ishitani L et al (2015) Trends in mortality due to diabetes in Brazil, 1996–2011. Diabetol Metab Syndr 7: 109
Tinetti ME, Ginter SF (1988) Identifying mobility dysfunctions in elderly patients. Standard neuromuscular examination or direct assessment? JAMA 259: 1190–3

Related content
Genetic insights into type 2 diabetes and some cancers
Free for all UK & Ireland healthcare professionals

Sign up to all DiabetesontheNet journals


By clicking ‘Subscribe’, you are agreeing that are able to email you periodic newsletters. You may unsubscribe from these at any time. Your info is safe with us and we will never sell or trade your details. For information please review our Privacy Policy.

Are you a healthcare professional? This website is for healthcare professionals only. To continue, please confirm that you are a healthcare professional below.

We use cookies responsibly to ensure that we give you the best experience on our website. If you continue without changing your browser settings, we’ll assume that you are happy to receive all cookies on this website. Read about how we use cookies.