ABSTRACT
Background
Caffeine, identified as a central nervous system stimulant in foods, beverages (coffee, tea, chocolate), and medications, has been focused on its ergogenic properties, enhancing physical performance. The aim of this study was to investigate the association between the caffeine intake (from coffee) and fat-free mass index (FFMI).
Materials and Methods
We carried out a cohort study that included 3,466 women and 3,145 men aged ≥20 years who were intaking caffeine. Caffeine intake from coffee were obtained from two 24-hour dietary recall interviews. The FFMI was calculated as FFM (kg) divided by height in m2. The caffeine intake was classified into quartiles and combined into 4 groups. Multiple linear regression model analysis and multiple logistic regression model analysis were used to assess associations between the caffeine and FFMI adjusted for potential confounders.
Results
Among the 2,427 participants, males accounted for 52.4%, and females 47.6%. In multiple linear regression model, Model 1 (unadjusted Model (p = 0.041)) and Model 2 (adjusted for age, race, and BMI (p = 0.006)) in women showed a significant relationship between caffeine intake and FFMI. In multivariable models, caffeine intake and FFMI were significantly different (p < 0.05). In sex subgroups, among females, each quartile of caffeine intake was positively correlated with FFMI levels in the average FFMI group in Model 3 (p < 0.001). In age subgroups, each quartile of caffeine intake was positively correlated with FFMI levels in the average FFMI group in Model 3 for individuals aged 20–40 (p = 0.039) and those aged above 40 (p = 0.016). In drinking status subgroups, if they drunk alcohol, each quartile was positively correlated with FFMI levels in the average FFMI group in Model 3 (p < 0.001).
Conclusion
Caffeine intake was mainly positively associated with FFMI, especially in women with above levels of FFMI. Longitudinal studies and randomized controlled trials are needed to establish causality and provide evidence-based recommendations regarding caffeine intake to optimize muscle health.
KEYWORDS: Caffeine, fat-free mass index, metabolic equivalent of task, cohort study
1. Introduction
The relationship between dietary components and physical fitness indicators has long been of interest to the scientific community [1]. Caffeine, identified as a central nervous system stimulant in foods, beverages (coffee, tea, chocolate), and medications, has undergone extensive study for its ergogenic properties, enhancing physical performance [2]. Caffeine is favored by athletes and non-athletes alike for its ability to stimulate the central system, increase alertness and concentration, and potentially improve physical performance in a short period of time [3,4]. However, beyond the immediate effects on athletic performance and metabolism, the long-term effects of caffeine intake on muscle development and body composition have not been fully discovered or finalized. Compared to traditional assessment tools such as the body mass index (BMI), the fat-free mass index (FFMI) provides a more nuanced perspective on assessing a person’s body composition by accurately measuring a person’s lean muscle mass minus fat [5]. Especially among people with different levels of physical activity and muscle development, FFMI can distinguish lean muscle mass from overall shape (including fat content), thus effectively assessing nutritional status, physical fitness, and the potential for muscle growth [6,7].
Recent studies have revealed that moderate caffeine intake has a positive effect on promoting muscle enlargement [8]. Caffeine activates the central nervous system and promotes the release of neurotransmitters, allowing muscles to utilize energy more efficiently, which in turn improves the strength and durability of exercise and benefits muscle growth [4,9]. Caffeine may indirectly contribute to muscle growth by regulating fat oxidation and metabolic processes. A study published in the Journal of the International Society of Sports Nutrition utilized a double-blind, placebo-controlled crossover design to investigate the effects of caffeine intake on fat oxidation during a graded exercise test. The results indicated that participants in the caffeine group exhibited significantly higher fat oxidation rates across all exercise intensities compared to the placebo group [10,11]. A review from the United States conducted a systematic review and meta-analysis of 94 studies and found that caffeine ingestion increases fat metabolism [12]. Additionally, caffeine has been found to increase basal metabolic rate at rest, prompting the collective to utilize fat more than muscle as a source of energy, according to the study [13,14].
However, the effects of caffeine on muscle production are influenced by a variety of factors and may vary depending on an individual’s physiological response and intake [2]. Some studies have shown that high doses of caffeine intake negatively affect muscle growth by producing sleep disturbances, increasing anxiety and tension or affecting calcium absorption in the bones [15,16]. In addition, individual tolerance to caffeine may vary, with some individuals being more sensitive to its effects while others may be less affected. This research is intended to inform evidence-based dietary guidelines and help individuals make informed choices about caffeine consumption in relation to their fitness and health goals [17,18].
2. Materials and methods
The National Health and Nutrition Examination Survey (NHANES) use a stratified, multistage, probabilistic clustering design to collect a representative sample of the non-institutionalized U.S. population. The database contains comprehensive dietary intake records and accurate body composition measurements. This study followed the Standards for Reporting Enhanced Reporting of Observational Studies in Epidemiology (STROBE) reporting guidelines.
2.1. Study population
All the databases could be obtained from the NHANES website (https://wwwn.cdc.gov/nchs/nhanes/Default.aspx). The independent validation sample used here was from NHANES including 39,156 participants during the 2011–2018. This study excluded those who aged under 20 years (n = 16,539), without caffeine intake (n = 5,486) and FFMI data (n = 8,221), as well as those without covariate data (including BMI (n = 14), physical activity (n = 1,709), drinking status (n = 572), smoking status (n = 4)). The final study population was 6,611 adult participants (Figure 1).
Figure 1.
Flowchart of participants selection from the NHANES 2011–2018.
2.2. Anthropometry and variable conversion
Caffeine intake from coffee were obtained from two 24-hour dietary recall interviews. All NHANES participants were eligible to receive two 24-hour dietary recall interviews. The first face-to-face recall interview was conducted in a private room at the Mobile Examination Center (MEC) by trained interviewers. The MEC Diet interview room provided a set of measurement guides, including various glasses, bowls, mugs, drink boxes and bottles, household spoons, measuring cups and spoons, rulers, thickness bars, beanbags, and circles, for participants to report food quantities. At the end of the MEC dietary interview, the interviewers would schedule them for a phone follow-up (PFU) interview 3–10 days later. In this study, caffeine intake was divided into quartiles as the independent variable based on previous literature [19,20]: Q1 = 0 mg/day, Q2 = 0 mg/day ~40 mg/day, Q3 = 40 mg/day ~400 mg/day, and Q4 > 400 mg/day. As the dependent variable, FFMI has been increasingly recognized as a superior metric compared to BMI for evaluating muscle mass and physical fitness, particularly in populations with varying levels of physical activity. FFMI was calculated as FFM (kg) divided by height in m2, which allows for a more accurate assessment of body composition by focusing on lean mass rather than overall weight [6]. To categorize FFMI, sex-specific thresholds were applied, with men classified into three groups: skinny (<18.0 kg/m2), average (18.0–20.0 kg/m2), and above average (>20.0 kg/m2). For women, the categories were: skinny (<15.0 kg/m2), average (15.0–17.0 kg/m2), and above average (>17.0 kg/m2). BF% was determined through dual-energy X-ray absorptiometry (DXA), a gold-standard method known for its accuracy in measuring body composition, particularly in distinguishing between fat and lean mass. The application of these methods ensures a robust and scientifically sound assessment of body composition, critical for analyzing the potential impacts of dietary and exercise interventions on muscle development [21]. Fat-free mass (FFM) was calculated by multiplying body weight by the FFM percentage, with FFM% derived by subtracting the body fat percentage (BF%) from 100%. The body measures data were collected, in the Mobile Examination Center (MEC), by trained health technicians. Standing height is measured using a stadiometer with a fixed vertical backboard and an adjustable head piece. Participants were required to wear a strict dress code and use a standard scale to measure their weight. For more information on the data measurement process, please refer to the NHANES Anthropometric Procedure Manual (https://wwwn.cdc.gov/nchs/data/nhanes/2011-2012/manuals/anthropometry_procedures_manual.pdf).
The covariates were mainly demographic data such as age (in years) BMI, physical activity (metabolic equivalent of task (MET) ≤960 hr/week, 960 hr/week − 2480 hr/week, 2480 hr/week − 6720 hr/week, >6720 hr/week), drinking status (yes, no) and smoking status (current, former, never), race (Mexican American or Hispanic, Non-Hispanic White, Non-Hispanic Black, Other Race). BMI is a commonly used international indicator to measure the degree of fatness and thinness of the human body and whether it is healthy; BMI = weight (kg)/height (m)2. BMI <25 kg/m2 was considered normal, 25 kg/m2 ≤ BMI <29.9 kg/m2 was considered overweight, and BMI ≥30 kg/m2 was considered obese. Physical activity referred to any body movement that results in higher energy expenditure than at rest, and the MET could measure the intensity of physical activity [22].
2.3. Statistical analyses
Data were presented as mean (standard deviation) or median (interquartile range [IQR]) for continuous variables and as numbers (percentages) for categorical variables. In this study, the odds ratio (OR) effect size was used to measure the correlation between two events, and 95% confidence intervals (CI) were used to quantify the uncertainty in the parameter estimates to assess the reliability and precision of the results. Because of differences in the distribution of FFMI between the sexes, men and women were analyzed separately. Caffeine intake was categorized into four quartile groups: Q1 (lowest group), Q2, Q3, and Q4 (highest group). Multivariable logistic regression models were performed to explore the independent association of caffeine intake concentration with FFMI after adjusting for potential confounding factors. Subgroup analyses stratify by sex, age, MET and drinking status were performed. Three models were constructed: an unadjusted model, a minimally adjusted model, and a fully adjusted model. A general linear Model was used for the analysis of continuous variables. For categorical variables, chi-squared tests were used to analyze whether smoking and drinking status differed significantly between FFMI quartile groups. R software was employed for the all statistical analysis. p values for linear trends were calculated based on quartiles of FFMI. P-value represents the probability that the observed association is due to chance. In this study, a P-value of less than 0.05 was considered statistically significant, indicating that the observed associations are unlikely to have occurred by chance.
3. Results
3.1. Basic characteristics
Participant characteristics, as depicted in Table 1. A total of 3,145 men and 3,466 women met all inclusion criteria, and their data were included in the analysis. The mean age of all participants was 38.98 (±11.86) and the caffeine intake level Q1 group consisted of 531 individuals with a mean age of 32.94 (±10.95). The Q2 group consisted of 944 individuals with a mean age of 35.57 (±12.28). There were 3, 582 individuals in the Q3 group with a mean age of 38.03 (±11.60). The Q4 group had 1, 554 individuals with a mean age of 43.18 (±10.97).
Table 1.
Baseline characteristics of participants according to caffeine intake levels in NHANES 2011–2018.
| Caffeine intake levels,mg (N = 6,611) |
P | |||||
|---|---|---|---|---|---|---|
| Total | Q1 | Q2 | Q3 | Q4 | ||
| Number of participants | 6,611 | 531 | 944 | 3,582 | 1,554 | |
| Age | 38.98 (11.86) | 32.94 (10.95) | 35.57 (12.28) | 38.03 (11.60) | 43.18 (10.97) | <0.001 |
| Gender | <0.001 | |||||
| Female | 3145 (47.6) | 229 (43.2) | 498 (52.8) | 1787 (49.9) | 662 (42.6) | |
| Male | 3466 (52.4) | 302 (56.8) | 446 (47.2) | 1795 (50.1) | 892 (57.4) | |
| Race/ethnicity | <0.001 | |||||
| Mexican American or Hispanic | 1101 (16.6) | 115 (21.8) | 199 (21.1) | 688 (19.2) | 147 (9.5) | |
| Non-Hispanic White | 4178 (63.2) | 204 (38.5) | 455 (48.2) | 2128 (59.4) | 1253 (80.6) | |
| Non-Hispanic Black | 708 (10.7) | 148 (27.8) | 189 (20.0) | 401 (11.2) | 44 (2.8) | |
| Other Race | 624 (9.4) | 64 (12.0) | 101 (10.7) | 365 (10.2) | 110 (7.1) | |
| BMI | 0.021 | |||||
| <25.0 | 2114 (32) | 174 (32.7) | 348 (36.9) | 1163 (32.5) | 452 (29.1) | |
| 25.0–29.9 | 2193 (33.2) | 176 (33.3) | 244 (25.8) | 1177 (32.9) | 567 (36.5) | |
| ≥30.0 | 2304 (34.8) | 181 (34.0) | 352 (37.3) | 1242 (34.7) | 535 (34.4) | |
| MET | 0.041 | |||||
| ≤960 | 1634 (24.7) | 89 (16.7) | 227 (24.0) | 938 (26.2) | 374 (24.1) | |
| 960–2480 | 1621 (24.5) | 111 (20.9) | 237 (25.1) | 856 (23.9) | 407 (26.2) | |
| 2480–6720 | 1746 (26.4) | 178 (33.5) | 261 (27.7) | 892 (24.9) | 419 (27.0) | |
| >6720 | 1610 (24.4) | 153 (28.9) | 219 (23.2) | 896 (25.0) | 354 (22.8) | |
| Marital status | <0.001 | |||||
| Married | 3331 (50.4) | 159 (29.9) | 430 (45.6) | 1741 (48.6) | 925 (59.5) | |
| Separated | 854 (12.9) | 39 (7.3) | 87 (9.2) | 448 (12.5) | 252 (16.2) | |
| Never married | 2426 (36.7) | 333 (62.8) | 427 (45.2) | 1393 (38.9) | 377 (24.2) | |
| Drinking status | <0.001 | |||||
| Yes | 5638 (85.3) | 410 (77.3) | 687 (72.8) | 3070 (85.7) | 1414 (91.0) | |
| No | 973 (14.7) | 121 (22.7) | 257 (27.2) | 512 (14.3) | 140 (9.0) | |
| Smoking status | <0.001 | |||||
| Current | 1343 (20.3) | 89 (16.7) | 116 (12.3) | 694 (19.4) | 399 (25.7) | |
| Former | 1316 (19.9) | 67 (12.7) | 126 (13.3) | 647 (18.1) | 421 (27.1) | |
| Never | 3952 (59.8) | 375 (70.6) | 702 (74.3) | 2241 (62.6) | 734 (47.2) | |
| FFMI | 0.117 | |||||
| Skinny | 1454 (22.0) | 111 (20.9) | 254 (26.9) | 799 (22.3) | 308 (19.8) | |
| Average | 1924 (29.1) | 137 (25.8) | 252 (26.7) | 1067 (29.8) | 457 (29.4) | |
| Above | 3233 (48.9) | 283 (53.3) | 438 (46.5) | 1716 (47.9) | 789 (50.8) | |
| Height | 169.59 (9.36) | 169.45 (9.12) | 168.36 (9.44) | 168.90 (9.52) | 171.25 (8.90) | <0.001 |
| Weight | 82.22 (20.04) | 82.77 (21.23) | 80.77 (21.85) | 81.69 (20.28) | 83.56 (18.53) | 0.07 |
| BF% | 0.32 (0.08) | 0.31 (0.08) | 0.33 (0.09) | 0.33 (0.08) | 0.32 (0.08) | 0.021 |
| FMI | 9.57 (4.19) | 9.37 (4.47) | 9.63 (4.52) | 9.72 (4.39) | 9.34 (3.61) | 0.124 |
Values are weighted mean (SE) for continuous variables or numbers (weighted %) for categorical variables.BMI,body mass index; BF%, body fat percentage; FMI, fat mass index; FFMI, Fat-free mass index; MET, Metabolic equivalent of task.
Meanwhile, as shown in Table 1, the BMI values of Q1, Q2, and Q3 groups showed that obese people were predominant, and the MET values of Q1, Q2, and Q3 groups were the most in the range of 2,480–6,720. The Q3 group had the highest percentage of smokers and drinkers, the median FFMI value was higher in the Q3 group than in the other three groups.The data indicates significant variation in caffeine intake levels across different racial/ethnic groups (p < 0.001). The majority of participants were Non-Hispanic White (63.2%). There were no significant differences in FFMI (p = 0.117) and FMI (p = 0.124) between quartiles, while BMI, MET, smoking, and drinking status were statistically significant (p < 0.05).
3.2. Multiple linear regression model analysis
The results of the multiple linear regression Model analysis of caffeine intake (quartiles) and FFMI were as follows (Table 2). Overall, caffeine intake and FFMI were significantly different (p < 0.05). Model 1 (unadjusted Model (β: −0.000368, 95%CI: −0.000721 ~ −0.000015, p = 0.041)) and Model 2 (adjusted for age, race, and BMI (β: 0.000179, 95%CI: 0.000053 ~ 0.000305, p = 0.006)) in women showed a significant relationship between caffeine intake and FFMI, whereas Model 3 (adjusted for marital status, smoking status, drinking status, height, weight, and MET based on Model 2 (β: 0.000128, 95%CI: −0.000001 ~ 0.000257, p = 0.051)) showed no statistical relationship. The three Models in men did not show a statistically significant relationship (p > 0.05).
Table 2.
Multivariable linear regression models between caffeine intake and fat-free mass index.
| Total |
Female |
Male |
||||
|---|---|---|---|---|---|---|
| β [95% CI] | P value | β [95% CI] | P value | β [95% CI] | P value | |
| Model 1 | 0.000324(0.000087,0.000562) | 0.007 | −0.000368(−0.000721,-0.000015) | 0.041 | 0.000146(−0.000111,0.000403) | 0.265 |
| Model 2 | 0.000567(0.000423,0.000711) | <0.0001 | 0.000179(0.000053,0.000305) | 0.006 | 0.000071(−0.000031,0.000174) | 0.172 |
| Model 3 | 0.000128(0.000011,0.000245) | 0.032 | 0.000128(−0.000001,0.000257) | 0.051 | 0.00007(−0.000033,0.000174) | 0.183 |
β partial regression coefficient, CI confidence interval, FFMI fat-free mass index. Model 1: Unadjusted. Model 2: Adjusted for age, race, BMI. Model 3: Adjusted for marital status, smoking status, drinker, Height, Weight and metabolic equivalent of task in addition to model 2.
3.3. Multiple logistic regression model analysis
The results of the multiple logistic regression Model analysis of caffeine intake (quartiles) and FFMI are as follows.
3.4. Sex
In sex subgroups, caffeine is positively correlated with FFMI levels in the women’s average FFMI group for the Q2 (OR: 1.5900, 95%CI: 1.5890 ~ 1.5910), Q3 (OR: 1.5710, 95%CI: 1.5690 ~ 1.5730) and Q4 (OR: 2.7350, 95%CI: 2.7320 ~ 2.7390) cohorts in Model 3 (based on Model 2 adjusting for marital status, smoking status, drinking status, height, weight, and MET) (p = 0.001). Caffeine is positively correlated with FFMI levels in the women’s above FFMI group for the Q3 (OR: 1.4810, 95%CI: 1.4800 ~ 1.4830) and Q4 (OR: 2.5890, 95%CI: 2.5870 ~ 2.5910) cohorts in Model 3 (p = 0.001). Caffeine is positively correlated with FFMI levels in the men’s average FFMI group for the Q2 (OR: 1.0240, 95%CI: 1.0230 ~ 1.0250) and Q4 (OR: 1.3400, 95%CI: 1.3390 ~ 1.3410) cohorts in Model 3, for Q3 (OR: 0.9640, 95%CI: 0.9630 ~ 0.9650) is negatively correlated (p = 0.072). Caffeine is negatively correlated with FFMI levels in the men’s above FFMI group for the Q2 (OR: 0.75525, 95%CI: 0.75518 ~ 0.75533) and Q3 (OR: 0.7250, 95%CI: 0.7240 ~ 0.7260) in Model 3, for the Q4 (OR: 1.5070, 95%CI: 1.5060 ~ 1.5070) is positively correlated (p = 0.001) (Table 3).
Table 3.
Associations between caffeine intake and fat-free mass index: multivariable logistic regression analyses by sex.
| Caffeine intake, mg |
P for trend | ||||
|---|---|---|---|---|---|
| Q1 | Q2 | Q3 | Q4 | ||
| Female | |||||
| FFMI (Average) | |||||
| Model 1 | 1 (ref) | 1.293(0.823,2.031) | 1.148(0.767,1.721) | 1.918(1.230,2.992) | <0.001 |
| Model 2 | 1 (ref) | 1.472(0.795,2.725) | 1.633(0.937,2.817) | 2.940(1.582,5.466) | <0.001 |
| Model 3 | 1 (ref) | 1.590(1.589,1.591) | 1.571(1.569,1.573) | 2.735(2.732,2.739) | 0.001 |
| FFMI (Above) | |||||
| Model 1 | 1 (ref) | 0.845(0.573,1.246) | 0.909(0.647,1.277) | 1.142(0.777,1.679) | 0.060 |
| Model 2 | 1 (ref) | 0.869(0.377,2.006) | 1.633(0.777,3.433) | 3.320(1.439,7.662) | <0.001 |
| Model 3 | 1 (ref) | 0.905(0.905,0.905)† | 1.481(1.480,1.483) | 2.589(2.587,2.591) | 0.001 |
| Male | |||||
| FFMI (Average) | |||||
| Model 1 | 1 (ref) | 1.026(0.673,1.564) | 1.173(0.823,1.671) | 1.277(0.873,1.869) | 0.164 |
| Model 2 | 1 (ref) | 1.074(0.597,1.932) | 1.013(0.612,1.675) | 1.464(0.833,2.575) | 0.047 |
| Model 3 | 1 (ref) | 1.024(1.023,1.025) | 0.964(0.963,0.965) | 1.340(1.339,1.341) | 0.072 |
| FFMI (Above) | |||||
| Model 1 | 1 (ref) | 0.921(0.637,1.331) | 1.048(0.770,1.426) | 1.235(0.886,1.723) | 0.039 |
| Model 2 | 1 (ref) | 0.774(0.364,1.649) | 0.793(0.416,1.507) | 1.667(0.817,3.401) | 0.001 |
| Model 3 | 1 (ref) | 0.755(0.755,0.755)‡ | 0.725(0.724,0.726) | 1.507(1.506,1.507) | 0.001 |
Model 1: Unadjusted. Model 2: Adjusted for age, race, BMI. Model 3: Adjusted for marital status, smoking status, drinker, Height, Weight and MET in addition to model 2. †: 0.9054 (0.9051, 0.9057), ‡: 0.75525 (0.75518, 0.75533). FFMI, Fat-free mass index.
3.5. Age
In age subgroups, aged between 20 and 40, caffeine is positively correlated with FFMI levels in the average FFMI group for the Q2 (OR: 1.4745, 95%CI: 1.4741 ~ 1.4748), Q3 (OR: 1.1699, 95%CI: 1.1688 ~ 1.1709) and Q4 (OR: 1.7741, 95%CI: 1.7733 ~ 1.7748) cohorts in Model 3 (Model 2 adjusted for marital status, drinking status, smoking status, height, weight, and MET) (p = 0.039). Caffeine is negatively correlated with FFMI levels in the above FFMI group for the Q2 (OR: 0.9401, 95%CI: 0.9400 ~ 0.9402) and Q3 (OR: 0.8574, 95%CI: 0.8569 ~ 0.8780) cohorts in Model 3, for the Q4 (OR: 1.6923, 95%CI: 1.6919 ~ 1.6927) is positively correlated (p = 0.008). When aged over 40 years, caffeine is negatively correlated with FFMI levels in the average FFMI group for the Q2 (OR: 0.7081, 95%CI: 0.7079 ~ 0.7084) and Q3 (OR: 0.9561, 95%CI: 0.9555 ~ 0.9567), for the Q4 (OR: 1.3392, 95%CI: 1.3386 ~ 1.3398) is positively correlated in Model 3 (p = 0.016). Caffeine is positively correlated with FFMI levels in the above FFMI group for the Q3 (OR: 1.0053, 95%CI: 1.0048 ~ 1.0058) and Q4 (OR: 1.7672, 95%CI: 1.7666 ~ 1.7678) in Model 3 (Table 4) (p < 0.001).
Table 4.
Associations between caffeine intake and fat-free mass index: multivariable logistic regression analyses by age.
| Caffeine intake,mg |
P for trend | ||||
|---|---|---|---|---|---|
| Q1 | Q2 | Q3 | Q4 | ||
| 20–40 | |||||
| FFMI (Average) | |||||
| Model 1 | 1 (ref) | 1.2040(0.8397,1.7265) | 1.1413(0.8327,1.5644) | 1.4479(1.0058,2.0844) | 0.055 |
| Model 2 | 1 (ref) | 1.3660(0.8440,2.2111) | 1.2022(0.7899,1.8298) | 1.7698(1.0798,2.9007) | 0.034 |
| Model 3 | 1 (ref) | 1.4745(1.4741,1.4748) | 1.1699(1.1688,1.1709) | 1.7741(1.7733,1.7748) | 0.039 |
| FFMI (Above) | |||||
| Model 1 | 1 (ref) | 0.8312(0.6068,1.1386) | 1.0439(0.7983,1.3651) | 1.1778(0.8586,1.6156) | 0.049 |
| Model 2 | 1 (ref) | 0.8782(0.4608,1.6737) | 0.9241(0.5299,1.6116) | 1.7993(0.9397,3.4454) | 0.005 |
| Model 3 | 1 (ref) | 0.9401(0.9400,0.9402) | 0.8574(0.8569,0.8580) | 1.6923(1.6919,1.6927) | 0.008 |
| >40 | |||||
| FFMI (Average) | |||||
| Model 1 | 1 (ref) | 0.9197(0.5081,1.6647) | 0.9046(0.5388,1.5188) | 1.1377(0.6658,1.9439) | 0.093 |
| Model 2 | 1 (ref) | 0.6876(0.2987,1.5824) | 0.9337(0.4506,1.9346) | 1.3325(0.6203,2.8624) | 0.015 |
| Model 3 | 1 (ref) | 0.7081(0.7079,0.7084) | 0.9561(0.9555,0.9567) | 1.3392(1.3386,1.3398) | 0.016 |
| FFMI (Above) | |||||
| Model 1 | 1 (ref) | 0.8238(0.4875,1.3923) | 0.6839(0.4324,1.0817) | 0.8389(0.5212,1.3502) | 0.414 |
| Model 2 | 1 (ref) | 0.4793(0.1659,1.3854) | 0.9983(0.3944,2.5271) | 1.7680(0.6681,4.6787) | <0.001 |
| Model 3 | 1 (ref) | 0.4944(0.4943,0.4946) | 1.0053(1.0048,1.0058) | 1.7672(1.7666,1.7678) | <0.001 |
Model 1: Unadjusted. Model 2: Adjusted for gender, race, BMI. Model 3: Adjusted for marital status, smoking status, drinker, Height, Weight and MET in addition to model 2. FFMI, Fat-free mass index.
3.6. MET
In MET subgroups, MET ≤ 960, caffeine is positively associated with FFMI levels in the average FFMI group for the Q2 (OR: 1.8900, 95%CI: 1.0400 ~ 3.4500) and Q4 (OR: 1.8900, 95%CI: 1.0600 ~ 3.3700) cohorts in Model 1 (unadjusted Model) (p = 0.1430); The Q4 cohort is positively associated with FFMI levels in the above FFMI group in both Model 1 (unadjusted Model (OR: 1.6600, 95%CI: 1.0200 ~ 2.6900, p = 0.0250)) and Model 2 (adjusted for age, ethnicity, and sex (OR: 1.9500, 95%CI: 1.1300 ~ 3.3700, p = 0.0080)). When 2,480<MET ≤6,720, the Q4 (OR: 3.3800, 95%CI: 1.2000 ~ 9.5000, p = 0.0020) cohort is positively associated with FFMI levels in the above FFMI group in Model 3 (Table 5).
Table 5.
Associations between caffeine intake and fat-free mass index: multivariable logistic regression analyses by MET.
| Caffeine intake,mg |
|||||
|---|---|---|---|---|---|
| Q1 | Q2 | Q3 | Q4 | P for trend | |
| MET (≤960) | |||||
| FFMI (Average) | |||||
| Model 1 | 1 (ref) | 1.89(1.04,3.45) | 1.34(0.79,2.27) | 1.89(1.06,3.37) | 0.143 |
| Model 2 | 1 (ref) | 1.77(0.96,3.27) | 1.24(0.72,2.15) | 1.61(0.87,2.98) | 0.408 |
| Model 3 | 1 (ref) | 1.70(0.70,4.09) | 1.14(0.53,2.48) | 1.69(0.70,4.07) | 0.316 |
| FFMI (Above) | |||||
| Model 1 | 1 (ref) | 1.32(0.79,2.19) | 1.14(0.74,1.76) | 1.66(1.02,2.69) | 0.025 |
| Model 2 | 1 (ref) | 1.30(0.75,2.27) | 1.29(0.79,2.09) | 1.95(1.13,3.37) | 0.008 |
| Model 3 | 1 (ref) | 0.65(0.19,2.27) | 0.93(0.31,2.81) | 1.75(0.52,5.95) | 0.031 |
| MET (960–2480) | |||||
| FFMI (Average) | |||||
| Model 1 | 1 (ref) | 0.73(0.39,1.39) | 1.00(0.57,1.76) | 1.46(0.80,2.65) | 0.004 |
| Model 2 | 1 (ref) | 0.81(0.42,1.56) | 1.06(0.60,1.89) | 1.37(0.73,2.57) | 0.045 |
| Model 3 | 1 (ref) | 0.67(0.27,1.68) | 0.91(0.41,2.03) | 1.29(0.54,3.10) | 0.084 |
| FFMI (Above) | |||||
| Model 1 | 1 (ref) | 0.84(0.47,1.48) | 0.99(0.60,1.64) | 1.10(0.63,1.90) | 0.292 |
| Model 2 | 1 (ref) | 1.01(0.55,1.86) | 1.16(0.67,2.00) | 1.21(0.67,2.21) | 0.483 |
| Model 3 | 1 (ref) | 0.76(0.22,2.58) | 0.99(0.33,2.90) | 1.38(0.43,4.48) | 0.202 |
| MET (2480–6720) | |||||
| FFMI (Average) | |||||
| Model 1 | 1 (ref) | 0.92(0.52,1.62) | 1.05(0.64,1.73) | 1.27(0.73,2.20) | 0.171 |
| Model 2 | 1 (ref) | 0.92(0.52,1.64) | 1.03(0.62,1.71) | 1.17(0.65,2.10) | 0.368 |
| Model 3 | 1 (ref) | 1.12(0.51,2.45) | 1.27(0.63,2.56) | 1.93(0.86,4.36) | 0.061 |
| FFMI (Above) | |||||
| Model 1 | 1 (ref) | 0.76(0.46,1.28) | 0.87(0.56,1.36) | 1.24(0.76,2.03) | 0.028 |
| Model 2 | 1 (ref) | 0.82(0.48,1.40) | 1.00(0.63,1.60) | 1.55(0.91,2.66) | 0.005 |
| Model 3 | 1 (ref) | 1.24(0.45,3.41) | 1.20(0.49,2.93) | 3.38(1.20,9.50) | 0.002 |
| MET (>6720) | |||||
| FFMI (Average) | |||||
| Model 1 | 1 (ref) | 1.43(0.73,2.79) | 1.35(0.76,2.40) | 1.60(0.87,2.95) | 0.249 |
| Model 2 | 1 (ref) | 1.33(0.68,2.61) | 1.34(0.75,2.42) | 1.64(0.85,3.16) | 0.195 |
| Model 3 | 1 (ref) | 1.42(0.58,3.50) | 1.53(0.69,3.39) | 2.53(1.03,6.24) | 0.033 |
| FFMI (Above) | |||||
| Model 1 | 1 (ref) | 0.80(0.45,1.42) | 1.04(0.64,1.66) | 0.98(0.59,1.64) | 0.829 |
| Model 2 | 1 (ref) | 0.76(0.42,1.35) | 1.06(0.65,1.74) | 1.08(0.62,1.88) | 0.464 |
| Model 3 | 1 (ref) | 0.52(0.16,1.66) | 0.91(0.34,2.46) | 1.69(0.55,5.18) | 0.035 |
Model 1: Unadjusted. Model 2: Adjusted for age, race, gender. Model 3: Adjusted for BMI, marital status, smoking status, drinking status, Height, Weight in addition to model 2. FFMI, Fat-free mass index; MET, Metabolic equivalent of task.
3.7. Drinking status
In drinking status subgroups, if they drink alcohol, each quartile is positively correlated with FFMI levels in the average FFMI group in Model 3 (adjusting for marital status, smoking status, height, weight, and MET based on Model 2) (p < 0.001). The Q3 (OR: 1.1904, 95%CI: 1.1898 ~ 1.1910) and Q4 (OR: 2.2921, 95%CI: 2.2914 ~ 2.2929) cohorts are positively correlated with FFMI levels in the above FFMI group in Model 3 (p < 0.001). If they do not drink alcohol, the Q2 (OR: 1.1558, 95%CI: 1.1541 ~ 1.1575) cohorts is positively correlated with FFMI levels in the average FFMI group in Model 3, and the Q3 (OR: 0.6928, 95%CI: 0.6919 ~ 0.6937) and Q4 (OR: 0.9346, 95%CI: 0.9336 ~ 0.9357) cohorts are negatively correlated (p = 0.719). Caffeine is negatively correlated with FFMI levels growth in the above FFMI group for each quartile except Q1 in Model 3 (Table 6) (p = 0.433). However, the two negative associations mentioned above were not statistically significant.
Table 6.
Associations between caffeine intake and fat-free mass index: multivariable logistic regression analyses by drinking status.
| Caffeine intake,mg |
P for trend | ||||
|---|---|---|---|---|---|
| Q1 | Q2 | Q3 | Q4 | ||
| Drinker | |||||
| FFMI (Average) | |||||
| Model 1 | 1 (ref) | 1.0976(0.7584,1.5884) | 1.1119(0.8106,1.5252) | 1.4025(1.0008,1.9655) | 0.008 |
| Model 2 | 1 (ref) | 1.0852(0.7466,1.5774) | 1.1225(0.8111,1.5534) | 1.3773(0.9637,1.9683) | 0.023 |
| Model 3 | 1 (ref) | 1.1971(1.1969,1.1974) | 1.3075(1.3066,1.3085) | 1.9908(1.9899,1.9917) | <0.001 |
| FFMI (Above) | |||||
| Model 1 | 1 (ref) | 0.8648(0.6270,1.1927) | 0.8629(0.6573,1.1328) | 1.0748(0.8016,1.4411) | 0.030 |
| Model 2 | 1 (ref) | 0.9135(0.6531,1.2778) | 1.0310(0.7720,1.3769) | 1.3659(0.9914,1.8818) | <0.001 |
| Model 3 | 1 (ref) | 0.9629(0.9628,0.9630) | 1.1904(1.1898,1.1910) | 2.2921(2.2914,2.2929) | <0.001 |
| No-dinker | |||||
| FFMI (Average) | |||||
| Model 1 | 1 (ref) | 1.2615(0.7305,2.1785) | 1.0578(0.6400,1.7485) | 1.7730(0.9604,3.2731) | 0.075 |
| Model 2 | 1 (ref) | 1.2869(0.7335,2.2581) | 0.9881(0.5873,1.6625) | 1.4331(0.7496,2.7398) | 0.412 |
| Model 3 | 1 (ref) | 1.1558(1.1541,1.1575) | 0.6928(0.6919,0.6937) | 0.9346(0.9336,0.9357) | 0.719 |
| FFMI (Above) | |||||
| Model 1 | 1 (ref) | 0.9294(0.5753,1.5015) | 1.2748(0.8306,1.9566) | 1.4454(0.8390,2.4901) | 0.080 |
| Model 2 | 1 (ref) | 0.9753(0.5818,1.6350) | 1.2490(0.7834,1.9912) | 1.3537(0.7448,2.4606) | 0.233 |
| Model 3 | 1 (ref) | 0.4950(0.4948,0.4953) | 0.4633(0.4628,0.4638) | 0.8911(0.8904,0.8918) | 0.433 |
Model 1: Unadjusted. Model 2: Adjusted for age, gender, race. Model 3: Adjusted for marital status, smoking status, Height, Weight and MET in addition to model 2. FFMI, Fat-free mass index.
4. Discussions
Large-scale cohort analyses utilizing the NHANES database reveal from 2011 to 2018, caffeine intake was found to be positively associated with FFMI overall, but at the same time, individual characteristics and lifestyle exerted a more complex influence on caffeine intake.
When analyzed by sex, caffeine intake was mainly positively associated with FFMI, especially in women with higher FFMI, suggesting a possible sex-specific effect. Previous studies have shown that there may be a real difference in the rate of caffeine metabolism between men and women, with women typically metabolizing caffeine at a slower rate than men [23]. The reason for this may involve several factors, and there are a number of current explanations. The first is sex hormone levels. CYP1A2, the main enzyme responsible for caffeine clearance, is more active in men than in women [24,25]. Estrogen can affect the activity of enzymes in the liver that metabolize caffeine [26]. Estrogen plays a regulatory role in the female body and may inhibit the activity of certain enzymes that metabolize caffeine, resulting in a slower rate of caffeine metabolism in the female body [27]. Next is the difference in body fat. In general, women tend to have a higher body fat content than men and adipose tissue may play a role in caffeine metabolism [28]. Adipose tissue, which is predominantly composed of fat cells, can indeed influence the metabolism of substances like caffeine. This is because fat cells can absorb and store caffeine, affecting its distribution and metabolism in the body [29]. As a result, women with more body fat may tend to metabolize caffeine more slowly [30,31]. However, it has also been found that in severely obese subjects, caffeine half-life and clearance are less affected by obesity, while caffeine volume of distribution is increased, and the effect is more marked in women [32]. The conclusions in the latter part are consistent with the results presented in the data of this study. The third factor is genetic, where an individual’s genetic background may also affect the rate of caffeine metabolism [5]. A number of genetic variants have been found to be associated with caffeine metabolism, and there may be differences in the distribution of these genetic variants in males and females, which in turn affects differences in the rate of metabolism [33]. Consequently, these factors collectively account for the observed results in the data.
Age-stratified analyses revealed different patterns of caffeine-FFMI associations in different age groups. The association between caffeine intake and FFMI was stronger in younger people than in older people based on data result, suggesting age-related metabolic differences. With aging, changes in the body’s metabolic processes involving several aspects, including energy metabolism, lipid metabolism, and glucose metabolism, may affect the absorption, distribution, metabolism, excretion of caffeine intake, etc. Firstly, When the gastrointestinal function of the elderly gradually fails, gastric acid secretion and gastrointestinal peristalsis become slower, thus affecting its absorption of a caffeine [34]. In addition, older adults typically have increased body fat content, which may result in an increased volume of distribution of caffeine, affecting its pharmacokinetics [35,36]. Secondly, aging may lead to a progressive decline in liver function, characterized by reduced activity of liver enzymes such as CYP1A2, which plays a crucial role in the metabolism of substances like caffeine, potentially impairing the efficiency of caffeine metabolism [25,37]. In addition, older adults may be more sensitive to the pharmacodynamics of caffeine and may experience greater caffeine-induced stimulation or adverse effects than younger adults [38,39]. Older adults may be at increased risk for drug interactions [40], that may affect the metabolism and pharmacodynamic response to caffeine. Age-related metabolic changes may also affect muscle synthesis and catabolism [41]. It is worth thinking about that, if metabolic changes result in decreased muscle synthesis or accelerated degradation, then even if caffeine intake has some facilitating effect on FFMI, it may be attenuated by the metabolic changes.
It should be noted that due to the focus on a single caffeine source, our study did not include other sources such as energy drinks or caffeine tablets. However, numerous studies have found that younger individuals are more likely to consume caffeine through energy drinks compared to other age groups. For instance, a study by Miller highlights that energy drinks are particularly popular among college students, with more than 50% reporting regular consumption [42]. This trend is driven by the need for quick energy boosts, especially during activities like studying, driving, and socializing. Another study by Evens et al. confirms that energy drink use is prevalent among young adults, often exceeding traditional coffee consumption, partly due to the targeted marketing and the convenience of these products [43]. Energy drink may affect the level of caffeine consumption, this may have affected the distribution of different age groups in the quartiles.
From the above results, it appeared that lifestyle factors such as smoking, alcohol consumption, and physical activity (MET levels) may influence the relationship between caffeine and FFMI. After adjusting for these confounders, the relationship between caffeine intake and FFMI was nuanced. Caffeine may increase alertness and physical endurance [3,4]. Given that caffeine can positively impact exercise performance to some extent, individuals engaging in higher levels of physical activity, such as long-distance runners, swimmers, etc. may tend to increase their caffeine intake to enhance energy levels and endurance. High levels of physical activity can promote muscle growth and increased muscle mass [44], so individuals who regularly engage in high-intensity physical activity may have higher levels of FFMI. Regarding the effects of drinking status, some studies suggest that drinkers may be more likely to consume caffeine and that caffeine induces alcohol intake [45–47]. This may be due to the tendency of drinkers to consume caffeinated beverages more frequently in social settings or because they have developed a higher tolerance to caffeine, requiring larger amounts to achieve the same effects. Moderate alcohol consumption may contribute to the maintenance or increase of muscle mass, potentially due to its effects on blood flow and hormone levels. However, excessive alcohol consumption is likely associated with muscle mass loss, which could result in lower FFMI levels [48,49]. Caffeine consumption is correlated with smoking, but whether this association is causal remains unclear [50–52]. Ware et al conducted a study examining the relationship between tobacco use and caffeine intake, and from their data smokers tended to consume less caffeine [53], but further MR analysis UK Biobank found no evidence of causal link. This may be because tobacco can be a substitute for caffeine in certain scenarios [54], so you don’t need to consume a lot of caffeine to get the same effect. Other studies have shown that tobacco users may consume more caffeine [55]. This may be because tobacco users have a higher tolerance to caffeine and need to consume more caffeine to get the same effect. In addition, tobacco use may be associated with health consciousness and lifestyle, which may also influence caffeine intake choices. In addition to these factors, race may also affect the relationship between FFMI and body composition [56] due to known differences muscle mass, and fat distribution [57]. Further, cultural variations influencing eating habits and physical activity levels may also impact FFMI and lead to racial differences [58]. When a single person becomes married, the different eating habits of the husband and wife may also affect each other.
An important strength of our study is the large sample size, which is representative of the US adult population. In addition, we adequately adjusted for a number of potential confounders to ensure an independent association between caffeine intake and FFMI. However, our findings should be interpreted with caution because the present study also has several limitations. Firstly, although this study used rigorous statistical methods to control for potential confounders, residual confounding, and measurement error may still exist. Secondly, the original database only quantified caffeine intake through coffee and did not take into account other sources such as caffeine tablets or energy drinks. Third, due to the loss of certain racial data during a specific time period, we were unable to include race as a covariate in further analysis. Future research should aim to address these limitations by incorporating a broader range of caffeine sources, ensuring more comprehensive data on racial and ethnic diversity, and employing additional methods to reduce residual confounding. To build on this, future work is needed to further elucidate the mechanistic pathways between caffeine intake and FFMI and to explore inter-individual differences in response to caffeine. Longitudinal studies and randomized controlled trials are essential for establishing causality and providing evidence-based recommendations regarding caffeine intake to optimize muscle health.
5. Conclusion
Currently, caffeine intake was mainly positively associated with FFMI, especially in women with above levels of FFMI. The findings contribute to our understanding of the role of caffeine in muscle growth and performance, particularly in diverse populations.
Funding Statement
The author(s) reported there is no funding associated with the work featured in this article.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Data availability statement
All data in the study are available at: https://www.cdc.gov/nchs/nhanes/index.htm.
References
- 1.Mahumud RA, Sahle BW, Owusu-Addo E, et al. Association of dietary intake, physical activity, and sedentary behaviours with overweight and obesity among 282,213 adolescents in 89 low and middle income to high-income countries. Int J Obes (Lond). 2021;45(11):2404–17. doi: 10.1038/s41366-021-00908-0 [DOI] [PubMed] [Google Scholar]
- 2.Lowery LM, Anderson DE, Scanlon KF, et al. International society of sports nutrition position stand: coffee and sports performance. J Int Soc Sports Nutr. 2023;20(1):2237952. doi: 10.1080/15502783.2023.2237952 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Kim H, Kang SH, Kim SH, et al. Drinking coffee enhances neurocognitive function by reorganizing brain functional connectivity. Sci Rep. 2021;11(1):14381. doi: 10.1038/s41598-021-93849-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.McLellan TM, Caldwell JA, Lieberman HR.. A review of caffeine’s effects on cognitive, physical and occupational performance. Neurosci Biobehav Rev. 2016;71(294):294–312. doi: 10.1016/j.neubiorev.2016.09.001 [DOI] [PubMed] [Google Scholar]
- 5.Liu P, Ma F, Lou H, et al. The utility of fat mass index vs. body mass index and percentage of body fat in the screening of metabolic syndrome. BMC Public Health. 2013;13(1). doi: 10.1186/1471-2458-13-629 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Ji W, Liu X, Liu P, et al. The efficacy of fat-free mass index and appendicular skeletal muscle mass index in cancer malnutrition: a propensity score match analysis. Front Nutr. 2023;10(1172610). doi: 10.3389/fnut.2023.1172610 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Lu Y, Shu H, Zheng Y, et al. Comparison of fat-free mass index and fat mass index in Chinese adults. Eur J Clin Nutr. 2012;66(9):1004–1007. doi: 10.1038/ejcn.2012.86 [DOI] [PubMed] [Google Scholar]
- 8.Grgic J, Trexler ET, Lazinica B, et al. Effects of caffeine intake on muscle strength and power: a systematic review and meta-analysis. J Int Soc Sports Nutr. 2018;15(1):11. doi: 10.1186/s12970-018-0216-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Fiani B, Zhu L, Musch BL, et al. The neurophysiology of caffeine as a central nervous system stimulant and the resultant effects on cognitive function. Cureus. 2021;13(5):e15032. doi: 10.7759/cureus.15032 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Gutiérrez-Hellín J, Aguilar-Navarro M, Ruiz-Moreno C, et al. Effect of caffeine intake on fat oxidation rate during exercise: is there a dose–response effect? Eur J Nutr. 2023;62(1):311–319. doi: 10.1007/s00394-022-02988-8 [DOI] [PubMed] [Google Scholar]
- 11.Ramírez-Maldonado M, Jurado-Fasoli L, Del Coso J, et al. Caffeine increases maximal fat oxidation during a graded exercise test: is there a diurnal variation? J Int Soc Sports Nutr. 2021;18(1):5. doi: 10.1186/s12970-020-00400-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Conger SA, Tuthill LM, Millard-Stafford ML. Does caffeine increase fat metabolism? A systematic review and meta-analysis. Int J Sport Nutr Exerc Metab. 2022;33(2):112–120. doi: 10.1123/ijsnem.2022-0131 [DOI] [PubMed] [Google Scholar]
- 13.Acheson KJ, Zahorska-Markiewicz B, Pittet P, et al. Caffeine and coffee: their influence on metabolic rate and substrate utilization in normal weight and obese individuals. Am J Clin Nutr. 1980;33(5):989–997. doi: 10.1093/ajcn/33.5.989 [DOI] [PubMed] [Google Scholar]
- 14.Zhang H, Speakman JR. The complexity of coffee and its impact on metabolism. J Endocrinol. 2024;262(3):e240075. doi: 10.1530/JOE-24-0075 [DOI] [PubMed] [Google Scholar]
- 15.Clark I, Landolt HP. Coffee, caffeine, and sleep: a systematic review of epidemiological studies and randomized controlled trials. Sleep Med Rev. 2017;31(70):70–78. doi: 10.1016/j.smrv.2016.01.006 [DOI] [PubMed] [Google Scholar]
- 16.Winston AP, Hardwick E, Jaberi N. Neuropsychiatric effects of caffeine. Adv Psychiatric Treat. 2005;11(6):432–439. doi: 10.1192/apt.11.6.432 [DOI] [Google Scholar]
- 17.Bchir F, Dogui M, Ben Fradj R, et al. Differences in pharmacokinetic and electroencephalographic responses to caffeine in sleep-sensitive and non-sensitive subjects. C R Biol. 2006;329(7):512–519. doi: 10.1016/j.crvi.2006.01.006 [DOI] [PubMed] [Google Scholar]
- 18.Reissig CJ, Strain EC, Griffiths RR. Caffeinated energy drinks—A growing problem. Drug Alcohol Depend. 2009;99(1–3):1–10. doi: 10.1016/j.drugalcdep.2008.08.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.van Dam RM, Hu FB, Willett WC, et al. Coffee, caffeine, and health. N Engl J Med. 2020;383(4):369–378. doi: 10.1056/NEJMra1816604 [DOI] [PubMed] [Google Scholar]
- 20.Walter K. Caffeine and health. JAMA. 2022;327(7):693. doi: 10.1001/jama.2021.21452 [DOI] [PubMed] [Google Scholar]
- 21.Flegal KM, Shepherd JA, Looker AC, et al. Comparisons of percentage body fat, body mass index, waist circumference, and waist-stature ratio in adults. Am J Clin Nutr. 2009;89(2):500–508. doi: 10.3945/ajcn.2008.26847 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Herrmann SD, Willis EA, Ainsworth BE, et al. Adult compendium of physical activities: a third update of the energy costs of human activities. J Sport Health Sci. 2024;13(1):6–12. doi: 10.1016/j.jshs.2023.10.010 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Jee HJ, Lee SG, Bormate KJ, et al. Effect of caffeine consumption on the risk for neurological and psychiatric disorders: sex differences in human. Nutrients. 2020;12(10):3080. doi: 10.3390/nu12103080 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Arnaud MJ. Pharmacokinetics and metabolism of natural methylxanthines in animal and man. Handb Exp Pharmacol. 2011200:33–91. doi: 10.1007/978-3-642-13443-2_3 [DOI] [PubMed] [Google Scholar]
- 25.Gunes A, Dahl ML. Variation in CYP1A2 activity and its clinical implications: influence of environmental factors and genetic polymorphisms. Pharmacogenomics. 2008;9(5):625–637. doi: 10.2217/14622416.9.5.625 [DOI] [PubMed] [Google Scholar]
- 26.Nehlig A, Alexander SPH. Interindividual differences in caffeine metabolism and factors driving caffeine consumption. Pharmacol Rev. 2018;70(2):384–411. doi: 10.1124/pr.117.014407 [DOI] [PubMed] [Google Scholar]
- 27.Kolahdouzan M, Hamadeh MJ. The neuroprotective effects of caffeine in neurodegenerative diseases. CNS Neurosci Ther. 2017;23(4):272–290. doi: 10.1111/cns.12684 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Blaak E. Gender differences in fat metabolism. Curr Opin Clin Nutr Metab Care. 2001;4(6):499–502. doi: 10.1097/00075197-200111000-00006 [DOI] [PubMed] [Google Scholar]
- 29.Heckman MA, Weil J, Gonzalez de Mejia E. Caffeine (1, 3, 7-trimethylxanthine) in foods: a comprehensive review on consumption, functionality, safety, and regulatory matters. J Food Sci. 2010;75(3):R77–87. doi: 10.1111/j.1750-3841.2010.01561.x [DOI] [PubMed] [Google Scholar]
- 30.Kamimori GH, Somani SM, Knowlton RG, et al. The effects of obesity and exercise on the pharmacokinetics of caffeine in lean and obese volunteers. Eur J Clin Pharmacol. 1987;31(5):595–600. doi: 10.1007/bf00606637 [DOI] [PubMed] [Google Scholar]
- 31.Guest NS, VanDusseldorp TA, Nelson MT, et al. International society of sports nutrition position stand: caffeine and exercise performance. J Int Soc Sports Nutr. 2021;18(1):1. doi: 10.1186/s12970-020-00383-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Caraco Y, Zylber-Katz E, Berry EM, et al. Caffeine pharmacokinetics in obesity and following significant weight reduction. Int J Obes Relat Metab Disord. 1995;19(4):234–239. [PubMed] [Google Scholar]
- 33.Cornelis MC, Kacprowski T, Menni C, et al. Genome-wide association study of caffeine metabolites provides new insights to caffeine metabolism and dietary caffeine-consumption behavior. Hum Mol Genet. 2016;25(24):5472–5482. doi: 10.1093/hmg/ddw334 [DOI] [PubMed] [Google Scholar]
- 34.Meena Harish Jain K, Hou HH, Siegel RA. An artificial gut/absorption simulator: description, modeling, and validation using caffeine. Aaps J. 2022;24(5):87. doi: 10.1208/s12248-022-00721-1 [DOI] [PubMed] [Google Scholar]
- 35.Arum O, Boparai RK, Saleh JK, et al. Specific suppression of insulin sensitivity in growth hormone receptor gene-disrupted (GHR-KO) mice attenuates phenotypic features of slow aging. Aging Cell. 2014;13(6):981–1000. doi: 10.1111/acel.12262 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Jura M, Kozak LP. Obesity and related consequences to ageing. Age (Dordr). 2016;38(1):23. doi: 10.1007/s11357-016-9884-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Waters DL, Ward AL, Villareal DT. Weight loss in obese adults 65 years and older: a review of the controversy. Exp Gerontol. 2013;48(10):1054–1061. doi: 10.1016/j.exger.2013.02.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Onder G, Pedone C, Landi F, et al. Adverse drug reactions as cause of hospital admissions: results from the Italian group of pharmacoepidemiology in the elderly (GIFA). J Am Geriatr Soc. 2002;50(12):1962–1968. doi: 10.1046/j.1532-5415.2002.50607.x [DOI] [PubMed] [Google Scholar]
- 39.Aymanns C, Keller F, Maus S, et al. Review on pharmacokinetics and pharmacodynamics and the aging kidney. Clin J Am Soc Nephrol. 2010;5(2):314–327. doi: 10.2215/CJN.03960609 [DOI] [PubMed] [Google Scholar]
- 40.Rochon PA, Gurwitz JH. The prescribing cascade revisited. Lancet. 2017;389(10081):1778–1780. doi: 10.1016/S0140-6736(17)31188-1 [DOI] [PubMed] [Google Scholar]
- 41.Wall BT, Gorissen SH, Pennings B, et al. Aging is accompanied by a blunted muscle protein synthetic response to protein ingestion. PLOS ONE. 2015;10(11):e0140903. doi: 10.1371/journal.pone.0140903 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Malinauskas BM, Aeby VG, Overton RF, et al. A survey of energy drink consumption patterns among college students. Nutr J. 2007;6(1):35. doi: 10.1186/1475-2891-6-35 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Svikis DS, Dillon PM, Meredith SE, et al. Coffee and energy drink use patterns in college freshmen: associations with adverse health behaviors and risk factors. BMC Public Health. 2022;22(1):594. doi: 10.1186/s12889-022-13012-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Miko H-C, Zillmann N, Ring-Dimitriou S, et al. Effects of physical activity on health. Gesundheitswes (Bundesverband der Arzte des Offentlichen Gesundheitsdienstes (Ger)). 2020;82(S03):S184–S195. doi: 10.1055/a-1217-0549 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Howland J, Rohsenow DJ, Arnedt JT, et al. The acute effects of caffeinated versus non-caffeinated alcoholic beverage on driving performance and attention/reaction time. Addiction. 2011;106(2):335–341. doi: 10.1111/j.1360-0443.2010.03219.x [DOI] [PubMed] [Google Scholar]
- 46.Okhuarobo A, Igbe I, Yahaya A, et al. Effect of caffeine on alcohol consumption and alcohol-induced conditioned place preference in rodents. J Basic Clin Physiol Pharmacol. 2018;30(1):19–28. doi: 10.1515/jbcpp-2018-0068 [DOI] [PubMed] [Google Scholar]
- 47.Marczinski CA, Fillmore MT, Henges AL, et al. Effects of energy drinks mixed with alcohol on information processing, motor coordination and subjective reports of intoxication. Exp Clin Psychopharmacol. 2012;20(2):129–138. doi: 10.1037/a0026136 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48.Skinner J, Shepstone L, Hickson M, et al. Alcohol consumption and measures of sarcopenic muscle risk: cross-sectional and prospective associations within the UK biobank study. Calcif Tissue Int. 2023;113(2):143–156. doi: 10.1007/s00223-023-01081-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49.Simon L, Bourgeois BL, Molina PE. Alcohol and skeletal muscle in health and disease. Alcohol Res. 2023;43(1):04. doi: 10.35946/arcr.v43.1.04 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 50.Freedman ND, Park Y, Abnet CC, et al. Association of coffee drinking with total and cause-specific mortality. N Engl J Med. 2012;366(20):1891–1904. doi: 10.1056/NEJMoa1112010 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 51.Swanson JA, Lee JW, Hopp JW. Caffeine and nicotine: a review of their joint use and possible interactive effects in tobacco withdrawal. Addict Behav. 1994;19(3):229–256. doi: 10.1016/0306-4603(94)90027-2 [DOI] [PubMed] [Google Scholar]
- 52.Hewlett P, Smith A. Correlates of daily caffeine consumption. Appetite. 2006;46(1):97–99. doi: 10.1016/j.appet.2005.10.004 [DOI] [PubMed] [Google Scholar]
- 53.Ware JJ, Tanner JA, Taylor AE, et al. Does coffee consumption impact on heaviness of smoking? Addiction. 2017;112(10):1842–1853. doi: 10.1111/add.13888 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 54.Alamneh AA, Endris BS, Gebreyesus SH, et al. Caffeine, alcohol, khat, and tobacco use during pregnancy in Butajira, South Central Ethiopia. PLOS ONE. 2020;15(5):e0232712. doi: 10.1371/journal.pone.0232712 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 55.Bjørngaard JH, Nordestgaard AT, Taylor AE, et al. Heavier smoking increases coffee consumption: findings from a Mendelian randomization analysis. Int J Epidemiol. 2017;46(6):1958–1967. doi: 10.1093/ije/dyx147 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 56.Gallagher D, Heymsfield SB, Heo M, et al. Healthy percentage body fat ranges: an approach for developing guidelines based on body mass index. Am J Clin Nutr. 2000;72(3):694–701. doi: 10.1093/ajcn/72.3.694 [DOI] [PubMed] [Google Scholar]
- 57.Wagner DR, Heyward VH. Measures of body composition in blacks and whites: a comparative review. Am J Clin Nutr. 2000;71(6):1392–1402. doi: 10.1093/ajcn/71.6.1392 [DOI] [PubMed] [Google Scholar]
- 58.Kumanyika SK. Environmental influences on childhood obesity: ethnic and cultural influences in context. Physiol Behav. 2008;94(1):61–70. doi: 10.1016/j.physbeh.2007.11.019 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
All data in the study are available at: https://www.cdc.gov/nchs/nhanes/index.htm.