Cross Sectional Study

Question 1

Describe a Cross Sectional Study:

Answer 1

1. A sample of, or total reference population is examined at a given point in time
2. Records information on disease/outcomes + exposures at a single point in time
   = snapshot of cohort study
3. Comparison of exposure in those with/without disease is equivalent to a case-control study
4. Cross sectional sample is like sampling those with /without disease at one point in time

Question 2

More about a Cross Sectional study

Answer 2

Can be population based or community based survey
Subjects are interviewed at a point in time without follow up
Can combine a cross sectional study with follow up to create a cohort study
Can conduct repeated cross sectional studies to measure changes against time

Question 3

Point vs Period Prevalence

Answer 3

Point Prevalence : do you currently have a backache?

Period Prevalence : Have you had a backache in the last 6 months?

Question 4

Non-equal probability sampling

Answer 4

Oversampling of particular populations in order to ensure adequate representation of those individual populations for analysis

Leads to a sample that doesn’t represent the general population

Question 5

Survey Weight + Calculation

Answer 5

Survey weights are used to compensate of over-/under- sampling of subject groups
Ex: If we double the size of our sample from minorities, each minority gets a weight of 0.5
- Makes the statistics representative of the population
- Only 1 weight per subject. Weights for different factors must be combined into one weight

Calculation:
Survey weight = 1 / probability of sampling, or “sampling fraction,” or oversampling amount for a given group

Ex: oversample 5x Asians than Whites –> weight of Asians = 1/5, weight of Whites = 1

Question 6

Bias in Cross Sectional Studies: Length-biased Sampling, Prevalence- incidence Bias

Answer 6

1. Study population often accrued through convenience sampling (non-probability) = not representative of general population
2. Systematic increase/decrease of cases
   - length- biased sampling - cases are overrepresented if illness has a long duration, underrepresented if short duration
3. Systematic increase or decrease of exposure?
   - prevalence-incidence bias: exposure doesn’t alter disease risk, but alters disease duration

Question 7

Strengths of Cross Sectional Studies

Answer 7

1. Relatively feasible, non-time consuming - no follow/up
2. Can study several diseases/exposures, thus useful for screening new hypotheses
3. We can describe disease frequency and health needs of a large population = useful for health planning
4. Serial surveys can be done so that we can monitor a trend of disease

Question 8

Limitations of Cross Sectional Studies

Answer 8

1. Rarely know when disease occurred among cases = temporal ambiguity, did E precede O?
2. Certain exposure info is vulnerable to measurement error, especially if collected retrospectively = bias in effect estimation
3. Prevalence cases occurred before study started = disease status can influence selection of subjects
   = bias in effect estimation
4. We don’t know duration of prevalent cases = can’t distinguish risk factors from prognostic factors
5. Inefficient for rare/ highly fatal / short duration diseases