Feb to March 5

Question 1

Descriptive studies vs analytical studies

Answer 1

Descriptive looks at distribution.
Analytical looks at determinants

Question 2

Define the following as descriptive or analytic: Case report, quasi-experiment, cohort studies (retro, pro), case series, case-control, cross-sectional, longitudinal, ecological

Answer 2

Case report and case series are descriptive

Question 3

What do analytic study designs test

Answer 3

one or more predetermined hypotheses about associations between an exposure and a health outcome

Question 4

Exposure vs outcome

Answer 4

• Exposure – any explanatory (“independent”) variable considered a possible health determinant
• Outcome – any health outcome (“dependent”)

Question 5

Two main types of analytic designs

Answer 5

Experimental and observational

Question 6

Cohort design

Answer 6

At baseline all participants are disease free and at risk of developing the outcome of interest
2nd observation point documents incidence

If incidence in exposed is higher than the unexposed that implies association in outcome

Question 7

One-sample cohort

Answer 7

Exposure is initially unknown and participants are assessed to determine exposure status

Question 8

Special cohort

Answer 8

Exposure status is known at the outset, then the comparison is selected to match as much as possible

Question 9

prospective cohort

Answer 9

 Exposure baseline in the present
 Follow-up period: present to future
 When the study begins, the outcomes have not yet developed (researchers wait for them to occur)

Question 10

Retrospective cohort study

Answer 10

 Exposure baseline in the past
 Follow-up period: past to present
 Both the exposures and outcomes have already occurred by the time the study begins

Question 11

Case-control design

Answer 11

• Participants are selected into the study and grouped on the basis of disease status: cases (disease) and controls (no disease)
• Exposure status is unknown at the beginning of the study
• Cases and controls are then compared regarding
  their history with regard to the exposure of interest

Question 12

Cross-sectional studies

Answer 12

Presence or absence of exposure and disease for each individual is determined at the same time

Question 13

How do you figure out the strongest association based on relative risk

Answer 13

The more the RR departs from 1, the stronger the association between exposure and outcome

Question 14

Rule of thumb for interpreting strength of ratios

Answer 14

Question 15

Which study uses pearsons correlation coefficient (r)

Answer 15

Ecologicla studies

Question 16

Issues in hypothesis testing

Answer 16

◼ We are trying to evaluate the role of chance as a potential explanation for observed results.
◼ Only assesses chance – there is the potential that other uncontrolled factors may be responsible for the observed association

Question 17

How do you interpret Confidence intervals for Risk Ratio

Answer 17

on average 19 out of every
20 (95%) such CI’s would contain the
true population value and one of every
20 (5%) would not

Question 18

What is a case study

Answer 18

Participants include cases and control
Exposure is assessed retrospectively

Question 19

Disadvantages of community controls

Answer 19

Time consuming and expensive to obtain
Low participation rates
Differential recall of exposures

Question 20

Advantages of a hospital control group

Answer 20

Easily accessible
Have time to participate
Motivated to cooperate
Differential recall is minimized

Question 21

Disadvantages of hospital control group

Answer 21

Greater potential for selection bias

Underestimate the study effect

Question 22

How to minimize differential misclassification

Answer 22

Use same procedures
Blind interviewers and participants
Objective markers of exposure
Cross-check

Question 23

How to minimized non-differential misclassification

Answer 23

High quality instruments (validity and reliability)
Objective measures
Ensure confidentiality of sensitive info standardized data collection
Use multiple measures
Train data collectors

Question 24

What is selection bias

Answer 24

Error due to systematic differences in characteristics between those who are selected for the study and those who are not.

Question 25

Strengths of case-control study design

Answer 25

Efficiency: precise estimates from smaller samples
Fit with studying rare diseases
Fit with long induction and latency period
Examines multiple exposures
less expensive and faster than some other designs

Question 26

Weaknesses of case-control

Answer 26

Vulnerable to selection and misclassification bias
Not good fit for rare exposures
Limited to estimating odds ratios
temporality issues: timing of exposure and disease may be unclear

Question 27

How to interpret a prevalence ration of 0.70

Answer 27

The prevalence of psychosis is .70 that of smokers compared to non- smokers.
The prevalence of psychosis is 30% lower for a smoker than for a non-smoker

Question 28

Limitations of cross-sectional studies

Answer 28

Difficulty establishing temporality
Unsuitable for rare exposures or diseases
Focuses on prevalence not incidence (cant estimate risk)

Question 29

What kind of study was Framingham

Answer 29

Cohort