Introduction to Study Design and RCTs Simon Thornley.
-
Upload
alexander-hawkins -
Category
Documents
-
view
236 -
download
1
Transcript of Introduction to Study Design and RCTs Simon Thornley.
![Page 1: Introduction to Study Design and RCTs Simon Thornley.](https://reader036.fdocuments.net/reader036/viewer/2022062320/56649d8c5503460f94a74962/html5/thumbnails/1.jpg)
Introduction to Study Design and RCTs
Simon Thornley
![Page 2: Introduction to Study Design and RCTs Simon Thornley.](https://reader036.fdocuments.net/reader036/viewer/2022062320/56649d8c5503460f94a74962/html5/thumbnails/2.jpg)
That sugar movie… http://
gameauland.com/that-sugar-film-teaser-trailer/
![Page 3: Introduction to Study Design and RCTs Simon Thornley.](https://reader036.fdocuments.net/reader036/viewer/2022062320/56649d8c5503460f94a74962/html5/thumbnails/3.jpg)
Cohort
Participants
UnexposedDisease
Unexposed No disease
ExposedDisease
Exposed no disease
Exposed
Unexposed
Participants Exposure Outcomes
By measurement
![Page 4: Introduction to Study Design and RCTs Simon Thornley.](https://reader036.fdocuments.net/reader036/viewer/2022062320/56649d8c5503460f94a74962/html5/thumbnails/4.jpg)
Cohort study
eg FraminghamPatients without diseaseGroup by exposureCan use a variety of exposuresFollow until disease develops
![Page 5: Introduction to Study Design and RCTs Simon Thornley.](https://reader036.fdocuments.net/reader036/viewer/2022062320/56649d8c5503460f94a74962/html5/thumbnails/5.jpg)
Cohort advantages
Exposure precedes disease Disease status does not influence selection Several outcomes possible Good for rare exposures Control group obvious (compare case-control)
![Page 6: Introduction to Study Design and RCTs Simon Thornley.](https://reader036.fdocuments.net/reader036/viewer/2022062320/56649d8c5503460f94a74962/html5/thumbnails/6.jpg)
Cohort disadvantages
Prospective costlyInefficient for rare diseases with long latencySeveral outcomes possibleExposed followed more closely than unexposed?Loss to follow up causes bias
![Page 7: Introduction to Study Design and RCTs Simon Thornley.](https://reader036.fdocuments.net/reader036/viewer/2022062320/56649d8c5503460f94a74962/html5/thumbnails/7.jpg)
Do computer screens cause spontaneous abortions?
1991
Participants
UnexposedDisease
Unexposed No disease
ExposedDisease
Exposed no disease
Exposed
Unexposed
Participants Exposure Outcomes
Computer screens
No computers
abortion No abortion
Female telephone operators
54 312
82 434
Time
![Page 8: Introduction to Study Design and RCTs Simon Thornley.](https://reader036.fdocuments.net/reader036/viewer/2022062320/56649d8c5503460f94a74962/html5/thumbnails/8.jpg)
Do computer screens cause spontaneous abortions?
Incidence in exposed Relative risk = ----------------------------
Incidence in unexposed 54/(54+312)
= ------------------- = 0.93 82/(82+434)
![Page 9: Introduction to Study Design and RCTs Simon Thornley.](https://reader036.fdocuments.net/reader036/viewer/2022062320/56649d8c5503460f94a74962/html5/thumbnails/9.jpg)
In pictures - Actual
![Page 10: Introduction to Study Design and RCTs Simon Thornley.](https://reader036.fdocuments.net/reader036/viewer/2022062320/56649d8c5503460f94a74962/html5/thumbnails/10.jpg)
Null hypothesis; No effect
![Page 11: Introduction to Study Design and RCTs Simon Thornley.](https://reader036.fdocuments.net/reader036/viewer/2022062320/56649d8c5503460f94a74962/html5/thumbnails/11.jpg)
Effect of Monitors on abortion
Risk ratio
Fre
qu
en
cy
0 1 2 3 4 5
02
00
40
06
00
80
01
00
01
20
01
40
0
![Page 12: Introduction to Study Design and RCTs Simon Thornley.](https://reader036.fdocuments.net/reader036/viewer/2022062320/56649d8c5503460f94a74962/html5/thumbnails/12.jpg)
Cross sectional
Participants sampled at one point or short duration
Exposures and outcomes assessed at same point in time
Participants
UnexposedDisease
Unexposed No disease
ExposedDisease
Exposed no disease
Exposed
Unexposed
Participants Exposure Outcomes
By measurement
![Page 13: Introduction to Study Design and RCTs Simon Thornley.](https://reader036.fdocuments.net/reader036/viewer/2022062320/56649d8c5503460f94a74962/html5/thumbnails/13.jpg)
Cross-sectional
AdvantagesDescribes pattern of
diseaseVariety of outcomes
and exposuresCheapInexpensive
DisadvantagesPrevalent rather than
incident casesCan not distinguish
cause and effectMust survive long
enough to be included in study
Short duration diseases under-represented (e.g. Influenza)
![Page 14: Introduction to Study Design and RCTs Simon Thornley.](https://reader036.fdocuments.net/reader036/viewer/2022062320/56649d8c5503460f94a74962/html5/thumbnails/14.jpg)
Cross-Sectional study - bias
Imagine... People with disease that are sedentary die
early Cross-sectional study of disease (outcome)
and exercise (exposure) Only sample survivors, so find high proportion
of people who exercise with disease What would you infer about causal
relationships?
![Page 15: Introduction to Study Design and RCTs Simon Thornley.](https://reader036.fdocuments.net/reader036/viewer/2022062320/56649d8c5503460f94a74962/html5/thumbnails/15.jpg)
Does wearing fluoro gear protect you from bike crashes?
Participants
UnexposedDisease
Unexposed No disease
ExposedDisease
Exposed no disease
Exposed
Unexposed
Participants Exposure Outcomes
Fluoro colours
No fluoro colours
Bike crash No bike crash
CyclistsTaupo bike race
162 323
588 1343
![Page 16: Introduction to Study Design and RCTs Simon Thornley.](https://reader036.fdocuments.net/reader036/viewer/2022062320/56649d8c5503460f94a74962/html5/thumbnails/16.jpg)
Do computer screens cause spontaneous abortions?
Cum. Incidence in exposed Relative risk = ----------------------------
Cum. Incidence in unexposed162/(162+323)
= ------------------- = 1.10 588/(588+1343)
![Page 17: Introduction to Study Design and RCTs Simon Thornley.](https://reader036.fdocuments.net/reader036/viewer/2022062320/56649d8c5503460f94a74962/html5/thumbnails/17.jpg)
In pictures- Actual
![Page 18: Introduction to Study Design and RCTs Simon Thornley.](https://reader036.fdocuments.net/reader036/viewer/2022062320/56649d8c5503460f94a74962/html5/thumbnails/18.jpg)
Independent
![Page 19: Introduction to Study Design and RCTs Simon Thornley.](https://reader036.fdocuments.net/reader036/viewer/2022062320/56649d8c5503460f94a74962/html5/thumbnails/19.jpg)
Case-control
Investigator selects cases and controls based on disease statusCarefully defined population (cases = control population)Exposure history examined Derive P(exposure | case status)
![Page 20: Introduction to Study Design and RCTs Simon Thornley.](https://reader036.fdocuments.net/reader036/viewer/2022062320/56649d8c5503460f94a74962/html5/thumbnails/20.jpg)
Case-control study: Cases
Ideally, pcase=pdisease
Prevalent From surveypeople with disease at particular point in timeselection bias/favours long lived, chronic cases
Incident from population registryexposure and disease tied only to development of disease, not duration or prognosis.
![Page 21: Introduction to Study Design and RCTs Simon Thornley.](https://reader036.fdocuments.net/reader036/viewer/2022062320/56649d8c5503460f94a74962/html5/thumbnails/21.jpg)
Case control study: Controls
Ideally, pcontrol=pundiseased
Population vs. hospital controlsHospital controls likely to have disease related to exposure, even if not disease of interest.Population controls, from source of cases, generally better approach, but $$ can be prohibitive
Electoral rollsRandom digit dialling
![Page 22: Introduction to Study Design and RCTs Simon Thornley.](https://reader036.fdocuments.net/reader036/viewer/2022062320/56649d8c5503460f94a74962/html5/thumbnails/22.jpg)
Reality More complex; rarely have matches, but
frequency matching more common. E.g. Cot death study Cases – infants who died from cot death
(area)
![Page 23: Introduction to Study Design and RCTs Simon Thornley.](https://reader036.fdocuments.net/reader036/viewer/2022062320/56649d8c5503460f94a74962/html5/thumbnails/23.jpg)
Method Sampling frame – all births in geographic area Frequency matched Control randomly allocated age for interview
similar to age distribution to cot deaths from previous years (about 3 months old)
DOB calculated and adjusted to fit day of week (weekends higher chance of becoming cases)
Obstetric hospital randomly chosen in proportion to number of births in previous financial year
![Page 24: Introduction to Study Design and RCTs Simon Thornley.](https://reader036.fdocuments.net/reader036/viewer/2022062320/56649d8c5503460f94a74962/html5/thumbnails/24.jpg)
![Page 25: Introduction to Study Design and RCTs Simon Thornley.](https://reader036.fdocuments.net/reader036/viewer/2022062320/56649d8c5503460f94a74962/html5/thumbnails/25.jpg)
![Page 26: Introduction to Study Design and RCTs Simon Thornley.](https://reader036.fdocuments.net/reader036/viewer/2022062320/56649d8c5503460f94a74962/html5/thumbnails/26.jpg)
CC - advantages
Good for long latency/ rare diseasesLook at many exposuresSmaller sample size
![Page 27: Introduction to Study Design and RCTs Simon Thornley.](https://reader036.fdocuments.net/reader036/viewer/2022062320/56649d8c5503460f94a74962/html5/thumbnails/27.jpg)
CC - disadvantages
Only one diseaseCan't estimate disease risk, because work backwards from disease to exposure*More susceptible to selection bias as exposure has already occurred.More susceptible to information biasNot efficient for rare exposures (Why?)
![Page 28: Introduction to Study Design and RCTs Simon Thornley.](https://reader036.fdocuments.net/reader036/viewer/2022062320/56649d8c5503460f94a74962/html5/thumbnails/28.jpg)
Case-Control study -example
Participants
UnexposedDisease
Unexposed No disease
ExposedDisease
Exposed no disease
Exposed
Unexposed
Participants Exposure Outcomes
Fenoterol
Ventolin/other
Cases Controls
Adults in hospital with asthma
Fenoterol study, Neil Pearce (guest lecturer)
60 189
57 279
![Page 29: Introduction to Study Design and RCTs Simon Thornley.](https://reader036.fdocuments.net/reader036/viewer/2022062320/56649d8c5503460f94a74962/html5/thumbnails/29.jpg)
Effect measure
odds of exposure in cases Odds ratio= ----------------------------
odds of exposure in controls 60/57= -------------- = 1.55 189/279
![Page 30: Introduction to Study Design and RCTs Simon Thornley.](https://reader036.fdocuments.net/reader036/viewer/2022062320/56649d8c5503460f94a74962/html5/thumbnails/30.jpg)
Actual
![Page 31: Introduction to Study Design and RCTs Simon Thornley.](https://reader036.fdocuments.net/reader036/viewer/2022062320/56649d8c5503460f94a74962/html5/thumbnails/31.jpg)
Null hypothesis; No effect
![Page 32: Introduction to Study Design and RCTs Simon Thornley.](https://reader036.fdocuments.net/reader036/viewer/2022062320/56649d8c5503460f94a74962/html5/thumbnails/32.jpg)
Questions Which study design is best for assessing
causation, assuming no other limitations are present? A) Cross-sectional study B) Randomised controlled trial C) Case-control study D) Cohort study E) Case-series
![Page 33: Introduction to Study Design and RCTs Simon Thornley.](https://reader036.fdocuments.net/reader036/viewer/2022062320/56649d8c5503460f94a74962/html5/thumbnails/33.jpg)
Questions In a cross sectional study of risk factors for
angina, a random sample of elderly subjects were asked the question “Do you smoke cigarettes?” Answers were used to classify respondents as smokers or non-smokers. Further, subjects were classified as positive for angina if they had, at some time in the past, been told by a doctor that they suffered from this condition.
When the data from the study was analysed, no statistically significant association was found between cigarette smoking status and angina status.
![Page 34: Introduction to Study Design and RCTs Simon Thornley.](https://reader036.fdocuments.net/reader036/viewer/2022062320/56649d8c5503460f94a74962/html5/thumbnails/34.jpg)
Has the study measured incidence or prevalence of angina? Explain your answer.
A considerable body of past evidence suggests that the risk of angina increases with increasing tobacco consumption. Suggest reasons why the study described here failed to find an association.
Suggest an alternative design of study that would be more suitable for investigating whether smoking causes angina. Consider the question(s) that you would ask the chosen subjects about their smoking habits.
![Page 35: Introduction to Study Design and RCTs Simon Thornley.](https://reader036.fdocuments.net/reader036/viewer/2022062320/56649d8c5503460f94a74962/html5/thumbnails/35.jpg)
SummaryCharacteristic Cross-
sectionalCase-control Cohort RCT
Selection bias Medium High Low Low
Recall bias High High Low Low
Loss to follow up NA NA High High
Confounding Medium Medium Medium Low
Time required Low Medium High High
Cost Medium Medium High High
![Page 36: Introduction to Study Design and RCTs Simon Thornley.](https://reader036.fdocuments.net/reader036/viewer/2022062320/56649d8c5503460f94a74962/html5/thumbnails/36.jpg)
Summary
Observational
Cohort
Many outcomes, exposures limited
Case- control
One outcome, many exposures
Cross – sectional
Many exposure, many outcomes;
Temporality limits causal inference
ExperimentalRandomised controlled
trial
Ethical constraints
Ideal design