I come across a study design named comparative cross-sectional design in few published articles. The authors developed two models in each study, and presented the findings in the result section separately an example in one of the papers was for rural and urban residents, one model for rural and the other for the urban residents. Even though the authors did a separate model for the population, the sample size was calculated together in the method section as a double population formula. Because of a separate model the sample size in the models is very low.
Does really comparative cross-sectional study design an acceptable epidemiological design? As to me, there is no difference with the ordinary cross-sectional design. If the authors think that there could be a difference in the point of their interest between urban and rural residents, they could have handled the variation using an appropriate statistical methodology. I brought the question because I thought cross-sectional design I misused. So, let us discuss if a comparative cross-sectional design is an appropriate design or where it is really different from a cross-sectional one.
Hi,
Such types of studies is done with large sample sizes. Here are few such studies.
Setia M. S. (2016). Methodology Series Module 3: Cross-sectional Studies. Indian journal of dermatology, 61(3), 261–264. https://doi.org/10.4103/0019-5154.182410
Imam, A., Farouk, Z.L., Hassan-Hanga, F. et al. A comparative cross-sectional study of prevalence and intensity of soil-transmitted helminthic infection between healthy and severe acutely malnourished pre-school aged children in Kano, Northern Nigeria. BMC Infect Dis 19, 121 (2019). https://doi.org/10.1186/s12879-019-3755-3
Jadhav AV. Comparative cross‑sectional study for understanding the burden of low back pain among public bus transport drivers. Indian J Occup Environ Med 2016;20:26‑30. (PDF) Available from: https://www.researchgate.
Zelalem Alamrew Anteneh
You need to be aware of some limitations in which their inability to demonstrate a temporal relationship, limits the ability to infer causation.
Cross-sectional study design has three forms
Types of cross-sectional studies
1. Single cross-sectional studies
–Determine single proportion/mean in a single population at a single point in time
2. Comparative cross-sectional studies
–Determine two proportions/means in two populations at a single point in time 3.Time-series cross-sectional studies –Determine a single proportion/mean in a single population at multiple points in time.
In the comparative cross-sectional study your initial hypothesis is "there is a difference b/n the two population proportions of the study of interest". Example there is a difference in contraceptive prevalence (CP) among urban and rural Reproductive age group women. In this case both case-control and cohort study designs are inappropriate because the outcome of interest is prevalence. However, during analysis the first thing that should come in to every body's mind is testing for any statistical difference b/n the two population proportions before doing regression analysis. Then if have significant difference, do separate regression model because they are statistically significantly different due to different influencing factors otherwise merge and do singe regression model including the variable used to divide the population (in my example place of residence) as a factor.
Taddese, I totally agree with your submissions. I m working on a topic involving comparative cross-sectional study between two countries with different characteristics
I would like to add more in this question that, how comparative cross-sectional study is different from case-control study?
In cross-sectional study whether single or comparative, the sample size has to be calculated using a single population proportion formula as well as double population proportion formula, and the higher sample size will be selected and will be used for the study group and the comparative group.
For me, there is a clear difference between the comparative cross sectional and ordinary/ simple cross sectional study. The later follows the assumption of analytical studies which required comparative groups. but a simple cross sectional study is purely descriptive. For instance how can we measure if we believe that there is a difference in disease outcome between sexes (male vs female)? such hypothesis needs a test, require analytical study....comparative cross sectional study.
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Sampling Design
- Sample Size