MCHRI Welcomes New Quantitative Science Faculty to Facilitate Maternal and Child Health Research
Thursday, March 12, 2020
By Roxanna Van Norman
The Stanford Maternal and Child Health Research Institute (MCHRI) partners with the Quantitative Sciences Unit (QSU) within the Department of Medicine at Stanford School of Medicine (SoM) to provide support in biostatistics and data management to investigators who are conducting research in maternal or child health. Through this partnership, the Department of Pediatrics recently appointed a new faculty member, Maya Mathur, PhD, to lead this effort.
“We are delighted to welcome Maya to the QSU faculty and MCHRI. She has a distinguished record of innovative research in statistics methods and is recognized for her work on psychosocial and behavioral correlates of health,” says Mary Leonard, MD, MSCE, Director of MCHRI and Professor and Chair of Pediatrics. “She embraces our spirit of team science and will provide expert advice on study design, statistical analyses and interpretation of results in partnership with maternal and child health investigators.”
Dr. Mathur has a joint appointment in Pediatrics and the Biostatistics unit within the Department of Medicine’s Division of Biomedical Informatics. She joined Stanford in 2019 after completing her postdoctoral fellowship at Harvard in the Department of Epidemiology, where she received her doctorate in biostatistics. She graduated from Stanford with a master’s degree in statistics in 2013 and has collaborated extensively with investigators at Stanford and Harvard to provide her expertise on study design, statistical analysis, and the interpretation of results.
Since stepping into this new role at QSU, Dr. Mathur already has plans for long-term projects and collaborations. Her current statistical and substantive research is geared toward solving problems relevant to maternal and child health. She has interests in developing methods for the analysis of meta-analyses and of replication studies. In particular, Dr. Mathur focuses on evidence syntheses, a systematic process for bringing together complex information from various sources. This is important for fields where sample sizes may be limited, which is often the case in maternal and child research.
An interdisciplinary collaborative approach
The QSU was founded in 2009 in the Department of Medicine by Manisha Desai, PhD, Professor of Medicine and of Biomedical Data Science. Originally, the QSU was developed to exclusively collaborate with investigators who had a primary appointment in Department of Medicine.
“There was incredibly interesting science going on throughout the School of Medicine that we wanted to be part of and that could benefit from data science expertise. We, therefore, developed a model to enable partnerships with entities outside of the Department of Medicine,” says Dr. Desai. “We have a collaborative philosophy and have established partnerships with leaders who adopt this philosophy which involves engaging us as peer co-scientists.”
An important role for Dr. Mathur will be to serve as the point person for facilitating all maternal and child health-related research on campus.
Today, QSU is part of a larger collection of data science resources within SoM and comprises of over 25 data scientists who provide expertise in biostatistics, health informatics, computational biology, data management, as well as mentoring and education in research methods for both clinical and data scientists. Through an annual subscription model, departments, centers, or institutes, like MCHRI, across campus may partner with the QSU to meet data science-related needs. This involves a deep-level collaboration between affiliated investigators and QSU data scientists on clinical and basic science initiatives.
“Dr. Leonard embraces this idea, and as part of our partnership, has invested in the career development of a data science faculty member, Dr. Mathur,” says Dr. Desai. “An important role for Dr. Mathur will be to serve as the point person for facilitating all maternal and child health-related research on campus.”
Biostatistics and data management support for MCHRI members
As part of the formal partnership between QSU and MCHRI, all MCHRI members and grant applicants may request a consultation session as needed to collaborate with on their studies, including grant proposal development, manuscript preparation, and project implementation.
“Thanks to Mary Leonard’s efforts as well as the leadership at the QSU, people who are affiliated with MCHRI are beginning to understand the value of this type of model,” says Dr. Mathur.
To initiate any scientific involvement from the QSU, investigators should complete a project initiation form through their online portal. MCHRI members or their mentees must also submit this request directly to the QSU. For grant proposals, this form must be done at least three months before the sponsor's deadline and at least six weeks before Stanford internal funding deadline. Once the request is received, Dr. Mathur will reach out to investigators to set up an initial meeting.
Through QSU’s approach, data scientists like Dr. Mathur are fully integrated into the research teams. Each works closely with investigators on the scientific details of the project. This type of deep-level collaboration is especially important in maternal and pediatric studies, where typically the sample sizes are fairly small, complicating the study design, analysis, and interpretation of findings.
“The goal for us as statisticians is to be involved from the earliest stages of the design, all the way through manuscript writing,” says Dr. Mathur. “Having statistical support early on means that the project is much stronger from the initiation.”
She looks forward to working closely with investigators of all disciplines who are working on projects related to maternal or child health research. One piece of advice she would give investigators is to thoroughly develop their reason for "why" they are conducting their research. Oftentimes, she says that investigators start the planning process for their study with wanting to find correlation among their datasets or address feasibility concerns right away.
“The first step is to be really clear on what is the question that you really would like to address and why,” says Dr. Mathur. “Keeping that structure clear in one's mind will help with addressing the scientific and clinical value of what you're trying to do.”
For more information about the QSU and its partnership with MCHRI, please visit the website.
Roxanna Van Norman is the marketing manager for the Stanford Maternal and Child Health Research Institute.