Email updates

Keep up to date with the latest news and content from BMC Pregnancy and Childbirth and BioMed Central.

Open Access Correspondence

The fetuses-at-risk approach: Clarification of semantic and conceptual misapprehension

K S Joseph

Author Affiliations

Perinatal Epidemiology Research Unit, Departments of Obstetrics & Gynaecology and Pediatrics, Dalhousie University and the IWK Health Centre, Halifax, Nova Scotia, Canada

BMC Pregnancy and Childbirth 2008, 8:11  doi:10.1186/1471-2393-8-11

Published: 26 March 2008



Although proponents of the fetuses-at-risk approach describe it as a causal model that resolves various conundrums, several areas of semantic and conceptual misapprehension remain. Differences in terminology include use of denominators such as 'ongoing pregnancies' and the need for an ad hoc 'correction factor' in order to calculate gestational age-specific rates. Further, there is conceptual disagreement regarding the proper candidates for neonatal death and related phenomena. Perhaps the most egregious misconception is the belief that rising rates of gestational age-specific perinatal mortality observed under the fetuses-at-risk model automatically imply the need for indiscriminate increases in iatrogenic preterm delivery.


The term 'fetuses at risk' addresses the plurality of candidates for stillbirth in a multi-fetal pregnancy, while the use of standard terminology such as 'cumulative incidence' and 'incidence density' harmonizes the language of perinatal epidemiology with that used in the general epidemiologic literature. On the conceptual side, it is necessary to integrate clinical insights regarding latent periods into models of neonatal morbidity and mortality. The contention that the fetuses-at-risk approach implies the need for indiscriminate iatrogenic preterm delivery is a non-sequitur (just as rising age-specific cancer death rates do not imply the need for routine chemotherapy and radiation for all middle aged people). Finally, the traditional and fetuses-at-risk models are better viewed in terms of function as prognostic (non-causal) and causal models, respectively.


A careful examination of terms and concepts helps situate the traditional perinatal and the fetuses-at-risk approaches within the broader context of non-causal and causal models within general epidemiology.