1. Université Paris Descartes, Paris Sorbonne Cité – Assistance Publique Hôpitaux de Paris, CHU Cochin, Department of Obstetrics and Gynecology, Division of Reproductive Medicine, Paris, France.
2. Department of Obstetrics and Gynecology, Kaplan Medical Center, Rehovot, Israel.
3. Summarry of a presentation made at: The 2nd IVF Worldwide Live Congress: In vitro fertilization clinics growing in the digital age, Oct 31 – Nov 2, 2013. Berlin, Germany.

Thirty years later, neither 0% nor 100%

An enduring characteristic of assisted reproductive technologies (ART) is the fact that its outcome has remained – as it was at outset – neither 0%, nor 100%. Had pregnancy rates (PR) in ART been either 0% or 100%, the issues discussed today wouldn't be issues anymore. With PRs of or nearing 0%, ART wouldn't be an option for treating infertility. At the opposite end if PRs constantly neared 100%, quality assessment would be greatly simplified, needing less discussion. In the latter case, one single failure would be in itself evidence of a problem in a realm where normally all pregnancy tests should be positive. In case of anesthesia for example, awakening the patient after surgery is expected to be 100% effective. In this context, one single failure is by itself evidence of a problem that needs close and immediate attention.

      But, years after – over three decades later – PRs in ART still hover between 0% and 100%. Of course improvements have been made over time and ART numbers have been steadily rising. In today's environment, a first attempt at ART in a patient of <37 years of age is expected to yield PRs of ≥50%. Whether actual outcome is either closer to 40%, or 60% does not fundamentally change the problem, which is that ART failures do exist and are thus part of the picture. But how do failures occur? For quality control purposes, it would be nice if a clinic having PRs of say 50% in a given subgroup of patients would witness a predictable alternation of positive and negative outcomes. We however know that this is never the case. Cases – positive and negative – tend to cluster for reasons that escape our understanding. Therefore the question is: How many successive negative pregnancy tests exceed what can be explained by chance alone and thus, constitute evidence that a problem occurred and affects ART outcome? Conversely, how many cases are needed in a series of above-average positive pregnancy tests for constituting evidence of an overall improvement in ART efficacy, rather than being an effect of chance alone? With small numbers, it is often hard to tell a true improvement from a bout of good fortune. If they are too hasty, interpretations can easily drift and lead to deceptive reports based on clusters of good cases. For all these very reasons data reported over limited period of time may be misleading. Considering the sensitive nature of these data and their opaque nature in the past, the public has long been demanding for more non-impeached visibility on ART outcome.

This or this other pregnancy rate?

      Too often in its history, ART has allowed the sometimes-misleading if not deceptive use of intermediate outcome parameters (e.g., clinical pregnancy rate – cPR), instead of the definitive live born baby rates (LBR). Sadly, this has fudged visibility on true ART outcome and eroded public confidence. The concerns were stressed by the fact that differences between intermediate outcome parameters (i.e. cPR) and LBR – the early pregnancy waste rate – tend to be inversely related to the overall quality of a given ART program. Intermediate outcome parameters are thus best avoided for describing ART outcome.

The obligation to report outcome

      For the reasons laid out above and because of the highly media sensitive character of infertility issues, ART has been required nearly worldwide to report results to government agencies in standardized fashion. This is notably the case in the US, Canada, most European countries, Israel, Australia and New Zeeland. In several countries, reports must be made prospectively online within a few days of starting the control ovarian stimulation (COS) cycle.
Issues exist on what ART outcome should be tied to. Classically, results have been computed on per embryo transfer (ET) or per oocyte retrieval bases. Certain claim however that ART outcome ought to be reported instead on per given periods of time during which infertility treatment is provided – i.e. yearly. This is meant to better reflect the relative benefit of the so-called minimal-stimulation ART, which provide smaller oocyte harvests. Indeed minimal-stimulation ART yields lower per ET results, but can be more frequently repeated over time. Ultimately therefore, minimal stimulation may provide an overall outcome that equals, or certain claim surpasses, classical treatments if it is computed on per-year bases.

      Irrespective of the enduring debate on the mode of reporting ART outcome it is important to note that ART is a nearly unique position by having to conduct those reports. Indeed, only limited other medical activities are required to report such detailed treatment outcomes. As discussed below, having to report outcome can be an asset.

Data management system for reporting outcome

      The obligation to report ART data to governmental agencies, which has generalized worldwide, created needs for data management systems capable of preparing and executing such reports. Several systems dedicated to ART are commercially available and chosen by an increasing large number of ART programs. Other groups prefer relying on 'home-made' data management systems specially crafted for them. Dedicated systems such as offered by notably, Baby Sentry® and others are gaining interest however. This is in part because they offer novel features such as notably, 'aps' that allow patients to check their results and receive their instructions directly on their smart phone.

Data reporting, a burden to take advantage of

      The obligation to report outcome to government agencies has been seen perceived as a burden if not an infringement on individual freedom and greeted with moderate enthusiasm in the world of ART. Today however, we must realize that there is no coming back and that the obligation to report outcome is here to stay. Hence, the need for ART reports and the necessity of funneling clinical activity through data management systems capable of conducting such reports should be seen as an opportunity rather than a strain that befell on ART. As discussed below, ART has indeed the unique possibility of showing the way to other areas of medicine on how data reporting systems can open to multiple new uses notably, in education.

Infertility and ART in pole position for leaping in digital-age medicine

Data management systems, an opportunity for ART

The need to report outcome actually puts infertility and ART in pole position for leaping forward into the modern realm of digital-age medicine. Infertility and ART are indeed in the privileged position of showing to other fields how to venture into this new frontier of medicine. As discussed below, one of the first possible applications of data management system conceived and implemented for reporting ART data to the government is the domain of education.

A library of virtual cases

ART data management systems – particularly, if coupled with electronic medical records (EMR) – constitute a mine of medical information capable of offering far more benefit than merely reporting data to the government. One of the opportunities offered by data management systems originally designed for reporting ART data is to serve as a source of clinical experience capable of illustrating training in this field. For this, we envision a system capable of extracting in an anonym way and storing actual clinical cases encountered in daily infertility and ART practice for further use in education.

A 3D Central Education Unit (CEU)

The clinical cases selected for their ability to illustrate a given segment of infertility treatment and ART treatment must be stored in a system allows easy on-demand retrieval in order to facilitate practical use in education. For this, we have conceived a 'Le DU' central education unit – it is already used by us for storing all classical teaching materials dedicated to infertility and ART – whereby all education materials are stored in individual compartments. As illustrated in Fig. 1, six functional categories exist. The categories are classified following a functional classification that shadows the clinical management of infertility patients, going from the 1. workup to 2. diagnosis, 3. risk assessment, 4. treatment, 5. complication and 6. final assessment steps. Each category contains six compartments in which education materials and virtual cases are stored according to three depths of increasing complexity, categorized as the must-know, nice-to-know and expert levels. The '3D'-CEU storage system developed for harboring teaching materials therefore constitutes an ideal basis for also hosting the library of virtual clinical cases meant to serve in education.

Conclusion

Rather than seeing the obligation of reporting data to government agencies, infertility and ART programs should the existence of data management systems dedicated to that task, as an opportunity rather than a burden. ART data management systems can indeed become the data basis for developing library of virtual cases aimed at exposing doctors in training to circumstances too uncommon for being encountered during their training program (fellowship).