Sarah C. Lotspeich
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
Sarah C. Lotspeich has a total of 88 co-authors. Here are the top co-authors by number of shared publications:
Bryan E. Shepherd
Department of Biostatistics Vanderbilt University Medical Center Nashville Tennessee USA
5 shared publications- Tanya P. Garcia
Department of Biostatistics, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA
4 shared publications Catherine C. McGowan
Vanderbilt Univ, Med Ctr, Dept Med, Div Infect Dis, Nashville, TN USA
3 shared publications- Gustavo Amorim
Department of Biostatistics, Vanderbilt University Medical Center
3 shared publications - Lucy D'Agostino McGowan
Wake Forest University, Mathematics and Statistics
3 shared publications - Ran Tao
Department of Biostatistics, Vanderbilt University Medical Center, Nashville, USA
3 shared publications
Latest Publications
- Open Access
- 1 July 2026
- Published by Oxford University Press (OUP) in Biometrics
- Vol. 82 (3)
Symmetric vaccine efficacy
Traditional measures of vaccine efficacy (VE) are inherently asymmetric, constrained above by 1 but unbounded below. As a result, VE estimates and corresponding confidence intervals (CIs) can extend far below zero, making interpretation difficult and potentially obscuring whether the apparent effect reflects true harm or simply statistical uncertainty. The proposed symmetric vaccine efficacy (SVE) is a bounded and interpretable alternative to VE that maintains desirable statistical properties while resolving these asymmetries. SVE is defined as a symmetric transformation of infection risks, with possible values within $[-1, 1]$, providing a common scale for both beneficial and harmful vaccine effects. This paper describes the relationship between SVE and traditional VE, considers inference about SVE, and illustrates the utility of the proposed measure by reanalyzing data from a randomized trial of a candidate HIV vaccine. Open-source tools for computing estimates of SVE and corresponding CIs are available in R through the sve package.- 1 July 2026
- Published by Oxford University Press (OUP) in Biometrics
- Vol. 82 (3)
Open AccessSymmetric vaccine efficacy
Traditional measures of vaccine efficacy (VE) are inherently asymmetric, constrained above by 1 but unbounded below. As a result, VE estimates and corresponding confidence intervals (CIs) can extend far below zero, making interpretation difficult and potentially obscuring whether the apparent effect reflects true harm or simply statistical uncertainty. The proposed symmetric vaccine efficacy (SVE) is a bounded and interpretable alternative to VE that maintains desirable statistical properties while resolving these asymmetries. SVE is defined as a symmetric transformation of infection risks, with possible values within $[-1, 1]$, providing a common scale for both beneficial and harmful vaccine effects. This paper describes the relationship between SVE and traditional VE, considers inference about SVE, and illustrates the utility of the proposed measure by reanalyzing data from a randomized trial of a candidate HIV vaccine. Open-source tools for computing estimates of SVE and corresponding CIs are available in R through the sve package. - Open Access
- 14 June 2026
- Published by Taylor & Francis in Statistics and Public Policy
- Vol. 13 (1)
Linking Potentially Misclassified Healthy Food Access to Diabetes Prevalence
- 14 June 2026
- Published by Taylor & Francis in Statistics and Public Policy
- Vol. 13 (1)
Open AccessLinking Potentially Misclassified Healthy Food Access to Diabetes Prevalence
- Open Access
- 5 May 2026
- Published by Oxford University Press (OUP) in JAMIA Open
- Vol. 9 (3) , ooag080
Using large language models to enhance clinically-driven missing data recovery algorithms in electronic health records
Objectives: Electronic health record (EHR) data are prone to missingness and errors. Previously, we devised an enriched chart review protocol where a “roadmap” of auxiliary diagnoses was used to recover missing values. Still, chart reviews are expensive and time-intensive, limiting the number of patients whose data can be reviewed. Now, we investigate the accuracy and scalability of a roadmap-driven algorithm, based on International Classification of Diseases, 10th revision (ICD-10) codes, to mimic expert chart reviews and recover missing values. Materials and Methods: In addition to the clinicians’ original roadmap from our previous work, we consider new versions that were iteratively refined using large language models (LLMs) in conjunction with clinical expertise to expand the list of auxiliary diagnoses. Using chart reviews for 100 patients from an extensive EHR, we examine algorithm performance. Results: Across 100 chart reviewed patients, there were 413 missing values in the EHR data. The expert chart reviews recovered 49 (12%), while the algorithms using LLMs-enhanced roadmaps recommended almost twice as many (83-89, 20%-22%). The final algorithm using clinician-approved LLMs’ additions offered a balance (73, 18%), expanding the original roadmap with LLMs’ suggestions but only when deemed clinically relevant. When applied to a larger study of 1000 patients from the same EHR, the per-patient median number of non-missing values increased from 6 to 7. Discussion: Clinically-driven algorithms (enhanced by LLMs) can recover missing EHR data with similar accuracy to chart reviews and feasibly be applied to large samples. Extending them to monitor other data quality dimensions is a promising future direction.
- 5 May 2026
- Published by Oxford University Press (OUP) in JAMIA Open
- Vol. 9 (3) , ooag080
Open AccessUsing large language models to enhance clinically-driven missing data recovery algorithms in electronic health records
Objectives: Electronic health record (EHR) data are prone to missingness and errors. Previously, we devised an enriched chart review protocol where a “roadmap” of auxiliary diagnoses was used to recover missing values. Still, chart reviews are expensive and time-intensive, limiting the number of patients whose data can be reviewed. Now, we investigate the accuracy and scalability of a roadmap-driven algorithm, based on International Classification of Diseases, 10th revision (ICD-10) codes, to mimic expert chart reviews and recover missing values. Materials and Methods: In addition to the clinicians’ original roadmap from our previous work, we consider new versions that were iteratively refined using large language models (LLMs) in conjunction with clinical expertise to expand the list of auxiliary diagnoses. Using chart reviews for 100 patients from an extensive EHR, we examine algorithm performance. Results: Across 100 chart reviewed patients, there were 413 missing values in the EHR data. The expert chart reviews recovered 49 (12%), while the algorithms using LLMs-enhanced roadmaps recommended almost twice as many (83-89, 20%-22%). The final algorithm using clinician-approved LLMs’ additions offered a balance (73, 18%), expanding the original roadmap with LLMs’ suggestions but only when deemed clinically relevant. When applied to a larger study of 1000 patients from the same EHR, the per-patient median number of non-missing values increased from 6 to 7. Discussion: Clinically-driven algorithms (enhanced by LLMs) can recover missing EHR data with similar accuracy to chart reviews and feasibly be applied to large samples. Extending them to monitor other data quality dimensions is a promising future direction.
- Open Access
- 26 December 2025
- Published by Springer Nature in Health Services and Outcomes Research Methodology
- Vol. 26 (1) , 108-138
Examining the impacts of measurement error in quantifying health disparities: a case study on type-2 diabetes and the food environment
Disparities in healthy eating relate to disparities in well-being, leading to disproportionate rates of diseases like type-2 diabetes in communities that face more challenges in accessing nutritious food. They can be driven by individual- and neighborhood-level factors, like a person’s distance from home to the nearest grocery store or the socioeconomic status of their community, respectively. Quantifying these disparities is key to developing targeted interventions, and there are limitations with the currently available methods and data that we are working to resolve. Namely, available data on disease rates are usually aggregate, which smooths over details about the individuals and communities within them. Further, aggregate disease data often comprise small area estimates, which carry additional uncertainty. In this project, we investigate the relationship between patients’ food environment and the risk of diabetes using individual-level data from electronic health records at a large academic medical center. Using various health disparities methods, we quantify whether patients with worse access to healthy food or more food-insecure households in their neighborhood face a higher burden of prevalent type-2 diabetes. Still, we have measurement error in the food environment variables (access and food insecurity) since they are collected using inaccurate distance calculations and survey data. Finally, we discuss the impact of using error-prone food environment measures to detect health disparities in these data.
- 26 December 2025
- Published by Springer Nature in Health Services and Outcomes Research Methodology
- Vol. 26 (1) , 108-138
Open AccessExamining the impacts of measurement error in quantifying health disparities: a case study on type-2 diabetes and the food environment
Disparities in healthy eating relate to disparities in well-being, leading to disproportionate rates of diseases like type-2 diabetes in communities that face more challenges in accessing nutritious food. They can be driven by individual- and neighborhood-level factors, like a person’s distance from home to the nearest grocery store or the socioeconomic status of their community, respectively. Quantifying these disparities is key to developing targeted interventions, and there are limitations with the currently available methods and data that we are working to resolve. Namely, available data on disease rates are usually aggregate, which smooths over details about the individuals and communities within them. Further, aggregate disease data often comprise small area estimates, which carry additional uncertainty. In this project, we investigate the relationship between patients’ food environment and the risk of diabetes using individual-level data from electronic health records at a large academic medical center. Using various health disparities methods, we quantify whether patients with worse access to healthy food or more food-insecure households in their neighborhood face a higher burden of prevalent type-2 diabetes. Still, we have measurement error in the food environment variables (access and food insecurity) since they are collected using inaccurate distance calculations and survey data. Finally, we discuss the impact of using error-prone food environment measures to detect health disparities in these data.
- Open Access
On Using Large Language Models to Enhance Clinically-Driven Missing Data Recovery Algorithms in Electronic Health Records
Objective: Electronic health records (EHR) data are prone to missingness and errors. Previously, we devised an "enriched" chart review protocol where a "roadmap" of auxiliary diagnoses (anchors) was used to recover missing values in EHR data (e.g., a diagnosis of impaired glycemic control might imply that a missing hemoglobin A1c value would be considered unhealthy). Still, chart reviews are expensive and time-intensive, which limits the number of patients whose data can be reviewed. Now, we investigate the accuracy and scalability of a roadmap-driven algorithm, based on ICD-10 codes (International Classification of Diseases, 10th revision), to mimic expert chart reviews and recover missing values. Materials and Methods: In addition to the clinicians' original roadmap from our previous work, we consider new versions that were iteratively refined using large language models (LLM) in conjunction with clinical expertise to expand the list of auxiliary diagnoses. Using chart reviews for 100 patients from the EHR at an extensive learning health system, we examine algorithm performance with different roadmaps. Using the larger study of $1000$ patients, we applied the final algorithm, which used a roadmap with clinician-approved additions from the LLM. Results: The algorithm recovered as much, if not more, missing data as the expert chart reviewers, depending on the roadmap. Discussion: Clinically-driven algorithms (enhanced by LLM) can recover missing EHR data with similar accuracy to chart reviews and can feasibly be applied to large samples. Extending them to monitor other dimensions of data quality (e.g., plausability) is a promising future direction.Open AccessOn Using Large Language Models to Enhance Clinically-Driven Missing Data Recovery Algorithms in Electronic Health Records
Objective: Electronic health records (EHR) data are prone to missingness and errors. Previously, we devised an "enriched" chart review protocol where a "roadmap" of auxiliary diagnoses (anchors) was used to recover missing values in EHR data (e.g., a diagnosis of impaired glycemic control might imply that a missing hemoglobin A1c value would be considered unhealthy). Still, chart reviews are expensive and time-intensive, which limits the number of patients whose data can be reviewed. Now, we investigate the accuracy and scalability of a roadmap-driven algorithm, based on ICD-10 codes (International Classification of Diseases, 10th revision), to mimic expert chart reviews and recover missing values. Materials and Methods: In addition to the clinicians' original roadmap from our previous work, we consider new versions that were iteratively refined using large language models (LLM) in conjunction with clinical expertise to expand the list of auxiliary diagnoses. Using chart reviews for 100 patients from the EHR at an extensive learning health system, we examine algorithm performance with different roadmaps. Using the larger study of $1000$ patients, we applied the final algorithm, which used a roadmap with clinician-approved additions from the LLM. Results: The algorithm recovered as much, if not more, missing data as the expert chart reviewers, depending on the roadmap. Discussion: Clinically-driven algorithms (enhanced by LLM) can recover missing EHR data with similar accuracy to chart reviews and can feasibly be applied to large samples. Extending them to monitor other dimensions of data quality (e.g., plausability) is a promising future direction. - Open Access
- 2 September 2025
- Published by Elsevier in Journal of Biomedical Informatics
- Vol. 170, 104904
Overcoming data challenges through enriched validation and targeted sampling to measure whole-person health in electronic health records
- 2 September 2025
- Published by Elsevier in Journal of Biomedical Informatics
- Vol. 170, 104904
Open AccessOvercoming data challenges through enriched validation and targeted sampling to measure whole-person health in electronic health records
- Open Access
- 5 August 2025
- Published by The Open Journal in Journal of Open Source Software
- Vol. 10 (112) , 7320
sleev: An R Package for Semiparametric Likelihood Estimation with Errors in Variables
Data with measurement error in the outcome, covariates, or both are not uncommon, particularly with the increased use of routinely collected data for biomedical research. With error-prone data, often only a subsample of study data is validated; such settings are known as two-phase studies. The sieve maximum likelihood estimator (SMLE), which combines the error-prone data on all records with the validated data on a subsample, is a highly efficient and robust method to analyze such data. However, given their complexity, a computationally efficient and user-friendly tool is needed to obtain the SMLEs. The R package sleev fills this gap by making semiparametric likelihood-based inference using the SMLEs for error-prone two-phase data in settings with binary and continuous outcomes. Functions from this package can be used to analyze data with error-prone binary or continuous responses and error-prone covariates.
- 5 August 2025
- Published by The Open Journal in Journal of Open Source Software
- Vol. 10 (112) , 7320
Open Accesssleev: An R Package for Semiparametric Likelihood Estimation with Errors in Variables
Data with measurement error in the outcome, covariates, or both are not uncommon, particularly with the increased use of routinely collected data for biomedical research. With error-prone data, often only a subsample of study data is validated; such settings are known as two-phase studies. The sieve maximum likelihood estimator (SMLE), which combines the error-prone data on all records with the validated data on a subsample, is a highly efficient and robust method to analyze such data. However, given their complexity, a computationally efficient and user-friendly tool is needed to obtain the SMLEs. The R package sleev fills this gap by making semiparametric likelihood-based inference using the SMLEs for error-prone two-phase data in settings with binary and continuous outcomes. Functions from this package can be used to analyze data with error-prone binary or continuous responses and error-prone covariates.
- Open Access
- 7 July 2025
- Published by Springer Nature in Gender Issues
- Vol. 42 (3) , 1-28
‘Who’ Makes a ‘Good’ Leader? Examining the Influence of Leader Gender with Perceptions of Leader Competency and Employee Outcomes
Gender stereotypes suggest men are better fit in roles of leadership. The multi-billion-dollar sport industry is an example of this trend. The extant literature is limited in its examinations of the validity of these stereotypes that negatively influence a woman’s ability to obtain leadership roles in highly men-dominated industries. To address this gap, the current study asked U.S. intercollegiate administrators and coaches to evaluate their supervisors’ leadership capabilities across 15 previously established and validated leadership competencies. The current study sought to better understand the association, if any, of a leader’s gender on their perceived leadership competencies as reported on by their employees. We employed ordinal regression modeling to examine the association between competency evaluations for supervisor gender and the outcome variables of subordinate job satisfaction and organizational commitment. Findings revealed women leaders were rated higher across all leadership competencies by subordinates. Thus, results suggest men are not inherently better fit than women to lead in sport industry settings despite their immense overrepresentation in these leadership roles.
- 7 July 2025
- Published by Springer Nature in Gender Issues
- Vol. 42 (3) , 1-28
Open Access‘Who’ Makes a ‘Good’ Leader? Examining the Influence of Leader Gender with Perceptions of Leader Competency and Employee Outcomes
Gender stereotypes suggest men are better fit in roles of leadership. The multi-billion-dollar sport industry is an example of this trend. The extant literature is limited in its examinations of the validity of these stereotypes that negatively influence a woman’s ability to obtain leadership roles in highly men-dominated industries. To address this gap, the current study asked U.S. intercollegiate administrators and coaches to evaluate their supervisors’ leadership capabilities across 15 previously established and validated leadership competencies. The current study sought to better understand the association, if any, of a leader’s gender on their perceived leadership competencies as reported on by their employees. We employed ordinal regression modeling to examine the association between competency evaluations for supervisor gender and the outcome variables of subordinate job satisfaction and organizational commitment. Findings revealed women leaders were rated higher across all leadership competencies by subordinates. Thus, results suggest men are not inherently better fit than women to lead in sport industry settings despite their immense overrepresentation in these leadership roles.
- Open Access
- 3 April 2025
- Published by Springer Nature in Research Square
Examining the Impacts of Measurement Error in Quantifying Health Disparities: A Case Study on Type-2 Diabetes and the Food Environment
Disparities in healthy eating relate to disparities in well-being, leading to disproportionate rates of diseases like type-2 diabetes in communities that face more challenges in accessing nutritious food. They can be driven by individual- and neighborhood-level factors, like a person’s distance from home to the nearest grocery store or the socioeconomic status of their community, respectively. Quantifying these disparities is key to developing targeted interventions, and there are limitations with the currently available methods and data that we are working to resolve. Namely, available data on disease rates are usually aggregate, which smooths over details about the individuals and communities within them. Further, aggregate disease data often comprise small area estimates, which carry additional uncertainty. In this project, we investigate the relationship between patients' food environment and the risk of diabetes using individual-level data from electronic health records at a large academic medical center. Using various health disparities methods, we quantify whether patients with worse access to healthy food or more food-insecure households in their neighborhood face a higher burden of prevalent type-2 diabetes. Still, we face measurement error in the food environment variables (access and food insecurity) since they are collected using inaccurate distance calculations and survey data. Finally, we discuss the impact of using error-prone food environment measures to detect health disparities in these data.
- 3 April 2025
- Published by Springer Nature in Research Square
Open AccessExamining the Impacts of Measurement Error in Quantifying Health Disparities: A Case Study on Type-2 Diabetes and the Food Environment
Disparities in healthy eating relate to disparities in well-being, leading to disproportionate rates of diseases like type-2 diabetes in communities that face more challenges in accessing nutritious food. They can be driven by individual- and neighborhood-level factors, like a person’s distance from home to the nearest grocery store or the socioeconomic status of their community, respectively. Quantifying these disparities is key to developing targeted interventions, and there are limitations with the currently available methods and data that we are working to resolve. Namely, available data on disease rates are usually aggregate, which smooths over details about the individuals and communities within them. Further, aggregate disease data often comprise small area estimates, which carry additional uncertainty. In this project, we investigate the relationship between patients' food environment and the risk of diabetes using individual-level data from electronic health records at a large academic medical center. Using various health disparities methods, we quantify whether patients with worse access to healthy food or more food-insecure households in their neighborhood face a higher burden of prevalent type-2 diabetes. Still, we face measurement error in the food environment variables (access and food insecurity) since they are collected using inaccurate distance calculations and survey data. Finally, we discuss the impact of using error-prone food environment measures to detect health disparities in these data.
- Open Access
- 30 March 2025
- Published by Wiley in Statistics in Medicine
- Vol. 44 (7) , e70054
Combining Straight‐Line and Map‐Based Distances to Investigate the Connection Between Proximity to Healthy Foods and Disease
Healthy foods are essential for a healthy life, but accessing healthy food can be more challenging for some people than others. This disparity in food access may lead to disparities in well‐being, potentially with disproportionate rates of diseases in communities that face more challenges in accessing healthy food (i.e., low‐access communities). Identifying low‐access, high‐risk communities for targeted interventions is a public health priority, but current methods to quantify food access rely on distance measures that are either computationally simple (like the length of the shortest straight‐line route) or accurate (like the length of the shortest map‐based driving route), but not both. We propose a multiple imputation approach to combine these distance measures, allowing researchers to harness the computational ease of one with the accuracy of the other. The approach incorporates straight‐line distances for all neighborhoods and map‐based distances for just a subset, offering comparable estimates to the “gold standard” model using map‐based distances for all neighborhoods and improved efficiency over the “complete case” model using map‐based distances for just the subset. Through the adoption of a measurement error framework, information from the straight‐line distances can be leveraged to compute informative placeholders (i.e., impute) for any neighborhoods without map‐based distances. Using simulations and data for the Piedmont Triad region of North Carolina, we quantify and compare the associations between two health outcomes (diabetes and obesity) and neighborhood‐level access to healthy foods. The imputation procedure also makes it possible to predict the full landscape of food access in an area without requiring map‐based measurements for all neighborhoods.
- 30 March 2025
- Published by Wiley in Statistics in Medicine
- Vol. 44 (7) , e70054
Open AccessCombining Straight‐Line and Map‐Based Distances to Investigate the Connection Between Proximity to Healthy Foods and Disease
Healthy foods are essential for a healthy life, but accessing healthy food can be more challenging for some people than others. This disparity in food access may lead to disparities in well‐being, potentially with disproportionate rates of diseases in communities that face more challenges in accessing healthy food (i.e., low‐access communities). Identifying low‐access, high‐risk communities for targeted interventions is a public health priority, but current methods to quantify food access rely on distance measures that are either computationally simple (like the length of the shortest straight‐line route) or accurate (like the length of the shortest map‐based driving route), but not both. We propose a multiple imputation approach to combine these distance measures, allowing researchers to harness the computational ease of one with the accuracy of the other. The approach incorporates straight‐line distances for all neighborhoods and map‐based distances for just a subset, offering comparable estimates to the “gold standard” model using map‐based distances for all neighborhoods and improved efficiency over the “complete case” model using map‐based distances for just the subset. Through the adoption of a measurement error framework, information from the straight‐line distances can be leveraged to compute informative placeholders (i.e., impute) for any neighborhoods without map‐based distances. Using simulations and data for the Piedmont Triad region of North Carolina, we quantify and compare the associations between two health outcomes (diabetes and obesity) and neighborhood‐level access to healthy foods. The imputation procedure also makes it possible to predict the full landscape of food access in an area without requiring map‐based measurements for all neighborhoods.