Skip to main content
. Author manuscript; available in PMC: 2019 Oct 11.
Published in final edited form as: Br J Nutr. 2018 Mar;119(5):570–580. doi: 10.1017/S0007114517003841

Table 3.

Concentrations of biomarkers by frequency of artificially sweetened beverage (ASB) intake among participants free of diabetes and CVD in the Nurses’ Health Study (n 8492)*

(Least-squares means and 95 % confidence intervals)

Frequency of ASB intake
Never or almost never Less than once per week Once to twice per week Three to six times per week Once per day or more
Biomarker concentrations Mean 95% CI Mean 95% CI Mean 95% CI Mean 95% CI Mean 95% CI Pfor trend
Hepatic biomarker
 Fetuin-A (μg/ml)
  n 564 236 132 315 162
  Model 1 458·2 4·5 465·7 6·9 468·4 9·2 461·5 6·0 475·7 8·4 0·12
  Model 2 461·5 4·9 462·4 7·2 465·1 9·3 457·5 6·4 468·9 8·7 0·55
 Alanine transaminase (U/l)
  n 537 220 114 289 158
  Model 1 19·1 0·4 19·0 0·6 18·8 0·9 19·0 0·5 19·5 0·7 0·56
  Model 2 19·4 0·4 18·9 0·6 18·1 0·9 18·3 0·6 18·8 0·8 0·61
γ-Glutamyl transferase (U/l)
  n 537 220 114 287 158
  Model 1 23·7 1·2 23·1 1·9 20·9 2·6 23·2 1·7 25·0 2·3 0·54
  Model 2 24·8 1·3 23·3 1·9 20·6 2·6 22·6 1·7 24·4 2·3 0·96
Lipid biomarkers
 TAG (mmol/l)
  n 1290 556 330 779 374
  Model 1 2·66 0·04 2·77 0·06 2·91 0·08 2·80 0·05 2·86 0·07 0·03
  Model 2 2·77 0·04 2·82 0·06 2·92 0·08 2·81 0·05 2·74 0·07
 Total cholesterol:HDL-cholesterol ratio
  n 1109 471 268 651 348
  Model 1 3·74 0·04 3·75 0·06 3·91 0·08 3·93 0·05 3·91 0·08 0·03
  Model 2 3·93 0·08 3·87 0·09 4·05 0·11 4·02 0·09 3·89 0·1 0·89
 HDL-cholesterol (mmol/l)
  n 1196 500 284 694 362
  Model 1 1·55 0·01 1·52 0·02 1·51 0·03 1·50 0·02 1·50 0·02 0·05
  Model 2 1·49 0·02 1·48 0·02 1·46 0·03 1·47 0·02 1·50 0·03 0·69
 LDL-cholesterol (mmol/l)
  n 686 277 147 393 194
  Model 1 3·23 0·04 3·12 0·05 3·26 0·07 3·37 0·05 3·19 0·06 0·73
  Model 2 3·24 0·05 3·11 0·06 3·24 0·08 3·30 0·06 3·14 0·07 0·57
 Total cholesterol (mmol/l)
  n 1547 654 391 947 495
  Model 1 5·51 0·03 5·47 0·04 5·54 0·05 5·52 0·03 5·56 0·04 0·26
  Model 2 5·55 0·03 5·49 0·04 5·55 0·05 5·53 0·04 5·55 0·05 0·78
Inflammatory biomarkers
 C-reactive protein (mg/l)
  n 2248 994 550 1347 748
  Model 1 1·62 0·04 1·64 0·05 1·62 0·07 1·72 0·05 2·23 0·09 < 0·0001
  Model 2 1·79 0·05 1·7 0·06 1·65 0·07 1·72 0·05 1·93 0·08 0·04
 Intracellular adhesion molecule 1 (ng/ml)
  n 892 373 224 495 287
  Model 1 230·6 1·9 228·5 2·9 229·2 3·7 225·7 2·5 233·7 3·4 0·47
  Model 2 238·9 2·1 239·3 3·1 241 0 3·9 236·3 2·7 239·6 3·5 1·00
 Vascular cell adhesion protein 1 (ng/ml)
  n 466 186 114 263 176
  Model 1 595·4 6·6 594·4 10·2 598·1 13·1 591·8 8·6 605·6 11·0 0·45
  Model 2 592·0 7·0 592·2 10·3 592·3 13·1 590·4 8·6 600·5 11·0 0·53
Metabolic biomarkers
 Adiponectin (μg/ml)
  n 2248 986 560 1406 808
  Model 1 10·7 0·1 11·0 0·2 10·9 0·2 11·0 0·1 10·4 0·2 0·15
  Model 2 10·1 0·1 10·7 0·2 10·6 0·2 10·8 0·2 10·8 0·2 0·01
 Insulin (μU/ml)
  n 1042 440 281 624 338
  Model 1 4·53 0·10 5·04 0·17 5·27 0·22 5·16 0·15 5·47 0·22 0·0003
  Model 2 4·86 0·12 5·02 0·17 4·87 0·2 4·89 0·15 4·86 0·19 0·86
 HbA1C (%)
  n 1708 699 406 985 560
  Model 1 5·39 0·01 5·38 0·01 5·38 0·02 5·37 0·01 5·41 0·01 0·15
  Model 2 5·42 0·01 5·41 0·01 5·41 0·02 5·39 0·01 5·42 0·01 0·55
*

Biomarker sample sizes vary: fetuin-A (n 1409), alanine transaminase (n 1318), γ-glutamyl transferase (n 1316), TAG (n 3329), total cholesterol:HDL-cholesterol ratio (n 2847), HDL-cholesterol (n 3036), LDL-cholesterol (n 1697), total cholesterol (n 4034), C-reactive protein (n 5887), intracellular adhesion molecule 1 (n 2271), vascular cell adhesion protein 1 (n 1205), adiponectin (n 6008), insulin (n 2725), HbA1c (n 4358). Values were determined using general linear models. Model 1 was adjusted for age at blood draw, fasting status (yes/no) and time of blood draw (blood collection cycle 1/2). Model 2 was adjusted for the variables in model 1 plus smoking status (current/former/never), alcohol intake (continuous), physical activity (tertile), total energy intake (tertile), Alternate Healthy Eating Index score excluding sugar-sweetened beverages (tertile), postmenopausal hormone use (yes/no) and BMI (continuous).