What is the job of a Biostatistician?
The job of a biostatistician consists of using statistics to answer questions about health and biology: whether a treatment works, whether a disease is distributed differently among groups of people, or whether a difference seen in a study could be due to chance. It is a support job for research and public health, and it is almost always done inside a team with doctors, nurses, biologists or epidemiologists.
A typical day combines meetings and work at the computer. The biostatistician talks with the people who pose the question of the study, checks how the data were collected, cleans and organizes the datasets, runs analyses with statistical software and writes up the results in reports, tables and charts. Part of the day goes into explaining in plain words what the numbers mean and what their limits are.
Typical tasks
- Helping to design studies and clinical trials, including how many people are needed and how the groups are formed.
- Checking, cleaning and organizing databases of patients, surveys or lab results.
- Choosing and applying statistical methods suited to each type of data.
- Programming the analyses in statistical software and keeping the code tidy so it can be repeated.
- Preparing tables, charts and reports that explain the results and their limitations.
- Taking part in writing scientific papers and in answering the comments of the people who review them.
- Advising other health professionals on how to interpret data and results.
Subjects and content usually covered
- Probability and statistical inference.
- Regression models and survival analysis.
- Design of experiments and clinical trials.
- Epidemiology and the basics of public health.
- Programming and data handling with languages such as R or Python and with spreadsheets.
- Introductory biology, physiology and genetics.
- Research ethics and the confidential handling of people's data.
Usual duration and type of training
There is no single route. A bachelor's degree in statistics, mathematics or a related field usually takes 4 to 5 years, and many positions also ask for a master's degree of 1 to 3 years in biostatistics, public health or applied statistics. In some countries and positions a doctorate of 3 to 6 years is asked for, especially for research or teaching. The length and the names of the qualifications vary by country and institution.
How you get there and what is usually required to practice
There are several ways in. The most common is to study statistics, mathematics, data science or another quantitative field and then specialize in biostatistics with a master's or graduate program. Some people start from a health or biology degree and add strong training in statistics and programming.
A professional license is not usually required for this job, although some positions, such as those in regulated clinical trials, ask for knowledge of good research practice rules and previous experience. The first job is usually found as an assistant or junior analyst within a research team.
Skills that help
- Applying statistical methods and explaining why each one was chosen.
- Programming analyses in R, Python or other software and documenting what was done.
- Spotting errors or odd values in a dataset before analyzing it.
- Translating technical results into clear language for people without statistical training.
- Working with professionals from other fields and understanding their questions.
- Keeping people's data confidential and being honest when reporting uncomfortable results.
Career paths and job opportunities
- Universities and biomedical research centers.
- Hospitals and health institutes with research units.
- Pharmaceutical companies and organizations that run clinical trials.
- Health ministries and departments, and public health agencies.
- International organizations and foundations dedicated to health.
- Consulting firms that provide data analysis services to research teams.
What you are not always told
- Much of the time goes into cleaning and organizing data, not into striking analyses.
- Real data often arrive incomplete or with errors, and sometimes the outcome is that nothing can be concluded.
- It takes patience to explain why a result does not show what someone hoped, and that can create tension in the team.
- The work is sedentary and demands a lot of concentration, with tight deadlines when a report or a paper is due.
- Getting the first job usually requires a graduate degree or project experience, and it can take a while.
How the field is changing
Open programming tools and software that automates part of the analysis mean fewer tasks are repeated by hand and more transparency is asked for about how each result was obtained. Large health databases, electronic records and genetic data are being used more, and with them machine learning techniques. At the same time, the requirements for protecting personal data and for making studies reproducible are increasing.
Want to know if this is the right career for you?
Take the free vocational test with 5 scientific methods — CHASIDE, Holland, Big Five, MMMG and VAK — to get your personalized profile. You can also explore all university careers or learn about the scientific methodology.