TABLE 2.
Summary of observational studies examining the association between high SSB consumption on metabolic parameters and chronic diseases1
| Reference, year | Subjects | Study duration | Dietary comparator | Main findings |
|---|---|---|---|---|
| Berkey et al., 2004 (112) | 11,755 adolescents aged 9–14 y from the US Growing Up Today study (43.1% boys) | 3 y | Consumption of SSBs | Before adjustment for total energy intake, consumption of SSBs was associated with increase in BMI in the corresponding year (boys: +0.03 kg/m2 per daily serving, P = 0.04; girls: +0.02 kg/m2, P = 0.096), compared with nondrinkers. Children who increased consumption by ≥2 servings/d from the prior year gained weight [boys: +0.14 kg (P = 0.01); girls: +0.10 kg (P = 0.046)], compared with those with unchanged intakes |
| After adjustment for total energy intake, the effects were not significant | ||||
| de Koning et al, 2012 (81) | 42,883 males aged 40–75 y in the Health Professionals Follow-Up Study | 22 y | SSB consumption (never vs. 2/mo vs. 1–3/wk vs. 3.7/wk to 9/d) | Higher SSB consumption was associated with increased risks of CHD (RR for never vs. 3.7/wk to 9/d: 1.18; 95%CI, 1.06–1.31; Ptrend < 0.01 after adjustment for confounders) |
| An increase in every serving of SSB per day was also associated with 12.7 (95% CI, 4.2–21.2) mg/dL higher TG (P = 0.01), 1.87 (95% CI, 1.03–2.70) mg/dL lower HDL (P < 0.01), 0.12 (95% CI, 0.02–0.23) mg/L higher CRP (P = 0.02), 0.16 (95% CI, 0.03–1.65) pg/mL higher IL-6 (P = 0.02), and 796 (95% CI, 149–1442) pg/mL lower leptin (P = 0.02) | ||||
| den Biggelaar et al., 2020 (149) | 2240 middle-aged subjects (mean ± SD age, 59.5 ± 8.1 y; 50.4% male) | NA (cross-sectional study) | Non-consumers vs. moderate or daily SSB consumers | No statistically significant difference in β-cell glucose sensitivity and potentiation factor, C-peptidogenic index, overall insulin secretion, and Matsuda index between nonconsumers vs. moderate or daily SSB consumers |
| Dhingra et al., 2007 (73) | Cross-sectional and longitudinal analyses of the Framingham Heart Study Cohort (6039 person- observations, 3470 in women; mean age 52.9 y) | 3 y | Consumption of sugar-sweetened soft drinks | Cross-sectionally, consumption of ≥1 serving/d of sugar-sweetened soft drink was associated with increased prevalence of MetSyn (OR, 1.81; 95% CI, 1.28–2.56), compared to intake of <1 serving/wk |
| Longitudinally, consumption of ≥1 serving/d was associated with increased incidence of MetSyn (OR, 1.62; 95% CI, 0.96–2.75), compared with infrequent drinkers (<1 serving/wk) | ||||
| Duffey et al., 2010 (70) | 2774 adults (mean ± SD age, 25.0 ± 3.6 y; females, 53.5% ± 0.8%) from the CARDIA study | 20 y | Consumption of SSBs across quartiles | Higher SSB consumption was associated with increased risks of high WC (adjusted RR, 1.09; 95% CI, 1.04–1.14; Ptrend < 0.001), high LDL cholesterol (adjusted RR, 1.18; 95% CI, 1.02–1.35; Ptrend = 0.018), high TG (adjusted RR, 1.06; 95% CI, 1.01–1.13; Ptrend = 0.033), and hypertension (adjusted RR, 1.06; 95% CI, 1.01–1.12; Ptrend = 0.023) across quartiles |
| Eny et al., 2020 (90) | 1778 preschool children aged 3–6 y (53.4% boys) | 9 y | Consumption of sugar-containing beverage | An increase in every serving of sugar-containing beverage per day was associated with 0.02 (95% CI, 0.01–0.03) mmol/L lower HDL (P = 0.01) and 1.05 (95% CI, 1.01–1.10) mmol/L higher TG (P = 0.03) after adjustment for confounders |
| No statistically significant association was observed between sugar-containing beverage consumption and blood glucose or systolic blood pressure | ||||
| Fagherazzi et al., 2013 (61) | 66,118 females (mean ± SD age, 52.6 ± 6.6 y) from the E3N cohort | 14 y | SSB consumption (nonconsumer vs. <86 vs. 86–164 vs. 165–359 vs. >359 mL/wk) | Higher SSB consumption was associated with increased risks of T2DM (HR for nonconsumer vs. >359 mL/wk: 1.30; 95% CI, 1.02–1.66; Ptrend = 0.021) after adjustment for confounders |
| Fung et al., 2009 (82) | 88,520 females from the Nurses’ Health Study aged 34–59 y | 24 y | SSB consumption in servings (<1/mo vs. 1–4/mo vs. 2–6/wk vs. 1 to <2/d vs. ≥2/d) | Higher consumption of SSBs was associated with increased risks of CHD (RR for <1/mo vs. ≥2/d: 1.35; 95% CI, 1.07–1.69; Ptrend < 0.001) |
| Funtikova et al., 2015 (68) | 2181 Spanish males and females aged 25–74 y | 9 y | Changes in soft drink consumption (maintenance of no consumption vs. decrease in consumption vs. increase in consumption vs. maintained consumption) | 100-kcal increase in soft drink consumption was associated with 1.1-cm increase in WC (P = 0.018), and higher soft drink consumption was associated increased odds of 10-year incidence of abdominal obesity (OR for no consumption vs. ≥200 ml/d, 1.77; 95% CI, 1.07–2.93) |
| Garduño-Alanís et al., 2020 (62) | 5205 Russian adults aged 45–69 y (47% males) from the Health, Alcohol and Psychosocial factors in Eastern Europe cohort | 3 y | Fruit juice or SSB consumers vs. nonconsumers | No statistically significant association between fruit juice consumption and unit change in BMI (drinkers vs. nondrinkers; OR, 0.92; 95% CI, 0.81–1.05; P = 0.203)SSB consumers had 26% (95% CI, 9%–45%) higher odds of having a 1-kg/m2 increase in their BMI in 3 years compared with nondrinkers |
| Hirahatake et al., 2019 (76) | 4719 Black and White males and females aged 18–30 y at baseline from the CARDIA study (45.3% males) | 30 y | SSB consumption in servings (none to ≤1/wk vs. 1 to ≤4/wk vs. 4 to ≤7/wk vs. 1–2/d vs. ≥2/d) | An increase in every serving/d of SSB was associated with a 6% (95% CI, 1%–10%) increase in the risk of T2DM (P = 0.009) |
| Harrington et al., 2020 (74) | 1075 boys and girls aged 8–11 y (66.1% boys) | NA (cross-sectional analysis) | SSB consumption | Compared with normal-weight children, children with overweight or obesity had significantly higher intake of SSBs per day (383 vs. 315 mL). Also, children who consumed >200 mL per day of SSBs had a higher risk of overweight or obesity compared with those consuming <200 mL per day (OR, 1.8; 95% CI, 1.0–3.5) |
| Haslam et al., 2020 (88) | The FOS (n = 3146; mean ± SD age, 54.8 ± 9.8 y; 46.9% males) and Generation Three cohorts (n = 3584; mean ± SD age, 40.3 ± 8.8 y; 45.7% males) | 12.5 y | SSB consumption from none or <1 serving per month to ≥6 servings/d | Compared with low consumption (<1 serving/mo), regular consumption (>1 serving/d) of SSBs was associated with a greater mean decrease in HDL cholesterol (β ± standard error, −1.6 ± 0.4 mg/dl; Ptrend < 0.0001) and increase in TG concentrations (β ± standard error: 4.4 ± 2.2 mg/dl; Ptrend = 0.003) |
| Imamura et al., 2019 (63) | 27,662 adults from the EPIC-InterAct case-cohort study [mean ± SD age, 52.0 ± 9.0 y (38% males) and 56 ± 7.7 y (50% males) for randomly selected subcohort and ascertained cases of T2DM, respectively] | 15 y | SSB consumption (per 250 g/d increase and 250 g/d vs. 0 g/d) | For every 250 g/d increase in SSB consumption, the risk of T2DM incidence increases by 18% (95% CI, 8%–28%)Comparing with nonconsumers (i.e., 0 g/d), those who consumed 250 g/d of SSB had a 7.4/10,000 person-years increase in T2DM rates |
| Janzi et al., 2020 (71) | 25,877 adults aged 45–74 y (mean age, 57.8 y; 37.6% males) from the Malmö Diet and Cancer Study | 19.5 y | Consumption of total added sugar and sugar-sweetened foods and beverages across categories | Added sugar intake > 20% daily energy intake was associated with increased risks of coronary events (HR, 1.39; 95% CI, 1.09–1.78) compared to the lowest intake category (<5% daily energy intake), and of stroke (HR, 1.31; 95% CI, 1.03–1.66), compared to 7.5%–10% daily energy intake |
| Lin et al., 2020 (89) | 6856 adults from the NHANES (50.5% males) | 3 y | SSB consumption [none vs. 1–350 (light) vs. 351–699 (medium) vs. ≥ 700 ml/d (heavy)] | Compared with nonconsumers, heavy SSB consumers had a 0.26 mg/l higher CRP level after adjusting for BMI. When taking into consideration the modifying effect of BMI, medium and heavy drinkers who were obese had 0.58 (P = 0.014) and 0.50 mg/l (P = 0.013) higher CRP levels than non-SSB drinkers, respectively |
| Ma et al., 2015 (64) | 2634 participants of the Framingham Heart Study (47.5% males) | NA (cross-sectional analysis) | SSB consumption in servings (0–1/mo vs. 1/mo to <1/wk vs. 1/wk to <1/d vs. ≥1/d) | Higher SSB consumption was associated with increased odds of NAFLD (OR for 0–1/mo vs. ≥ 1/d, 1.56; 95% CI, 1.03–2.36; Ptrend = 0.02) after adjustment for age, sex, dietary confounders, and smoking. The statistical significance was lost after additional adjustment for VAT |
| Malik et al., 2019 (84) | 37,716 men from the Health Professional's Follow-up Study and 80,647 women from the Nurses’ Health Study | Health Professional's Follow-Up Study (28 y)Nurses’ Health Study (34 y) | SSB consumption (number of times of consuming a standard portion of foods and beverages; <1/mo vs. 1–4/mo vs. 2–6/wk vs. 1 to <2/d vs. ≥2/d) | Across categories, high SSB consumption was associated with higher risks of total mortality in a dose-response relationship [HRs of 1.00 (reference), 1.01 (95% CI, 0.98–1.04), 1.06 (95% CI, 1.03–1.09), 1.14 (95% CI, 1.09–1.19), and 1.21 (95% CI, 1.13–1.28) for consumption frequencies of <1/mo, 1–4/mo, 2–6/wk, 1 to <2/d and ≥2/d, respectively; P < 0.0001] |
| High SSB consumption was also associated with increased risks of CVD mortality across categories [HRs of 1.00 (reference), 1.06 (95% CI, 1.00–1.12), 1.10 (95% CI, 1.04–1.17), 1.19 (95% CI, 1.08–1.31), 1.31 (95% CI, 1.15–1.50) for consumption frequencies of <1/mo, 1–4/mo, 2–6/wk, 1 to <2/d and ≥2/d, respectively; P < 0.0001] | ||||
| O'Conner et al., 2018 (104) | 9678 British adults (mean ± SD age, 47.8 ± 7.4 y; 46.6% males) | NA (cross-sectional analysis) | Sugar intake from liquid foods and solid foods (Q1: 0.5–8.0 vs. Q2: 8.0–10.4 vs. Q3: 10.4–12.6 vs. Q4: 12.6–15.5 vs. Q5: 15.5–46.4; % daily energy intake) | After correction for multiple testing (α = 0.003), sugars from liquid foods were positively associated with ln HOMA-IR (Q5 vs. Q1; β-coefficient, 0.11; 95% CI, 0.07–0.15; Ptrend < 0.001), ln-CRP (β-coefficient, 0.21; 95% CI, 0.13–0.28; Ptrend < 0.001), and metabolic risk z-score (β-coefficient, 0.18; 95% CI, 0.13–0.24; Ptrend < 0.001). No association was found for sugars from solid foods |
| Odegaard et al., 2010 (72) | 43,580 Chinese Singaporeans (mean ± SD age, 54.8 ± 7.5 y; 42.9% males) | 5 y | Consumption of soft drinks (almost never vs. 1–3 portions/mo vs. 1 portion/wk vs. 2 to ≥3 portions/wk) | Consumption of ≥2 soft drinks/wk was associated with an increased risk of T2DM (RR, 1.42; 95% CI, 1.25–1.62), compared to the lowest intake category |
| Palmer et al., 2008 (113) | 59,000 African American females aged 21–69 y at baseline | 6 y | Consumption of SSBs (<1 drink/mo vs. 1–7 drinks/mo vs. 2–6 drinks/wk vs. 1 drink/d vs. ≥2 drinks/d) | Increase in consumption was associated with increased risks of T2DM for sugar-sweetened soft drinks (Ptrend = 0.002) and sugar-sweetened fruit drinks (Ptrend = 0.001). Consuming ≥2 drinks/d was associated with increased risks of type 2 diabetes (incidence rate ratio, 1.24; 95% CI, 1.06–1.45) for soft drinks and fruit drinks (incidence rate ratio, 1.31; 95% CI, 1.13–1.52), compared with the lowest consumption category (<1 drink/mo) |
| Pacheco et al., 2022 (85) | 100,314 women aged 22–104 y at baseline (median age, 53 y) from the California Teachers Study | 20 y | SSB or its subtypes consumption (rare or never vs. >rare or never to <1 serving/wk vs. ≥ 1 to ≤6 servings/wk vs. ≥7 servings/wk) | For total SSBs, consumption of ≥7 servings/wk was not associated with total, CVD, or cancer mortality compared with rare or never consuming. For caloric soft drinks, a significant association was found between consumption frequency of ≥7 servings/wk and all-cause mortality (HR, 1.26; 95% CI, 1.10–1.46; Ptrend = 0.02) and cancer mortality (HR, 1.33; 95% CI, 1.08–1.63; Ptrend = 0.08), compared with rare or never consumption |
| Romaguera et al., 2013 (65) | 27,058 subjects [11,684 incident cases (unknown male:female ratio) and 15,374 controls (37.8% males)] from the EPIC-InterAct study | 16 y | Fruit juice and SSB consumption in glass (<1/mo vs. 1–4/mo vs. >1–6/wk vs. ≥1/d) | Higher SSB consumption was associated with higher risks of T2DM (HR for <1/mo vs. ≥1/d, 1.29; 95% CI, 1.02–1.63; Ptrend = 0.013) after adjustment for confounders |
| No statistically significant association between the risk of T2DM and fruit juice intake was observed (HR for <1/mo vs. ≥1/d, 1.06; 95% CI, 0.90–1.25; Ptrend = 0.21) after adjustment for confounders | ||||
| Schulze et al., 2004 (66) | 91,249 females from the Nurses’ Health Study II aged 24–44 y at baseline | 8 y | SSB consumption at baseline (<1/mo vs. 1–4/mo vs. 2–6/wk vs. ≥1/d) and change in SSB consumption between 1991–1995 (consistent ≤1/wk vs. consistent ≥1/d vs. changed from ≤1/wk to ≥1/d vs. changed from ≥1/d to ≤1/wk vs. other) | Weight gain over 4 years was higher in females who increased their consumption from ≤1/wk to ≥1/d (+4.69 kg for 1991 to 1995 and 4.20 kg for 1995 to 1999) compared with those who decreased their consumption (+1.34 and 0.15 kg for the 2 periods, respectively)Higher SSB consumption was dose-dependently associated with higher risks of T2DM (RR for <1/mo vs. ≥1/d, 1.83; 95% CI, 1.42–2.36, Ptrend < 0.001) after adjustment for confounders |
| Stern et al., 2017 (69) | 11,218 females from the Mexican Teachers’ Cohort (mean ± SD age, 43.3 ± 5.2 y) | 2 y | Changes in consumption of sugar-sweetened soda (servings/wk): decreased (<−1) vs. no change (−1 to +1) vs. increased (>+1) vs. increase in 1 serving/d | Compared with no change, decrease in consumption by >1 serving/week was associated with less weight gain (−0.4 kg; 95% CI, −0.6 to −0.2), and increase in consumption by >1 serving/wk was associated with weight gain of 0.3 kg (95% CI, 0.2–0.5). Increase in 1 serving/d was associated with weight gain of 1.0 kg (95% CI, 0.7–1.2; P < 0.001) |
| For change in WC, compared with no change, decrease in consumption by >1 serving/wk was associated with reduction in WC by 0.5 cm (95% CI, 0.9 to −0.1), increase in consumption by >1 serving/wk was associated with increase in WC by 0.3 cm (95% CI, 0.1–0.6). Increase in 1 serving/d was associated with change in WC by +0.9 cm (95% CI, 0.5–1.4) |
1
CARDIA, Coronary Artery Risk Development in Young Adults; CHD, coronary heart disease; CRP, C-reactive protein; CVD, cardiovascular disease; E3N, The French E3N Prospective Cohort Study; EPIC, European Prospective Investigation into Cancer and Nutrition; FOS, Framingham Offspring Study; MetSyn, metabolic syndrome; NA, not applicable; NAFLD, nonalcoholic fatty liver disease; Q, quintile; SSB, sugar-sweetened beverage; T2DM, type 2 diabetes mellitus; TG, triglyceride; VAT, visceral adipose tissue; WC, waist circumference.