Design and conduct research properly

Download a pdf of Design and conduct research properly


Stimulated by surveys revealing the poor quality of many reports of clinical trials, reporting standards have been developed and applied.  Such standards make clear how many patients have been asked to participate in a study and how many declined the invitation. Results are presented according to the various treatment groups selected at the outset.

But there is still a long way to go to improve: (a) the choice of questions being addressed in research; (b) the way that these questions are formulated to ensure that the outcomes of treatments chosen for assessment are those that patients regard as important; and (c) the information made available to patients.

To see whether a proposed trial might be feasible and acceptable, exploratory work involving groups of patients can be useful. This may highlight shortcomings in the design plans; or help to define outcomes that are more relevant; or even suggest that the concept is a non-starter. [5, 6]

This can save a lot of time, money, and frustration. The clinical trial in men with localized prostate cancer that we described elsewhere showed how the research design was improved by careful consideration of the terms used by clinicians to describe the trial’s purpose and the treatment options. Exploration of patients’ views led to an acceptable study because the concerns and information needs of the men being invited to participate had been identified, and the information provided to potential participants took account of these findings. [7]