Abstract
Background
Depression and gastrointestinal disease are prevalent conditions that often coexist, significantly impacting daily life and productivity. Recent studies suggest a potential link between the intake of dietary live microbe and the alleviation of depressive symptoms. However, the relationship between live microbe consumption and depressive symptoms in patients suffering from gastrointestinal diseases remains unexplored.
Methods
This study included participants with gastrointestinal diseases from the National Health and Nutrition Examination Survey (NHANES) spanning from 2005 to 2018. We utilized weighted multivariate logistic regression, subgroup analyses, and restricted cubic spline (RCS) analyses to investigate the association between live microbe consumption and depression. Additionally, the eXtreme Gradient Boosting (XGBoost) algorithm was implemented to develop a predictive model for depression based on individual characteristics.
Results
Of the 2,195 individuals, 472 (21.5%) exhibited depressive symptoms (PHQ-9 ≥ 10). Our findings indicate an inverse relationship between the live microbe intake and the incidence of depression in individuals with gastrointestinal diseases. In the most comprehensively adjusted model, patients with the highest level of microbe intake exhibited a 66.1% or 52.9% reduced risk of depressive symptoms compared to those with the minimal intake. An L-shaped dose-response relationship was observed in the RCS analysis (non-linear P = 4e-04). The XGBoost model demonstrated effective prediction capabilities for depression, with an area under the curve (AUC) of 0.897 (95% CI: 0.869–0.925).
Conclusions
This study provides evidence of an inverse, non-linear association between dietary live microbe and depression in individuals with gastrointestinal diseases, suggesting that higher intake levels may offer protective effects against depressive symptoms.
Supplementary Information
The online version contains supplementary material available at 10.1186/s12986-025-00976-3.
Keywords: Dietary live microbe, Gastrointestinal diseases, Depression, NHANES, Machine learning
Background
Globally, gastrointestinal diseases rank highly as major health concerns, significantly impacting national healthcare expenditures and diminishing health-related quality of life [1]. In the United States, annual healthcare expenditures for gastrointestinal diseases reached $111.8 billion in 2021 [2]. Globally, functional gastrointestinal disorders affect over 40% of the population, with higher prevalence among women, and are associated with lower quality of life and increased healthcare utilization [3]. Therefore, tackling and managing gastrointestinal diseases represents a major lobal challenge.
Depressive symptoms, which are prevalent in mental health contexts, frequently accompany gastrointestinal diseases. Research indicates that the incidence of depression is higher among those with gastrointestinal diseases than in the general population [4], and their depressive symptoms are more severe [5]. For example, one study reported that the prevalence of depressive symptoms was 49% among individuals with inflammatory bowel disease, compared with 23% in those without the condition [6]. Similarly, the incidence of gastrointestinal symptoms is nearly twice as high in depressed patients (54%) compared to non-depressed individuals (29%) [7]. The coexistence of gastrointestinal diseases and depression seriously impairs patients’ ability to work and carry out daily activities.
Diet quality and healthy dietary intake have been linked to a reduced risk of depressive symptoms [8, 9]. An increasing body of research has contributed to an emerging understanding of the beneficial effects of microbial dietary intake on human health [10–12]. Along with daily dietary intake, live microbes are transported to the intestines, where they collaborate with the immune system to enhance intestinal function and prevent the development of chronic diseases [13]. Probiotics have demonstrated efficacy in the treatment or prevention of various conditions such as acute gastroenteritis in children [14], antibiotic-associated diarrhoea [15], irritable bowel syndrome [16], and colorectal neoplasia [17]. Several studies have investigated the effects of probiotics on depressive symptoms, with many showing positive results in mood, anxiety, and cognition improvement [18]. Additionally, prebiotics and postbiotics have demonstrated promise in mitigating depression symptoms in both preclinical and clinical trials [19]. However, the relationship between dietary live microbe consumption and the risk of depression in individuals with gastrointestinal diseases remains unclear.
We hypothesize that the consumption of dietary live microbe is associated with a reduced frequency of depressive symptoms among patients with gastrointestinal diseases. The aim of this study was to investigate this association using data from the National Health and Nutrition Examination Survey (NHANES) from 2005 to 2018.
Methods and materials
Study population
To assess the nutritional and health status of the U.S. population, the National Center for Health Statistics (NCHS) and the Centers for Disease Control and Prevention conducted the NHANES study using a complex, multi-stage probability sampling design. Because only participants in NHANES 2005–2018 received the Patient Health Questionnaire-9 (PHQ-9) survey (a key instrument for assessing mental health and specifically depression), our analysis included seven independent cross-sectional waves of NHANES (2005–2006, 2007–2008, 2009–2010, 2011–2012, 2013–2014, 2015–2016, and 2017–2018). We systematically collected data on demographics, physical examinations, laboratory tests, and questionnaires. Participants were included if they reported stomach or intestinal illness with vomiting or diarrhoea in the past 30 days (n = 4,967). Exclusion criteria were as follows: (1) having incomplete PHQ-9 data (n = 2,145), or (2) missing complete covariate data (n = 627). As shown in Fig. 1, a total of 2,195 individuals were included in the final analysis.
Fig. 1.
Flow chart of the participant selection
Dietary live microbe consumption category
Dietary intake data were obtained through in-person interviews conducted by trained personnel using a 24-hour dietary recall method to document all food and beverage consumption. The NCHS linked these dietary records with the US Department of Agriculture (USDA) Food and Nutrient Database to estimate nutrient and energy intake. The amount of live microbe in 9,388 food items across 48 subcategories in the NHANES database was determined using a method developed by Marco et al. [20]. Live microbial content in foods was classified by four experts in the field into three groups: low (Lo; <104 CFU/g), medium (Med; 104–107 CFU/g), and high (Hi; >107 CFU/g). The specialists relied on scholarly sources, credible assessments, and established impacts of food treatments (including pasteurization) on microbial viability to assign the scores [21–23]. Conflicts were addressed through discussions among team members and with Fred Breidt of the USDA Research Service, a microbiologist, via external collaboration.
Participants were divided into different categories based on their consumption levels of foods containing live microbe. The grouping was initially done by classifying individuals into three distinct categories depending on their general consumption of microbe-rich foods: Low (all foods consumed were Lo); Medium (any foods consumed were Med but not Hi); and High (any foods consumed were Hi). Additionally, a separate category, termed “MedHi”, was established for those consuming either “Med” or “Hi” microbe content foods, but not “Lo”. Consequently, the participants were also categorized into three distinct groups based on the MedHi consumption to quantify the ingestion of live microbe. These were labelled as G1, consisting of individuals who did not consume any “MedHi” microbe content foods; G2, comprised of individuals who consumed “MedHi” foods in amounts greater than zero but less than the median consumption level of these foods; and G3 consisting of individuals whose consumption of “MedHi” foods was above the median level.
Assessment of depression symptoms
Depressive symptoms were assessed using the PHQ-9, a self-administered questionnaire covering nine core criteria for depressive disorders [24]. Each participant was queried to evaluate the occurrence of their symptoms for the two weeks prior, with the cumulative score for each individual ranging from 0 to 27. The severity of depressive symptoms was categorized into five levels: “none or minimal” (0–4), “mild” (5–9), “moderate” (10–14), “moderately severe” (15–19), and “severe” (20–27). Consistent with previous studies, a cutoff score of ≥ 10 was used to define the presence of depressive symptoms [25, 26].
Covariate assessment
This investigation included a variety of covariates, focusing on socio-demographic characteristics and biochemical blood markers. Considering known or hypothesized influences on depressive symptoms, our study gathered data on various potential confounding factors. These included participants’ age, gender, ethnic background (White, Black, Mexican American, and Other), education level (less than high school or equivalent, high school or equivalent, and college or above), the poverty-income ratio (PIR), and body mass index (BMI). Additionally, lifestyle factors such as smoking habits (categorized as never, former, or now) and alcohol consumption (never, former, mild-to-moderate, and heavy drinking) were assessed. Health conditions like hypertension, arteriosclerotic cardiovascular disease (ASCVD), hyperlipidemia, and diabetes mellitus (DM) were also considered as the variables potentially impacting depressive symptoms [27, 28].
Construction of the machine learning model
The dataset was randomly split into a training set (70%) and a test set (30%). We implemented an eXtreme Gradient Boosting (XGBoost) model to predict depression based on individual characteristics. The interpretation of the prediction model was performed by Shapley additive explanations (SHAP) methods, which is a unified approach to accurately calculate the contribution and influence of each variable on the predictions [29].
Statistical analyses
Subsample (mobile examination centres, MEC) weight was used to represent the complex survey design as suggested in the NHANES analysis guides (https://wwwn.cdc.gov/nchs/nhanes/tutorials/Module3.aspx). Continuous variables were presented as mean ± standard error (SE), and categorical variables were shown in terms of count and weighted percentages. The t-test (continuous variables) or chi-squared test (categorical variables) were used to identify the significant differences in population characteristics.
For evaluating the effect of dietary intake of live microbe on depression symptoms, weighted multivariate logistic regression analysis was employed. The crude model was not adjusted for confounders. Model 1 included adjustments for age and gender, while Model 2 added further adjustments for race, education, PIR, BMI, marital, smoking, and drinking status. Model 3 expanded on these adjustments by including hypertension, ASCVD, DM, and hyperlipidemia. Subgroup analyses and interaction tests were employed to investigate the interaction effect modification of depression symptoms by varying dietary living microbe intake and the subgroup stratified by various population features. Moreover, a restricted cubic spline (RCS) analysis was used to investigate the non-linear relationship between live microbe intake and depressive symptoms. All statistical analyses were conducted using R software, with P-values < 0.05 considered statistically significant.
Results
Baseline characteristics of individuals with gastrointestinal diseases
Table 1 presents the weighted baseline characteristics of individuals included in the study, categorized by the presence or absence of depressive symptoms. A total of 2,195 participants were analyzed, of whom 68.88% were female, and 21.5% (n = 472) were identified as having depressive symptoms. Depressive symptoms were more prevalent among females, unmarried individuals, those with a college education or higher, participants with lower PIR, obese individuals, smokers, and those who consume alcohol. Additionally, a higher prevalence of comorbid conditions such as hypertension, ASCVD, DM, and hyperlipidemia was observed in the depression group. Notably, participants with depressive symptoms had significantly lower levels of dietary live microbe intake (P = 0.0001).
Table 1.
Weighted baseline characteristics of 2,195 participants based on depression
| Variable | Total | Depressive symptoms | P value | |
|---|---|---|---|---|
| No | Yes | |||
| Age (years) | 46.548(0.479) | 46.569(0.589) | 46.457(0.916) | 0.924 |
| Sex | < 0.0001* | |||
| Female | 1289(58.141) | 972(55.640) | 317(68.883) | |
| Male | 906(41.859) | 751(44.360) | 155(31.117) | |
| Ethnicity | 0.084 | |||
| White | 1017(69.202) | 806(70.297) | 211(64.501) | |
| Mexican American | 352(8.234) | 268(7.843) | 84(9.914) | |
| Black | 442(10.853) | 343(10.265) | 99(13.378) | |
| Other | 384(11.710) | 306(11.595) | 78(12.206) | |
| Marital status | < 0.0001* | |||
| Married | 1025(50.959) | 864(54.140) | 161(37.292) | |
| Never married | 409(17.639) | 311(17.364) | 98(18.822) | |
| Living with partner | 190(8.647) | 145(8.491) | 45(9.316) | |
| Others | 571(22.755) | 403(20.005) | 168(34.570) | |
| Education | < 0.0001* | |||
| College or above | 1141(58.955) | 946(61.564) | 195(47.752) | |
| High school or equivalent | 523(25.149) | 418(25.301) | 105(24.498) | |
| Less than High school or equivalent | 531(15.895) | 359(13.135) | 172(27.750) | |
| PIR | < 0.0001* | |||
| 0-1.3 | 837(26.331) | 578(22.532) | 259(42.648) | |
| 1.3–3.5 | 807(37.243) | 652(37.520) | 155(36.053) | |
| >3.5 | 551(36.426) | 493(39.948) | 58(21.299) | |
| BMI | 0.005* | |||
| < 18.5 | 30(1.282) | 27(1.487) | 3(0.398) | |
| 18.5–25 | 553(26.413) | 452(27.186) | 101(23.091) | |
| 25–30 | 643(29.779) | 529(30.912) | 114(24.914) | |
| ≥ 30 | 969(42.526) | 715(40.414) | 254(51.597) | |
| Drinking status | < 0.001* | |||
| Never | 264(9.159) | 224(9.932) | 40(5.839) | |
| Former | 365(13.895) | 267(13.403) | 98(16.009) | |
| Mild to moderate | 1027(52.322) | 843(53.892) | 184(45.580) | |
| Heavy | 539(24.623) | 389(22.773) | 150(32.572) | |
| Smoking status | < 0.0001* | |||
| Never | 1044(47.047) | 877(50.222) | 167(33.413) | |
| Former | 545(25.552) | 442(26.304) | 103(22.323) | |
| Now | 606(27.400) | 404(23.474) | 202(44.265) | |
| Hypertension | < 0.001* | |||
| No | 1125(55.759) | 928(58.032) | 197(45.994) | |
| Yes | 1070(44.241) | 795(41.968) | 275(54.006) | |
| Hyperlipidemia | 0.017* | |||
| No | 597(27.888) | 488(29.299) | 109(21.827) | |
| Yes | 1598(72.112) | 1235(70.701) | 363(78.173) | |
| ASCVD | < 0.0001* | |||
| No | 1890(88.365) | 1523(90.152) | 367(80.691) | |
| Yes | 305(11.635) | 200(9.848) | 105(19.309) | |
| DM | < 0.001* | |||
| No | 1698(82.241) | 1364(84.084) | 334(74.326) | |
| Yes | 497(17.759) | 359(15.916) | 138(25.674) | |
| Live microbe intake | < 0.0001* | |||
| Low | 894(37.545) | 655(34.894) | 239(48.933) | |
| Medium | 838(37.429) | 684(38.891) | 154(31.148) | |
| High | 463(25.026) | 384(26.215) | 79(19.919) | |
| Live microbe intake (G) | < 0.0001* | |||
| G1 | 894(37.545) | 655(34.894) | 239(48.933) | |
| G2 | 650(29.902) | 519(29.843) | 131(30.153) | |
| G3 | 651(32.553) | 549(35.263) | 102(20.914) | |
| MediHi | 104.575(5.128) | 112.346(5.739) | 71.195(8.641) | < 0.0001* |
Mean (SE) for continuous variables; number (%) for categorical variables. Differences between groups were analyzed by nonparametric t-test or χ2 test
Abbreviations: ASCVD, arteriosclerotic cardiovascular disease; BMI, body mass index; DM, diabetes mellitus; PIR, poverty-income ratio
Characteristics by dietary live microbe groups
Table 2 shows the characteristics of participants across different levels of dietary live microbe intake. According to both classification methods, variables including race, education level, PIR, smoking status, and depression status differed significantly across intake groups. However, no significant differences were found in terms of age, sex, marital status, alcohol consumption, hypertension, ASCVD, DM, or hyperlipidemia.
Table 2.
Weighted baseline characteristics of 2,195 participants based on dietary live microbe intake
| Variable | Live microbe intake | P value | Live microbe intake | P value | ||||
|---|---|---|---|---|---|---|---|---|
| Low | Medium | High | G1 | G2 | G3 | |||
| Age (years) | 45.714(0.535) | 47.621(0.729) | 46.194(1.036) | 0.113 | 45.71(0.54) | 46.72(0.84) | 47.35(0.91) | 0.28 |
| Sex | 0.167 | 0.16 | ||||||
| Female | 489(54.885) | 513(61.260) | 287(58.361) | 489(54.88) | 397(60.52) | 403(59.71) | ||
| Male | 405(45.115) | 325(38.740) | 176(41.639) | 405(45.12) | 253(39.48) | 248(40.29) | ||
| Ethnicity | < 0.0001* | < 0.001* | ||||||
| White | 400(65.690) | 367(67.870) | 250(76.465) | 400(65.69) | 301(69.16) | 316(73.29) | ||
| Mexican American | 125(7.902) | 165(10.232) | 62(5.745) | 125(7.90) | 119(9.29) | 108(7.65) | ||
| Black | 221(14.546) | 161(10.577) | 60(5.728) | 221(14.55) | 123(10.07) | 98(7.31) | ||
| Other | 148(11.862) | 145(11.322) | 91(12.063) | 148(11.86) | 107(11.48) | 129(11.75) | ||
| Marital status | 0.266 | 0.08 | ||||||
| Married | 370(47.424) | 421(52.322) | 234(54.222) | 370(47.42) | 312(51.26) | 343(54.76) | ||
| Never married | 187(20.030) | 140(16.712) | 82(15.440) | 187(20.03) | 117(15.68) | 105(16.69) | ||
| Living with partner | 77(7.750) | 71(9.312) | 42(8.996) | 77(7.75) | 64(11.10) | 49(7.43) | ||
| Others | 260(24.796) | 206(21.654) | 105(21.342) | 260(24.80) | 157(21.97) | 154(21.13) | ||
| Education | < 0.0001* | < 0.0001* | ||||||
| College or above | 409(50.482) | 447(63.022) | 285(65.586) | 409(50.48) | 347(59.49) | 385(68.24) | ||
| High school or equivalent | 251(31.373) | 177(20.654) | 95(22.535) | 251(31.37) | 131(22.54) | 141(20.37) | ||
| Less than High school or equivalent | 234(18.145) | 214(16.324) | 83(11.878) | 234(18.15) | 172(17.97) | 125(11.40) | ||
| PIR | < 0.0001* | < 0.0001* | ||||||
| 0-1.3 | 396(33.085) | 300(23.328) | 141(20.690) | 396(33.08) | 233(23.22) | 208(21.40) | ||
| 1.3–3.5 | 332(37.844) | 310(37.603) | 165(35.802) | 332(37.84) | 235(39.17) | 240(34.78) | ||
| >3.5 | 166(29.071) | 228(39.069) | 157(43.509) | 166(29.07) | 182(37.61) | 203(43.82) | ||
| BMI | 0.028* | 0.06 | ||||||
| < 18.5 | 12(0.783) | 11(1.322) | 7(1.970) | 12(0.78) | 7(1.18) | 11(1.95) | ||
| 18.5–25 | 212(23.701) | 204(26.274) | 137(30.691) | 212(23.70) | 166(27.79) | 175(28.28) | ||
| 25–30 | 254(27.314) | 253(30.999) | 136(31.654) | 254(27.31) | 192(30.22) | 197(32.21) | ||
| ≥ 30 | 416(48.202) | 370(41.405) | 183(35.686) | 416(48.20) | 285(40.81) | 268(37.56) | ||
| Drinking status | 0.062 | 0.004* | ||||||
| Never | 105(8.863) | 114(10.676) | 45(7.335) | 105(8.86) | 82(10.65) | 77(8.13) | ||
| Former | 160(15.304) | 121(11.656) | 84(15.131) | 160(15.30) | 86(11.04) | 119(14.90) | ||
| Mild to moderate | 386(47.729) | 409(55.038) | 232(55.150) | 386(47.73) | 313(51.75) | 328(58.15) | ||
| Heavy | 243(28.104) | 194(22.629) | 102(22.384) | 243(28.10) | 169(26.56) | 127(18.83) | ||
| Smoking status | < 0.001* | < 0.0001* | ||||||
| Never | 383(40.633) | 435(51.668) | 226(49.759) | 383(40.63) | 335(51.19) | 326(50.64) | ||
| Former | 204(24.487) | 213(25.449) | 128(27.306) | 204(24.49) | 151(23.47) | 190(28.70) | ||
| Now | 307(34.880) | 190(22.883) | 109(22.935) | 307(34.88) | 164(25.35) | 135(20.66) | ||
| Hypertension | 0.169 | 0.1 | ||||||
| No | 445(53.554) | 424(55.198) | 256(59.907) | 445(53.55) | 324(53.82) | 356(60.09) | ||
| Yes | 449(46.446) | 414(44.802) | 207(40.093) | 449(46.45) | 326(46.18) | 295(39.91) | ||
| Hyperlipidemia | 0.309 | 0.43 | ||||||
| No | 236(25.805) | 217(27.922) | 144(30.963) | 236(25.80) | 178(28.84) | 183(29.41) | ||
| Yes | 658(74.195) | 621(72.078) | 319(69.037) | 658(74.20) | 472(71.16) | 468(70.59) | ||
| ASCVD | 0.69 | 0.71 | ||||||
| No | 755(87.433) | 733(89.191) | 402(88.528) | 755(87.43) | 567(89.00) | 568(88.86) | ||
| Yes | 139(12.567) | 105(10.809) | 61(11.472) | 139(12.57) | 83(11.00) | 83(11.14) | ||
| DM | 0.207 | 0.21 | ||||||
| No | 681(79.906) | 650(83.328) | 367(84.119) | 681(79.91) | 518(84.39) | 499(82.96) | ||
| Yes | 213(20.094) | 188(16.672) | 96(15.881) | 213(20.09) | 132(15.61) | 152(17.04) | ||
| Depression | < 0.0001* | < 0.0001* | ||||||
| No | 655(75.388) | 684(84.285) | 384(84.970) | 655(75.39) | 519(80.96) | 549(87.87) | ||
| Yes | 239(24.612) | 154(15.715) | 79(15.030) | 239(24.61) | 131(19.04) | 102(12.13) | ||
Mean (SE) for continuous variables; number (%) for categorical variables. Differences between groups were analyzed by nonparametric t-test or χ2 test
Abbreviations: ASCVD, arteriosclerotic cardiovascular disease; BMI, body mass index; DM, diabetes mellitus; PIR, poverty-income ratio
Association between dietary live microbe intake and depressive symptoms
The univariate and multivariate logistic regression models were used to assess the association between dietary live microbe intake and depressive symptoms. In the univariate logistic regression analysis (Supplementary Table 1), it was found that individuals with higher dietary live microbe consumption had a significantly lower risk of depressive symptoms compared to those with a lower intake (OR = 0.542, 95% CI, 0.382–0.769, P < 0.001). Similarly, individuals in the highest consumption group (G3) had a lower likelihood of depressive symptoms compared to the lowest (G1) group (OR = 0.423, 95% CI, 0.298–0.601, P < 0.0001). These associations remained robust in the multivariate models (Table 3). When considering dietary live microbe consumption as a categorical variable, individuals with medium to high intake levels showed a notable decrease in the prevalence of depression symptoms. Similarly, the G3 group showed a significant lower risk of depressive symptoms compared to the G1 group in three models. Additionally, when assessing dietary live microbe intake (MedHi) as a continuous variable, the increase in MedHi consumption correlated with consistent reduction in the risk of depression symptoms prevalence by 0.2%, 0.1%, and 0.1% in the respective models.
Table 3.
Association between different live microbe intake group and depression among 2,195 participants
| Character | Crude model | Model 1 | Model 2 | Model 3 | ||||
|---|---|---|---|---|---|---|---|---|
| OR (95% CI) | P value | OR (95% CI) | P value | OR (95% CI) | P value | OR (95% CI) | P value | |
| Low | Ref | Ref | Ref | Ref | ||||
| Medium | 0.571(0.440,0.742) | < 0.0001* | 0.547(0.425,0.704) | < 0.0001* | 0.663(0.501, 0.877) | 0.005* | 0.670(0.504, 0.890) | 0.006* |
| High | 0.542(0.382,0.769) | < 0.001* | 0.527(0.371,0.748) | < 0.001* | 0.670(0.459, 0.977) | 0.038* | 0.661(0.456, 0.959) | 0.03* |
| P for trend | < 0.001* | < 0.0001* | 0.017* | 0.014* | ||||
| G1 | Ref | Ref | Ref | Ref | ||||
| G2 | 0.720(0.549,0.946) | 0.019* | 0.694(0.531,0.908) | 0.008* | 0.810(0.599, 1.096) | 0.169 | 0.814(0.602, 1.099) | 0.177 |
| G3 | 0.423(0.298,0.601) | < 0.0001* | 0.407(0.289,0.574) | < 0.0001* | 0.530(0.369, 0.762) | < 0.001* | 0.529(0.365, 0.765) | < 0.001* |
| P for trend | < 0.0001* | < 0.0001* | < 0.001* | < 0.001* | ||||
| MediHi | 0.998(0.996,0.999) | 0.004* | 0.998(0.996,0.999) | 0.003* | 0.999(0.997,1.000) | 0.043* | 0.999(0.997,1.000) | 0.049* |
Crude Model was unadjusted; Model 1 was adjusted for age and gender; Model 2 was adjusted as for Model 1 plus race, education level, PIR, BMI, marital status, smoking status, and drinking status. Model 3 further added hypertension, ASCVD, DM, and hyperlipidemia
Abbreviations: ASCVD, arteriosclerotic cardiovascular disease; BMI, body mass index; CI, confidence interval; DM, diabetes mellitus; OR, odds ratio; PIR, poverty-income ratio; Ref, reference
Weighted multivariate logistic regression analysis was performed to examine the relationships between live microbe intake group and depression
Subgroup and interaction analyses
Additionally, we conducted stratified and interaction analyses to further investigate the relationship between dietary live microbe intake and depressive symptoms. With regard to the two methods of categorizing live microbe intake, the results indicated no significant differences for all predefined factors except for gender (Table 4). It was observed that, within groups with equivalent dietary live microbe consumption, men exhibited a more pronounced reduction in depressive symptom risk compared to women.
Table 4.
Subgroup analyses on the association between live microbe intake group and depression among 2,195 participants
| Character | Medium | P | High | P | P for interaction | G2 | P | G3 | P | P for interaction |
|---|---|---|---|---|---|---|---|---|---|---|
| Age | 0.586 | 0.689 | ||||||||
| < 50 | 0.519(0.354,0.760) | < 0.001* | 0.551(0.336,0.903) | 0.019* | 0.672(0.458,0.987) | 0.043* | 0.409(0.241,0.694) | 0.001* | ||
| ≥ 50 | 0.724(0.474,1.105) | 0.132 | 0.615(0.355,1.064) | 0.081 | 0.888(0.554,1.425) | 0.621 | 0.505(0.307,0.833) | 0.008* | ||
| Sex | 0.038* | 0.021* | ||||||||
| Female | 0.720(0.524,0.990) | 0.043* | 0.610(0.387,0.962) | 0.034* | 0.823(0.584,1.161) | 0.265 | 0.549(0.361,0.834) | 0.005* | ||
| Male | 0.302(0.178,0.513) | < 0.0001* | 0.446(0.255,0.779) | 0.005* | 0.540(0.334,0.873) | 0.012* | 0.214(0.121,0.378) | < 0.0001* | ||
| Ethnicity | 0.036* | 0.503 | ||||||||
| White | 0.453(0.301,0.680) | < 0.001* | 0.578(0.364,0.918) | 0.021* | 0.669(0.445,1.004) | 0.052 | 0.374(0.228,0.612) | < 0.001* | ||
| Mexican American | 0.684(0.385,1.216) | 0.187 | 0.311(0.111,0.870) | 0.028* | 0.628(0.324,1.215) | 0.159 | 0.514(0.273,0.966) | 0.04* | ||
| Black | 0.999(0.571,1.748) | 0.997 | 1.029(0.469,2.258) | 0.943 | 1.217(0.613,2.416) | 0.567 | 0.776(0.399,1.509) | 0.447 | ||
| Other | 0.895(0.443,1.807) | 0.753 | 0.386(0.163,0.913) | 0.031* | 0.814(0.380,1.742) | 0.591 | 0.539(0.240,1.209) | 0.131 | ||
| Education | 0.057 | 0.45 | ||||||||
| Less than High school or equivalent | 0.436(0.263,0.722) | 0.002* | 0.942(0.517,1.718) | 0.844 | 0.778(0.496,1.222) | 0.271 | 0.343(0.177,0.664) | 0.002* | ||
| College or above | 0.699(0.470,1.040) | 0.076 | 0.680(0.401,1.155) | 0.152 | 0.884(0.557,1.401) | 0.595 | 0.550(0.325,0.932) | 0.027* | ||
| High school or equivalent | 0.609(0.338,1.096) | 0.097 | 0.281(0.121,0.648) | 0.003* | 0.504(0.256,0.993) | 0.048* | 0.411(0.204,0.827) | 0.013* | ||
| PIR | 0.126 | 0.135 | ||||||||
| 0-1.3 | 0.632(0.444,0.899) | 0.011* | 0.604(0.420,0.869) | 0.007* | 0.814(0.584,1.135) | 0.222 | 0.456(0.303,0.688) | < 0.001* | ||
| 1.3–3.5 | 0.952(0.556,1.629) | 0.855 | 0.847(0.448,1.599) | 0.604 | 1.169(0.699,1.955) | 0.547 | 0.669(0.341,1.312) | 0.239* | ||
| >3.5 | 0.309(0.146,0.653) | 0.002* | 0.383(0.161,0.912) | 0.031* | 0.398(0.185,0.856) | 0.019* | 0.295(0.132,0.657) | 0.003* | ||
| BMI | 0.214 | 0.3 | ||||||||
| < 18.5 | 1.882(0.000, 46151.032) | 0.813 | 3.752(0.000,1081433.016) | 0.695 | 3.807(0.002,6777.153) | 0.523 | 0.000(0.000, 0.000) | 0.009* | ||
| 18.5–25 | 0.291(0.167,0.507) | < 0.0001* | 0.532(0.287,0.986) | 0.045* | 0.578(0.325,1.030) | 0.062 | 0.239(0.115,0.496) | < 0.001* | ||
| 25–30 | 0.690(0.370,1.287) | 0.240 | 0.558(0.258,1.205) | 0.135 | 0.756(0.410,1.394) | 0.365 | 0.535(0.234,1.220) | 0.135 | ||
| ≥ 30 | 0.730(0.500,1.066) | 0.102 | 0.531(0.318,0.887) | 0.016* | 0.832(0.570,1.213) | 0.335 | 0.496(0.312,0.789) | 0.003* | ||
| Marital status | 0.577 | 0.211 | ||||||||
| Married | 0.575(0.357,0.927) | 0.024* | 0.623(0.338,1.147) | 0.127 | 0.679(0.379,1.218) | 0.192 | 0.521(0.313,0.867) | 0.013* | ||
| Never married | 0.487(0.249, 0.954) | 0.036* | 0.474(0.226, 0.996) | 0.049* | 0.939(0.468, 1.885) | 0.858 | 0.188(0.083, 0.423) | < 0.001* | ||
| Living with partner | 0.251(0.096, 0.658) | 0.006* | 0.497(0.183, 1.348) | 0.163 | 0.430(0.175, 1.054) | 0.064 | 0.205(0.063, 0.670) | 0.01* | ||
| Others | 0.817(0.533,1.251) | 0.347 | 0.615(0.331,1.141) | 0.122 | 0.887(0.521,1.509) | 0.655 | 0.600(0.337,1.070) | 0.083 | ||
| Smoking status | 0.399 | 0.126 | ||||||||
| Never | 0.698(0.436,1.118) | 0.133 | 0.614(0.332,1.137) | 0.12 | 0.677(0.395,1.163) | 0.156 | 0.655(0.383,1.122) | 0.122 | ||
| Former | 0.528(0.274,1.019) | 0.057 | 0.339(0.153,0.751) | 0.008* | 0.664(0.328,1.343) | 0.251 | 0.294(0.149,0.578) | < 0.001* | ||
| Now | 0.632(0.401,0.998) | 0.049* | 0.887(0.521,1.508) | 0.654 | 1.063(0.717,1.577) | 0.758 | 0.436(0.262,0.724) | 0.002* | ||
| Drinking status | 0.082 | 0.617 | ||||||||
| Never | 0.394(0.149,1.040) | 0.06 | 0.705(0.203,2.453) | 0.575 | 0.500(0.180,1.387) | 0.178 | 0.470(0.174,1.267) | 0.132 | ||
| Former | 1.367(0.673,2.779) | 0.38 | 0.376(0.164,0.860) | 0.021* | 1.015(0.438,2.356) | 0.971 | 0.731(0.361,1.480) | 0.377 | ||
| Mild to moderate | 0.484(0.307,0.762) | 0.002* | 0.517(0.296,0.905) | 0.021* | 0.633(0.398,1.007) | 0.053 | 0.389(0.216,0.702) | 0.002* | ||
| Heavy | 0.594(0.356,0.989) | 0.045* | 0.697(0.380,1.280) | 0.24 | 0.840(0.502,1.404) | 0.501 | 0.406(0.212,0.781) | 0.008* |
Abbreviations: BMI, body mass index; CI, confidence interval; PIR, poverty-income ratio
RCS analysis
An RCS analysis, adjusted for multiple covariates, was conducted to visualize the dose-response relationship between MediHi and depression risk (Fig. 2). The results revealed that individuals with MediHi intake below 28.25 g had a higher risk of depressive symptoms. Beyond this threshold, increasing dietary live microbe intake appeared to offer a protective effect, indicating a non-linear relationship (P for non-linearity = 4e-04).
Fig. 2.
RCS analysis of the nonlinear correlation between live microbe intake and depression risk. The model was adjusted for age, gender, ethnicity, marital status, education level, PIR, BMI, drinking/smoking status, hypertension, ASCVD, DM, and hyperlipidemia. The solid blue line represents the smooth curve fit between variables
Model construction and interpretation
The predictive model developed using the XGBoost algorithm demonstrated good performance in both the training and validation sets. The AUC of the training and testing set was 0.834 and 0.897 respectively (Fig. 3). The SHAP plot provides insights into the impact of each variable within the XGBoost model on depression in the test datasets. The variables are ranked on the Y-axis according to their decreasing impact on the outcome. The X-axis shows the SHAP values, which represent the influence of each feature on the model prediction for the subject. According to the model, the five most influential factors are age, marital status, PIR, live microbe intake, and hypertension, listed in order of decreasing significance. (Fig. 4).
Fig. 3.
ROC curve of XGBoost machine learning algorithm in predicting depression risk for patients with gastrointestinal diseases
Fig. 4.
Summary plot of SHAP values for the model constructed by XGBoost algorithm. The dot represents the direction of contribution of each value of each variable, with orange representing higher values and purple representing lower values of each variable
Discussion
Using data from 2,195 individuals with gastrointestinal diseases in the NHANES dataset, our study revealed an inverse association between dietary live microbe consumption and depressive symptoms. This association remained robust after adjusting for sociodemographic factors and comorbidities such as hypertension, ASCVD, DM, and hyperlipidemia. Furthermore, the interpretable XGBoost prediction model based on multiple features demonstrates good performance in assessing depression risk in individuals suffering from gastrointestinal diseases. This novel finding suggests that higher dietary live microbe intake may serve as a protective factor against depressive symptoms in this population.
In present study, two methods were utilized to categorize dietary live microbe intake into low, medium, and high, or alternatively into G1, G2, and G3 groups. Individuals with medium or high intake exhibited significantly lower odds of depressive symptoms compared to those with low intake. Previous research indicates that probiotic supplementation can positively impact cognitive function and depression by altering the gut microbiota [30–32]. Nevertheless, probiotics constitute only one category within the vast array of live microbe. The consumption of live microbe offers a more comprehensive reflection of dietary factors. A substantial intake of dietary live microbe has been found to considerably lower the prevalence of diabetic kidney disease among DM patients [33]. The consumption of live microbe is also associated with various conditions, including cardiovascular diseases [34], chronic constipation [35], and arthritis [36]. Moreover, our study constitutes the first exploration of the link between live microbe intake and depression among individuals with gastrointestinal diseases.
Live microbes are naturally present in a variety of foods commonly consumed in our daily diet, including yogurt, fermented foods, mushroom-based food supplements, and fresh fruits and vegetables. These microbes can influence the composition and behavior of the resident gut microbiota [37]. The microbiota-gut-brain axis is believed to be a key pathway through which gut microbiota influences depression [38]. Microbiota and metabolites in the gastrointestinal tract communicate with the central and enteric nervous systems via neuroendocrine and enteroendocrine signaling pathways [39]. In addition, microbiota and gut metabolites can translocate through gut barrier damage, leading to dysregulation of inflammatory factors such as Treg, interleukin-6, and tumor necrosis factor-α, which are several hypothesized pathways contributing to the multifactorial etiology of depression [40–42]. Sudo et al. found that the gut microbiota regulate the hypothalamic-pituitary-adrenal axis in mice to combat stress, such as anxiety and depression [43]. These findings underscore the need for randomized controlled trials to establish causality between dietary live microbe intake and depression risk in gastrointestinal disease populations.
The potential therapeutic role of dietary live microbe, particularly those derived from mushrooms, in managing depression is gaining increasing attention. A recent review by Wang et al. [44] underscores the significance of mushroom and fungus extracts in preclinical models of depression, providing a robust foundation for our findings. The review systematically evaluated the antidepressant activity of various mushroom species, highlighting the diverse mechanisms through which these natural compounds may exert their effects. For instance, species such as Ganoderma lucidum and Hericium erinaceus [45] have been noted for their mood-improving effect, aligning with our observation of a reduced risk of depressive symptoms with increased live microbe intake.
This research possesses several key advantages. First, this investigation is the first to examine the relationship between live microbe intake and depression among individuals with gastrointestinal diseases. Second, using NHANES data, which follows a multi-stage probability sampling design, ensures the findings are relevant to the broader US population. Third, the study design took into account a range of confounding factors, such as sociodemographic characteristics and common comorbid diseases, to more precisely evaluate the association between exposure and outcome. Therefore, our research findings hold significant implications for public health initiatives targeting depression prevention in individuals with gastrointestinal diseases.
Nonetheless, the following limitations should be considered when generalizing the current research findings. First, the current study included a relatively small number of participants, and replication studies with larger sample sizes are needed to enhance the reliability and generalizability of the findings. Second, the cross-sectional nature of the study cannot establish the causality between live microbe intake and depression. Third, recall bias may exist since the participants’ depressive symptoms and dietary intake of live microbes were determined through self-report and 24-hour dietary recalls. The lack of a relatively objective measure of depression is an unavoidable limitation of this study. Lastly, because we performed a series of subgroup, trend, and dose–response analyses without formal correction for multiple comparisons, there remains a risk of inflated type I error. Future confirmatory studies should apply multiplicity adjustments or prespecified analysis plans to guard against false-positive findings.
Conclusion
In conclusion, our study highlights a potential protective role of live microbe intake in reducing depressive symptoms among patients with gastrointestinal diseases. Further longitudinal and interventional studies are warranted to establish causality and determine the optimal range of live microbe intake for depression prevention in this population.
Electronic supplementary material
Below is the link to the electronic supplementary material.
Supplementary Table 1. Associations between different live microbe intake group and depression through weighted univariate logistic regression analyses
Acknowledgements
Not applicable.
Abbreviations
- AUC
Area under the curve
- ASCVD
Arteriosclerotic cardiovascular disease
- BMI
Body mass index
- CI
Confidence interval
- DM
Diabetes mellitus
- NHANES
National Health and Nutrition Examination Survey
- OR
Odds ratio
- PHQ-9
Patient Health Questionnaire-9
- PIR
Poverty-income ratio
- RCS
Restricted cubic spline
- SE
Standard error
- SHAP
Shapley additive explanations
- USDA
U.S. Department of Agriculture
- XGBoost
Extreme gradient boosting
Author contributions
Cheng Zhang: Methodology, Software, Formal analysis, Data curation, Writing– original draft, Funding acquisition. Ying Liu: Validation, Investigation. Ke Li: Validation, Investigation. Shu-Ning Xu: Validation, Investigation. Ming-Hui Ma: Methodology, Data curation, Writing– review & editing, Funding acquisition.
Funding
This research was supported by the PhD Start-up Fund of Henan Cancer Hospital (No. zx1657) and Henan Provincial Medical Science and Technology Research Project (SBGJ202403015).
Data availability
No datasets were generated or analysed during the current study.
Declarations
Ethics approval and consent to participate
The National Center for Health Statistics Research Ethics Review Board approved the research protocols, and all participants provided written informed consent (Protocol #2005-06, #2011-17, and #2018-01).
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Peery AF, Crockett SD, Murphy CC, Jensen ET, Kim HP, Egberg MD, Lund JL, Moon AM, Pate V, Barnes EL, et al. Burden and cost of gastrointestinal, liver, and pancreatic diseases in the united states: update 2021. Gastroenterology. 2022;162:621–44. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Peery AF, Murphy CC, Anderson C, Jensen ET, Deutsch-Link S, Egberg MD, Lund JL, Subramaniam D, Dellon ES, Sperber AD, et al. Burden and cost of gastrointestinal, liver, and pancreatic diseases in the united states: update 2024. Gastroenterology. 2025;168:1000–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Sperber AD, Bangdiwala SI, Drossman DA, Ghoshal UC, Simren M, Tack J, Whitehead WE, Dumitrascu DL, Fang X, Fukudo S, et al. Worldwide prevalence and burden of functional Gastrointestinal disorders, results of Rome foundation global study. Gastroenterology. 2021;160:99–e114113. [DOI] [PubMed] [Google Scholar]
- 4.Mykletun A, Jacka F, Williams L, Pasco J, Henry M, Nicholson GC, Kotowicz MA, Berk M. Prevalence of mood and anxiety disorder in self reported irritable bowel syndrome (IBS). An epidemiological population based study of women. BMC Gastroenterol. 2010;10:88. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Yan M, Chen J, Liu F, Li H, Zhao J, Guo W. Abnormal default mode network homogeneity in major depressive disorder with Gastrointestinal symptoms at rest. Front Aging Neurosci. 2022;14:804621. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Bhandari S, Larson ME, Kumar N, Stein D. Association of inflammatory bowel disease (IBD) with depressive symptoms in the united States population and independent predictors of depressive symptoms in an IBD population: A NHANES study. Gut Liver. 2017;11:512–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Hillila MT, Hamalainen J, Heikkinen ME, Farkkila MA. Gastrointestinal complaints among subjects with depressive symptoms in the general population. Aliment Pharmacol Ther. 2008;28:648–54. [DOI] [PubMed] [Google Scholar]
- 8.Marx W, Lane M, Hockey M, Aslam H, Berk M, Walder K, Borsini A, Firth J, Pariante CM, Berding K, et al. Diet and depression: exploring the biological mechanisms of action. Mol Psychiatry. 2021;26:134–50. [DOI] [PubMed] [Google Scholar]
- 9.Wu T, Liu R, Zhang L, Rifky M, Sui W, Zhu Q, Zhang J, Yin J, Zhang M. Dietary intervention in depression - a review. Food Funct. 2022;13:12475–86. [DOI] [PubMed] [Google Scholar]
- 10.Mukherjee A, Breselge S, Dimidi E, Marco ML, Cotter PD. Fermented foods and Gastrointestinal health: underlying mechanisms. Nat Rev Gastroenterol Hepatol 2023. [DOI] [PubMed]
- 11.Roselli M, Natella F, Zinno P, Guantario B, Canali R, Schifano E, De Angelis M, Nikoloudaki O, Gobbetti M, Perozzi G, Devirgiliis C. Colonization ability and impact on human gut microbiota of foodborne microbes from traditional or Probiotic-Added fermented foods: A systematic review. Front Nutr. 2021;8:689084. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Valentino V, Magliulo R, Farsi D, Cotter PD, O’Sullivan O, Ercolini D, De Filippis F. Fermented foods, their Microbiome and its potential in boosting human health. Microb Biotechnol. 2024;17:e14428. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Marco ML, Hill C, Hutkins R, Slavin J, Tancredi DJ, Merenstein D, Sanders ME. Should there be a recommended daily intake of microbes?? J Nutr. 2020;150:3061–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Sindhu KN, Sowmyanarayanan TV, Paul A, Babji S, Ajjampur SS, Priyadarshini S, Sarkar R, Balasubramanian KA, Wanke CA, Ward HD, Kang G. Immune response and intestinal permeability in children with acute gastroenteritis treated with Lactobacillus rhamnosus GG: a randomized, double-blind, placebo-controlled trial. Clin Infect Dis. 2014;58:1107–15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.D’Souza AL, Rajkumar C, Cooke J, Bulpitt CJ. Probiotics in prevention of antibiotic associated diarrhoea: meta-analysis. BMJ. 2002;324:1361. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Li B, Liang L, Deng H, Guo J, Shu H, Zhang L. Efficacy and safety of probiotics in irritable bowel syndrome: A systematic review and Meta-Analysis. Front Pharmacol. 2020;11:332. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Kim CE, Yoon LS, Michels KB, Tranfield W, Jacobs JP, May FP. The impact of prebiotic, probiotic, and synbiotic supplements and yogurt consumption on the risk of colorectal neoplasia among adults: A systematic review. Nutrients 2022, 14. [DOI] [PMC free article] [PubMed]
- 18.Wallace CJK, Milev R. The effects of probiotics on depressive symptoms in humans: a systematic review. Ann Gen Psychiatry. 2017;16:14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Chudzik A, Orzylowska A, Rola R, Stanisz GJ. Probiotics, prebiotics and postbiotics on mitigation of depression symptoms: modulation of the Brain-Gut-Microbiome Axis. Biomolecules 2021, 11. [DOI] [PMC free article] [PubMed]
- 20.Marco ML, Hutkins R, Hill C, Fulgoni VL, Cifelli CJ, Gahche J, Slavin JL, Merenstein D, Tancredi DJ, Sanders ME. A classification system for defining and estimating dietary intake of live microbes in US adults and children. J Nutr. 2022;152:1729–36. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Jeddi MZ, Yunesian M, Gorji ME, Noori N, Pourmand MR, Khaniki GR. Microbial evaluation of fresh, minimally-processed vegetables and bagged sprouts from chain supermarkets. J Health Popul Nutr. 2014;32:391–9. [PMC free article] [PubMed] [Google Scholar]
- 22.Rezac S, Kok CR, Heermann M, Hutkins R. Fermented foods as a dietary source of live organisms. Front Microbiol. 2018;9:1785. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Valentin-Bon I, Jacobson A, Monday SR, Feng PC. Microbiological quality of bagged cut spinach and lettuce mixes. Appl Environ Microbiol. 2008;74:1240–2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Kroenke K, Spitzer RL, Williams JB. The PHQ-9: validity of a brief depression severity measure. J Gen Intern Med. 2001;16:606–13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Manea L, Gilbody S, McMillan D. A diagnostic meta-analysis of the patient health Questionnaire-9 (PHQ-9) algorithm scoring method as a screen for depression. Gen Hosp Psychiatry. 2015;37:67–75. [DOI] [PubMed] [Google Scholar]
- 26.Moriarty AS, Gilbody S, McMillan D, Manea L. Screening and case finding for major depressive disorder using the patient health questionnaire (PHQ-9): a meta-analysis. Gen Hosp Psychiatry. 2015;37:567–76. [DOI] [PubMed] [Google Scholar]
- 27.Alwhaibi M. The impact of anxiety and depression on Health-Related quality of life in hyperlipidemic adults in the united States. J Clin Med 2025, 14. [DOI] [PMC free article] [PubMed]
- 28.Wang X, Wang H, Yu Q, Fu S, Yang Z, Ye Q, Lin F, Cai G. High dietary live microbe intake is correlated with reduced risk of depressive symptoms: A cross-sectional study of NHANES 2007–2016. J Affect Disord. 2024;344:198–206. [DOI] [PubMed] [Google Scholar]
- 29.Chen H, Lundberg SM, Lee SI. Explaining a series of models by propagating Shapley values. Nat Commun. 2022;13:4512. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Kim CS, Cha L, Sim M, Jung S, Chun WY, Baik HW, Shin DM. Probiotic supplementation improves cognitive function and mood with changes in gut microbiota in Community-Dwelling older adults: A randomized, Double-Blind, Placebo-Controlled, multicenter trial. J Gerontol Biol Sci Med Sci. 2021;76:32–40. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Shi S, Zhang Q, Sang Y, Ge S, Wang Q, Wang R, He J. Probiotic Bifidobacterium longum BB68S improves cognitive functions in healthy older adults: A randomized, Double-Blind, Placebo-Controlled trial. Nutrients 2022, 15. [DOI] [PMC free article] [PubMed]
- 32.Suda K, Matsuda K. How microbes affect depression: underlying mechanisms via the Gut-Brain Axis and the modulating role of probiotics. Int J Mol Sci 2022, 23. [DOI] [PMC free article] [PubMed]
- 33.Wang M, Huang ZH, Zhu YH, Li S, Li X, Sun H, He P, Peng YL, Fan QL. Association of dietary live microbe intake with diabetic kidney disease in patients with type 2 diabetes mellitus in US adults: a cross-sectional study of NHANES 1999–2018. Acta Diabetol 2024. [DOI] [PMC free article] [PubMed]
- 34.Han L, Wang Q. Association of dietary live microbe intake with cardiovascular disease in US adults: A Cross-Sectional study of NHANES 2007–2018. Nutrients 2022, 14. [DOI] [PMC free article] [PubMed]
- 35.Yang C, Hong Q, Wu T, Fan Y, Shen X, Dong X. Association between dietary intake of live microbes and chronic constipation in adults. J Nutr. 2024;154:526–34. [DOI] [PubMed] [Google Scholar]
- 36.Liu Y, Xing Y, Kong X. Exploratory analysis on the relationship between dietary live microbe intake and arthritis: a National population based cross-sectional study. Front Nutr. 2023;10:1328238. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.David LA, Maurice CF, Carmody RN, Gootenberg DB, Button JE, Wolfe BE, Ling AV, Devlin AS, Varma Y, Fischbach MA, et al. Diet rapidly and reproducibly alters the human gut Microbiome. Nature. 2014;505:559–63. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Mlynarska E, Gadzinowska J, Tokarek J, Forycka J, Szuman A, Franczyk B, Rysz J. The role of the Microbiome-Brain-Gut Axis in the pathogenesis of depressive disorder. Nutrients 2022, 14. [DOI] [PMC free article] [PubMed]
- 39.Furness JB, Rivera LR, Cho HJ, Bravo DM, Callaghan B. The gut as a sensory organ. Nat Rev Gastroenterol Hepatol. 2013;10:729–40. [DOI] [PubMed] [Google Scholar]
- 40.Alvarez-Mon MA, Gomez AM, Orozco A, Lahera G, Sosa MD, Diaz D, Auba E, Albillos A, Monserrat J, Alvarez-Mon M. Abnormal distribution and function of Circulating monocytes and enhanced bacterial translocation in major depressive disorder. Front Psychiatry. 2019;10:812. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Alvarez-Mon MA, Gomez-Lahoz AM, Orozco A, Lahera G, Diaz D, Ortega MA, Albillos A, Quintero J, Auba E, Monserrat J, Alvarez-Mon M. Expansion of CD4 T lymphocytes expressing Interleukin 17 and tumor necrosis factor in patients with major depressive disorder. J Pers Med 2021, 11. [DOI] [PMC free article] [PubMed]
- 42.Alvarez-Mon MA, Gomez-Lahoz AM, Orozco A, Lahera G, Sosa-Reina MD, Diaz D, Albillos A, Quintero J, Molero P, Monserrat J, Alvarez-Mon M. Blunted expansion of regulatory T lymphocytes is associated with increased bacterial translocation in patients with major depressive disorder. Front Psychiatry. 2020;11:591962. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Sudo N. Microbiome, HPA axis and production of endocrine hormones in the gut. Adv Exp Med Biol. 2014;817:177–94. [DOI] [PubMed] [Google Scholar]
- 44.Wang CK, Kim G, Aleksandrova LR, Panenka WJ, Barr AM. A scoping review of the effects of mushroom and fungus extracts in rodent models of depression and tests of antidepressant activity. Front Pharmacol. 2024;15:1387158. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Fijalkowska A, Jedrejko K, Sulkowska-Ziaja K, Ziaja M, Kala K, Muszynska B. Edible mushrooms as a potential component of dietary interventions for major depressive disorder. Foods 2022, 11. [DOI] [PMC free article] [PubMed]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplementary Table 1. Associations between different live microbe intake group and depression through weighted univariate logistic regression analyses
Data Availability Statement
No datasets were generated or analysed during the current study.