Protective and Risk Factors in Daily Life Associated with Cognitive Decline of Older Adults

doi:10.21203/rs.3.rs-4675489/v1

Download PDF

Research Article

Protective and Risk Factors in Daily Life Associated with Cognitive Decline of Older Adults

https://doi.org/10.21203/rs.3.rs-4675489/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background: Cognitive decline is a chronic condition which is characterized by a loss of the ability to remember, learn, and pay attention to complex tasks. Many older people are now suffering from cognitive decline, which decreases life quality and leads to disability. This study aimed to identify the risk and protective factors for cognitive decline of the older people from daily life and establish a predictive model using logistic regression.

Methods: We investigated 3,790 older people with health examination and questionnaires which included information associated with physical condition, lifestyle factors, and cognitive status. Single-factor comparison, principal component analysis with a Manova-Wilk test, and logistic regression were performed to filter the risk and protective factors regarding cognitive decline of older individuals.

Results: We found a significant separation along the coordinate axis between people with normal and declined cognition by principal component analysis, as confirmed by the Manover-Wilk test. Single-factor comparison and logistic regression demonstrated significant difference in gender, age, hypertension level, height, dietary habit, physical-exercise duration, physical-exercise history, and smoking history between the two groups. We also observed significant differences in height, physical exercise duration, physical-exercise years, and smoking years between the male and female of the participants. ROCs of the predictive model by logistic regression were plotted, with AUC values of 0.683 and 0.682 respectively for the training and testing sets.

Conclusion: The protective factors of cognitive decline for older people were male gender, height, keeping exercising, and nicotine stimulation, and the risk factors included age, female gender, vegetarianism, hypertension, and over-exercise. Except for the genetic factor, differences in lifestyle, such as smoking and exercise habits, may contribute to the observed differences in cognitive function between genders.

Protective and risk factors

Cognitive decline

Older adults

Cognitive function

Cognitive decline is a chronic condition which is characterized by a loss of the ability to remember, learn, and pay attention to complex tasks, incomparable with the educational level (1, 2). The clinical manifestation of cognitive decline is minor and difficult to observe in the early stage, and will eventually proceed to dementia without medical intervention. Globally, approximately 47 million people have been diagnosed with dementia and this number is increasing by 8 million annually (2). The dementia population is expected to be close to 140 million in 2050 (3). Aging is proved to be positively associated with cognitive decline. For older people, cognition determines life expectancy, as well as life quality, including independent living, effective communication, finance management, drug administration and safe driving. Age-related cognitive decline is a serious issue which human beings must confront in this century. Nowadays, the life expectancy of most people is over 60 (4), and children alive are extremely optimistic that they could live beyond 100 years (5). Many older people are suffering from cognitive decline, which is decreasing life quality and may lead to disability (6). Alzheimer’s disease (AD) and vascular dementia are the top causes of dementia in older people (7, 8). Unfortunately, no effective method has been established to treat dementia. Thus, risk and protective factors of cognitive decline should be disclosed to make appropriate prevention.

Cognitive performance can be assessed using various psychological scales, such as the Mini-Mental State Examination (MMSE), Montreal Cognitive Assessment (MoCA), Rowland Universal Dementia Assessment Scale (RUDAS), and Cognitive Impairment Scale (PAS). Among these scales, MMSE is widely used in clinical and research settings due to its brevity and usefulness. It is a brief questionnaire consisting of 30 points that can sensitively test cognitive decline (9). MMSE is utilized to gauge the severity and progression of cognitive decline, as well as to monitor the evolution of cognitive changes in an individual over time (10). The examination is made up of 8 sections which are orientation to time, orientation to place, registration, attention and calculation, recall, language, repetition, and complex commands. People with normal cognition have an MMSE score over 24, and scores below this indicate cognitive decline of different degrees: severe (0–9 points), moderate (10–18 points), and mild (19–23 points) (11).

Risk and protective factors of cognitive decline are complicated and multifaceted, including aging, living habits, nutrition, educational level, vascular pathology, sleep-disordered breathing, Apolipoprotein E (APOE) genotype, lack of Vitamin D, and earlier severe diseases (12, 13). Advanced age is an independent cause for cognitive decline, and other risk factors are biological, socioeconomic, and environmental, which can be intervened to reduce the incidence rate (14). This study conducted in the Youyi community of Shanghai investigated 3,790 older people with health examination and questionnaires which included information associated with physical condition, lifestyle factors, and cognitive status. We aimed to identify the risk and protective factors for cognitive decline from daily life and establish a predictive model using logistic regression based on the results of the questionnaire.

Participants and Study Design

The study included participants who took part in a health examination organized by a community hospital in the Baoshan district of Shanghai. The data were collected from senior citizens residing in the Youyi Community in November 2020, during a period of strict Covid-19 policy enforcement in Shanghai. This ensured that the impact of virus exposure could be excluded. Participants provided informed consent prior to the health examination. The survey, conducted by experienced clinical doctors from the hospital, comprised three parts: general physical condition (including blood pressure, respiratory rate, height, weight, BMI), lifestyle habits (diet, physical exercise frequency and duration, years of exercise, smoking habits, daily cigarette consumption, years of smoking, alcohol consumption, years of drinking), and cognitive status (assessed using the MMSE). The questionnaires have never been published before and could be attached in the supplementary material 1. Questionnaires with missing content or apparent logical errors were excluded from the analysis. A total of 3708 samples were returned, yielding an effective response rate of 97.83%. Figure 1 presented a flowchart illustrating the inclusion procedure. The study received approval from the Ethic Review Board of Shanghai University of Medicine and Health Sciences and Shanghai Medical College of Fudan University.

Definition, Classification, and Assignment

Cognitive status was defined and classified based on the MMSE score. Participants with an MMSE score above 24 were assigned to the normal cognition group (assigned as 0), while those with a score between 0 and 23 were categorized as having cognitive decline (assigned as 1). Gender was assigned as 0 for females and 1 for males. Blood pressure (BP) was classified according to the American Heart Association’s BP category standards and assigned different values from 0 to 4. Self-assessment of health condition, dietary habits, physical exercise frequency, smoking addiction level, and alcohol addiction level were also classified and assigned specific values. More information on the classification and assignment of variables can be found in Supplementary Materials 2.

Statistical Analysis

Characteristics were statistically described as mean ± standard error mean (SEM). Basic statistical analysis and graph generation were performed using GraphPad Prism 10.0 software (GraphPad Software, San Diego, CA). Principal Component Analysis (PCA) with Manova-Wilk test, logistic regression, and nomograph were performed based on R Studio (software version 4.2.2). P value thresholds are set to 0.1 (weak evidence), 0.05 (moderate evidence), 0.01 (strong evidence), 0.001 (very strong evidence) and 0.0001 (extremely strong evidence). For the predictive model, samples were randomly assigned to a training set (70%) and a test set (30%). The model was built based on the training set and the accuracy was checked on the test set. We share all the codes associated with analysis on R studio in the Supplementary Material 3.

Sample distribution and variables

Figure 2A illustrated the distribution of samples from two groups: those with normal cognition and those experiencing cognitive decline. A significant separation along the coordinate axis was observed, as confirmed by Manover-Wilk test (P < 0.0001). Figure 2B presented the principal components of the two dimensions. Dimension 1 contrasted individuals with a strongly positive coordinate on the axis (right side of the graph) against those with a strongly negative coordinate (left side of the graph). Variables such as years of smoking, smoking addiction, daily cigarette consumption, years of drinking, drinking addiction, gender, height, weight, frequency of physical exercise, and self-assessment of health condition significantly influence the sample position on the positive coordinate axis (indicating normal cognition). Conversely, variables like body temperature and age significantly influence the sample position on the negative coordinate axis (indicating cognitive decline). Dimension 2 contrasted individuals with a strongly positive coordinate on the axis (top of the graph) against those with a strongly negative coordinate (bottom of the graph). Variables such as frequency of physical exercise, years of physical exercise, duration of physical exercise, and pulse rate significantly influence the sample position on the positive coordinate axis (indicating normal cognition). Again, variables like body temperature and age significantly influenced the sample position on the negative coordinate axis (indicating cognitive decline). For more details on the correlation values of these variables, please refer to Supplementary Materials 4.

Figure 2A illustrated the distribution of samples from groups with normal cognition and cognitive decline. A significant separation along the coordinate axis was observed, as confirmed by the Manover-Wilk test (P < 0.0001). Figure 2B demonstrated the principal components of the two dimensions. Dimension 1 contrasts individuals with a strongly positive coordinate on the axis (right side of the graph) against those with a strongly negative coordinate (left side of the graph). Variables such as years of smoking, smoking addiction, daily cigarette consumption, years of drinking, drinking addiction, gender, height, weight, frequency of physical exercise, and self-assessment of health condition significantly influenced the sample position on the positive coordinate axis (indicating normal cognition). Conversely, variables like body temperature and age significantly influence the sample position on the negative coordinate axis (indicating cognitive decline). Dimension 2 contrasted individuals with a strongly positive coordinate on the axis (top of the graph) against those with a strongly negative coordinate (bottom of the graph).

Characteristics of samples

Table 1 presented the baseline characteristics of all samples, categorized by normal cognition and cognitive decline. Of the 3708 individuals, 3048 exhibited normal cognition, while 660 showed signs of cognitive decline. The group with cognitive decline had a higher proportion of females (P < 0.0001) and were generally older (P < 0.0001) than those with normal cognition. This group also had a lower proportion of individuals with normal and elevated blood pressure, but a higher proportion with high blood pressure (P < 0.0001). Individuals with normal cognition had higher pulse rates (P < 0.05), as well as greater height and weight (P < 0.0001 for both) than those with cognitive decline. Dietary habits between the two groups showed weakly significant differences (P < 0.1), with the cognitive decline group having a higher proportion of individuals consuming ‘vegetables only’ and ‘more vegetables than meat’. In terms of physical exercise, the cognitive decline group had a higher proportion of individuals who ‘never exercise’ (P < 0.01) and a longer average duration of physical exercise, although the latter was only weakly significant (P < 0.1). The normal cognition group, on the other hand, had more years of physical exercise (P < 0.05). Regarding smoking habits, 92.09% of the normal cognition group declared ‘never smoke’, compared to 96.97% in the cognitive decline group (P < 0.0001). The normal cognition group smoked approximately 1.007 cigarettes per day, which was higher than the 0.3985 per day in the cognitive decline group (P < 0.0001). People from the group with normal cognition also had around 2 more years of smoking (P < 0.0001) than the other. In terms of alcohol addiction, 93.73% of the normal cognition group reported never having smoked, compared to 96.97% in the cognitive decline (P = 0.0013). The ‘drinking years’ for the normal cognition group was approximately 2.83, compared to 1.455 for the cognitive decline group (P < 0.01). No significant differences were observed in body temperature, pulse rate, self-assessment of health, respiratory rate, and BMI between the two groups.

Table 1

Comparison of the characteristics a Chi-square test is used for the comparison of the classification variables.
Characteristics	Total (N = 3708)	Normal Cognition (n = 3048)	Cognitive Decline (n = 660)	P value
gender				< 0.0001^***
male	1466 (39.54%)	1285 (42.16%)	181 (27.42%)
female	2242 (60.46%)	1763 (57.84%)	479 (72.58%)
Age (years)	72.05 (0.09199)	71.52 (0.09533)	74.51 (0.2496)	< 0.0001^***
Hypertension level ^a
no hypertension	197 (5.31%)	148 (4.86%)	31 (4.70%)	< 0.0001^***
elevated	1366 (36.84%)	1167 (38.29%)	199 (30.15%)
high blood pressure stage 1	1777 (47.92%)	1408 (46.19%)	369 (55.91%)
high blood pressure stage 2	382 (10.30%)	323 (10.60%)	59 (8.94%)
high blood pressure crisis	4 (0.11%)	2 (0.066%)	2 (0.30%)
body temperature (ºC)	36.55 (0.003612)	36.55 (0.003939)	36.55 (0.009002)	0.7808
pulse rate (/min)	73.48 (0.06783)	73.52 (0.07385)	73.33 (0.17)	0.0265^*
self-assessment of health ^a
dissatisfactory	1 (0.03%)	1 (0.03%)	0 (0%)	0.8612
not quite satisfactory	9 (0.24%)	8 (0.26%)	1 (0.15%)
basically satisfactory	3489 (94.09%)	2871 (94.19%)	618 (93.64%)
satisfactory	204 (5.50%)	165 (5.41%)	39 (5.91%)
respiratory rate (/min)	18.12 (0.0205)	18.12 (0.02258)	18.12 (0.04898)	0.9004
Height (cm)	161.6 (0.1354)	162.1 (0.1492)	159.0 (0.3035)	< 0.0001^***
Weight (kg)	63.48 (0.1641)	63.90 (0.182)	61.55 (0.3696)	< 0.0001^***
BMI	24.35 (0.07598)	24.36 (0.08822)	24.32 (0.1274)	0.9135
dietary habit ^a
vegetable only	42 (1.13%)	29 (0.95%)	13 (1.97%)	0.054^#
more vegetables than meat	6 (0.16%)	4 (0.13%)	2 (0.30%)
balanced diet	3635 (98.03%)	2991 (98.13%)	644 (97.58%)
more meat than vegetables	1 (0.03%)	1 (0.03%)	0 (0%)
meat only	24 (0.65%)	23 (0.75%)	1 (0.15%)
physical-exercise frequency				0.0044**
never do exercise	1091 (29.42%)	859 (28.18%)	232 (35.15%)
do exercise occasionally	69 (1.86%)	59 (1.94%)	10 (1.52%)
more than once a week	38 (1.02%)	33 (1.08%)	5 (0.76%)
every day	2510 (67.69%)	2097 (68.80%)	413 (62.58%)
physical-exercise duration (min)	34.62 (0.6196)	34.60 (0.6203)	34.72 (1.979)	0.0685^#
physical-exercise years	7.392 (0.1081)	7.484 (0.1182)	6.968 (0.2653)	0.013^*
smoking addiction ^a				< 0.0001^***
never smoke	3447 (92.96%)	2807 (92.09%)	640 (96.97%)
used to smoke, not smoke now	69 (1.86%)	62 (2.03%)	7 (1.06%)
smoke now, but not every day	7 (0.19%)	5 (0.16%)	2 (0.30%)
smoke everyday	185 (4.99%)	174 (5.71%)	11 (1.67%)
daily-cigarette number	0.8986 (0.06543)	1.007 (0.07496)	0.3985 (0.1219)	< 0.0001^***
smoking years	2.934 (0.1851)	3.332 (0.2164)	1.092 (0.2766)	< 0.0001^***
alcohol addiction ^a				0.0013^***
never drink	3497 (94.31)	2857 (93.73)	640 (96.97)
drink occasionally	63 (1.70)	60 (1.97)	3 (0.45)
always drink but not every day	21 (0.57)	18 (0.59)	3 (0.45)
drink everyday	127 (3.43)	113 (3.71)	14 (2.12)
drinking years	2.585 (0.1809)	2.830 (0.2073)	1.455 (0.3385)	0.0024^***

Risk and protective factors and predictive model

Table 2 presented the risk and protective factors of cognitive decline, as determined by a univariate logistic regression model. Significant risk factors included age (P < 0.0001, OR = 1.9399, 95%CI: 1.71–2.21), hypertension level (P < 0.1, OR = 1.1377, 95%CI: 0.986–1.31), and physical-exercise duration (P < 0.01, OR = 1.2578, 95%CI: 1.07–1.48). Significant protective factors were male gender (P < 0.1, OR = 0.72442, 95%CI: 0.519–1.01), dietary habit (P < 0.05, OR = 0.1888, 95%CI: 0.0421–0.846), physical-exercise years (P < 0.1, OR = 0.82723, 95%CI: 0.665–1.03), and smoking years (P < 0.1, OR = 0.147, 95%CI: 0.0202–1.07). These results are visualized in a nomograph (Fig. 3). The single-factor comparison and logistic regression shared common positive results for gender, age, hypertension level, height, dietary habit, physical-exercise duration, physical-exercise history, and smoking history. These variables were used as risk factors to build a predictive model using logistic regression. The ROCs of the training and testing sets were plotted, with AUC values of 0.683 and 0.682 respectively (Fig. 4). De Long tests showed that the 95%CI of the training set was 0.6561-0.71 and the testing set was 0.6399–0.7241 (Fig. 4). A calibration curve of the predictive model was plotted (Fig. 5), with the mean absolute error of 0.012, mean squared error of 0.00018, and 0.9 quantile of absolute error of 0.024.

Table 2

Results from logistic regression
Characteristics	Coefficient	S.E	P Value		Odds Ratio	95% CI
gender	-0.3224	0.1697	0.0575^#		0.72442	(0.519, 1.01)
age	0.0947	0.0094	< 0.0001^***		1.9399	(1.71, 2.21)
hypertension level	0.129	0.0729	0.0767^#		1.1377	(0.986, 1.31)
body temperature	-0.0955	0.2456	0.6974		0.97175	(0.841, 1.12)
pulse rate	-0.013	0.0128	0.308		0.94918	(0.859, 1.05)
self-assessment of health	0.0958	0.1838	0.6021	1.4672		(0.347, 6.20)
respiratory rate	-0.0121	0.0412	0.7684	0.86465		(0.328, 2.28)
height	-0.0388	0.017	0.0224^*	0.62785		(0.421, 0.936)
weight	0.0165	0.0156	0.291	1.2592		(0.821, 1.93)
BMI	-0.0384	0.0331	0.2458	0.86093		(0.669, 1.11)
dietary habit	-0.8335	0.3827	0.0294^*	0.1888		(0.0421, 0.846)
physical exercise
physical-exercise frequency	-0.0706	0.0524	0.1779	0.80904		(0.594, 1.10)
daily-physical-exercise duration	0.0038	0.0014	0.0052^**	1.2578		(1.07, 1.48)
physical-exercise years	-0.019	0.0111	0.0876^#	0.82723		(0.665, 1.03)
smoking
smoking addiction	0.2144	0.2166	0.3223	1.9027		(0.532, 6.80)
daily-cigarette number	0.0387	0.0279	0.1656	4.7078		(0.527, 42.1)
smoking years	-0.032	0.0169	0.0585^#	0.147		(0.0202, 1.07)
drinking
alcohol addiction	0.2072	0.2137	0.3322	1.8621		(0.530, 6.54)
drinking years	-0.0074	0.0115	0.5185	0.54772		(0.0881, 3.41)

The ROCs of the training and testing sets were plotted, with AUC values of 0.683 and 0.682 respectively. De Long tests showed that the 95%CI of the training set was 0.6561-0.71 and the testing set was 0.6399–0.7241.

A calibration curve of the predictive model was plotted, with the mean absolute error of 0.012, mean squared error of 0.00018, and 0.9 quantile of absolute error of 0.024.

Comparison of the risk and protective factors between genders

Table 3 presented the gender differences in the risk and protective factors. The males were found to be older than females in this study (P < 0.05). The males also had greater height (P < 0.0001) and spent more time on physical exercise (P < 0.001). The number of physical-exercise years was greater in the males than females, although the difference was weakly significant (P < 0.1). The difference in smoking years between genders was highly significant (P < 0.0001). No significant differences were observed in hypertension level and dietary habit between genders.

Table 3

Gender differences in the risk and protective factors
Characteristics	Male	Female	P value
Characteristics	(n = 1466)	(n = 2242)	P value
Age (years)	72.39(0.1507)	71.83(0.1157)	0.0131^*
Hypertension level ^a
no hypertension	65 (4.43%)	114 (5.08%)	0.5932
elevated	527 (35.95%)	839 (37.42%)
high blood pressure stage 1	713 (48.64%)	1064 (47.46%)
high blood pressure stage 2	160 (10.91%)	222 (9.90%)
high blood pressure crisis	1 (0.068%)	3 (0.13%)
Height (cm)	168.7(0.1700)	156.9(0.1149)	< 0.0001^***
dietary habit ^a
vegetable only	11 (0.75%)	31 (1.38%)	0.2666
more vegetables than meat	1 (0.068%)	5 (0.22%)
balanced diet	1445 (98.57%)	2190 (97.68%)
more meat than vegetable	0 (0%)	1 (0.045%)
meat only	9 (0.61%)	15 (0.67%)
daily-physical-exercise duration (min)	36.56(0.9251)	33.35(0.8263)	0.0007^***
physical-exercise years	7.696(0.1775)	7.194(0.1358)	0.0578^#
smoking years	7.369(0.4419)	0.03345(0.02758)	< 0.0001^***
a Chi-square test is used for the comparison of the classification variables.

Comparison of the risk and protective factors in the male and female participants

Table 4 presented the differences in risk and protective factors among the male participants. Out of 1466 individuals, 1285 displayed normal cognition, while 181 exhibited cognitive decline. Males with cognitive decline were significantly older (P < 0.0001) than those with normal cognition. Individuals with normal cognition were taller than those with cognitive decline (P < 0.01). The duration of physical exercise was longer in the normal group compared to the cognitive decline group (P < 0.05). Males with normal cognition had more years of physical exercise than those with cognitive decline (P < 0.01). Individuals with normal cognition had a longer history of smoking than those with cognitive decline (P < 0.01). No significant differences were found in hypertension levels and dietary habits between the groups.

Table 4

Comparison of the risk and protective factors in the male participants
Characteristics	Total (N = 1466)	Normal Cognition (n = 1285)	Cognitive Decline (n = 181)	P value
Age (years)	72.39 (0.1507)	72.00 (0.1550)	75.11 (0.4839)	< 0.0001^***
Hypertension level ^a
no hypertension	65 (4.433%)	54 (4.202%)	11 (6.077%)	0.5959
elevated	527 (35.948%)	468 (36.42%)	59 (32.597%)
high blood pressure stage 1	713 (48.636%)	617 (48.016%)	96 (53.039%)
high blood pressure stage 2	160 (10.914%)	145 (11.284%)	15 (8.287%)
high blood pressure crisis	1 (0.0682%)	1 (0.078%)	0 (0%)
Height (cm)	168.7 (0.17)	168.9 (0.1805)	167.4 (0.4942)	0.0016^*
dietary habit ^a
vegetable only	11 (0.75%)	11 (0.0856%)	0 (0%)	0.8348
more vegetables than meat	1 (0.0682%)	1 (0.078%)	0 (0%)
balanced diet	1445 (98.568%)	1264 (98.366%)	181 (100%)
more meat than vegetables	0 (0%)	0 (0%)	0 (0%)
meat only	9 (0.614%)	9 (0.7%)	0 (0%)
physical-exercise duration (min)	36.56 (0.9251)	37.05 (0.9861)	33.09 (2.663)	0.0402^*
physical-exercise years	7.696 (0.1775)	7.876 (0.19)	6.420 (0.487)	0.0028^**
smoking years	7.369 (0.4419)	7.846 (0.4836)	3.983 (0.9786)	0.0033^**
a Chi-square test is used for the comparison of the classification variables.

Table 5 illustrated the differences in risk and protective factors among the female participants. Of the 2242 individuals, 1763 displayed normal cognition, while 479 showed cognitive decline. Females with cognitive decline were significantly older (P < 0.0001) than those with normal cognition. Females with cognitive decline had a higher proportion of high blood pressure (P < 0.0001). Individuals with normal cognition were taller than those with cognitive decline (P < 0.0001). Significant differences were observed in dietary habits among female participants (P < 0.05). The proportion of females with cognitive decline following a ‘balanced diet’ was lower, and they consumed more vegetables in their diet. No significant differences were observed in ‘physical-exercise duration’, ‘physical-exercise years’, and ‘smoking years’.

Table 5

Comparison of the risk and protective factors in the female participants
Characteristics	Total (N = 2242)	Normal Cognition (n = 1763)	Cognitive Decline (n = 479)	P value
Age (years)	71.83 (0.1157)	71.17 (0.1194)	74.28 (0.2908)	< 0.0001^****
Hypertension level ^a
no hypertension	114 (5.085%)	94 (5.332%)	20 (4.175%)	< 0.0001^****
elevated	839(37.422%)	699 (39.648%)	140 (29.228%)
high blood pressure stage 1	1064 (47.458%)	791 (44.867%)	273 (56.994%)
high blood pressure stage 2	222 (9.902%)	178 (10.096%)	44 (9.186%)
high blood pressure crisis	3 (0.134%)	1 (0.057%)	2 (0.418%)
Height (cm)	156.9 (0.1149)	157.2 (0.1284)	155.8 (0.2514)	< 0.0001^****
dietary habit ^a
vegetable only	31 (1.383%)	18 (1.021%)	13 (2.714%)	0.0254^*
more vegetables than meat	5 (0.223%)	3 (0.17%)	2 (0.418%)
balanced diet	2190 (97.681%)	1727 (97.958%)	463 (96.66%)
more meat than vegetables	1(0.0446%)	1 (0.057%)	0 (0%)
meat only	15 (0.669%)	14 (0.794%)	1 (0.209%)
physical-exercise duration (min)	33.35 (0.8263)	31.54 (0.8116)	35.33 (2.536)	0.4382
physical-exercise years	7.194 (0.1358)	7.199 (0.15)	7.175 (0.3156)	0.4302
smoking years	0.0335 (0.0276)	0.0425 (0.0351)	0 (0)	0.4614
a Chi-square test is used for the comparison of the classification variables.

Numerous factors in daily life are closely linked to the cognitive health of older adults. Our research found that individuals with normal cognition and those with cognitive decline were distributed differently on the scale by PCA. This suggests that cognitive decline is not a random occurrence, but rather the result of complex factors in daily life. Upon further analysis, we identified several potential risk and protective factors associated with cognitive decline in older adults. These include age, dietary habits, hypertension, height, daily physical exercise duration, years of physical exercise, and years of smoking. Most of these factors, which encompass physical condition, disease, and lifestyle, are modifiable or preventable. Therefore, appropriate interventions could be performed to potentially improve and maintain brain function in older adults, thereby reducing the risk of cognitive decline (15).

Age, the only unmodifiable factor among the results mentioned, is positively associated with cognitive decline. The aging process in the brain differs from that in other organs of the human body, as neurons are highly differentiated and incapable of regeneration during the lifespan (16, 17). Aging can significantly impair neurogenesis, further damaging the cognitive function of older adults (16). While age cannot be changed, recent findings on Neural Stem Cells (NSCs) suggest that brain aging could be intervened. NSC treatment could maintain neurogenesis at a youthful state during aging, offering a promising approach to preventing age-related cognitive decline (16).

Dietary and nutritional elements are factors which can be adjusted to delay or prevent cognitive decline. It is widely recognized that an unhealthy diet is a major risk factor for cognitive decline (18). For instance, after Japan transitioned from a traditional dietary pattern to a Western one, the Alzheimer’s Disease (AD) rate increased from 1% in 1985 to 7% in 2008 (19). So, what constitutes a diet which is beneficial for cognitive function? People are often advised to consume more vegetables and fruits because these plant foods are low in calories and high in fiber, which is beneficial to the body and could reduce the chances of moderate or severe cognitive decline (20). Therefore, it seems reasonable to suggest that a vegan diet could prevent cognitive decline by potentially exerting neuroprotective effects. However, there is no solid evidence to confirm that a vegetarian diet can lower the risk of cognitive decline more than a diet that includes meat. A study conducted by Giem et al in 1993, which included 2984 participants, investigated the relationship between dietary patterns (vegetarian and meat-eating) and the incidence of dementia, reporting no significant differences between the two groups (21). From our perspective, a vegetarian diet does not decrease the risk of cognitive decline in older individuals. According to our findings, a diet with a very high proportion of vegetables can actually increase the risk of cognitive decline. Therefore, we suggest increasing the proportion of meat in the diet to ensure a diverse intake of nutrients from food. The combination of nutrients can have synergistic and/or antagonistic effects beyond single components, which may have antioxidant and anti-inflammatory effects on the brain and improve cognitive function (14, 22).

Physical activity is favored among older adults, as many believe that it can slow down body aging. Physical activity is also thought as an effective way to enhance cognitive function. Studies suggest that increasing physical activity can lower dementia risk by 3% (23, 24). Evidence shows that the progression of cognitive decline can be hindered or slowed by physical activity (25). Older individuals who engage in more physical exercise are more likely to maintain cognitive function and reduce dementia risk compared to those who are sedentary (23, 26). Furthermore, the progression from mild cognitive impairment (MCI) to dementia can be delayed by physical activity. (27). This study showed a close relationship between physical activity/exercise and cognitive function. On one hand, individuals with a longer history of exercise are less likely to experience cognitive decline. On the other hand, prolonged exercise duration may increase the risk of cognitive decline. The findings suggest that maintaining physical activity/exercise can help preserve the cognitive function of older people. However, the duration of exercise should be regulated to prevent overloading the body. It is well-known that heart function diminishes with aging. Therefore, the heart function of the older cannot sustain extended periods of exercise, which could lead to brain hypoxia. Based on our findings, we recommend that each exercise session should not exceed 30 minutes.

Vascular conditions, particularly hypertension, play a significant role in the progression of cognitive decline and dementia (28). Postmortem examinations have revealed neurovascular pathology in over 50% of Alzheimer’s disease (AD) patients (29, 30). If the genetic predisposition for dementia is not taken into account, hypertension during midlife can increase the likelihood of dementia in later life (31, 32). It is well understood that chronic hypertension can lead to neurovascular remodeling, reducing cerebral blood flow and ultimately contributing to tissue damage and cognitive decline. This study further substantiates that hypertension is a contributing factor to cognitive decline. Moreover, the risk of cognitive decline is directly proportional to the level of hypertension. It is crucial for older adults to monitor their blood pressure regularly, and lifelong treatment is necessary if hypertension is diagnosed.

It is widely recognized that smoking is a detrimental habit, linked to a multitude of diseases such as lung cancer, hypertension, hyperlipidemia, pneumonia, and chronic obstructive pulmonary disease. It seems reasonable to assume that smoking could lead to cognitive decline due to neurovascular lesions. However, this study suggests a contrasting narrative: older adults with a longer history of smoking appear to have a lower risk of cognitive decline. This unexpected finding could be attributed to the neurological effects of nicotine, which has been shown to enhance cognitive function at typical smoking levels (33). Nicotine exerts its influence by interacting with nicotinic acetylcholine receptors (nAChRs), ligand-gated ion channels, which are formed by various pentameric combinations (34). These combinations arrange a central pore that allows the passage of sodium, potassium, and calcium ions (35). The majority of neuronal nAChRs in the brain are excitatory and fast-acting, modulating the release of other neurotransmitters including acetylcholine (ACh), dopamine (DA), serotonin, glutamate, GABA, and norepinephrine (34, 35). The prefrontal cortex and hippocampus, regions with a high concentration of nAChRs, are associated with the cognitive effects of nicotine (36, 37). Cognitive function is enhanced due to improvements in signal-to-noise ratios or the facilitation of synaptic plasticity in specific neural circuits within these two regions. (36, 38). It is important to note that we do not advocate for older adults to take up smoking as a means to improve cognitive function. This is due to the other toxic effects of cigarettes, which can ultimately lead to diseases in the vascular and respiratory systems.

This study revealed an interesting result: height is positively associated with cognition in older adults, which aligns with findings from previous studies (39, 40). Additionally, height is considered a protective factor against age-related cognitive impairment (41). Surprisingly, genetic factors do not account for the relationship between height and cognition, as demonstrated by twin studies (39). The association between height and cognition may be influenced by social and environmental factors. Height is often biasedly linked to positive attributes such as intelligence, competence, attractiveness, and social status (39). Consequently, taller individuals tend to receive more favorable treatment from a young age. As a result, taller people are likely to have better education, higher income, and greater access to medical resources, all of which positively impact cognition. These, altogether, exert positive influence on cognition. Diet during early childhood could play a crucial role in determining height. Nutritional adequacy during youth may have lasting effects on cognitive abilities throughout the lifespan. Notably, taller children often demonstrate enhanced cognitive abilities, which could be attributed to sufficient nutrient intake (42). Collectively, the associations between height and cognition are extensive, resulting from a combination of social and environmental factors.

The differences between males and females span various aspects, including behavior, thought patterns, emotions, and cognition. Our study revealed a gender bias in cognitive decline, with females more likely to exhibit cognitive decline in their later years. This could be attributed to the differing brain structures between the two genders (43). Studies based on Magnetic Resonance Imaging (MRI) have shown that the volumes of the amygdala and thalamus in males are larger than those in females (44–46), while females have a larger hippocampus (46, 47). It is worth noting that amygdala contains high concentrations of androgen receptors (48), while hippocampus have high number of estrogen receptors (49). As we know, the estrogen level in females drastically decreases with aging, impacting hippocampal function and contributing to cognitive decline. In contrast, the androgen level in males decreases gradually with aging, potentially helping to maintain cognitive function in later years. We compared the risk and protective factors among participants of both genders respectively. Our findings revealed that men with normal cognition tend to exercise for longer durations, a result that contradicts the findings from the entire participant group. This discrepancy could be attributed to the physiological fact that males typically have stronger cardiorespiratory functions than females. Furthermore, the cognition of older females is more likely to be influenced by hypertension, which could be associated with the rapid decrease in estrogen levels. Studies have shown that the absence of ovarian-produced E2 and P4 can accelerate the development of cardiovascular diseases(50). The resulting decline in vascular function could lead to brain damage and subsequent cognitive decline. Beyond genetic differences, our results indicated that men have a longer history of exercise than women, and most women do not smoke. We also discovered a significant difference in the dietary habits of older females between the two groups, but not in males. These lifestyle differences could also contribute to the observed cognitive differences between the genders.

We developed a predictive model using logistic regression, based on the most significant protective and risk factors. The AUCs for the training and testing sets were 0.683 and 0.682, respectively. These results suggest that the model’s sensitivity and specificity are acceptable. The strong performance of this predictive model further validates the protective and risk factors identified in our study.

In conclusion, the protective factors of cognitive decline for older people were male gender, height, keeping exercising, and nicotine stimulation, and the risk factors included age, female gender, vegetarianism, hypertension, and over-exercise (Fig. 6). Except for the genetic factor, differences in lifestyle, such as smoking and exercise habits, may contribute to the observed differences in cognitive function between genders.

Data availability statement

The datasets used and/or analyzed during the current study available from the corresponding author on reasonable request.

Acknowledgments

We thank all the participants included in this research for their support.

Author contributions

FT: Data curation, Conceptualization, Formal analysis, Writing – original draft, Funding acquisition. YH: Data curation, Writing – review and editing. HD-Y: Data curation, Investigation, Writing – review & editing. LW-S: Formal analysis, Methodology, Software, Visualization. JY-Y: Formal analysis, Methodology, Software, Visualization. CL-S: Formal analysis, Methodology, Software, Visualization. QY-Y: Methodology, Software. Mf-S: Writing – review and editing. CL-Q: Writing – review and editing. GL: Investigation, Methodology, Writing – review and editing. CZ: Funding acquisition, Conceptualization, Data curation, Writing – review and editing. HJ-W: Funding acquisition, Conceptualization, Data curation, Writing – review and editing. All the authors have reviewed the manuscript.

Funding

This work was supported by the grants from National Natural Science Foundation of China (Grant No.82174158, and 82301577), National Key Research and Development Project of China (2018YFC2000200), Project for the Discipline Leader of Shanghai Municipal Health Commission (2022XD051), Shanghai Pudong New Area Leading Health System Personnel Training Program (PWR12023-06), and Shanghai Pudong New Area Health Commission Clinical Characteristic Discipline Construction Plan (PWYts2021-02).

Data availability

The datasets are available from the corresponding author upon reasonable request.

Ethics approval and consent to participate

All procedures involving human participants were in accordance with the ethical standards of the institution or practice at which the study was performed. We followed relevant guidelines to ensure that the study was voluntary and confidential. Because the subjects conducted annual physical examination in Youyi Road Community Health Service Centre for Baoshan District, verbal informed consent was obtained before completing the questionnaire. This study was approved by the Ethics Committee of the Shanghai University of Medicine and Health Sciences and Fudan University.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

McKhann GM, Knopman DS, Chertkow H, Hyman BT, Jack CR, Jr., Kawas CH, et al. The diagnosis of dementia due to Alzheimer's disease: recommendations from the National Institute on Aging-Alzheimer's Association workgroups on diagnostic guidelines for Alzheimer's disease. Alzheimers Dement. 2011;7(3):263-9.
Dominguez LJ, Veronese N, Vernuccio L, Catanese G, Inzerillo F, Salemi G, et al. Nutrition, Physical Activity, and Other Lifestyle Factors in the Prevention of Cognitive Decline and Dementia. Nutrients. 2021;13(11).
Prince M, Ali GC, Guerchet M, Prina AM, Albanese E, Wu YT. Recent global trends in the prevalence and incidence of dementia, and survival with dementia. Alzheimers Res Ther. 2016;8(1):23.
Beard JR, Officer A, de Carvalho IA, Sadana R, Pot AM, Michel JP, et al. The World report on ageing and health: a policy framework for healthy ageing. Lancet. 2016;387(10033):2145-54.
Vaupel JW. Biodemography of human ageing. Nature. 2010;464(7288):536-42.
Prince MJ, Wu F, Guo Y, Gutierrez Robledo LM, O'Donnell M, Sullivan R, et al. The burden of disease in older people and implications for health policy and practice. Lancet. 2015;385(9967):549-62.
Scheltens P, De Strooper B, Kivipelto M, Holstege H, Chételat G, Teunissen CE, et al. Alzheimer's disease. Lancet. 2021;397(10284):1577-90.
Murman DL. The Impact of Age on Cognition. Semin Hear. 2015;36(3):111-21.
Arevalo-Rodriguez I, Smailagic N, Roqué-Figuls M, Ciapponi A, Sanchez-Perez E, Giannakou A, et al. Mini‐Mental State Examination (MMSE) for the early detection of dementia in people with mild cognitive impairment (MCI). Cochrane Database of Systematic Reviews. 2021(7).
Shim YS, Yang DW, Kim H-J, Park YH, Kim S. Characteristic differences in the mini-mental state examination used in Asian countries. BMC neurology. 2017;17:1-9.
Crum RM, Anthony JC, Bassett SS, Folstein MF. Population-based norms for the Mini-Mental State Examination by age and educational level. Jama. 1993;269(18):2386-91.
Yaffe K, Laffan AM, Harrison SL, Redline S, Spira AP, Ensrud KE, et al. Sleep-disordered breathing, hypoxia, and risk of mild cognitive impairment and dementia in older women. Jama. 2011;306(6):613-9.
Iwashyna TJ, Ely EW, Smith DM, Langa KM. Long-term cognitive impairment and functional disability among survivors of severe sepsis. Jama. 2010;304(16):1787-94.
Dominguez LJ, Veronese N, Vernuccio L, Catanese G, Inzerillo F, Salemi G, et al. Nutrition, physical activity, and other lifestyle factors in the prevention of cognitive decline and dementia. Nutrients. 2021;13(11):4080.
Kivipelto M, Mangialasche F, Ngandu T. Lifestyle interventions to prevent cognitive impairment, dementia and Alzheimer disease. Nature Reviews Neurology. 2018;14(11):653-66.
Isaev NK, Stelmashook EV, Genrikhs EE. Neurogenesis and brain aging. Reviews in the Neurosciences. 2019;30(6):573-80.
Isaev N, Stelmashook E, Stelmashook N, Sharonova I, Skrebitsky V. Brain aging and mitochondria-targeted plastoquinone antioxidants of SkQ-type. Biochemistry (Moscow). 2013;78:295-300.
Norton S, Matthews FE, Barnes DE, Yaffe K, Brayne C. Potential for primary prevention of Alzheimer's disease: an analysis of population-based data. The Lancet Neurology. 2014;13(8):788-94.
Grant WB. Using multicountry ecological and observational studies to determine dietary risk factors for Alzheimer's disease. Journal of the American College of Nutrition. 2016;35(5):476-89.
Yuan C, Fondell E, Bhushan A, Ascherio A, Okereke OI, Grodstein F, et al. Long-term intake of vegetables and fruits and subjective cognitive function in US men. Neurology. 2019;92(1):e63-e75.
Giem P, Beeson WL, Fraser GE. The incidence of dementia and intake of animal products: preliminary findings from the Adventist Health Study. Neuroepidemiology. 1993;12(1):28-36.
Jacobs Jr DR, Orlich MJ. Diet pattern and longevity: do simple rules suffice? A commentary. The American journal of clinical nutrition. 2014;100(suppl_1):313S-9S.
Livingston G, Sommerlad A, Orgeta V, Costafreda SG, Huntley J, Ames D, et al. Dementia prevention, intervention, and care. The lancet. 2017;390(10113):2673-734.
Liang J-h, Lu L, Li J-y, Qu X-y, Li J, Qian S, et al. Contributions of modifiable risk factors to dementia incidence: a Bayesian network analysis. Journal of the American Medical Directors Association. 2020;21(11):1592-9. e13.
Vancampfort D, Solmi M, Firth J, Vandenbulcke M, Stubbs B. The impact of pharmacologic and nonpharmacologic interventions to improve physical health outcomes in people with dementia: a meta-review of meta-analyses of randomized controlled trials. Journal of the American Medical Directors Association. 2020;21(10):1410-4. e2.
Sofi F, Valecchi D, Bacci D, Abbate R, Gensini GF, Casini A, et al. Physical activity and risk of cognitive decline: a meta‐analysis of prospective studies. Journal of internal medicine. 2011;269(1):107-17.
Nuzum H, Stickel A, Corona M, Zeller M, Melrose RJ, Wilkins SS. Potential benefits of physical activity in MCI and dementia. Behavioural neurology. 2020;2020.
Santisteban MM, Iadecola C, Carnevale D. Hypertension, neurovascular dysfunction, and cognitive impairment. Hypertension. 2023;80(1):22-34.
Azarpazhooh MR, Avan A, Cipriano LE, Munoz DG, Sposato LA, Hachinski V. Concomitant vascular and neurodegenerative pathologies double the risk of dementia. Alzheimer's & Dementia. 2018;14(2):148-56.
Santisteban MM, Iadecola C. Hypertension, dietary salt and cognitive impairment. Journal of Cerebral Blood Flow & Metabolism. 2018;38(12):2112-28.
McGrath ER, Beiser AS, DeCarli C, Plourde KL, Vasan RS, Greenberg SM, et al. Blood pressure from mid‐to late life and risk of incident dementia. Neurology. 2017;89(24):2447-54.
Littlejohns TJ, Collister JA, Liu X, Clifton L, Tapela NM, Hunter DJ. Hypertension, a dementia polygenic risk score, APOE genotype, and incident dementia. Alzheimer's & Dementia. 2023;19(2):467-76.
Waisman Campos M, Serebrisky D, Mauricio Castaldelli-Maia J. Smoking and cognition. Current drug abuse reviews. 2016;9(2):76-9.
Clader JW, Wang Y. Muscarinic receptor agonists and antagonists in the treatment of Alzheimer's disease. Current pharmaceutical design. 2005;11(26):3353-61.
Dani JA, Bertrand D. Nicotinic acetylcholine receptors and nicotinic cholinergic mechanisms of the central nervous system. Annu Rev Pharmacol Toxicol. 2007;47:699-729.
Wallace T, Bertrand D. Importance of the nicotinic acetylcholine receptor system in the prefrontal cortex. Biochemical pharmacology. 2013;85(12):1713-20.
Kutlu MG, Gould TJ. Nicotinic receptors, memory, and hippocampus. The neurobiology and genetics of nicotine and tobacco. 2015:137-63.
Couey JJ, Meredith RM, Spijker S, Poorthuis RB, Smit AB, Brussaard AB, et al. Distributed network actions by nicotine increase the threshold for spike-timing-dependent plasticity in prefrontal cortex. Neuron. 2007;54(1):73-87.
Amin V, Fletcher JM. What is driving the relationship between height and cognition? Evidence from the Twins Early Development Study. Economics & Human Biology. 2022;47:101174.
Case A, Paxson C. Stature and status: Height, ability, and labor market outcomes. Journal of political Economy. 2008;116(3):499-532.
Guven C, Lee WS. Height, aging and cognitive abilities across Europe. Econ Hum Biol. 2015;16:16-29.
Guo J, Song S. Associations of Height Loss With Cognitive Decline and Incident Dementia in Adults Aged 50 Years and Older. J Gerontol A Biol Sci Med Sci. 2023;78(8):1445-52.
Li R, Singh M. Sex differences in cognitive impairment and Alzheimer’s disease. Frontiers in neuroendocrinology. 2014;35(3):385-403.
Bramen JE, Hranilovich JA, Dahl RE, Forbes EE, Chen J, Toga AW, et al. Puberty influences medial temporal lobe and cortical gray matter maturation differently in boys than girls matched for sexual maturity. Cerebral Cortex. 2011;21(3):636-46.
Koolschijn PCM, Crone EA. Sex differences and structural brain maturation from childhood to early adulthood. Developmental cognitive neuroscience. 2013;5:106-18.
Neufang S, Specht K, Hausmann M, Güntürkün O, Herpertz-Dahlmann B, Fink GR, et al. Sex differences and the impact of steroid hormones on the developing human brain. Cerebral cortex. 2009;19(2):464-73.
Giedd JN, Snell JW, Lange N, Rajapakse JC, Casey B, Kozuch PL, et al. Quantitative magnetic resonance imaging of human brain development: ages 4–18. Cerebral cortex. 1996;6(4):551-9.
Clark AS, Maclusky NJ, Goldman-Rakic PS. Androgen binding and metabolism in the cerebral cortex of the developing rhesus monkey. Endocrinology. 1988;123(2):932-40.
Morse JK, Scheff SW, DeKosky ST. Gonadal steroids influence axon sprouting in the hippocampal dentate gyrus: a sexually dimorphic response. Experimental neurology. 1986;94(3):649-58.
Gersh F, O'Keefe JH, Elagizi A, Lavie CJ, Laukkanen JA. Estrogen and cardiovascular disease. Prog Cardiovasc Dis. 2024.

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

Protective and Risk Factors in Daily Life Associated with Cognitive Decline of Older Adults

Status:

Version 1

Abstract

Figures

Introduction

Methods

Statistical Analysis

Results

Discussion

Conclusion

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1