Study design types

Author: Dr. Nida Nadeem

Vascular Laboratory, Kings College Hospital London

Importance of choosing a good study design

Evidence based research is all about applying the most suitable study design to fit the particular research question. Study design is one of the most crucial aspects of planning a clinical study. You can have the most amazing research question which everyone wants to know the answer to, but without a proper study design, the research might not be considered robust or the conclusions of the study may not reflect the truth. You do not want to weaken your study by not thinking about the study design properly. Many different quantitative and qualitative research designs exist, each with their own strengths and limitations. This research article explores some study design types, when they might be utilized and the strengths/weakness of each.  

Qualitative vs Quantitative research 

Figure 1: Examples of qualitative and quantitative research studies

Randomized Controlled Trial (RCT)

RCT is the most robust study design in healthcare to answer questions that evaluate the effectiveness of interventions. The basic structure of a RCT is displayed in figure 2. A RCT is a true experiment in which participants are randomly allocated to receive a new intervention / conventional intervention (experimental groups) or no intervention at all (control group).

The groups are treated and observed identically apart from the actual intervention, therefore any differences in outcome are attributed to the trial intervention.

The process of randomization eliminates selection bias. It should in theory balance out baseline confounding factors, both known and unknown.

RCTs that are blind are considered better to avoid bias in reporting results.  There are two types:

Single blind: When the participants do not know whether they are in the control or treatment group, but the researchers do.

Double blind: Neither participants nor researchers know who is in the control or treatment group, this can also include data collectors and statisticians.

Figure 2: Randomized controlled trial study design. Source: Emergency Medicine Journal 2003; 20: 164-168.

The main strengths of RCTs are

• The random allocation of participants which helps to ensure that the groups are similar in all respects except for exposure to the treatment.
• The longitudinal nature of the study which means that exposure to the treatment precedes the development of the outcome. These study design features ensure that differences in the outcome can be attributed to the intervention.

Disadvantages of RCTs include:

• Costly to conduct and run.
• Follow up period can be lengthy.
• Sometimes strict inclusion and exclusion criteria mean that participants are not necessarily representative of the patients seen in daily clinical practice, this can affect the generalisation of the final results.
• RCTs cannot study harmful exposure events because it would be unethical to expose participants to harm

Cohort Studies

When the evaluation of an intervention using a RCT is unethical or unfeasible, we can think of using an alternative study design such as a cohort study (figure 3). This study design is similar to the RCT in that there are comparison groups and they are followed up to determine who experiences the outcome of interest. The important difference between the two designs is the absence of random allocation to study groups. Cohort studies can be used to compare two different forms of treatment over the long term.

Cohort defines a population group followed over a period of time. Investigators compare the experience of one group exposed to a factor (exposed group) with that of the other which was not exposed to the factor (control group). If the exposed group has a higher or lower frequency of an outcome than the unexposed, then an association between exposure and outcome is evident. It is important that the study population is free of the disease of interest at entry point into study.

Cohort study is generally an observational study, but it can be either prospective or retrospective. Prospective design allows exposure to risk factors to be assessed directly and confounding variables to be considered. A retrospective design is effective for diseases with a long development time. 

Figure 3: Cohort study design. Source: Plast Reconstr Surg. 2010 Dec; 126(6): 2234–2242

Advantages:

• Can study rare exposures and multiple effects of one exposure.
• Allows for direct measurement of incident of disease.
• Can calculate risk such as the extent to which those exposed to the risk factor are likely to get the disease and absolute risk which is the incident rate for the group exposed to the risk factor.

Disadvantages:

• Validity of results can be seriously affected by losses to follow up.
• Can be very time consuming.

Case Controlled Studies

When the outcome of interest is rare or takes a long time to develop, neither RCTs nor cohort studies may be feasible. In these circumstances it may be better to think about using a case controlled study design. In a case control design, patients with the outcome of interest (cases) and patients without the outcome of interest (controls) are identified and then the investigator determines whether they have had previous exposure to the causative agent (figure 4). The investigator is able to match the case and control patients on variables that may influence the outcome (e.g. age, sex, and other health conditions). This ensures the groups are as similar as possible and the specific effect of the causative agent on the outcome can be more confidently determined. This is usually a retrospective study design type.

Advantages:

• Cost effective and shorter duration, this is because cases are identified at the start of study, so there is no lengthy follow up waiting for disease to develop.
• Particularly useful for studying rare diseases because researchers start with a population of target disease rather than waiting for disease to develop (as with cohort studies).

Disadvantage:

• Due to retrospective nature it can be subject to more bias, example recall bias.
• Limited to examining one outcome at a time.

Figure 4: Case-control study design. Source: Plast Reconstr Surg. 2010 Dec; 126(6): 2234–2242

Systematic reviews (with/without meta-analysis)

Systematic reviews are a robust review of the evidence from relevant primary research studies.

The aim is to:

1. Identify all available research relevant to the review question.

2. Use explicit inclusion and exclusion criteria for studies to be accepted into the systematic review.

3. Apply established standards to critically appraise the quality of each study included in the review.

4. Extract and analyze the data from each included study. This may include a meta-analyses (a statistical analyses that combines the results of the multiple studies to identify common results and trends).

5. The results of the review should address the overall strength of evidence in support of the original question (i.e. weak, moderate, and strong).

Advantages:

• This is an exhaustive review of the published and unpublished literature.
• Results can be generalized more than individual studies.
• More reliable than individual studies.
• This is considered an evidence based resource.

Disadvantages:

• This is a very time consuming process
• Studies can be difficult to include because they must all have the same variables/outcomes, however study design can vary.

Qualitative research methods

Although the focus of this article has been to discuss quantitative study designs, we do not wish to underplay the importance of qualitative research. Qualitative research such as surveys, patient focus groups, interviews, telephone discussions can all help us to understand the patterns of health behaviors, capture illness experiences and give understanding to the beliefs and motivations of various groups of people. Qualitative research provides powerful insight into a problem and can inform hypotheses for quantitative research. Qualitative methods are a more holistic and person-centered way of uncovering thoughts and actions of human beings and definitely have a place in healthcare related research.

References

Understanding randomised controlled trials | Archives of Disease in Childhood (bmj.com)

Identifying the best research design to fit the question. Part 1: quantitative designs | Evidence-Based Nursing (bmj.com)

Designing a research project: randomised controlled trials and their principles | Emergency Medicine Journal (bmj.com)

Observational Studies: Cohort and Case-Control Studies - PMC (nih.gov)

Presenting and Evaluating Qualitative Research | American Journal of Pharmaceutical Education (ajpe.org)