Relationship between coffee concentration and bowel motility using bowel-sound-based stimulus-response plots in healthy individuals

Abstract

Coffee is consumed worldwide and exerts multifaceted physiological effects, yet evidence on its short-term impact on intestinal motility remains inconsistent. We investigated 40-min changes in motility after coffee prepared at three concentrations (weak, strong, extra strong) in 10 healthy young men using bowel-sound–based stimulus–response plots (BSSRPs) derived from BS segments per minute. Concentrations were created by precisely adjusting instant-coffee powder per 150 mL serving at 45 °C to control total dissolved solids. Motility increased versus baseline in both post-ingestion phases (Gastric 0–10 min; Intestinal 20–40 min). Planned pairwise comparisons did not detect differences among concentrations within phase, whereas exploratory BSSRPs suggested a concentration-ordered inverse pattern (WC ≥ SC ≥ Ex.SC) that persisted across 0–40 min. These findings indicate that coffee acutely modulates intestinal motility under fixed temperature/volume conditions, while any concentration dependence remains provisional. This work provides a basis for more systematic evaluation of coffee’s physiological effects; future studies should standardize composition, temperature, and volume, include a matched hot-water control, and be powered to test concentration effects and mechanisms directly.

Introduction

Coffee is an extensively consumed beverage worldwide. It is a complex mixture of approximately 1,000 physiologically active compounds. Previous studies have suggested that coffee possesses antioxidant, anti-inflammatory, and anticancer properties and that long-term consumption may contribute to the prevention of cardiovascular diseases and neurodegenerative disorders^1–3. By contrast, short-term coffee intake activates sympathetic nervous system activity by increasing blood pressure through peripheral vasoconstriction⁴. Furthermore, heart rate increases significantly immediately after coffee consumption, and this effect has been reported to persist for at least 30 min⁵. Transient increases in heart rate and cardiac output during the ingestion phase can augment postprandial intestinal blood flow (postprandial hyperemia) and may contribute to enhanced intestinal motility⁶.

Furthermore, interest in the effects of coffee on the digestive system, particularly on intestinal motility, has grown. Previous studies have found that caffeine and chlorogenic acid, components of coffee, stimulate intestinal peristalsis^7,8. Additionally, caffeine has been reported to stimulate directly intestinal smooth muscle and activate the enteric nervous system, thereby promoting intestinal peristalsis. Furthermore, coffee has been shown to induce approximately a 30% contraction of the gallbladder via stimulation of cholecystokinin secretion, and bile acids have been suggested to promote intestinal peristalsis through receptors such as TGR5⁹.

However, conflicting results regarding the effects of coffee on intestinal motility have been reported. For example, the beneficial effects of coffee on intestinal function recovery following gynecological surgery or caesarean sections have been confirmed in several randomized controlled trials^8,10,11. However, comparisons between caffeinated and decaffeinated coffee on intestinal function recovery after colon surgery did not confirm beneficial effects^12–15. These inconsistencies can be attributed to differences in the types of coffee used, coffee concentrations, postconsumption observation periods, and, possibly, the effects of caffeine on the autonomic nervous system.

In previous studies that investigated coffee and intestinal motility, various metrics, such as the time to first flatus and the time to first defecation, have been used as evaluation indicators^10–13,15. However, it is difficult to assess intestinal motility in detail using these indicators after coffee consumption. While some studies have used manometry to investigate intestinal motility in detail¹⁶, the method is invasive and poses challenges, such as the requirement of advanced techniques and high costs.

A noninvasive approach based on bowel sound (BS) analysis was developed. For example, the development of BS detection methods^17,18, evaluation of intestinal motility based on BSs during beverage consumption^19–22, evaluation of intestinal motility based on BSs during food intake²³, and the application of BS-based methods to intestinal diseases²⁴ have been reported. Compared with conventional invasive approaches, BS-based methods are cost-effective, simple to perform, less burdensome for subjects, and enable detailed evaluations of intestinal motility.

Recently, our research group developed a BS-based approach to evaluate intestinal motility in response to beverage stimulation²⁰. This approach can be used to assess intestinal motility by considering the influences of preconsumption intestinal motility on postconsumption motility responses.

The inconsistencies in the previous research results may have been caused by the possibility that motility differed based on the concentration of coffee used. The importance of this study lies in its ability to clarify the effects of different coffee concentrations on intestinal motility using a noninvasive and state-of-the-art approach, thereby addressing previously inconsistent results and debated findings. In particular, this study provides clues on how coffee consumption contributes to the improvement of digestive function by evaluating short-term responses and sustained changes in intestinal motility caused by different concentrations of coffee within the normal range using the Brewing Control Chart established by the Specialty Coffee Association of America (SCAA)^25,26.

This study investigated the short-term and sustained effects of different coffee concentrations on intestinal motility by evaluating the stimulus–motor responses of the intestine. To achieve this goal, a coffee-loading test was conducted on a group of healthy young men, and the effects of ingesting three different concentrations of coffee (weak coffee (WC), strong coffee (SC), and extra strong coffee (Ex. SC)) on intestinal motility were compared and examined noninvasively.

Methods

Coffee concentration measurements

The SCAA developed the SCAA Brewing Control Chart^25,26 to evaluate objectively the quality of brewed coffee. This chart enables an objective assessment of coffee using two indicators: total dissolved solids (TDS), which represents the concentration of inorganic salts and organic compounds dissolved in water, and extraction yield (EY), which indicates the extent to which coffee bean components have dissolved. Based on the SCAA Brewing Control Chart, this study focused on the TDS owing to its emphasis on coffee concentrations. We set TDS to 0.90% in WC, TDS to 1.60% in SC, and TDS to 2.10% in Ex. SC.

In the experiments of this study, we used “Chotto Zeitakuna Coffee Shop Special Blend” instant coffee from Ajinomoto AGF Inc. to provide coffee with stable TDS and EY. As the relationship between the amount of instant coffee powder of this product and TDS was unknown, we used PAL-COFFEE from Atago Co., Ltd. to estimate TDS, and an electronic scale (Scout Pro SP123FJP) manufactured by OHAUS to weigh the powder amount, thereby investigating the relationship between the two.

The TDS measurements were performed as follows. Before making the coffee, the mass of the instant coffee powder was measured and dissolved in a water volume of 150 mL to prepare the coffee. From the prepared coffee, one spoonful (equivalent to 1 mL) was obtained, and the concentration was measured using PAL-COFFEE. Based on the measured results, we interpreted the relationship between the powder amount and TDS and determined the amount of instant coffee powder used so that the coffee concentrations used in the experiment corresponded to WC, SC, and Ex. SC.

Subjects and participation in coffee intake tests

Participants were recruited from May 2023 to December 2024. All procedures were performed in accordance with the Declaration of Helsinki and approved by the Tokushima University Graduate School of Technology, Industrial and Social Sciences, Division of Science and Technology and Division of Bioscience and Bioindustry Research Ethics Committee (14011). Informed consent was obtained from all participants before their inclusion in the study.

The study included 10 healthy young men (mean age ± standard deviation: 20.5 ± 2.5 years, height: 173.0 ± 9.0 cm, weight: 62.0 ± 10.0 kg, body mass index: 21.2 ± 3.2). Once informed consent was obtained from all participants, it was confirmed that they did not have irritable bowel syndrome based on the Rome IV criteria.

The coffee provided (volume = 150 mL) was prepared at 45 °C, and adjusted to the three concentrations of WC, SC, and Ex. SC. The participants were required to fast for approximately 12 h starting on the day before the experiment until the time of measurement, and recordings were performed in the following morning. To enhance data reliability, three experiments were conducted for each concentration. Additionally, it has been suggested that caffeine in coffee may result in the development of tolerance through continuous oral intake over four consecutive days²⁷. Therefore, in this study, the experiments were limited to one per day and were conducted in this manner to avoid performing experiments on three consecutive days.

During the first trial of coffee consumption (at all three concentrations), the experiments were conducted thrice at each concentration only if the subject had consumed a small amount of coffee; subsequently, it was determined that it was safe and could be consumed without any problems. Regarding Ex. SC, one participant determined that they could not consume it, which resulted in 27 experiments at this concentration. In total, 30, 30, and 27 experiments were conducted for the WC, SC, and Ex. SC concentration cases, respectively, and the data were analyzed. Notably, this study employed a within-subject concentration–response design under fixed temperature (45 °C) and volume (150 mL) conditions. In principle, each participant completed nine trials (three concentrations × three replicates), except for the one individual noted above. To minimize participant burden, we did not include a temperature- and volume-matched water arm—which would have increased the total trial number by three—instead, we used the lowest-concentration coffee (WC) as an internal quasi-control.

The coffee provided to the participants was prepared immediately before consumption using an electronic kettle with temperature control (HAGGOGI GEK-1801) and its temperature was confirmed to be 45 °C using a compact thermographic camera (FLIR C3). During the experiment, the participants were placed in the supine position, and an electronic stethoscope (Cardioics E-Scope) was positioned 7 cm to the right of the navel and 2 cm inferiorly from the navel. The recordings were performed before and after coffee consumption. The electronic stethoscope was connected to an audio interface (Roland OCTA-CAPTURE), and the recorded data were saved on a personal computer at a sampling frequency of 44,100 Hz and a digital resolution of 16 bits. Additionally, the use of smartphones by the participants was permitted during the recordings. The recorded data used for the analysis contained 10 min of data to observe the resting state before coffee consumption, and 40 min of data to observe the effects after consumption.

Automatic BS detection method for coffee intake tests

Upon downsampling the recorded data to 4,000 Hz, an automatic BS detection method, which was developed by our research group to modify the feature set used in the BS automatic detection method²⁰, was employed. This method first divided the recording data into 64 ms segments with a segmentation shift width of 16 ms. Subsequently, 22-dimensional Mel-frequency cepstral coefficients (MFCC) features and 10-dimensional linear predictive cepstral coefficients (LPCC), along with ΔLPCC and ΔΔLPCC, were extracted from each segment. Using these features and an artificial neural network (ANN), we determined whether the BS was present within the segment. To train the ANN, we used the recording dataset¹⁹ from our previous study, supplemented with an additional 40-min recording dataset.

BS motility evaluation using BS-based stimulus–response plots (BSSRPSs)

We calculated the automatically detected number of BS segments per minute to evaluate the intestinal motility before and after coffee consumption. We were interested in the immediate stimulus response following coffee intake and the duration of this response. Therefore, we divided the 40-min period after coffee consumption into 10-min intervals and created a BSSRP based on the average number of BS segments per minute in each analysis interval (interval 1:0–10 min, interval 2:10–20 min, interval 3:20–30 min, interval 4:30–40 min). The BSSRPs were used to assess the intestinal stimulus response in each analysis interval as the baseline using the average number of BS segments per minute during the 10 min before consumption (before coffee intake, BCI)²⁰.

We defined the average number of BS segments per minute in a given analysis interval after coffee consumption as , ​and the average number of BS segments per minute during the 10 min before consumption as . Using ​​ and , we calculated the ratio using Eqs. (1),

1

We created BSSRPs with on the x-axis and the ratio values on the y-axis. To evaluate this correlation, we also created log–log BSSRPs using the natural logarithms of and Ratio.

Statistical analysis

Because the 10-min serial measurements were not independent, data were aggregated into two physiologically defined postprandial summary measures—Gastric (0–10 min) and Intestinal (20–40 min)—and analyzed them using a summary-measures approach²⁸.

The analysis set comprised participant-level means for each phase, calculated from three repeated measurements (observations with missing values for all three phases were excluded). To evaluate temporal changes in BS segments per min across phases and differences among coffee concentration groups within each phase, we performed planned pairwise comparisons based on a repeated-measures framework. Specifically, we tested (i) all pairwise phase comparisons (BCI vs. Gastric, BCI vs. Intestinal, Gastric vs. Intestinal) and (ii) all between-group comparisons within each phase (WC vs. SC, WC vs. Ex.SC, SC vs. Ex.SC). Bonferroni adjustments were applied to control the family-wise error rate. All tests were two-sided with α = 0.05, and adjusted P-values and 95% confidence intervals (CI) were computed.

We also evaluated temporal changes in the ratio across phases and differences among coffee-concentration groups within each phase using the same framework. For the ratio, phase comparisons were limited to Gastric vs. Intestinal, and the same between-group comparisons were assessed within each phase (WC vs. SC, WC vs. Ex.SC, SC vs. Ex.SC). Bonferroni adjustments were applied to control the family-wise error rate. All tests were two-sided with α = 0.05, and adjusted P-values and 95% CIs were reported.

Correlations were calculated using Pearson’s correlation coefficient, which represents a linear relationship. To assess the presence of a correlation, a t-test was conducted to test the null hypothesis that there was no correlation, with the significance level set at 0.05. If the obtained P-value was less than 0.05, the null hypothesis was rejected and a correlation was considered to exist.

Results

Relationship between the amounts of coffee powder and TDS

Following the method described in the previous section, the amounts of coffee powder used for WC, SC, and Ex. SC were calculated in this experiment. We plotted coffee powder quantity (g) on the x-axis and TDS (%) on the y-axis and fitted a linear regression. The results are shown in Fig. 1.

Fig. 1.

Relationship between coffee powder quantity and total dissolved solids (TDS).

The linearity of the relationship between the amount of coffee powder and the TDS was confirmed based on the results of Fig. 1. Based on the derived regression line, the amounts of coffee powder for the WC, SC, and Ex. SC were calculated. The coffee powder amounts for each concentration determined from the regression line are listed in Table 1.

Table 1.

Relationship between total dissolved solids (TDS) and coffee powder quantity at all tested coffee concentrations.

Label	TDS [%]	Coffee powder quantity [g]
Weak coffee (WC)	0.90	1.34
Strong coffee (SC)	1.60	2.02
Extra-strong coffee (Ex. SC)	2.10	2.51

In the experiment, coffee was prepared using the amounts of coffee powder listed in Table 1 at 45 °C at a volume of 150 mL and was consumed by the participants. Each 2 g serving contained approximately 62 mg of caffeine, as stated on the manufacturer’s product label.

Conventional bowel motility evaluation based on coffee intake tests

The performance evaluation of the BS automatic detection method was conducted using 5-fold cross-validation. The findings included sensitivity: 81.0 ± 1.41%, specificity: 97.3 ± 0.52%, Positive Predictive Value (PPV): 86.0 ± 1.82%, Negative Predictive Value (NPV): 96.1 ± 0.81%, accuracy: 94.5 ± 0.79%, and F1-score: 83.4 ± 0.71%.

Using this method, the average number of BS segments per minute was calculated from the pre-coffee consumption recording and from the recordings in each analysis after coffee consumption (interval 1:0–10 min; interval 2:10–20 min; interval 3:20–30 min; and interval 4:30–40 min) were calculated. The average numbers of BS segments per minute for each concentration are shown in Fig. 2A–C.

Fig. 2.

Number of bowel sound (BS) segments per minute before and after coffee intake in the following cases: (A) weak coffee (WC), (B) strong coffee (SC), and (C) extra-strong coffee (Ex. SC).

As shown in Fig. 2, during Interval 1, the number of BS segments per minute after ingestion increases compared with those before ingestion at all coffee concentrations, demonstrating the largest variation among all the analyzed intervals. Similarly, an upward trend is observed from Interval 2 onward, and increases in the numbers of BS segments per min after ingestion are observed at all concentrations. Planned pairwise comparisons indicated higher BS segments per minute in the Intestinal phase than in BCI (difference = 209.9 segments/min, 95% CI 32.6–387.1; Bonferroni-adjusted P = 0.016). The number of BS sounds in the Gastric phase also exceeded baseline (difference = 346.3, 95% CI 161.4–531.2; adjusted P = 1.7 × 10⁻⁴), whereas Gastric vs. Intestinal did not differ (difference = 136.4, 95% CI − 43.9 to 316.8; adjusted P = 0.192). Within each phase, no between-group differences across coffee concentrations were detected (all Bonferroni-adjusted P-values ≥ 0.677). To mitigate the influence of baseline heterogeneity on post-ingestion comparisons, we derived BSSRPs and conducted complementary analyses.

BSSRPs-based stimulus–motor response evaluation in coffee intake tests

In Interval 1, which exhibited the greatest variation across all analyzed intervals following coffee consumption, Fig. 3A, D,G displays the log–log BSSRPs obtained from the first ingestion experiment at each coffee concentration. Similarly, Fig. 3B, E,H presents the log–log.

Fig. 3.

Log–log BS-based stimulus-response plots (BSSRPs) obtained from Interval 1 in each experiment: (A) first WC intake test, (B) first and second WC intake tests, (C) first, second, and third WC intake tests, (D) first SC intake test,, (E) first and second SC intake tests, (F) first, second, and third SC intake tests, (G) first Ex. SC intake test, (H) first and second Ex. SC intake tests, and (I) first, second, and third Ex. SC intake tests.

BSSRPs from the first and second ingestion experiments, and Fig. 3C, F,I shows the log–log BSSRPs from all ingestion experiments up to the third intake tests.

Based on the findings in Fig. 3, it is suggested from the log–log BSSRPs obtained in the first intake experiment and those obtained in all intake experiments (up to the third) that there exists a linear relationship between log() and log(ratio) as a function of the number of plot points. Based on this observation, the correlation coefficient R and the P-value between the two were calculated, and a linear approximation of the plot points was performed. To achieve this, the coefficient of determination () adjusted for the degrees of freedom was calculated; this metric indicates the accuracy of the linear approximation. A value closer to one indicates a higher accuracy of the linear approximation outcome. The results obtained using the log–log BSSRPs from the first up to the third intake experiments are presented in Tables 2, 3 and 4.

Table 2.

, , and adjusted values obtained from the first coffee intake test.

Label			adjusted
WC	− 0.521	0.459	0.180
SC	− 0.265	0.459	− 0.046
Ex. SC	− 0.339	0.372	− 0.011

Table 3.

, , and adjusted obtained from the first and second coffee intake tests.

Label			adjusted
WC	− 0.456	0.043	0.164
SC	− 0.467	0.038	0.174
Ex. SC	− 0.437	0.067	0.140

Table 4.

, , and adjusted obtained from the first, second, and third coffee intake tests.  

Label			adjusted
WC	− 0.543	0.002	0.270
SC	− 0.641	0.000	0.390
Ex. SC	− 0.775	0.000	0.584

Comparison of Tables 2, 3 and 4, indicates that the adjusted values adjusted for the degrees-of-freedom tend to increase as a function of the number of trials across SC, and Ex. SC, whereas the WC shows a decrease followed by an increase. This suggests that the precision of the linear approximation improved with each additional intake experiment. The correlation coefficient and -values showed that regardless of the condition, WC, SC, or Ex. SC, the correlation coefficient tended to increase and the P-value tended to decrease as the number of intake experiments increased. The same trend was observed across the entire range of the analyses. These findings indicate that multiple trials clarify the relationship between log( and log(), enhancing the significance value. Based on this observation, additional analyses should be conducted using the log BSSRPs obtained from all intake experiments.

Evaluation of the time dependence of the correlation coefficient based on log–log BSSRPs

To investigate the effects of coffee intake over time, the correlation coefficient R was calculated for each analysis interval using the log–log BSSRPs obtained from all intake experiments up to the third for each coffee concentration. The results are plotted in Fig. 4.

Fig. 4.

Relationship between log–log BSSRP-based correlation coefficient and Intervals 1–4.

As shown in Fig. 4, during Interval 1 (0–10 min), an inverse correlation became more pronounced with increasing coffee concentration (WC → SC → Ex.SC). The same ordering was apparent at Interval 4, indicating that this pattern persisted over time. Thus, log–log BSSRPs depict the intestinal stimulus–response relationship and suggest that responsiveness aligns with stimulation intensity (indexed by coffee concentration), complementing the phase-wise results. Using the ratio as the outcome, the planned pairwise comparisons (Gastric vs. Intestinal) indicated higher ratios in the Gastric phase than in the Intestinal phase (Bonferroni-adjusted P = 0.00531). Within each phase, no between-group differences across coffee concentrations were detected (WC vs. SC, WC vs. Ex.SC, SC vs. Ex.SC; all Bonferroni-adjusted P > 0.05).

Discussion

In this study, we used log–log BSSRPs to examine the short-term and sustained effects of varying coffee concentrations on human intestinal motility. Log–log BSSRPs accounted for each participant’s pre-ingestion motility. A previous study²⁰ showed that log–log BSSRP correlations were higher after ingestion of carbonated water—presumed to provide a stronger stimulus—than after water, which is considered weaker. Under our experimental conditions (coffee at 45 °C, 150 mL), log–log BSSRPs displayed a concentration-ordered inverse association (WC ≥ SC ≥ Ex.SC) that persisted from 0 to 10 to 30–40 min. These visual patterns complement the planned pairwise comparisons, which showed that both Gastric (0–10 min) and Intestinal (20–40 min) phases exceeded baseline (BCI), whereas no between-concentration differences were detected within either phase.

Although beverages that provide a strong stimulus, such as carbonated water, have been reported to yield marked increases in correlation values, our BSSRPs suggested concentration-ordered responsiveness under the present protocol; however, even Ex.SC did not reach values comparable to those reported for carbonated water. A direct comparison of stimulatory potency with a prior study²⁰ is not appropriate because ingestion temperatures and volumes differed (water in the prior study: 10 °C, 200 mL; coffee in the present study: 45 °C, 150 mL), factors that can influence gastric accommodation and (in turn) intestinal motility.

In a previous study²⁹, intestinal motility was compared for 60 min after ingestion of hot water and hot coffee; it was shown that coffee promoted intestinal motility more than hot water. This indicates that the short-term intestinal-stimulatory effect of coffee may depend greatly on its composition and concentration. Among these, physiologically active components, such as caffeine and chlorogenic acid, are thought to play crucial roles. Caffeine antagonizes adenosine receptors and enhances nerve activity, thereby promoting intestinal smooth muscle contraction and peristalsis^7,8,30, which may be one of the factors that contribute to the increased intestinal motility caused by coffee. However, there are divergent views regarding heart rate variability after coffee consumption, and numerous factors have been considered, including the balance between the sympathetic and parasympathetic nervous systems and the differences in the composition of coffee components^20,31. Specifically, in situations where the parasympathetic nervous system is dominant, peristaltic intestinal movements are promoted^32,33. Therefore, if coffee intake enhances parasympathetic nerve activity under specific conditions, it may contribute to increased intestinal motility. Given that caffeine uptake is limited in the first 10 min and that the response magnitude does not increase with time, the effect is most plausibly a chemosensory/cephalic response—likely initiated by orosensory/bitter stimulation—rather than one driven by systemic uptake of coffee constituents, consistent with the pattern reported by McMullen et al.⁵.

Additionally, coffee is rich in polyphenols, and may thus alter postprandial hemodynamics through bitter-taste stimulation, and promote congestion responses within the intestinal tract³⁴. The congestion response in the intestine is believed to result from the complex interaction of various regulatory mechanisms, including the vascular smooth muscle and autonomic nervous system³⁵. It has been hypothesized that these multifaceted mechanisms act collectively to produce a concentration-dependent enhancement of intestinal motility by coffee. In the present study, this mechanistic interpretation remains provisional, consistent with the planned pairwise comparisons showing no significant between-concentration differences within either phase.

The instant coffee used in this experiment contained approximately 410 mg of polyphenols per serving (2.0 g) based on the ingredient label. By contrast, a typical cup of coffee (approximately 140 mL) contains approximately 200 mg of polyphenols³⁶.

The instant coffee used in this study had a higher polyphenol content than regular coffee, suggesting that the bitter polyphenol-derived components may have contributed considerably to the experimental results.

Improvement in intestinal motility is typically considered to contribute to a reduction in food retention time within the intestinal tract and to the maintenance and improvement of normal bowel rhythms^37–39. Therefore, by adjusting the concentration of coffee—that is, the content of physiologically active components such as caffeine and polyphenols—it is believed that additional improvement of bowel function can be achieved through intestinal motility. No differences among concentrations were detected; accordingly, any concentration dependence remains provisional and warrants confirmation in larger, matched-control studies.

In previous coffee-based studies, the amounts of instant coffee powder used were inconsistent; accordingly, the preparations varied considerably, including 111 mL of coffee (without specifying the powder amount)⁸, 150 mL of coffee^10–12, and 2.46 g of powder in 300 mL of coffee⁴⁰. One of the factors contributing to the variability in the interpretation of the results of these studies is the lack of uniformity in coffee concentrations (and in component amounts). Our results emphasize standardized dosing, temperature, and volume, and the value of planned pairwise comparisons alongside BSSRPs.

This study was associated with various limitations. First, the arabica/robusta composition of the instant coffee was undisclosed; caffeine doses were estimated from the manufacturer’s label, and the effects of other constituents could not be isolated. Second, a temperature- and volume-matched, hot noncarbonated water control, was not included, precluding isolation of water-only effects under identical conditions. Although prior work findings in responders reported that black coffee (regular/decaffeinated) increased the rectosigmoid motility index within 4 min and for at least 30 min, whereas hot water did not²⁹ (45 °C,200 mL), we did not include a water arm owing to participant burden. Future studies should incorporate a matched water control. Third, exposure to the highest concentration was not feasible for one participant, which may have limited power for between-concentration comparisons. Fourth, participants were healthy young men; thus, potential sex- or age-related differences were not assessed. Chronic constipation is more prevalent among the elderly because of physiological changes and a decrease in food intake with aging⁴¹. Additionally, reports are indicating that in young Japanese women, higher coffee consumption is associated with a reduced risk of constipation⁴². Therefore, future research that expands the range of subjects could provide more general guidelines on the effects of coffee consumption in improving digestive function. Furthermore, this study did not isolate or examine the individual effects of coffee components. In particular, research on the short-term effects of polyphenol intake on the body has not been exhaustive. Additional studies are required to elucidate the specific effects of these components on intestinal motility. Overall, BSSRPs here offer complementary, interval-wise visualization of stimulus–response patterns, whereas the planned pairwise comparisons provide the primary inferential results.

Conclusions

Using BS-based stimulus–response plots derived from BS segments per minute, together with planned pairwise comparisons, we observed acute increases versus baseline in both phases (Gastric 0–10 min and Intestinal 20–40 min) over 0–40 min. BSSRPs suggested concentration-ordered inverse patterns, but the planned pairwise comparisons did not detect differences among concentrations within phase. These findings support coffee’s short-term modulation of intestinal motility under fixed temperature/volume conditions (45 °C, 150 mL), and provide a basis for more systematic and precise evaluations of coffee’s effects on physiological functions. Future studies should standardize composition, temperature, and volume, include a matched hot-water control, and be powered to test concentration effects and mechanisms directly.

Author contributions

T.M., T.H. and T.E. conceptualized the study, T.H., T.E. and T.M. developed the methodology, and conducted the investigation. T.M., T.H. and K. T. performed the data analysis, T.M. and T.E. wrote the original draft. M. M. and J. I. acted as medical advisors and evaluated the study. T.E. is responsible for funding, project supervision, and T.E. and J.I. reviewed and edited the manuscript.

Funding

This work was supported by a specific grant from JSPS Grant-in-Aid for Scientific Research: Scientific Research (C): KAKEN 20K12755.

Data availability

The data analyzed in this study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

This study was approved by the Research Ethics Committee (14011) of Tokushima University Graduate School of Technology, Industrial and Social Sciences, Division of Science and Technology, and the Division of Bioscience and Bioindustry. All participants provided written informed consent prior to their participation.

Consent for publication

All authors have approved the publication.

Competing interests

The authors declare no competing interests.