Abstract
Background
Some pathophysiological effects of physical frailty and cognitive impairment might be similar; therefore, finding the associations in epidemiologic studies could guide clinicians and researchers to recognize effective strategies for each type of frailty such as frailty phenotype and frailty index, which in turn will result in a preventive approach. The study aimed to reveal which components of frailty phenotype are more associated with cognitive impairment. The findings of this study may help other researchers clarify the related pathways.
Methods
This is a cross-sectional analysis of the results of the second phase of Bushehr Elderly Health Program; a community-based elderly prospective cohort study conducted in 2015–2016. The participants were selected through a multistage stratified cluster random sampling method. Frailty was assessed based on the Fried frailty phenotype criteria. Cognitive impairment was assessed by the Mini-Mental State Examination (MMSE), the Mini-Cog, and the Category Fluency Test (CFT). Multiple logistic regression models were applied to determine the association between frailty and cognitive impairment. Depression trait was assessed using the Patient Health Questionnaire-9 (PHQ-9). Activities of daily living were assessed using the Barthel Index and Instrumental Activities of Daily Living (IADLs) using Lawton’s IADL.
Results
The studyp conducted among people ≥ 60 years old (N = 2336) with women consisting 51.44% of the sample group. The mean age of the participants was 69.26 years old. The prevalence of pre-frailty and frailty were 42.59% and 7.66%, respectively. In the fully adjusted model, the odds ratio of the association between pre-frailty and frailty with cognitive impairment was 1.239, 95% CI: 1.011 – 1.519 and 1.765, 95% CI: 1.071 – 2.908, respectively (adjusted for age, sex, education, body mass index, smoking, diabetes mellitus, PHQ- 9, Barthel Index, and IADLs). In the fully adjusted multiple logistic regression models, all of the components of Fried frailty phenotype were significantly related to cognitive impairment except weight loss.
Conclusion
Cognitive impairment may be associated with frailty phenotype. Moreover, low strength and function of muscles had a stronger association with cognitive impairment. It seems that a consideration of cognitive impairment assessment in older people along with frailty and vice versa in clinical settings is reasonable.
Keywords: Aged, Category fluency test, Cognitive impairment, Depression, Frailty
Introduction
Frailty syndrome is a condition in both geriatric care and research settings, which leads to a decrease in vital capacities and increases the susceptibility to develop several diseases [1]. Studies have shown that physical frailty is associated with cognitive impairment and dementia [2–6]. Some similar mediators or pathways have been suggested to have a role in developing both physical frailty and cognitive impairment [2].
Several shared mechanisms may have simultaneous roles in the development of frailty and cognitive impairment. Over 20 neuro-inflammatory markers have been reported to possibly have an association with both physical frailty and a decrease in cognitive functions. Some inflammatory processes may play a role in the development of low body muscle mass, and as a result, lead to sarcopenia, an important component of physical frailty. The inflammatory markers are also related to the vascular inflammatory processes, which lead to atherosclerosis and involvement of cerebral small vessel disease, that may impair cognitive function. Moreover, the interaction between hypothalamic–pituitary–adrenal axis hormones and inflammatory factors can be another mechanism that brings about this co-occurrence [7].
Although an association between cognitive impairment and frailty phenotype is indicated in several studies, this association has not been revealed in Iranian population. Furthermore, few studies have reported the association between components of frailty and cognitive impairment. Moreover, the domains of cognitive impairment that are associated with physical frailty are not completely clear either. In this study the association between frailty phenotype and cognitive impairments was assessed and we determined which components of frailty were associated with cognitive functions in a cross-sectional analysis of a large community-based sample of older adults. Besides, we evaluated the relationship between the domains of cognitive functioning and physical frailty in this study.
Methods
Ethical considerations
Bushehr Elderly Health Program phase II was approved by the Research Ethical Committee of Endocrinology and Metabolism Research Institute, affiliated to Tehran University of Medical Sciences (Ethical Code: IR.TUMS.EMRI.REC.1394.0036). All the participants or their legal proxies signed informed consents.
Sampling and setting
This study is a cross-sectional analysis approach to phase II of Bushehr Elderly Health Program’s (BEHP) data, which was carried out in 2015–2016 among people ≥ 60 years old, who were residents of Bushehr city; located in the south of Iran. A representative sample of older adults was selected using a multistage stratified cluster random sampling method. Community-dwelling older adults aged 60 years and over who lived in Bushehr at least one year before their recruitment, and who were also able to participate, were considered as eligible subjects. We excluded those who did not want to participate. Furthermore, the data of subjects who were suspected of having severe depression (PHQ-9 > 19) were deleted from the analyses. The sampling and settings of phases I and II of the Bushehr Elderly Health program were explained elsewhere [8, 9].
Data collection
The participants were interviewed by trained researchers to collect data on their socio-demographic status, lifestyle, general health, medical history, mental health, and medication use. All the assessments were carried out by experts in the Persian Gulf Tropical Medicine Research Center, Bushehr University of Medical Sciences.
Demographic and medical history
Demographic data and medical history were collected by asking elderly participants and their caregivers. Anthropometric measures were carried out using the NEHENS III protocol [10]. Body Mass Index (BMI) was calculated by dividing weight (in kilograms) by height squared (in meters). Smoking habits were obtained, using standard self-reported questionnaires. Overnight fasting blood samples were collected using vacuum tubes. Fasting plasma glucose was measured using calorimetric enzymatic methods and by an auto-analyzer (Roche 904, Osaka, Japan). Diabetes mellitus was defined based on ADA Clinical Guidelines-2021 as FBS ≥ 126 mg/dl, or HbA1c ≥ 6.5% [11] or if there was a history of diabetes mellitus along with consuming anti-diabetes. Parkinson's disease was identified based on the self-report history of the diagnosis of this condition by neurologist.
Assessment of physical health
Handgrip strength was assessed 6 times, three times in the left hand and three times in the right hand, using Saehan digital hand dynamometer (Korea). The mean of recorded values was considered as the handgrip’s strength of the participants. The time in seconds it took the participants to walk a 4.57-m distance twice were measured and the mean value of the measurements was considered as the walking time. By dividing the 4.57 m by the time of walk, the gait speed was calculated (m/s). Physical activity was evaluated using self-report physical activity instruments and the energy cost of each physical activity was calculated by considering the duration and the intensity based the Metabolic Equivalents (METs) table [12].
Frailty phenotype features assessment
Frailty was defined based on Fried frailty phenotype. Its features include weakness, exhaustion, slow gait speed, low physical activity, and unintentional weight loss. Weakness was specified based on the lowest quintile of handgrip strength (by gender and body mass index). Exhaustion was defined by self-reporting using two questions from the CES–D scale. Slow gait speed was defined as the longest quintile of 4.5 m walking time (by gender and height). Unintentional weight loss was considered as a loss of ≥ 4.5 kg or 5% of normal body weight over the last 12 months without an intention to reduce weight. Low activity was defined based on the lowest quintile of Metabolic Equivalents (METs). Individuals who did not have any features of the frailty phenotype were categorized as the “robust” group, those who had one or two criteria were considered as “pre-frail”, and those with three or more criteria were defined as “frail” subjects. [13].
Cognitive status assessment
Three tools were utilized to assess the cognitive status of the participants; the first tool applied for cognitive assessment in this study was the Mini-cog, which has two parts (three-word registrations and recalls and the clock drawing test). When the subjects cannot recall all three words or cannot draw the specified time in a clock, they are considered as cognitively impaired subjects. If the participants could recall one or two words, cognitive impairment was defined if who could not carry out soundly the clock drawing test. The psychometric property of this instrument has been sufficient in an Iranian study [14]. Meanwhile, the Category Fluency Test (CFT) (the animal category) was also applied to categorize the participants into ‘impaired cognition’ and ‘normal-cognitive group’ [15]. The cutoff point 12 was determined for subjects with an education level of ≤ 5 years and the cutoff point 14 for subjects with an education level higher than primary school. The Persian version of the Mini-Mental State Examination (MMSE) was utilized for subjects who were literate. This tool has 6 domains including orientation (time and place), registration, attention and calculation, recall, language, and copying. The cut-off point 20 was considered for those with primary school education, and 24 for those with secondary school or higher education [16]. In our study, the participants were categorized as cognitively impaired if they had a problem in the Mini-cog or CFT in case they were illiterate and in case they were literate, they were considered to have cognitive impairment if they had problems in the Mini-cog or CFT or MMSE. Those who did not have problems in either test (or three tests) were considered as a normal cognitive group.
Activities of daily living
Difficulty performing some activities of daily living can also be present in people with mild cognitive impairment. The Modified Barthel Index is an ordinal scale used to measure performance in activities of daily living (ADLs). Each performance item is rated on this scale with a given number of points assigned to each level or ranking. It uses ten variables describing ADLs and mobility. A higher score indicates less dependence while performing daily activities. A score of 95–100 represents complete independence, a score between 65–95 represents partial dependence, a score of 45–64 represents half Independence, and a score of 0–44 is interpreted as complete dependence [17]. This instrument was validated in Iranian older population[12]
Instrumental activities of daily living (IADLs) are those activities that allow an individual to live independently in a community. The tool includes 8 questions including "Ability to use the phone", "Ability to shop", "Ability to prepare food", "ability to do chores", "Ability to wash clothes", "Ability to travel and use vehicles", "Ability to take medicine", and "ability to account along with books, money and financial handling". The score range is between 0–8, where a higher score indicates a status of independence in performing IADL activities. A score of 8 is interpreted as complete independence and a score below 8 is interpreted as partially or semi-independent [18, 19]. The reliability and validity of this tool was approved in Iranian older population[19].
Mood assessment
Depression was assessed using the Patient Health Questionnaire (PHQ-9) that included nine items, a maximum of 27 and the least score is zero. In this tool, a score less than 4 indicates no depressive symptom, 5–9 are interpreted as mild depression, 10–14 moderate depression, 15–19 moderate to severe depression, and 20–27 severe depression [19, 20]. The psychometric properties of this tool also were evaluated in the Persian language[19].
Statistical analyses
Normality tests were used to determine if a data set is well-modeled by a normal distribution. Parametric continuous variables were presented as mean and standard deviation. Proportions were shown in numbers and percentages. The t-test was used to compare the means between the two given groups. Pearson's chi-squared test was used to evaluate how likely it was for any observed difference between the groups to arise by chance. In order to differentiate pseudo-dementia for cognitive impairment, the subjects with PHQ-9 score > 19 were excluded from the analyses.
Associations between cognitive impairment as the dependent variable and frailty phenotype along with demographic factors and medical history were established by univariate and multiple logistic regression models, and were presented as odds ratios. The forward approach to perform the multiple regression was chosen. The new models were formed by adding the new known related variables (based on published documents) to the previous model. If the new variables had a significant association, they were preserved; otherwise, they were removed from the model. Significant statistical values were considered as α < 0.05. All analyses were performed using Stata software version 12 (USA, Texas).
Sensitivity analyses
To assess the effect of using the different tools to diagnose cognitive impairment on the strength of an association between frailty and cognitive impairment, we considered three definitions of cognitive impairment. First, cognitive impairment was defined only by the Mini-Cog, the second model was only based on the CFT and the third model defined cognitive impairment by the MMSE. The univariate and final multiple logistic regression models were run. The results were then compared with the results of the main analyses.
Results
The data of 2,336 individuals (1202 women and 1124 men) aged 60 years and more were analyzed. The data of the subjects with severe depression were excluded and the data of 2336 subjects were finally analyzed in this study. The mean age of the participants was 69.26 (SD = 6.29) years and 51.44% of them were female. General characteristics of the participants according to their cognitive status are illustrated in Table 1.
Table 1.
Characteristics of the participants according to cognitive status
| Total Participants No = 2336 |
Normal Cognition No = 955 |
Impaired Cognition N0 = 1381 |
P value | ||
|---|---|---|---|---|---|
| Mean Year Age (SD) | 69.26 (6.29) | 67.75 (5.25) | 70.31 (6.73) | < 0.001 | |
| Gender female N (%) | 1223 (51.44) | 371 (38.49) | 852 (60.47) | < 0.001 | |
| Mean BMI kg/m2 (SD) | 27.54 (4.90) | 27.60 (4.66) | 27.50 (5.06) | 0.636 | |
| Mean Education years (SD) | 4.87 (4.98) | 4.92 (5.02) | 4.84 (4.58) | 0.695 | |
| Smoking N (%) | 400 (17.12) | 149 (15.60) | 251 (18.18) | 0.105 | |
| PHQ-9 mean (SD) | 3.00 (4.02) | 2.28 (3.46) | 3.49 (4.29) | < 0.001 | |
| Diabetes Mellitus N (%) | 753 (32.23) | 271 (28.38) | 482 (34.90) | 0.001 | |
| Mean Barthel’s ADL score (SD) | 96.00 (6.02) | 96.75 (5.33) | 95.47 (6.41) | < 0.001 | |
| Mean Lawton IADL score (SD) | 6.75 (1.54) | 7.23 (1.09) | 6.41 (1.71) | < 0.001 | |
| Frailty Syndrome N (%) | Robust | 1168 (49.22) | 584 (60.58) | 584 (41.45) | < 0.001 |
| Pre-frail | 1019 (42.94) | 351 (36.41) | 668 (47.41) | ||
| Frail | 186 (7.84) | 29 (3.01) | 157 (11.14) | ||
| Slow Gate Speed 4.8 m N (%) | 418 (17.61 | 110 (11.41) | 308 (21.86) | < 0.001 | |
| Low Hand Grip Strength N (%) | 469 (19.76) | 113 (11.72) | 356 (25.27) | < 0.001 | |
| HX of Weight loss ≥ 10 Ib N (%) | 144 (6.07) | 40 (4.15) | 104 (7.38) | 0.007 | |
| Exhaustion N (%) | 431 (18.16) | 130 (14.42) | 292 (20.72) | < 0.001 | |
| Low Physical Activity N (%) | 469 (19.51) | 129 (13.38) | 334 (23.70) | < 0.001 | |
SD: Standard Deviation, No: Number, BMI: Body Mass Index, ADL: Activity of Daily Living, PHQ-9: Patient Health Questionnaire 9-item, HX: History
The crude frequency of frailty, according to Fried frailty phenotype was calculated as 179 (7.66%) and the crude frequency of pre-frailty was equal to 995(42.59%). According to our definition, the prevalence of cognitive impairment among the participants was 59.12% (1381 subjects). According to the Mini-Cog tool categorization, the prevalence of cognitive impairment was 51.93%; according to the CFT definition of cognitive impairment, it was equal to 27.05; and according to MMSE, the prevalence was 30.55%. In univariate analyses, there was an association between cognitive impairment and all components of frailty phenotype. There was a significant association between cognitive impairment and age; (OR = 1.074; 95% CI: 1.058 – 1.090), male gender; (OR = 0.417; 95%CI: 0.352 – 0.4984), diabetes mellitus; (OR = 1.353; 95% CI: 1.131 – 1.618), PHQ score; (OR = 1.086; 95% CI: 1.061 – 1.111), pre-frailty; (OR = 1.884; 95% CI: 1.583—2.242, frailty; OR = 5.172; 95% CI: 3.419—7.824, Barthel index (OR = 0.962; 95% CI: 0.948 – 0.977), and IADLs score (OR = 0.650; 95% CI: 0.606 – 0.699) in univariate analysis. In the final multiple models, after adjustment for age, gender, BMI, smoking, years of classic education, diabetes mellitus, PHQ-9 score, Barthel Index, and IADLs, there was a significant association between cognitive impairment and frailty phenotype (OR = 1.239; 95% CI: 1.011—1.519 for pre-frail and OR = 1.765; 95% CI: 1.071—2.908 for frail subjects) [Table 2]. Besides, the association between frailty features and cognitive impairment was assessed by a cumulative approach. As the number of frailty features increased, the odds ratios increased in the univariate analyses (P trends for odds ratios = < 0.01). However, in multivariable analyses, this increasing trend continued only up to three features [Table 3]. Moreover, there was a significant association between cognitive impairment and the following components: low physical activity; (OR = 1.356; 95% CI 1.053 – 1.746, P = 0.018), low handgrip; (OR = 1.817; 95% CI 1.409 – 2.343, P < 0.001), and low walk speed; (OR = 1.547; 95% CI 1.182 – 2.024) after adjustment for age, gender, current smoker, education year, PHQ-9 score, diabetes mellitus, and Barthel Index. There was a negative association between cognitive impairment and exhaustion (OR = 0.643 95% CI: 0.464 – 0.891, P = 0.008) and weight loss during the previous year (OR = 1.455; 95% CI 0.980 – 2.158, P = 0.063) (Table 4).
Table 2.
Association between frailty syndrome and cognitive impairment
| Variable | Odds Ratio | 95% CI OR | P value | |
|---|---|---|---|---|
| Univariate Analyses | ||||
| Frailty Syndrome | Robust (0) | Reference Group | ||
| Pre-frail (1 – 2 criteria) | 1.884 | 1.583—2.242 | < 0.001 | |
| Frail (3 – 5 criteria) | 5.172 | 3.419—7.824 | < 0.001 | |
| Age (each year increases) | 1.074 | 1.058 – 1.090 | < 0.001 | |
| Gender (Male/Female) | 0.417 | 0.352 – 0.494 | < 0.001 | |
| Years of classic education (each year increases) | 0.997 | 0.980 – 1.013 | 0.695 | |
| BMI (each kg/m2 increases) | 0.996 | 0.979 – 1.013 | 0.546 | |
| Smoking (yes/no) | 1.208 | 0.968 – 1.509 | 0.095 | |
| PHQ_9 score (each score increases) | 1.086 | 1.061 – 1.111 | < 0.001 | |
| Diabetes Mellitus (yes/no) | 1.353 | 1.131 – 1.618 | 0.001 | |
| Parkinson’s Disease(yes/no) | 1.003 | 0.528 – 1.905 | 0.992 | |
| Barthel’s Index (each score increases) | 0.962 | 0.948 – 0.977 | < 0.001 | |
| IADL (each score increases) | 0.650 | 0.606 – 0.699 | < 0.001 | |
| Low Physical Activity (yes/no) | 1.999 | 1.596 – 2.497 | < 0.001 | |
| Exhaustion (yes/no) | 1.483 | 1.184 – 1.858 | < 0.001 | |
| Low Hand Grip Strength (yes/no) | 2.516 | 1.996 – 3.172 | < 0.001 | |
| Slow Gate Speed (yes/no) | 2.399 | 1.868 – 3.082 | < 0.001 | |
| Weight loss ≥ 10 Ib (yes/no) | 1.686 | 1.152 – 2.469 | 0.007 | |
| Low Speed of Gate | 2.145 | 1.692 – 2.718 | < 0.001 | |
| First Multivariable Model | ||||
| Fried’s Frailty Phenotype | Robust (0) | Reference Group | ||
| Pre-frail (1 – 2 criteria) | 1.434 | 1.191 – 1.725 | < 0.001 | |
| Frail (3 – 5 criteria) | 2.852 | 1.835 – 4.434 | < 0.001 | |
| Second Multivariate Model | ||||
| Fried’s Frailty Phenotype | Robust (0) | Reference Group | ||
| Pre-frail (1 – 2 criteria) | 1.397 | 1.159 – 1.685 | < 0.001 | |
| Frail (3 – 5 criteria) | 2.892 | 1.848 – 4.525 | < 0.001 | |
| Third Multivariable Model | ||||
| Fried’s Frailty Syndrome | Robust (0) | Reference Group | ||
| Pre-frail (1 – 2 criteria) | 1.239 | 1.011 – 1.519 | 0.039 | |
| Frail (3 – 5 criteria) | 1.765 | 1.071 – 2.908 | 0.026 | |
First Model: adjusted for age, sex, and years of classic education
Second Model: adjusted for age, sex, years of classic education, smoking, and diabetes mellitus
Third Model: adjusted for age, sex, years of classic education, smoking, diabetes mellitus, Barthel’s index, and IADL score
Table 3.
Univariate and multivariable association between cognitive impairment with frailty phenotype as cumulative approach of features
| Cumulative approach to features of phenotype of frailty | Univariate | Multivariable | ||||
|---|---|---|---|---|---|---|
| OR | 95% CI OR | P value | OR | 95% CI OR | P value | |
| Without any Feature | Reference group | Reference group | ||||
| Existed One Feature | 1.728 | 1426 – 2.094 | < 0.001 | 1.336 | 1.081—1.651 | 0.007 |
| Existed Two Features | 2.356 | 1.797—3.088 | < 0.001 | 1.563 | 1.144—2.136 | 0.005 |
| Existed Three Features | 6.010 | 3.602—10.029 | < 0.001 | 3.485 | 1.977 – 6.144 | < 0.001 |
| Existed Four Features | 4.257 | 1.954—9.275 | < 0.001 | 1.541 | 0.644 – 3.691 | 0.332 |
| Existed Five Features | Empty | Empty | ||||
*Adjusted for age, gender, classic education year, current smoking, PHQ-9 score, Barthel’s index, and diabetes mellitus
Table 4.
Univariate and multivariable association between cognitive impairment with components of frailty phenotype
| Univariate | Multivariable | ||||||
|---|---|---|---|---|---|---|---|
| OR | 95% CI OR | P value | OR | 95% CI OR | P value | ||
| Low physical activity | 1.996 | 1.596 – 2.497 | < 0.001 | 1.356 | 01.053 – 1.746 | 0.018 | |
| Exhaustion | 1.483 | 1.184– 1.858 | 0.001 | 0.643 | 0.464—0.891 | 0.008 | |
| Low handgrip force | 2.516 | 1.996—3.172 | < 0.001 | 1.817 | 1.409—2.343 | < 0.001 | |
| Weight loss 10 Ib | 1.853 | 1.269—2.708 | 0.001 | 1.447 | 0.971 – 2.155 | 0.069 | |
| Low walk speed | 2.145 | 1.692 – 2.718 | < 0.001 | 1.547 | 1.182 – 2.024 | 0.001 | |
*Adjusted for age, gender, classic education year, current smoking, PHQ-9 score, Barthel’s index, and diabetes mellitus
Subsequently, the association between impairment in some domains of cognition and frailty phenotype was assessed. In the linear regression analysis, after adjustment for years of schooling, Barthel scores and other relevant risk factors including age, gender, PHQ-9 score, and diabetes mellitus, the orientation score of MMSE was related to pre-frailty and frailty (standardized β = -0.074, P = 0.016). Moreover, CFT was associated with frailty as well (standardized β = -0.129, P < 0.001). In the multiple logistic regression analysis, memory was not related but executive functions were related to frailty (Table 5).
Table 5.
Univariate and multivariable association between domains of cognitive function and frailty phenotype
| Univariate | Multivariable | ||||||
|---|---|---|---|---|---|---|---|
| β | B (Se B) | P value | β | B (Se B) | P value | ||
| Orientation (MMSE) | Robust | ||||||
| Pre-frail | -0.089 | -0.166 (0.055) | 0.003 | -0.078 | -0.146 (0.061) | 0.016 | |
| Frail | -0.155 | -0.750 (0.143) | < 0.001 | -0.137 | -0.662 (0.161) | < 0.001 | |
| Calculation & Attention (MMSE) | Robust | ||||||
| Pre-frail | -0.045 | -0.165 (0.109) | 0.131 | 0.027 | 0.099 (0.117) | 0.401 | |
| Frail | -0.106 | -1.019 (0.284) | < 0.001 | -0.031 | -0.300 (0.312) | 0.337 | |
| Language Fluency (CFT) | Robust | ||||||
| Pre-frail | -0.174 | -1.764 (0.210) | < 0.001 | -0.083 | -0.844 (0.224) | < 0.001 | |
| Frail | -0.235 | -4.437 (0.390) | < 0.001 | -0.129 | -2.441 (0.457) | < 0.001 | |
| OR | 95% CI OR | P value | OR | 95% CI OR | P value | ||
| Memory (Min-Cog) | Robust | ||||||
| Pre-frail | 1.218 | 1.016 – 1.461 | 0.033 | 0.951 | 0.719 – 1.259 | 0.725 | |
| Frail | 1.535 | 1.075 – 2.192 | 0.018 | 0.878 | 0.515 – 1.501 | 0.637 | |
|
Execution Function ( Min-Cog) |
Robust | ||||||
| Pre-frail | 2.334 | 1.943 – 2.804 | < 0.001 | 1.457 | 1.171 – 1.813 | 0.001 | |
| Frail | 9.543 | 5.463 – 16.668 | < 0.001 | 3.655 | 1.943 – 6.877 | < 0.001 | |
*Adjusted for age, gender, classic education year, current smoking, PHQ-9 score, Barthel’s index, and diabetes mellitus
Sensitivity analyses
In order to evaluate the role of how to define cognitive impairment on the reported results, cognitive impairment was defined based on each of the tools separately and the analyzes were repeated. When cognitive impairment was defined in terms of mini-gags and CFTs, it was still associated with phenotype. When cognitive impairment was defined in terms of Mini-Cog and CFT, it was still associated with frailty phenotype, while this association was not existed with cognitive impairment defined only by MMSE. [Table 6 appendix].
Table 6.
Appendix- Association between cognitive impairment with different definition and frailty phenotype A in multivariable logistic regression models*. A. Defined by Mini-Cog, B. Defined by CFT, C. Cognitive impairment and normal cognition defined by both Mini-Cog and CFT, D. Defined by MMSE
| Cognitive impairment defined by only using Mini-Cog | ||||
|---|---|---|---|---|
| Frailty Syndrome | Robust | Reference group | ||
| Pre-frail | 1.296 | 1.073—1.567 | 0.007 | |
| Frail | 1.980 | 1.323—2.963 | 0.001 | |
| Cognitive impairment defined by only using CFT | ||||
| Robust | ||||
| Pre-frail | 1.419 | 1.167 – 1.724 | < 0.001 | |
| Frail | 2.669 | 1.795 – 3.971 | < 0.001 | |
| Cognitive impairment if the subjects had problem in both Mini-Cog and CFT and they were considered as normal cognitive if subjects no have problem in both tests | ||||
| Robust | ||||
| Pre-frail | 1.603 | 1.248 – 2.057 | < 0.001 | |
| Frail | 4.702 | 2.618 – 8.445 | < 0.001 | |
| Cognitive impairment defined by only using MMSE | ||||
| Robust | ||||
| Pre-frail | 1.533 | 1.021 – 2.302 | 0.039 | |
| Frail | 2.939 | 1.296 – 6.667 | 0.010 | |
*Adjusted for age, gender education levels, BMI, current smoking, depression, diabetes mellitus, and Parkinson’s disease
Discussion
We found that cognitive impairment was related to frailty phenotype in a cross-sectional analysis. This relationship was observed with the different definitions of cognitive impairment. We also found that low handgrip and low gait speed had the strongest associations with cognitive impairment. Moreover, with an increase in the number of existing components of frailty, the association has become stronger only up to three components. Our findings confirm the results of another study [21] that reported cognitive performance is related to the motor/muscle-skeletal systems (grip strength and gait speed). Physical frailty and cognitive dysfunction are common aging processes [22–24]. Physical frailty and cognitive impairment may have common biological pathways [25].
Moreover, in our study, there was an increasing trend of odds ratio for cognitive impairment with adding several features of frailty phenotype (up to three features), which were concordant with frailty’s definition. Some studies have reported that combined features such as grip strength and gait speed, as objective performance-based measures, have a greater association with cognitive status than a single feature, namely balance or strength [25, 26]. Gait speed is considered a powerful predictor of outcomes (mobility, functional status, and mortality) in the older adult population [27]. Age-related decline in gait speed is accompanied by multiple organs’ failure [28]. A study also has observed that slow gait speed, low physical activity, weight loss, and cognitive impairment can be used as good indicators for determining frailty, while self-reported features may have less value for the diagnosis of this syndrome [29]. The decline of gait speed and grip strength before the onset of cognitive impairment may occur in conjunction with vascular dementia [30]. Older people with dementia have shorter step lengths, more step-to-step variations, and slower gait speed [31]. Gait speed is affected by episodic memory impairments or executive function disorders [32]. Gray matter atrophy in the prefrontal cortex and hippocampal regions is associated with slower gait speed [33]. In a cohort study, decreased muscle strength was associated with a mild increase in cognitive impairment (MCI) incident and with a higher rate of decline in cognitive functions[34].
We observed a statistically significant association between cognitive impairment and weight loss as a component of frailty phenotype in multivariable models. Yessed et al. could not find any relationship between cognitive impairment, detected by using the Mini-Mental State Examination (MMSE) and weight loss and exhaustion [25]. A Taiwanese longitudinal aging study also reported that weight loss was the least sensitive component of frailty and had the weakest association with cognitive function [35]. Furthermore, another study has reported that BMI is not a good predictor of cognitive impairment [21]. In contrast, weight decline has been reported by some studies to accompany cognitive impairment [36, 37].
Also, we found that exhaustion is not related to cognitive impairment. Exhaustion could be a symptom of severe depression; these subjects were excluded from our analysis. On the other hand, exhaustion is a subjective complaint and it is possible that subjects with cognitive impairment did not correctly state the complaint. We found that the female sex is a risk factor for cognitive impairment. This could be explained by fewer social roles that older women have in the community, which require higher neurocognitive functioning levels; Another reason could also be an abrupt decrease of sexual hormones in women, which has a significant role in normal brain functioning. Cognitive impairment was 37% more common in subjects with diabetes mellitus than those without diabetes in our study in multiple analysis. Higher blood glucose may be associated with an increased risk of poor cognitive functioning [38].
According to our observation, there is a higher risk of cognitive impairment in a group with frailty than those with pre-frailty. Other studies have reported that frail older adults had a higher rate of cognitive impairment than those who were pre-frail and robust [39]. The susceptibility of cognitive domains in different frailty status may provide a new view to assess the pathogenesis of both frailty and cognitive impairment.
This study had a large sample and the participants of this study were representative of community-dwelling older adults in an urban region. This sampling decreased the selection bias in our study. Moreover, we assessed cognitive status by three tools and we performed sensitive analyses for the different definitions of cognitive impairment. We enrolled a high proportion of illiterate subjects (due to the representativeness of the population) and it may limit the interpretation of our results compared to other higher-educated aged groups. The results of this study are limited for causality interference because of the cross-sectional approach in our analyses. Moreover, some components of frailty might be with information bias, including the weight loss during the previous year and feeling of exhaustion. In addition, physical activity was defined based on a self-report during the last week and this may have caused bias.
Conclusion
Cognitive impairment might be related to frailty phenotype and the probability of cognitive impairment in pre-frail subjects was higher than normo-cognitive older adults. In addition, the prevalence of cognitive dysfunction was higher in people who were frail than in those who were pre-frail. Cognitive impairment may be a clinical feature of frailty or common background factors may predispose people to both of these health conditions. Therefore, pre-frail elderly individuals should also be considered as a group who are at risk of cognitive deterioration and preventive strategies should be considered for them too. The geriatric multidisciplinary approach is likely to be the most effective way to meet the needs of pre-frail elderly individuals. Longitudinal approaches can help the researchers to make a better judgment on the temporal precedence of frailty and cognitive impairment. Furthermore, other studies that assess cellular and molecular pathways may reveal common pathways for development of cognitive impairment and frailty.
Appendix
Declarations
Conflict of interest
On behalf of all authors, the corresponding author states that there is no conflict of interest.
Footnotes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Hoogendijk EO, et al. Frailty: implications for clinical practice and public health. The Lancet. 2019;394(10206):1365–1375. doi: 10.1016/S0140-6736(19)31786-6. [DOI] [PubMed] [Google Scholar]
- 2.Gómez-Gómez ME, Zapico SC. Frailty, cognitive decline, neurodegenerative diseases and nutrition interventions. Int J Mol Sci. 2019;20(11):2842. doi: 10.3390/ijms20112842. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Feng L, et al. Physical frailty, cognitive impairment, and the risk of neurocognitive disorder in the Singapore longitudinal ageing studies. J Gerontol A. 2017;72(3):369–375. doi: 10.1093/gerona/glw050. [DOI] [PubMed] [Google Scholar]
- 4.Calleja MR, et al. Sarcopenia, frailty, cognitive impairment and mortality in elderly patients with non-valvular atrial fibrillation. Rev Clín Española (English Edition) 2019;219(8):424–432. doi: 10.1016/j.rceng.2019.04.002. [DOI] [PubMed] [Google Scholar]
- 5.Son Y-J, Lee K, Kim B-H. Gender differences in the association between frailty, cognitive impairment, and self-care behaviors among older adults with atrial fibrillation. Int J Environ Res Public Health. 2019;16(13):2387. doi: 10.3390/ijerph16132387. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Borges MK, et al. Frailty as a predictor of cognitive disorders: a systematic review and meta-analysis. Front Med. 2019;6:26. doi: 10.3389/fmed.2019.00026. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Sargent L, et al. Shared biological pathways for frailty and cognitive impairment: a systematic review. Ageing Res Rev. 2018;47:149. doi: 10.1016/j.arr.2018.08.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Ostovar A, et al. Bushehr Elderly Health (BEH) Programme, phase I (cardiovascular system) BMJ Open. 2015;5(12):e009597. doi: 10.1136/bmjopen-2015-009597. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Shafiee G, et al. Bushehr Elderly Health (BEH) programme: study protocol and design of musculoskeletal system and cognitive function (stage II) BMJ Open. 2017;7(8):e013606. doi: 10.1136/bmjopen-2016-013606. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Control, C.f.D. and Prevention, Anthropometry Procedures Manual. National Health and Nutrition Examination Survey (NHANES) available from: https://www.cdcgov/nchs/data/nhanes/nhanes_07_08/manual_an.pdf. 2007.
- 11.Classification and Diagnosis of Diabetes Standards of medical care in diabetes—2021. Diabetes Care. 2021;44(Supplement 1):S15–S33. doi: 10.2337/dc21-S002. [DOI] [PubMed] [Google Scholar]
- 12.Aadahl M, Jørgensen T. Validation of a new self-report instrument for measuring physical activity. Med Sci Sports Exerc. 2003;35(7):1196–1202. doi: 10.1249/01.MSS.0000074446.02192.14. [DOI] [PubMed] [Google Scholar]
- 13.Fried LP, et al. Frailty in older adults: evidence for a phenotype. J Gerontol A Biol Sci Med Sci. 2001;56(3):M146–M156. doi: 10.1093/gerona/56.3.M146. [DOI] [PubMed] [Google Scholar]
- 14.Rezaei M, et al. Psychometric properties of the Persian adaptation of mini-cog test in Iranian older adults. Int J Aging Human Dev. 2018;86(3):266–280. doi: 10.1177/0091415017724547. [DOI] [PubMed] [Google Scholar]
- 15.Brucki SMD, Rocha MSG. Category fluency test: effects of age, gender and education on total scores, clustering and switching in Brazilian Portuguese-speaking subjects. Braz J Med Biol Res. 2004;37(12):1771–1777. doi: 10.1590/S0100-879X2004001200002. [DOI] [PubMed] [Google Scholar]
- 16.Wu J, et al. Relation of cigarette smoking and alcohol drinking in midlife with risk of cognitive impairment in late life: The Singapore chinese health study. Age and ageing. 2019;48(1):101–107. doi: 10.1093/ageing/afy166. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Hormozi S et al. Iranian version of barthel index: validity and reliability in outpatients' elderly. Int J Prevent Med. 2019;10. [DOI] [PMC free article] [PubMed]
- 18.Gallo JJ and Paveza G. Activities of daily living and instrumental activities of daily living assessment. Handb Geriatr Assess. 2006;193–240.
- 19.Mirzadeh FS et al. Validity and reliability: The Iranian version of Lawton IADL in elderly community dwellers. J Public Mental Health. 2020.
- 20.Manea L, Gilbody S, McMillan D. Optimal cut-off score for diagnosing depression with the Patient Health Questionnaire (PHQ-9): a meta-analysis. Cmaj. 2012;184(3):E191–E196. doi: 10.1503/cmaj.110829. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Boyle PA, et al. Physical frailty is associated with incident mild cognitive impairment in community-based older persons. J Am Geriatr Soc. 2010;58(2):248–255. doi: 10.1111/j.1532-5415.2009.02671.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Langlois F, et al. The multiple dimensions of frailty: physical capacity, cognition, and quality of life. Int Psychogeriatr. 2012;24(9):1429–1436. doi: 10.1017/S1041610212000634. [DOI] [PubMed] [Google Scholar]
- 23.Rolfson DB, et al. An assessment of neurocognitive speed in relation to frailty. Age Ageing. 2013;42(2):191–196. doi: 10.1093/ageing/afs185. [DOI] [PubMed] [Google Scholar]
- 24.Han ES, Lee Y, Kim J. Association of cognitive impairment with frailty in community-dwelling older adults. Int Psychogeriatr. 2014;26(1):155. doi: 10.1017/S1041610213001841. [DOI] [PubMed] [Google Scholar]
- 25.Yassuda MS, et al. Frailty criteria and cognitive performance are related: data from the FIBRA study in Ermelino Matarazzo, São Paulo, Brazil. J Nutr Health Aging. 2012;16(1):55–61. doi: 10.1007/s12603-012-0003-6. [DOI] [PubMed] [Google Scholar]
- 26.Holtzer R, et al. Neuroimaging of mobility in aging: a targeted review. J Gerontol A Biomed Sci Med Sci. 2014;69(11):1375–1388. doi: 10.1093/gerona/glu052. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Middleton A, Fritz SL, Lusardi M. Walking speed: the functional vital sign. J Aging Phys Act. 2015;23(2):314–322. doi: 10.1123/japa.2013-0236. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Rosso AL, et al. Multisystem physiologic impairments and changes in gait speed of older adults. J Gerontol A Biomed Sci Med Sci. 2015;70(3):319–324. doi: 10.1093/gerona/glu176. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Rothman MD, Leo-Summers L, Gill TM. Prognostic significance of potential frailty criteria. J Am Geriatr Soc. 2008;56(12):2211–2216. doi: 10.1111/j.1532-5415.2008.02008.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Buchman AS, Bennett DA. Loss of motor function in preclinical Alzheimer’s disease. Expert Rev Neurother. 2011;11(5):665–676. doi: 10.1586/ern.11.57. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Rosano C, Rosso AL, and Studenski SA. Aging, brain, and mobility: progress and opportunities. Oxford University Press. 2014. [DOI] [PubMed]
- 32.Beauchet O, et al. Episodic memory and executive function impairments in non-demented older adults: which are the respective and combined effects on gait performances? Age. 2015;37(4):70. doi: 10.1007/s11357-015-9812-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Makizako H, et al. The association between decline in physical functioning and atrophy of medial temporal areas in community-dwelling older adults with amnestic and nonamnestic mild cognitive impairment. Arch Phys Med Rehabil. 2011;92(12):1992–1999. doi: 10.1016/j.apmr.2011.07.195. [DOI] [PubMed] [Google Scholar]
- 34.Boyle PA, et al. Association of muscle strength with the risk of Alzheimer disease and the rate of cognitive decline in community-dwelling older persons. Arch Neurol. 2009;66(11):1339–1344. doi: 10.1001/archneurol.2009.240. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Wu Y-H, et al. Cognitive function in individuals with physical frailty but without dementia or cognitive complaints: results from the i-lan longitudinal aging study. J Am Med Directors Assoc. 2015;16(10):899.e9–899.e16. doi: 10.1016/j.jamda.2015.07.013. [DOI] [PubMed] [Google Scholar]
- 36.Alhurani RE, et al. Decline in weight and incident mild cognitive impairment: Mayo Clinic Study of Aging. JAMA Neurol. 2016;73(4):439–446. doi: 10.1001/jamaneurol.2015.4756. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Cova I, et al. Weight loss predicts progression of mild cognitive impairment to Alzheimer’s disease. PloS one. 2016;11(3):e0151710. doi: 10.1371/journal.pone.0151710. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Crane PK, et al. Glucose levels and risk of dementia. New England J Med. 2013;369(6):540–548. doi: 10.1056/NEJMoa1215740. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Ávila-Funes JA, et al. Cognitive impairment improves the predictive validity of the phenotype of frailty for adverse health outcomes: The three-city study. J Am Geriatr Soc. 2009;57(3):453–461. doi: 10.1111/j.1532-5415.2008.02136.x. [DOI] [PubMed] [Google Scholar]