Multivariate Relative Risk Research Articles

Alcohol consumption and breast cancer risk by family history of breast cancer and folate intake: A prospective cohort study

ObjectivesTo evaluate the association between alcohol consumption and breast cancer risk by family history of breast cancer and folate intake.MethodsWe followed 90,433 premenopausal women aged 26 to 46 years in the Nurses’ Health Study II who provided alcohol intake data at baseline of 1991. Cumulative alcohol and folate intake was measured by a food frequency questionnaire, which was repeated every 4 years.ResultsDuring 18 years of follow‐up, we documented 2,420 incident cases of invasive breast cancer. Alcohol consumption was not significantly associated with breast cancer risk overall. However, when the association was stratified by family history of breast cancer, alcohol consumption was associated with risk of breast cancer among those with a family history of breast cancer in a 1st degree relative. Compared with non‐drinkers, the multivariate relative risks (RR) of breast cancer for alcohol consumption of ≥ 10g/day was 1.54 (95% confidence interval [CI], 1.09–2.16) among those with family history and 1.01 (95% CI, 0.86–1.18) among those without family history. Among those with positive family history, the positive association was largely limited to those with low folate intake (<400mcg/d) (RR, 2.02; 95% CI, 1.13–3.60 for alcohol consumption ≥ 10g/day vs. non‐drinkers).ConclusionsIn this population of largely premenopausal women, alcohol consumption was associated with an increased breast cancer risk among those with family history of breast cancer, especially when folate intake was low.

Intake of specific fruits and vegetables in relation to risk of estrogen receptor-negative breast cancer among postmenopausal women

In previous studies of postmenopausal women, overall intake of fruits and vegetables groups has been inversely associated with estrogen receptor-negative (ER-) breast cancer. In this analysis, we prospectively examined the associations of specific fruits and vegetables with risk of ER- postmenopausal breast cancer among 75,929 women aged 38-63 years at baseline and followed for up to 24 years. Dietary data were collected seven times during this period. Cox proportional hazard models were used, adjusting for potential confounders, including a modified Alternate Mediterranean Diet score. We ascertained 792 incident cases of ER- postmenopausal breast cancer. The multivariate relative risk (RR) for every 2 servings/week consumption for total berries was 0.82 (95 % CI = 0.71-0.96, p = 0.01), and the RR for women who consumed at least one serving of blueberries a week was 0.69 (95 % CI = 0.50-0.95, p = 0.02) compared with non-consumers. Also, the RR for consuming at least 2 servings of peaches/nectarines per week was 0.59 (95 % CI = 0.37-0.93, p = 0.02). Risk of ER- breast cancer was not associated with intakes of other specific fruits or vegetables. In conclusion, higher intake of berries and peaches was associated with lower risk of ER- breast cancer among postmenopausal women. These results are considered exploratory and need to be confirmed in further studies.

Glycaemic index, carbohydrate substitution for fat and risk of CHD in men

We have examined the associations between dietary glycaemic index (GI), substitutions of total, low-, medium- and high-GI carbohydrates for fat and the risk of CHD. The study consisted of 21 955 male smokers, aged 50-69 years, within the Alpha-Tocopherol, Beta-Carotene Cancer Prevention Study. The diet was assessed at baseline using a validated FFQ. During a 19-year follow-up, 4379 CHD cases (2377 non-fatal myocardial infarctions and 2002 CHD deaths) were identified from national registers. Relative risks (RR) and CI for CHD were analysed using Cox proportional hazards modelling, and multivariate nutrient density models were applied to examine the associations between the substitutions of macronutrients and the risk of CHD. Dietary GI was inversely associated with CHD risk: multivariate RR in the highest v. lowest quintile was 0·89 (95 % CI 0·81, 0·99). Replacement of higher-GI carbohydrates with lower-GI carbohydrates did not associate with the risk. Replacing saturated and trans-fatty acids with carbohydrates was associated with decreased CHD risk: RR for substitution of 2 % of energy intake was 0·97 (95 % CI 0·94, 0·99). On the contrary, replacing MUFA with carbohydrates was associated with an increased risk: RR for substitution of 2 % of energy intake was 1·08 (95 % CI 1·01, 1·16). We conclude that in the present study population, contrary to the hypothesis, a lower GI does not associate with a decreased risk of CHD. The associations of carbohydrates with CHD risk depend on the fatty acid composition of the diet.

Dietary Calcium from Dairy and Nondairy Sources, and Risk of Symptomatic Kidney Stones

Because of high correlations between dairy intake and total dietary calcium, previously reported associations between lower calcium intake and increased kidney stone risk represent de facto associations between milk products and risk. We examined associations between dietary calcium from nondairy and dairy sources, and symptomatic nephrolithiasis. We performed prospective studies in the Health Professionals Follow-up Study (HPFS) in 30,762 men, and in the Nurses' Health Study (NHS) I and II in 94,164 and 101,701 women, respectively. We excluded men 60 years old or older because we previously reported inverse associations between calcium intake and risk only in men younger than 60 years. Food frequency questionnaires were used to assess calcium intake every 4 years. We used Cox proportional hazards regression to adjust for age, body mass index, supplemental calcium, diet and other factors. We documented 5,270 incident kidney stones during the combined 56 years of followup. In participants in the highest vs the lowest quintile of nondairy dietary calcium the multivariate relative risk of kidney stones was 0.71 (95% CI 0.56-0.92, p for trend 0.007) in HPFS, 0.82 (95% CI 0.69-0.98, p trend 0.08) in NHS I and 0.74 (95% CI 0.63-0.87, p trend 0.002) in NHS II. When comparing the highest to the lowest quintile of dairy calcium, the multivariate relative risk was 0.77 (95% CI 0.63-0.95, p trend 0.01) for HPFS, 0.83 (95% CI 0.69-0.99, p trend 0.05) for NHS I and 0.76 (95% CI 0.65-0.88, p trend 0.001) for NHS II. Higher dietary calcium from nondairy or dairy sources is independently associated with a lower kidney stone risk.

Carotid Artery Plaque and Progression of Coronary Artery Calcium: The Multi-Ethnic Study of Atherosclerosis

Carotid and coronary atherosclerosis are associated with each other in imaging and autopsy studies. The aim of this study was to evaluate whether carotid artery plaque seen on carotid ultrasound can predict incident coronary artery calcification (CAC). Agatston calcium score measurements were repeated in 5,445 participants of the Multi-Ethnic Study of Atherosclerosis (MESA; mean age, 57.9 years; 62.9% women). Internal carotid artery lesions were graded as 0%, 1% to 24%, or >25% diameter narrowing, and intima-media thickness (IMT) was measured. Plaque was present for any stenosis >0%. CAC progression was evaluated with multivariate relative risk regression for CAC scores of 0 at baseline and with multivariate linear regression for CAC score > 0, adjusting for cardiovascular risk factors, body mass index, ethnicity, and common carotid IMT. CAC was positive at baseline in 2,708 of 5,445 participants (49.7%) and became positive in 458 of 2,837 (16.1%) at a mean interval of 2.4 years between repeat examinations. Plaque and internal carotid artery IMT were both strongly associated with the presence of CAC. After statistical adjustment, the presence of carotid artery plaque significantly predicted incident CAC with a relative risk of 1.37 (95% confidence interval, 1.12-1.67). Incident CAC was associated with internal carotid artery IMT, with a relative risk of 1.13 (95% confidence interval, 1.03-1.25) for each 1-mm increase. Progression of CAC was also significantly associated (P < .001) with plaque and internal carotid artery IMT. In individuals free of cardiovascular disease, subjective and quantitative measures of carotid artery plaques by ultrasound imaging are associated with CAC incidence and progression.

The effect of lipid restriction on the prevention of parenteral nutrition-associated cholestasis in surgical infants

PurposeSurgical infants requiring long-term parenteral nutrition (PN) are at risk for parenteral nutrition-associated liver disease (PNALD). The purpose of this study was to determine the effect of a lipid restricted PN regimen in preventing the development of PNALD in surgical infants. MethodsIn 2009, we implemented a lipid restricted strategy in surgical infants expected to be on long-term PN using a soy-based lipid emulsion at a goal provision of 1g/kg/day throughout a patient's entire PN course. An experimental cohort of surgical infants treated with lipid restriction from 2009 to 2011 (n=82) was retrospectively compared to a control cohort of infants from 2005 to 2008 receiving standard intravenous lipid dosing (n=132). A multivariable relative risk regression model was constructed analyzing the association between lipid restriction and PNALD. ResultsPatients admitted during the lipid restriction era had reduced daily lipid provisions compared to the control group (p<0.001). There were no significant differences in demographic or measured clinical characteristics between the two groups. A significant reduction in the incidence of PNALD was demonstrated in the lipid restricted group compared to the control group (22% vs. 43%, p=0.002). On multivariable relative risk regression, patients treated with standard lipid provisions were 1.77 times more likely to develop PNALD than patients who were lipid restricted (95% CI: 1.2–2.7; p=0.007). ConclusionRestriction of intravenous soy-based lipid in PN-fed surgical infants is associated with a reduction in the incidence of liver disease. Early lipid restriction should be considered in all surgical infants who require PN as a preventative measure against PNALD.

Fruit and Vegetable Intake and Risk of Breast Cancer by Hormone Receptor Status

Estrogen receptor-negative (ER(-)) breast cancer has few known or modifiable risk factors. Because ER(-) tumors account for only 15% to 20% of breast cancers, large pooled analyses are necessary to evaluate precisely the suspected inverse association between fruit and vegetable intake and risk of ER(-) breast cancer. Among 993 466 women followed for 11 to 20 years in 20 cohort studies, we documented 19 869 estrogen receptor positive (ER(+)) and 4821 ER(-) breast cancers. We calculated study-specific multivariable relative risks (RRs) and 95% confidence intervals (CIs) using Cox proportional hazards regression analyses and then combined them using a random-effects model. All statistical tests were two-sided. Total fruit and vegetable intake was statistically significantly inversely associated with risk of ER(-) breast cancer but not with risk of breast cancer overall or of ER(+) tumors. The inverse association for ER(-) tumors was observed primarily for vegetable consumption. The pooled relative risks comparing the highest vs lowest quintile of total vegetable consumption were 0.82 (95% CI = 0.74 to 0.90) for ER(-) breast cancer and 1.04 (95% CI = 0.97 to 1.11) for ER(+) breast cancer (P (common-effects) by ER status < .001). Total fruit consumption was non-statistically significantly associated with risk of ER(-) breast cancer (pooled multivariable RR comparing the highest vs lowest quintile = 0.94, 95% CI = 0.85 to 1.04). We observed no association between total fruit and vegetable intake and risk of overall breast cancer. However, vegetable consumption was inversely associated with risk of ER(-) breast cancer in our large pooled analyses.

Body mass index, height and risk of lymphoid neoplasms in a large United States cohort

Results from epidemiologic studies examining associations between body size and risk of non-Hodgkin lymphoma (NHL) are inconsistent, and etiology may vary by histologic subtype of disease. Using Cox proportional hazards regression, multivariable relative risks (RRs) and 95% confidence intervals (CIs) were computed for associations of body mass index (BMI) and height with NHL in the prospective American Cancer Society Cancer Prevention Study-II Nutrition Cohort. From 1992 to 2007, 2074 incident cases of NHL were identified among 152 423 men and women. Obese individuals (BMI ≥ 30 kg/m2) had 23% higher incidence of NHL (95% CI 1.08–1.40) compared to those with normal weight (BMI 18.5–< 25 kg/m2). Height was positively associated with NHL (RR = 1.25, 95% CI 1.10–1.43, sex-specific quintile 5 vs. 1). BMI associations were strongest for diffuse large B-cell lymphoma. Height was most strongly associated with chronic lymphocytic leukemia/small lymphocytic lymphoma and to a lesser extent with multiple myeloma. These findings provide further evidence that body size may play a role in the etiology of NHL, which is of public health importance given the rapid rise in obesity worldwide.

Consumption of dairy and meat in relation to breast cancer risk in the Black Women’s Health Study

Dairy and meat consumption may impact breast cancer risk through modification of hormones (e.g., estrogen), through specific nutrients (e.g., vitamin D), or through products formed in processing/cooking (e.g., heterocyclic amines). Results relating meat and dairy intake to breast cancer risk have been conflicting. Thus, we examined the risk of breast cancer in relation to intake of dairy and meat in a large prospective cohort study. In the Black Women's Health Study, 1,268 incident breast cancer cases were identified among 52,062 women during 12 years of follow-up. Multivariable (MV) relative risks (RRs) and 95 % confidence intervals (CIs) were calculated using Cox proportional hazards models. Null associations were observed for total milk (MV RR = 1.05, 95 % CI 0.74-1.46 comparing ≥1,000-0 g/week) and total meat (MV RR = 1.04, 95 % CI 0.85-1.28 comparing ≥1,000 < 400 g/week) intake and risk of breast cancer. Associations with intakes of specific types of dairy, specific types of meat, and dietary calcium and vitamin D were also null. The associations were not modified by reproductive (e.g., parity) or lifestyle factors (e.g., smoking). Associations with estrogen receptor (ER) positive (+), ER negative (-), progesterone receptor (PR) +, PR-, ER+/PR+, and ER-/PR- breast cancer were generally null. This analysis of African-American women provides little support for associations of dairy and meat intake with breast cancer risk.

Total white blood cell count or neutrophil count predict ischemic stroke events among adult Taiwanese: report from a community-based cohort study

BackgroundEvidence about whether white blood cell (WBC) or its subtypes can act as a biomarker to predict the ischemic stroke events in the general population is scanty, particularly in Asian populations. The aim of this study is to establish the predictive ability of total WBC count or subtypes for long-term ischemic stroke events in the cohort population in Taiwan.MethodsThe Chin-Shan Community Cohort Study began from 1990 to 2007 by recruiting 1782 men and 1814 women of Chinese ethnicity. Following a total of 3416 participants free from ischemic stroke events at baseline for a median of 15.9 years; we documented 187 new incident cases.ResultsThe multivariate relative risk for the comparison of the participants in the fifth and first WBC count quintiles was 1.67 (95% confidence interval [CI], 1.02–2.73; P for trend=0.03), and the corresponding relative risk for neutrophil count was 1.93 (95% CI, 1.13–3.29; P for trend=0.02). The discriminative ability by WBC and neutrophil counts were similar (area under the receiver operating characteristic curve, 0.600 for adding WBC, 0.610 for adding neutrophils, 0.595 for traditional risk factor model). In addition, the net reclassification improvement (NRI) values between the neutrophil and white blood cell count models were not significant (NRI, =-2.60%, P=0.35), indicating the similar discrimination performance for both WBC and neutrophil counts.ConclusionsWBC and neutrophil count had a similar ability to predict the long-term ischemic stroke events among Taiwanese.

Black tea consumption and risk of stroke in women and men

PurposeOur aim was examine the association between black tea consumption and risk of total stroke and stroke types in a prospective study. MethodsA total of 74,961 Swedish women and men who were free of cardiovascular disease and cancer at baseline in 1997 were followed up through December 2008. Tea consumption was assessed with a questionnaire at baseline. Stroke cases were ascertained from the Swedish Hospital Discharge Registry. ResultsDuring a mean follow-up of 10.2 years, we ascertained 4089 cases of first stroke, including 3159 cerebral infarctions, 435 intracerebral hemorrhages, 148 subarachnoid hemorrhages, and 347 unspecified strokes. After adjustment for other risk factors, high tea consumption was associated with a significantly lower risk of total stroke; however, there was no dose–response relation (P for trend = .36). Compared with no tea consumption, the multivariable relative risk for four or more cups per day (median, 5) was 0.79 (95% confidence interval [CI], 0.62–0.998). The corresponding relative risks were 0.80 (95% CI, 0.61–1.04) for cerebral infarction and 0.68 (95% CI, 0.35–1.30) for hemorrhagic stroke. ConclusionsThese findings suggest that daily consumption of four or more cups of black tea is inversely associated with risk of stroke.

Total and specific fruit and vegetable consumption and risk of stroke: A prospective study

BackgroundFruit and vegetables is a heterogeneous food group with different content of dietary fiber, vitamins, minerals, carotenoids, and bioactive phytochemicals. Our objective was to examine the relation between specific consumption of fruit and vegetable subgroups and stroke risk in a cohort of Swedish women and men. Methods and resultsWe prospectively followed 74,961 participants (34,670 women and 40,291 men) who had completed a food frequency questionnaire in the autumn of 1997 and were free from stroke, coronary heart disease, and cancer at baseline. Diagnoses of stroke in the cohort during follow-up were ascertained from the Swedish Hospital Discharge Registry. A total of 4089 stroke cases, including 3159 cerebral infarctions, 435 intracerebral hemorrhages, 148 subarachnoid hemorrhages, and 347 unspecified strokes, were ascertained during 10.2 years of follow-up. The multivariable relative risk (RR) of total stroke for the highest vs. lowest category of total fruit and vegetable consumption was 0.87 (95% confidence interval [CI] 0.78–0.97; P for trend = 0.01). The association was confined to individuals without hypertension (corresponding RR, 0.81; 95% CI, 0.71–0.93; P for trend = 0.01). Among individual fruits and vegetable subgroups, inverse associations with total stroke were observed for apples/pears (RR, 0.89; 95% CI, 0.80–0.98; P for trend = 0.02) and green leafy vegetables (RR, 0.92; 95% CI, 0.81–1.04; P for trend = 0.03). ConclusionThis study shows an inverse association of fruit and vegetable consumption with stroke risk. Particularly consumption of apples and pears and green leafy vegetables was inversely associated with stroke.

Abstract P4-12-03: Towards a risk prediction model for breast cancer that utilizes breast tissue risk features

Abstract Background: Optimal early detection and prevention strategies for breast cancer are predicated on our ability to accurately identify women at increased risk for disease. Current breast cancer risk prediction models rely on clinical and epidemiologic features. As demonstrated with other cancers, such as those of the cervix and colon, cancer risk assessment is enhanced when the tissue at risk is examined for evidence of premalignant abnormalities. Here, we present a multivariate model for individualized breast cancer risk prediction that incorporates tissue features identified from biopsies from women with benign breast disease (BBD). Methods: We used our cohort of more than 9000 women diagnosed with BBD at the Mayo Clinic from 1967–1991, and identified a nested case:control series of 377 women who developed cancer (cases) matched with 734 women who did not (controls). We identified a set of variables that either alone or in two-way interactions contributed to a multivariable risk prediction model. We combined this relative risk model with the baseline incidence of breast cancer and the baseline hazard of death from our full BBD cohort to develop a risk prediction tool that computes absolute risk estimates for breast cancer. We calculated 5-year risk predictions for these cases and controls using our model and using the Gail model and compared c-statistics from these risk predictions. We validated our model using an independent set of 378 breast cancer cases and 728 matched controls drawn from our BBD cohort. Results: Cases and controls in the model development data set differed significantly on a number of features when examined individually. Those included in a multivariable relative risk model for breast cancer among women with BBD are shown below. After including interactions between lobular involution, and histologic impression, with age at first live birth/number of children, and degree of involution with sclerosing adenosis, we devised a model that predicts the absolute risk of breast cancer. Predicted five-year risks in the model development series, we observed a c-statistic of 0.66, higher than the c-statistic of 0.61 for the Gail model. With risk prediction estimates from our validation series of cases and matched controls, the c-statistic of our model decreased to 0.59 and the c-statistic for the Gail model was 0.55. Conclusion: We have identified an initial risk prediction model for breast cancer in women with BBD. The identified model includes basic demographic information as well as a number of histologic features of the tissue samples. Its initial performance is promising, providing improved discrimination over the Gail model in our replication data set. Further work is needed to identify features which might provide additional predictive information for breast cancer risk. Citation Information: Cancer Res 2012;72(24 Suppl):Abstract nr P4-12-03.

Calcium intake and risk of primary hyperparathyroidism in women: prospective cohort study

Objective To examine the association between calcium intake and risk of primary hyperparathyroidism in women.Design Prospective cohort study.Setting Nurses’ Health Study I, which originally recruited participants from the 11 most...

Obesity, waist circumference, weight change and the risk of psoriasis in US women

To evaluate the associations between body mass index (BMI), weight change, waist circumference, hip circumference and risk of incident psoriasis. A prospective study of female nurses who were followed up over a 12-year period (1996-2008) in Nurses' Health Study, a cohort of 121,700 US women at the inception in 1976. The study included 67,300 women who responded to a question about a history of physician-diagnosed psoriasis in last 12 years in 2008 (mean age at 1996, 62 years). The primary outcome was self-reported, physician-diagnosed psoriasis. During the 12 years of follow-up, there were a total of 809 incident psoriasis cases. There was a graded positive association between BMI (both baseline and updated) and the risk of psoriasis (both P values for trend <0.0001). Compared to women with updated BMI of <25, the multivariate relative risks (RRs) of incident psoriasis were 1.21 (95% CI, 1.03-1.43) for a BMI of 25.0-29.9, 1.63 (95% CI, 1.33-2.00) for a BMI of 30.0-34.9 and 2.03 (95% CI, 1.58-2.61) for a BMI of 35.0 or greater. Higher waist circumference, hip circumference and waist-hip ratio were associated with a higher risk of incident psoriasis, but became non-significant after additionally adjusting for BMI. The BMI at age of 18 years was not associated with the risk of psoriasis. Weight gain since the age of 18 years was associated with an increased risk of psoriasis, and RR of 10 lb gain was 1.08 (95% CI, 1.06-1.11; P < 0.0001). This large prospective study indicates that higher BMI and weight gain are risk factors for incident psoriasis in older US women.

Depression and the risk of psoriasis in US women

Depression is a common mental health condition that has been associated with psoriasis. In the absence of prospective data, it remains unclear whether depression precedes psoriasis as a risk factor. To examine the association between depression and the risk of new-onset psoriasis. A prospective cohort of 86 880 US female nurses, The Nurses' Health Study II, was followed up from 1993 to 2005. Participants reported anti-depressant use and completed the Mental Health Index (MHI), a subscale of the Short-Form 36 in 1993. The MHI assessed for depression and scores was categorized into four strata: 0-52, 53-75, 76-85 and 86-100, with lower scores associated with increasing depressive symptoms. We excluded participants with a history of psoriasis prior to 1993. A self-report of incident physician-diagnosed psoriasis constituted the main outcome measure. For a sensitivity analysis, we had a subset of confirmed psoriasis cases. Depression was associated with an increased risk of incident psoriasis. Compared to women in the non-depressed group (MHI 86-100), women who reported either having high depressive symptomatology (MHI scores < 52) or who were on anti-depressants had a multivariate relative risk (RR) of 1.59 for developing subsequent psoriasis (95% confidence interval [CI], 1.21-2.08). These associations became stronger among confirmed psoriasis cases. We found that depression was independently associated with an increased risk of psoriasis in this population of US women.

Plasma Adiponectin and the Risk of Hypertension in White and Black Postmenopausal Women

Adiponectin may have a protective role in the development of obesity-related metabolic and vascular disorders, including hypertension. We conducted a prospective, nested case control study to investigate the relation between baseline plasma adiponectin, measures of adiposity, and subsequent risk of hypertension. We selected 400 white and 400 black postmenopausal women, age <70 years, who developed incident hypertension during 5.9-year follow-up and an equal number of age- and race-matched controls in the Women's Health Initiative Observational Study. We measured plasma concentrations of total adiponectin in their baseline blood samples. In crude matched models, plasma adiponectin was inversely associated with risk of hypertension among both white and black women. The association appeared to be nonlinear in white women but dose related in black women. Adjustment for lifestyle factors, measures of obesity, and obesity-related clinical factors attenuated these associations. The multivariable relative risk (95% CI) of hypertension across increasing quartiles of plasma adiponectin were 1.00, 0.98 (0.66-1.46), 0.63 (0.41-0.97), and 0.92 (0.60-1.42) in white women (P(trend): 0.38) and 1.00, 0.96 (0.64-1.46), 0.83 (0.53-1.29), and 0.58 (0.36-0.94) in black women (P(trend): 0.02). Further adjustment for inflammatory markers and endothelial markers eliminated the association in white, but not black, women. In this prospective, nested case control study, we found an inverse association between plasma adiponectin and risk of hypertension in white and black postmenopausal women. The reduced risk of hypertension was limited to only intermediate concentrations of adiponectin in white women whereas it was graded across quartiles of adiponectin in black women.

Use of Nonsteroidal Anti-inflammatory Drugs and Risk of Ovarian and Endometrial Cancer: The Multiethnic Cohort

Chronic inflammation may play an etiologic role in ovarian and endometrial cancer, and it is hypothesized that nonsteroidal anti-inflammatory drugs (NSAID) decrease the risk of developing these malignancies. No prospective study with a large multiethnic population has explored this hypothesis. We investigated whether NSAID use was associated with risks of ovarian and endometrial cancer in the Multiethnic Cohort Study. Medication use of at least twice a week for ≥1 month was assessed at baseline. Multivariable relative risks (RR) and 95% confidence intervals (CI) were estimated using Cox proportional hazards models. During 13.3 years of follow-up, 275 ovarian and 620 endometrial incident cases were identified among approximately 64,000 women included in this analysis (16.5% African Americans, 30.8% Japanese, 7.7% Native Hawaiians, 18.9%, Latinas, and 26.0% whites). The RR (95% CI) for ovarian cancer associated with aspirin, non-aspirin NSAIDs, and acetaminophen were 0.87 (0.68-1.14), 0.97 (0.74-1.26), and 0.86 (0.67-1.12), respectively. The RR (95% CI) for endometrial cancer associated with aspirin, non-aspirin NSAIDs, and acetaminophen were 0.93 (0.79-1.10), 0.88 (0.74-1.05), and 0.96 (0.81-1.13), respectively. No heterogeneity across ethnic groups (P ≥ 0.29) or dose-response relation with increased duration of use (P(trend) ≥ 0.16) was observed. The results did not differ by tumor histology. We found no compelling evidence to support an association between the use of NSAIDs and risk of ovarian and endometrial cancers in a multiethnic population. It is unlikely that NSAID is involved in the etiology of endometrial and ovarian cancer.

Chocolate consumption and risk of stroke

To investigate the association between chocolate consumption and risk of stroke in men and conduct a meta-analysis to summarize available evidence from prospective studies of chocolate consumption and stroke. We prospectively followed 37,103 men in the Cohort of Swedish Men. Chocolate consumption was assessed at baseline using a food-frequency questionnaire. Cases of first stroke were ascertained from the Swedish Hospital Discharge Registry. For the meta-analysis, pertinent studies were identified by searching the PubMed and EMBASE databases through January 13, 2012. Study-specific results were combined using a random-effects model. During 10.2 years of follow-up, we ascertained 1,995 incident stroke cases, including 1,511 cerebral infarctions, 321 hemorrhagic strokes, and 163 unspecified strokes. High chocolate consumption was associated with a lower risk of stroke. The multivariable relative risk of stroke comparing the highest quartile of chocolate consumption (median 62.9 g/week) with the lowest quartile (median 0 g/week) was 0.83 (95 % CI 0.70-0.99). The association did not differ by stroke subtypes. In a meta-analysis of 5 studies, with a total of 4,260 stroke cases, the overall relative risk of stroke for the highest vs lowest category of chocolate consumption was 0.81 (95% CI 0.73-0.90), without heterogeneity among studies (p = 0.47). These findings suggest that moderate chocolate consumption may lower the risk of stroke.

Use of aspirin, other nonsteroidal anti-inflammatory drugs, and acetaminophen and postmenopausal breast cancer incidence.

The associations between use of aspirin, other nonsteroidal anti-inflammatory drugs (NSAIDs), and acetaminophen and breast cancer incidence in postmenopausal women are uncertain. We examined these associations with breast cancer, both overall and by molecular subtype. We observed 84,602 postmenopausal women, free of cancer in 1980, until June 2008 and prospectively collected data on analgesic use, reproductive history, and other lifestyle factors using biennial questionnaires. Proportional hazards models were used to estimate multivariable relative risks (RRs) and 95% CIs. We documented 4,734 cases of incident invasive breast cancer. Compared with nonuse of aspirin, multivariable RRs of regular aspirin use (≥ two tablets per week) for more than 20 years were 0.91 for overall breast cancer (95% CI, 0.81 to 1.01; P(trend) = 0.16), 0.90 for estrogen receptor (ER) -positive progesterone receptor (PR) -positive breast cancer (95% CI, 0.77 to 1.06; P(trend) = 0.17), and 0.91 for ER-negative PR-negative breast cancer (95% CI, 0.68 to 1.22; P(trend) = 0.97). Results did not vary appreciably by past or current use, days per week of use, or dosage of use. Use of other NSAIDs and acetaminophen was largely not significantly associated with breast cancer risk. Additionally, use of higher doses of each analgesic (≥ six tablets per week) for more than 10 years was generally not significantly associated with risk of breast cancer, either overall or by subtype. Furthermore, largely no substantial associations were noted for breast cancer molecular subtypes, including luminal A, luminal B, triple negative, basal-like, human epidermal growth factor receptor 2 positive, cyclooxygenase-2 (COX-2) negative, and COX-2 positive. Our study suggests that use of aspirin, other NSAIDs, and acetaminophen is not importantly associated with risk of postmenopausal breast cancer, either overall or by specific subtype.