How can we best evaluate, treat, and assess long-term risks for children with kidney disease? Who is at risk of developing end-stage renal disease (ESRD) in childhood or young adulthood? Clinicians are often faced with questions such as these with uncertain answers in the practice of pediatric nephrology. Parents ask, "Why did my child get this disease?" "What is the most effective method to treat this condition?" "What's the prognosis of this condition in my child?" Frequently, these answers are not known, and these questions are the inspiration for high-quality clinical research. The first step in developing a valuable clinical study is determining whether the initial query can be translated into a good research question.
Hallmarks of a Good Research QuestionA good research question gives useful information, is interesting to the researcher, builds on what is known, and can be answered with available resources. Research is a labor of love, demanding attention to detail, perseverance, honesty, and imagination. Developing a good research question is an iterative process. One needs input from knowledgeable colleagues and collaborators. The researcher must become thoroughly familiar with what is already known about the topic by reviewing the literature and consulting with experts in the area. Investigating what is already known has several benefits. First, it can reveal that the candidate research question has already been answered adequately. Second, learning what is already known provides insight into potentially useful methods for addressing a research question. For example, previous studies may demonstrate good ways to measure a variable of interest or provide background information for determining sample size. Third, a literature review may suggest ways to frame the research question at hand. For example, a literature review may reveal that particular risk factors are consistently associated with a disease process, and an intervention to modify these risk factors may form a sound basis for a clinical trial.Finally, a good research question needs to be answerable with available resources. These include subjects available for study, technical expertise of the research staff, and the time and money that can be devoted to the project. Once a question is framed, the researcher needs to outline the study protocol or methods, which include specifying the recruitment method, number of subjects and how they will be recruited, how each variable will be defined, and the plan for data analysis. A poorly designed study is worse than no study at all because, like imprecise measurements and an improper analytic plan, it can also lead to false conclusions.