soft drinks increase depression by microbiome change

 An impressive study found that the soft drink consumption was associated with major depression disorder as well as the intensity of symptom severity, specifically in women (see depression soft drink change in microbiome JAMApsych2025 in dropbox, or doi:10.1001/jamapsychiatry.2025.2579)

 

Details:

-- a multi-center cohort study conducted in Germany using cross-sectional data from the Marburg-Munster Affective Cohort

-- 405 patients with major depressive disorder (MDD) were compared to 527 healthy controls

-- mean age 36 (nonsignificant difference of 36.37 if MDD and 35.33 in controls), 66% female (nonsignificant difference of 67.9% if MDD and 65.5% in controls)

-- Hamilton Rating Scale for Depression (HAMD) 8.75 if MDD and 1.56 in healthy controls (p<0.001); Beck Depression Inventory (BDI) 16.83 if MDD and 4.17 in healthy controls (p<0.001)

-- BMI 25 (26.04 if MDD and 24.08 in healthy controls, p<0.001); total calorie intake 21.52 kcal (no difference between groups)

-- soft drink consumption 237 grams/day if MDD and 162 grams/d in healthy controls, p=0.006

-- years of education 13.45 if MDD and 14.10 in healthy controls (p<0.001)

-- patients with MDD: 1.4 prior inpatient treatments for depression, 53.8 previous depressive episodes, 10.86 weeks total duration of previous inpatient treatments, 44 months total duration of all previous depressive episodes

-- comorbid psych conditions in patients with MDD: anxiety disorder 31%, eating disorder 6%, dysthymic disorder 16%, substance use disorder 2%, somatic symptom disorder 7%, psychotic disorder 3%

Main outcomes:

-- multivariable regression and analysis of variance (ANOVA) models assessing the association between soft drink consumption and both MDD diagnosis and symptom severity

-- association of gut microbiome changes to assess if microbiota changes mediated a link between soft drinks and  MDD, assessing specifically in the abundance of Eggerthella and Hungatella species 

 Results:

-- soft drink consumption was associated with:

    -- MDD (major depressive disorder) diagnosis: 

        -- female patients (n=932): odds ratio (OR) 1.081 (1.008-1.159), p = .03

        -- male patients (n = 312): no effect was observed, OR 1.020 (0.924-1.127), p=0.69

    -- MDD symptom severity in patients with MDD:

        -- female patients (n = 271): ηp² 0.023 (0.003-0.058), p=0.01 [ηp² is a statistical measure of the effect size, using analysis of variance]

            -- higher soft drink consumption was associated with higher symptom severity

        -- male (n = 129) no significant effect: ηp² 0.019 (0.000-0.069), p= 0.12

-- these associations between soft drink consumption and MDD diagnosis, as well as symptom severity, remained significant in female participants (all p < 0.05) when additionally controlling for BMI, while the effects were not significant in male participant

-- therefore, further analyses below included female patients only

-- microbiota changes (based on a prior study of this same group, finding that Eggerthella and Hungatella abundance were causally related to MDD):

    -- females: soft drink consumption was associated with Eggerthella abundance (p = 0.007; ηp² 0.017 (0.0002-0.068); n = 435), controlling for BMI

        -- but not associated with Hungatella abundance (p = 0.57; ηp² 0.000 (0.000-0.012); n = 435)

        -- higher soft drink intake was associated with lower microbiome alpha-diversity (p = 0.001; ηp² 0.025 (0.002-0.048); n = 435) [decreased microbiome alpha-diversity reflects fewer bacterial species and a less healthy microbiome]

    -- males: the results of microbiome analysis did not find significant changes in the abundance of:

        -- Eggerthella abundance (F(1,214)=1.10, p=0.295, ηp²=0.005

        -- Hungatella abundance (F(1,)=0.54, p=0.464, ηp²=0.003

        -- no effect on alpha-diversity (p = 0.30; ηp² 0.005 (0.000-0.023); n = 219).

-- association between soft drinks and MDD diagnosis mediated by Eggerthella species:

    -- mediation effect of Eggerthella between soft drink consumption and MDD effect: statistically significant 3.82% of the total effect of soft drink consumption on MDD diagnosis (p=0.01), controlling for BMI as well as total calorie intake (all with p<0.05)

    -- mediation effect of Eggerthella between soft drink consumption and depressive symptom severity: statistically significant 5.00% of the total effect of soft drink consumption on MDD diagnosis (p=0.005), controlling for BMI as well as total calorie intake (all with p<0.05)

Commentary:

A recent CDC report assessed sugar consumption by age, demographics, and the area of the US (https://www.cdc.gov/nutrition/php/data-research/added-sugars.html ), found that:

    -- "added sugars" included foods packaged as sweeteners, sugars from syrup and honey, and sugars from concentrated fruit or vegetable juices; this category did not include naturally occurring sugars in milk, fruit, and vegetables

    -- the leading sources of sugar in the US population aged 1 and older was 24% by sugar-sweetened beverages (16% by soft drinks, 5% by fruit drinks), 19% by desserts and sweet snacks, 11% by coffee and tea, 9% candy and sugars, 7% sandwiches, and 4% higher fat milk and yogurt

    -- 57.1% of children aged 1 to 5 years old consumed at lest one sugar-sweetened beverage in the past 7 days

    -- a 2017-2018 report found:

        -- average daily intake of added sugars was 18 teaspoons for boys and 15 teaspoons for girls

        -- average daily intake for children aged 2 to 5:

            -- 13 teaspoons for non-Hispanic black children

            -- 12 teaspoons for non-Hispanic white children

            -- 11 teaspoons for Hispanic children

            -- 7 teaspoons for non-Hispanic Asian children

        -- average daily intake for children aged 6 to 11:

            -- 19 teaspoons for non-Hispanic black children

            -- 18 teaspoons for non-Hispanic white children

            -- 16 teaspoons for Hispanic children

            -- 12 teaspoons for non-Hispanic Asian children

        -- average daily intake for children aged 12 to 19:

            -- 20 teaspoons for non-Hispanic black children

            -- 20 teaspoons for non-Hispanic white children

            -- 15 teaspoons for Hispanic children

            -- 14 teaspoons for non-Hispanic Asian children

        -- average daily intake for adults:

            -- men: 19 teaspoons

            -- women: 15 teaspoons

    -- mean percent of kcal from added sugars among children and adolescents aged 2-19 (https://www.cdc.gov/nchs/products/databriefs/db87.htm )

        -- Male: about 16%

        -- Female: about 15%

        -- non-Hispanic individuals consumed about 14%, non-Hispanic black individuals 16% and non-Hispanic white 17%

-- soft drink consumption has been shown to be associated with several medical conditions including obesity, type II diabetes, cardiovascular disease, and cancer. These associated diseases raise considerable concern, given that there is increasing soft drink consumption in children and adolescents

-- several studies in the past have found negative impact of soft drinks on mental health. 

-- one large trial of 31,712 females aged 42 to 62 from the Nurses' Health Study II found that ultra-processed foods were associated with a 49% increased risk of depression, independent of age, BMI, physical activity, or smoking. in particular, they found that that the most significant component of ultra-processed foods associated with depression were artificially sweetened beverages (37% increase) and "other artificial sweeteners" (26%); this study did not find that sugar-sweetened beverages themselves were associated with incident depression (https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2809727)

-- a few studies found that high-sugar solid foods were not associated with depression, as opposed to soft drinks. for example the population-based UK Biobank cohort of 188,426 participants found that there was a linear relationship between the extent of free sugar consumption from beverages and depression, vs intrinsic sugars in foods [perhaps this is related to the faster gastric emptying of beverages, overwhelming the small intestine's absorptive capacity and affecting bacterial growth there]

    -- the large simple-sugar load from beverages may affect the gut microbiome by promoting the growth of pro-inflammatory bacteria, weakening the gut integrity and impairing mucosal immunity. several beneficial bacteria, for example, ones that produce butyrate, elaborate short-chain fatty acids that are anti-inflammatory (https://pmc.ncbi.nlm.nih.gov/articles/PMC7284805/ highlights that sugar consumption increases the abundance of pro-inflammatory bacteria in the gut microbiome and decrease the anti-inflammatory ones)

    -- in mice, the microbiome changes from soft drinks induce glucose intolerance and dyslipidemia, independent of weight gain

    -- an interesting study found that metformin increased the growth of Akkermansia in the microbiome, and this bacterium itself increases glucose tolerance and decreases adipose tissue inflammation; this suggests that a perhaps quite important mechanism of metformin's effect in patients with diabetes may be mediated through its effect on the microbiome: https://gmodestmedblogs.blogspot.com/2024/07/microbiome-in-cardiovascular-disease.html

-- interestingly, transplanting human fecal material from depressed patients into microbiota-deplete rats led to a depressive phenotype and anxiety behavior in the rats (https://www.sciencedirect.com/science/article/abs/pii/S0022395616301571?via%3Dihub ). and one might imagine that an anxious rat might be especially concerning.......

-- this current study focused on specific microbiota changes (Eggerthella and Hungatella) since those bacteria were found to be associated with MDD in women in a prior study by this same group. However, other bacteria have been found in patients with MDD in other studies, including abundances of Atopium, Bifidobacterium, Enterococcus, Flavonifractor, Holdemania, and Streptococcus, and reduced abundance in Faecalibacterium .                                                                       

-- the researchers in the current study found that in women soft drink consumption was associated with an abundance of Eggerthella as a likely mediator of both MDD diagnosis and MDD symptom severity (though neither was found in men)

-- these findings add to the medical literature concerning the presence of the gut-brain axis:

    -- as noted in a 2023 blog https://gmodestmedblogs.blogspot.com/2023/07/gastroparesis-high-placebonocebo.html, "the gut/brain axis (see microbiome gut brain axis JClinInvest2015 in dropbox, or doi:10.1172/JCI76304) has been pretty well described showing that the gut microbiome can affect emotional behavior, stress- and pain-modulation, and brain neurotransmitters (the gut microbiota elaborates many neurotransmitters; animal studies have found major effects on the expression of brain signaling systems; some of these changes could affect appetite/diet/weight as well as the placebo effect (see https://www.frontiersin.org/articles/10.3389/fpsyt.2022.824468/full )"

    -- proinflammatory mediators in the gut microbiome (as happens with sugar intake) are associated with neuroinflammation, anxiety, and learning deficits, particularly during adolescence (which is the leading age associated with excessive soda intake)

    -- this current article adds to the gut-brain axis more information about depression risk and diet, finding changes in microbiota diversity (lower richness and evenness of the microbial taxa) as well as increases in Eggerthella with soft drink consumption being associated with the MDD

-- an interesting question is why there are such differences between women and men, highlighted by the finding that men drink more soft drinks than women. it certainly seems likely that this is hormonal -- https://www.sciencedirect.com/science/article/abs/pii/S0378512217306503 

-- since i have strayed back into the microbiome (i have had numerous blogs on it over the past several years: https://gmodestmedblogs.blogspot.com/search?q=microbiome ), several articles have come out recently on the gut-bone axis, some being animal studies, finding that there may be heterogeneity in the relationship between different types of osteoporosis (menopausal and that induced by aging itself) and changes in the microbiome (see osteoporosis microbiome change Bone2025 in dropbox, or doi.org/10.1038/s41413-025-00432-1, and osteoporosis microbiome MicrobioSpectrum2023 in dropbox or doi.org/10.1128/spectrum.00322-2), which suggest:

    -- estrogen deficiency leading to menopausal osteoporosis can induce bone loss through changes in the gut microbiota, in a mouse study: Estrogen deficiency induces bone loss through the gut microbiota in dropbox, or https://www.sciencedirect.com/science/article/pii/S1043661823002864?via%3Dihub

        -- ovariectomized mice had microbiota dysbiosis, especially with increases in the gut Firmicutes:Bacteroides ratio and serum lipopolysaccharides (microbiome depletion and the attendant decreased microbiome diversity are associated with changes in the intestinal barrier and lymphoid tissue, leading to absorption of lipopolysaccharides and systemic inflammation)

        -- gut microbiota depletion caused by antibiotic treatment improves bone mass, bone microstructure, and bone biomechanics in ovariectomized mice 

            -- postmenopausal mice with antibiotic-induced gut microbiota depletion have increased bone mass and mineral content, all suggesting that there are microbiome effects on estrogen metabolism. does that explain why the association with MDD and microbiome changes are present in women and not men??

            -- depletion of the microbiome has also been shown to decrease osteoarthritis progression, presumably by decreased bone mass in subchondral bone and improved bone microscopic outcomes

        

Limitations:

-- the situation with soft drinks is a bit complex and perhaps beyond the specific effects on the growth of Eggerthella. Soft drinks also contain additives, including artificial sweeteners and preservatives that can disrupt the microbiome. for example, in the Nurses' Health Study cited above, it was the artificially sweetened drinks that were the most associated with depression risk. the researchers in the current study note that for artificially sweetened beverages "our data did not show a specific effect on depressive symptoms, likely due to insufficient detail in data collection"

-- the increments in depression associated with soft drinks in this study were quite small, but the authors note that as mediators of increased depression risk, these are perhaps easier to control (low-hanging fruit) and they do give us some insight into further interventions (eg studying different artificial sweeteners, specific sugars such as fructose vs glucose vs more complex sugars)

-- these studies were associations between dietary sugar intake and microbiome changes and depression, raising several issues:

    -- self-reported diet may not be accurate

    -- diet is quite complex: how were the interaction between foods and soft drinks assessed (eg, did those drinking soft drinks more often have sweet cookies/pastries/etc with them or other foods with more sugars, natural or artificial?, and perhaps these researchers falsely attributed the depression to the soft drinks?? there was that Nurses' study cited above that only found MDD only with the artificially-sweetened soft drinks)

    -- did those drinking more soft drinks have other issues that affect the microbiome (eg, were they less likely to eat green vegetables, or do exercise, or had a more stressful life, or live in a food desert and mostly have access to small stores with limited healthy food selection, or be subject to discrimination or other stressors...???:  all of these affect the microbiome)

    -- did those drinking more soft drinks take more antibiotics that affect the microbiome (we don't know, since this was not assessed)

    -- since there is no possibility of doing a randomized trial (eg, getting lots of participants who are randomized to drinking varying amounts of soft drinks but otherwise eating the same as those in the control group, getting the same exercise, no difference in stress levels, ..., and following them for a few decades to assess their outcomes), we are therefore left with observational trials. some of these are better than others (eg the Nurses' Health Studies had rigorous regular assessments of diet that were taken into account in their studies vs this current study with minimal background data)

    -- and observational trials can therefore only detect associations and cannot determine causality, since there may well be undetected confounders (a large randomized trial, because of its randomness, will statistically have the 2 groups equal in terms of confounders)

so,

-- another study stressing the importance of a healthy gut microbiome (this is the largest of the microbiomes by far. very little information on the systemic effects of other microbiomes, such as the ones associated with the upper respiratory tract, the lower respiratory tract, skin, vagina..)

-- this all reinforces the importance of a healthy lifestyle, focusing on diet, exercise, appropriate weight, low levels of stress, good emotional support systems, etc.

    -- there is evidence that a healthy diet does decrease depressive symptoms in depressed young adults, and anti-inflammatory meds might also help: https://gmodestmedblogs.blogspot.com/2019/10/healthy-dietantiinflammatories-dec.html

-- this also suggests that other microbiome disruptors, especially antibiotics, should be prescribed/taken judiciously and appropriately (and trying to avoid the broad spectrum ones that are the most disruptive). we do know that antibiotics are overprescribed

    -- https://gmodestmedblogs.blogspot.com/2023/06/antibiotic-resistant-microorganisms-can.html is one of several blogs on antibiotic overprescribing and resistance, with reference to lots of other blogs on inappropriate prescribing. there was even a study finding that urgent care providers were more likely to prescribe antibiotics inappropriately in the afternoon than the morning (presumably related to their feeling more tired and wanting to finish their session more quickly??)

-- very fortunately, even dysfunctional microbiomes are usually reparable through reinstatement of healthy diet, exercise, etc

-- and, this current study found that the gut microbiome is likely a real contributor to both depression incidence and its severity.  this is a real issued especially since major depressive disorders are so very common, the meds for them do help but not in many cases, there are such profound effects of MDD on a person's life (home, work, etc) as well as on their families and their communities, and  healthy lifestyle changes may help significantly for MDD as well as most chronic diseases we encounter

geoff

-----------------------------------

If you would like to be on the regular email list for upcoming blogs, please contact me at gmodest@bidmc.harvard.edu

to get access to all of the blogs:  go to http://gmodestmedblogs.blogspot.com/ to see the blogs in reverse chronological order

or you can just click on the magnifying glass on top right, then type in a name in the search box and get all the blogs with that name in them