Dataset Information

Estimating Risk with Time-to-Event Data: An Application to the Women's Health Initiative.

ABSTRACT: Accurate and individualized risk prediction is critical for population control of chronic diseases such as cancer and cardiovascular disease. Large cohort studies provide valuable resources for building risk prediction models, as the risk factors are collected at the baseline and subjects are followed over time until disease occurrence or termination of the study. However, for rare diseases the baseline risk may not be estimated reliably based on cohort data only, due to sparse events. In this paper, we propose to make use of external information to improve efficiency for estimating time-dependent absolute risk. We derive the relationship between external disease incidence rates and the baseline risk, and incorporate the external disease incidence information into estimation of absolute risks, while allowing for potential difference of disease incidence rates between cohort and external sources. The asymptotic properties, namely, uniform consistency and weak convergence, of the proposed estimators are established. Simulation results show that the proposed estimator for absolute risk is more efficient than that based on the Breslow estimator, which does not utilize external disease incidence rates. A large cohort study, the Women's Health Initiative Observational Study, is used to illustrate the proposed method.

SUBMITTER: Liu D

PROVIDER: S-EPMC4091861 | biostudies-literature | 2014 Jun

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Estimating Risk with Time-to-Event Data: An Application to the Women's Health Initiative.

Liu Dandan D Zheng Yingye Y Prentice Ross L RL Hsu Li L

Journal of the American Statistical Association 20140601 506

Accurate and individualized risk prediction is critical for population control of chronic diseases such as cancer and cardiovascular disease. Large cohort studies provide valuable resources for building risk prediction models, as the risk factors are collected at the baseline and subjects are followed over time until disease occurrence or termination of the study. However, for rare diseases the baseline risk may not be estimated reliably based on cohort data only, due to sparse events. In this p ...[more]

PMID: 25018574

Similar Datasets

Project description:Nutrient-related anemia among postmenopausal women is preventable; recent data on prevalence are limited.To investigate the association between nutrient intakes and anemia prevalence, in relation to both incidence and persistence, in a longitudinal sample of postmenopausal women. We hypothesized that anemia prevalence, incidence, and persistence would be greater among women reporting lower intake of vitamin B-12, folate, and iron.Prospective cohort analysis.The observational cohort of the Women's Health Initiative, including 93,676 postmenopausal women, aged 50 to 79 years, who were recruited across the United States at 40 clinical study sites. Women were enrolled between 1993 and 1998; data collection for these analyses continued through 2000.Anemia was defined as a blood hemoglobin concentration of <12.0 g/dL (120.0 g/L). Persistent anemia was defined as anemia present at each measurement time point. Diet was assessed by food frequency questionnaire for iron, folate, B-12, red meat, and cold breakfast cereal; inadequacies were based on dietary reference intakes for women older than age 50 years.Descriptive statistics (mean ± standard deviation) were used to characterize the population demographics, anemia rates, and diet. Unconditional logistic regression was used to investigate associations between diet and incident and persistent anemia. Associations are presented as odds ratio and 95% confidence intervals.Anemia was identified in 3,979 (5.5%) of the cohort. Inadequate intakes of multiple anemia-associated nutrients were less frequent in non-Hispanic whites (7.4%) than other race/ethnic groups (inadequacies demonstrated in 14.6% to 16.3% of the sample). Age, body mass index, and smoking were associated with anemia. Women with anemia reported lower intakes of energy, protein, folate, vitamin B-12, iron, vitamin C, and red meat. Multiple (more than a single nutrient) dietary deficiencies were associated with a 21% greater risk of persistent anemia (odds ratio 1.21, 95% confidence interval 1.05 to 1.41) and three deficiencies resulted in a 44% increase in risk for persistent anemia (odds ratio 1.44, 95% confidence interval 1.20 to 1.73).Inadequate nutrient intake, a modifiable condition, is associated with greater risk for anemia in postmenopausal women participating in the Observational Study of the Women's Health Initiative. Efforts to identify and update incidence estimates for anemia-associated nutrient deficiencies in aging women should be undertaken.

Dataset Information

Estimating Risk with Time-to-Event Data: An Application to the Women's Health Initiative.

Publications

Estimating Risk with Time-to-Event Data: An Application to the Women's Health Initiative.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets