Lifetime Smoking History and Prevalence of Osteoporosis and Low Bone Density in U.S. Adults, National Health and Nutrition Examination Survey 2005-2010.

Background: Osteoporosis is common among older adults. Women are more likely to have osteoporosis than men. The prevalence varies with race/ethnicity, with the highest prevalence observed among non-Hispanic, Asian women. Prior studies identified a negative association between smoking and bone mineral density (BMD). The association between smoking and osteoporosis has not been investigated according to race/ethnicity. Materials and Methods: We included 4,226 U.S. adults aged 50 years or older with complete information on smoking history, BMD, and other independent variables from the 2005-2010 National Health and Nutrition Examination Surveys. Design-based multinomial logistic regression was utilized to estimate prevalence odds ratios (POR) of osteoporosis (T-score ≤ -2.5) and of low bone density (T-score between -1.0 and -2.5) in relation to lifetime smoking pack-years, stratified by sex and race/ethnicity. Results: Participants were 61.5 (standard error 0.21) years old on average and 48% women (n = 2,027). Among women, a smoking history ≥30 pack-years was positively associated with osteoporosis (POR: 2.40; 95% confidence interval [CI]: 1.42-4.06). Similar POR were observed among non-Hispanic White, non-Hispanic Black, and Mexican American women. However, POR for ≥30 pack-years and low bone density were positive but not statistically significant. Among men, null associations of smoking history, osteoporosis, and low bone density were observed, except for a positive association of ≥30 pack-years and low bone density among non-Hispanic Black men. Conclusion: Osteoporosis was twice as prevalent among women who smoked ≥30 pack-years than among women who never smoked, regardless of race/ethnicity. Smoking history and osteoporosis were not associated among men.

History of Breastfeeding and Stroke Prevalence Among Parous Females: Findings from the National Health and Nutrition Examination Survey, 1999-2012.

Introduction: Given the known relationships between breastfeeding and hypertension, and between hypertension and stroke, we examined the association between breastfeeding a child and stroke later in life. We hypothesized a history of breastfeeding is associated with lower prevalence of stroke. Methods: Our study population included parous females 20 years and older. Potential confounders included family income-to-poverty ratio, education, smoking status, parity, and age at first live birth. Potential effect measure modifiers included race/ethnicity and age at survey. Multivariable logistic regression models were used to estimate crude and adjusted prevalence odds ratios (ORs) and 95% confidence intervals (CIs) for the association between breastfeeding and stroke. Results: The odds of stroke among those who ever breastfed were lower than those who never breastfed (OR = 0.68, 95% CI: 0.55-0.86; p < 0.01). Adjusting for education decreased the strength of the association (OR = 0.80, 95% CI: 0.63-1.01; p = 0.06). In analyses stratified by age, the inverse association between breastfeeding status was significant among those ages 20-64 years (OR = 0.69, 95% CI: 0.47-1.00) but not among those 65 years and older (OR = 0.99, 95% CI: 0.73-1.34). Conclusions: Ever breastfeeding a child was associated with lower prevalence of stroke in a nationally representative U.S. sample of parous females 20 years and older. Results were strongest among those <65 years old.