Study Design Flashcards
What is a RCT?
intervention study where subjects are randomly allocated to treatment options
What happens during RCT?
- patients given particular drug/intervention/programme of care they don’t usually receive
- then typically randomised to new vs current/placebo treatment
Why are RCTs accepted as ‘gold standard’ of individual
research studies?
They provide sound evidence about treatment efficacy which is only bettered when several RCTs are pooled in a meta-analysis.
Why is choice of comparison group important?
affects how we interpret evidence from a trial
What is sig of comparison of active agent with inert substance/placebo?
likely to give a more favourable result than comparison with another active agent.
When can RCTs be unethical?
Comparison of an active agent against placebo when an existing active agent is available
What happens when an intervention in RCT is programme of care?
common practice for the comparison group to receive the usual care.
What are the benefits of randomisation in RCTs?
- ensures subjects’ characteristics do not affect which treatment they receive.
- allocation to treatment unbiased - treatment groups balanced by subject characteristics in long run and
differences between groups in trial outcome can be attributed as being caused by treatments alone - provides fair test of efficacy for treatments - not confounded by patient characteristics
- makes blindness possible
What is the usual way to do random allocation in RCTS?
using a computer programme based on random numbers.
What is blinding in RCTs?
treatment allocation is concealed from either the subject or assessor or both
Why is blinding done in RCTs?
avoid conscious or unconscious bias in reported outcomes
What is a double-blind trial?
neither the subject nor the assessor knows which treatment is being given
What is a single-blind trial?
if the treatment allocation is concealed from either the subject or the assessor but not both
Why is blindness sig for subjects and accessors?
- subject who knows they’re receiving a new treatment for pain which they expect to be beneficial may perceive/actually feel less pain than he would if he thought he was receiving the old treatment.
- assessor who knows that a subject is receiving new treatment which he expects to work better than the old one, may tend to round up measurements
- If treatment allocation concealed, both patient + assessor will make unbiased assessments of effects of treatments being tested.
What is a placebo?
- inert treatment which is indistinguishable from the active treatment
- In drug trials, often possible to use placebo drug for control which looks + tastes exactly like active drug
Why is placebo important?
makes it possible for both subject + assessor to be blinded
In which situations is blindness not possible in RCT?
- in trials of technologies e.g. trials comparing different types of ventilators - impossible to blind the
clinician - in trials of surgery versus chemotherapy.
When is statistical analysis of RCTs straightforward?
where there are complete data
What is intention to treat analysis?
- direct comparison of treatment groups with subjects being included in the group to which they were originally allocated
- e.g. patient in group A analysed in group A even if they don’t comply/switch treatment
What is sig of ITT analysis?
- only way in which there can be certainty about the balance of treatment groups with respect to characteristics of subjects.
- therefore provides unbiased comparison of the treatments.
What should happen if patients change treatments during RCT?
- should still be analyzed together with patients in their original, randomly allocated group since change of treatment may be related to the treatment itself.
- If patient’s data are analyzed as if they were in their new treatment group, balance in patient characteristics present after random allocation will be lost.
What is per-protocol analysis?
- patients are analysed according to the treatment they have actually received
- may be useful in addition to ITT analysis if some patients have stopped or changed treatment.
What are observational studies?
- subjects receive no additional intervention beyond what would normally constitute usual care.
- Subjects observed in their natural state (in the real world)
What is a case-control study?
investigates causes of disease, or factors associated with a condition
Who are ‘cases’ in case-control studies?
patients with disease/cond of interest selected for inclusion,
Who are ‘controls’ in case-control studies?
- A comparison group without the disease is then selected
- cases and controls are compared to identify possible risk/causal factors
What is different about case-control studies?
usually retrospective in that the data relating to risk factors are collected after the disease has been identified.
What are the limitations of case-control studies?
- choice of control group affects comparisons between cases + controls e.g. may choose patients with another cond/healthy patients
- Exposure to risk factor data usually collected retrospectively either reported by patients/from records and may be incomplete, inaccurate/biased (people with cond have more acute recollection of past events compared to those without)
What are benefits of case-control studies?
- quick
- expensive
What is a cohort study?
observational study that aims to investigate causes of disease/factors related to condition but is longitudinal and starts with unselected group of individuals who are followed up for set period of time
When are cohort studies sometimes used?
confirm the findings of case-control studies
How are cohort studies set up?
- starts with unselected group of ‘healthy’ individuals
- subjects followed up to monitor the disease/condition of interest + potential risk factors
- length of follow up chosen to allow sufficient subjects to get the disease + risk factors to be explored
What is outcome in the simplest case of cohort study?
where there is a single risk factor that is either present or absent, the incidence of disease can be related directly to the presence of the risk factor
What is different about cohort studies compared to case-control studies?
- longitudinal - for set time period
- Usually prospective, with the risk factor data being recorded before the disease is confirmed
Can cohort studies be retrospective?
requires that full risk factor data is obtained on all individuals with and without the disease of interest using data which was recorded prospectively
What are the limitations of cohort studies?
- large number subjects needed to obtain enough individuals who get the disease/condition, particularly if it is uncommon
- length of follow up may be substantial to get enough diseased individuals (needs long follow up) and so cohort study not feasible for rare diseases
- difficulty in maintaining contact with subjects, particularly if follow-up is lengthy
- resources required may be v. high - expensive
What is benefit of cohort studies?
Less biased answer than case-control as data collected prospectively
What are cross-sectional studies?
sample is chosen and data on each individual is collected at one point in time (may not be exactly the same time point for each subject)
When are cross-sectional studies used?
- Surveys of prevalence of disease
- Surveys of attitudes/views, e.g. studies of patient satisfaction, patient/professional knowledge; studies of behaviour e.g. alcohol use, sexual behaviour etc
- When inter-relationships between variables are of interest, e.g. study to determine characteristics of heavy drinkers where a cross-sectional study allows comparisons by sex, age etc