Intake of whole grain foods and risk of coronary heart disease in US men and women.

BACKGROUND: Epidemiological studies have demonstrated a favorable association of whole grain intake with coronary heart disease (CHD) risk, although whether such an inverse association holds true for individual whole grain foods that have various nutritional profiles has not been examined. METHODS: We followed 74,244 women from Nurses' Health Study since 1986, 91,430 women from Nurses' Health Study II since 1991, and 39,455 men from the Health Professionals Follow-Up Study since 1984, who did not have a history of cardiovascular disease or cancer at baseline. Intake of seven individual whole grain foods was repeatedly assessed using a validated semi-quantitative food frequency questionnaire every 2-4 years since baseline. CHD diagnoses were ascertained through review of medical records or death certificates. RESULTS: We documented 9461 CHD cases during an average of 25.8 years' follow-up. In the multivariable-adjusted model, the pooled hazard ratio (HR) (95% CI) of CHD risk corresponding to each one serving/day consumption of total whole grains was 0.93 (0.90-0.95; p trend <0.0001). Higher consumption of most individual whole grain foods was associated with significantly lower risk of CHD. Comparing participants consuming ≥1 serving/day with those consuming < 1 serving/month, the multivariable-adjusted pooled HRs (95% CIs) of CHD were 0.83 (0.78-0.89) for whole grain cold breakfast cereal, 0.92 (0.86-0.99) for dark bread, and 1.08 (0.96-1.22) for popcorn. For other whole grain foods with lower overall intake levels, comparing intake level of ≥2 servings/week with < 1 serving/month, the pooled hazard ratios (95% CIs) were 0.79 (0.74-0.84) for oatmeal, 0.79 (0.71-0.87) for brown rice, 0.84 (0.78-0.90) for added bran, and 0.87 (0.77-0.99) for wheat germ. Cubic spline regression suggested non-linear associations for certain whole grain foods: the risk reduction plateaued approximately over 2 servings/day for total whole grains, 0.5 serving/day for both cold breakfast cereal and dark bread, 0.5 serving/week for oatmeal, 1 serving/week for brown rice, and 2 serving/week for added bran (p for non-linearity <0.01 for all associations). CONCLUSIONS: These data suggest that higher consumption of total whole grains, as well as individual whole grain foods except popcorn, were significantly associated with lower CHD risk. The inverse associations may plateau at various intake levels for total whole grain and individual whole grain foods. This study provides further evidence in support of increasing whole grain intake for the prevention of CHD in US populations.

Urinary concentrations of phthalate biomarkers and weight change among postmenopausal women: a prospective cohort study.

BACKGROUND: Some phthalates are endocrine disrupting chemicals used as plasticizers in consumer products, and have been associated with obesity in cross-sectional studies, yet prospective evaluations of weight change are lacking. Our objective was to evaluate associations between phthalate biomarker concentrations and weight and weight change among postmenopausal women. METHODS: We performed cross-sectional (N = 997) and longitudinal analyses (N = 660) among postmenopausal Women's Health Initiative participants. We measured 13 phthalate metabolites and creatinine in spot urine samples provided at baseline. Participants' weight and height measured at in-person clinic visits at baseline, year 3, and year 6 were used to calculate body mass index (BMI). We fit multivariable multinomial logistic regression models to explore cross-sectional associations between each phthalate biomarker and baseline BMI category. We evaluated longitudinal associations between each biomarker and weight change using mixed effects linear regression models. RESULTS: In cross-sectional analyses, urinary concentrations of some biomarkers were positively associated with obesity prevalence (e.g. sum of di (2-ethylhexyl) phthalate metabolites [ΣDEHP] 4th vs 1st quartile OR = 3.29, 95% CI 1.80-6.03 [p trend< 0.001] vs normal). In longitudinal analyses, positive trends with weight gain between baseline and year 3 were observed for mono-(2-ethyl-5-oxohexyl) phthalate, monoethyl phthalate (MEP), mono-hydroxybutyl phthalate, and mono-hydroxyisobutyl phthalate (e.g. + 2.32 kg [95% CI 0.93-3.72] for 4th vs 1st quartile of MEP; p trend < 0.001). No statistically significant associations were observed between biomarkers and weight gain over 6 years. CONCLUSIONS: Certain phthalates may contribute to short-term weight gain among postmenopausal women.

Relation of statin use with non-melanoma skin cancer: prospective results from the Women's Health Initiative.

BACKGROUND: The relationship between statin use and non-melanoma skin cancer (NMSC) is unclear with conflicting findings in literature. Data from the Women's Health Initiative (WHI) Observational Study and WHI Clinical Trial were used to investigate the prospective relationship between statin use and NMSC in non-Hispanic white (NHW) postmenopausal women. METHODS: The WHI study enrolled women aged 50-79 years at 40 US centres. Among 133,541 NHW participants, 118,357 with no cancer history at baseline and complete medication/covariate data comprised the analytic cohort. The association of statin use (baseline, overall as a time-varying variable, duration, type, potency, lipophilicity) and NMSC incidence was determined using random-effects logistic regression models. RESULTS: Over a mean of 10.5 years of follow-up, we identified 11,555 NMSC cases. Compared with participants with no statin use, use of any statin at baseline was associated with significantly increased NMSC incidence (adjusted odds ratio (ORadj) 1.21; 95% confidence interval (CI): 1.07-1.35)). In particular, lovastatin (OR 1.52; 95% CI: 1.08-2.16), simvastatin (OR 1.38; 95% CI: 1.12-1.69), and lipophilic statins (OR 1.39; 95% CI: 1.18-1.64) were associated with higher NMSC risk. Low and high, but not medium, potency statins were associated with higher NMSC risk. No significant effect modification of the statin-NMSC relationship was found for age, BMI, smoking, solar irradiation, vitamin D use, and skin cancer history. CONCLUSIONS: Use of statins, particularly lipophilic statins, was associated with increased NMSC risk in postmenopausal white women in the WHI cohort. The lack of duration-effect relationship points to possible residual confounding. Additional prospective research should further investigate this relationship.

The association of walking pace and incident heart failure and subtypes among postmenopausal women.

BACKGROUND: To investigate the association between walking pace and the risk of heart failure (HF) and HF sub-types. METHODS: We examined associations of self-reported walking pace with risk of incident HF and HF subtypes of preserved (HFpEF) and reduced (HFrEF) ejection fractions, among 25,183 postmenopausal women, ages 50-79 years. At enrollment into the Women's Health Initiative cohort in 1993-1998, this subset of women was free of HF, cancer, or the inability to walk one block, with self-reported information on walking pace and walking duration. Multivariable Cox regression was used to examine associations of walking pace (casual <2 mph [referent], average 2-3 mph, and fast >3 mph) with incident HF. We also examined the joint association of walking pace and duration with incident HF. RESULTS: There were 1455 incident adjudicated acute decompensated HF hospitalization cases during a median of 16.9 years of follow-up. There was a strong inverse association between walking pace and overall risk of HF (HR = 0.73, 95% CI [0.65, 0.83] for average vs. casual walking; HR = 0.66, 95%CI [0.56, 0.78] for fast vs. casual walking). There were similar associations of walking pace with HFpEF (HR = 0.73, 95%CI [0.62, 0.86] average vs. casual; HR = 0.63, 95%CI [0.50, 0.80] for fast vs. casual) and with HFrEF (HR = 0.72, 95%CI [0.57, 0.91] for average vs. casual; HR = 0.74, 95%CI [0.54, 0.99] for fast vs. casual). The risk of HF associated with fast walking with less than 1 h/week walking duration was comparable with the risk of HF among casual and average walkers with more than 2 h/week walking duration. CONCLUSION: Walking pace was inversely associated with risks of overall HF, HFpEF, and HFrEF in postmenopausal women. Whether interventions to increase the walking pace in older adults will reduce HF risk and whether fast pace will compensate for the short duration of walking warrants further study.