TABLE 3.
Summary of observational studies examining the association between high total, free, and added sugar consumption on metabolic parameters and chronic diseases1
| Reference, year | Subjects | Study duration | Dietary comparator | Main findings |
|---|---|---|---|---|
| Ahmadi-Abhari et al., 2014 (91) | 25,639 adults aged 40–79 y (mean ± SD age, 61.2 ± 8.3 y, 56.5% males) | 10 y | Intakes of total sugars, sucrose, and fructose | Intakes of total sugars and sucrose were not associated with risk of T2DM [HR per 40 g/d, 0.95 (95% CI, 0.83–1.08); HR per 27 g/d, 1.00 (95% CI, 0.88–1.12) for total sugars and sucrose, respectively]. Fructose intake was inversely associated with risk of T2DM (HR per 10 g/d, 0.88; 95% CI, 0.78–0.99) |
| Assy et al., 2008 (60) | 31 patients with NAFLD (mean ± SD age, 30 ± 13 y; 53.0% males) vs. 30 healthy controls (age- and sex-matched) | NA (cross-sectional study) | Intake and sources of added sugars | Patients with NAFLD had 125% higher intake of added sugar (P = 0.001), and a higher proportion of their added sugar intake came from soft drink and juice (43% vs. 8%; P = 0.001) when compared with healthy controls, respectively |
| Barclay et al., 2007 (92) | 4477 Australians aged 49+ y | 10 y | Sugar intake (per 100 g/d) | Intake of sugar (per 100 g/d) was not associated with increased risk of T2DM [HR, 1.02 (95% CI, 0.62–1.67; P = 0.949) and HR, 1.09 (95% CI, 0.63–1.88; P = 0.767) for age- and sex-adjusted and multivariate-adjusted models, respectively] |
| Bergeron et al., 2021 (150) | 1019 adults aged 18–65 y at baseline (50% males) from the PREDISE study | NA (cross-sectional analysis) | State of sugar-containing foods (solid vs. liquid) and form of sugar (free sugars vs. naturally occurring sugar) | High intake of free sugar from soft drinks was associated with higher fasting insulin level (1.06%; 95% CI, 0.30%–1.84%; P = 0.006) and HOMA-IR insulin resistance (1.01%; 95% CI, 0.19%–1.84%; P = 0.02) compared with the lowest intake category. Intake of naturally occurring sugar from solid foods was not associated with any outcome. After adjusting for covariates, associations were not significant for all states and forms of sugars |
| Burger et al., 2011 (101) | 8855 males (mean ± SD age, 43.0 ± 11.0 y) and 10,753 females (42.1 ± 11.3 y) aged 21–64 y | 11.9 y | Sugar intake | Sugar intake was not significantly associated with risk of CHD [HR per SD increases, 1.17 (95% CI, 0.99–1.38) and 1.10 (95% CI, 0.86–1.41) for males and females, respectively] and stroke [HRs, 1.00 (95% CI, 0.70–1.44) and 0.96 (95% CI, 0.65–1.44) for males and females, respectively] |
| Hodge et al., 2004 (97) | 36,787 males and females aged 40–69 y without T2DM at baseline (41.0% males) | 4 y | Sugar intake (per 100 g/d) | Intake of sugar (OR per 100 g/day, 0.61; 95% CI, 0.47–0.79; P < 0.001) was inversely associated with the incidence of T2DM. The association was weaker after adjusting for BMI and waist-to-hip ratio (OR per 100 g/d, 0.72; 95% CI, 0.56–0.93; P = 0.01) |
| Janket et al., 2003 (93) | 39,345 women from the Women's Health Study (mean ± SD age, 53.3 ± 6.6 y) | 6 y | Intakes of total sugar, sucrose, and fructose across quintiles | Intakes of total sugar, sucrose, and fructose were not significantly associated with increased risks of T2DM, compared with the lowest intake category [RRs, 0.86 (95% CI, 0.69–1.06), 0.84 (95% CI, 0.67–1.04), and 0.96 (95% CI, 0.78–1.19) for total sugars, sucrose, and fructose respectively] |
| Liu et al., 2000 (99) | 75,521 women aged 38–63 y | 10 y | Consumption of sucrose and fructose across quintiles | Intakes of sucrose and fructose in the highest quintile were not significantly associated with increased risks of CHD compared with the lowest quintile [RRs, 1.22 (95% CI, 0.94–1.60) and 1.07 (95% CI, 0.82–1.40) for sucrose and fructose, respectively] |
| Meyer et al., 2000 (78) | 35,988 older Iowa women aged 55–69 y at baseline | 6 y | Intakes of glucose, sucrose, and fructose across quintiles | High sucrose intake was associated with a lower incidence of T2DM (RR, 0.81; 95% CI, 0.67–0.99), compared to the lowest quintile. Higher glucose and fructose intakes were associated with increased incidences of T2DM, compared to the lowest quintile [RRs, 1.30 (95% CI, 1.08–1.57) and 0.81 (95% CI, 0.67–0.99) for glucose and fructose, respectively] |
| Montonen et al., 2007 (77) | 4304 males and females aged 40–60 y [mean ± SD ages, 51.7 ± 8.0 y (53.8% males) and 57.5 ± 7.0 y, (37.3% males) for noncases and cases of T2DM, respectively) | 5 y | Intakes of total sugar, fructose, soft drinks, and sucrose across quartiles | High total sugar intake was modestly associated with an increased incidence of T2DM (Quartile 4 vs. Quartile 1; RR, 1.56; 95% CI, 0.99–2.46; Ptrend = 0.10). High fructose and soft drink intakes were associated with increased incidences of T2DM [Quartile 4 vs. Quartile 1; RRs, 1.68 (95% CI, 1.06–2.65; Ptrend = 0.009) and 1.60 (95% CI, 0.93–2.76; Ptrend = 0.01) for fructose and soft drinks, respectively). High sucrose intake was not associated with the incidence of T2DM (Quartile 4 vs. Quartile 1; RR, 1.22; 95% CI, 0.77–1.92; Ptrend = 0.35) |
| Olsson et al., 2021 (86) | 26,622 participants from the MDCS (39% males) | 18 y | Intakes of sucrose, added sugar, SSBs, and table sugar by quintiles (percentages of daily energy intake) | Intakes of sucrose and added sugar were not associated with the risk of T2DM [Quintile 5 vs. Quintile 1; HRs, 1.03 (95% CI, 0.92–1.15; Ptrend = 0.41) and 0.95 (95% CI, 0.85–1.07; Ptrend = 0.65) for sucrose and added sugar, respectively]. Consumption of SSBs and table sugar were also not associated with risks of T2DM [Quintile 5 vs. Quintile 1; HRs, 1.05 (95% CI, 0.96–1.14; Ptrend = 0.23) and 1.03 (95% CI, 0.93–1.14; Ptrend = 0.81) for SSBs and table sugar, respectively) |
| Ramne et al., 2019 (79) | 24,272 participants from the MDCS (mean age, 57.6 y; range, 44–73 y; 38.6% males) & 24,475 participants from the NSHDS (mean age, 48.6 y; 36–64 y; 46.3% males) | ∼20 y | Intakes of free sugars and added sugars (<5% vs. 5%–7.5% vs. 7.5%–10% vs. 10%–15% vs. 15% to <20% vs. ≥20% of daily energy intake), and sugar sources (treats vs. SSBs) | Added and free sugar intakes of ≥20% of daily energy intake were associated with increased risks of all-cause mortality, compared with intake between 7.5% and <10% of daily energy intake [MDCS HR, 1.30 (95% CI, 1.12–1.51; P < 0.001) and NSHDS HR, 1.31 (95% CI, 1.01–1.70; P = 0.005) for added sugar; MDCS HR, 1.26 (95% CI, 1.10–1.44; P < 0.001) and NSHDS HR, 1.29 (95% CI, 1.03–1.63; P = 0.337) for free sugar] |
| Intake of treats was inversely associated with all-cause mortality [>14 vs. ≤2 servings/wk; MDCS HR, 0.83 (95% CI, 0.74–0.93; P < 0.001); NSHDS HR, 0.66 (95% CI, 0.56–0.78; P < 0.001)], whereas intake of SSBs was positively associated with all-cause mortality [>8 vs. ≤1 servings/wk; MDCS HR, 1.14 (95% CI, 1.03–1.26; P = 0.035); NSHDS HR, 1.10 (95% CI, 0.90–1.35; P = 0.549)] | ||||
| Seo et al., 2019 (87) | 7005 Korean adults aged between 40–69 y (53.5% males) | NA (cross-sectional study) | Energy from total sugar intake (≤20% kcal vs. >20% kcal) | Males who consumed >20% kcal from total sugar had 49.1% (95% CI, 16.2%–91.4%), 31.3% (95% CI, 3.8%–66.0%) and 33.2% (95% CI, 3.8%–70.9%) higher odds of obesity, low HDL, and MetSyn respectively, compared with those who derived ≤20% kcal from total sugar, after adjustment for confounders. No statistically significant association between these outcomes and energy from total sugar was observed in women |
| Schulze et al., 2008 (94) | 9702 males and 15,365 females aged 35–65 y from the EPIC- Potsdam cohort | 9 ± 2 y | Intakes of sucrose and fructose across quintiles (g/d) | Intakes of sucrose and fructose were not significantly associated with risks of T2DM in men [Quintile 5 vs. Quintile 1; RRs 0.72 (95% CI, 0.50–1.04; P = 0.063) and 1.00 (95% CI, 0.74–1.35; P = 0.987) for sucrose and fructose, respectively] and women [Quintile 5 vs. Quintile 1; RRs 1.13 (95% CI, 0.74–1.74; P = 0.492) and 1.09 (95% CI, 0.75–1.58; P = 0.877) for sucrose and fructose, respectively] |
| Sieri et al., 2010 (100) | 13,637 males (35–64 y) and 30,495 females (35–74 y) | 7.9 y | Sugar intake across quartiles | Participants in the highest quartile of sugar intake did not have increased risk of CHD compared to those in the lowest quartile [RRs, 1.10 (95% CI, 0.69–1.76; Ptrend = 0.83) and 0.97 (95% CI, 0.69–1.38; Ptrend = 0.75) for females and males, respectively] |
| Sluijs et al., 2010 (98) | 37,846 participants aged 21–70 y at baseline of EPIC-NL cohort (25.6% males) | 10 y | Sugar intake | High sugar intake was associated with a lower incidence of T2DM (HR per SD increase, 0.87; 95% CI, 0.81–0.93; P < 0.001) |
| Sluijs et al., 2013 (95) | 12,403 incident T2DM cases & 16,835 subcohort participants (37.8% males) | 12 y | Sugar intake across quartiles | Sugar intake was not associated with the risk of T2DM (Quartile 4 vs. Quartile 1; HR, 0.96; 95% CI, 0.86–1.07) |
| Tapanee et al., 2021 (75) | 524 young adults aged 18–31 y (17.4% males) | NA (cross-sectional study) | Consumption of total sugar, added sugar, sucrose, fructose, and glucose (g/d or percentage of daily energy intake) | Compared with the normal-weight controls, overweight or obese subjects had significantly higher intakes of total sugar (F2251 = 7.156; P < 0.001), added sugar (F2251 = 7.742; P < 0.001), fructose (F2251 = 8.243; P < 0.001), glucose (F2251 = 9.249; P < 0.001), and sucrose (F2251 = 3.592; P = 0.028), and higher percentage energy intakes from total sugar (F2251 = 11.779; p < 0.001) and added sugar (F2251 = 10.198; P < 0.001) |
| Tasevska et al., 2014 (103) | 353,751 participants aged 50–71 y from the NIH-AARP Diet and Health Study (58.3% males) | 13 y | Intakes of total and added sugar, total and added fructose, and total and added sucrose | No statistically significant association was observed between added sugars, total sucrose, or added sucrose for all-cause mortality or mortality from cancer, CVD, and other causes in females (Quartile 5 vs. Quartile 1) |
| Compared with females with the lowest total fructose intake, those who had the highest intake had a 10% (95% CI, 4%–17%) increased risk of all-cause mortality (Ptrend < 0.0001). The association was weaker in males (6% increase in risk; Ptrend = 0.002) | ||||
| In males, high added sugar intake was not statistically significantly associated with the risks of all-cause mortality or mortality from cancer and CVD (Quartile 5 vs. Quartile 1). Interestingly, high added sucrose intake was found to be associated with reduced risks of CVD mortality (Quartile 5 vs. Quartile 1; HR, 0.93; 95% CI, 0.86–1.01; Ptrend = 0.02) | ||||
| Tasevska et al., 2018 (96) | 82,254 postmenopausal women aged 50–79 y from the Women's Health Initiative Observational Study | 16 y | Total sugar intake | High total sugar intake was not associated with increased risks of T2DM [HRs per 20% increase in calibrated total sugars, 0.94 (95% CI, 0.77–1.15) and 1.00 (95% CI, 0.85–1.18) in multivariable energy substitution and partition models, respectively], total CVD [HRs, 0.97 (95% CI, 0.87–1.09) and 0.91 (95% CI, 0.80–1.04)], total CHD [HRs, 0.96 (95% CI, 0.86–1.07) and 0.90 (95% CI, 0.78–1.04)], and total stroke [HRs, 1.00 (95% CI, 0.85–1.18) and 0.97 (95% CI, 0.85–1.10)]. |
| Warfa et al., 2016 (80) | 26,190 participants from the MDCS (mean ± SD age, 58 ± 7.6 y; range, 44.3–73.6 y; 38.0% males) | 17 y | Sucrose intake (<5% vs. 5%–7.5% vs. 7.5%–10% vs. 10%–15% vs. >15% of daily energy intake) | Sucrose intake in the highest category (>15% of daily energy intake) was associated with an increased risk of a coronary event (HR, 1.37; 95% CI, 1.13–1.66; Ptrend = 0.008), compared with the lowest intake category (<5% of daily energy intake) |
| Yamakawa et al., 2020 (67) | 13,229 residents aged 35–69 y of Yamakawa City in Japan (44.4% males) | 10 y | Intakes of free sugars, glucose, fructose, sucrose, and lactose | No statistically significant association between the intake of sugars and weight change was observed in females across quartiles |
| High intakes of sucrose and fructose were associated with weight gain in males across quartiles (Ptrends = 0.018 for sucrose and 0.001 for fructose) | ||||
| Yang et al., 2014 (83) | 11,733 adult respondents aged 20 y and older of the NHANES (48.1% males) | 18 y | Percentage of daily energy intake from added sugars (0 to <9.6 vs. 9.6 to <13.1 vs. 13.1 to <16.7 vs. 16.7 to <21.3 vs. ≥21.3%) | Percentage daily energy intake from added sugar was dose-dependently associated with CVD mortality (HR for 0 to <9.6% vs. ≥21.3%, 2.03; 95% CI, 1.26–3.27; Ptrend = 0.04) after adjustment for confounders |
1
AARP, American Association of Retired Persons; CHD, coronary heart disease; CVD, cardiovascular disease; EPIC, European Prospective Investigation into Cancer and Nutrition; EPIC-NL, European Prospective Investigation into Cancer and Nutrition–Netherlands; MDCS, Malmö Diet and Cancer Study; MetSyn, metabolic syndrome; NA, not applicable; NAFLD, nonalcoholic fatty liver disease; NSHDS, Northern Swedish Health and Disease Study; PREDISE, Rédicteurs Individuels, Sociaux et Environnementaux; SSB, sugar-sweetened beverage; T2DM, type 2 diabetes mellitus.