3.8
Are there any other epidemiology study types or scenarios that you think that similar calculators would be useful for?
Results
(9 Answers)
-
Expert 3
Any study design except randomized trials would benefit from these calculators. Randomized trials break the link between exposure and outcome. -
Expert 7
Similar calculators would be useful for innovative designs – eg oversampling on high environmental exposure, and then increasing number of bio measurements for highly exposed? -
Expert 4
For Calculator 1, would it be able for the users to determine the optimal combination of m and n by inputting the cost for each measurement and cost for each participant recruited? -
Expert 6
- Cluster-randomized trials
- Within-person variability in outcome measurements
- Meta-analysis across multiple studies
-
Expert 1
Yes, I think more general sample size calculators for RCT, Case-control studies, Cohort studies etc... I realize that there are a lot already out there, but if I had a single website I could go to and do multiple calculations on a single site, it would make life easier but importantly, if the site was professionally developed, I would trust the output I obtained from such an online resource. -
Expert 9
Calculators like this might be useful for any epidemiology studies that collect multiple measures on individuals. Some potential examples follow:- Behavioral epidemiology (epi) studies or educational studies where several measures are taken on each individual, such as each subject responding to the same question over time, or each subject being rated by different evaluators;
- Clinical epi studies with multiple evaluations of the same person. Lab studies are included as biomarkers. But other studies may have the same individual evaluated by different clinicians regarding diagnosis, severity of illness, etc.
- Non-biomarker occupational or environmental epi studies where individuals self-report exposures at different times.
-
Expert 8
I would say these are not only specific for 'biomonitoring' - maybe change the terminology -
Expert 5
Your group could likely create a calculator for categorical data - i.e. using kappa vs intraclass correlation.
Also, often in environmental epidemiology studies, including biomonitoring studies, continuous exposures are categorized into ordinal categories for data analysis, and depending on the shape of the exposure distribution (i.e. skewed), this can introduce additional error that can result in unpredictable bias in different directions. It would be interesting to see the effect of categorization on the bias estimates in the presence of exposure measurement error. -
Expert 2
- Although the calculator doesn't specify a particular type of epidemiological study design, I believe it can be broadly applied to population-based observational study designs, including cohort studies and retrospective cohort designs. Cross-sectional study designs may also benefit from the use of the calculators introduced in the white paper.
- For case-control study designs, where biomonitoring for exposure assessment is based on historical records, a similar type of calculator could be highly useful.
- Additionally, for more advanced study designs, such as those involving stratified or clustered randomization, multilevel data, or Mendelian randomization, similar calculators could also be beneficial.
Expert 4
Expert 3