Table 4.
Relative odds of weight gain (≥10 % from baseline weight vs. <10 %) by baseline weight status and quintile (Q) of adherence to a low-fat, reduced-carbohydrate, Mediterranean-style or Dietary Guidelines for Americans (DGA) diet pattern in postmenopausal women who participated in the Women’s Health Initiative Observational Study* (Odds ratios and 95 % confidence intervals)
| Normal weight | Overweight | Obese class I | Obese class II | Obese class III | ||||||
|---|---|---|---|---|---|---|---|---|---|---|
|
|
|
|
|
|
||||||
| OR | 95 % CI | OR | 95 % CI | OR | 95 % CI | OR | 95 % CI | OR | 95 % CI | |
| Low-fat diet | ||||||||||
| Q1 (high adherence) | 1·28 | 1·13, 1·46 | 1·60 | 1·40, 1·83 | 1·73 | 1·43, 2·09 | 1·44 | 1·08, 1·92 | 1·09 | 0·72, 1·65 |
| Q2 | 1·02 | 0·90, 1·15 | 1·22 | 1·07, 1·39 | 1·34 | 1·12, 1·59 | 1·35 | 1·04, 1·75 | 0·93 | 0·64, 1·35 |
| Q3 | 0·94 | 0·82, 1·07 | 1·19 | 1·04, 1·36 | 1·23 | 1·02, 1·48 | 1·07 | 0·81, 1·40 | 0·81 | 0·56, 1·17 |
| Q5 | 0·89 | 0·78, 1·02 | 1·09 | 0·95, 1·25 | 1·10 | 0·92, 1·31 | 1·06 | 0·82, 1·39 | 0·76 | 0·53, 1·09 |
| Q5 (low adherence) | 1·00 | Ref. | 1·00 | Ref. | 1·00 | Ref. | 1·00 | Ref. | 1·00 | Ref. |
| Ptrend† | <0·001 | <0·001 | <0·001 | 0·005 | 0·727 | |||||
| Reduced-carbohydrate diet | ||||||||||
| Q1 (high adherence) | 0·72 | 0·63, 0·81 | 0·67 | 0·59, 0·76 | 0·63 | 0·53, 0·76 | 0·79 | 0·60, 1·05 | 0·97 | 0·65, 1·46 |
| Q2 | 0·72 | 0·65, 0·81 | 0·73 | 0·65, 0·82 | 0·59 | 0·49, 0·70 | 0·83 | 0·63, 1·11 | 0·65 | 0·43, 0·99 |
| Q3 | 0·75 | 0·68, 0·83 | 0·74 | 0·66, 0·82 | 0·78 | 0·67, 0·91 | 1·05 | 0·81, 1·37 | 0·99 | 0·66, 1·49 |
| Q5 | 0·82 | 0·76, 0·89 | 0·81 | 0·74, 0·89 | 0·86 | 0·75, 0·99 | 1·05 | 0·83, 1·33 | 0·98 | 0·67, 1·43 |
| Q5 (low adherence) | 1·00 | Ref. | 1·00 | Ref. | 1·00 | Ref. | 1·00 | Ref. | 1·00 | Ref. |
| Ptrend† | <0·001 | <0·001 | <0·001 | 0·040 | 0·416 | |||||
| Mediterranean-style diet | ||||||||||
| Q1 (low adherence) | 1·00 | Ref. | 1·00 | Ref. | 1·00 | Ref. | 1·00 | Ref. | 1·00 | Ref. |
| Q2 | 0·91 | 0·83, 0·99 | 1·02 | 0·93, 1·12 | 1·06 | 0·93, 1·22 | 1·16 | 0·94, 1·43 | 0·79 | 0·57, 1·10 |
| Q3 | 0·99 | 0·90, 1·08 | 0·98 | 0·89, 1·08 | 1·05 | 0·90, 1·21 | 1·11 | 0·88, 1·41 | 1·19 | 0·85, 1·67 |
| Q4 | 0·93 | 0·84, 1·03 | 1·05 | 0·94, 1·18 | 1·09 | 0·92, 1·29 | 0·92 | 0·69, 1·25 | 1·12 | 0·73, 1·72 |
| Q5 (high adherence) | 0·90 | 0·80, 1·01 | 0·95 | 0·83, 1·09 | 1·15 | 0·93, 1·42 | 0·86 | 0·60, 1·23 | 1·08 | 0·61, 1·90 |
| Ptrend† | 0·159 | 0·831 | 0·206 | 0·531 | 0·403 | |||||
| DGA diet | ||||||||||
| Q1 (low adherence) | 1·00 | Ref. | 1·00 | Ref. | 1·00 | Ref. | 1·00 | Ref. | 1·00 | Ref. |
| Q2 | 0·99 | 0·88, 1·12 | 1·06 | 0·94, 1·20 | 1·01 | 0·86, 1·19 | 1·09 | 0·86, 1·38 | 1·04 | 0·75, 1·46 |
| Q3 | 1·00 | 0·89, 1·13 | 1·13 | 1·00, 1·28 | 1·04 | 0·88, 1·23 | 1·09 | 0·84, 1·41 | 1·65 | 1·15, 2·36 |
| Q4 | 1·10 | 0·98, 1·24 | 1·15 | 1·02, 1·30 | 1·28 | 1·08, 1·52 | 1·12 | 0·84, 1·49 | 1·09 | 0·73, 1·65 |
| Q5 (high adherence) | 1·13 | 1·00, 1·28 | 1·30 | 1·15, 1·48 | 1·41 | 1·17, 1·70 | 1·33 | 0·99, 1·80 | 1·86 | 1·18, 2·95 |
| Ptrend† | 0·005 | <0·001 | <0·001 | 0·096 | 0·019 | |||||
Ref., referent values.
Adjusted models included age (continuous) at baseline, alcohol intake at baseline, race/ethnicity, annual family income, physical activity at baseline (continuous), smoking status at baseline and energy intake (continuous) at baseline. All non-continuous variables were modeled using disjoint indicator variables.
Ptrend corresponds to a Wald test statistic when a linear term for quintile of diet pattern was substituted in the model.