1. Introduction
China has the largest number of people with dementia, accounting for nearly 25% of the world’s dementia cases and increasing at a rate of over 360,000 per year [
1]. Identifying, and subsequently managing risk, is critical to slowing the progression of Alzheimer’s disease (AD)-related cognitive impairment. One modifiable factor in dementia is diet, and healthy eating habits are associated with a reduced risk of dementia [
2]. Dietary intervention is essential to prevent age-related cognitive decline [
3].
Adequate water intake is required to maintain the hydrated state, which benefits human health [
4]. Dehydration or other hydration imbalances can lead to the dysfunction of physical or cognitive activities [
4]. A German study suggested that adequate water intake improves short-term memory [
5]. Despite the body’s water balance mechanism, even minor systemic water loss can impair cognitive performance [
6]. However, a systematic review concluded that underhydration might not significantly impair cognitive function [
7].
The relationship between tea, and coffee consumption and cognitive performance has been inconsistent in the results of previous clinical trials. A study found that green tea intake may reduce the risk of dementia and improve cognitive performance [
8]. However, another study found a cognitive protective effect of black tea intake in older Chinese adults but did not find a correlation between green tea and cognitive function [
9]. A 21-year longitudinal study reported that coffee consumption in middle age was associated with a reduced risk of AD/dementia in later life, and no association was found between tea consumption and AD/dementia [
10]; the possible reason for this result is that coffee drinking is more common in the Finnish population and tea drinking is relatively uncommon. Several studies have also denied a significant cognitive effect of habitual coffee [
10,
11,
12] or tea [
12] consumption.
The results of two retrospective meta-analyses suggested that higher milk intake may be associated with better cognitive performance [
13,
14], but this may be limited to Asian populations [
14]. The current evidence on the relationship between milk intake and cognitive performance is inconsistent, and no firm conclusions can be drawn about their relationship [
15,
16].
The relationship between water, tea, coffee, and pure milk consumption and cognitive impairment remains to be determined. Previous studies have found that the effects of beverage consumption on chronic diseases such as hypertension and Parkinson’s differ across gender groups [
17,
18], and there may also be differences in beverage preferences across gender groups [
19]. At present, there are few studies on the relationship between beverages consumption and cognitive impairment in different gender groups.
Amyloid beta (Aβ) deposition increases the risk of cognitive decline [
20]. Compared with cerebrospinal fluid biomarkers, blood-based biomarkers are safe, minimally invasive, less resource-intensive, and widely available. Compared to normal cognitive individuals, individuals with cognitive impairment had lower plasma Aβ42/40 levels [
21], higher plasma neurofilament light chain (NfL) levels [
22], and lower plasma tau phosphorylated at threonine 181 (p-Tau-181) [
23]. In addition, it was found that the increase of plasma p-tau-181 was related to extensive cortical Aβ lesions, helping to identify people at high risk of AD [
23,
24]. Plasma biomarkers can not only assist in diagnosing AD but can also be used clinically to monitor the effects of lifestyle factors such as diet on cognition in people at greater risk of cognitive impairment through regular plasma biomarker testing [
25]. However, few studies have paid attention to the relationship between beverage consumption and plasma biomarkers in Aβ-PET-positive populations.
The purpose of this study was to assess the relationship between water, tea, coffee, and pure milk consumption and cognitive status in middle-aged and elderly populations of different genders. We also explored the relationship between different consumptions of water, tea, pure milk, and coffee and plasma biomarkers in participants with Aβ-PET-positive(+) and Aβ-PET-negative(−) genotypes.
3. Results
A comparison of participants’ demographic data, basic physical condition, drug history, smoking habits, and drinking habits and the analysis of the results can be found in the previous article [
26].
Table 1 shows the results of the univariate analysis. The SCD, AD, and objective cognitive unimpaired groups had more APOEε4 allele carriers compared with the NC, MCI, and objective cognitive impairment groups. Compared to the NC group, the SCD group consumed coffee infrequently. Compared with the MCI group, the AD group had less daily water consumption, consumed green tea or pure milk infrequently, and had less weekly milk intake. The objective cognitive impairment group had less daily water consumption and consumed less coffee, green tea, or yogurt compared to the objective cognitive unimpaired group. The “daily milk intake” and “frequency of coffee consumption” items were not found to be associated with the severity of cognitive impairment.
Table 2 shows the results of the multivariate analysis. Compared with those with infrequent coffee consumption, females in NC group had a lower risk of SCD from regular coffee consumption (OR = 0.500, 95%CI, 0.305–0.820,
p = 0.006). This relationship remained statistically significant after adjusting for education, age, diarrhea, and pro-cognitive drugs use (OR = 0.479, 95%CI, 0.287–0.801,
p = 0.005). After further adjustment for daily water consumption, long-term drinking water, pure milk consumption, tea consumption, tea drinking frequency, yogurt consumption, and Apolipoprotein E genotype in Model c, the results remained unchanged (OR = 0.459, 95%CI, 0.248–0.851,
p = 0.013).
Table 3 shows the results of multivariate analysis. Compared with those drinking >1500 mL per day, individuals with MCI who drank 1000–1500 mL, 500–1000 mL, and <500 mL per day had a higher risk of AD (OR = 1.998, 95%CI, 1.119–3.568,
p = 0.019; OR = 2.283, 95%CI, 1.317–3.959,
p = 0.003; OR = 4.791, 95%CI, 2.295–10.005,
p = 1.567, respectively). This relationship remained statistically significant after adjusting for education, age, dieting and weight loss, diarrhea, allergic history, and pro-cognitive drugs use in Model b (OR = 2.083, 95%CI, 1.076–4.032,
p = 0.029; OR = 2.015, 95%CI, 1.071–3.792,
p = 0.030; OR = 3.911, 95%CI, 1.721–8.887,
p = 0.001, respectively). After further adjustment for long-term drinking water, tea consumption, tea frequency, pure milk consumption, yogurt consumption, and APOE genotype in Model c, individuals with MCI who drank <500 mL per day had a higher risk of developing AD (OR = 3.387, 95%CI, 1.447–7.927,
p = 0.005).
Compared to those with infrequent tea consumption, females with MCI had a lower risk of AD from regular tea consumption (OR = 0.322, 95%CI, 0.177–0.585, p < 0.001). This relationship remained statistically significant in Model b (OR = 0.379, 95%CI, 0.197–0.729, p = 0.004). After further adjustment of daily water consumption, long-term drinking water, coffee consumption, pure milk consumption, yogurt consumption, and APOE genotype in Model d, the results remained unchanged (OR = 0.430, 95%CI, 0.195–0.948, p = 0.037).
Compared to those with infrequent pure milk consumption, females with MCI had a lower risk of AD from regular pure milk consumption (OR = 0.392, 95%CI, 0.339–0.644, p = 0.000). This relationship remained statistically significant in Model b (OR = 0.410, 95%CI, 0.234–0.718, p = 0.002). After further adjustment of daily water consumption, long-term drinking water, tea consumption, tea frequency, coffee consumption, yogurt consumption, and APOE genotype in Model e, the results remained unchanged (OR = 0.441, 95%CI, 0.231–0.843, p = 0.013).
Table 4 shows the results of multivariate analysis. Compared to those with infrequent tea consumption, males with objective cognitive unimpaired had a lower risk of objective cognitive impairment from regular tea consumption (OR = 0.314, 95%CI, 0.185–0.533,
p < 0.001). This relationship remained statistically significant after adjusting for education, age, marriage, allergic history, and pro-cognitive drugs use in Model b (OR = 0.294, 95%CI, 0.161–0.537,
p < 0.001). After further adjustment of daily water consumption, coffee consumption, pure milk consumption, yogurt consumption, and APOE genotype in Model c, the results remained unchanged (OR = 0.344, 95%CI, 0.182–0.647,
p = 0.016).
Compared to those who basically did not drink tea per day, males with objective cognitive unimpaired with tea frequency ≥3 cups/day, 1–2 cups/day, and <1 cup/day had a lower risk of objective cognitive impairment (OR = 0.367, 95%CI, 0.201–0.671, p = 0.001; OR = 0.419, 95%CI, 0.216–0.812, p = 0.010; OR = 0.331, 95%CI, 0.141–0.770, p = 0.010, respectively). This relationship remained statistically significant in Model b (OR = 0.296, 95%CI, 0.146–0.692, p = 0.001; OR = 0.374, 95%CI, 0.176–0.792, p = 0.010; OR = 0.245, 95%CI, 0.090–0.669, p = 0.006, respectively). After further adjustment in Model c, the results remained unchanged (OR = 0.336, 95%CI, 0.174–0.772, p = 0.008; OR = 0.442, 95%CI, 0.201–0.974, p = 0.043; OR = 0.261, 95%CI, 0.093–0.733, p = 0.011, respectively).
Table 5 shows that in the PET (+) group, pure milk or green tea consumption (compared to none) was statistically associated with lower p-Tau-181 levels (
p = 0.042,
p = 0.014, respectively).
Table 6 shows that coffee or daily water consumption were statistically associated with AD-related biomarkers.
4. Discussion
This study examined the intrinsic relationship between beverage consumption and cognitive abilities. The results suggest that the relationship between water, tea, coffee, and pure milk consumption and cognitive performance in middle-aged and elderly adults is related to basic cognitive status, gender, and Aβ deposition.
Water has many roles in the human body. Water in the body is the solvent that carries nutrients, the reactants and products of metabolic processes, and the main component of cells and tissues [
30]. Therefore, optimal hydration is essential to maintain a range of normal physiological functions required for the health of the body [
31]. A Chinese randomized controlled trial of youth water drinking reported that water restriction for 36 h impaired situational memory and mood [
32]. Compared to the young population, the older population has a greater risk of dehydration due to a more inadequate water balance regulation mechanism [
6]. A study found that human hydration status and water intake follow a U-shaped curve with cognition and that dehydration or overhydration can impair cognition in older adults [
33]. Available data suggest that good hydration is associated with better results on cognitive tests and that mild dehydration impairs cognitive performance [
34]. Our study found that daily water consumption <1500 mL (especially <500 mL) was a risk factor for middle-aged and elderly people whose cognitive abilities have been impaired. Our study provided further support for the cognitive impairment of insufficient water consumption in the middle-aged and elderly, especially those with pre-existing cognitive impairment. In contrast to our results, a study [
35] found no significant relationship between hydration status and cognitive function. A possible reason for this result is that the study was conducted in normal cognitive older adults and defined inadequate water consumption in older adults as less than the appropriate intake (AI) (2000 mL for females and 2500 mL for males).
Our study found that for objective cognitive unimpaired males and females with MCI, green tea consumption was a protective factor for cognitive impairment. The mechanism was that green tea consumption improved AD pathology in participants with Aβ deposition. A Japanese meta-analysis found that green tea consumption reduced all-cause mortality in middle-aged and older adults, but there were differences between males and females [
36]. A longitudinal study reported a 28% or greater reduction in the risk of cognitive decline among older adults with higher green tea consumption [
37]. Another longitudinal study found that tea intake slowed the rate of cognitive decline in females, but it did not find an effect of tea consumption on cognitive performance in males [
38]. It is hypothesized that there is variability in the effects of green tea intake on the cognitive profiles of gender-specific groups. This variability may be due to the gender-dependent effects of epigallocatechin gallate (EGCG) on the metabolic regulation of the organism [
39]. Tea drinking originated in Asia and is particularly revered by the Chinese. Tea is usually made from the leaves of the tea tree, an economic plant, and is one of the most popular beverages due to its satisfying sensory experience and many health-promoting properties [
40]. The main polyphenol present in green tea is EGCG, and previous studies have found that the improvement in cognitive performance associated with green tea intake may be the result of the combined effects of EGCG, caffeine, L-theanine, and flavonoids [
8,
41,
42,
43]. These active ingredients prevent and modulate AD pathology through multiple mechanisms, including the reduction of Aβ production, aggregation and inhibition of tau aggregation, and hyperphosphorylation [
42,
43]. An experimental animal study revealed that EGCG and flavonoids reduced Aβ-induced mitochondrial dysfunction in AD transgenic mice [
44]. A systematic review reported that caffeine and L-theanine, alone or in combination, affected various neurotransmitter systems, improving attention, cognition, and mood [
45].
Green tea has a higher concentration of polyphenols and a higher antioxidant potential than black tea [
46]. Black tea is made by promoting the enzymatic oxidation of fresh leaves. Most catechins are converted to oxidized forms called theaflavins and theobromine [
43]. The total catechin levels are reduced from 35−50% in green tea to 10% in black tea [
43]. Meta-analyses results have suggested that the cognitive benefits of tea (green, black) may increase with increasing daily consumption [
47,
48,
49]. Another study [
9] reported a cognitive protective effect of black tea consumption. Our study did not find a correlation between black tea consumption and cognitive impairment in middle-aged and older adults. We speculate that this result may be due to the small sample of study subjects who filled in the black tea consumption or the limited extent of the effect of black tea on cognitive impairment.
Caffeine is the most widely consumed mental stimulant in the world. Our study found that for middle-aged and older females with NC, coffee consumption was a protective factor for cognitive impairment. Previous studies have found that coffee consumption improves cognitive health [
47,
50,
51]. The 2011–2014 National Health and Nutrition Survey conducted by XueDong et al. reported that coffee and caffeine were associated with better cognitive performance [
50]. MRI parameter analysis found that moderate coffee consumption (two cups per day) was associated with better white matter preservation and cerebral blood flow, especially in the elderly with normal cognitive function [
52]. Jee Wook Kim et al. [
53] found that coffee intake can reduce the risk of AD or related cognitive decline in older subjects with normal cognition. Marilyn et al. [
54] found that women with higher levels of coffee intake scored higher on cognitive tests [
54]. A clinical study found that caffeine had a more significant neuroprotective effect on women with normal cognition [
55]. That is, the neuroprotective effect of coffee may be related to gender [
54,
55] and cognitive status [
52,
53,
55]. Our study did not find that coffee reduced AD pathological changes in participants with Aβ deposition. However, previous studies have found that the cognitive protective effect of coffee may be associated with reduced Aβ deposition in the brain [
53]. Various types of coffee, including roasted coffee and instant coffee, can reduce Aβ levels in the body, and the mechanism may be to reduce the production of Aβ [
56,
57]. Notably, the effect of coffee in inhibiting Aβ and tau aggregation and providing neuroprotection is not only dependent on caffeine but is also associated with Phenylindanes [
58].
Our study found that for middle-aged and older females with MCI, pure milk consumption was a protective factor for cognitive impairment. For participants with Aβ deposition, the regular consumption of pure milk may improve AD pathology. Similar to our results, a prospective cohort study in a Japanese older population suggested that higher milk and dairy consumption may reduce the risk of dementia in older adults [
59]. A Spanish study found that whole milk consumption was associated with higher MMSE scores in people at high cardiovascular risk [
60]. However, other studies [
61,
62,
63] have concluded that milk consumption is detrimental to cognitive function in older populations. A study found that higher milk and dairy consumption among older adults over 60 was negatively related to cognitive function, although protein consumption was positively related to cognitive performance [
61]. Maija et al. [
62] found that whole milk may cause cognitive decline. A prospective cohort study in France found that eating more than the recommended amount of dairy products may damage the cognitive function of older women [
63]. A precise conclusion has not been reached regarding the relationship between milk intake and cognition, which may be influenced by the fat content and intake of milk. The protective effect of pure milk on cognitive impairment in middle-aged and older adults may be attributed to its content of protein, minerals, vitamins, and essential amino acids. Metabolic diseases such as type 2 diabetes, hypertension, dyslipidemia, and obesity are associated with an increased risk of cognitive impairment [
3]. Milk consumption can promote cognitive function in middle-aged and elderly adults by improving neurovascular dysfunction, as well as reducing weight and the risk of metabolic diseases [
64,
65].
The limitation of this study can be found in the previous article [
26]. The study was conducted among a specific population of middle-aged and elderly adults in Chinese, and the results may not be generalizable to other populations or age groups.