Does a study use the right design to answer the study question?

Would you use a screwdriver to brush your teeth? You could try, but it probably wouldn’t turn out as you hoped (and could, in fact, be disastrous) since screwdrivers are not meant for brushing your teeth. Likewise, using the wrong study design to answer a research question could give you poor results. As a clinician that practices evidence-based medicine, you’ll want to make sure that the study’s authors used the best tools to answer the research question.

Understanding the type of research question in a study will help you evaluate whether the chosen study design in the paper was the most appropriate. Let’s look at the main types of research questions and then discuss the most commonly used study designs in research.

There are two main types of research questions:

1. Qualitative
2. Quantitative

Qualitative research involves focus groups and participants who provide answers in questionnaire, video, and audio formats. Sometimes, the beliefs and thoughts of a specific group of people can provide great insight for clinicians. Their answers can point to areas where more research is needed for a specific topic.

For example, let’s say that a pediatric nutritionist wanted to enhance her patient’s healthy eating habits. It would be useful to know how parents defined healthy eating, and some of the challenges they face when making healthy meals. This type of information would be best represented in a long-form format where parents can share their feelings.

Surveys with only multiple-choice answers may not fully capture the problem. For example, a study illustrating how many fruit and veggies children ate in a week would not explain why the parents were choosing the meals for their kids.

Now, what if you needed to know the average amount of sugar that a child consumed in a day? Would an in-depth qualitative study detailing the parent’s beliefs on sugar consumption give you the answer you’re looking for? Not really! On the other hand, quantitative studies rely on numerical data and statistics to draw conclusions.

To quantify a child’s daily sugar intake, a survey would give the respondents the option to choose the most accurate answer (e.g., one spoonful, two spoonfuls, etc). Then, the researchers could look at the data and run statistics on the average sugar consumption of children. Depending on the questionnaire’s details, the researchers could also find an association between a child’s background, age, sex, and daily sugar intake. Quantitative data can be used to find associations that can be useful for making conclusions about the general population.

In a clinical setting, you will probably be searching for quantitative data more often than qualitative data. So, how do you know what type of quantitative study would best answer the questions you have? Well, there are different study designs in medicine, and each design is suitable for a specific type of research question.

Let’s take a quick look at some of the most common epidemiological study designs!

Randomized controlled trials

Randomized controlled trials (RCTs) are considered the gold standard in scientific research and are often used to analyze new treatments and medications.

In this type of study, participants are usually placed in two groups: an experimental group that receives the new treatment and a control group that receives a placebo or the standard treatment option. Participants are chosen randomly, and usually, the participants and / or the physicians don’t know who is in each group. This is called blinding.

What if you had a patient with a specific type of lung cancer and you wanted to know whether a new experimental drug would increase their chances of survival? An RCT would best answer this type of research question. From the results, you’d be able to assess whether the new drug increased survival rates compared to the standard of care.

Cohort study designs

Cohort studies are used to measure the association between exposure and disease status over a long period of time.

Participants in cohort studies are chosen based on exposure status. They are divided into two groups: exposed and unexposed participants. Each group is then followed for an extended period of time to see which group develops the disease being studied.

Cohort studies are useful for determining the association between lifestyle habits (e.g., diet, smoking, alcohol consumption) and chronic diseases (e.g., cancer, cardiovascular disease, diabetes).

For example, if you wanted to know a patient’s risk of developing throat cancer from vaping, you may want to look at a cohort study. A cohort study would follow people who vape for a length of time to see whether they develop throat cancer.

Case-control study designs

Case-control studies also measure the association between exposure and disease status.

Unlike cohort studies, participants are chosen based on whether they have a specific disease or condition. The participants are placed into diseased (e.g., case) and non-diseased (e.g., control) groups. Then, each group is analyzed to see if they had or hadn’t been exposed to a specific substance.

Case-control studies are also useful for studying rare diseases.

For example, if one of your patients had a rare form of cancer and you wanted to find the latest research, then you may want to search for case-control studies. A case-control study is more appropriate since the participants are chosen based on whether or not they have the disease. In contrast, cohort studies on a rare disease would be more difficult to come across since it would require finding participants and seeing whether or not they develop this rare disease after some time.

In short, when choosing a scientific study, make sure that the study design is appropriate for the research question asked in the paper.

This is just a quick summary of the most common study types you may come across in your searches. Be sure to check out our Epidemiology Essentials Course to learn more about different types of studies and how to interpret their findings.