Biostatistics FAQ

I submitted a request ticket, what happens now?

What should I expect from the initial consultation?

How should my data be formatted for the analysis?

What services are available for database management?

What is involved in the consultation process?

What if I have a last minute request?

How quickly can I expect to have my results delivered?

What if I have a follow-up analysis request?

What if I have questions about my analysis plan or reported results?

What if my results were not statistically significant?

What if I would like to conduct an interim analysis?

What constitutes clinical research?

What are the expectations regarding authorship for any collaboration?

What educational services are offered beyond statistical support?

How do I provide feedback about my collaboration with a statistician?

 

 

I submitted a request ticket, what happens now?

We assign projects to statisticians on a weekly basis.  You should expect to receive an email from a statistician within 5 business days of your request.

What should I expect from the initial consultation?

At the initial consultation, we will review the statistical services you requested and ensure they meet your study objectives.   Ahead of this meeting, please provide your statistician with your study protocol and data, if collected. Optionally, you may provide a set of analysis ‘table shells’ (i.e., templates of the proposed manuscript tables).

How should my data be formatted for the analysis?

The structure and format of the dataset is very important and needs to be tailored to your research aims. Any concerns regarding data cleaning, coding, and tracking should be addressed in the initial consult. As part of the collaboration, your assigned biostatistician will conduct an initial review of your data for quality assurance. However, if there are systemic errors or inconsistencies, it remains the responsibility of the research team to clean and further prepare the data. A formatted dataset should also include a list of variable names, variable definitions, and all assigned labels.  This data dictionary facilitates the analysis and ensures results are accurate and interpretable.

What services are available for database management?

Ideally, the statistician is involved from the design phase of the study and will help design the database.  We recommend using REDCap for data collection as it provides a secure method of storing data with validation methods to help prevent data entry errors.  [see How should my data be formatted for the analysis?].

What is involved in the consultation process?

We consult on all aspects of services available from our core [see Biostatistics Services]. In general, we welcome collaboration from initial study design and planning to manuscript preparation and review.

What if I have a last minute request?

Depending on the size and scope of the project, the Biostatistics Collaborative Core requires two to four weeks’ notice to meet an external deadline. If the need arises, you should contact the Director of the Biostatistics Collaborative Core, Cara Joyce, to discuss the project’s urgency and desired timeframe (cjoyce6@LUC.edu).

How quickly can I expect to have my results delivered?

At any given time, statisticians are managing 10-20 active projects. Barring exigent circumstances, analyses are prioritized in the order they are received. However, Loyola investigators who fund our statisticians receive immediate priority.  The biostatistician assigned to your project will work with you closely to set a reasonable and well-articulated timeline based on the scope of your study, while remaining sensitive to other investigators’ deadlines.

What if I have a follow-up analysis request?

Follow-up analyses may require an amendment to your IRB approved protocol and an additional project request form. You can submit a CRO Request Form.

What if I have questions about my analysis plan or reported results?

For any questions regarding your analysis or results, please contact your assigned statistician. Should any concerns or questions arise that cannot be addressed by your statistician, contact the Director of the Biostatistics Collaborative Core, Cara Joyce (cjoyce6@LUC.edu).

What if my results were not statistically significant?

If you wish to explore new hypotheses after the analysis is complete, please make sure the new hypotheses are in your protocol aims with an IRB amendment if necessary.  You can then submit a new project request using the CRO Request Form.

What if I would like to conduct an interim analysis?

Hunting for significance is a serious threat to statistical validity.  For prospective data collection for studies that do not pose unknown or significant risks for patients, we believe you and your research team should strive to require as few “looks” at your results as possible, and recommend a single analysis performed after all data is collected.

However, some research requires interim analyses.  This is particularly true in clinical research where the design requires two or more stages of enrollment or where a design requires interim safety monitoring.  It is our policy that all interim analyses are planned before the very first subject is enrolled.  Your consulting statistician will collaborate with you on the importance of the interim analysis plan (IAP) and will generate the IAP for your protocol.  The IAP will list the specific and narrow hypotheses evaluated during an interim analysis.  Typically, these analyses exclude efficacy and instead focus on safety (e.g., adverse event rates, withdrawal rates, etc.).

What constitutes clinical research?

The Biostatistics Collaborative Core supports investigator-initiated clinical and translational research that focuses on health and illness in humans. These include, but are not limited to, efforts to better prevent, screen, diagnose, and manage human outcomes. If you have any questions as to whether your study constitutes clinical research, please request our services using the CRO Request Form, and your assigned statistician will follow-up with you.

What are the expectations regarding authorship for any collaboration?

We welcome co-authoring with you and your team when appropriate.  We recommend authorship be determined as outlined by the International Committee of Medical Journal Editors: http://www.icmje.org/recommendations/browse/roles-and-responsibilities/defining-the-role-of-authors-and-contributors.html

It is advisable the research team agrees to authorship order and manuscript responsibilities before beginning any work on the project or manuscript.  It is important to point out that using our provided methods, results, tables, or figures verbatim and without our permission may be considered plagiarism by some journals.

What educational services are offered beyond statistical support?

Staff in the Biostatistics Collaborative Core support educational and mentorship opportunities across the health sciences campus and often provide basic statistical seminars for students, residents, and clinicians upon request. For more information, please see a full list of core services on the Biostatistics Services page.

How do I provide feedback about my collaboration with a statistician?

As part of our department’s ongoing efforts to improve our services, we welcome your feedback on all collaborations. The information you provide is considered confidential and anonymous, though aggregate information may be shared internally or with department administrators to further improve our core. 

You can provide feedback using the Biostatistics Feedback Form. Otherwise, if you have any questions or concerns, please feel free to contact the Director of the Biostatistics Collaborative Core, Cara Joyce (cjoyce6@LUC.edu).