Week 6 - Cross-sectional Studies

cross-sectional studies are ____

no forward/backward = exposure and outcome happen at the same time

2 types of cross-sectional studies

1. descriptive

   • suggest an outcome prevalence

   • point prevalence = single point in time

   • period prevalence = period of time

2. analytical

   • compare prevalence of outcome between exposed vs. unexposed

characteristics of cross-sectional studies

• observational

• conducted at a single point-in-time or at successive periods of time

• simultaneous

  • selection

  • assignment

  • assessment of subjects

when to use cross-sectional studies

• evaluate a new test/new application of an old test

• evaluate predictive capabilities

• determine prevalence of a prob

• narrow a list of possible etiological agents/causes

pros of cross-sectional studies

• fast

• relatively inexpensive

• efficient (time, resources)

• can result in prospective study

useful when:

• there is an unchanging expo over time

• short time between expo & disease

• there’s a need for prevalence determination

cons of cross-sectional studies

• can’t pinpoint etiological factors

• can’t determine nature of relationship

  • only show associations

• data collection errors

  • no loss to f/u but usually suffer from non-responders

• transient effects

• antecedent-consequent bias

sequential cross-sectional study

• = pseudo-longitudinal study

• use combined data to determine trends

• assume same popu

• probs:

  • changed reference popu

  • changed sampling technique

  • changed responses pattern

components of a cross-sectional study

1. title, abstract, keywords

   • include the timeframe

   • can include the design

   • abstract should include what this cross-sectional study tries to do

2. intro

   • background + objective

3. methods

   • if said “the study has been described in detail elsewhere” → need to include reference → have to go back and look at the reference source

   • popu & sampling technique well defined

     • have inclusion + exclusion criteria

   • samples representative + well defined (age, demographics…)

   • same non-response minimized

   • participants defined objectively + unbiased

   • how they collect data (surveys, forms…)

   • outcomes & expo were objective + unbiased + well-defined

   • stats - which tests were used

4. results + conclusions

   • include flow diagram (participants left…) —> numbers need to match

   • prevalence + OR calculated

   • response rate presented

   • timeline between data collection & publication is short

   • numeric consistency

5. discussions

   • interpretation

   • strengths + limitations

   • do the results match with the objective

6. supplemental

   • author affilations

   • lacking peer review dates

   • statement of conflict

   • statement of funding

   • authors note