A paper claims to measure incidence and prevalence in one study on end-stage renal dialysis, so patients go on and don't come off. Any new cases that appeared (started dialysis) in the most recent year are classified as an incident case, but every case before the current year a prevalent case. All cases are presumably still alive at the time of analysis (not mentioned if they died). Is this really measuring prevalence and incidence? Then all cases are combined to assess differences between treatment centers.
Interesting, thank you. So they're measuring one or the other, not incidence and prevalence? However, in Hennekens (Epidemiology in Medicine) it states: "this measure [period prevalence] is not frequently used since it combines both point prevalence and incidence in a single parameter." So seems there's no point in distinguishing incident patients and prevalent patients. Both are calculated in the same way, as a proportion of the total population, which is the same total population for both groups.
From the report: "
"The characteristics of our study group of 587 patients are shown in Table 2. More than two thirds of the patients were Yemani (70.7%), and most were males (62.5%). Ages ranged widely in all health sectors, from 18 to 95 years old; the mean (SD) age was 50.8 (16) years. The incidence of hemodialysis (those who started their first regular hemodialysis sessions during 2012) accounted for only (11.9%) as compared with (88.1%) prevalent patients."
They seem to be defining prevalence as 517/587 (517 are the cases that are not incident cases), then the incidence is 70/587. They haven't calculated a prevalence or incidence at all, which ought to be 587/? some figure for the population that the cases are drawn from.
Dear John;
Thank you for sharing with me this point. I think it is unusual to calculate incidence and prevalence on one study. This might be possible if the design is nested cross-sectional cohort study, in your problem it seem difficult to do this design. As we know prevalence contain incident cases inside but not the inverse, so with this design only permit you to calculate the percentage. As was been stated by Martin, this design will not supportive to calculate the incidence (Problem of exposure).
- Badges
- Science topic
- Similar topics
- Medicine
- Disease
- Infectious Diseases