Association of Vegetable and Fruit Consumption Patterns with Cognitive Function in Older People with Different BMI Ranges: Findings from China

Introduction

China is facing population aging.¹ Under this challenging situation, cognitive impairment in older people has become a public health concern that needs to be addressed.^2,3 It has been reported that Alzheimer’s disease (AD) affects over 55 million people worldwide, and this figure is anticipated to rise to a new peak by 2050.⁴ Despite great efforts, there are no clinically effective means and drugs to treat Alzheimer’s disease.⁵ Previous studies have shown that mild cognitive impairment (MCI) is an early stage in a patient’s development of AD.^6,7 Therefore, early prevention for MCI has become a new perspective and approach to addressing AD effectively.⁸

Recently, the impact of vegetable and fruit consumption on health in later life has been documented.^9–11 For example, a meta-analysis showed that vegetable and fruit consumption related to a reduced risk of cardiovascular disease and cancer mortality.¹² Besides, a prospective Australian study of older women showed that increased intake of vegetables, particularly cruciferous and allium vegetables, may prevent fractures in older women.¹³ Current research has shown that the cognitive function of older adults is influenced by many factors such as lifestyle habits,¹⁴ illnesses (depression,¹⁵ diabetes,¹⁶ etc.), religious beliefs,¹⁷ and socioeconomic status.¹⁸ Moreover, vegetable and fruit consumption has emerged as a crucial concern for preventing and delaying cognitive impairment due to the potential impact the minerals, fibre, and vitamins in vegetables and fruits may have on brain health.^19,20 As reported in a previous study, vegetable and fruit consumption has a protective role in slowing cognitive decline.²¹ Another study also showed that sufficient consumption of vegetables and fruits was associated with a reduced risk of dementia.²²

Currently, relevant studies have confirmed that older people with different weight statuses as assessed by body mass index (BMI) exhibit different cognitive function statuses.²³ Older people with a BMI defined as underweight were significantly more likely to experience a risk of cognitive impairment.²⁴ Another study suggested that obesity reduces the risk of dementia compared to people with a normal weight.²⁵ However, evidence from Poland showed that overweight and obesity in older people were not associated with cognitive decline.²⁶ Individuals categorized according to BMI may exhibit different cognitive capabilities,²⁷ and these differences could alter the impact of eating vegetables and fruits on cognitive function. For example, obesity was linked to a higher risk of cognitive decline, which may be due to the effects of obesity-related inflammation and metabolic syndrome on brain health.²⁸ The link between vegetable and fruit consumption and cognitive function appears multifaceted and may be influenced by various factors, including BMI. Besides, gender differences further complicate this relationship. Men and women typically exhibit different dietary patterns, nutritional needs, and physiological responses to the diet, which may impact cognitive function differently.^29,30 For instance, studies suggested that women have a higher intake of vegetables and fruits than men, potentially correlating with enhanced cognitive function.³¹

Nevertheless, evidence regarding the impact of vegetable and fruit consumption on cognitive function in older people with different BMIs, as well as gender differences, is insufficient. Therefore, this study aims to elucidate the relationship among older adults with different BMIs while focusing on gender differences. By elucidating these relationships, policymakers identify targeted dietary interventions to support cognitive function and improve older people’s overall health and quality of life.

Materials and Methods

Study Design and Sample

This cross-sectional survey is part of the Anhui Health and Longevity Survey (AHLS) conducted in China from July to September 2019.^7,32 To obtain a well-representative sample, we designed a multi-stage purposive sampling method. The sampling process of this study can be described as follows. First, four cities from Anhui Province were selected based on geographic location: Chuzhou (East), Xuancheng (South), Lu’an (West), and Fuyang (North). Second, 3 to 5 urban and rural communities were randomly selected from these four cities. Third, residents were invited to participate in the questionnaire survey in each selected urban and rural community. The following criteria were used to select the research participants: Residents aged 60 years and older, with no severe hearing or speech impairments, who were able to cooperate with the survey. The investigator conducted face-to-face interviews on-site. Prior to the interview, the investigator explained the purpose and procedures of the study to the participants, and informed consent was obtained. A total of 6211 questionnaires were collected in this survey. This study excluded participants who were unable to complete the assessment of cognitive function (N=32), had missing BMI-related data (N=381), and a final 5798 participants were eligible for data analysis (details are shown in Supplementary Figure S1). Ethical approval for this study was obtained from the Biomedical Ethics Committee of Anhui Medical University (No. 2020H011).

Cognitive Function

This study assessed the participants’ cognitive function through the Chinese version of the Mini-Mental State Examination (MMSE), a validated and commonly used method for detecting cognitive impairment in clinical and research settings.^33,34 The Chinese version of MMSE consists of orientation, memory, attention and calculation, recall, and language abilities, scoring 30 points. Several questions in this scale are influenced by the participant’s educational background,³⁵ so the criteria for assessing MCI in participants of this study are as follows: illiterate individuals scoring below 18 points; those with primary education but scoring below 21 points; those with more than six years of education but scoring below 25 points, in line with other studies.⁷

Vegetable and Fruit Consumption Patterns

To obtain the frequency of vegetable and fruit consumption of the respondents, the question “How many days a week do you usually consume vegetables/fruits?” was set in the questionnaire. The available answer choices include “every day”, “4–6 days”, “2–3 days”, and “less than 1 day”. Consistent with other studies,^32,36 respondents’ vegetable and fruit consumption was categorized into the following four patterns: V+/F+: daily or often intake of both vegetables and fruits; V+/F-: daily or often intake of vegetables, but sometimes or never intake of fruits; V-/F+: daily or often intake fruits, but sometimes or never intake vegetables; V-/F-: Sometimes or never intake of either vegetables or fruits.

BMI

The investigator determined BMI (kg/m²) by taking on-site measurements of height and weight for each participant and dividing the measured weight by the square of the height. According to the BMI standards in China,^37,38 respondents were classified as follows: underweight: BMI <18.5, normal weight: 18.5 ≤ BMI < 24, overweight: 24 ≤ BMI < 28, obese: BMI ≥ 28.

Other Variables

The following information about the respondents was collected for this study: age (years), gender, place of residence (urban, rural), living status (living alone, living with others), marital status (married, other), education level (illiterate, primary school, above primary school), annual income (<6500 CNY, ≥6500 CNY), smoking status (smoker, non-smoker), and drinking status (drinker, non-drinker). Depressive symptoms in participants were evaluated with the Patient Health Questionnaire (PHQ-9), covering nine questions out of 27 points, classifying participants with scores of five and above as depressed, otherwise not depressed.³² The question “Do you suffer from one of the following chronic diseases (hypertension, hyperlipidemia, diabetes, chronic hepatitis, cancer, heart disease, stroke, lung disease, psychiatric disease, etc.)” was used to determine if the respondent suffered from a chronic disease. In the current study, we used the activities of daily living (ADL) scale, which includes ten dimensions, to measure disability in ADL. Respondents were classed as “No” if they scored 100, ie, indicated “self-care” on any of the items. Otherwise, they were categorized as “Yes”, consistent with another study.³⁹

Statistical Analysis

Data for categorical variables were described in the form of frequencies and percentages. First, we used simple descriptive analysis to describe the characteristics of the sample. Second, a binary logistic regression model was performed to explore gender differences in the correlation between vegetable and fruit consumption patterns and cognitive function. Univariate logistic regression analyses were conducted to produce Model 1, which was then adjusted for the covariates of age, place of residence, living status, marital status, education, annual income, smoking status, drinking status, chronic disease, depression status, disability in ADL, and vegetable and fruit consumption patterns to produce Model 2. The regression models yielded odds ratios (OR) and adjusted odds ratios (AOR) and the corresponding 95% confidence intervals (95% CI). This study used SPSS 23.0 software for data analysis, with statistical significance set at P<0.05.

Results

Basic Characteristics of the Sample

Participants were categorized by gender, resulting in Table 1. A total of 5798 participants, of whom more were women (3147, 54.3%), most of the participants were between 60–69 years old (2740, 47.3%), had a BMI in the normal weight range (2463, 42.5%), lived in urban areas (2864, 49.4%). Most were married (4159, 71.7%), lived with others (4733, 81.6%), and almost half were illiterate (2849, 49.2%). The majority had an annual income of less than 6,500 CNY (3469, 59.8%). Notably, 79.0% and 61.3% were non-smokers and non-drinkers, respectively. Regarding health status, the vast majority of participants suffered from chronic diseases (4139, 71.4%), a few were depressed (1887, 32.5%), and more than half of the participants had disability in ADL (3384, 58.4%). In addition, the majority of participants’ vegetable and fruit consumption patterns were classified as V+/F- (3910, 67.4%).

Table 1 Characteristics of Participants by Gender (N=5798)

Binary Logistic Regression Analysis of Vegetable and Fruit Consumption Patterns and Cognitive Function in Underweight Participants

Using binary logistic regression analysis (Table 2), we found no statistically significant relationship between vegetable and fruit consumption patterns and cognitive function among underweight men, either before or after adjustment (P>0.05). In contrast, unadjusted results among underweight women showed that the V-/F- pattern correlated with lower cognitive function (OR=11.31; 95% CI: 1.25–102.72), but no significant correlation was found after adjustment.

Table 2 Binary Logistic Analysis Examining Relationships Between Vegetable and Fruit Consumption Patterns and Cognitive Function in Underweight Participants (N=274)

Binary Logistic Regression Analysis of Vegetable and Fruit Consumption Patterns and Cognitive Function in Normal Weight Participants

Table 3 illustrates the findings of the binary logistic regression model following variable adjustments. In normal weight men, cognitive impairment is more likely to be found in the V+/F- (AOR=1.65; 95% CI: 1.16–2.35) and V-/F- (AOR=3.95; 95% CI: 1.86–8.42) groups, respectively, compared to the V+/F+ group. However, in women of normal weight, vegetable and fruit consumption patterns were not statistically correlated with cognitive function after adjustment.

Table 3 Binary Logistic Analysis Examining Relationships Between Vegetable and Fruit Consumption Patterns and Cognitive Function in Normal Weight Participants (N=2463)

Binary Logistic Regression Analysis of Vegetable and Fruit Consumption Patterns and Cognitive Function in Overweight Participants

As presented in Table 4, after adjusting for confounders, only V+/F- (AOR=1.54; 95% CI: 1.12–2.11) was related to an increased probability of cognitive impairment among overweight women (Model 2). Furthermore, statistical significance did not show other vegetable and fruit consumption patterns in the overweight participants (P>0.05).

Table 4 Binary Logistic Analysis Examining Relationships Between Vegetable and Fruit Consumption Patterns and Cognitive Function in Overweight Participants (N=2072)

Binary Logistic Regression Analysis of Vegetable and Fruit Consumption Patterns and Cognitive Function in Obese Participants

Table 5 shows the results of the analyses of the obese participants. The adjusted results (Model 2) reveal that vegetable and fruit consumption patterns were not significantly correlated with cognitive function in either males or females (P>0.05).

Table 5 Binary Logistic Analysis Examining Relationships Between Vegetable and Fruit Consumption Patterns and Cognitive Function in Obese Participants (N=989)

Discussion

As far as we know, this study is the first to explore the link between vegetable and fruit consumption patterns and cognitive function in older adults with varying BMIs and to examine gender differences in this association. According to the data analysis, the association between the two varied among older people with different BMIs.

This study found that 31.4% of older individuals had MCI, surpassing the typical prevalence range of 15.2–15.9% for MCI among older people in China.⁴⁰ The present study showed no correlation between vegetable and fruit consumption patterns and cognitive function in underweight older people. One possible explanation is that older people suffering from being underweight may face overall malnutrition.⁴¹ They may be deficient in multiple nutrients, not just those found in vegetables and fruits.⁴² This deeper nutritional deficiency may have a greater impact on the brain and weaken this relationship.⁴³

Conversely, among normal weight men, older people with a higher frequency of vegetable and fruit consumption reported better cognitive function, consistent with other studies.^21,36,44 In other words, individuals in the V+/F- and V-/F- groups were 1.65 and 3.95 times more likely to have poorer cognitive function than those in the V+/F+ group, respectively. Previous studies also supported this conclusion.^21,45,46 A similar association between vegetable and fruit intake and cognitive function was observed in a Brazilian cohort of older adults,⁴⁷ where higher vegetable and fruit intake was associated with better cognitive function. Still, the effect size was not the same in Western populations.^48,49 This geographic heterogeneity may reflect differences in diet composition, food preparation methods, or lifestyle confounders across cultures. Another prospective cohort study also showed that higher vegetable and fruit consumption was positively associated with cognitive function in older people.⁵⁰ The reasons why vegetable and fruit intake can prevent cognitive decline may include the following. Vegetables and fruits are rich in various vitamins and minerals,^51,52 which are essential for brain health. For instance, folic acid can prevent hyperhomocysteinemia, reducing the risk of cognitive impairment.⁵³ In addition, many vegetables and fruits contain plentiful antioxidants (flavonoids, anthocyanins, carotenoids, etc.) that help reduce oxidative stress and protect brain cells from damage.^54,55 Moreover, vegetables and fruits often have anti-inflammatory properties that reduce chronic inflammation in the body, considered to be associated with AD and other cognitive impairments.⁵⁶

Interestingly, this association was not observed in older women. There are physiological and hormonal differences between older men and women. Especially after menopause, older women’s estrogen levels drop sharply, which may affect their eating habits and the utilization of nutrients from vegetables and fruits, thereby impacting their cognitive function in later life.^57,58 For overweight older people, our study found gender differences in the association between vegetable and fruit consumption patterns and cognitive function. Specifically, among overweight women, older adults with low vegetable and fruit consumption (V+/F-) have a higher likelihood of developing MCI, but not in men. This discrepancy may be attributed to differences in overweight men’s and women’s diets. For example, overweight women may be more inclined to improve the overall quality of their diet by increasing their vegetable and fruit intake, especially if they need to control their weight.⁵⁹ This self-regulation of diet may help them improve their overall health and cognitive function.⁶⁰ Overweight men, on the other hand, may rely more on high-protein, high-fat foods, and increased vegetable and fruit intake may not significantly improve cognitive function.⁶¹

Our study found no association between vegetable and fruit consumption patterns and cognitive function in obese older people. This may be attributed to the poor cognitive function of obese older people. Current research suggests that obesity negatively correlates with cognitive function.⁶² Besides, obese older people commonly have chronic diseases such as diabetes, hypertension, and hyperlipidemia,⁶³ which may harm their cognitive function,⁶⁴ so vegetable and fruit intake, while it may improve diet quality and nutrient intake, may not counteract these adverse effects.

This study has several advantages. First, this study differs from previous research by focusing on different BMIs of older people and examining the gender differences in the association between vegetable and fruit consumption patterns and cognitive function. Second, this study employed effective and reliable tools to measure participants’ cognitive function and obtained a large representative sample, reducing the impact of random factors on the results.

However, it is crucial to be aware of some research limitations. The present study was cross-sectional and could not determine the exact causal relationship, suggesting the need to explore this more fully in future cohort or intervention studies. Besides, a self-reported questionnaire on the frequency of vegetable and fruit consumption was used in this study, leading to possible misclassification of vegetables and fruits. Furthermore, China-specific evidence dominated our analysis. For example, the traditional consumption attitudes of older Chinese adults (who may be reluctant to spend money on fruits) resulted in a small sample size of vegetable and fruit consumption patterns defined as the V-/F+ group, which may limit the generalizability of the findings. The complex interplay between regional dietary habits, genetic susceptibility, and environmental factors needs to be validated by comparative studies of different ethnic populations. Finally, the variables considered in this study were limited, and some factors, such as physical activity patterns, mental health status, and medication taken by participants, were not adequately emphasized. Therefore, a longitudinal design with multidimensional assessments will be utilized in subsequent investigations to help elucidate nutritional factors’ independent and synergistic effects on cognitive function.

Conclusion

This study shows how consuming vegetables and fruits is linked to cognitive function in older Chinese people with different BMIs. No correlation was found between vegetable and fruit consumption patterns and cognitive function in underweight and obese older adults. However, among normal weight and overweight older adults, those with a higher frequency of vegetable and fruit consumption reported better cognitive function, with gender differences. This study emphasizes the significance of vegetable and fruit consumption and contributes to formulating more effective policy approaches to protect cognitive function in older adults with different BMIs, aiming for healthy aging.