Quantitative study designs

Introduction

Throughout our lives, we are exposed to many factors that can affect our health and wellbeing. But what kinds of factors influence specific health outcomes? And what do we do when we become ill? Health researchers dedicate their working lives to answering a huge range of different clinical research questions and they do so by carrying out very specific studies. The findings from these studies eventually lead to the development of interventions that can help save lives.

Flowchart adapted from: Grimes, D. A., & Schulz, K. F. (2002). “An overview of clinical research: The lay of the land”. The Lancet, 359(9300), 57-61

So many study designs – what’s the difference?

In clinical research, a study design is a plan for selecting study subjects and for obtaining data. These study designs fall into two different categories:

• Experimental
• Observational

These categories are based on whether or not the investigators assign a particular exposure to a cohort.

Experimental trials

Investigators assign exposures (for example, a trial to test the effectiveness of a new medication) and these are categorised into randomized (studies with a control, or comparison, group) and non-randomized trials (those without a control group).

Observational studies

These studies focus on exposures that are already present in a population and assess the effects of the exposure on that cohort. These studies are further categorised into analytical and descriptive.

Analytical studies

• Include a control (or comparison group)
• In Cohort studies people are tracked forward in time from exposure to outcome.
• Case-control studies, by contrast, trace back from outcome to exposure.
• Cross-sectional studies are like a snapshot in time, measuring both exposure and outcome at a particular time point.

Descriptive studies

• Include case reports, case-series and case studies. Cross-sectional studies can also be descriptive.
• Do not have a control or comparison group
• Cannot examine associations between an exposure and an outcome.

Which study type will answer my clinical question?

Not all study types will be appropriate for answering a particular clinical question. For example, if you wanted to investigate the impact of maternal smoking on foetal development, then a randomized-controlled trial would not be appropriate as it is not ethical to assign a disease or potentially harmful exposure to an individual. In this case, an observational study would be more appropriate.
This table illustrates the most appropriate study designs for answering specific types of clinical questions.
 

In this learning series, we will examine each study design type in more detail, the types of clinical questions they investigate and the methodologies applied in each study design.