Now that we have discussed confirmation bias (part one of this series) and finding what you need (part two), we will dig deeper in this final article into sources and types of dental research available. It is important to assess if the research you have found is valid before applying it as a standard in your practice.

pen and dental research paper, part 3 overlay

4 Key Factors to Consider in Dental Research:

  1. Source of the Information
  2. Types of Dental Research and Size of the Study
  3. Authors and Reviewers
  4. Study Design and Findings

1. Source of the Information

To critically analyze the source of the information you are considering is an important step in the research process. For example, journals produced and edited by respected academies and organizations usually are more critical of the information submitted. They are also subjected to the peer review process. Peer review is a way to validate the study's trustworthiness by sending it to other researchers or topic experts to evaluate the article. While many journals, including open access and online-only journals, claim to be peer reviewed, it is important to examine who is doing the reviewing. Many journals require authors to pay them to have their research published, and the reviewers and review process are not clearly disclosed. If you can't validate this process, you should not put much faith in the findings.

Grey literature is another source of clinical information. This is information produced outside of traditional publishing and distribution channels, and includes reports, policy literature, working papers, newsletters, government documents, speeches, and white papers. Grey literature can be of value but may not be indexed on the major academic search sites. Education websites (.edu), government sites (.gov), and academies and other nonprofit organizations (.org), are all great alternatives if you can confirm that the site is valid. You can generally find the authoring organization at the bottom of the website's home page to confirm this source's validity.

2. Types of Dental Research and Size of the Study

Once you have your sources, using the evidence pyramid is a great way to determine the value of your findings. The pyramid outlines where evidence generally ranks according to quality, credibility, and generalizability of a study. There are different forms of the research pyramid, but all basically follow the same hierarchy. As you move from the top to the bottom of the pyramid, the amount of evidence will increase, and the quality of the evidence will decrease. The top first three levels are considered “filtered” studies, which means the authors are evaluating the evidence for you. The levels below and moving towards the base of the pyramid are “unfiltered” meaning you need to determine the study's credibility on your own.

colorful blue, gray, and yellow evidence pyramid showing the best (systematic review) to least (editorials & expert opinions) compelling types of evidence

The Evidence Pyramid (Best to Least Compelling Evidence):

1. Systematic Review:

When conducting a systematic review, the authors will ask and then study a specific clinical question, perform a comprehensive review of the literature, eliminate poorly designed studies, and make recommendations based on those reports. They include only experimental or quantitative studies, and often include only randomized controlled trials. The Cochrane Database of Systematic Reviews is considered the “Gold Standard” of systematic review research.

2. Critically appraised Topics:

Authors of these studies evaluate and synthesize multiple research studies. They are essentially a short systematic review focused on a specific topic.

3. Critically appraised Individual Studies:

The authors of these studies evaluate and synthesize individual research studies, evaluating the quality, credibility, and generalizability of articles for you so you do not have to make those determinations on your own.

4. Randomized Controlled Trials (RCTs):

RCT's are studies which include a randomized group of patients divided into an experimental group and a control group. Those in the experimental group are followed for the intervention being studied with the variables and outcomes compared with those in the control group not receiving the intervention.

5. Cohort Study:

In this type of study, two groups (cohorts) of patients, one which did receive the treatment/exposure of interest, and one which did not, are followed to determine the outcome of the treatment being studied.

6. Case-Control Study:

This type of study identifies patients who have the outcome of interest (cases) and control patients without the same outcome, to see if they have the same treatment/exposure being studied.

7. Background Information / Expert Opinion:

Handbooks, encyclopedias, and textbooks often provide a good foundation or introduction of a topic of interest and include generalized information. Background information presents a convenient summary of the topic of interest. It often takes three or more years before the information is published, so it may not be as up to date as other sources.

8. Animal Research / Lab Studies:

This information is positioned at the bottom of the pyramid as this is where ideas and laboratory research take place. Ideas for therapies or diagnostic tools are tested with lab models and animals. Success with models or in animals does not always transfer to humans, so this type of information should be considered a starting point for further research.

The sample size of a study is also an important aspect of credibility. The more patients, materials, or techniques that are being studied, the more likely the conclusions are not merely due to chance and are statistically significant. Authors with very small sample sizes will sometimes state that their results are statistically significant when in fact they are not when compared to a larger population study. The larger the study population, the more likely the results will transfer to the general population with the same intervention.

3. Authors and Reviewers

The authors and the reviewers of a study also reflect on its validity. Consider the authors and those reviewing the study and ask yourself:

  • Are they respected in their field?
  • Do they take a very one-sided view of a particular topic?
  • Have other experts spoken out against the author's previous studies?
  • Do the reviewers have the background and knowledge to evaluate the study design and findings?

Depending on the answer to these questions, you may want to consider the information presented by the study through a filtered lens for bias.

Authors often want to highlight the best conclusion possible to improve the chance of publication, and so skim over the limitations their study may have. You can determine whether the author has identified the flaws or limitations within their work by evaluating the following:

  • Failure to evaluate past similar published studies or a lack of previous research on the topic.
  • Issues associated with their population or samples selected.
  • Insufficient sample size which can affect the statistical analysis.
  • The methods, instruments or techniques used to collect and measure the data.
  • Limited access to critical information or data which potentially affects the study outcome.
  • Time constraints which may have impacted the results.

While limitations do not necessarily mean the study is not valid, it is important you know what they are so you can better assess the results provided within the context of the study limits. This information is often found at the very end of the study in the conclusions section. You often must consciously seek it out.

4. Study Design and Findings

If you are having doubt about the findings, it is important to remember that correlation does not equal causation. For example, just because there is an association between eating an apple each day and developing caries, does not mean that eating an apple each day will lead to fewer cavities. The researcher should be exploring other explanations for their findings than just those studied. These other explanations are known as confounding variables. For example, the researchers should also consider whether people who eat carrots and brush regularly after meals also develop fewer cavities and do additional research to see if there is a proven cause and effect or if additional study is necessary.

Determine whether there are any major flaws or bias in the study by asking yourself:

  • Was the study conducted by a specific clinician who is proficient in the technique being studied?
  • Was the study conducted by the company producing the product under review?
  • Were the patients for the study “cherry picked” to ensure the desired result was obtained?

While bias may be obvious in some studies, it may be difficult to determine if you do not have a research background. When looking at the results of the study, consider whether the conclusions reached are in line with similar studies by asking yourself:

  • Is there a reason to doubt the conclusions reached?
  • Have other “experts” on the topic spoken out or criticized the study findings?
  • Are the “experts” well respected and knowledgeable about the study topic?

If there is a preponderance of opposing evidence on the topic, you may want to question a study's validity.

Critically Analyze Types of Dental Research

Making changes to your clinical practice is an important decision and must be evaluated thoroughly. Identifying personal and professional biases when looking for research articles, evaluating the source and types of dental research you gather, and the critically analyzing your search process are all key factors in this process.

Robert Winter, D.D.S., is a member of Spear Resident Faculty.