Search tips
Search criteria

Results 1-25 (50)

Clipboard (0)

Select a Filter Below

more »
Year of Publication
Document Types
author:("Cai, jiande")
1.  Semiparametric Inference for Data with a Continuous Outcome from a Two-Phase Probability Dependent Sampling Scheme 
Multi-phased designs and biased sampling designs are two of the well recognized approaches to enhance study efficiency. In this paper, we propose a new and cost-effective sampling design, the two-phase probability dependent sampling design (PDS), for studies with a continuous outcome. This design will enable investigators to make efficient use of resources by targeting more informative subjects for sampling. We develop a new semiparametric empirical likelihood inference method to take advantage of data obtained through a PDS design. Simulation study results indicate that the proposed sampling scheme, coupled with the proposed estimator, is more efficient and more powerful than the existing outcome dependent sampling design and the simple random sampling design with the same sample size. We illustrate the proposed method with a real data set from an environmental epidemiologic study.
PMCID: PMC3984585  PMID: 24737947
Empirical likelihood; Missing data; Semiparametric; Probability sample
2.  Kappa statistic for the clustered dichotomous responses from physicians and patients 
Statistics in medicine  2013;32(21):10.1002/sim.5796.
The bootstrap method for estimating the standard error of the kappa statistic in the presence of clustered data is evaluated. Such data arise, for example, in assessing agreement between physicians and their patients regarding their understanding of the physician-patient interaction and discussions. We propose a computationally efficient procedure for generating correlated dichotomous responses for physicians and assigned patients for simulation studies. The simulation result demonstrates that the proposed bootstrap method produces better estimate of the standard error and better coverage performance compared to the asymptotic standard error estimate that ignores dependence among patients within physicians with at least a moderately large number of clusters. An example of an application to a coronary heart disease prevention study is presented.
PMCID: PMC3844626  PMID: 23533082
Kappa statistic; a dichotomous response for clusters; bootstrap resampling for clusters
3.  A Randomized Trial of a Facebook-based Physical Activity Intervention for Young Adult Cancer Survivors 
Over half of young adult cancer survivors do not meet physical activity (PA) guidelines. PA interventions can enhance health and quality of life among young adult cancer survivors. However, few exercise interventions have been designed and tested in this population. This study evaluated the feasibility and preliminary efficacy of a 12-week, Facebook-based intervention (FITNET) aimed at increasing moderate-to-vigorous intensity PA compared to a Facebook-based self-help comparison (SC) condition.
Young adult cancer survivors (n=86) were randomly assigned to the FITNET or SC group. All participants were asked to complete self-administered online questionnaires at baseline and after 12 weeks.
Seventy-seven percent of participants completed post-intervention assessments, and most participants reported using intervention components as intended. Participants in both groups would recommend the program to other young adult cancer survivors (FITNET: 46.9% vs. SC: 61.8%; p=0.225). Over 12 weeks, both groups increased self-reported weekly minutes of moderate-to-vigorous PA (FITNET: 67 minutes/week, p=0.009 vs. SC: 46 minutes/week, p=0.045), with no significant difference between groups. Increases in light PA were 135 minutes/week greater in the FITNET group relative to the SC group (p=0.032), and the FITNET group reported significant weight loss over time (−2.1 kg, p=0.004; p=0.083 between groups).
Facebook-based intervention approaches demonstrated potential for increasing PA in young adult cancer survivors.
Implications for Cancer Survivors
Social networking sites may be a feasible way for young adult cancer survivors to receive health information and support to promote PA and healthy behaviors.
PMCID: PMC3737370  PMID: 23532799
young adults; cancer survivors; physical activity; social networking site; randomized trial; intervention
4.  Long- and Short-term Weight Change and Incident Coronary Heart Disease and Ischemic Stroke 
American Journal of Epidemiology  2013;178(2):239-248.
Weight gain increases the prevalence of obesity, a risk factor for cardiovascular disease. Nevertheless, unintentional weight loss can be a harbinger of health problems. The Atherosclerosis Risk in Communities Study (1987–2009) included 15,792 US adults aged 45–64 years at baseline and was used to compare associations of long-term (30 years) and short-term (3 years) weight change with the risks of coronary heart disease (CHD) and ischemic stroke. Age-, gender-, and race-standardized incidence rates were 4.9 (95% confidence interval (CI): 4.6, 5.2) per 1,000 person-years for CHD and 2.5 (95% CI: 2.3, 2.8) per 1,000 person-years for stroke. After controlling for baseline body mass index and other covariates, long-term weight gain (since age 25 years) of more than 2.7% was associated with elevated CHD risk, and any long-term weight gain was associated with increased stroke risk. Among middle-aged adults, short-term (3-year) weight loss of more than 3% was associated with elevated immediate CHD risk (hazard ratio = 1.46, 95% CI: 1.18, 1.81) and stroke risk (hazard ratio = 1.45, 95% CI: 1.10, 1.92). Risk tended to be larger in adults whose weight loss did not occur through dieting. Avoidance of weight gain between early and middle adulthood can reduce risks of CHD and stroke, but short-term, unintentional weight loss in middle adulthood may be an indicator of immediate elevated risk that has not previously been well recognized.
PMCID: PMC3816338  PMID: 23645623
body mass index; body weight changes; coronary heart disease; ischemic stroke
5.  Associations of body mass index with incident hypertension in American white, American black and Chinese Asian adults in early and middle adulthood: the Coronary Artery Risk Development in Young Adults (CARDIA) study, the Atherosclerosis Risk in Communities (ARIC) study and the People’s Republic of China (PRC) study 
The association of body mass index (BMI) with blood pressure may be stronger in Asian than non-Asian populations, however, longitudinal studies with direct comparisons between ethnicities are lacking. We compared the relationship of BMI with incident hypertension over approximately 9.5 years of follow-up in young (24-39 years) and middle-aged (45-64 years) Chinese Asians (n=5354), American Blacks (n=6076) and American Whites (n=13451). We estimated risk differences using logistic regression models and calculated adjusted incidences and incidence differences. To facilitate comparisons across ethnicities, standardized estimates were calculated using mean covariate values for age, sex, smoking, education and field center, and included the quadratic terms for BMI and age. Weighted least-squares regression models with were constructed to summarize ethnic-specific incidence differences across BMI. Wald statistics and p-values were calculated based on chi-square distributions. The association of BMI with the incidence difference for hypertension was steeper in Chinese (p<0.05) than in American populations during young and middle-adulthood. For example, at a BMI of 25 vs 21 kg/m2 the adjusted incidence differences per 1000 persons (95% CI) in young adults with a BMI of 25 vs those with a BMI of 21 was 83 (36-130) for Chinese, 50 (26-74) for Blacks and 30 (12-48) for Whites; among middle-aged adults it was 137 (77-198) for Chinese, 49 (9-88) for Blacks and 54 (38-69) for Whites. Whether hypertension carries the same level of risk of stroke or cardiovascular disease across national or ethnic groups remains uncertain.
PMCID: PMC4053207  PMID: 24231024
blood pressure; multi-ethnic; obesity; African American; cardiovascular disease
6.  Patient-Physician Agreement on the Content of CHD Prevention Discussions 
Little is known about agreement between patients and physicians on content and outcomes of clinical discussions. A common perception of content and outcomes may be desirable to optimize decision making and clinical care.
To determine patient-physician agreement on content and outcomes of coronary heart disease (CHD) prevention discussions.
Cross-sectional survey nested within a randomized CHD prevention study.
Setting and participants
University internal medicine clinic; 24 physicians and 157 patients.
Following one clinic visit, we surveyed patients and physicians on discussion content, decision making, and final decisions about CHD prevention. For comparison, we audio-recorded, transcribed, and coded 20 patient-physician visits. We calculated percent agreement between patient/physician reports, patient/transcription reports, and physician/transcription reports. We calculated Cohen’s kappas to compare patient/physician perspectives.
Patients and physicians agreed on whether CHD was discussed in 130 visits (83%; kappa=0.55; 95% CI .40–.70). When discussions occurred, they agreed about discussion content (pros versus cons) in 53% of visits (kappa=0.15; 95% CI −.01–.30), and physicians’ recommendations in 73% (kappa=0.44; 95% CI .28–.66). Patients and physicians agreed on final decisions to take medication in 78% (kappa=.58; 95% CI .45–.71) and change lifestyle in 69% (kappa=.38; 95% CI .24–.53). They agreed less often, 43%, (kappa=.13; 95% CI −.11–.37) about degree of involvement in decision making. Audio-recorded results were similar, but showed very low agreement between transcripts and patients’ and physicians’ self-report on discussion content and decision making.
Disagreements about clinical discussions and decision making may be common. Future work is needed to determine: how widespread such agreements are; whether they impact clinical outcomes; and the relative importance of the subjective experience versus objective steps of shared decision making.
PMCID: PMC4043219  PMID: 20673244
shared decision making; patient-physician agreement
7.  Primary Care Providers' Self-Efficacy and Outcome Expectations for Childhood Obesity Counseling 
Childhood Obesity  2013;9(3):208-215.
Primary care providers have a role in the prevention and management of childhood obesity. We explored the relationship of providers' self-efficacy, outcome expectations, and practice level support with childhood obesity counseling frequency.
Providers (n=123) completed a survey that assessed their self-efficacy, outcome expectations, and reported obesity counseling frequency. A practice level assessment tool was used to characterize the practices. We analyzed data using frequencies and proportional odds modeling.
Providers were confident or very confident (78.5–93.5%) in their ability to counsel about healthy eating, physical activity, and weight and agreed or strongly agreed (64.2–86.2%) that their counseling would result in actual changes. Providers with higher outcome expectations were more likely [odds ratio (OR)=3.4] to report providing obesity counseling. Female providers were more likely to report counseling about obesity (OR=2.3) than males. Providers in practices with resources for healthy eating and physical activity reported higher levels of self-efficacy and counseling frequency.
In our study, providers were confident in their ability to provide obesity counseling and expected changes from their efforts, suggesting that future studies should build on the high level of outcome expectations as well as self-efficacy. The gender difference found regarding obesity counseling may need further exploration.
PMCID: PMC3675833  PMID: 23635310
8.  Sample size estimation in educational intervention trials with subgroup heterogeneity in only one arm 
Statistics in medicine  2012;32(12):2140-2154.
We present closed form sample size and power formulas motivated by the study of a psycho-social intervention in which the experimental group has the intervention delivered in teaching subgroups while the control group receives usual care. This situation is different from the usual clustered randomized trial since subgroup heterogeneity only exists in one arm. We take this modification into consideration and present formulas for the situation in which we compare a continuous outcome at both a single point in time and longitudinally over time. In addition, we present the optimal combination of parameters such as the number of subgroups and number of time points for minimizing sample size and maximizing power subject to constraints such as the maximum number of measurements that can be taken (i.e. a proxy for cost).
PMCID: PMC3615113  PMID: 23172724
Sample size; heterogeneous subgroups; clinical trials; longitudinal data
9.  Effect of Adding a Values Clarification Exercise to a Decision Aid on Heart Disease Prevention: A Randomized Trial 
Experts have called for the inclusion of values clarification (VC) exercises in decision aids (DA) as a means of improving their effectiveness, but little research has examined the effects of such exercises.
To determine whether adding a VC exercise to a DA on heart disease prevention improves decision making outcomes.
Randomized trial.
UNC Decision Support Laboratory.
Adults ages 40–80 with no history of cardiovascular disease.
A web-based heart disease prevention DA with or without a VC exercise.
Pre and post-intervention decisional conflict and intent to reduce CHD risk. Post-intervention self-efficacy and perceived values concordance.
We enrolled 137 participants (62 in DA; 75 in VC) with moderate decisional conflict (DA 2.4; VC 2.5) and no baseline differences among groups. After the interventions, we found no clinically or statistically significant differences between groups in decisional conflict (DA 1.8; VC 1.9; absolute difference VC-DA 0.1, 95% CI −0.1 to 0.3), intent to reduce CHD risk (DA 98%; VC 100%; absolute differences VC-DA: 2%, 95% CI −0.02% to 5%), perceived values concordance (DA 95%, VC 92%; absolute difference VC-DA −3%, 95% CI −11 to +5%), or self efficacy for risk reduction (DA 97%, VC 92%; absolute difference VC-DA −5%, 95% CI −13 to +3%). However, VC tended to change some decisions about risk reduction strategies.
Use of a hypothetical scenario. Ceiling effects for some outcomes.
Adding a VC intervention to a DA did not further improve decision making outcomes in a population of highly educated and motivated adults responding to scenario-based questions. Work is needed to determine the effects of VC on more diverse populations and more distal outcomes.
PMCID: PMC3996504  PMID: 20484089
10.  Childhood Obesity Prevention: Fathers' Reflections with Healthcare Providers 
Childhood Obesity  2013;9(2):137-143.
To prevent childhood obesity, parents and their children's healthcare providers need to engage in effective dialogue. We know much about mothers' experiences, but very little about fathers' experiences.
We explored African-American, Caucasian, and Latino fathers' perceptions and experiences communicating with their children's provider during clinic visits regarding weight, diet, and physical activity. Focus groups (n=3), grouped by race/ethnicity, including a total of 24 fathers, were conducted. The men were asked open-ended questions; responses were recorded and transcribed, and analyzed using ATLAS.ti.
Findings revealed that these fathers were involved in their children's healthcare and found providers to be helpful partners in keeping their children healthy, yet they generally felt “left out” during clinic appointments. The quality of the relationship with their children's provider influenced how receptive fathers were to discussing their children's weight, diet, and physical activity behaviors. Fathers made suggestions to help improve communication between providers and fathers, such as personalizing the discussion.
These fathers expressed strong feelings about the provider–parent relationship when discussing weight, diet, and physical activity.
PMCID: PMC3621357  PMID: 23472966
11.  The effect of a decision aid intervention on decision making about coronary heart disease risk reduction: secondary analyses of a randomized trial 
Decision aids offer promise as a practical solution to improve patient decision making about coronary heart disease (CHD) prevention medications and help patients choose medications to which they are likely to adhere. However, little data is available on decision aids designed to promote adherence.
In this paper, we report on secondary analyses of a randomized trial of a CHD adherence intervention (second generation decision aid plus tailored messages) versus usual care in an effort to understand how the decision aid facilitates adherence. We focus on data collected from the primary study visit, when intervention participants presented 45 minutes early to a previously scheduled provider visit; viewed the decision aid, indicating their intent for CHD risk reduction after each decision aid component (individualized risk assessment and education, values clarification, and coaching); and filled out a post-decision aid survey assessing their knowledge, perceived risk, decisional conflict, and intent for CHD risk reduction. Control participants did not present early and received usual care from their provider. Following the provider visit, participants in both groups completed post-visit surveys assessing the number and quality of CHD discussions with their provider, their intent for CHD risk reduction, and their feelings about the decision aid.
We enrolled 160 patients into our study (81 intervention, 79 control). Within the decision aid group, the decision aid significantly increased knowledge of effective CHD prevention strategies (+21 percentage points; adjusted p<.0001) and the accuracy of perceived CHD risk (+33 percentage points; adjusted p<.0001), and significantly decreased decisional conflict (-0.63; adjusted p<.0001). Comparing between study groups, the decision aid also significantly increased CHD prevention discussions with providers (+31 percentage points; adjusted p<.0001) and improved perceptions of some features of patient-provider interactions. Further, it increased participants’ intentions for any effective CHD risk reducing strategies (+21 percentage points; 95% CI 5 to 37 percentage points), with a majority of the effect from the educational component of the decision aid. Ninety-nine percent of participants found the decision aid easy to understand and 93% felt it easy to use.
Decision aids can play an important role in improving decisions about CHD prevention and increasing patient-provider discussions and intent to reduce CHD risk.
PMCID: PMC3943405  PMID: 24575882
Decision support techniques; Medication adherence; Heart disease; Primary prevention
12.  Nonparametric estimation of the mean function for recurrent event data with missing event category 
Biometrika  2013;100(3):10.1093/biomet/ast016.
Recurrent event data frequently arise in longitudinal studies when study subjects possibly experience more than one event during the observation period. Often, such recurrent events can be categorized. However, part of the categorization may be missing due to technical difficulties. If the event types are missing completely at random, then a complete case analysis may provide consistent estimates of regression parameters in certain regression models, but estimates of the baseline event rates are generally biased. Previous work on nonparametric estimation of these rates has utilized parametric missingness models. In this paper, we develop fully nonparametric methods in which the missingness mechanism is completely unspecified. Consistency and asymptotic normality of the nonparametric estimators of the mean event functions accommodate nonparametric estimators of the event category probabilities, which converge more slowly than the parametric rate. Plug-in variance estimators are provided and perform well in simulation studies, where complete case estimators may exhibit large biases and parametric estimators generally have a larger mean squared error when the model is misspecified. The proposed methods are applied to data from a cystic fibrosis registry.
PMCID: PMC3887139  PMID: 24415792
Cystic fibrosis; Local polynomial regression; Nelson–Aalen estimation; Pseudomonas aeruginosa infection; Rate proportion
13.  Marginal additive hazards model for case-cohort studies with multiple disease outcomes: an application to the Atherosclerosis Risk in Communities (ARIC) study 
In the case-cohort studies conducted within the Atherosclerosis Risk in Communities (ARIC) study, it is of interest to assess and compare the effect of high-sensitivity C-reactive protein (hs-CRP) on the increased risks of incident coronary heart disease and incident ischemic stroke. Empirical cumulative hazards functions for different levels of hs-CRP reveal an additive structure for the risks for each disease outcome. Additionally, we are interested in estimating the difference in the risk for the different hs-CRP groups. Motivated by this, we consider fitting marginal additive hazards regression models for case-cohort studies with multiple disease outcomes. We consider a weighted estimating equations approach for the estimation of model parameters. The asymptotic properties of the proposed estimators are derived and their finite-sample properties are assessed via simulation studies. The proposed method is applied to analyze the ARIC Study.
PMCID: PMC3520500  PMID: 22826550
Additive hazards model; ARIC study; Case-cohort study; Multivariate failure times; Weighted estimating equations
14.  Semiparametric additive marginal regression models for multiple type recurrent events 
Lifetime data analysis  2012;18(4):10.1007/s10985-012-9226-4.
Recurrent event data are often encountered in biomedical research, for example, recurrent infections or recurrent hospitalizations for patients after renal transplant. In many studies, there are more than one type of events of interest. Cai and Schaubel (2004) advocated a proportional marginal rate model for multiple type recurrent event data. In this paper, we propose a general additive marginal rate regression model. Estimating equations approach is used to obtain the estimators of regression coefficients and baseline rate function. We prove the consistency and asymptotic normality of the proposed estimators. The finite sample properties of our estimators are demonstrated by simulations. The proposed methods are applied to the India renal transplant study to examine risk factors for bacterial, fungal and viral infections.
PMCID: PMC3844629  PMID: 22899088
additive model; empirical process; multiple type recurrent events; recurrent events
15.  Interactions Between Obesity, Parental History of Hypertension, and Age on Prevalent Hypertension: The People’s Republic of China Study 
Age, family history, and body mass index (BMI) influence the prevalence of hypertension, but very little is known about the interplay of these factors in Chinese populations. The authors examined this issue in Chinese adults (n = 4104) in the People’s Republic of China Study. In young adults (24–39 years), the prevalence of hypertension/1000 persons (95% confidence interval [CI]) at the referent BMI was greater among subjects with a parental history of hypertension (35; 15–54) compared with those without (7; 3–11). Among middle-aged (40–71 years) adults, the prevalence of hypertension was similar regardless of parental history; however, the effect of BMI was modified by parental history status. For example, at BMI = 25 kg/m2, the prevalence difference/1000 persons was 375 (95% CI = 245–506) and 97 (95% CI = 51–144) among subjects with and without a parental history, respectively. These large differences call for further investigation of the genetic and environmental factors that could be driving this interaction.
PMCID: PMC3298639  PMID: 21653607
Asian; blood pressure; body mass index; Chinese; family history
16.  Prevalence of Major Cardiovascular Risk Factors and Cardiovascular Diseases Among Hispanic/Latino Individuals of Diverse Backgrounds in the United States 
Major cardiovascular diseases (CVDs) are leading causes of mortality among US Hispanic and Latino individuals. Comprehensive data are limited regarding the prevalence of CVD risk factors in this population and relations of these traits to socioeconomic status (SES) and acculturation.
To describe prevalence of major CVD risk factors and CVD (coronary heart disease [CHD] and stroke) among US Hispanic/Latino individuals of different backgrounds, examine relationships of SES and acculturation with CVD risk profiles and CVD, and assess cross-sectional associations of CVD risk factors with CVD.
Design, Setting, and Participants
Multicenter, prospective, population-based Hispanic Community Health Study/Study of Latinos including individuals of Cuban (n =2201), Dominican (n = 1400), Mexican (n=6232), Puerto Rican (n=2590), Central American (n=1634), and South American backgrounds (n = 1022) aged 18 to 74 years. Analyses involved 15 079 participants with complete data enrolled between March 2008 and June 2011.
Main Outcome Measures
Adverse CVD risk factors defined using national guidelines for hypercholesterolemia, hypertension, obesity, diabetes, and smoking. Prevalence of CHD and stroke were ascertained from self-reported data.
Age-standardized prevalence of CVD risk factors varied by Hispanic/Latino background; obesity and current smoking rates were highest among Puerto Rican participants (for men, 40.9% and 34.7%; for women, 51.4% and 31.7%, respectively); hypercholesterolemia prevalence was highest among Central American men (54.9%) and Puerto Rican women (41.0%). Large proportions of participants (80% of men, 71% of women) had at least 1 risk factor. Age- and sex-adjusted prevalence of 3 or more risk factors was highest in Puerto Rican participants (25.0%) and significantly higher (P<.001) among participants with less education (16.1%), those who were US-born (18.5%), those who had lived in the United States 10 years or longer (15.7%), and those who preferred English (17.9%). Overall, self-reported CHD and stroke prevalence were low (4.2% and 2.0% in men; 2.4% and 1.2% in women, respectively). In multivariate-adjusted models, hypertension and smoking were directly associated with CHD in both sexes as were hypercholesterolemia and obesity in women and diabetes in men (odds ratios [ORs], 1.5–2.2). For stroke, associations were positive with hypertension in both sexes, diabetes in men, and smoking in women (ORs, 1.7–2.6).
Among US Hispanic/Latino adults of diverse backgrounds, a sizeable proportion of men and women had adverse major risk factors; prevalence of adverse CVD risk profiles was higher among participants with Puerto Rican background, lower SES, and higher levels of acculturation.
PMCID: PMC3777250  PMID: 23117778
17.  Health-Related Quality of Life Differences between African-Americans and Non-Hispanic Whites with Head and Neck Cancer 
Head & neck  2012;35(9):1255-1264.
Cancers of the head and neck are associated with detriments in health-related quality of life (HRQOL), however little is known about different experiences between African Americans and non-Hispanic whites.
HRQOL was measured by the Functional Assessment of Cancer Therapy – Head and Neck approximately five months post diagnosis among 222 cancer patients from North Carolina. Higher scores represent better HRQOL. Regression models included sociodemographic characteristics and clinical factors.
African Americans reported higher Physical Well-Being than Caucasians (adjusted means 23.1 vs 20.9). African Americans with incomes <$20,000 reported higher Emotional Well-Being (21.4) and fewer head and neck symptoms (22.0). Non-Hispanic whites making <$20,000 reported the poorest Emotional Well-Being (17.3) while African Americans making >$20,000 reported the most head and neck symptoms (18.7).
Further investigation is needed to explore variation in HRQOL experiences among different race and socio-economic groups that may inform resource allocation to improve cancer care.
PMCID: PMC3750061  PMID: 22907719
health-related quality of life; head and neck cancer; African Americans
18.  Body Mass Index at age 25 and All-cause Mortality in Whites and African Americans: The Atherosclerosis Risk in Communities Study 
The Journal of Adolescent Health  2011;50(3):221-227.
Approximately 20% of young adults in the United States are obese, and most gain weight between young and middle adulthood. Few studies have examined the association between elevated BMI in early adulthood and mortality or examined such effects independent of changes in weight. We know of no studies in African American samples.
We used data from 13,941 African American and White adults who self-reported their weight at age 25 and had weight and height measured when they were 45-64 years of age (1987-89). Date of death was ascertained from 1987 to 2005. Hazard ratios and hazard differences for the effects of BMI at age 25 on all-cause mortality were determined using Cox proportional hazard and additive hazard models, respectively.
In the combined ethnic-gender groups, the hazard ratio associated with a 5 kg/m2 increment in BMI at age 25 was 1.28 (95% CI: 1.22, 1.35) and the hazard difference was 2.75 (2.01, 3.50) deaths/1,000 person-years. Associations were observed in all four ethnic-gender groups. Models including weight change from age 25 to age in 1987-89 resulted in null estimates for BMI in African American men, while associations were maintained or only mildly attenuated in other ethnic-gender groups.
Excess weight during young adulthood should be avoided as it contributes to increases in death rates that may be independent of changes in weight experienced in later life. Further study is needed to better understand these associations in African American men.
PMCID: PMC3279701  PMID: 22325126
BMI; young adulthood; mortality; White Americans; African Americans; risk ratio; risk difference
19.  Marginal Hazard Regression for Correlated Failure Time Data with Auxiliary Covariates 
Lifetime Data Analysis  2011;18(1):116-138.
In many biomedical studies, it is common that due to budget constraints, the primary covariate is only collected in a randomly selected subset from the full study cohort. Often, there is an inexpensive auxiliary covariate for the primary exposure variable that is readily available for all the cohort subjects. Valid statistical methods that make use of the auxiliary information to improve study efficiency need to be developed. To this end, we develop an estimated partial likelihood approach for correlated failure time data with auxiliary information. We assume a marginal hazard model with common baseline hazard function. The asymptotic properties for the proposed estimators are developed. The proof of the asymptotic results for the proposed estimators is nontrivial since the moments used in estimating equation are not martingale-based and the classical martingale theory is not sufficient. Instead, our proofs rely on modern empirical theory. The proposed estimator is evaluated through simulation studies and is shown to have increased efficiency compared to existing methods. The proposed methods are illustrated with a data set from the Framingham study.
PMCID: PMC3259288  PMID: 22094533
Marginal hazard model; Correlated failure time; Validation set; Auxiliary covariate
20.  Semiparametric Transformation Rate Model for Recurrent Event Data 
Statistics in biosciences  2011;3(2):187-207.
In this article, we propose a class of semiparametric transformation rate models for recurrent event data subject to right-censoring and potentially stopped by a terminating event (e.g., death). These transformation models include both additive rates model and proportional rates model as special cases. Respecting the property that no recurrent events can occur after the terminating event, we model the conditional recurrent event rate given survival. Weighted estimating equations are constructed to estimate the regression coefficients and baseline rate function. In particular, the baseline rate function is approximated by wavelet function. Asymptotic properties of the proposed estimators are derived and a data-dependent criterion is proposed for selecting the most suitable transformation. Simulation studies show that the proposed estimators perform well for practical sample sizes. The proposed methods are used in two real-data examples: a randomized trial of rhDNase and a community trial of Vitamin A.
PMCID: PMC3325027  PMID: 22505954
recurrent event data; transformation model; additive rates model; proportional rates model; terminating event; wavelet approximation
21.  Additive Mixed Effect Model for Clustered Failure Time Data 
Biometrics  2011;67(4):1340-1351.
We propose an additive mixed effect model to analyze clustered failure time data. The proposed model assumes an additive structure and include a random effect as an additional component. Our model imitates the commonly used mixed effect models in repeated measurement analysis but under the context of hazards regression; our model can also be considered as a parallel development of the gamma-frailty model in additive model structures. We develop estimating equations for parameter estimation and propose a way of assessing the distribution of the latent random effect in the presence of large clusters. We establish the asymptotic properties of the proposed estimator. The small sample performance of our method is demonstrated via a large number of simulation studies. Finally, we apply the proposed model to analyze data from a diabetic study and a treatment trial for congestive heart failure.
PMCID: PMC3139827  PMID: 21418052
Additive models; Clustered survival; Goodness of fit; Hazards rate; Moment methods; Random effects
22.  Impact of body mass index levels on lipid abnormalities in Chinese Asians, American Blacks and American Whites: The People’s Republic of China (PRC) and Atherosclerosis Risk in Communities (ARIC) Studies 
Atherosclerosis  2011;218(2):517-523.
Several researchers have reported that Chinese adults may have a greater chronic disease burden than Whites, especially at lower body mass index (BMI) levels.
To compare the incidence of lipid abnormalities in Chinese (n=5,303), White (n=10,752) and Black (n=3,408) middle-aged adults and the effect of BMI on these incidences.
Data were from the People’s Republic of China (PRC) and the Atherosclerosis Risk in Communities (ARIC) studies. In each ethnic group, we calculated the adjusted cumulative incidence for high total cholesterol (≥240 mg/dL), LDL-cholesterol (≥160 mg/dL), and triglycerides (≥200 mg/dL) and low HDL-cholesterol (≤40 in men and ≤50 mg/dL in women) adjusted for age, gender, education, field site, smoking and drinking status. Risk differences associated with BMI (referent=18.5–22.9 kg/m2) were calculated using weighted linear regression and slopes compared using the Wald test.
Chinese had lower incidence of abnormal total cholesterol, LDL-cholesterol and triglycerides than Whites in most BMI groups and had lower incidence of abnormal HDL-cholesterol and triglycerides than Blacks. Across the range of 18.5 to <30, BMI was more strongly associated with the incidence of having high total cholesterol in Chinese and Whites than in Blacks. Similar trends were seen for LDL-cholesterol and triglycerides, but were not always statistically significant. In contrast, BMI was more highly associated with incidence of low HDL-cholesterol in Whites than in Chinese or Blacks.
Although differences in the incidence of lipid abnormalities and the impact of BMI were identified, results varied by lipid type indicating no consistent ethnic/national pattern.
PMCID: PMC3234685  PMID: 21802083
Obesity; Total cholesterol; LDL-cholesterol; HDL-cholesterol; Triglycerides; Ethnicity
23.  Semiparametric inference for a 2-stage outcome-auxiliary-dependent sampling design with continuous outcome 
Biostatistics (Oxford, England)  2011;12(3):521-534.
Two-stage design has long been recognized to be a cost-effective way for conducting biomedical studies. In many trials, auxiliary covariate information may also be available, and it is of interest to exploit these auxiliary data to improve the efficiency of inferences. In this paper, we propose a 2-stage design with continuous outcome where the second-stage data is sampled with an “outcome-auxiliary-dependent sampling” (OADS) scheme. We propose an estimator which is the maximizer for an estimated likelihood function. We show that the proposed estimator is consistent and asymptotically normally distributed. The simulation study indicates that greater study efficiency gains can be achieved under the proposed 2-stage OADS design by utilizing the auxiliary covariate information when compared with other alternative sampling schemes. We illustrate the proposed method by analyzing a data set from an environmental epidemiologic study.
PMCID: PMC3114654  PMID: 21252082
Auxiliary covariate; Kernel smoothing; Outcome-auxiliary-dependent sampling; 2-stage sampling design
24.  Differences in Cardiovascular Disease Risk Factors by Weight History: The Aerobics Center Longitudinal Study 
Obesity (Silver Spring, Md.)  2011;19(10):2063-2068.
The objective of this study was to compare cardiovascular disease (CVD) risk factor levels in adults with a history of weight loss to levels in adults who did not lose weight, after both groups subsequently experienced an approximate 1-year interval of weight maintenance. Extant data from the Aerobics Center Longitudinal Study (ACLS) were used to identify 5,151 adults who were weight maintainers (maintained weight within ±3.0% over two consecutive periods of ~1 year) or weight-loss maintainers (lost >3.0–<5.0% or ≥5.0% of body weight in the first interval and maintained that loss in the second interval). Mixed models regression was used to accommodate repeated measures and adjust for gender, age, smoking, cardiorespiratory fitness, decade of clinic visit, interval length, and BMI at the time of risk factor measurement. Coefficients from the model were used to calculate the adjusted risk factor levels in the three groups. Differences in total cholesterol (−3.8 mg/dl, 95% confidence interval: −5.5, −2.0), low-density lipoprotein (LDL) cholesterol (−3.0 mg/dl, confidence interval: −4.8, −1.1), triglycerides (−6.1 mg/dl, confidence interval: −10.6, −1.7) and diastolic blood pressure (−0.8 mg/dl, confidence interval: −1.4, −0.3) indicated that levels were slightly more favorable in the ≥5.0% weight-loss maintenance group than weight maintenance group. Levels were similar for glucose, high-density lipoprotein (HDL) cholesterol and systolic blood pressure. This work indicates that, when adjusted for covariates including current BMI, adults with a history of weight loss may have CVD risk factors to levels as good, or perhaps even better than, those observed in adults who maintain their weight.
PMCID: PMC3234676  PMID: 21394090
25.  Asymptotic results for fitting marginal hazards models from stratified case-cohort studies with multiple disease outcomes 
In stratified case-cohort designs, samplings of case-cohort samples are conducted via a stratified random sampling based on covariate information available on the entire cohort members. In this paper, we extended the work of Kang & Cai (2009) to a generalized stratified case-cohort study design for failure time data with multiple disease outcomes. Under this study design, we developed weighted estimating procedures for model parameters in marginal multiplicative intensity models and for the cumulative baseline hazard function. The asymptotic properties of the estimators are studied using martingales, modern empirical process theory, and results for finite population sampling.
PMCID: PMC3308729  PMID: 22442642
marginal hazards model; multivariate failure times; stratified case-cohort design; survival analysis; weighted estimating equations

Results 1-25 (50)