The prevalence and predictors of geriatric giants in community-dwelling older adults: a cross-sectional study from the Middle East


We used data from the second wave of the Bushehr Elderly Health Program (BEHP). Participants were urban residents from Bushehr in the south of Iran. They were included using a multistage cluster random sampling method. The detail of the first and second wave of this study was published elsewhere17. In the first wave, 3000 older adults (age > 60 years) who lived in a community in Bushehr were enrolled. The participants were chosen from 75 separate neighborhoods defined by the local municipality. The number of participants selected from each neighborhood was proportional to the number of households registered in the last census. In the second wave, 2480 subjects from the first wave were enrolled again. They were invited to a health center, and their health status was evaluated. Data on demographic characteristics, lifestyle factors, general health, medical history, and mental and functional status were collected through a questionnaire. Assessment of anthropometric indices, performance-based tests, muscle strength tests, blood pressure evaluation, and body composition measurements were performed. Because of incomplete data, finally, 2392 subjects were included for analysis. The reason for incomplete data was death or insufficient compliance of patients for completing the assessments.

Data collection

The phenotype of frailty was defined according to the Fried phenotype. Unintentional weight loss ≥ 4.5 kg during the past year was asked. Exhaustion was defined as positive answers to two questions of the Center for Epidemiological Studies Depression (CES-D). Slowness was defined based on sex and height. Handgrip strength and physical activity were determined using the lowest quintile of mean handgrip measures after three times of measurements in each hand. Low physical activity was defined as the lowest metabolic equivalent during 1 week based on WHO physical activity questionnaire. The presence of one or two of the aforementioned criteria was defined as pre-frailty and the presence of more than two criteria was defined as frailty42.

Demographic data were also collected during the interview. Anthropometric data such as weight, height, and waist circumference were measured based on NHANCE-3 guidelines. Body mass index (BMI) was calculated by dividing weight (kg) by squared height (meter). Overnight fasting blood samples were collected and fasting blood sugar, Hb A1C, and complete blood count (CBC) were measured using automatic standardized devices. Diabetes mellitus and hypertension were defined according to ADA 2018 and JNC-8 criteria.

Urinary and fecal incontinence was defined as the incidence of any episode of incontinence during the last 6 months. Mobility and instability were evaluated by asking the participants and their partners about their ability to move. In addition, timed up and go test and short physical performance battery (SPPB) scores were assessed43. The incidence and number of falls during the last year were recorded. Anorexia in the past 3 months was evaluated by the Mini Nutritional Assessment (MNA) questionnaire44. Furthermore, information regarding drug history and significant weight loss (≥ 4.5 kg) during the last year was also collected. Taking more than five drugs per day was considered polypharmacy.

Katz ADL instrument was utilized for assessing independence in basic daily activities, including bathing, feeding, urinary and fecal continence, dressing, toileting, and transferring45. Lawton instrumental activity of daily living was used for assessing intermediate daily function. This instrument has eight domains. A score of eight shows independence in all areas and a score of zero shows complete dependence.

Mood was measured using patient health questionnaire 9 (PHQ-9)46. This questionnaire has nine items and each item is a Likert-type question with a score from zero to three. The maximum score is 27 showing the worst mood and zero showing the best mood. The validity and reliability of this tool were approved in the Iranian population47.

Cognitive function was assessed using two instruments; the Mini-Cog score and the category fluency test. Mini-Cog had two stages48. In the first stage, the ability to recall three words was evaluated. Remembering all words was considered normal cognition and recalling none of them was considered impaired cognition. Participants who recalled one or two words, entered the second stage, a clock drawing test. Those who correctly drew the clock were considered cognitively intact; otherwise, the subjects were considered cognitively impaired. Furthermore, the category fluency test was used for naming animals. Subjects were categorized into impaired cognition and normal cognition groups based on the number of years they spent in school. Participants with impaired cognition in at least one test were categorized into the impaired cognition group.

We also measured the frailty index comprising 35 items. These items were needing assistance with bathing, dressing, getting in and out of a chair, walking around the house, eating, grooming, using toilet, going up and down stairs, shopping, housework, meal preparation, taking medication, managing finances, experiencing weight loss of more than 4.5 kg in the last year, walking outside, feeling that everything is an effort, feeling depressed or lonely, having trouble getting going, high blood pressure, heart attack, congestive heart failure, stroke, cancer, diabetes, arthritis, chronic lung disease, chronic kidney disease, gait speed, BMI, grip strength, cognitive impairment (Mini-Cog), seizures, urinary incontinence, and bowel incontinence.

The frailty index did not include peak flow, shoulder strength, and timed normal pace due to the lack of clear cut-points in the original article49.

Recorded procedures were used to transform categorical, ordinal, and interval variables into a common scale ranging from 0 to 1. In this mapping, 0 represents the absence of a deficit, while 1 represents the full presence of the deficit. Individual deficit scores were combined to create an index, where a score of 0 indicated the absence of deficits and a score of 1 indicated the presence of all 40 deficits49. Although we report the prevalence of frailty based on two measures, frailty index and Fried frailty phenotype, the latter was used for constructing regression models and interpreting the associations between frailty and other variables.

All questionnaires used in this study were previously translated into Persian language and validated in Iran17.

Ethics approval

The Ethics Committee of Tehran University of Medical Sciences and Bushehr University of Medical Sciences approved the protocol of this study (IR.TUMS.EMRI.REC.1394.0036). Written informed consent was obtained before participation. The protocol of this study followed the Ethics standards defined by the 2013 version of the Declaration of Helsinki. This study has been reported following the STROBE Statement.

Statistical analysis

The prevalence of variables was calculated as a survey analysis with the weighting of the Iranian population census 2016. For comparing our findings with those from other countries, the data were age-standardized based on the World Health Organization (WHO) population 2000–2025.

Quantitative variables are expressed as mean (SD). The prevalence of pre-frailty, frailty, and other components of geriatric giants was assessed in males and females and across different ranges of age. We also used the Chi-square (χ2) test to assess the effect of monthly income or wealth quantile on the prevalence of frailty.

A binary logistic regression model was used to identify the independent factors associated with frailty. The following variables were adjusted in the binary logistic regression model: age, gender, physical activity, cancer, MNA score, PHQ-9 score, waist circumference, BMI, WBC, smoking, marital status, number of comorbidities, and polypharmacy. The presence of the following comorbidities was assessed to calculate the number of comorbidities: hypertension, diabetes, hypo- and hyperthyroidism, chronic kidney disease, rheumatoid arthritis, osteoarthritis, Alzheimer’s disease, liver disease, lung disease, epilepsy, and Parkinson’s. We also used multinomial logistic regression model to investigate the associations between the number of geriatric giants and age, sex, low physical activity, cancer, MNA score, PHQ-9 score, waist circumference, BMI, and smoking.

In addition, we measured the co-incidence of 4 components of geriatric giants, including frailty, sarcopenia, cognitive impairment, and anorexia. Moreover, a multinomial logistic regression model was developed to assess the effect of different variables on the co-existence of frailty, sarcopenia, cognitive impairment, and anorexia. Age, sex, physical activity, cancer, MNA score, PHQ-9 score, waist circumference, BMI, smoking, WBC count, alcohol consumption, marital status, and the number of drugs were entered into this analysis as independent variables.

Stata package version 12 (StataCorp Texas, USA) was utilized for analysis. P values < 0.05 were considered statistically significant.


Leave a Reply

Your email address will not be published. Required fields are marked *