We used data from the NHANES and NDI databases from 2009 to 2018 and combined them to examine the relationship between tea consumption, PA, and all-cause mortality. Initially, we collected data on 49,693 participants. After excluding those with missing data on crucial variables such as tea consumption, mortality outcomes, and physical activity, the cohort was reduced to 26,544 participants. These exclusions were essential to maintain the reliability of our analyses. Subsequently, due to missing covariate data and non-compliance with study criteria, an additional 5,194 participants were removed. Ultimately, 21,350 participants were included in the further analysis. A χ2 test was conducted to compare the baseline characteristics between the original 49,693 participants and the analyzed sub-cohort of 21,350 (Supplementary Table 5). Notably, the results showed a significant difference in age (χ² = 4694.03, P < 0.001), which may be due to the larger number of participants aged 20–40 who were excluded (23,361 participants) compared to those aged 40–60 (2,249 participants) and over 60 (2,733 participants). This disparity in age distribution could introduce bias in age-stratified analyses, thus, age-related results should be approached with caution, especially concerning the 20–40 age group. To test the consistency of our findings, we excluded this age group and achieved similar results (Supplementary Table 6).
We identified tea drinks based on the food/beverage category given by WWEIA and calculated each person’s weekly MET score based on the recommended MET score given by NHANES. According to our study, in addition to the health benefits of PA and tea consumption, adding adequate tea consumption to PA has further health benefits.
We first examined the effects of PA and tea consumption on all-cause mortality. Consistent with previous studies7,8,33, both PA and tea consumption reduced all-cause mortality, possibly related to their cardiovascular, cancer, diabetes, and respiratory benefits5,9,34,35,36,37. On this basis, we separately evaluated the effects of tea consumption in individuals categorized by different levels of physical activity. We found that, after adjustment, tea consumption did not significantly affect all-cause mortality in the inactive group, whereas it demonstrated a significant beneficial effect in the active group. This differential impact underscores the potential moderating role of physical activity in the relationship between tea consumption and mortality risk. Although most studies have focused on how tea consumption reduces oxidative stress caused by PA, there is a lack of research on how PA affects tea itself. However, in Arabzadeh et al.‘s study, it was observed that markers (HIF-1α, BNIP3, and IGFBP3) of cardiomyocyte apoptosis in aging rats were lower in the group that used green tea extract combined with PA compared to those using green tea extract alone38. This implies that there might be mechanisms through which PA could potentially amplify the health benefits of tea consumption, suggesting a need for further exploration.
Afterward, we conducted a stratified analysis and found that tea consumption had a more significant effect on reducing all-cause mortality in the physically active group. Specifically, tea consumption demonstrated a significant effect in 12 out of 10 strata analyzed within the physically active group, including age > = 60 years, male gender, Non-Hispanic White ethnicity, Other Race-Including Multi-Racial, normal weight status, nonsmokers, former smokers, no alcohol consumption, no high cholesterol, absence of diabetes, presence of hypertension, and absence of cancer. In contrast, in the physically inactive group, tea consumption significantly reduced all-cause mortality only within the subgroup with hypertension. Interaction analysis showed a significant interaction between tea consumption and age in the active group, inactive group, and total group. In the study by Chen et al., it was similarly found that among individuals aged > 60 years, daily consumption of 2–4 cups of tea or > = 5 cups of tea can lead to a greater reduction in all-cause mortality compared to those aged < 60 years5. However, in the study by Wu et al. on the impact of tea consumption on all-cause mortality in populations with metabolic syndrome4, it was found that the effects of tea are more pronounced in individuals aged < 60 years. This difference may partly be attributed to the unique characteristics of the population with metabolic syndrome; however, it is more likely that age is indeed a critical factor affecting the impact of tea consumption on all-cause mortality rates. Additionally, they also found a significant interaction between tea consumption and age in their study (P < 0.1). Therefore, more research is needed to confirm the effects of tea consumption in different age groups.
As for our findings, the result indicating that tea consumption increases the risk of all-cause mortality among the 20–40 age group is not entirely convincing for several reasons:
-
1.
The number of deaths within this group is relatively small, with only 70 cases of mortality recorded among 7,131 individuals, which could introduce significant bias.
-
2.
The age distribution in the population was significantly different between the original cohort and the final inclusion cohort (χ² = 4694.03, P < 0.001), suggesting potential selection bias.
-
3.
An interaction between age and tea consumption was observed (P value of interaction test: Inactive 0.0081, Active 0.02, Total 0.0014), which might alter the impact of tea on mortality rates.
These factors should be carefully considered when interpreting the effect of tea consumption on mortality among younger adults.
After that, we divided tea consumption into three groups: None (0 g), Low (0–300 g), and High (> 300 g). This division was meticulously chosen to elucidate the differential impacts on all-cause mortality. The 0 g category served as a baseline to evaluate the effects of any tea consumption versus none. The 300 g threshold was chosen as it approximates the median consumption among tea drinkers, which is actually 333 g. The corresponding percentile for 300 g is 46.3%, a variation we considered acceptable. Furthermore, 300 g is a more practical reference for public understanding compared to 333 g. In the physically active group, individuals in the High consumption category showed a significant reduction in all-cause mortality compared to those in the None category, whereas the effect was not significant in the Low consumption category. This suggests that a sufficient amount of tea consumption may be necessary to significantly benefit from its mortality-reducing effects. Similarly, in Model 2, among the physically inactive group, neither the Low nor High consumption categories showed a significant reduction in all-cause mortality. In the study by Chen et al., it was also found that the threshold consumption for the lowest mortality risk was approximately 3 cups per day of tea5. Specifically, consuming 6 cups per day of tea was associated with the lowest risk of digestive disease mortality5. After adjusting for confounding factors, participants who consumed 2–4 cups of tea per day had the lowest risks of all-cause mortality, cardiovascular disease mortality, and respiratory disease mortality5. Consumption of > = 5 cups per day showed the lowest mortality risk for digestive system diseases5. In the study by Shin et al., it was found that there is a significant association between green tea consumption and reducing cardiovascular-specific mortality39. Furthermore, this effect increases with higher levels of tea intake39.
In the joint effect model, compared with the inactive and non-tea drinking group, the high tea group had the lowest HR in all three PA groups. Although the result was not statistically significant in the low MET group, which was consistent with previous analyses, it still reflected the trend of tea consumption in reducing all-cause mortality. Although high tea consumption was associated with the lowest HR in the medium MET group, low tea consumption was associated with increased HR compared with no tea consumption. However, HR tended to decrease with increasing tea consumption in the high tea group, which further suggests that adequate PA may play an integral role in the beneficial effects of tea consumption. Tokuda et al. also found that supplementing with tea catechins (TCCs), beneficial compounds found in tea, after resistance exercise (RE), improves skeletal muscle mass in older adults with sarcopenia, compared to RE alone40. Another study found that leg muscle mass and usual walking speed did not improve or even decreased in the TCCs intake group. In contrast, these parameters were significantly improved in the exercise group and further enhanced in the exercise plus TCCs group41.
In the study of tea consumption and cancer-specific mortality, we opted for this binary categorization as we did not observe a significant difference in cancer-specific mortality rates between those consuming more than 300 g daily and those consuming 0–300 g (Supplementary Table 7). Instead, the presence or absence of tea consumption appeared more impactful. Thus, we reported the outcomes using tea consumption as a binary variable to reflect these findings more accurately.
We found that although tea consumption can reduce cancer-specific mortality rate in the overall population, its effect on reducing cancer-specific mortality rate is not significant in the physically inactive group, whereas in the physically active group, this effect is significant. This result is innovative compared to previous studies on the protective effects of tea consumption against cancer, suggesting that when considering using tea to assist in improving cancer prognosis, the critical role of exercise should not be overlooked. There is abundant research on the benefits of tea against cancer. Deng et al.‘s study on dark tea indicates that tea’s role in cancer prevention and control primarily involves anti-oxidation, anti-inflammation, inhibiting cancer cell proliferation, inducing cancer cell apoptosis, inhibiting tumor metastasis, and regulating intestinal flora42. Singh et al.‘s research on black tea points out that regular consumption of phytochemical-rich black tea is associated with the regulation of several molecular targets, including COX-2, 5-LOX, AP-1, JNK, STAT, EGFR, AKT, Bcl2, NF-κB, Bcl-xL, caspases, p53, FOXO1, TNFα, PARP, and MAPK, which may form the basis of black tea’s preventive and curative effects against cancer43. Farhan’s study highlights that catechins are the effective antioxidants among the physiologically active compounds found in Camellia sinesis44, and its derivative EGCG has the greatest anti-inflammatory and anticancer potential44. EGCG has been demonstrated to have a chemopreventive effect by inhibiting processes such as carcinogenesis initiation, promotion, and progression45. Moreover, this catechin plays a role in cancer management by modulating various cellular signaling pathways, including regulation of proliferation, apoptosis, angiogenesis, and induction of apoptosis in various cancer cell types45.
Further health benefits of tea consumption may be explained by oxidative stress (Supplementary Fig. 2). Muscular contraction has been shown to generate several reactive radicals, such as superoxide, hydrogen peroxide, nitric oxide, and hydroxyl radicals, while unscavenged oxidants can modify macromolecules in the cell including nucleic acids, proteins, and lipids17. This oxidation of cellular components (oxidative stress) can occur when an imbalance exists between oxidants and antioxidants17. In general, the body can neutralize exercise-induced oxidative stress through antioxidant defense, and long-term exercise can also enhance this ability17, but this defense may be overwhelmed by exercise-induced ROS production14. Green tea has obvious antioxidant activity, especially green tea catechins can fight oxidative stress through direct or indirect pathways in the body, thereby protecting the body from oxidative damage14, which may reduce the possible negative effects of exercise to a certain extent, so as to further improve health.
This study has several shortcomings. First, tea consumption was determined by only two dietary questionnaires, which is uncertain for determining whether a person is a habitual tea drinker. Secondly, there is a lack of mechanistic studies on the interaction between tea drinking and PA, which is worthy of further investigation in our present study. Thirdly, due to missing data, we excluded 28,343 interviewees. Chi-square tests indicated significant differences in the distribution of baseline data before and after exclusion, which may lead to potential selection bias. Therefore, further randomized controlled trials are needed to determine the associations between tea consumption, PA and all-cause mortality, as well as cancer-specific mortality. Further animal studies are needed to determine the interaction between tea consumption and PA.
link