Week 2- Research Design Overview Flashcards
Levels of evidence- systematic reviews
Secondary study
Level 1 bias
Description: combines all lower level primary studies to get a conclusion
Strengths: summary of evidence. Good validity. Greatly reduced bias.
Weaknesses: time consuming. Only as strong as the weakest study. Means it can be prone to biases. Relies on data being available.
Levels of evidence- RCT (randomised control trial)
Primary study: experimental
Level 2 bias
‘Randomised’ refers to how we allocate people to each group. Two types, parallel trials and crossover trials.
Strengths: strongest primary design. Used to determine the effectiveness of an intervention. Good reliability and validity.
Weaknesses: time consuming. Expensive. Only looks at one or two variables. Lots of resources required.
Levels of evidence- Quasi (pseudo trial)
Primary: experimental study
Level 3.1 bias
Description: almost identical to RCT but isn’t randomised. Eg. testing genders or races against each other.
Strengths: easy to set up. Realistic (can take place outside a lab). Useful when it’s unethical to manipulate the IV.
Weaknesses: subject to allocation bias. Could be difficult to determine causation.
Levels of evidence- cohort
Primary: observational study
Level 3.2 bias
Description: two types, prospective and retrospective. Prospective looks at a group of people exposed to something and follows them over a period of time.
Strengths: suitable for rare exposures. Can assess exposure on multiple outcomes. Less prone to recall bias.
Weaknesses: not suitable for rare outcomes if the subject doesn’t develop the desired outcome, time is wasted. High dropout rates
Levels of evidence- case control
Primary: observational study
Level 3.3 bias
Description: always retrospective= looking back to see if there’s a relationship with an exposure.
Strengths: suitable for rare outcomes. More cost effective because less resource and time intensive. Useful for looking at multiple exposures.
Weaknesses: very prone to recall bias. Can not determine causation.
Levels of evidence- cross sectional (prevalence study)
Primary: observational study
Level 4 bias
Description: “snapshot”. Descriptive study of a group of people at one point in time. Eg. census
Strengths: cost effective. Quick and easy.
Weaknesses: relies on self report. Only valid at one point of time.
Levels of evidence- correlational
Primary: observational study
Level 4 bias
Description: used to assess the relationship between two variables. Testing a relationship.
Strengths: useful to determine an indicative relationship. Stimulates further study.
Weaknesses: correlation does not equal causation
Levels of evidence- case series
Primary: observational study
Level 4 bias
Description: group of cases presenting similarly in disease or symptoms. No control groups. Hypothesis generating. Eg. HIV first discovered
Strengths: lots of in depth info on a small group. Realistic. Useful for measuring naturally occurring things.
Weaknesses: small sample which limits validity. Can’t generalise. Mixture of self reports and observations
Levels of evidence- case study
Primary: qualitative study
Level 5 bias
Description: in depth study on either a person or disease
Strengths: hypothesis generating. Rich and in depth info on topic. High internal validity
Weaknesses: limits reliability. No generalisation
Levels of evidence- grounded theory
Primary: qualitative study
Level 5 bias
Description: creating a theory from data gathered from a group of people. The theory is “grounded in actual data”
Levels of evidence- phenomenology
Primary: qualitative study
Level 5 bias
Description: aims to understand the lived experience of a person in relation to a concept. Focuses on that persons experience in a certain situation
Levels of evidence- ethnography
Primary: qualitative study
Level 5 bias
Description: researchers observe or interact with a study’s participants in their real life environment. Creates shared understanding.
When to use PICO/ PEO
When identifying the question you want to address, depending on weather you’re looking at qualitative or quantitative research
Purpose of PICO/ PEO
Provides a framework for formulating patient-specific clinical questions.
PICO stands for and study type used for
Quantitative studies
P- patient problem (or population). What are the characteristics of the population or patient
I- intervention. How do we wish to intervene
C- control/ comparison. Alternative intervention
O- outcomes. What are the possible outcomes
Eg. (P) in middle aged male amputees suffering from phantom limb pain, (I) is gabapentin, (C) compared with placebo, (O) effective in decreasing pain.
