The Master of Biostatistics program requires a total of 50 credits. 44 credits of graded coursework, a Practicum experience, a Proficiency Examination, and a Master’s Project, for which 6 credits are awarded.
Tracks and Timeline
There are four tracks in the Master of Biostatistics program:
What is the Clinical and Translational Research Track, and what types of students are interested in it?
The Clinical and Translational Research Track can be selected in Year 2 of the program. This track combines instruction in the core of statistical knowledge required of biostatisticians working in interdisciplinary scientific teams with an emphasis on collaboration and communication. It especially appeals to students interested in applying statistics to public health and medicine, including the design and analysis of observational studies and clinical trials.
What kinds of research opportunities are available for students on the Clinical and Translational Research Track?
As an applied track, a common research opportunity involves working under the guidance of a faculty member as the biostatistician on a collaborative research project. The Department of Biostatistics and Bioinformatics benefits from its association with Duke’s School of Medicine, as our faculty, staff, and students offer statistical support to many researchers in biology, medicine, and biomedical informatics, resulting in numerous project opportunities.
Previous students have worked on projects ranging from enhancing MRI sensitivity for breast cancer survivors to conducting statistical analyses on the impact of smoking on a community’s lifespan.
What types of jobs would be available to a Clinical and Translational Track student upon graduation?
Graduates pursue positions in a variety of industries, including academic research centers and hospitals, contract research organizations, pharmaceutical companies, small biotech startups, and federal health agencies. Common job titles are biostatistician and clinical research data analyst.
What types of doctoral programs would be a good fit for Clinical and Translational Track graduates?
Biostatistics and statistics PhD programs that focus on applied or hybrid (applied and theoretical) studies are ideal. Graduates of this track also make strong candidates for PhD programs in related interdisciplinary fields such as epidemiology, computational biology, or population health sciences, where their in-depth knowledge of statistics sets them apart from other applicants.
What is the Biomedical Data Science Track, and what types of students are interested in it?
The Biomedical Data Science (BDS) Track can be chosen in Year 2 of the program. This track attracts students interested in the intersection of computer science, data science, and statistics, especially those focused on applications in health research. Key features of the BDS track include an emphasis on computation, along with topics like workflow best practices, software tools for large biomedical datasets, and data structures and algorithms for data-heavy applications. Although Health Artificial Intelligence (AI) is not a primary focus of this track, BDS students may take elective courses in Health AI with permission from the Director of Graduate Studies. Unlike the Health AI track, no prior experience in data science is necessary; students join the BDS track in their second year after completing prerequisite coursework in their first year.
What kinds of research opportunities might students on the Biomedical Data Science Track have?
Students in the BDS track typically work on health research studies that involve manipulating and processing large datasets, such as Electronic Health Record data, data collected through wearable devices, or genomic data. Many of these studies are observational and require knowledge of causal inference. Data analytic approaches often include techniques like machine learning. While most students do applied work in this area, there are also opportunities for research in developing new methods.
What types of jobs are available to a Biomedical Data Science Track student after graduation?
Employers include academic medical centers, pharmaceutical companies, contract research organizations (CROs), healthcare systems, and technology companies involved in the fields of medicine and public health. Common job titles are data scientist, data analyst, statistical analyst, bioinformatician, or AI/ML engineer.
What types of doctoral programs would be a good fit for Biomedical Data Science Track graduates?
Depending on a student’s areas of focus, they could be well-suited for PhD programs in biostatistics, computational biology, bioinformatics, genomics, computational epidemiology, or even computer science.
What is the Health AI Track, and what types of students are interested in it?
The Health AI Track is for students interested in pursuing a career in developing, evaluating, and deploying AI tools in a healthcare environment, or in pursuing advanced doctoral training in the theoretical underpinnings of AI-based solutions. Coursework focuses on the methodology behind machine learning, managing complex datasets, and developing a better understanding of how to apply this knowledge in a clinical environment. The Health AI (HAI) Track should be selected during the application and enrollment process, as some HAI-specific courses are required in the MB program's first semester.
What kind of background should Health AI Track applicants have to be competitive in the admissions process?
Competitive applicants have a strong foundation in quantitative and computational skills, which are crucial for excelling in the Health AI Track. Prior training in data science and programming is required. Practical experience, gained through working in a research lab or a professional environment, is recommended.
What is the course sequence for the Health AI track and how does it differ from the larger MB program?
Health AI students begin their first semester with two required data science courses. These classes will include some second-year students from the other three tracks who are taking the course as an elective.
Like the larger MB program, students take more elective courses and have the opportunity to take more advanced – often PhD-level – courses in topics ranging from reinforcement learning to the theory of high dimensional data analysis and natural language processing. Students select courses that reflect their goals and interests in consultation with their faculty advisor. Finally, all Health AI students take a second-semester course in Deep Learning for Health Data and a fourth-semester course in Ethical Evaluation and Governance of Clinical AI in Healthcare, and also complete a capstone project.
What kinds of research opportunities are available for students on the Health AI Track?
Faculty work across the spectrum of AI methods and health domains. Methodological research and expertise include: large language models, predictive modeling/clinical decision support, reinforcement learning, causal machine learning, and methods for imaging, natural language processing, and wearables. Application fields span both clinical areas (e.g., mental health, children’s health, women’s health, cardiology) and basic science (e.g., protein modeling, single-cell sequencing, metabolomics). Students may engage in projects exploring the theoretical properties of an AI method, developing new approaches to improve existing AI algorithms, or applying AI techniques to address important health questions and generate new insights. Students have successfully published their work, presented at conferences, and implemented their algorithms within the Duke University Health System.
What types of jobs would be available to a Health AI Track student upon graduation?
With a background in machine learning, health informatics, and predictive analytics, graduates can pursue roles at various health tech companies, startups specializing in medical imaging, diagnostics, or digital health, larger technology firms with AI or health divisions, or hospitals and health systems that develop AI applications to enhance patient outcomes. Common job titles include machine learning scientist, health informatics data scientist, predictive analytics specialist, applied ML engineer, or AI research scientist.
What types of doctoral programs would be a good fit for HAI Track graduates?
Students emerging from the HAI track would be good candidates for Biostatistics programs with an AI/ML research emphasis, programs in computer science, biomedical informatics, health informatics, computational data science, or occasionally for engineering PhD programs with a focus in biomedical engineering or computer engineering.
How does the HAI track differ from the Biomedical Data Science (BDS) track?
The Biomedical Data Science track is another option for students interested in Health AI. During the first year, the BDS track more closely follows the larger MB curriculum, providing a broader foundation in biostatistics methods. In the second year, students choose specialty courses in health data science. As such, it is more appropriate for students who either want a broader training in biostatistics (in conjunction with a focus on data science methods) or those who have not taken as many computing and analytic classes previously. All the same research opportunities are available to BDS (and all MB) students.
What is the Mathematical Statistics Track, and what types of students are interested in it?
The Mathematical Statistics Track can be selected in the second year of the program. Students in this track often enjoy mathematical theory, proofs, and show an interest in evaluating or developing new statistical methods. While the other tracks are more focused on the application of statistics, this track focuses on the theoretical and methodological side of the field. Most students who select this track have a broad background in mathematics at the undergraduate level. Some students who select this track demonstrate an interest in pursuing additional doctoral training, particularly in statistics, applied mathematics, or biostatistics. Others genuinely enjoy the mathematical underpinnings of statistical methods and plan to work after completing their master’s degree.
What kinds of research opportunities might students on the Mathematical Statistics Track have?
Students in this track have ample opportunities to work with faculty members who conduct theoretical research in statistics. For example, previous students have worked on projects comparing statistical methods using both mathematical and simulation-based approaches, while others have collaborated with faculty to derive new theoretical methods.
What types of jobs would be available to a Mathematical Statistics Track student upon graduation?
Various organizations value statistical methodology, so graduates may work in academia, technology, insurance, pharmaceuticals, biotechnology, or government agencies. Companies that have a methodology research group would be good matches for someone with a mathematical statistics background. Common job titles are statistician, quantitative research scientist, statistical consultant, statistical modeling analyst, methodological statistician, or mathematical statistician.
What types of doctoral programs would be a good fit for Mathematical Statistics Track graduates?
Doctoral programs that are mathematically intensive and that focus on developing new statistical methodology are ideal for students in this track. Graduates have pursued PhD training in Statistics, Biostatistics, Applied Mathematics, and Quantitative Sciences.
Practicum
All Master of Biostatistics students must complete a Practicum, providing hands-on experience analyzing a real dataset (not simulated data). The purpose is to strengthen:
- Analytical skills
- Biological understanding
- Communication abilities
Most students complete the Practicum during the summer after Year 1, but it can be scheduled in Year 2 if needed.
Proficiency Examination
Students must pass a written Proficiency Exam demonstrating mastery of core concepts from first-year courses (BIOSTAT 701–706).
- Taken after Year 1 and before starting electives
Master's Project
A two-semester Master’s Project (6 credits) completed in Year 2.
- Demonstrates mastery of statistical concepts and biostatistics practice
Note: The Core foundational courses are required of all degree-seeking students. Full-time Master of Biostatistics students will select six three-credit elective courses during the second year of study.
Course Planning
First Year
(26 graded credit hours of the required 50 graded credit hours)
| Fall | Spring |
|---|---|
Statistical Theory Sequence BIOSTAT 701 (3) or BIOSTAT 701A (3) OR BIOSTAT 830 (3)*
| Statistical Theory Sequence BIOSTAT 704 (3) or BIOSTAT 704A (3) OR BIOSTAT 709 (3) OR BIOSTAT 724 (3)* |
Applied Data Analysis Sequence BIOSTAT 702 (3) or BIOSTAT 702A (3) OR BIOSTAT 707 (3)** | Applied Data Analysis Sequence BIOSTAT 705 (3) or BIOSTAT 705A (3) OR BIOSTAT 826 (3)** |
Practice of Biostatistics Sequence BIOSTAT 703 (3) or BIOSTAT 703A (3) + BIOSTAT 703L (0)*** | Practice of Biostatistics Sequence BIOSTAT 706 (3) or BIOSTAT 706A (3)*** |
Computational Sequence BIOSTAT 721 (3) or BIOSTAT 721A (3) [R] OR BIOSTAT 726 (3)**** | Computational Sequence BIOSTAT 722 (3) or BIOSTAT 722A (3) [SAS] OR BIOSTAT 821 (3) [PYTHON]**** |
Professional Development Sequence BIOSTAT 801 (1)***** | Professional Development Sequence BIOSTAT 802 (1)***** |
| Total: 13 credit hours | Total: 13 credit hours |
Summer
- Master's Proficiency Examination (Required)
Covers content from the first and second semesters of training. - Practicum (Required)
May be completed at any point after the first year.
*Students must take two courses to fulfill the Theory Sequence requirement. In the first year, students choose from BIOSTAT 701/701A or 830 in the fall semester. In the spring semester, students choose 1 of 3 courses (BIOSTAT 704/704A, 709, or 724) to complete their Theory Sequence requirements. Any course not selected may be taken for elective credit in the second year with the Director of Graduate Studies’ approval.
**Students must take two courses to fulfill the Applied Data Analysis sequence. In the first year, students choose from BIOSTAT 702/702A and 707. In the spring semester, students choose between BIOSTAT 705/705A OR 826. The Health AI students take BIOSTAT 707 and 826 to complete their Applied Data Analysis requirements. Students from other tracks may take BIOSTAT 707 for 826 for elective credit in the second year with the Director of Graduate Studies’ approval.
***Students must take BIOSTAT 703/703A and 703L in the fall semester and BIOSTAT 706/706A in the spring semester to fulfill the Practice of Biostatistics sequence requirements.
****Students must take two courses to fulfill the Computing Sequence requirement. In the first year, students choose from BIOSTAT 721/721A or 726 in the fall semester. In the spring semester, students choose BIOSTAT 722/722A or 821. Any computing course not selected may be taken for elective credit in the second year with the Director of Graduate Studies’ approval.
*****To meet program requirements for the Professional Development sequence, students must take BIOSTAT 801 and 802.
Second Year
Second Year (24 credit hours total)
[BIOSTAT 720 - Master’s Project and Statistical Consulting (6 credit hours) plus 18 credit hours of graded coursework]
| Fall | Spring |
|---|---|
BIOSTAT 720 (3) *HAI Students: BIOSTAT 707 is required to meet track requirements. + 3 of the following:
| BIOSTAT 720 (3) *HAI Students: BIOSTAT 826 and 827 are required to meet track requirements. + 3 of the following:
|
| Total: 3 required credit hours plus 9 elective credit hours | Total: 3 required credit hours plus 9 elective credit hours |
Other Duke Departments
- Students must obtain permission from the course instructor to register for courses outside of the department in addition to following all policies and procedures described in the Bulletin and Student Handbook. Enrollment in courses outside the department may be subject to that department’s requirements, policies, and procedures.
Biology:
- BIOLOGY 790S: Graphic Design for Biologists
Computational Biology:
- Any 500 or 600 level except: 510S, 511, or 591
Computer Science:
- Any 500 or 600 level
Engineering:
- Biomedical Engineering: Any 500 or 600 level
- Electrical and Computer Engineering: Any 500 or 600 level except 631
Environmental Sciences (NSOE courses require NSOE department approval in addition to that of the student's Director of Graduate Studies and the course instructor before submitting the NP-1 form):
- ENVIRON 537: Environmental Health and Epidemiology
- ENVIRON 761: Geospatial Analysis for Land and Water Mgmt
Global Health (GLHLT courses require permission from the Director of Graduate Studies):
- GLHLTH 531, 562, 590, or 690
Math
- MATH 531: Real Analysis I
- MATH 574: Quantitative Methods for Biomedical Data
- MATH 721: Linear Algebra and Application
- MATH 731: Introduction to Real Analysis
- MATH 765: Introduction to High Dimensional Data Analysis
- MATH 766: Statistical Learning and Bayesian Nonparametrics
Statistical Sciences
- Any 500 level except 501S
- Any 600 level except 693
- Any 700 level except 701S, 790, 791
- Any 800 level except 851
- Any 900 level except 993, 994, 995
First Year
26 graded coursework credit hours
| Fall | Spring |
|---|---|
| BIOSTAT 701 (3) or BIOSTAT 701A (3) | BIOSTAT 704 (3) or BIOSTAT 704A (3) or BIOSTAT 709 (3) OR BIOSTAT 724* (3) |
BIOSTAT 702 (3) or OR BIOSTAT 707** (3) | BIOSTAT 705 (3) or OR BIOSTAT 826** (3) |
| BIOSTAT 703 (3) or BIOSTAT 703A (3) | BIOSTAT 706 (3) or BIOSTAT 706A (3) |
| BIOSTAT 703L (0) | BIOSTAT 722 (3) or OR BIOSTAT 821*** (3) |
| BIOSTAT 721 (3) or BIOSTAT 721A (3) | BIOSTAT 802 (1) |
| BIOSTAT 801 (1) | N/A |
| Total: 13 credit hours | Total: 13 credit hours |
Summer
- Master's Proficiency Examination (Required)
Covers content from the first and second semesters of training. - Practicum (Required)
May be completed at any point after the first year. - Internship (Optional)
Summer external internships can be completed between the first and second years of training. Students participating in a summer external internship are required to enroll in BIOSTAT 829 (1). This course is permission only. Internships at Duke do not require course registration.
**Students must choose 1 of these 3 courses to complete their Theory Sequence requirements. Any course not selected can be taken for elective credit in the second year with approval from the director of graduate studies.
**Required for Health AI Track to complete the Applied Data Analysis sequence. Students in other tracks may take these courses as electives in their second year with approval from the director of graduate studies.
***To meet program requirements for the Computing sequence, students must take either BIOSTAT 722/722A or BIOSTAT 821.
Second Year
24 credit hours – Master’s project (6) plus graded coursework credit hours (18)
[BIOSTAT 720 - Master’s Project and Statistical Consulting (6 credit hours) plus 18 credit hours of graded coursework]
| Fall | Spring |
|---|---|
| BIOSTAT 720 (3) | BIOSTAT 720 (3) |
+ 3 of the following:
| + 3 of the following:
*Students on the Health AI Track are required |
| Total: 3 required credit hours plus 9 elective credit hours | Total: 3 required credit hours plus 9 elective credit hours |
Other Duke Departments
- Biology: BIOLOGY 790S: Graphic Design for Biologists (3)
- Computational Biology: Any 500 or 600 level except: 510S, 511, or 591
- Computer Science: Any 500 or 600 level
- Electrical and Computer Engineering: Any 500 or 600 level
- Environmental Sciences:
- ENVIRON 537: Environmental Health and Epidemiology (3)
- ENVIRON 761: Geospatial Analysis for Land and Water Mgmt (3)
- **All NSOE courses require NSOE department approval in addition to that of the student’s director of graduate study and the course instructor before submitting the NP-1 form.
- Global Health: GLHLTH 562: Data Science and Data Visualization (3)
- Math:
- MATH 531: Real Analysis I (3)
- MATH 574: Quantitative Methods for Biomedical Data (3)
- MATH 721: Linear Algebra and Application (3)
- MATH 731: Introduction to Real Analysis (3)
- MATH 765: Introduction to High Dimensional Data Analysis (3)
- MATH 766: Statistical Learning and Bayesian Nonparametrics (3)
- Statistical Sciences:
- Any 500 level except 501S
- Any 600 level except 693
- Any 700 level except 701S, 790, 791
- Any 800 level except 851
- Any 900 level except 993, 994, 995