8. MTB Step 3 - Study Design, Analysis of Results, Bias Flashcards Preview

1.06) BIOSTATISTICS > 8. MTB Step 3 - Study Design, Analysis of Results, Bias > Flashcards

Flashcards in 8. MTB Step 3 - Study Design, Analysis of Results, Bias Deck (12)
Loading flashcards...
1
Q

STUDY DESIGN

RANDOMIZED CONTROLLED TRIAL (RCT)

  1. What does “Randomization” mean?
  2. Which type of Bias does a Randomized Controlled Trial (RCT) help avoid?
A
  1. Randomization = persons, animals, or samples are sorted into Different Arms of the Study by Computer or a Randomly Generated list of assignments.
  2. Selection Bias - on the basis of the patient being enrolled, on economic or insurance considerations, or by the investigating physician. For the purposes of sorting, all that matters is that at the end, the Same Number of patents ends up in Each Group of the Trial
2
Q

STUDY DESIGN

RANDOMIZED CONTROLLED TRIAL (RCT)

Which type of Study is a Randomized Controlled Trial (RCT)?

A

PROSPECTIVE Study

3
Q

STUDY DESIGN

RANDOMIZED CONTROLLED TRIAL (RCT)

What is a Benefit of Prospective Studies (i.e., RCTs) that Retrospective Studies do not share?

A

With Prospective Studies, if Clear Harm or Clear Benefit is observed BEFORE the end of the RCT, an independent Data Monitoring group Stops the Study

4
Q

STUDY DESIGN

COHORT STUDY

When would you choose a Cohort Study Design?

A

A Cohort Study is undertaken in order to observe over time what happens to groups of patients with certain exposures or underlying illnesses.

Example:

  • To look at HRs and Eye Disease in those who Smoke or have HTN or Diabetes, a study can observe these groups of patients (“cohorts”) over many years.
5
Q

STUDY DESIGN

COHORT STUDY

Which type of Study is a Cohort Study?

A

Observational

and

PROSPECTIVE

Observational = NO intervention. Just observing the effect of Exposures over a period of time.

Used to assess the Risk of Disease. Relative Risk (RR) is used to assess Cohort Study results.

6
Q

ANALYSIS OF RESULTS

How is Relative Risk (RR) calculated?

A

RR = a / (a + b)

c / (c + d)

RR looks at the risk of a disease based on who was exposed to a potential danger in the past. RR for a cohort study starts with an asymptomatic group and calculates the comparative risk of developing disease either with the exposure or without the exposure.

7
Q

STUDY DESIGN

What type of study is a Case-Control Study?

A

Retrospective Study

  • Case-Control studies are looking for the Odds of a previous exposure on the development of a rare disease manifestation.
  • Case-Control studies start with people who have a disease and look BACKWARD at other groups that are otherwise matched to assess for risks of exposure.
  • Case-Control studies are subject to Recall Bias about what people may have been exposed to in the past.
8
Q

ANALYSIS OF RESULTS

How do you calculate an Odds Ratio (OR)?

A

OR = Cases Exposed / Cases Not

Controls Exposed / Controls Not

OR starts with those who Have a Disease and then looks for the chance of Having an Exposure.

9
Q

BIAS

Describe SELECTION Bias?

A

Selection Bias occurs, for instance, when an investigator chooses Less Ill patients for the Drug Side of the trial and Sicker patients for the Placebo side – contructing outcomes that make the drug look more successful than it really is.

  • For instance, a trial of Antidepressants in which previously suicidal patients or those with psychiatric hospitalization are chosen for the drug side can make the drug look less successful if compared to less depressed persons.
10
Q

BIAS

Describe BERKSON Bias

A

Berkson Bias is when Hospitalized patients are used as trial subjects instead of the general population.

  • This type of Bias is solved by Random Selection of trial subjects
11
Q

BIAS

What is the HAWTHORNE Effect?

A

The Hawthorne Effect is when those being studied know they are being watched for the effect of a drug or intervention.

  • This is solved by using a Placebo Control and Blinding both the investigator and the participants.
12
Q

BIAS

Describe LEAD-TIME Bias?

A

In Lead-Time bias, EARLY DETECTION is confused with Increased Survival based on treatment.

  • For instance, early detection of minor cancerous cells such as prostate cancer can make it look like screening was a benefit.