Diane Feskanich

Reproducibility and validity of food intake measurements from a semiquantitative food frequency questionnaire

Few food frequency questionnaires have been evaluated for their ability to assess intakes of individual foods that may be related to disease independently of their nutrient content. The reproducibility and validity of food intake measurements by a 131-item semiquantitative food frequency questionnaire were evaluated in a sample of 127 men from the Health Professionals Follow-up Study, a large longitudinal study of diet and disease. Each subject completed two questionnaires 1 year apart and two 1-week diet records 6 months apart during the intervening year. Pearson correlations assessing reproducibility between food intakes from the two questionnaires ranged from .31 for pie to .92 for coffee (mean = .59). Validity was measured by comparing food intakes from the second questionnaire with those from the diet records. Pearson correlations corrected for within-person weekly variation in diet record data ranged from .17 for other nuts to .95 for bananas (mean = .63). Large within-person variation precluded the calculation of accurate validity correlations for 29 foods. As we previously observed in women, the foods most often overreported were fruits and vegetables, and meats and dairy products were most often underreported. With few exceptions, reasonable levels of reproducibility and validity were observed for intake of individual foods in this extensive food frequency questionnaire.

Ovarian Conservation at the Time of Hysterectomy and Long-Term Health Outcomes in the Nurses' Health Study

OBJECTIVE: To report long-term health outcomes and mortality after oophorectomy or ovarian conservation. METHODS: We conducted a prospective, observational study of 29,380 women participants of the Nurses’ Health Study who had a hysterectomy for benign disease; 16,345 (55.6%) had hysterectomy with bilateral oophorectomy, and 13,035 (44.4%) had hysterectomy with ovarian conservation. We evaluated incident events or death due to coronary heart disease (CHD), stroke, breast cancer, ovarian cancer, lung cancer, colorectal cancer, total cancers, hip fracture, pulmonary embolus, and death from all causes. RESULTS: Over 24 years of follow-up, for women with hysterectomy and bilateral oophorectomy compared with ovarian conservation, the multivariable hazard ratios (HRs) were 1.12 (95% confidence interval [CI] 1.03–1.21) for total mortality, 1.17 (95% CI 1.02–1.35) for fatal plus nonfatal CHD, and 1.14 (95% CI 0.98–1.33) for stroke. Although the risks of breast (HR 0.75, 95% CI 0.68–0.84), ovarian (HR 0.04, 95% CI 0.01–0.09, number needed to treat=220), and total cancers (HR 0.90, 95% CI 0.84–0.96) decreased after oophorectomy, lung cancer incidence (HR=1.26, 95% CI 1.02–1.56, number needed to harm=190), and total cancer mortality (HR=1.17, 95% CI 1.04–1.32) increased. For those never having used estrogen therapy, bilateral oophorectomy before age 50 years was associated with an increased risk of all-cause mortality, CHD, and stroke. With an approximate 35-year life span after surgery, one additional death would be expected for every nine oophorectomies performed. CONCLUSION: Compared with ovarian conservation, bilateral oophorectomy at the time of hysterectomy for benign disease is associated with a decreased risk of breast and ovarian cancer but an increased risk of all-cause mortality, fatal and nonfatal coronary heart disease, and lung cancer. In no analysis or age group was oophorectomy associated with increased survival. LEVEL OF EVIDENCE: II

Smoking and mortality--beyond established causes.

BACKGROUND Mortality among current smokers is 2 to 3 times as high as that among persons who never smoked. Most of this excess mortality is believed to be explained by 21 common diseases that have been formally established as caused by cigarette smoking and are included in official estimates of smoking-attributable mortality in the United States. However, if smoking causes additional diseases, these official estimates may significantly underestimate the number of deaths attributable to smoking. METHODS We pooled data from five contemporary U.S. cohort studies including 421,378 men and 532,651 women 55 years of age or older. Participants were followed from 2000 through 2011, and relative risks and 95% confidence intervals were estimated with the use of Cox proportional-hazards models adjusted for age, race, educational level, daily alcohol consumption, and cohort. RESULTS During the follow-up period, there were 181,377 deaths, including 16,475 among current smokers. Overall, approximately 17% of the excess mortality among current smokers was due to associations with causes that are not currently established as attributable to smoking. These included associations between current smoking and deaths from renal failure (relative risk, 2.0; 95% confidence interval [CI], 1.7 to 2.3), intestinal ischemia (relative risk, 6.0; 95% CI, 4.5 to 8.1), hypertensive heart disease (relative risk, 2.4; 95% CI, 1.9 to 3.0), infections (relative risk, 2.3; 95% CI, 2.0 to 2.7), various respiratory diseases (relative risk, 2.0; 95% CI, 1.6 to 2.4), breast cancer (relative risk, 1.3; 95% CI, 1.2 to 1.5), and prostate cancer (relative risk, 1.4; 95% CI, 1.2 to 1.7). Among former smokers, the relative risk for each of these outcomes declined as the number of years since quitting increased. CONCLUSIONS A substantial portion of the excess mortality among current smokers between 2000 and 2011 was due to associations with diseases that have not been formally established as caused by smoking. These associations should be investigated further and, when appropriate, taken into account when the mortality burden of smoking is investigated. (Funded by the American Cancer Society.).

Lung cancer occurrence in never-smokers: an analysis of 13 cohorts and 22 cancer registry studies.

Background Better information on lung cancer occurrence in lifelong nonsmokers is needed to understand gender and racial disparities and to examine how factors other than active smoking influence risk in different time periods and geographic regions. Methods and Findings We pooled information on lung cancer incidence and/or death rates among self-reported never-smokers from 13 large cohort studies, representing over 630,000 and 1.8 million persons for incidence and mortality, respectively. We also abstracted population-based data for women from 22 cancer registries and ten countries in time periods and geographic regions where few women smoked. Our main findings were: (1) Men had higher death rates from lung cancer than women in all age and racial groups studied; (2) male and female incidence rates were similar when standardized across all ages 40+ y, albeit with some variation by age; (3) African Americans and Asians living in Korea and Japan (but not in the US) had higher death rates from lung cancer than individuals of European descent; (4) no temporal trends were seen when comparing incidence and death rates among US women age 40–69 y during the 1930s to contemporary populations where few women smoke, or in temporal comparisons of never-smokers in two large American Cancer Society cohorts from 1959 to 2004; and (5) lung cancer incidence rates were higher and more variable among women in East Asia than in other geographic areas with low female smoking. Conclusions These comprehensive analyses support claims that the death rate from lung cancer among never-smokers is higher in men than in women, and in African Americans and Asians residing in Asia than in individuals of European descent, but contradict assertions that risk is increasing or that women have a higher incidence rate than men. Further research is needed on the high and variable lung cancer rates among women in Pacific Rim countries.

Computerized collection and analysis of dietary intake information.

A computer system has been developed which facilitates standardized and in-depth collection and analysis of dietary intake information required for nutritional research. The system uses a hierarchically arranged database of food descriptions. A user identifies a consumed food by traversing a hierarchy and reaching a leaf node food item via a succession of selections from menu displays. If a food contains ingredients or was cooked with additional fat or salt, multiple traversals of the food hierarchies are necessary. The system assigns one or more food codes and determines the gram weight of the food consumed using everyday measures such as cups, slices, ounces, etc. It then uses these food codes and gram weights to calculate nutrient content. Utilizing binary trees and physical pointers to expedite data retrieval from this nearly 10 Mbyte database permits the systems use in conducting real-time interactive interviews.

Dietary Carotenoids and Risk of Lung Cancer in a Pooled Analysis of Seven Cohort Studies

Intervention trials with supplemental β-carotene have observed either no effect or a harmful effect on lung cancer risk. Because food composition databases for specific carotenoids have only become available recently, epidemiological evidence relating usual dietary levels of these carotenoids with lung cancer risk is limited. We analyzed the association between lung cancer risk and intakes of specific carotenoids using the primary data from seven cohort studies in North America and Europe. Carotenoid intakes were estimated from dietary questionnaires administered at baseline in each study. We calculated study-specific multivariate relative risks (RRs) and combined these using a random-effects model. The multivariate models included smoking history and other potential risk factors. During follow-up of up to 7–16 years across studies, 3,155 incident lung cancer cases were diagnosed among 399,765 participants. β-Carotene intake was not associated with lung cancer risk (pooled multivariate RR = 0.98; 95% confidence interval, 0.87–1.11; highest versus lowest quintile). The RRs for α-carotene, lutein/zeaxanthin, and lycopene were also close to unity. β-Cryptoxanthin intake was inversely associated with lung cancer risk (RR = 0.76; 95% confidence interval, 0.67–0.86; highest versus lowest quintile). These results did not change after adjustment for intakes of vitamin C (with or without supplements), folate (with or without supplements), and other carotenoids and multivitamin use. The associations generally were similar among never, past, or current smokers and by histological type. Although smoking is the strongest risk factor for lung cancer, greater intake of foods high in β-cryptoxanthin, such as citrus fruit, may modestly lower the risk.

Vitamin D intake and risks of systemic lupus erythematosus and rheumatoid arthritis in women

Objectives: Vitamin D has immune-modulating effects and may protect against the development of systemic lupus erythematosus (SLE) and rheumatoid arthritis (RA). Methods: We identified incident cases of SLE and RA among 186 389 women followed from 1980 to 2002 in the Nurses’ Health Study and Nurses’ Health Study II cohorts. We excluded subjects where SLE or RA was not confirmed by medical record review, and those who failed to return questionnaires. Semi-quantitative food frequency questionnaires assessed vitamin D intake from food and supplements. We used cumulative-updated total energy-adjusted dietary exposures for each 2-year cycle. Relationships between vitamin D intake and incident SLE and RA were examined in age-adjusted and Cox proportional hazards models, adjusted for confounders. Results were pooled using meta-analysis random effects models. Results: We confirmed 190 incident cases of SLE and 722 of RA with dietary information. Increasing levels of vitamin D intake had no relationship to the relative risk of developing either SLE or RA. Conclusions: Vitamin D intake was not associated with risk of SLE or RA in these large prospective cohorts of women.

Fruits, vegetables and lung cancer: A pooled analysis of cohort studies

Inverse associations between fruit and vegetable consumption and lung cancer risk have been consistently reported. However, identifying the specific fruits and vegetables associated with lung cancer is difficult because the food groups and foods evaluated have varied across studies. We analyzed fruit and vegetable groups using standardized exposure and covariate definitions in 8 prospective studies. We combined study‐specific relative risks (RRs) using a random effects model. In the pooled database, 3,206 incident lung cancer cases occurred among 430,281 women and men followed for up to 6–16 years across studies. Controlling for smoking habits and other lung cancer risk factors, a 16–23% reduction in lung cancer risk was observed for quintiles 2 through 5 vs. the lowest quintile of consumption for total fruits (RR = 0.77; 95% CI = 0.67–0.87 for quintile 5; p‐value, test for trend < 0.001) and for total fruits and vegetables (RR = 0.79; 95% CI = 0.69–0.90; p‐value, test for trend = 0.001). For the same comparison, the association was weaker for total vegetable consumption (RR = 0.88; 95% CI = 0.78–1.00; p‐value, test for trend = 0.12). Associations were similar between never, past, and current smokers. These results suggest that elevated fruit and vegetable consumption is associated with a modest reduction in lung cancer risk, which is mostly attributable to fruit, not vegetable, intake. However, we cannot rule out the possibility that our results are due to residual confounding by smoking. The primary focus for reducing lung cancer incidence should continue to be smoking prevention and cessation.

Risk Factors and Individual Probabilities of Melanoma for Whites

PURPOSE Incidence and mortality of cutaneous melanoma is rising rapidly in the United States; therefore, identifying risk factors for melanoma and integrating them into a clinical and population risk estimation tool may help guide prevention efforts and identify participants for preventive interventions. METHODS We examined risk factors for melanoma in three large prospective studies of women and men. We observed 152,949 women and 25,206 men free of cancer at baseline for up to 14 years. RESULTS A total of 535 incident cases of invasive melanoma (444 women and 91 men) were included in the analysis. We combined the three studies to examine risk factors and to build a risk model to calculate melanoma risk score. Older age, male sex, family history of melanoma, higher number of nevi, history of severe sunburn, and light hair color were each associated with significantly elevated risk of melanoma and were included in the final risk prediction. Participants at the highest decile of risk had a more than three-fold increase in risk of melanoma compared with those in the lowest decile (observed relative risk, 3.61; expected relative risk, 4.20). The measure of discriminatory accuracy as summarized by an age-and sex-adjusted concordance statistic of 0.62 (95% CI, 0.58 to 0.65) indicated that the model had reasonable ability to differentiate those who will develop melanoma and those who will remain free from the disease. CONCLUSION We identified several risk factors for melanoma and developed statistical models with adequate performance and discriminatory accuracy.

Milk, dietary calcium, and bone fractures in women : A 12-year prospective study

OBJECTIVES This study examined whether higher intakes of milk and other calcium-rich foods during adult years can reduce the risk of osteoporotic fractures. METHODS This was a 12-year prospective study among 77761 women, aged 34 through 59 years in 1980, who had never used calcium supplements. Dietary intake was assessed with a food-frequency questionnaire in 1980, 1984, and 1986. Fractures of the proximal femur (n = 133) and distal radius (n = 1046) from low or moderate trauma were self-reported on biennial questionnaires. RESULTS We found no evidence that higher intakes of milk or calcium from food sources reduce fracture incidence. Women who drank two or more glasses of milk per day had relative risks of 1.45 for hip fracture (95% confidence interval [CI] = 0.87, 2.43) and 1.05 for forearm fracture (95% CI = 0.88, 1.25) when compared with women consuming one glass or less per week. Likewise, higher intakes of total dietary calcium or calcium from dairy foods were not associated with decreased risk of hip or forearm fracture. CONCLUSIONS These data do not support the hypothesis that higher consumption of milk or other food sources of calcium by adult women protects against hip or forearm fractures.