You will need to carefully plan the analysis and write up of your study findings. It may follow these phases.
Analysis should usually be conducted by, or in conjunction with, a qualified statistician. It is good practice for a Statistical Analysis Plan (SAP) to be prepared in advance. It should detail the trial objectives and outcomes and state exactly how each outcome will be analysed and reported. The SAP can be built into the Protocol or written as a separate document.
The SAP should say:
- How data will be summarised
- What descriptive statistics will be presented
- How each research question or hypothesis will be tested
- Whether and how statistical significance will be assessed
The SAP should usually be agreed by the Chief Investigator. If your study involves other researchers, a study management group or oversight committees then they should also review and agree the SAP
Following analysis of the data, results should be written up for publication. You must always check with your funding body and any bodies that reviewed the study about the requirements for results publication. Some may only require publication of findings in a peer-reviewed journal. Others may require a formal report. When preparing research findings for publication, please consider the following:
a) Methodology
Explain your approach. You need to specify your hypothesis and describe the methodology. Provide details of the research design, sample size and sampling procedure. Justify any changes from the study protocol.
b) Results
The results section for clinical trials is likely to include:
- Information on participant flow (e.g. in a CONSORT flow diagram).
- Recruitment
- Baseline data
- Numbers analysed
- Treatment received
- Outcomes and estimation
- Ancillary analyses (sensitivity and subgroup analyses, etc)
- Adverse events.
c) Interpretation
Research findings should be interpreted in terms of the statistical and clinical significance of the treatment estimates. Include:
- Whether findings are generalisable to the broader population
- Whether findings support or contradict previous research
- The strengths and limitations of the research
With special thanks and appreciation to Caroline Fairhurst – Senior Statistician, York Trials Unit, for the content of this section
