top of page
8626a064701a059ba3b19a63dbb24433 (1).jpg

Glossary of Research Terms - PART 1


  • Absolute vs Relative Data

  • Accolades vs Evidence

  • All-Cause Mortality

  • Anecdotal Testimonials

  • Arm

  • Burden of Proof

  • Cause and Effect

  • Cohort Studies

  • Confirmation Bias

  • Correlations

  • Double Blind

  • End Point of Interest

  • Informed Consent

  • Intervention / Treatment

  • Hypothesis

  • Participant Flow

  • Preponderance

  • Reporting Group

  • Scientific Methodology

  • Surrogate Biomarkers

Absolute vs Relative Data - The best way to describe this difference is an example found in study reporting:


  • A clinical trial outcome is reported and states that “Acme-Pill decreased the risk of heart disease by 54%.”

  • The pill’s result was a decrease of 3.9% and a placebo’s result was a decrease of 1.8%.

  • The difference between these two percentages is 2.1%.

  • The RELATIVE term of 54% was calculated by dividing the difference (2.1%) by the pill’s result (3.9%) which equals 53.8%.

  • The ABSOLUTE term is calculated by subtracting the Acme-Pill’s decrease of 3.9% from the placebo’s decrease of 2.1%, which would result in an absolute decrease of 1.8%.

  • When the public reads about a 54% benefit vs a 1.8% benefit, potentially harmful choices might be made.

  • Absolute data is accurate while relative data, which is often used in marketing campaigns, is less accurate and more exaggerated.


Accolades vs Evidence - although this is technically not a research term, it's important to point out the significant difference between someone with higher-education accolades versus someone who might (or might not) have a more conventional background, but they speak and educate based on highly scrutinized research data outcomes. An example of "empty" accolades would be someone who got a PhD at Harvard or the University of Tokyo but has a conflict of interest via getting funded by special interest groups or corporations (this is much more common than people realize). People with accolades without sound research can potentially cause issues when their impressive backgrounds are trusted yet their so-called research is unfounded or faulty. Consider that evidence-based data is What It So. For example, an apple peel has beneficial effects on our immune system, and cheese is addictive and increases the risk of prostate and breast cancer. These facts are indisputable, no matter who is talking or how many letters appear after their last names. When in doubt, follow the evidence all the way down the rabbit holes where truths and/or deceptions live.  

All-Cause Mortality - a measure of all deaths due to any cause, which occur during a clinical study.

Anecdotal Testimonials - for example, if enough people report that eating Indian gooseberries (amla) lowered their LDL cholesterol levels, these stories might get the attention of researchers who may try to find funding for further study. This is the positive aspect of testimonials. On the flip side, testimonials are based on human memory and behaviors, which are not likely to be sound enough to consider evidence-based information worthy of further investigation. Plus, marketing campaigns include patient stories in ads and actively look for people who agree with the results of their brands. This practice can sway the public without evidence-based research to back up claims.

Arm - a group (or sub-group) of clinical trial participants who receive a specific intervention or no intervention (see intervention below). This approach adheres to the trial’s protocol/methodology.

Burden of Proof - in clinical research terms, this falls on the person or group who brings an existing claim into question. The party who doesn’t carry the burden of proof is presumed to be correct, that is until the burden changes after evidence has been presented by those who have questioned the claim.

Cause and Effect - this shows a study outcome which has a direct connection from one thing to another. An example would be a tobacco study, in which thousands of participants were followed for a long period of time, and those who quit smoking developed lung cancer at a significantly lower rate than those who did not. The study approach, clinical setting, and other factors such as diet, fitness levels, age, gender, and lifestyle would also be considered.

Cohort Studies - this is a type of research design which follows groups of people over time (also called longitudinal studies). Data is used to understand human health and the social and environmental factors that would influence this. Typically, participants share traits such as geographical location, health history, and/or age.


Confirmation Bias - this is the tendency to interpret new evidence-based information within the lens of confirming an existing belief or theory versus within the scope of the actual evidence itself. This term is used in both psychological as well as clinical research settings.

Correlations - not all study correlations are created equal and some may have merit. However, in most cases, correlations are likely not strong enough to be true. In nutrition research, a correlation shows a study outcome which may or may not be related to an actual health benefit. Some examples can be absurd, such as correlating people born in July who have an increased risk of type-2 diabetes (this is a real study). Other examples may show some veracity, such as correlating vegetarianism with lower risks of heart disease, which might inspire researchers to engage in further research. Often, these studies are not based on exhaustive, meta-analysis, long-term, or clinically controlled treatments. Outcome correlations could also be based on subjective, anecdotal stories.

Double-Blind - this is a type of clinical trial when the participants, lab assistants/scientists, trial managers, and others involved are not aware of which particular treatments or interventions were implemented until after the clinical trial is complete. This ensures that study results are not biased by human observations and/or behaviors. Blind studies are when only the participants are unaware of the treatment implemented. The opposite are open label studies in which everyone involved has been informed of the drug or treatment being administered.

End Point of Interest - this refers to an outcome measured in an objective manner, which can determine if an intervention has actually been beneficial. An example: I want to prevent heart disease, so I’m prescribed a tablet to lower my LDL cholesterol levels. When I get a heart attack, my end point of interest is not connected to the prescription, which only offers improved blood lipid panels. My diet, lifestyle, stress, and sleep have not changed; I’ve only taken a pill. Lab results are perhaps beneficial to know, however, medications to “fix” any abnormalities don’t necessarily address a desire to prevent a medical incident or disease (my end point of interest). Drugs or surgeries might only “cure” a symptom versus actually offering a beneficial health outcome by significantly lowering the risk of heart attacks, strokes, or heart disease via diet and lifestyle treatments.

Informed Consent - this is a document which study participants should be given before agreeing to a clinical research study. This concept also applies to non-study environments such as visits to a doctor's office and any information given to patients. First, these five points MUST be present:


  • A participant must have the capacity and ability to make a healthcare decision.

  • A medical provider/researcher must disclose information on the treatment, test, or procedure in question.

  • A medical provider/researcher must disclose expected benefits and risks, and the likelihood or probability that the benefits and risks will occur.

  • Participants must comprehend the relevant information.

  • Participants must voluntarily grant consent, without coercion or duress, by signing a legal document.


There are five main requirements within the scope and details of treatments / interventions:


  1. Defining the nature of the process.

  2. Defining the risks and benefits of the process.

  3. Offering reasonable alternatives.

  4. Defining the risks and benefits of reasonable alternatives.

  5. Assessing a participant’s understanding of the process and requirements.


Intervention / Treatment - examples are giving a participant a new drug, vaccine, or dietary supplement, doing a diagnostic or therapeutic procedure, offering a different type of counselling, and/or introducing an educational tool.

Hypothesis - this is a tentative statement about the connection between two or more variables. It’s a specific, clear, and measurable prediction about what to expect in a study. A hypothesis should always include: an explanation of what’s going to happen, a clear and understandable outline, and a testable variable (both dependent and independent variables). Examples: Researcher X thinks that...


  • apple peels lower the risk of influenza.

  • the toxins in fish and seafood contribute to higher rates of mental development issues in children.

  • seven to eight hours of healthy sleep, unplugging for one hour each day, or dancing for an hour each day... can significantly decrease the risk of dementia.


Participant Flow - a summary of participant’s progress through the stages in a clinical study. This is categorized by study arm, group, or cohort. It includes the number of people who started, completed, and dropped out of the study.

Preponderance of Evidence - is a standard used in a kind of “burden of proof” analysis. Proof is reached when there is more than a 50% chance of a claim to be true and is based on more than one study. In some medical communities, meta-analyses results are considered more valuable since they gather a number of related studies, collect the various data into one database, and analyze this meta-data to offer evidence-based, cause and effect conclusions.

Reporting Group - this is a grouping of participants in a clinical study which is used for summarizing the data gathered. This group might be the same as or different from a study arm or group.


Scientific Methodology - skeptical critical thinking is the single most integral aspect of scientific methodology. Since the 17th Century, this is a method of procedure that consists of systematic observation, measurement, and experimentation, as well as the development, testing, and modification (if needed) of hypotheses. According to the Oxford Dictionary: "criticism is the backbone of the scientific method. Study Design includes the methods, strategies, and settings used in an investigative clinical study." Examples include:


  • Meta-Analysis: a statistical process that combines the findings from other individual studies, which are related to one core hypothesis.

  • Systematic Review: a summary of the past and current clinical literature available for study.

  • Randomized Controlled Trial: a controlled clinical trial that assigns participants to two or more groups by random chance such as alphabetically by last name or by postal code.

  • Cohort Study: a clinical research study where people who currently have a specific condition or receive a particular treatment are followed over time. They are compared with another group who are not affected by the condition.

  • Case-Control Study: these begin with the outcomes and do not follow people over time. Researchers choose people with a particular result (the cases) and interview the groups or check their records to ascertain what different experiences, environments etc they had.

  • Crossover Study: a type of clinical trial when participants receive each treatment in a random order. They're used when researchers sense it might be challenging to recruit participants, who are willing to take a risk by not doing a new treatment that may be beneficial. 

  • Cross-Sectional Study: this is an observation of a defined population at a single point in time or interval of time. Exposure and outcomes are determined at the same time.

  • Case Report and Case Series: these report on a series of patients with an outcome of interest. No control group is involved.

  • Ideas, Editorials, Opinions: these are offered by pioneers and experts in the field, with an adherence to evidence-based research.

  • Animal Research Study: these are conducted with animals as study subjects. The most common are mice, rats, fish, rabbits, guinea pigs, hamsters, birds, cats, dogs, farm animals including pigs, and non-human primates.

  • Test-Tube Lab Research: these are test-tube experiments within a controlled laboratory setting.

  • Blind, Double-Blind, and Open Label Studies: see Double-Blind above.

  • Observational Study: this is a study when people are observed or certain outcomes are measured. There are no treatments given, and there isn't an attempt to affect the outcome. Regarding the field of nutrition studies, these can show some reliable results. This is important because it would be considered unethical if humans were purposefully given something or asked to be in a situation where their health is at risk. Hence, observational studies are vital in understanding food science, and their credibility has proven to be founded in research. 


Surrogate Biomarkers - this is defined as “an indirect indicator of a disease state or of its response to therapy or other interventions.” Surrogate biomarkers can support some healthcare decisions, however, they are not an absolute indicator of actual health outcomes. For example, a patient can have normal lab and/or stress test results and still die of a heart attack. Surrogate biomarkers were an unfortunate factor in the case of Tim Russert, the Meet the Press moderator who had excellent medical insurance, so-called “best” physicians, normal lab and stress test results, and no outward symptoms. Nonetheless, he suffered from heart disease and died of “sudden cardiac arrest” (heart attack). Two additional examples would be:


  • Vitamins: taking a calcium supplement to strengthen bone health is a surrogate biomarker. Since the “endpoint of interest” is increasing bone strength, the blood lipid panel results may show a positive effect of taking a calcium tablet, but this may not connect to a relationship between the tablet and actually stronger bones. Side note: calcium supplements have been shown to increase the risk of having a heart attack or stroke.

  • Medications: taking a medication to reduce cholesterol such as a simvastatin is a surrogate biomarker. Since the “endpoint of interest” is reducing the risk of heart disease and pre-mature death, the medication may indeed reduce LDL cholesterol levels, but does not necessarily show that this medication prevents heart disease and early death.

bottom of page