Table 3.
Associations between quartiles of UPF consumption (in servings/d) and obesity indicators by diabetes status in the CACTI cohort followed up for over 14 years *
| T1DM | Non-diabetic controls | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Q2 | Q3 | Q4 | Q2 | Q3 | Q4 | |||||||
| β | se | β | se | β | se | β | se | β | se | β | se | |
| Weight (kg) | ||||||||||||
| Model 1 | 0·61 | 0·97 | 1·44 | 1·20 | 4·19 | 1·23 | 0·23 | 0·81 | 2·32 | 0·89 | 3·74 | 1·22 |
| Model 2 | 0·64 | 0·97 | 1·46 | 1·22 | 3·08 | 1·20 | −0·17 | 0·75 | 1·32 | 0·82 | 3·13 | 1·21 |
| Model 3 | 0·65 | 0·97 | 1·47 | 1·22 | 3·11 | 1·21 | −0·18 | 0·74 | 1·30 | 0·82 | 3·08 | 1·27 |
| Model 4 | 0·36 | 1·01 | 1·21 | 1·27 | 3·07 | 1·27 | −0·08 | 0·76 | 1·13 | 0·80 | 3·36 | 1·27 |
| WC (cm) | ||||||||||||
| Model 1 | 1·42 | 1·25 | 2·63 | 1·44 | 4·15 | 1·45 | 2·04 | 1·08 | 2·64 | 1·15 | 5·04 | 1·20 |
| Model 2 | 2·06 | 1·13 | 2·38 | 1·37 | 3·37 | 1·40 | 1·30 | 0·93 | 1·61 | 1·09 | 3·30 | 1·14 |
| Model 3 | 2·13 | 1·13 | 2·48 | 1·37 | 3·56 | 1·43 | 1·39 | 0·93 | 1·33 | 1·10 | 3·68 | 1·21 |
| Model 4 | 2·18 | 1·23 | 2·04 | 1·49 | 2·89 | 1·59 | 1·46 | 0·93 | 1·12 | 1·07 | 3·80 | 1·19 |
| BMI (kg/m2) | ||||||||||||
| Model 1 | 0·35 | 0·34 | 0·43 | 0·40 | 1·07 | 0·42 | 0·11 | 0·28 | 0·77 | 0·33 | 1·17 | 0·41 |
| Model 2 | 0·43 | 0·32 | 0·71 | 0·40 | 0·97 | 0·38 | 0·02 | 0·26 | 0·59 | 0·30 | 1·04 | 0·41 |
| Model 3 | 0·45 | 0·32 | 0·73 | 0·40 | 1·02 | 0·39 | 0·03 | 0·26 | 0·60 | 0·30 | 1·07 | 0·43 |
| Model 4 | 0·42 | 0·33 | 0·71 | 0·40 | 1·02 | 0·40 | 0·07 | 0·26 | 0·54 | 0·30 | 1·15 | 0·43 |
| Overweight | ||||||||||||
| Model 1 | 0·07 | 0·25 | −0·16 | 0·23 | 0·11 | 0·23 | 0·06 | 0·18 | 0·17 | 0·16 | 0·65 | 0·21 |
| Model 2 | 0·13 | 0·25 | −0·17 | 0·25 | 0·01 | 0·24 | 0·12 | 0·19 | 0·18 | 0·19 | 0·62 | 0·23 |
| Model 3 | 0·17 | 0·27 | −0·13 | 0·27 | 0·02 | 0·26 | 0·13 | 0·19 | 0·21 | 0·19 | 0·69 | 0·24 |
| Model 4 | 0·19 | 0·27 | −0·14 | 0·27 | 0·05 | 0·27 | 0·12 | 0·19 | 0·22 | 0·19 | 0·70 | 0·25 |
| Obesity | ||||||||||||
| Model 1 | 0·14 | 0·33 | 0·64 | 0·31 | 0·44 | 0·28 | 0·04 | 0·22 | 0·29 | 0·21 | 0·27 | 0·26 |
| Model 2 | 0·25 | 0·36 | 0·71 | 0·33 | 0·40 | 0·31 | −0·03 | 0·24 | 0·27 | 0·22 | 0·25 | 0·26 |
| Model 3 | 0·28 | 0·36 | 0·75 | 0·34 | 0·49 | 0·35 | −0·04 | 0·24 | 0·24 | 0·22 | 0·18 | 0·27 |
| Model 4 | 0·32 | 0·36 | 0·66 | 0·35 | 0·34 | 0·36 | 0·01 | 0·25 | 0·27 | 0·23 | 0·21 | 0·28 |
UPF: ultra-processed food; Q: quartiles; T1DM: type 1 diabetes mellitus; WC: waist circumferences.
Generalised estimating equations were used to measure parameter estimates and se of obesity indicators. Q1 is reference category. Model 1 was unadjusted model. Model 2 adjusted for sex, age, race, education, smoking status, physical activity, and duration of follow-up. Model 3 is model 2 further adjusted for total energy intake. Model 4 is model 3 further adjusted for diabetes duration, antihypertensive and lipid-lowering drugs for T1DM; antihypertensive and lipid-lowering drugs for non-diabetic control.
Significant difference at P < 0·05 (bold).