The future of Biostatistics

By Dimitris Rizopoulos and Jeff Leek

Starting in January we took over as co-editors of the journal Biostatistics. We are really excited about the opportunity to give back to our community and help shape the future of our field. We would like to thank Butch Tsiatis and Geert Molenberghs especially for giving us this opportunity and setting such an excellent example as editors of the journal. We would also like to thank the entire editorial board for being so supportive.

Biostatistics was founded in 2000 by editors Scott Zeger and Peter Diggle who wanted to build a journal where biostatisticians could contribute methods and ideas that would substantively advance scientific fields. Some incredible papers have been published in the journal including:

  • The paper introducing one of the most popular normalization methods for microarrays of all time, RMA
  • The paper introducing circular binary segmentation, one of the most widely used methods for DNA copy number inference ever created.
  • The paper that introduced the concept of batch effects and proposed the first solution to this critical problem
  • The paper that introduced an approach for estimating negative binomial dispersions for SAGE data — anticipating the key problems with next generation sequencing.
  • The paper introducing the graphical lasso, one of the most widely used methods for inferring graphs from data.
  • The paper introducing a measure to validate surrogate endpoints in meta analysis of randomized experiments.

Biostatistics was also the first journal to have a reproducibility policy and badge that could be earned for making a paper fully reproducible.

Yet we still think that now is an even more exciting time for the field of biostatistics and for the journal. Health and biomedical data have never been more abundant. Statisticians are playing key roles in the discussions of how to model big data from sequencers and brain scans. Cleaning and processing data remains one of the largest open challenges for biostatistics and data science. As we examine electronic health records and data from wearable computing, the modeling of longitudinal and survival data have never been more important or more difficult.

Biostatisticians are also central to many of the most important scientific problems of our time: reproducibility and replicability, the appropriate use of screening tests, the health impacts of climate change, and the importance of methods and software in scientific research.

We want to make Biostatistics a home for the diversity of contributions that statisticians and data scientists can make in this era. We want to be a home for real solutions that are well documented, well evaluated, and available to the world. We want to be a home to practical solutions even if they aren’t technical and technical solutions that are practical or conceptual advances. We want to be the home to papers that introduce methods, or combine methods in novel ways to solve real problems. We want to be home to papers about the best biostatistical software.

We will do our best to continue the incredible stewardship started by Scott, Peter, Geert, Butch and the entire editorial staff. At the same time we are excited to announce some new initiatives to take advantage of new developments in the world of publishing and science.

Our philosophy as editors is to try to be a home for statistics papers that solve real problems for the greater scientific community. We are open to any major practical or conceptual advance across disciplines from clinical trials, to EMRs, to genomics, to natural language processing and everything in between. We are actively looking for papers that: 1. illustrate a key real problem with current data, 2. propose a solution to that problem and 3. evaluate the solution rigorously and carefully. We are also both major proponents of open access to data and reproducible research so will be looking for papers that drive these standard forward.

Stay tuned as we will be announcing several major initiatives over the coming months to make Biostatistics a place where streamlined submission, fast and fair review, and wide, rapid dissemination are the norm. As a start we are announcing our new Twitter handle @biostatistics and this new blog where we will be posting our updates about the journal. Follow us to hear about all the new developments first — and don’t forget to tell your friends.