Part One: How to Find, Evaluate, and Understand Scientific Research
The nutrition space can be extremely noisy. At Needed, we are committed to highlighting nutrition fundamentals over fads, and we want to empower you to do the same.
Today, we are sharing the first in a three-part series on how to identify, read, and evaluate scientific research articles. Part 1: how to identify and evaluate appropriate research on topics of interest. To do so, we’ve partnered with experts on nutrition research from the Needed community, Dr. Ari Calhoun, and Certified Clinical Nutritionist, Julie Jurenka.
What is a Clinical Research Study and How Can I Find Reliable Research?
A clinical research study is a methodical investigation designed to develop or advance the knowledge on a specific topic, through a clearly defined question that is focused and concise. Research studies then develop a hypothesis, which is a proposed explanation or answer to the question at hand based on limited evidence. The focus of the study is then to prove or disprove the hypothesis through a clearly defined systematic framework for data collection and analysis that ensures reproducible and valid results. Within the results section, data are reported and examined through statistical analysis to determine if the results were significant and if the hypothesis was supported or not. The study’s discussion explains the author’s interpretation of the findings in relation to prior evidence from other studies, highlights areas of bias, and poses future questions. The study conclusion explains how well the research findings answer the study’s questions, if it supports the hypothesis, and how the findings add to our knowledge about the particular topic.
How Do I Know if a Scientific Paper Represents Clinically Sound Research?
When you are searching for research papers on health-related topics, the goal is to find reliable, peer reviewed research on the topic of interest. Since science is fluid and ever-changing, it is important to remember that any scientific paper can have inaccuracies, whether intended or not. Regardless of the journal of publication, the prestige of the researchers/authors, or the public support by the results, any scientific paper may be found to contain inaccuracies at some point. All hypotheses, study materials, data, results, claims, and even conclusions will be open to re-examination as future evidence on the subject becomes available. This is the very nature of science – we are always learning more, and nothing is static.
What is Peer Review?
Peer review is a process by which the article is reviewed by other experts in the field, from which the authors gain suggested edits and constructive critiques.
Does Peer Review of Research Publications Ensure Good Science?
Peer review is an essential part of the scientific process, but it doesn't necessarily guarantee that published papers are 100 percent accurate.
Peer review is a hallmark of a credible scientific article and is undertaken to ensure:
- its methods and results are logical and reproducible.
- it accurately represents a worthy addition to the existing literature.
- conclusions drawn represent the study results.
Goals of the peer review process are to:
- suggest ways to improve the paper so it is more easily read and understood.
- point out questionable results or conclusions.
- possibly recommend additional research.
In order to determine the reliability of a peer-review, examine the peer-review board page of a journal publication to ensure it has multiple members with varied credentials from various scientific disciplines. If there are only a few individuals on the journal’s peer-review board, the article may have been subject to less scrutiny than with a journal that has many reviewers. Generally, peer reviewers take this role seriously and do a great job, but it is important to remember that they cannot possibly check every aspect of the research, may not have time to delve too deeply into the data, and generally cannot uncover well-concealed acts of scientific misconduct. Therefore, peer review doesn’t universally guarantee the reader that a scientific paper is accurate, but it is a vital component of a credible scientific article and the journal of publication.
Features of a legitimate research paper:
While peer review may not guarantee a paper is accurate, you can still determine if a paper is legitimate by reviewing whether it:
- was written in good faith without significant conflicts of interest.
o If the research is paid for by a company marketing the product, it may be biased.
o If the research team members are also working for a company manufacturing the product, they may be incentivized to skew the study results.
- was carried out using appropriate methodologies and materials.
- provides a list of references/sources for statements made in the paper.
- was taken seriously by the relevant scientific community and cited by other authors.
Types of Scientific Papers and the Information They Provide
There are three main types of scientific papers and they each provide a different type of information. They all have value, but it is important to understand the nature of the investigation, as this will allow you to appropriately determine a valid conclusion.
- Primary research means the research was conducted by the author themselves. This includes several sub-types of research designs, including, experimental, correlational, and descriptive – which are further explained below. It is generally published in peer-reviewed journals, usually features a treatment or intervention, and provides the best look at detailed science on a specific topic. It should include a hypothesis, measurable endpoints, materials and methods used, detailed results, and conclusions. This is the gold standard, as it reduces room for bias and mis-interpretation of results, but other types of papers can present valuable information as well. Information on the type of scientific paper can generally be found in the study title, abstract, or introduction.
- Experimental research describes a process whereby the researcher conducts an intervention by manipulating a single variable and observes the respected outcome. The researcher is careful to control for confounding variables, in order to ensure that the variable being studied is the only explanation for the observed outcome (i.e. cause and effect). Conclusions drawn from this type of study are considered to be the benchmark, while conclusions from all other primary research study designs are considered less reliable. While experimental research is the ideal way to collect reliable information, not all information can or should be studied in a single variable format. Furthermore, using findings from a controlled clinical trial and applying those results to an uncontrolled environment doesn’t always translate. In reality, there are many factors that determine efficacy or a desired outcome, and purely relying solely on experimental research is a reductionistic approach, leaving behind other useful sources of information.
- Correlational designs include case-control studies and observational studies. In contrast to experimental research, correlational designs look at naturally occurring relationships between two variables. For instance, a study may look at the association between smoking and birth outcomes. While these studies do attempt to control for other variables in the process of statistical analysis, inevitably, there are variables which are not controlled for. Thus, the outcomes in these studies are not as trustworthy as those in found in experimental designs. Unfortunately, these types of studies are often the most popularized by the media.
- Descriptive designs include case studies, which provide a detailed explanation of a particular outcome for an individual or group of people in a real-life situation. This type of study design does not control for variables in the way other studies do. While these studies do shed light on potential therapeutics, the strength of the findings are much weaker due to the limited number of study participants and lack of control for outside variables.
- Meta-analyses look at the bulk of available research on a topic over a given time frame that meet specific criteria, provide statistical analysis, and identify overall trends based on a body of research (covers multiple primary research studies). They help summarize the breadth of the literature and sift out poor quality primary research, yet can be a source of bias through poor collection or misinterpretation of the gathered data.
- Systematic Review articles can provide a useful synthesis of information to answer a well-defined question on a topic through a systematic means of gathering and analyzing information across a variety of studies. These are the most likely to be influenced by the author’s opinion.
Once you have identified a legitimate, peer-reviewed, scientifically-sound publication, the next step is to learn how to recognize key features of the paper that will allow you to efficiently navigate and understand the author’s work. This information will be presented in Part 2 of our blog series.
trusted education is needed.