Defining Case Control Studies

6 04 2011

We read with great interest the article “Obstetric outcome of women with in vitro fertilization pregnancies hospitalized for ovarian hyperstimulation syndrome: a case-control study” by Courbiere et al. (1). It is our belief that, despite the title, the study did not adhere to the principles of a case-control study.

The defining feature of a case-control study is the selection of study participants on the basis of disease (or outcome) status. In the current study, the “cases” were women who were pregnant through in vitro fertilization (IVF) and were hospitalized for severe ovarian hyperstimulation syndrome (OHSS), while the “controls” were similar women without OHSS. OHSS morbidity and pregnancy outcomes and obstetric complications were then evaluated between these two groups. Thus, OHSS represents the exposure (or risk factor) of interest and the selection of study participants with respect to this characteristic who are then followed-up for specific outcomes of interest means that this is a cohort study, not a case-control study. For it to be a case-control study, the study participants would have been selected based on the presence and absence of single or multiple obstetric outcomes and the frequency of OHSS between those groups would have been compared.

Another important issue relates to the authors’ description of the cases being matched to the controls; it is not clear what characteristics the groups were matched. Assuming they did indeed match OHSS (exposed) and non-OHSS (unexposed) participants, the authors should have conducted the appropriate analysis for a matched cohort study. The authors are strongly urged to conduct a reanalysis of their study data and provide readers with the appropriate measures of association and p-values, amending their interpretations as warranted.

Asundep Ntui, D.V.M., M.S.
Aditi Shendre, M.P.H.
Lesleigh Lee, M.P.H.
Gerald McGwin, Jr, M.S., Ph.D.
Department of Epidemiology
University of Alabama at Birmingham
Birmingham, Alabama

References
1. Courbiere B, Oborski V, Braunstein D, Desparoir A, Noizet A, Gamerre M. Obstetric outcome of women with in vitro fertilization pregnancies hospitalized for ovarian hyperstimulation syndrome: a case-control study. Fertil Steril 2011; 95:1629-32.

2. Mayo Foundation for Medical Education and Research (MFMER). (2010, July 8). Mayo Clinic. Retrieved January 15, 2011, from Mayoclinic.com: http://www.mayoclinic.com/health/ovarian-hyperstimulation-syndrome-ohss/DS01097/DSECTION=risk-factors

Published online in Fertility and Sterility doi:10.1016/j.fertnstert.2011.04.012

The Authors Respond:

We thank Ntui and colleagues for pointing out a problem in the presentation of our study (1).

This exploratory study was initially designed to be a case-control study of risk factors of ovarian hyperstimulation syndrome (OHSS) (comparing OHSS IVF- pregnancies versus non-OHSS IVF-pregnancies, as illustrated by Table 1). It was later extended with information gathering about obstetrical outcomes of those pregnant women, which became the main goal of the study. We agree with Ntui and colleagues that the principal results of our study are not part of a case-control study, as mistakenly stated in the article title. This problem was not detected during the reviewing process, and we apologize to the readers for this confusion.

However, we do not think that this terminology problem has any effect on the results or on the conclusion of this article. No formal hypothesis has been tested, no measure of association (odds ratio or relative risk) has been given, and we avoided giving the false impression of a formal case-control or cohort study. We only used tests of difference between the two groups (Chi square, Fisher exact test when Chi square was not applicable for rates, Student t test for difference between means) to identify potential risk factors (Table 1) or adverse outcomes linked to OHSS (Table 2). We avoided using unmitigated assertions in the redaction of the discussion and the conclusion of the article to reflect the exploratory nature of this study.

Regarding the question of matching, each patient presenting with OHSS was matched with two patients without OHSS, based at first on the type of pregnancy (singletons, twins, triplets), secondly on ART technique (IVF or ICSI) and finally on age. However, because of limits in recruitment, this rule could not be implemented strictly. Two triplets had discrepancies on method of ART and age differences within triplets (intended to be less than two years) ranged as far as eleven years in one case. This lack of regularity in matching and above all the exploratory nature of the study lead us to adopt a conservative nonmatched analysis, rather than the more powerful matched analysis (2). In the context of an exploratory study without a formal single a priori hypothesis, using a matched analysis would increase the chances to find spurious significant relations between variables.

Blandine Courbiere, M.D., Ph.D.a,b
David Braunstein, M.D.b,c
Virginie Oborski, M.S.d
Anne Desparoir, M.D.a,b
Agnes Noizet, M.D.a
Marc Gamerre, M.D.a,b

aDepartment of Gynaecology and Obstetrics
Unit of Reproductive Medicine
AP-HM La Conception
bFaculte de Medecine de Marseille
Universite de la Mediterranee
cEcole Universitaire de Maieutique Marseille Mediterranee
Universite de la Mediterranee
dPublic Health Department
AP-HM Sainte Marguerite
Marseille, France

References
1. Courbiere B, Oborski V, Braunstein D, Desparoir A, Noizet A, Gamerre M. Obstetric outcome of women with in vitro fertilization pregnancies hospitalized for ovarian hyperstimulation syndrome: a case-control study. Fertil Steril 2011; 95:1629-32.

2. Bland JM, Altman DG. Matching. BMJ 1994;309:1128.

Published online in Fertility and Sterility doi:10.1016/j.fertnstert.2011.04.011

Advertisements

Actions

Information

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s




%d bloggers like this: