Replies: 3 comments 2 replies
-
|
@Lidya1888 good question. It depends on the context, but probably, no. Generally in epidemiology, risk means the probability of something bad happening during some time interval. Here's a definition in Modern Epi 4e (p57): (see also https://en.wikipedia.org/wiki/Risk). In a cross-sectional study, there's usually no time interval, and you can't measure incidence/risk, only prevalence, unless you're making some strong assumptions and/or using data fusion. Do you have a particular use case in mind? |
Beta Was this translation helpful? Give feedback.
-
|
Thank you! In the note, it mentions that if a cross-sectional study is a probability sample of a population (which it rarely is), then we can estimate risks. So, is the note specifically referring to acute and short-lived diseases, and not all types of diseases even if the sampling is probabilistic? |
Beta Was this translation helpful? Give feedback.
-
|
@Lidya1888 are you referring to https://d-morrison.github.io/rme/logistic-regression.html#odds-ratios-in-cross-sectional-studies ? Using the word "risk" in that section was probably not ideal. I'll change it to "prevalence": https://github.com/d-morrison/rme/pull/229/files. "Risk" sometimes gets used to mean the probability of having an ongoing infection at a given time point (in other words, prevalence), but usually it refers to the probability of a new infection during a given time period (i.e., the cumulative incidence proportion). Under some assumptions, there is a relationship between incidence, prevalence, and duration of disease, as you alluded to (see section "Prevalence, Incidence, and Mean Duration" in Modern Epi 4, p. 73-74, for more details), but generally, when analyzing cross-sectional studies, I would recommend avoiding the term "risk". Does that help at all? |
Beta Was this translation helpful? Give feedback.
-
@Lidya1888 PS - maybe this excerpt from https://en.wikipedia.org/wiki/Cross-sectional_study will help more:
See also Lee, James (1994). "Odds Ratio or Relative Risk for Cross-Sectional Data?". International Journal of Epidemiology. 23 (1): 201–3. doi:10.1093/ije/23.1.201. PMID 8194918. I think "prevalence risk ratios" is a good word choice for cross-sectional studies. |
Beta Was this translation helpful? Give feedback.
-
|
Yes I was referring to that section. Thank you, the explanation is clear and is indeed helpful. |
Beta Was this translation helpful? Give feedback.
Uh oh!
There was an error while loading. Please reload this page.
-
In the case of probabilistic sampling in a cross-sectional study, can the prevalence ratio be interpreted as a risk ratio?
Beta Was this translation helpful? Give feedback.
All reactions