health effects of non-sugar sweeteners

 

The WHO recently published a tome (210 pages) on the health effects of non-sugar sweeteners (see https://www.who.int/publications/i/item/9789240046429 for access). Thanks to Jorge Finke for bringing this to my attention.

 

Details:

-- 283 studies were included in the systematic review/meta-analysis, including adults (and pregnant women) and kids

 

Results:

 

Adults, comparing those with higher intake of non-sugar sweeteners (NSS):

-- weight/adiposity (32 randomized controlled trials (RCTs) and 13 prospective cohort studies):

    -- body weight: in RCTs, there was a 0.71 kg loss (-1.13 to -0.28) in NSS consumers

    -- BMI:  in RCTs, there was a non-significant decrease of 0.14 kg/m² (-0.30 to 0.02) in NSS consumers; in prospective cohort studies, there was a there was a significant increase of 0.14 kg/m² (0.03 to 0.25), p=0.01

    -- incident obesity: in prospective cohort studies, there was a 76% increased risk in NSS consumers, HR 1.76 (1.25-2.49), p<0.01, though only 2 studies assessed this outcome

 

-- type 2 diabetes (12 prospective cohort studies):

    -- risk of developing diabetes:

        -- NSS consumed as beverages: 23% increased risk, HR 1.23 (1.14 to 1.32), p<0.01, in prospective cohort studies

        -- NSS added to foods or beverages: 34% increased risk, HR 1.34 (1.21 to 1.48), p<0.01, in prospective cohort studies

            -- to minimize reverse causation (eg overweight people consumed more NSS): all studies were adjusted for relevant confounders, including BMI; most performed a number of relevant sensitivity analyses including the exclusion of diabetes cases the first 3 to 7 years of follow-up

    -- intermediate markers of diabetes, 21 RCTs and 3 prospective cohort studies:

        -- high fasting glucose: higher intake of NSS was associate with 21% increased risk, HR 1.21 (1.01 to 1.45)

        -- no statistically significant difference in fasting insulin, hemoglobin A1c, or HOMA-IR (a mathematical composite reflecting insulin resistance)

 

-- all-cause mortality (7 prospective cohort studies):

    -- 12% increased in risk, HR 1.12 (1.05-1.19), p<0.01.  all but 1 studies found some increase

 

-- cardiovascular disease

    -- cardiovascular mortality (4 prospective cohort studies):

        -- 19% increased risk with NSS, HR 1.19 (1.07 to 1.32), p<0.01. all studies found an increase

    -- cardiovascular events (3 prospective cohort studies):

        -- 32% increased risk, HR 1.32 (1.17-1.50), p<0.01

    -- coronary heart disease (4 prospective cohort studies):

        -- nonsignificant trend to worse outcome: HR 1.16 (0.97-1.39), p=0.09

    -- stroke (5 prospective cohort studies):

        -- 19% increased risk of any type of stroke, HR 1.19 (1.09 to 1.29), p<0.01

            -- 33% increase in hemorrhagic stroke, HR 1.33 (1.03 to 1.72)

            -- 22% increase in ischemic stroke, HR 1.22 (1.04 to 1.44)

    -- hypertension (4 prospective cohort studies):

        -- 13% increased risk, HR 1.13 (1.09-1.17), p<0.01)

    -- intermediate markers of cardiovasc disease (19 RCTs and 4 prospective cohort studies):

        -- no clear effect in RCTs on systolic or diastolic blood pressure, and slight but significant increase in total cholesterol:HDL ratio in NSS consumers

 

-- cancer

    -- cancer mortality (3 prospective cohort studies):

        -- no significant difference

    -- cancer incidence (39 case-controlled and 9 cohort studies):

        -- no significant difference for any type of cancer

 

-- chronic kidney disease (2 RCTs and 2 prospective cohort studies):

     -- only one prospective cohort study with 15,348 participants found an association: 64% increase risk of end-stage renal disease, HR 1.64 (1.18 to 2.28). no association in other studies

 

-- eating behavior (26 RCTs included in meta-analysis)

    -- energy intake: higher intake of NSS intake was associated with 560 kJ/day decrease in total energy intake in RCTs, but in subgroup analysis of those overweight/obese, the difference was not significant

        -- hunger: no significant difference

        -- satiety: higher NSS intake was associated with small but significant decrease

        -- appetite and desire to eat: higher NSS intake was associated with small but significant increase

    -- sweet preference (12 RCTs): higher NSS was associate with a reduction of sugar intake by 39g/d, when NSS was used to replace sugars

 

-- dental caries (one 6-month RCT): no acid erosion of the enamel or caries developed in either group

 

-- mood (2 RCTs, 2 prospective cohort studies): higher intake of NSS (aspartame or stevia) found no difference in short-term trials (20 days, 6 weeks). one prospective cohort study found after 6 years there was a nonsignificant lower cognitive function with NSS; the other found that after 10 years, there was increased risk of dementia/Alzheimers in those consuming NSS, adjusting for prevalent hypertension, cardiovascular disease, diabetes and risk factors for those diseases

 

Children:

-- adiposity (2 RCTs, 14 cohort studies): no difference in meta-analysis of RCTs (but only small number of studies reporting appropriate data). one study did find BMI, waist circumference, and body fat mass did decrease when NSS-sweetened beverages replaced sugar-sweetened beverages (SSBs)

-- type 2 diabetes (no studies on development of diabetes):  1 prospective cohort study found no difference in chronic consumers of NSS-sweetened beverages in postprandial glucose when compared to sucrose, or most markers of diabetes. A1c did increase more in chronic NSS consumers

-- cardiovascular disease (1 cohort study): 12-18yo overweight children followed one year: those drinking NSS-sweetened beverages had no changes in total cholesterol, HDL, LDL, or triglycerides

-- cancer (2 case-controlled studies): no associated between maternal NSS consumption and brain cancer in kids

-- energy intake (4 studies): contradictory results

-- hunger (1 RCT): no difference

-- satiety (1 RCT): no difference, though preference to drink SSBs

-- sugar intake (3 studies): mixed results, from 100 g/d less sugar if had NSS, but up to 14 g/d more in other studies

-- dental carries (4 studies): mixed results. the best trial was comparing stevia vs placebo for 6 months, finding less plaque and gingiva problems with stevia, and fewer cavitated lesions

-- mood (2 studies): a 3 week study found no differences in mood with aspartame or saccharine; another study using EEG found indication of a negative impact on mood

-- behavior (1 study): in kids described by parents as sensitive to sugars, no differences in behavior, conduct, attention deficit, deviation, attention, hyperactivity, social skills or oppositional behavior with sucrose, aspartame or saccharine for 3 weeks

-- neurocognition (2 studies): an RCT of children drinking sucralose or sucrose for 8.5 months found no difference in cognition measures; a prospective cohort study following kids in utero to age 7 found their cognition scores were inversely associated with maternal intake of NSS-sweetened beverages during pregnancy, but no association between their own cognition and NSS-sweetened beverages

-- asthma (1 study): cross-sectional analysis found NSS-sweetened beverages in 11-year olds was associated with higher but nonsignificant odds of asthma

-- allergies: no statistically significant difference

 

Pregnant women:

-- maternal outcomes:

    -- gestational diabetes (1 cohort study, 1 cross-sectional study): no relationship with NSS sweetened beverages

        -- another cohort study found that in women with gestational diabetes, drinking 2 or more NSS-sweetened beverages during and after pregnancy was associated with significantly higher A1c, fasting glucose and obesity

    -- gestational weight gain (3 studies; 1 RCT, 2 cohort studies): 1 RCT and 1 prospective cohort study found increased gestational weight gain with NSS; a cohort study did not find a difference

--birth outcomes:

    -- preterm birth (3 prospective cohort studies): NSS intake during pregnancy associated with 25% increased risk, with a dose-response relationship (esp between weeks 34-37)

    -- birthweight (2 studies): one with no relationship with birthweight or BMI, the other of NSS-sweetened products before conception was association with increased birthweight

    -- LGA infants (1 cohort study): no association

    -- adiposity in offspring  (3 prospective cohort studies): 2 of the 3 studies found small increases in BMI of offspring when pregnant women consumed NSS beverages

    -- neurocognition (1 prospective cohort study): children in early and mid-childhood (til age 7) had cognition inversely related to maternal intake of NSS beverages when pregnant

    -- asthma (1 birth cohort study): women consuming >1 NSS-sweetened beverage during  pregnancy had children assessed at 1.5 and 7 years old, finding significant increases in asthma

    -- allergies (1 birth cohort study): same as asthma study, but with nonsignificant increase in allergic rhinitis by age 7

    -- others: no increase in congenital heart disease in offspring, or spontaneous abortion

   

Commentary:

-- several studies have documented the consumption of free sugars as important in the current epidemics of overweight/obesity as well as in several diet-related noncommunicable diseases such as dental caries, type 2 diabetes, cardiovascular diseases, and cancer. Non-sugar sweeteners (NSS) were developed with the expectation that they would decrease these problems

 

--in this review, NSS consumption in RCTs have been associated with a small reduction in body weight and BMI in adults, especially when NSS was compared with sugar, likely related to a reduction in energy intake. Prospective cohort studies suggested that higher NSS intake has been associated with increased body weight and increased clinical outcomes of type II diabetes, cardiovascular diseases, and all-cause mortality. Saccharine may be related to increased incidence of bladder cancer. Studies in children have been largely inconclusive, though an RCT did find replacing sugar with NSS sweetened beverages was associated with a small reduction in adiposity. Pregnant women do seem to have an increased risk of preterm birth and possibly adiposity in offspring

    -- In the RCTs of NSS consumption (all short-term studies) did find weight loss and decreased total energy intake, the much longer-term results found in the prospective cohort studies (up to 10 years) found increases in obesity, as well as type II diabetes, cardiovascular diseases, and overall mortality.

    -- these differences between RCTs and the cohort studies could be related to reverse causation: high consumers of NSS might be doing so because of underlying disease or higher risk for developing disease (e.g. diabetes). The authors of these studies did take pains to reduce this likelihood (e.g. excluding patient outcomes for the first few years of follow-up); the fact that some studies found a dose-response curve further makes this less likely, but it still remains a possibility. On the other hand, RCTs also have an issue: it is unclear whether the construction of the RCT (e.g. giving participants NSS sweetened beverages as a replacement of sugar sweetened ones) may not actually reflect real-world consumption of beverages in terms of the quantity or their mix.

 

-- the mechanism whereby NSS might be detrimental include their interaction with both oral sweet taste receptors (the high intensity of the sweetness of the NSS can augment the desire to have both more NSS and more sugar-containing beverages), as well as on other receptors besides sweet taste ones (extragustatory tissues with sweet receptors include the gut, pancreas, bladder, brain, bone and adipose tissue: https://pubmed.ncbi.nlm.nih.gov/24763065/ ). also there can be profound changes in the gut microbiome by NSS

    -- Of note, high fructose corn syrup is a remarkably potent food additive (which is really abundantly used in the food chain) but leads to hyperuricemia, and this may well be a cause of increased cardiovascular disease: http://gmodestmedblogs.blogspot.com/2023/05/allopurinol-decreases-heart-disease.html 

 

-- other relevant blogs of the past:

http://gmodestmedblogs.blogspot.com/2020/11/cardiovasc-disease-inc-with-sweet.html : a French study (NutriNet-Sante Study) finding more cardiovascular disease in people drinking more sweet beverages, including those with artificial sweeteners

http://gmodestmedblogs.blogspot.com/2022/05/artificial-sweeteners-and-cancer-risk.html : another NutriNet-Sante Study  finding that artificial sweeteners (especially aspartame and acesulfame-K) were associated with increased cancer

http://gmodestmedblogs.blogspot.com/2018/04/non-nutritive-food-additives-and-c.html : a study finding that the artificial sugar trehalose was associated with a more virulent strain of C difficile intestinal infections, reflecting a change in the microbiome

http://gmodestmedblogs.blogspot.com/2017/01/artificial-sweeteners-microbiome1.html : more information on the affect of artificial sweeteners on the microbiomes

http://gmodestmedblogs.blogspot.com/2019/02/artificial-sweeteners-and-stroke.html : Women's Health Initiative Observational Study finding that artificially sweetened beverages are associated with increased risk of stroke, coronary artery disease, and all-cause mortality

http://gmodestmedblogs.blogspot.com/2017/05/stroke-and-dementia-and-artificial.html : two Framingham Heart Study studies finding association between artificially-sweetened beverages and both stroke and dementia

http://gmodestmedblogs.blogspot.com/2019/10/soft-drinks-increase-mortality.html : increased mortality with either sugar-sweetened or artificially-sweetened soft drinks

http://gmodestmedblogs.blogspot.com/2016/09/sugar-industrys-role-in-creating-policy.html : sugar industry’s role in creating nutrition policy and receiving validation by including key public health experts (highly paid) in supporting the sugar industry (eg Frederick Stare, chair of Harvard School of Public Health)

 

Limitations:

-- they did not measure foods supplemented with NSS in many of the studies (most of the studies really focused on beverages, which are likely a major NSS component). But this could distort the above results in people eating NSS supplemented foods 

-- other studies (including the one referenced in this review) have found that NSS may be associated with decreased body weight and energy intake when substituted for sugary beverages. This is expected since there is less caloric intake. But I think the real issue is not substituting NSS for sugar sweetened beverages but substituting water for these beverages; there was no benefit, but actual harm, in the few studies comparing NSS to water consumption

    -- for people who "do not like water", i have suggested using a small amount of a natural juice of their choice as a flavoring device

-- limited information on subjective outcomes, such as food tastes and changes in peoples' likes and dislikes in the RCTs

-- limited information on the different NSSs used. much of the data was from older studies and may not including the newer sugar substitutes

-- many of the RCTs were very short-term studies and may not reflect the true clinical effects on longer term NSS ingestion

-- some of the longer-term cohort studies did not include multiple assessments of diet throughout the study: the initial assessment may not reflect changes during the study that might alter the results

 

so, the issue here, of course, is that societies around the world (especially in the resource-rich countries but increasingly in poorer ones) have evolved so that the major causes of morbidity/mortality have changed from infectious diseases to chronic diseases (diabetes, cardiovascular disease, etc etc). And these chronic diseases are predominantly as a result of the associated "lifestyle" changes: lack of exercise (and using more vehicles to get around, with the attendant pollution and global warming effects), changes in diet (more fast foods, calorie-dense and sweetened products, more artificial ingredients such as NSS, high-fructose corn syrup, preservatives, and in general "more ingredients that your grandmother had never heard of"), more massive industrialized food production (leading to more economic inequities in rural areas with loss of family farming, more contamination of the food chain from massive numbers of animals in confined spaces), more globalization (with its attendant global inequities, well-described in NY Times editorial: https://www.nytimes.com/2023/06/18/business/economy/global-economy-us-china.html?smid=nytcore-ios-share&referringSource=articleShare ), more pollution (air pollution from big factories in some previously relatively pristine areas, more contamination of the oceans and local waterways eg with plastics, agricultural and other pesticides/fertilizers, more excreted medications prescribed to treat many of the lifestyle-related chronic conditions (eg lots of fluoxetine in the local waters leading to weird anatomic changes in tadpoles, etc; toxic chemical effluents from large factories contaminating the water, air and soil), etc.

 

the WHO summary above elucidates the adverse effects of one significant component of diet: NSS (also referred to as non-nutritive sweeteners). The food chain overall has been transformed by nutrition/chemical engineers to produce better tasting foods ("better living through chemistry"), aggressive advertising for these foods, complicity in local stores in promoting these often nutritionally-deficient and ultimately harmful products (calorie-dense foods are typically cheaper and featured prominently in large supermarket chains, as well as often being the only foods available in  local small bodegas/convenience stores, a huge issue in poorer communities with less access to large supermarkets). And many of these foods (junk food) are really difficult "addictions" making it difficult for people to stop easily.

 

bottom line: overall not a great direction we are moving in. i personally have had some success with patients in focusing on soda (both sugar and non-sugar sweetened), one of the lower hanging fruits, and promoting water as the liquid of choice. 

 

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 (2 options):

 

1. go to http://gmodestmedblogs.blogspot.com/ to see the blogs in reverse chronological order

  -- click on 3 parallel lines top left, if you want to see blogs by category, then click on "labels" and choose a category​

  -- 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

 

2. go to https://bucommunitymed.wpengine.com/, a website from the Community Medicine section at Boston Medical Center.  This site does have a very searchable and accessible list of my blogs and is the easiest to view blogs and displays more at a time.

 

if you would like to see the article, please email me. 

 

please feel free to circulate this to others. also, if you send me their emails (gmodest@bidmc.harvard.edu), i can add them to the list