Biostatistics
JHBC collaborates with research faculty and staff within the Johns Hopkins University, primarily from the Johns Hopkins Medical Institutions.
These collaborations range from the conceptualization of study design and clarification of study hypotheses for grant submission to dissemination of study results. We offer several avenues for this collaboration depending on the researcher’s needs and interests. The Biostatistics arm of JHBC includes PhD and Masters level biostatisticians and administrative staff. As the practice arm of the Johns Hopkins Department of Biostatistics, we have access to the department’s faculty for additional expertise. We also coordinate with the Data Informatics arm of JHBC (DISC) to enlist their expertise in data capture and management for the best interests of each research study.
What We Offer
- Study design, protocol review and development
- Sample size justification
- IRB submission assistance
- Randomization (from simple to block to complex minimization/covariate-adaptive randomization algorithms)
- Sampling (e.g. nested, case-cohort, matched case-control studies)
- Review of data collection instruments
- General statistical consulting
- Statistical methods selection and implementation
- Statistical analysis and interpretation
- Independent statistical reviews for manuscripts
- Professional and scientific writing (manuscripts, abstracts, presentations)
- Independent DSMB report preparation
- Grant proposal development
Study Design and Protocol Development
- Clinical trials (Phase II, II/III, and III)
- Basic science experiments
- Observational studies (cohort and case-based designs)
- Program evaluation projects
Why and When to Engage Biostatistics
Researchers consider many aspects in developing their study design and its conduct. Since the quality of the research derives from these aspects, it is important to engage biostatistical counsel in this process. JHBC biostatisticians can help a researcher review the design and implementation to ensure that the research questions can be answered as intended, to consider what limitations might result from resources or other study constraints, and to better stand up to critical review.
To be effective, the best time to begin involving the biostatistician in the research study is in the process of study design. This allows for critical thinking and discussion between the researcher and consultant on the study protocol and on the endpoints to be measured in the study. We recognize that starting the discussion process this early is not always feasible, and our biostatisticians work with researchers at various stages of their studies, including identifying confounding factors and potential biases. However, involving a biostatistician too late in the study may not allow for important input to study design, data capture and analyses that could be made to meet the intended study goals.
JHBC offers a range of options to provide biostatistical collaboration and consulting for JHU faculty and staff.
Institute of Clinical and Translational Research (ICTR) - Free Access
Johns Hopkins has been a recipient of a National Institutes of Health (NIH) Clinical and Translational Science Award (CTSA) since 2007. This award allows JHBC to make limited free consultations available to JHU research faculty and staff performing clinical and translational research.
Walk-in Clinic
Clinics are available to Johns Hopkins researchers (Faculty, Staff, Pre-MD, and Post Docs) with short (20-30 minute) questions regarding study design, statistical analysis, or statistical software.
Effective March 16, 2020, and until further notice: Biostatistics Program’s walk-in sessions are moved on-line.
The virtual walk-in clinics will be conducted via MS Teams – Visit the Microsoft Teams Training Site at the regular standing clinic times:
Tuesdays, 1:30- 2:30 pm (Statistical software expertise: SAS)
Wednesdays, 11 am- 12 pm (Statistical software expertise: R)
Thursdays 11 am- 12 pm (Statistical software expertise: STATA)
We ask that researchers email jhbc@jhu.edu one hour prior to the standing clinic time. The names of the first three sign-ups will be forwarded to the expert running the clinic on that day and an email confirmation will be sent to these three individuals. “Session-full” notices will be sent to others. The expert running the clinic will contact the researchers “attending” clinic through MS Teams at the respective times. As usual, we plan to provide an approximately 20-minute consultation to each researcher.
Please, note that our biostatistics and data management team will continue to provide consultations initiated through the web portal. Everything except the consultation itself will continue to be managed via email. If necessary, the consultants will be in contact with the researchers to arrange the best time to meet using one of the university’s suggested on-line meeting platforms: MS Teams or Zoom.
Connection Request
An ICTR request provides a researcher with up to 5 free hours of biostatistics or data management consultation/analysis per research project. These requests can be submitted as a follow-up to a walk-in clinic visit, or when a researcher knows that their questions require more substantive discussion than afforded by the clinic. A separate ICTR request should be submitted for biostatistics and for data management support on the same project.
For a biostatistics request, a biostatistician will be assigned to the request based on the information provided (within 3 business days). If a data management request is also submitted, the biostatistician and data manager will work together to ensure consistency in approach. Biostatisticians are assigned on a rotating schedule . A researcher has the option of requesting a biostatistician with whom they have previously worked through the walk-in clinic or a previous ICTR request. Once the assignment has been made, the biostatistician will contact the researcher (within a week) to set up an initial meeting at which they will discuss and clarify the needs of the research project, and an approach for supporting the request.
If the researcher's request requires more than the 5 free hours, the researcher can coordinate with JHBC for support through a Fee-for-Service or long-term collaboration agreement.
Acknowledgement of NIH
To recognize the assistance provided by the grant, NIH requests that the statement on the ICTR website be included as an acknowledgement in any dissemination of the supported research.
Fee-for-Service Collaborations
When the amount of limited free ICTR support is not adequate for the project, researchers can take advantage of our fee-for-service contracts. In discussion with a senior biostatistician (the JHBC project leader), a project task order (PTO) is developed which includes an estimate for the projected amount of effort (in hours and cost) required between senior and associate level biostatisticians, and data manager, if needed, and delineates the type of tasks that will be provided for that estimate. The researcher will be asked to sign the PTO and to include the financial information needed to bill for the work provided. The PTO is an estimate, and the project is only charged for hours worked by the biostatisticians. The estimate can be amended over time in discussion with the assigned project leader as requirements change in terms of effort/tasks required, the period of time over which the tasks are requested, or funding vehicles. Work will begin once the PTO has been signed and is invoiced monthly. Rates are evaluated annually to defray the costs of operating JHBC as a not-for-profit center within JHSPH.
Long-term Collaborations
For some projects, a researcher may want to have a biostatistician assigned to a project for a period of time on a percent/level of effort (LOE) basis. In grant proposals, these collaborators may be considered key personnel. This type of arrangement is coordinated with the researcher and JHBC's Executive Director to determine a best fit with JHBC's resources and the researcher's needs. Once this has been discussed, the researcher will need to submit a faculty collaboration request, which will then be approved by the assigned biostatistician and the JHBC Director. The researcher will then be contacted by JHBC administrative staff regarding the costs (salary, benefits and computing fees) associated with this type of arrangement. For new grant applications, the biostatistician will coordinate with the researcher to provide an appropriate and most up-to-date biosketch, if needed.
The LOE request should match the effort to be expended and by the size and complexity of the study design and analytical support expected. LOEs can vary across the years/phases of the project.
Collaboration and Consulting Expectations
Consultation and collaboration require a mutually agreed-upon understanding of, and respect for, the expectations that will guide this relationship. These expectations will be part of the initial discussions and may include the following categories.
Making Efficient Use of a Collaboration
The first meeting between the researcher and biostatistician is an opportunity to begin to clearly define the tasks and timelines required. Good communication is important. The better prepared a researcher is in the initial meetings, the more efficient the collaboration can proceed. The following include examples of information the researcher should be prepared to discuss with the biostatistician.
Study Design
Research hypotheses, target, source and sampled population, outcomes and how they are to be measured, potential predictors and confounders
Sample size/Power
Source populations, clinically meaningful effect size for the primary hypothesis, level of precision or power required, estimates of variability and or correlation from the literature or pilot data
Statistical analysis – plan or actual
Study design, research hypotheses, outcomes and predictors to be measured and their distributions, missing data
Timeframe Guidelines
To provide quality work to all of our researchers, the following are guidelines of time to be allowed for a research request. Recognizing that projects and requests differ in their complexity and needs, and that there are competing demands for support, specific timeframe requirements will be discussed with the biostatistician in the initial meetings. The timeframe may be re-evaluated as the request is refined.
- Study design: 2 to 6 weeks
- Statistical analysis: 2 to 8 weeks, depending on the study design, type of analyses and clean data
- Abstract: 2 to 4 weeks, depending on type of analyses and clean data
- Grant preparation: 1 to 3 months, depending on the complexity of study design and required elements
- Manuscript or other dissemination preparation: 3 to 8 weeks after analysis is complete
- Protocol review: 1 to 2 weeks
- Development of data capture/management approach and implementation: 2 to 6 weeks
Authorship
As partners with our collaborators in providing quality dissemination of their research results, it is important to be recognized for the intellectual contributions and scientific assistance provided by the biostatistician(s). The International Committee of Medical Journal Editors (ICJME) has issued guidelines for when the efforts of the biostatistician(s) should be recognized with co-authorship as follows:
- Substantial contributions to the conception or design of the work; or the acquisition, analysis, or interpretation of data for the work
- Drafting the work or revising it critically for important intellectual content
- Final approval of the version to be published
- Agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved
Payment for services is not a substitute for such recognition. Some journals request acknowledgement of each co-author's role, and other institutions have instituted policies for such acknowledgement (e.g., Vanderbilt University). The discussion of co-authorship is part of the collaboration/consultation process.
An additional note about acknowledgements relates to whether the research project's efforts were supported by the NIH CTSA grant through an ICTR request. To recognize the assistance provided by this grant, NIH requests that the following statement be included as an acknowledgement in any dissemination of the supported research.
We would like to acknowledge support for the statistical analysis from the National Center for Research Resources and the National Center for Advancing Translational Sciences (NCATS) of the National Institutes of Health through Grant Number 1UL1TR001079.
Ethics
Research collaboration is based on a set of mutually-shared values between the researcher and the biostatisticians, one of which is ethical statistical practice. We abide by the ethical principles and guidelines for statistical practice as described by the American Statistical Association.