Statistics and Methodology Consulting
To request statistical and methodology consulting, please complete this Consult Request Form.
Advice and Planning
Our biostatisticians collaborate with investigators in planning scientific studies, providing advice on:
- Effective experimental study design
- Efficient data collection and database management
- Appropriate analytical techniques
If you want to ensure that your study will collect useful data and produce publishable results, let us help you on the front end of your study when the critical decisions are being made. Seeking a biostatistician’s advice during the study-planning stage can also lead to quicker, more productive analyses when the study ends.
Data Management and Analysis
Data Management – depending on the needs of your study, we can manage your data collection process from start to finish (see our Clinical Trials page) or we can take a finished database from you (or anything in between).
Analysis – Preventive Medicine is home to 5 PhD-level biostatisticians and numerous other statistical and computer personnel to support the analysis of study data. Our biostatisticians are highly trained in range of analytical methods, including:
- Longitudinal data analysis
- Bayesian data analyses
- Statistical genomics
- Survival analyses
- Mixed effects models
- Generalized Estimating Equations
Although we recommend that you contact us when planning your scientific study, to ensure that the data collected will be useful for testing study hypotheses, in many cases we can also provide support services for data that have already been collected. Many researchers publish medical studies using incorrect statistical tests or misinterpreted statistics. Our experienced and knowledgeable biostatisticians can offer very useful advice to ensure the best and most appropriate analysis of your data.
What is Experimental Study Design?
An experimental study tries to determine whether a treatment has an effect on patients. To do this, it is necessary to have both a treatment and a control group. By having both a group that received the treatment and group that does not, researchers control for the possibility that other factors not related to the treatment are responsible for the difference between the two groups. It is also important that both the treatment and control groups are of adequate size to determine whether an effect took place or not. Finally, it is important to make sure that both groups are statistically similar (except for the treatment).
Before you begin any research, it is important to think through the following types of questions:
- What is your research hypothesis or study question?
- What defines your ‘treatment’?
- What is your primary outcome? Do you have any secondary outcomes?
- Who is your target population? Do you want to look at any population subgroups?
- Who will be your ‘control group’?
- How will you select your treatment and control groups?
- How will you recruit study subjects?
How Can a Biostatistician Help Me with Efficient Data Collection?
Based on their experience with hundreds of research studies, our biostatisticians can help you determine:
- Appropriate primary and secondary outcomes
- Sample size
- Data collection strategies
- Analytical approaches
When these are tailored to your study hypotheses, you won’t waste time or money.
Why is it Important to Select the Appropriate Analytical Technique Before I Collect Data?
Not all analytical techniques are created equal. In many cases, higher impact journals are looking for more sophisticated analyses (e.g. not just a simple t-test). In all cases, journals expect investigators to design studies that are able to truly test their hypotheses. This requires that everyone take the time to think about their analyses ahead of time.
In many cases, the analytical technique to be used placed specific ‘demands’ on the data to be collected. For example, it may require:
- Specific data elements be collected (to be comparable to previous studies)
- Data to be collected at specific points in time
- Data be collected for at least some minimum number of patients (sample size)
- Data be collected for specific types of patients (sample strata)
For all these reasons and more, collecting data before selecting the analytical technique is often similar to ‘…putting the cart before the horse.”
What is the Difference between Consulting and Collaboration?
Consulting requests tend to be relatively focused in nature, often with very specific endpoints in mind, such as an analysis or a paper; as such, consultations are usually of limited duration, ranging from a few weeks to a few months. The University provides some support to the Department of Preventive Medicine so that we can offer these consulting services to faculty and students across the campus on a first come, first serve basis. The intention of consulting support is to stimulate and support research at multiple levels in the university.
Collaborations are generally less focused and more ongoing in nature. They also tend to be directed towards and/or supported by funded research projects. Biostatisticians (and other faculty members of Preventive Medicine) serve as co-investigators, co-PIs and PIs on numerous funded research projects with investigators from various UTHSC schools and departments as well as investigators from around the world. We welcome the opportunity to explore collaborative opportunities with investigators from a wide range of scientific backgrounds.
Karen C. Johnson, M.D., M.P.H., Interim Chair
66 N. Pauline Street, Suite 633
Memphis, TN 38163