Browse by Medical Category
Massachusetts General Hospital provides resources to help clinical researchers establish and refine their preliminary study designs.
Excellent research begins with a clear hypothesis, which can be either supported or rejected through the generation and analysis of new data collected through clinical research.
This data should be subjected to statistical analysis, and the results should either confirm or reject the original hypothesis.
To create a strong hypothesis, start with a compelling and scientifically important research question—one for which the answer will matter to other researchers in the field, practicing clinicians or patients.
Convert your research question into a hypothesis by asserting a position. This will lead directly to a consideration of measures, both of exposure and outcome.
The null hypothesis is a statement a researcher wants to test. In general, the null hypothesis is the same as a theoretical expectation. Analysis is performed to reject or accept the null hypothesis.
A review of the pertinent literature can help you generate your expected measures of exposure and outcome. Here are some things to consider when you conduct a literature review:
The effort to generate meaningful measures will generally require a return to the hypothesis for refinement and narrowing (i.e., expressing the hypothesis in terms of the specific exposure of interest and the specific outcome anticipated).
Once a reasonable hypothesis is developed, it is important to consider how it will be tested.
The first requirement is that a comparison be made.
Choosing the right control group is a challenging and subtle task. Once the comparison group is chosen, the magnitude of expected difference should be estimated as a basis for determining sample size (power calculation).
Engage the support of a statistician to ensure a proper approach to this process. A statistician will help determine measures, sample size, what endpoints can be reasonably expected and what tests should be used for your analysis.
Once your measures are established, determine the appropriate sample size and your methods of analysis.
Now is the time to assess feasibility. Specifically, you must ask: Can enough patients or study subjects be recruited for the study? Consultation with a statistician is necessary at this stage in the process.
Back to Top