Search tips
Search criteria 


Logo of epihEpidemiology and HealthThis ArticleAims and ScopeInstructions to Authorse-Submission
Epidemiol Health. 2017; 39: e2017019.
Published online 2017 May 10. doi:  10.4178/epih.e2017019
PMCID: PMC5543293

Postpartum modern contraceptive use in northern Ethiopia: prevalence and associated factors - methodological issues in this cross-sectional study

Dear Editor,

We read the paper entitled ‘Postpartum modern contraceptive use in northern Ethiopia: prevalence and associated factors,’ written by Abraha et al. [1], which was published in Epidemiology and Health in March 2017. The aim of the study was to assess postpartum modern contraceptive use and associated factors among postpartum women in northern Ethiopia. A multivariable logistic regression model showed that the following independent variables were the most important determinants of postpartum modern conception use in the town of Aksum: maternal educational level (secondary and tertiary education level), family planning counseling during pregnancy and during prenatal and postnatal care, having postnatal care, resuming sexual activity, menses returning after birth, and experiencing problems with previous contraceptive use [1]. However, although this research was valuable and the results are interesting, some methodological issues should be considered relating to this cross-sectional study.

Regardless of the results obtained from the model, it should be explained that accurate predictors or determinants of a dependent variable cannot be reliably identified by a cross-sectional study because predictors must be identified based on cohort studies [2,3]. In other words, predictive or casual inferences cannot be made from cross-sectional studies because of the associations between variables measured at the same time point in such studies. Without the temporality assumption (the dependent variable must occur after the independent variable) there is no way of determining whether a factor is a risk factor, is predictive/causal, or is a consequence of the outcome [4]. Therefore, longitudinal studies are essential for developing assumptions to be used in clinical prediction models, whereas in this study [1], a cross-sectional study was used to identify the independent predictors of postpartum modern contraceptive use. Therefore, it is essential to interpret the results of this study in light of the above explanation.


The authors have no conflicts of interest to declare for this study.


1. Abraha TH, Teferra AS, Gelagay AA. Postpartum modern contraceptive use in northern Ethiopia: prevalence and associated factors. Epidemiol Health. 2017;39:e2017012 [PMC free article] [PubMed]
2. Steyerberg EW. Clinical prediction models: a practical approach to development, validation, and updating. Dordrecht: Springer; 2008. p. 38.
3. Ayubi E, Sani M. Carotid atherosclerosis is associated with left ventricular diastolic function: methodological issue. J Echocardiogr. 2016;14:181. [PubMed]
4. Kamper SJ, Hancock MJ, Maher CG. Optimal designs for prediction studies of whiplash. Spine (Phila Pa 1976) 2011;36:S268–S274. [PubMed]

Articles from Epidemiology and Health are provided here courtesy of Korean Society of Epidemiology