Research Designs - Strengths/Weaknesses

0.0(0)
Studied by 7 people
call kaiCall Kai
learnLearn
examPractice Test
spaced repetitionSpaced Repetition
heart puzzleMatch
flashcardsFlashcards
GameKnowt Play
Card Sorting

1/28

encourage image

There's no tags or description

Looks like no tags are added yet.

29 Terms

1
New cards

What are the strengths of a meta-analysis/systematic review?

  • summarize all data on the topic

  • increase power to look for association

  • reveal peculiarities of different studies


2
New cards

What are the weaknesses of a meta-analysis/systematic review?

  • time and effort

  • publication and reporting bias

  • garbage in, garbage out


3
New cards

What are the strengths of a RCT?

  • weigh the benefits/harms

  • minimize bias due to co-interventions (if blinded)

  • balance confounding at baseline


4
New cards

What are the weaknesses of a RCT?

  • time consuming

  • costly

  • not good for rare outcomes


5
New cards

What are the strengths of a before-after study?

  • evaluate effect of a discrete, widespread intervention or exposure

  • real world setting


6
New cards

What are the weaknesses of a before-after study?

  • lack of blinding (performance and detection bias are substantial risks)

  • non-concurrent control group creates risk of confounding by temporal trends


7
New cards

external control group

Use of an _____ ______ _____ allows for adjustment for secular and seasonal trends

8
New cards

secular trends are minimized by…

short time intervals

9
New cards

seasonal trends are minimized by…

timing measurement with seasons

10
New cards

What are the strengths of a before-after study (with the same people in the before and after groups)?

  • including everyone in the population is ideal (no sampling)

  • many variables (age, gender, race, etc) are the same before and after, reducing risk of confounding


11
New cards

What are the weaknesses of a before-after study (with the same people in the before and after groups)?

  • potential confounding by time (non-concurrent control)

    • change in variables over time (e.g. age)

    • unintentional learning can confound the association

    • regression to the mean can confound the association if participants were chosen based on extreme values

    • attrition bias is a risk


12
New cards

What are the strengths of a before-after study (with different people in the before and after groups)?

No risk of unintentional learning, regression to the mean, or attrition bias

13
New cards

What are the weaknesses of a before-after study (with different people in the before and after groups)?

  • If there is any sampling of the population in obtaining the before and after groups, selection bias is a risk

  • Lack of randomization, so confounders are not balanced at baseline


14
New cards

What are the strengths of a prospective cohort study?

  • establish time sequence

  • Allow assessment of predictors unbiased by knowledge of outcome (avoid recall bias)

  • Good for rare exposures

  • Can study several outcomes


15
New cards

What are the weaknesses of a prospective cohort study?

  • Unmeasured confounders

  • Costly & Inefficient

  • Often requires large N

  • Hard to assess rare outcomes

  • Loss to follow-up


16
New cards

What are the strengths of a retrospective cohort study?

  • Similar strengths to the prospective cohort

design.

  • Less expensive

  • Less time-consuming


17
New cards

What are the weaknesses of a retrospective cohort study?

  • Similar to prospective cohort (except they are not costly and inefficient)

  • limited control over choice, quality, completeness, and accuracy of variables


18
New cards

What are the strengths of a multiple cohort study?

  • Only feasible approach for studying rare exposure to occupational/environ hazard

  • Using data from census/registry as external control group is population based & economical


19
New cards

What are the weaknesses of a multiple cohort study?

  • Accentuate problem of confounding because cohorts are assembled from separate populations


20
New cards

What are the strengths of a case-control study?

  • useful for studying conditions with long latent periods (because everything has already happened at the start of the study)

  • need relatively small number of participants

  • can be good for hypothesis generation


21
New cards

What are the weaknesses of a case-control study?

  • At risk for confounding and differential biases depending on RQ and methods used to collect data

    • potential for survivor bias in studies in which subjects need to be alive to answer questions

    • potential for recall bias in studies in which the predictor variables are measured retrospectively

  • limited to one outcome variable

  • does not establish a sequence of events


22
New cards

What are the strengths of a nested case-control study?

  • advantages of a cohort design

  • can be less costly if measuring predictor would have been costly

  • avoid sampling and recall biases


23
New cards

What are the weaknesses of a nested case-control study?

  • need a pre-existing defined cohort

  • they can be subject to differential recall bias if one or more is assessed retrospectively


24
New cards

What are the strengths of a case crossover study?

  • same as case control

  • avoid effect-cause with time component, confounding, and selection bias by using the same person as their own control


25
New cards

What are the weaknesses of a case crossover study?

  • only for short-term effects of intermittent exposures


26
New cards

What are the strengths of a cross-sectional study?

  • fast; don’t need to wait for follow up

  • no loss to follow up (though can have issue of item non-response)

  • can be converted to cohort if time component is added (but the question will be different)


27
New cards

What are the weaknesses of a cross-sectional study?

  • no time component, so at risk for effect-cause

  • At risk for confounding and differential biases depending on RQ and methods used to collect data (e.g. recall bias, detection bias)

  • Risk for sample bias


28
New cards

What are the strengths of a case series/case report?

  • can generate hypothesis

  • can increase awareness

  • inexpensive

  • can be converted to a case-control with the addition of a control group

  • most often used for rare outcomes, but can be used for rare exposures (occupational or environmental) or new interventions (e.g. new surgical procedure)


29
New cards

What are the weaknesses of a case series/case report?

  • No control group

  • Not valid for cause-effect interpretation

  • Concerns are the greatest for case-series

which are evaluating an intervention

  • When used to establish cause-and-effect,

can result in false leads, scares, and

wasted resources