Table 8.
Weight change over four year periods according to four year change in chocolate subtype intake, stratified by baseline BMI
| No of participants | Increase | Decrease | P for interaction‡ | ||||||
|---|---|---|---|---|---|---|---|---|---|
| β* | SE (95% CI) | P value† | β* | SE (95% CI) | P value† | ||||
| Dark chocolate | |||||||||
| BMI: | |||||||||
| <25 | 39 108 | −0.06 | 0.04 (−0.14 to 0.01) | 0.10 | 0.06 | 0.04 (−0.02 to 0.15) | 0.14 | <0.001 | |
| 25-30 | 29 007 | −0.13 | 0.06 (−0.25 to −0.01) | 0.03 | −0.10 | 0.06 (−0.22 to 0.02) | 0.11 | ||
| ≥30 | 17 690 | −0.07 | 0.14 (−0.34 to 0.19) | 0.60 | −0.45 | 0.14 (−0.72 to −0.17) | 0.001 | ||
| Milk chocolate | |||||||||
| BMI: | |||||||||
| <25 | 39 108 | 0.33 | 0.04 (0.24 to 0.42) | <0.001 | −0.05 | 0.04 (−0.12 to 0.02) | <0.001 | <0.001 | |
| 25-30 | 29 007 | 0.27 | 0.06 (0.14 to 0.39) | <0.001 | −0.37 | 0.06 (−0.49 to −0.26) | <0.001 | ||
| ≥30 | 17 690 | 0.68 | 0.14 (0.42 to 0.95) | <0.001 | −0.85 | 0.13 (−1.10 to −0.59) | <0.001 | ||
Multivariable generalized linear regression models (with independent correlation matrix and robust variance) were used.
All models adjusted for age, ethnicity (white, African American, Asian, and others), baseline BMI (continuous), family history of diabetes, baseline hypertension, hypercholesterolemia, baseline and change in total energy intake, change in smoking status (stayed never smoker, stayed former smoker, stayed current smoker, changed from former to current smoker, changed from never to current smoker, and changed from current to former smoker), baseline and change in physical activity (MET-h/week), change in alcohol consumption (continuous, g/day), postmenopausal hormone use (women only; premenopausal, never, former, current, or missing), oral contraceptive use (NHSII only; yes, no), changes in AHEI (five groups), and study origin (NHS, NHSII, HPFS).
AHEI=Alternate Healthy Eating Index; BMI=body mass index; CI=confidence interval; NHS=Nurses’ Health Study; NHSII=Nurses’ Health Study II; HPFS=Health Professionals Follow-up Study; MET-h=metabolic equivalent tasks per hour; SE=standard error.
Interpreted as four year weight change in kilograms (95% CI) comparing participants with increased or decreased total chocolate intake to those with no change in total chocolate intake over four year periods. Data from three cohorts were combined to run this analysis.
Calculated using Wald test.
Calculated using generalized score tests.