Yulei He
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Yulei He has a total of 31 co-authors. Here are the top co-authors by number of shared publications:
- Benmei Liu
National Cancer Institute 1 Bethesda, Md
13 shared publications - Bill Cai
National Center for Health Statistics, U.S. Centers for Disease Control and Prevention, Hyattsville, Maryland
13 shared publications - Diba Khan
Centers for Disease Control and Prevention, Atlanta, Georgia
13 shared publications - Eric J. Feuer
National Cancer Institute 1 Surveillance Research Program, Division of Cancer Control and Population Sciences Bethesda, Maryland
13 shared publications - John R. Pleis
VA Pittsburgh Healthcare System, Pittsburgh
13 shared publications - Machell Town
Division of Population Health, Centers for Disease Control and Prevention, Atlanta, Georgia
13 shared publications
Latest Publications
- Open Access
- 4 August 2026
- Published by SAGE Publications in Journal of Official Statistics
- Vol. 42 (3) , 414-445
Applying Multiple Imputation to Improve Estimates from Web Surveys: An Alternative to the Pseudoweighting Method
To improve the timeliness of data products, survey researchers and practitioners have increasingly used web surveys and alike to collect information for population health research and dissemination. From the statistical inferential perspective, these data are often referred to as nonprobability samples due to the lack of a well-defined probability sampling structure, or they come from probability panel surveys yet are subject to high nonresponse and/or coverage errors. Certain statistical adjustments are therefore needed to make proper inferences using web surveys. With a high-quality reference probability survey available, one popular adjustment approach is to create pseudoweights that properly “weight” the web survey samples back to the target population underlying the reference survey in order to produce population-weighted estimates of the target of interest. When the variable of interest is collected in the web survey but not in the reference survey, the analytical question can also be framed as a missing data problem. Thus we propose to multiply impute the missing variable on the reference survey which combines the information from both data sources. We illustrate main features and performances of the multiple imputation strategy using a simulation study. The simulation results have shown that the multiple imputation analysis method is comparable to the pseudoweighting method, both of which can adequately account for the nonprobability feature of web surveys to improve the statistical inference. We also present a real data analysis based on the web-based Research and Development Survey and the interviewer-administered National Health Interview Survey. Our research shows that results from different imputation and pseudoweighting models can be compared to better understand features of the web survey data and improve the analysis.
- 4 August 2026
- Published by SAGE Publications in Journal of Official Statistics
- Vol. 42 (3) , 414-445
Open AccessApplying Multiple Imputation to Improve Estimates from Web Surveys: An Alternative to the Pseudoweighting Method
To improve the timeliness of data products, survey researchers and practitioners have increasingly used web surveys and alike to collect information for population health research and dissemination. From the statistical inferential perspective, these data are often referred to as nonprobability samples due to the lack of a well-defined probability sampling structure, or they come from probability panel surveys yet are subject to high nonresponse and/or coverage errors. Certain statistical adjustments are therefore needed to make proper inferences using web surveys. With a high-quality reference probability survey available, one popular adjustment approach is to create pseudoweights that properly “weight” the web survey samples back to the target population underlying the reference survey in order to produce population-weighted estimates of the target of interest. When the variable of interest is collected in the web survey but not in the reference survey, the analytical question can also be framed as a missing data problem. Thus we propose to multiply impute the missing variable on the reference survey which combines the information from both data sources. We illustrate main features and performances of the multiple imputation strategy using a simulation study. The simulation results have shown that the multiple imputation analysis method is comparable to the pseudoweighting method, both of which can adequately account for the nonprobability feature of web surveys to improve the statistical inference. We also present a real data analysis based on the web-based Research and Development Survey and the interviewer-administered National Health Interview Survey. Our research shows that results from different imputation and pseudoweighting models can be compared to better understand features of the web survey data and improve the analysis.
- Open Access
- 9 January 2025
- Published by American Association for Cancer Research (AACR)
Supplementary Figure S2 from Small Area Estimation of PSA Testing in US States and Counties
Model diagnosis plots from the 2018-2019 county-level theta model. This figure shows the trace plot, the autocorrelation plot, and the density plot of the MCMC values for the parameter associated with covariate “% of rural areas”.
- 9 January 2025
- Published by American Association for Cancer Research (AACR)
Open AccessSupplementary Figure S2 from Small Area Estimation of PSA Testing in US States and Counties
Model diagnosis plots from the 2018-2019 county-level theta model. This figure shows the trace plot, the autocorrelation plot, and the density plot of the MCMC values for the parameter associated with covariate “% of rural areas”.
- Open Access
- 9 January 2025
- Published by American Association for Cancer Research (AACR)
Data from Small Area Estimation of PSA Testing in US States and Counties
Background:In 2012, the US Preventive Services Task Force recommended against prostate cancer screening using the PSA test for all age groups. In 2018, the US Preventive Services Task Force’s recommendation shifted from a “D” (not recommended) to a “C” (selectively offering PSA-based screening based on professional judgment and patient preferences) in men ages 55 to 69. Limited reliable county-level prostate cancer screening data are available for cancer surveillance purposes.Methods:Utilizing data from the National Health Interview Survey (NHIS) and Behavioral Risk Factor Surveillance System (BRFSS) collected in 2012 to 2019, state- and county-level small area models were developed for estimating PSA testing. Model diagnosis, internal validation, and external validation examining associations of PSA testing and prostate cancer incidence were conducted.Results:Model-based estimates of PSA testing rates were produced for all US states and 3,142 counties for two data periods: 2012 to 2016 and 2018 to 2019. Geographic variations across counties were demonstrated through maps. Moderate positive correlations between PSA-based screening and prostate cancer incidence were observed, e.g., the state-level weighted Pearson’s correlation coefficients were 0.5025 (P value = 0.0002) and 0.3691 (P value = 0.0077) for 2012 to 2016 and 2018 to 2019, respectively.Conclusions:These modeled estimates showed improved precision and adjusted for the differences between the BRFSS and NHIS. The approach of combining the NHIS and BRFSS utilized strengths of the larger sample size of the BRFSS and generally higher response rates and better household coverage from the NHIS.Impact:The resulting small area estimates offer a valuable resource for the cancer surveillance community, aiding in targeted interventions, decision making, and further research endeavors.
- 9 January 2025
- Published by American Association for Cancer Research (AACR)
Open AccessData from Small Area Estimation of PSA Testing in US States and Counties
Background:In 2012, the US Preventive Services Task Force recommended against prostate cancer screening using the PSA test for all age groups. In 2018, the US Preventive Services Task Force’s recommendation shifted from a “D” (not recommended) to a “C” (selectively offering PSA-based screening based on professional judgment and patient preferences) in men ages 55 to 69. Limited reliable county-level prostate cancer screening data are available for cancer surveillance purposes.Methods:Utilizing data from the National Health Interview Survey (NHIS) and Behavioral Risk Factor Surveillance System (BRFSS) collected in 2012 to 2019, state- and county-level small area models were developed for estimating PSA testing. Model diagnosis, internal validation, and external validation examining associations of PSA testing and prostate cancer incidence were conducted.Results:Model-based estimates of PSA testing rates were produced for all US states and 3,142 counties for two data periods: 2012 to 2016 and 2018 to 2019. Geographic variations across counties were demonstrated through maps. Moderate positive correlations between PSA-based screening and prostate cancer incidence were observed, e.g., the state-level weighted Pearson’s correlation coefficients were 0.5025 (P value = 0.0002) and 0.3691 (P value = 0.0077) for 2012 to 2016 and 2018 to 2019, respectively.Conclusions:These modeled estimates showed improved precision and adjusted for the differences between the BRFSS and NHIS. The approach of combining the NHIS and BRFSS utilized strengths of the larger sample size of the BRFSS and generally higher response rates and better household coverage from the NHIS.Impact:The resulting small area estimates offer a valuable resource for the cancer surveillance community, aiding in targeted interventions, decision making, and further research endeavors.
- Open Access
- 9 January 2025
- Published by American Association for Cancer Research (AACR)
Supplementary Data from Small Area Estimation of PSA Testing in US States and Counties
The supplementary material includes details on the rational of input data grouping, the small area models, and model diagnosis and selection.
- 9 January 2025
- Published by American Association for Cancer Research (AACR)
Open AccessSupplementary Data from Small Area Estimation of PSA Testing in US States and Counties
The supplementary material includes details on the rational of input data grouping, the small area models, and model diagnosis and selection.
- Open Access
- 9 January 2025
- Published by American Association for Cancer Research (AACR)
Supplementary Table S3 from Small Area Estimation of PSA Testing in US States and Counties
Empirical percentiles for county-level estimates of Prostate Specific Antigen (PSA) screening rate among men aged 55-69. This table presents summary statistics for county-level NHIS direct estimates, BRFSS direct estimates, and final modeled estimated for the two data periods, 2012-2016 and 2018-2019.
- 9 January 2025
- Published by American Association for Cancer Research (AACR)
Open AccessSupplementary Table S3 from Small Area Estimation of PSA Testing in US States and Counties
Empirical percentiles for county-level estimates of Prostate Specific Antigen (PSA) screening rate among men aged 55-69. This table presents summary statistics for county-level NHIS direct estimates, BRFSS direct estimates, and final modeled estimated for the two data periods, 2012-2016 and 2018-2019.
- Open Access
- 9 January 2025
- Published by American Association for Cancer Research (AACR)
Supplementary Figure S3 from Small Area Estimation of PSA Testing in US States and Counties
Model diagnosis plots from the 2018-2019 county-level delta model. This figure shows the trace plot, the autocorrelation plot, and the density plot of the MCMC values for the parameter associated with covariate “Population density”.
- 9 January 2025
- Published by American Association for Cancer Research (AACR)
Open AccessSupplementary Figure S3 from Small Area Estimation of PSA Testing in US States and Counties
Model diagnosis plots from the 2018-2019 county-level delta model. This figure shows the trace plot, the autocorrelation plot, and the density plot of the MCMC values for the parameter associated with covariate “Population density”.
- Open Access
- 9 January 2025
- Published by American Association for Cancer Research (AACR)
Supplementary Table S1 from Small Area Estimation of PSA Testing in US States and Counties
Final response rates (RR) for the National Health Interview Survey (NHIS) and the Behavioral Risk Factor Surveillance System (BRFSS). This table summarizes the final household-level and adult RR for NHIS across 2013, 2015, 2018 and 2019, along with the state-level minimum, maximum and median RR for BRFSS in 2012, 2014, 2016 and 2018.
- 9 January 2025
- Published by American Association for Cancer Research (AACR)
Open AccessSupplementary Table S1 from Small Area Estimation of PSA Testing in US States and Counties
Final response rates (RR) for the National Health Interview Survey (NHIS) and the Behavioral Risk Factor Surveillance System (BRFSS). This table summarizes the final household-level and adult RR for NHIS across 2013, 2015, 2018 and 2019, along with the state-level minimum, maximum and median RR for BRFSS in 2012, 2014, 2016 and 2018.
- Open Access
- 9 January 2025
- Published by American Association for Cancer Research (AACR)
Supplementary Figure S1 from Small Area Estimation of PSA Testing in US States and Counties
Trend of single year direct estimates of PSA Screening rates (%) and prostate cancer incidence rates for men ages 55-69 from 2012 to 2019. The plots include direct estimates from the NHIS (blue solid line) and the BRFSS (black dash line), the age-adjusted prostate cancer incidence rates (per 100 K population(red dash line), and the aggregated modeled estimates from the county-level model to the national level (green solid line, data presented in the mid-point of each data period, i.e., 2014 and 2018.5, respectively). USPSTF made Grade D recommendation in May 2012 and Grade C recommendation in May 2018 for Prostate cancer screening.
- 9 January 2025
- Published by American Association for Cancer Research (AACR)
Open AccessSupplementary Figure S1 from Small Area Estimation of PSA Testing in US States and Counties
Trend of single year direct estimates of PSA Screening rates (%) and prostate cancer incidence rates for men ages 55-69 from 2012 to 2019. The plots include direct estimates from the NHIS (blue solid line) and the BRFSS (black dash line), the age-adjusted prostate cancer incidence rates (per 100 K population(red dash line), and the aggregated modeled estimates from the county-level model to the national level (green solid line, data presented in the mid-point of each data period, i.e., 2014 and 2018.5, respectively). USPSTF made Grade D recommendation in May 2012 and Grade C recommendation in May 2018 for Prostate cancer screening.
- Open Access
- 9 January 2025
- Published by American Association for Cancer Research (AACR)
Supplementary Table S2 from Small Area Estimation of PSA Testing in US States and Counties
State and county-level covariates pool. This tables lists all potential covariates and their sources, with indicators for the final covariates selected for each theta-model and delta-model in the state- and county-level models, respectively.
- 9 January 2025
- Published by American Association for Cancer Research (AACR)
Open AccessSupplementary Table S2 from Small Area Estimation of PSA Testing in US States and Counties
State and county-level covariates pool. This tables lists all potential covariates and their sources, with indicators for the final covariates selected for each theta-model and delta-model in the state- and county-level models, respectively.
- Open Access
- 9 January 2025
- Published by American Association for Cancer Research (AACR)
Supplementary Figure S6 from Small Area Estimation of PSA Testing in US States and Counties
Model diagnosis plots from the 2018-2019 county-level theta model. This figure shows the trace plot, the autocorrelation plot, and the density plot of the MCMC values for the model precision parameter A1.
- 9 January 2025
- Published by American Association for Cancer Research (AACR)
Open AccessSupplementary Figure S6 from Small Area Estimation of PSA Testing in US States and Counties
Model diagnosis plots from the 2018-2019 county-level theta model. This figure shows the trace plot, the autocorrelation plot, and the density plot of the MCMC values for the model precision parameter A1.