Instructor in Medicine
Brigham and Women's Hospital
Brigham and Women's Hospital
OBC Suite 3030
75 Francis St
Boston MA 02115
My current research focuses on the utilization and outcomes of both the commonly used and newly marketed medical products in pregnant women and their offspring. I have a special interest in linking large existing databases to obtain rich information across multiple real-world data sources (e.g., claims databases, electronic health records, and state birth certificates), and establishing advanced methods (e.g., machine learning-based gestational age prediction and non-live birth identification algorithms), to generate evidence on the use of medical products during pregnancy for clinical practice and post-marketing surveillance.
Available: 12/01/22, Expires: 03/31/23
Project overview: Given the challenges of including pregnant individuals in clinical trials due to ethical considerations and the relatively small sample size when they do, researchers are increasingly using healthcare utilization data to evaluate the magnitude of potential safety concerns related to medication exposure in pregnancy. While knowing the start of pregnancy or the gestational age at birth is essential to determine when during pregnancy the medication exposure occurs, this information is generally not recorded in healthcare utilization databases. International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) coding-based algorithms to estimate gestational age have been developed. Since the US switched to ICD-10-CM codes in October 2015, a new set of gestational age-related ICD-10-CM codes are available, which provides the opportunity to modify the current ICD-9-CM based algorithms, or develop new ICD-10-CM based algorithm for more accurate estimation of gestational age, for both live birth and non-live birth in healthcare utilization databases.
Using the Medicaid claims database, we aim to develop ICD-10-CM based gestational age algorithms, and compare the estimated gestational age to the ICD-9-CM based algorithms that have been converted into ICD-10 coding for both live and non-live birth. We will further identify a subset of pregnancies in the Mass General Brigham Research Patient Data Registry (MGB RPDR) system using a previously established linkage between the Medicaid data and the MGB RPDR for patients receiving care within the MGB system. Based on the gestational age extracted from the medical records, we will estimate the differences between the algorithm estimated gestational age and the gold standard.
Expected student role: Under the guidance of the faculty mentor and by working with the research team (including other investigators, research scientists and programmers under the H4P research group), the student will be responsible for (1) developing the study protocol to estimate gestational age using ICD 10 codes, and (2) working with chart reviewers to extract gestational age from the medical records. Dependent on the involvement and availability of the student, the student may also be responsible for conference abstract and/or manuscript drafting.
Available: 12/01/22, Expires: 02/28/23
Project overview: Healthcare utilization databases have been routinely used to evaluate medication safety in pregnant and pediatric populations. While algorithms for identification of some neonatal outcomes have been well established (i.e., cardiac malformations, neural tube defects, oral clefts, and clubfoot), algorithms for many other clinically important neonatal outcomes have not been previously developed or validated, including the ones that have been reported to be associated with antibiotics and other medications that are frequently used in pregnancy.
Using the linked Medicaid claims database and Mass General Brigham Research Patient Data Registry (MGB RPDR), we aim to develop and validate claims-based algorithms for neonatal outcomes, including urinary tract, genital, gastrointestinal (including pyloric stenosis) and musculoskeletal malformations, and hyperbilirubinemia/ neonatal jaundice. First, we will develop diagnosis and procedure code-based algorithms to identify select neonatal outcomes in Medicaid claims data. As the next step, we will further identify the subset of potential cases in the MGB RPDR system using a previously established linkage between the Medicaid data and the MGB RPDR for patients receiving care within the MGB system. Based on the medical records and standardized clinical criteria, trained chart reviewers will determine the presence or absence of the outcomes. The positive predictive values and their confidence intervals for all outcomes will be estimated.
Expected student role: Under the guidance of the faculty mentor and via working with the research team (including clinicians, research scientist and programmers), the student will be responsible for (1) developing the study protocol to identify neonatal outcomes using claims data, and (2) working with chart reviewers to determine the presence or absence of the outcomes. Depending on the involvement and availability of the student in the project, the student has the opportunity to be responsible for drafting conference abstract, and/or the manuscript.
The research activities and funding listed below are automatically derived from
NIH ExPORTER and other sources, which might result in incorrect or missing items.
to make corrections and additions.
Sep 2, 2022 - May 31, 2027
Comparative Safety of Antibiotics for Common Bacterial Infections During Pregnancy
Role: Principal Investigator
Local representatives can answer questions about the Profiles website or help with editing a profile or issues with profile data. For assistance with this profile: HMS/HSDM faculty should contact contactcatalyst.harvard.edu. For faculty or fellow appointment updates and changes, please ask your appointing department to contact HMS. For fellow personal and demographic information, contact HMS Human Resources at human_resourceshms.harvard.edu. For faculty personal and demographic information, contact HMS Office for Faculty Affairs at facappthms.harvard.edu.