Lectures Flashcards
What is peer review
Critical assessment of manuscripts submitted to journals by experts who are not part of the editorial staff
List the degrees of crystallisation
-informal conversation between two people
-discussion at a lab meeting
-public discussion at a conference
-scientific blog entry
-conference presentation
-peer reviewed paper
Types of journal articles
Research article
Clinical trial
Review article
Sources of information
Journal articles
Textbooks
Government documents
Ethics applications
Research funding applications
Purpose of peer review
• Selection of quality articles for publication
• Scientificmethodology
• Relevance of article to journal readership
• Interest to reader
• Presentationofmanuscript
• Improve the manuscript
• Check against cheating…
• Provide advice to editors on whether manuscript meets criteria for their journal
Equation for journal quality and ranking
=total number of Times the article were cited during the two previous years/total number of citadel articles in the journal during those 2 years
What are reviewers looking for
• Title – clear and reflects the content of the paper
• Abstract – concise with enough detail
• Introduction – sets the scene of current literature with hypothesis and aims
• Methods – clear with enough detail to repeat work, good experimental design
• Results - well presented with appropriate analysis
• Discussion and conclusions – in the context of
published work and limitations
What is not scientific evidence
-opinion and hearsay
-wipikedia
-newspapers and magazines
-social media
Why do we need a hierarchy of evidence
Not all research designs are equal!
Results are prone to different risk of error and levels of bias What is Bias?
Unfair inclination or prejudice for or against one result Hierarchies of evidence first popularised in 1979
The quality of a studies design determines the validity and applicability of the data
Hierarchy of research(bottom to top)
-animal and in vitro studies
-editorials and expert opinion
-cross sectional studies
-case control studies
-cohort studies
-randomised control trails
-meta analysis ans systematic reviews
Animal and in vitro studies
-tests initial studies
-paves the way for further research
-allows development of therapies and diagnostic tools
In vitro studies
-lab based studies using tissues and cells from humans or animals
-experimental versus in Vivo conditions
-use of physiological conditions to avoid bias such as experimental solutions or temperature
-some laboratory techniques are more open to bias
What is immunohistochemistry(IHC) staining
-determines presence of a protein
-produces an immunoreactive score
% of cells stained + intensity
How can you remove bias in in vitro studies
- blind studies
-agreement from multiple researchers
-computational image analysis
Qualitative data
-descriptive data
- an example Method is the western blot procedure(bio-rad)which can indicate the presence or absence of a protein
Quantitative data
-numeric data
-less dependent on interpretation
-quantitative above qualitative methodologies in the hierarchy
-an example method is densitometry which can determine the relative amount of protein
Animal studies
-higher in hierarchy than laboratory studies
- but are findings in animal studies always representative of human condition ?
-uses comparative physiology research in order to :
>determine hoe species differ
>helps to identify which animal models are most physiologically relevant in a study
An example of a comparative physiology
-research question:how is calcium handling In Heart muscle cells effected by inflammation
-you cannot obtain liber heart muscle cells from humans
-the method would involve calcium removal from cytoplasm which is facilitated by an ATPase puma and an exchanger
-picking an animal which has similar conditions eg Heart rate
Editorials and expert opinion(2 from the bottom)
-expert opinion from authorities/expert committees
-evidence based on physiology or laboratory research such as:
>committee reports
>textbooks
>narrative reviews
-it is a good convenient summary of background information,provides a foundation to develop understanding
-it is very basic level
-takes several years to be published
Observational studies examples
-case studies
-surveys
-cross sectional studies
-case control studies
-Cohort studies
Case study
Studies an individual
Case series
-Studies a group of individuals
-lacks a control group
Case studies and series
-provide a good insight inti conditions or interventions in specific populations
-too simplistic though
-doesn’t account for other populations.
Surveys
+ Potentially very representative of the population
(with a large cohort)
- Limited detail
- Based on recollection
- Participants may lie
Cross sectional studies
-takes a snapshot of the satiation
-compares different groups at a specific point in time
-groups can be chosen based on exposure/outcome or neither
-it directly compares between Groups however does not consider how this may change over time
Case control studies
-retrospective study
-assessing risk factors in cohort already of interest
-more robust than a case series
-analytical as opposed to descriptive
+ Direct comparisons between groups
+ Can evaluate multiple risk factors simultaneously
+ Useful for investigating latent outcomes of risk factors
- Potential for recall and selection bias
- Cannot determine incidence of outcome - Inefficient at identifying rare risk factors
Cohort studies
-Prospective and Longitudinal study
-Exposed cohort observed for outcomes
-Prospective above retrospective in hierarchy
-it’s simple and straight forward as well as it provides accurate results from large populations
-however it is too simplistic
-doesn’t account for other populations and type of intervention a well as it also largely dependent on size of cohort
Cross over studies
A type of longitudinal cohort study Intervention and control group switch over Widely used in physiology and medical trials
↓ Experimental bias if randomised or placebo-controlled
+ Allows within participant comparison between the intervention and control states. + Limits inter-subject variability
+ Fewer participants required
- Effects can carry over to the next period
- Participants may drop-out before second period - Not suitable for all situations
Correlation
Observational studies can only demonstrate a correlation between different events
Causation
Randomised control trials can demonstrate causation
Randomised control trials
-Prospective Cohort studies
-With randomised allocation of participants
-outcome measured from both intervention and control group
Randomised control trails
-Heavily used in physiology and medical trials
-+ Randomisation eliminates the influence of other variables + Can determine causality
- Ethical restrictions
- Selection criteria for study sample may cause it not to be representative
Factors that affect the quality of randomised control trails
Placebo controlled? Blinded?
Cohort size? Confidence interval size?
What are unfiltered studies
-primary research
-collection of primary data
-original evidence through experimentation or initial observations
Examples of unfiltered studies
Randomised control trials Cohort studies Case-control studies Cross-sectional studies
Filtered information
-secondary research which has passed through an evaluation process
-collates already published data
-it can provide the basis for official clinical practice guideliness
Examples of filtered data
Meta-analysis Systematic reviews Critically-appraised topics/articles
Systematic reviews
A form of literature review
Key terms and strict inclusion and exclusion criteria
A systematic search for all of the literature on a specific topic
Information is then reviewed and summarised
+ A good overview and summary of work in an area of research + Thorough and accurate review of current literature
Author is normally highly regarded in that area Sometimes written by a research group
Meta-analysis
A systematic review that uses quantitative methods to summarize the results
Based on previously published data All data collated and analysed RevMan Software
+ Very large sample sizes
+ A good indication of current thought development in the field
Ideally based on > 3 RCT’s More studies = More reliable
Relevance of hierarchy
This hierarchy of research is now widely accepted
However,
The study designs or techniques lower in the hierarchy are not unnecessary!
There are some situations where some studies may be
more relevant
(regardless of rank)
Journals
Journals – Dissemination of results/findings from research studies (original articles, reviews, case studies etc).
Research – The advancement of knowledge within a specific field/expertise/discipline.
