1 - Formulating a Research Question Flashcards
What is pre-clinical evidence?
- research that is now systematic (ie experiments are designed, data is collected/analysed/conclusions are made)
- at this phase research is not done on humans (animals)
- depending on animal used, differs in applicability to humans
- can be goodor bad (thalidomide)
The research question is based on what 3 things? (describe)
- population: who?
- intervention/exposure: therapies, diagnostic tests, patient characteristics, life choices
- outcome: surrogate or lab-based measure, survival, rare events, cost
Describe the typical research question for a diagnostic study
- intervention/exposure: the diagnostic test
- outcome: if trying to predict accuracy of new diagnostic test, presence or absence of disease you are trying to diagnose = outcome
What research questions does the hierarchy of evidence answer?
All of them! (therapies, harm, prognosis, diagnosis)
What is unsystematic observation?
- anecdotal experiental type of evidence
- example: a clinician who thinks they know things bc of years of practice
- this is the weakest type of evidence!
- should be treated as a hypothesis, but in recent decades has been where research has started/stopped
Describe the typical research question for a therapy study
- intervention/exposure: active, placebo, cham
- outcome: if trying to decrease death, death. If trying to increace QOL,QOL
For therapy/harm, what 2 things muat we consider on the evidence hierarchy?
- pragmatic (effectiveness) and explanitory (efficacy)
- pragmatic more important than explanitory
- higher pragmatic = higher applicability = increased level of evidence = increased external validity
Name 3 main things that determine the accuracy of reviewed research
1) Study must be designed properly in a way that prevents other explanations from being possible.
2) Interventions, analysis and interpretation must have been carried out with integrity.
3) Editor of journal/peer reviewrs must also have a sufficient toolkit with which to assess whether the study was designed/executed with the highest standards (often missing!)
Name the 5 types od research questions
- Therapy
- Harm
- Differential Diagnosis X
- Diagnosis
- Prognosis
what is observational evidence?
- now working with patients who have the disease of interest
- stronger if less likely to have other explainations for the findings
- allwoing someone to select the treatment they receive
- if a clinician selects treatment for patient, this is not randomized! - the treatment likely worked for them bc the clinician chose it (tailored to them)
What is harm?
- evaluates rare bad events following different treatments
- compares btw other treatments
- ascertaning the effects potentially harmful agents (includinghterapies from the first type of question) on patient-important ourcomes
Describe the typical research question for a prognostic study
- intervention/exposure: patient characteristics (risk or prognostic factors), life choices
- outcome: if you are trying to predict the future, future disease you are trying to predict = outcome
what is the structure of the test used for therapeutic issues?
what is the structure of the test used for diagnostic issues?
What is therapy?
- generally evaluates outcomes following differnt treatments
- whether one treatment can cause reduction in a bad outcome more than other different treatments
- determining the effect of interventions on patient-important outcomes (symptoms, function, morbidity, mortality, costs)
what is the structure of the test used for prognostic issues?
what is the structure of the test used for harm issues?
Name the order on the hierarchy of evidence
