Glossary of terms and jargon buster

Scientific article types explained! (click on each to go directly)

For more information why not check out this helpful overview and see what they say is ‘best evidence’.

  1. Clinical Practice Guidance
  2. Rapid evidence Guidance
  3. Systematic review (with meta-analyses)
  4. Systematic review (without meta-analyses)
  5. Randomised Controlled Trial (standard parallel)
  6. Randomised Controlled Trial (cluster)
  7. Randomised Controlled Trial (cross-over)
  8. Feasibility or exploratory Randomised Controlled Trial
  9. Controlled Clinical Trial
  10. Economic evaluation
  11. Systematic review of non-randomised or observational studies (with meta-analyses)
  12. Systematic review of non-randomised or observational studies (without meta-analyses)
  13. Prospective cohort study
  14. Retrospective cohort study
  15. Other longitudinal study (including survival analysis)
  16. Case-control study
  17. Cross-sectional or prevalence study
  18. Clinical prediction tool
  19. Pharmaco-epidemiological safety study
  20. Mediation analyses from a trial or observational study
  21. Moderation analyses from a trial or observational study
  22. Ecological or population level study
  23. Measure development study
  24. Diagnostic checklist study
  25. Qualitative synthesis (or systematic review of qualitative studies)
  26. Semi-structured interview study
  27. Focus group
  28. Ethnography
  29. Text analysis
  30. N-of-1 study
  31. Ecological momentary assessment study
  32. Case series study
  33. Case report/study
  34. Pre-experimental study (qualitative outcomes, e.g., chemistry flame tests)
  35. Pre-experimental study (quantitative outcomes/units, e.g., temperature)
  36. Quasi-experimental study (qualitative outcomes, e.g., chemistry flame tests)
  37. Quasi-experimental study (quantitative outcomes/units, e.g., temperature)
  38. True experimental study (qualitative outcomes, e.g., chemistry flame tests)
  39. True experimental study (quantitative outcomes/units, e.g., temperature)
  40. Mixed methods study
  41. Expert or expert service user perspectives
  42. Narrative review
  43. Theoretical/framework or position piece
  44. Scoping review
  45. Delphi study
  46. Editorial
  47. Commentary
  48. Other expert opinion
  49. Published study protocol
  50. Method (‘how to’) article

Glossary of Scientific Article Types Made Easy

Click on study names below to download lay versions of Quality Assessment checklists used by scientists, where available. Authors are encouraged to upload these versions with their lay summary to be open about just how well the original scientific article did in terms of its overall quality to help their reader! 

  1. Clinical Practice Guidance

‘Clinical practice guidance’ (e.g., written by the National Institute of Clinical Excellence (NICE), or other professional organiosations) is a review of evidence either from a larger number of clinical trials of a given treatment(s) or for a given problem or condition, but it can also be guidance for a particular aspect of clinical work (e.g., clinical supervision) based on a range of evidence (so not just trials). The guidance is supposed to support healthcare (and other) professionals work out the best treatment of a given problem/condition or inform best practice in a broader sense. When putting clinical practice guidance together, those involved in completing the review carefully check the quality of the evidence that they decide to include, which is usually (but not only limited to) several high quality randomised controlled trials. 

For more detailed information why not check out: National Institute of Clinical Excellence (NICE) Guidance

2. Rapid evidence Guidance

‘Rapid evidence assessment’ or ‘rapid review’, is a review of published and unpublished evidence either from a larger number of trials of a given treatment (or several treatments) or for a given problem or condition. Compared with systematic reviews, they are very similar but provide a faster way of providing information for decision making quickly (usually within less than five weeks!) However, the way they are done can vary. Rapid reviews are supposed to help healthcare (and other) professionals work out the best treatment of a given problem/condition. When putting rapid reviews together, those involved in completing the review check the quality of the trials that they decide to include, which is usually (but not only limited to) several (ideally!) high quality randomised controlled trials.

 For more detailed information why not check out: National Institute of Clinical Excellence (NICE) Rapid Guidance example with Covid-19

  1. Systematic review (with meta-analyses)
  2. Systematic review (without meta-analyses)

A ‘Systematic review’ of randomised controlled trials or controlled clinical trials involves researchers searching several online scientific databases (and other places for unpublished studies) to find all relevant clinical trials to answer a focused question(s). Usually the question(s) will look at a given trial of a treatment(s) for a particular condition or group of people: e.g., A systematic review of all exercise treatments for people living with Multiple Sclerosis who experience symptoms of tiredness/ fatigue. The search for the trials is done and reported in such a way that it can easily be repeated by other researchers should they wish to at a later stage. Once the authors have worked out what trials should be included (as not all trials found will be relevant), they give a quality rating score for each trial in (ideally!) a fair way. Depending on how many trials are included in the review and what they compare the treatment of interest to (e.g., exercise programmes versus general practitioner care only), authors will sometimes try to take all the results of the given measure(s) of interest (e.g., changes in a person’s tiredness/fatigue) and put them into a large calculation called a “meta-analysis”. This helps us to work out on average how much the measure of interest (e.g., fatigue) improves (or not!) across all the different trials. The key idea here is that including more trials (and with it more people) is much better than just looking at only one small trial. Sometimes doing a meta-analysis is not possible though at the time of writing a review because there isn’t enough of the same kind of clinical trials and measures of interest in the scientific literature. So, this means that some systematic reviews are written in more of an essay or summary format called a “narrative synthesis”.

For more detailed information why not check out: The Cochrane Library

  1. Randomised Controlled Trial (standard parallel)
  2. Randomised Controlled Trial (cluster)
  3. Randomised Controlled Trial (cross-over)

A ‘definitive randomised control trial’ (or ‘standard parallel RCT’) is one of the final stages in testing out a treatment for a given group of people or health condition. It seeks to show if the treatment really does cause positive change on given measures of interest over time. All of the people taking part in the trial are randomly placed in usually one of two treatment groups, where ideally using something like a computer random number sequence generator works best (‘randomisation’). If two groups, this means some people will receive the treatment being tested and others won’t. Randomly placing people into the groups means that there should be an even number of people with roughly the same demographic, illness or other characteristics in each group. If you don’t do this, placing people in groups means ‘bias’ can happen, where for example there may be more depressed people in the tested treatment group because the researchers believe they will benefit more or because the person has asked to be in this group.

Usually, measures of interest are taken before randomisation and the treatment goes ahead, at the end of treatment, and then hopefully over a longer period of follow-up. The measures of the groups are then compared. Ideally, the trial authors, people ‘crunching the numbers’ or statisticians and, if possible, people taking part in the trial should not know what group they are in and so ideally are ‘blinded’ to knowing what treatment they get. This is because simply the act of knowing which group people fall into can hugely influence how they do in a trial.

It’s also important to note that not all definitive RCTs are the same. Some may be more focused on looking at cause and effect and will carefully pick people with say, one type of condition or symptoms (called an ‘efficacy RCT’), whilst others are based more in real-life clinics or services where people have a number of different conditions (called an ‘effectiveness or naturalistic RCT’). To make matters slightly more confusing, some will offer a particular treatment to one group and not much else to the other, but then offer the group that received very little the treatment later on (called a ‘cross-over RCT’), whilst others will be more focused on randomly putting whole clinics or participating health centres into the treatment or no treatment groups (called a ‘cluster RCT’).

 For more detailed information why not check out: ClinicalTrials.gov

  1. Feasibility or exploratory Randomised Controlled Trial

An ‘Exploratory or Feasibility randomised control trial’ (or ‘feasibility RCT’) is one or perhaps a few steps before doing a ‘definitive randomised control trial. It is usually more interested in seeing whether a treatment for a given group of people or health condition is doable (i.e., ‘feasible’) and if people like it and find it acceptable or not. It also gives researchers and health professionals a ROUGH IDEA of whether the treatment might work and has a positive change on given measure/s of interest over time but is less interested in saying if the treatment ‘causes’ changes. Usually, there’s not enough people taking part in feasibility trials to get any clear answers about cause. All of the people taking part in the trial will be randomly placed in usually one of two treatment groups, but it could be more, ideally using a computer random number sequence generator. If two, this means some people will receive the treatment being tested and others won’t. Randomly placing people into the groups means that there should be an even number of people with roughly the same demographic, illness or other characteristics in each group. If you don’t do this, placing people in groups means ‘bias’ can happen, where for example there may be more depressed people in the tested treatment group because the researchers believe they will benefit more or because the person has asked to be in this group.

Usually, measures of interest are taken before randomisation and the treatment goes ahead, at the end of treatment, and then hopefully over a longer period of follow-up. The measures of the groups are then compared. Ideally, the trial authors, people ‘crunching the numbers’ or statisticians and, if possible, people taking part in the trial should not know what group they are in and so ideally are ‘blinded’ to knowing what treatment they get. This is because simply the act of knowing which group people fall into can hugely influence how they do in a trial.

It’s also important to note that not all feasibility RCTs are the same. Some may be more focused on carefully picking people with say, one type of condition or symptoms, with the next step being an ‘definitive efficacy’ randomised controlled trial to look at cause and effect, whilst others are based more in real-life clinics or services where people have a number of different health conditions where the next step might be an ‘effectiveness or naturalistic’ randomised controlled trial. To make matters slightly more confusing, some will offer a particular treatment to one group and not much else to the other, but then offer the group that received very little the treatment later on (called a ‘cross-over’ randomised controlled trial), whilst others will be more focused on randomly putting whole clinics or participating health centres into the treatment or no treatment groups (called a ‘cluster’ randomised controlled trial).

For more detailed information why not check out this article

  1. Controlled Clinical Trials

A ‘controlled clinical trial’ or clinical quasi-experimental study usually tests out a treatment for a given group of people or health condition. It seeks to show if the treatment MAY result in a positive change on given measures of interest over time. All of the people taking part in the trial will be placed in usually one of two treatment groups, but it could be more. If two, this usually means some people will receive the treatment being tested and others won’t. The controlled clinical trial can look pretty similar to a randomised controlled trial, but there is a very important difference, namely that people are not randomly placed into two or more groups. By not randomly placing people into the different groups this means that there may NOT be an even number of people with roughly the same demographic, illness or other characteristics in each group. This means ‘bias’ can happen, where for example there may be more depressed people in the tested treatment group because the researchers believe they will benefit more or because the person has asked to be in this group, for instance. Also, other lurking or what are called ‘confounding’ characteristics may affect the results. In controlled clinical trials we CANNOT strictly say the treatment ‘causes’ this change. Usually, they are done before an RCT or when an doing one simply is not possible, e.g., in real life clinics, where randomly offering the person the tested treatment or not is unethical and cause harm.

Usually, measures of interest are taken before the treatment goes ahead, at the end of treatment, and then hopefully over a longer period of follow-up. The measures of the groups are then compared. Unlike RCTs, the trial authors, people ‘crunching the numbers’ called statisticians and, if possible, people taking part in the trial know what group people are in and so are NOT ‘blinded’. Simply the act of knowing which group people are in, can hugely influence how they do in a trial. However, blinding in these trials is not always done or achievable.

It’s also important to note that not all controlled clinical trials are the same.  Some may be more focused on carefully picking people with say one type of condition with the next step being an ‘definitive efficacy’ randomised controlled trial to look at cause and effect, whilst others are based more in real-life clinics or services where people have a number of different conditions and the next step might be an ‘effectiveness or naturalistic’ randomised controlled trial.

  1. Economic evaluation

An ‘economic evaluation’, which is usually a randomised controlled trial that tests out how helpful a treatment for a given group of people or health conditions is, tries to work out the cost and (hopefully beneficial!) health outcomes of a treatment compared with other existing treatments.

For more detailed information why not check out: The International Society for Pharmacoeconomics and Outcomes Research (ISPOR)

  1. Systematic review of non-randomised or observational studies (with meta-analyses)
  2. Systematic review of non-randomised or observational studies (without meta-analyses)

A ‘Systematic review’ of non-randomised or observational studies (i.e., not controlled clinical trials or randomised controlled trials) involves the author searching several online scientific databases (and other places for unpublished studies) to find all relevant studies to try to answer a focused question(s). For instance, the question(s) will look at a given set of a things related for a particular problem, condition or group of people: e.g., A systematic review exploring if past abuse is a risk factor (i.e., possible cause) for depression. The search for the studies is done and reported in such a way that it can easily be repeated by other researchers should they wish to. Once the authors have worked out what studies should be included (as not all studies found will be relevant), they give a quality rating score for each study in (ideally!) a fair way. Depending on how many studies are included in the review and what they look at, authors will sometimes try to take out all the results of the given measure(s) of interest (e.g., rates of abuse and depression) and put them into a large calculation called a ‘meta-analysis’. This helps us to work out on average how much the measure of interest (e.g., rates of abuse) is related to depression rates (or not!) across all the different studies. The key idea here is that including more studies (and with it more people) is much better than just looking at only one small study. Sometimes doing a meta-analysis is not possible though at the time of writing a review because there isn’t enough of the same kind of measures of interest in the scientific literature. So, this means that some systematic reviews are written in more of an essay or summary format called a “narrative synthesis”. Observational studies can include cohort studies, case-control studies, cross-sectional studies, clinical prediction tools, among others.

For more detailed information why not check out: The MOOSE (Meta-analysis Of Observational Studies in Epidemiology) guidelines

  1. Prospective cohort study
  2. Retrospective cohort study
  3. Other longitudinal studies (including survival analysis)

A ‘cohort study’ seeks to follow the journey of people over time to answer important research questions. For instance, if you wanted to see how much smoking cigarettes causes lung cancer, a ‘prospective cohort’ study will try to follow-up a large group of people or ‘cohort’ from a certain age or time to see how often they smoke over a given period of time (the ‘exposure’) and see how they fare in terms of getting lung cancer or not (the ‘outcome’). As you can imagine, in this large group there will be people that decide to smoke, while others decide not to. The task of the researcher here is to compare people who have and haven’t smoked to see if rates of lung cancer are higher in the smoking group or not. A ‘retrospective cohort” is often cheaper to organise and takes much less time if the data is already available because, for example, it starts with getting records of people who have died from lung cancer and then looks back to their old GP records to see if they had smoked to explore the same question about whether smoking leads to lung cancer-related deaths. Strictly speaking, we CANNOT say that one thing caused another in cohort studies. This is because there may be other lurking or ‘confounding’ factors at play. For instance, we might have initially thought that drinking coffee might lead to lung cancer but did not take into account the possibility that most people drink coffee also smoke cigarettes, where smoking cigarettes in the ‘confound’.

Other longitudinal studies might follow people up for a given amount of time (e.g., six months) but be more focused on say, for example, changes in symptoms of their health condition over time or perhaps in relation to changes in their mood. Some include a calculation that tries to work out a person’s exposure to something (e.g., radioactivity) and the time until which an event happens (e.g., onset of condition) called a ‘survival analysis’.

For more detailed information why not check out: The STROBE Statement

  1. Case-control studies

A ‘case-control study’ is where a group of people are invited to take part based on particular characteristic(s) or ‘exposure’ to something (e.g., child abuse or a given health condition) and are directly compared on given measures of interest to those without the condition or exposure.

For more detailed information why not check out: The STROBE Statement

  1. Cross-sectional or prevalence study

A ‘cross-sectional study’ or ‘prevalence study’ is simply a snapshot of a group or ‘sample’ of people (e.g., brain scans) or their experience (e.g., mood), or other types of non-human data measured at one point in time. This allows researchers to look at rates of things (e.g., unemployment rates in the UK) and/or relationships between different measures of interest to answer particular questions such as, “Does being unemployed relate to low mood?” However, because it is a snapshot in time, we CANNOT say that one thing (unemployment) causes another (low mood) or vice-versa.

For more detailed information why not check out: The STROBE Statement

  1. Clinical prediction tool

A ‘clinical prediction rule’ study is mathematical tool(s) that is intended to guide health or other care professionals in their everyday clinical decision making. They can be tested to see if they can predict a person’s diagnosis or their expected journey with the condition over time (i.e., prognosis), or when working out if a person will respond well to a particular treatment or not.

For more detailed information why not check out: The TRIPOD (Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis)

  1. Pharmaco-epidemiological safety study

A ‘Pharmaco-epidemiological safety study’ tries to work out how effective medications are in large numbers of people with e.g., a particular health condition. It tries to work out the benefits of taking the drug in the population but also the harms or what we call ‘adverse events’.

 For more detailed information why not check out: The STROBE Statement

  1. Mediation analyses from trials or observational studies
  2. Moderation analyses from trials or observational studies

A ‘mediation analysis’ study is a calculation that a researcher does (usually on a single randomised controlled trial, controlled clinical trial, or cohort study or can be a systematic review including several of these studies) to work out HOW the treatment worked over a period of time. Quite often researchers do mediation analyses in cross-sectional studies but really these are not proper mediation analyses because they are done only at one point in time. So, for example, the research team might believe that a given psychological therapy for depression works by making people feel more confident in managing their depression symptoms. So, this would mean the person in the trial completes questionnaire measures about the level of confidence in managing the symptoms at the start, during, and at the end of the treatment, and (ideally!) at longer term follow-up. Then the researcher looks to see if positive changes in depression measures happened after seeing improvements in the person’s confidence about managing their symptoms.

  1. Ecological or population level study

An ‘ecological study’ or ‘population level study’ links together available data usually at the national level. For instance, you could look at the number of suicide rates in Europe and how many of the citizens were in poverty in the same year to see if there was a link between the two things. It’s important to remember though that the ecological study does not collect data from individuals, and so makes NO DIRECT LINK between the people that were in poverty and those who took their lives. We might believe these are likely to be the same people in the population, but we CANNOT actually say poverty causes suicide. What we can say is that at the population level at least there is a trend showing a relationship between these two things.

  1. Measure development study

A ‘measure development study’ is when a researcher wants to measure something (e.g., depression) either for the first time, or measure something with greater accuracy because the currently used or “gold standard” measure (e.g., of depression) has some problems with it. Like the Diagnostic checklist study, the researchers need to look at their new measure and see how it works by giving it to people to complete alongside a pervious “gold standard” measure. The researchers have to go through a careful process and calculations to ensure that what they’re measuring is actually what they think they’re measuring, and that it can be measured with accuracy over time and across different people or professionals. Some measures are created to be completed by professionals or by people with a given health condition themselves, and their parents, carers or teachers. As you can imagine, making sure that we’ve got good enough measures is important when you think about how we work out if people improve in clinical trials, for instance. It is also important that people with the health condition, for example, are involved in offering feedback to researchers when they are putting the measure together because if the person needs to fill out themselves it needs to be easy to understand and relevant to them.

  1. Diagnostic checklist study

A ‘diagnostic checklist study’ is simply a way to use numbers to work out if a new test (say for a given health condition) is any better above and beyond tests that are currently available. These tests need to be good at working out if the condition or diagnosis is present or not (see overlap with Measure Development Study).

  1. Qualitative synthesis (or systematic review of qualitative studies)

A ‘qualitative synthesis’, or sometimes called a systematic review of qualitative studies, involves the author searching several online scientific databases (and other places for unpublished studies) to find all relevant qualitative studies to answer a focused question(s). The key idea here is that including more studies (and with it more people) and thinking about their findings is much better than just looking at only one small study. Qualitative studies could include semi-structured interview studies, focus groups, ethnographies and/or text analyses.  For instance, the author might ask questions like “What is it like for people to have chronic fatigue?”, “What is it like working as a paramedic?” or “How do the general public understand COVID-19?” The search for the studies is done in such a way that it can easily be repeated by other researchers should they wish to. Once the authors have worked out what studies should be included (as not all studies found will be relevant), they will often read through all the different study articles (or related data from those studies) and go through a repeated process of assigning codes to different parts of the text, which are then grouped together and summarised as key themes (‘Integrated reviews’). Or the authors will interpret the data and come up with completely new themes or understandings that emerge from the text of the included studies, which can help lead to new ideas or theories (‘Interpretative reviews’).

  1. Semi-structured interview study
  2. Focus group
  3. Ethnography
  4. Text analysis

A ‘qualitative study’ (or the group of studies that make up this category) is the process of collecting and carefully analysing and making sense of written text or spoken information or language (including video, audio recordings and photographs), and unlike other studies are much less interested in numbers. You could say that the information that qualitative studies are interested in is more descriptive in nature and the aim is to offer a more detailed understanding of things that numbers simply can’t get at as easily. Like for example “Exploring the lived experience of working as a paramedic in X hospital in the UK”. Qualitative studies give us a much richer sense of people’s lived experience compared to looking at only numbers. It can tell us a lot more about how things work in lots of detail, and we might then go on to formally test out some of the ideas expressed in other “quantitative” studies using numbers.

 Not all qualitative studies are the same. For instance, the researcher might want to do in-depth individual interviews with people in person, online or over the telephone using some guided open questions (e.g., “Can you tell me about your experience of working as a paramedic?”), these are called ‘Semi-structured Interviews’. Researchers might also want to do something similar but ask people to meet in a ‘focus group’ to discuss their experience together. They might also want to look at people’s diaries or Twitter posts to do a ‘text analysis’. Or as with the case of the paramedics, the researcher might decide to spend a given amount of time observing the paramedics in their natural environment and make notes about the different comments that are made and the things that happen, which is called an ’ethnography’ (so a bit like a fly on the wall).

Once collected, the researcher often changes the data into text format and begins to code the different things that have been talked about, usually on a line-by-line basis, before starting capture core themes or ideas in larger chunks of the text. Themes usually emerge after the researcher has looked at the data several times, including re-reading and coding those texts that were already coded. There is no strict number of people or pieces of information that need to be analysed for qualitative studies, it depends mostly on the nature of the question the researcher is trying to answer, and the type of analysis used. One way of knowing when it might be important to stop collecting more data is if the researcher notices there are no new themes coming out of the data, which is called ‘saturation’.

  1. N-of-1 study
  2. Ecological momentary assessment study
  3. Case series study
  4. Case report/study

‘Single-case’ studies are doing exactly what their name suggests. They are mostly focused on looking at one individual’s journey, be that through a treatment or procedure, or capturing their lived experience in day-to-day life, for instance. Although it focuses only on individual data, there is the possibility of combining data of several individuals to make up a group in a study. Apart from researchers not always having the money available to do a big randomised controlled trial (and needing to show their treatment is worthy of funding to do one!), a very important reason for using single-case designs is because randomised controlled trials tend to use averages to work out if people have benefited or not. The problem with this is that whenever you use averages you can easily lose sight of the individual, and if they benefited a lot, not at all or even if it caused harm. The same goes for other group studies like cohort studies, where it becomes quite difficult to see the unique journey that the person has been on.

Not all single-case studies are the same. Some are interested in testing a given treatment in a way where they can use a computer number generator to randomly (‘randomisation’) give the person treatment or number of treatments at different times, and then randomly give someone else a different order of the same treatments – this is called an ‘N-of-1 study’. This can give us more confidence that if the given treatment caused this change, we might see the same changes in both individuals at the time it is given.

Other single-case studies can look at a number of people and track their individual measures regularly (e.g., once daily or several times a day) over a given period of time, which again can also be before and after receiving a treatment (‘Case series’). Sometimes the author has produced a more detailed descriptive report on solely one individual but goes into a lot of detail and uses some numbers or measures (‘Case report / study’). Some Case report studies have been useful to explore possible cause in a lot of detail when done in a particular way, but like Case series, most do not use randomisation and so we CANNOT say they the tested treatment or procedure caused the change.

Another powerful way to look at individual’s experience (of say, a particular symptom) is to use things like the person’s smartphone as a prompt or remind them to track different things on their phone over a given period of time usually on a daily (or even an hourly) basis and repeated e.g., a few times (or ‘waves’) over a year. The more data points that you have over time, the more detail you get, and the more confident you can be that the person is not relying on their memory alone to give you a sense of how their symptoms have been. The other really useful thing about this study called ‘ecological momentary assessment’ is that you can explore whether there are notable changes in say the severity of people’s symptoms (and other things) when people go about their daily lives across their natural home and/or work environment.

  1. Pre-experimental study (qualitative outcomes, e.g., chemistry flame tests)
  2. Pre-experimental study (quantitative outcomes/units, e.g., temperature)
  3. Quasi-experimental study (qualitative outcomes, e.g., chemistry flame tests)
  4. Quasi-experimental study (quantitative outcomes/units, e.g., temperature)
  5. True experimental study (qualitative outcomes, e.g., chemistry flame tests)
  6. True experimental study (quantitative outcomes/units, e.g., temperature)

‘Field and natural experiments’ are used to work out the causes of things in the lab setting rather than the clinic. They can be experiments with people (‘in human’) or animals, cells, or other substance ‘samples’ (‘non-human’).

 In-human and animal experiments are not always focused on testing out treatments or medical procedures but can also test so called ‘manipulations’ to the lab environment and how this changes a person’s or animal’s behaviour. For instance, researchers might want to understand more social aspects of our human nature, such as “Will people taking part in the study start to actively punish others if they are told repeatedly by an authority figure to do so?” (e.g., does giving a firm and repeated instruction from a university professor to electrocute another person for getting answers to general knowledge test wrong make them conform by harming another person?) – The short answer is “yes”, unfortunately it does. In-human (‘human in vitro wet lab’) and non-human experiments (‘non-human wet lab’) are slightly different in that they look at the changes with a particular focus on cells or other substances. Other experiments can look at computers and micro-chip experiments in labs, called ‘Computational In silico or tech-based dry lab’ studies.

Not all experiments are the same, some use numbers to work out whether things change over time and others use observation using the senses (e.g., flame tests in chemistry). There are broadly three types of experiment types:

‘Pre-experiments’ are where the research is set up to answer the question by making sure the lab environment and the things the researchers do with the person, animal and substances is tightly controlled (e.g., telling them to repeatedly electrocute another person who gets answers to questions on a test wrong; or adding substances). However, these studies are usually just a single group of people, animals or substances and doesn’t make any comparisons with another ‘control’ group who did not receive the manipulation (e.g., firm instruction to electrocute the person, adding a substance of interest).

Other experiments tend to compare two or more groups or samples. One will test the condition of, say for example, the firm and authoritative person being present, while the other may have a researcher who is far less firm or authoritative or another with no researcher present at all. Like, randomised control trials, it’s best if people or samples are randomly put into different groups, this is called a ‘True experiment’ to make sure that the characteristics of those taking part or samples are evenly balanced between the groups, otherwise it may be that the researchers are biased and favour and select people or samples either consciously or unconsciously to go in one of the groups. By comparison, a ‘quasi-experiment’, much like a controlled clinical trial, looks a lot like a true experiment, but people or samples are not randomly assigned to the groups so there are a lot more problems with bias creeping in.

  1. Mixed methods study (you can use more than one of the lay quality assessment checklist listed here)

A ‘mixed methods’ study simply means using a number of different study types or methods to answers questions. For instance, it is not uncommon for randomised clinical trials to also ask people to take part to do qualitative in-depth interviews afterwards to find out more about how they found the treatment.

  1. Expert or expert service user perspectives

‘Expert or expert service user perspective’ are usually a brief opinion by experts, be that academics, clinicians or other professionals or people living with a health condition, that are (hopefully!) based on other evidence or ideas in the scientific literature. But sometimes the views can be quite radical or even deemed controversial to inspire or shake-up the way we currently think about things. Therefore, they can be (though not always) quite reflective and ground-breaking in their way.

  1. Narrative review

A ‘narrative review’ is probably the closest thing to what a student does for an essay that they submit as part of their degree. The idea is that the research and draws together lots of information on a certain topic, be it clinical trial, e.g., The effectiveness of virtual reality treatments in helping people with chronic dizziness, or some other area like, e.g., Do bicycle helmets really prevent head injuries? The important thing to remember here is that although this can be a really interesting read with lots of key information, it is not the same thing as a systematic review. This means that the things the author talks about in the review are mostly included based on their own personal choice and ‘bias’. Also, the way they searched the scientific databases for information to include data is not done in a way where someone else could easily repeat this. As you can imagine, this can result in researchers only including information that they like, or talk about only from a particular angle, rather than including other important data information in a more balanced way. It’s not to say narrative reviews are ‘bad’, but it’s important to remember they have some significant limitations.

  1. Theoretical/framework or position piece

A ‘Theoretical framework / position piece’ article it’s kind of a deep dive into thinking more about understanding the underlying ideas around things, such as Einstein’s theory of relativity. The piece is designed to put forward some ideas to try to move things forwards in terms of understanding and to make some calculated guesses about where to go next. Hopefully, most of these articles will include some evidence to say whether the theory they’re talking about is necessarily helps our understanding of things or not. Much like the narrative review, it is important to remember here is that although this can be a really interesting read with lots of key information, it is not the same thing as a systematic review. This means that the things the author talks about are mostly included based on their own personal choice and ‘bias’. Also, the way they searched the scientific databases for information to include data was not done in a way where someone else could easily repeat this. As you can imagine, this can result in researchers only including information that they like, or talk about only from a particular angle, rather than including other important data information in a more balanced way. It’s not to say the Theoretical framework / position pieces are ‘bad’, but it’s important to remember that they have some significant limitations.

  1. Scoping review

A ‘Scoping review’ it’s not really trying to answer a focussed question you might see in a systematic review or even a narrative review. Instead, it is more interested in exploring the landscape of an area of research to see what is out there. For instance, it might be the case that there are not enough studies or trials looking at how to treat symptoms of severe tiredness/fatigue in Multiple Sclerosis, so it makes sense for the researcher to broaden out their search to look for studies or trials looking at how to treat symptoms of fatigue in other health conditions. The idea is that these findings might have some use in people with Multiple Sclerosis. Much like other expert opinion articles, it is important to remember here is that although this can be a really interesting read with lots of key information, it is not the same thing as a systematic review. This means that the things the author talks about are mostly included based on their own personal choice and ‘bias’. Also, the way they searched the scientific databases for information to include data was not done in a way where someone else could easily repeat this. As you can imagine, this can result in researchers only including information that they like, or talk about only from a particular angle, rather than including other important data and information in a more balanced way. It’s not to say that scoping reviews are ‘bad’ – they are really useful to give us some new ideas – but it’s important to remember that they have some significant limitations.

  1. Delphi studies

A ‘Delphi study’ involves trying to answer a research question by asking the views of several chosen experts (“panellists”) in the area to reach a consensus or agreement on something, e.g., key features of a health condition. Usually, the process has a series of rounds, which allows the experts to reflect and re-think their opinion based on the opinions of others, which are anonymised. For example, a researcher might want to know what clinical psychologists think the most important psychological issues with chronic tinnitus are. As you can imagine, getting a sense of what is being thought about or going ‘on the ground’ in clinic can be useful to inform new research. A Delphi study, alongside other research studies, can help the researcher to decide what the most important, e.g., psychological issues, might be and support the development of guidance for clinical practice.

  1. Editorial

An ‘editorial’ article is a short summary usually the editor of the scientific journal (sometimes an expert in the field) who introduces or makes comment on other articles within the current volume or edition of the journal. Much like other expert opinion articles, it is important to remember here is that although this can be a really interesting read with lots of key information, it is not the same thing as a systematic review. This means that the things the author talks about are mostly included based on their own personal choice and ‘bias’. So, they may talk about the work only from a particular angle, rather than including other important data and information in a more balanced way. It’s not to say that editorials are ‘bad’ – they are really useful to give us some new ideas – but it’s important to remember that they have some significant limitations.

  1. Commentary

A ‘commentary’ put very simply is when a researcher who is fairly well known offers some thoughts on other people’s research or ideas, whether this is a book/s or published scientific articles. Much like other expert opinion articles, it is important to remember here is that although this can be a really interesting read with lots of key information, it is not the same thing as a systematic review. This means that the things the author talks about are mostly included based on their own personal choice and ‘bias’. So, they may talk about the work only from a particular angle, rather than including other important data and information in a more balanced way. It’s not to say that editorials are ‘bad’ – they are really useful to give us some new ideas – but it’s important to remember that they have some significant limitations.

  1. Other expert opinion
  2. Published study protocol

A ‘published study protocol’ (hopefully!) a good job of describing in some detail what a researcher or team plan to do in their future study, usually a clinical trial. The researchers are then expected to stick to this plan when completing the study, so it’s a really good way to see if any issues came up or if ‘bias’ crept in and they decided to do something different, like not clearly report on the main measure of interest. This is important because as researchers sometimes we really want our treatments to work and will do things consciously or unconsciously to make our results look better than they actually are. So maybe refer back to this later once the study has been done!

  1. Method (‘how to’) article

A ‘Method’ article is exactly what it says on the tin. It aims to describe clearly how to use a particular tool or scientific method, can be quite technical in this regard. It also includes articles where treatments have been described in detail to help the reader understand exactly what is done so it can be repeated if necessary. Usually, it’s written for other researchers and other professionals who might need a ‘how to’ guide to move their own research forwards.

  1. Other (please specify):

Any other article we have not captured above.

Other terms worth knowing about

Carryover effect – this is when the influence of a previous condition in an experiment on a person, animal or cell etc. lingers and can overshadow or influence the next condition or experiment.

Cross-level bias – this is when it is falsely assumed that one type of aggregate data collected (e.g., rates of violent crime in the UK) causes another (rates of suicide of victims of violent crimes in UK). This may be true at the individual level but there might be other reasons why victims of violent crimes take their lives.

E-print – this can either be a scientific pre-print published on an online repository, a post-print/MS Word version of a peer-reviewed scientific article accepted for publication on (usually an academic) repository, or a published typeset (usually open-access) article from a scientific publishing journal website.

Intra-subject correlation – intra-subject correlations are a measure of how stable over time a measured characteristic is. For example, you might expect ‘intelligence’ to be stable over time, whilst ’tiredness’ will vary considerably over time and depends on other environmental factors, such as lack of sleep etc.

Period effects – This is the impact of how living during a certain historical period has on data in outcomes in scientific studies.

Regulatory documents:

These are documents that separately or together allow for a clear way of knowing that the way a clinical trial was done and data and results it showed. They show if and how the trial investigator, sponsor, and monitor have stuck to standards of good research in clinical practice and fulfilled regulatory requirements. This will vary according to the country a clinical trial was completed in but can include for example applications, modifications, amendments, supplements, revisions, reports, submissions, authorisations and approvals, non-disclosure agreements.

Structured Data Methods and Results:

These are datasets that the researcher has collected and/or calculations of these datasets used as part of the testing of the trial for a given treatment.

Trial Registry:

Is usually a website archive where researchers can register their clinical trial to show what they intend to do and what they’re planning, which hopefully would include something about what treatment they’re looking to test, what they will compare this to and what measures they will you use to see if it works or is effective.

Jargon Buster Guide for Lay Summary Authors

Captivate your audience with easy-to-follow words: Now you’ve started getting down your initial ideas for your amazing lay summary, you might have realised that after being engulfed by lots of scientific and technical jargon throughout your education, you’re running into some challenges with still using overly complicated words or phrases. Or worst still, you still think the words or phrases your using are ‘perfectly normal/understandable’ for lay audiences, when in reality they’re probably aren’t.

With this in mind, we have pulled together a growing list of a glossary of terms and turns of phrases that you can easily translate into lay language fast below to help out your audience (you’re welcome!)

Needless to say, if you have any other good ones you want to share with us, please get in touch with us on Hello@TheCollaborativeLibrary.com and we will happily add them to this list after reviewing them to help others in similar boat.

Happy writing!

The Collaborative Library Team

 

Turns of phrase translated:

  • Accordingly, consequently – so
  • Discontinue – stop
  • Duration – time
  • For the purpose of – to
  • If this is the case – if so
  • In the event of – if
  • Inform – tell
  • Participate in – take part
  • Prior to – before
  • Scheduled to undergo – due to have
  • With reference to, with regard to – about

Simple explanations of scientific terms:

  • Adverse drug reaction – Side effect
  • Apoptosis – How cells die
  • Axon – Also known as a nerve fibre, an axon is a long, slender projection of a nerve cell that transmits information to other surrounding nerve cells / Long ‘arms’ of cells “that help transmit signals from cell to cell”
  • Autoimmune – When cells of the immune system mistakenly recognize and attack substances naturally present in the body
  • Aggregates/aggregation – Clumps “abnormal proteins stick together, in an ordered way like Lego bricks, creating clumps”
  • Allele – Version of a gene
  • Aβ, Amyloid, A-beta peptide, A-beta 40, A-beta 42, A-beta, beta-amyloid, β-amyloid deposits, aggregates or aggregation – Amyloid “A protein which can build up into toxic clumps in the brain called amyloid plaques”
  • Angina – Chest pain
  • Astrocytes – Brain support cells
  • Atrophy – shrinkage
  • Bioinformatics – “when researchers use computer algorithms to study large amounts of data”
  • Biomarker – Biological signature/fingerprint “which can be an indicator of disease”
  • Blood Brain Barrier (BBB) – A protective layer that prevents most large molecules and cells found in the blood from entering the brain tissue
  • Cell – The basic structural, functional and biological unit of all known living organisms
  • Cell line – A sample of cells that originally came from a patient/mouse/etc that’s been carefully expanded in the lab until they’re all exactly the same.
  • Central nervous system (CNS) – A network of nerve cells that make up the brain and spinal cord. This system is responsible for neurological processes which govern basic activities such as thinking, feeling, learning, seeing, and moving
  • Clinical trial – A test in medical research and drug development that collects data on the safety and efficacy of a particular health intervention
  • Cognition – A group of mental processes that includes attention, memory, producing and understanding language, learning, reasoning, problem solving, and decision making
  • Cytokine – A small molecule that directs the movement and actions of cells in the immune system
  • Cognitive ability, cognitive impairment, cognitive decline, cognition – Memory and thinking skills
  • Cohort – Group of people
  • Cerebral Spinal Fluid (CFS) – Spinal fluid “that bathes our brain and spinal cord”
  • CT, MRI, PET scans – Brain scan
  • Culture – Cells grown in a dish
  • Demyelination – Process during which myelin is stripped from nerve fibres
  • Dendrite, neurite, neuronal process – Arms of nerve cells
  • Differentiation –The process by which a cell undergoes development to exhibit specialized properties and actions
  • DNA – the cell’s instruction manual – organised into genes, which tell the cell what to do and how to look.
  • Drug target – Something in the body that is changed by a drug to give a desirable effect
  • Experimental autoimmune encephalomyelitis (EAE) – An MS-like disease created in laboratory mice
  • Efficacy – The ability to produce a beneficial effect / How X works
  • Expression – How genes encode molecules / how genes make products (e.g., proteins) that can be used by cells
  • End-point – something that is measured in a clinical trial and is the goal of the trial
  • Extracellular – Found outside cells
  • Genotype – Gene
  • Heterogeneous – More diverse
  • Homeostasis – Housekeeping systems “to keep cells working well”
  • Homogeneous – More similar
  • Hyperglycaemia – High blood sugar
  • Hypertension – High blood pressure
  • In the event of – if
  • If this is the case – if so
  • inform – tell
  • Intracellular – Found inside cells
  • Inflammation – A complex biological response that is initiated by the body’s immune system to protect it against harmful agents / Swelling
  • Lesion – A wound to body tissues. In MS, a lesion which occurs in myelin of the central nervous system is called a plaque
  • Leucocytes Blood cells that fight infection
  • Magnetic resonance imaging (MRI) –technological tool that shows images of soft tissue in the body in greater detail
  • Metastasis – Spread of cancer
  • Mutation – A sudden and permanent change in the genetic makeup of a cell
  • Myelin – A fatty protein that covers and protects nerve fibres in central nervous system
  • MCI – Mild cognitive impairment “A term used to describe thinking and memory problems which are not severe enough to be diagnosed as dementia”
  • Microglia – Brain immune cells
  • Model (noun) (mouse/rat/fly etc.) with features of X disease
  • Model (verb) “recreate features of the disease in the lab”
  • Molecular biology – The nuts and bolts/cogs and wheels/ins and outs of how the cell works
  • Morphology – Shape
  • Mitochondria – Power generators “units within cells that generate energy”
  • Mutation (causative gene) – Faulty gene
  • Mutation (risk variant) “variation in a gene that increases/decreases the likelihood of disease”
  • Neurons – nerves in the brain
  • Neurodegeneration – Deterioration of nerve cells
  • Neurotransmitter – Chemical messenger “helps cells to communicate”
  • Novel – New
  • Oligodendrocytes – Cells in the CNS that make and maintain myelin
  • Organelles – Subunit inside cells
  • Pathway – A cascade of chemical reactions that instructs a cell on how to behave / a series of chemical reactions
  • Peripheral nerves – can be thought of as electrical cables: the nerves are the wires that run down the middle and they are wrapped in a fatty layer of insulating material called the myelin sheath’.
  • Probability – how likely X is to happen
  • Proteins – Biological molecules that can perform an array of functions within living organisms / Small building blocks of cells / / The cell’s action molecules. The DNA gives instructions, and the proteins carry them out: ‘grow, move, spread, die etc.’
  • Pathology – Disease process
  • Phenotype – Appearance or characteristics
  • Progressive – Gets worse over time
  • Randomised-controlled trial – a clinical trial where people are put into different groups randomly. One group is given the best current treatment, no treatment at all, or a placebo and their progress is compared to people having the treatment that is being tested. People are usually selected for each group by a computer.
  • Remyelination – A process by which myelin is reformed following injury
  • RNA – another type of genetic material, like DNA, mostly acts as a messenger between the DNA and the protein to carry the instruction from one to the other
  • Scheduled to undergo – due to have
  • Signalling pathway – Ways that/ cells communication / how cells talk with each other / Molecular chain of events
  • SNP – Genetic variation “Natural variation in the code that makes up a gene”
  • Synapse – Junction “point of contact and communication between two nerve cells”
  • Tau, tau fibrils, fibular tangles, tau tangles, phosphorylated tau, hyper-phosphorylated tau. Tau – “A protein which can behave abnormally and form toxic clumps called tau tangles”
  • Target – “biological molecule involved with disease process that a drug could act upon”
  • Therapy/ intervention – Treatment
  • Vascular -Blood vessels

 

Fancy joining The Collaborative Library Working group?

We welcome people from diverse backgrounds.

No science background needed.

You have Successfully Subscribed!