This week, I have been attending the ESHRE conference online. Despite my fervent hope, the long-awaited moment to return to in-person conferences has not transpired and I had to revert to virtual participation. Being able to attend remotely was, unexpectedly, a great opportunity. The conference platform was highly effective, and I could follow all the presentations undisturbed, while sitting comfortably at my desk sipping coffee.
With our poster (you can find the abstract here) we wanted to shed light on an overlooked issue related to sharing Time-Lapse (TL) videos with patients. As our last published work shows, patients often watch the videos of their embryos for the first time at home, without the support of an embryologist to answer any questions that might arise. Even if patients are asked whether they want to receive embryo videos, what these videos show is not very clear to them, and they often do not know what to expect. As patients often associated embryo videos with a potential future baby, watching these videos without an appropriate explanation can cause confusion and potential distress. This potential is even higher when embryo videos are shared just after the transfer and before patients are even able to take a pregnancy test. To improve patients’ experience, we urge professionals to consider further how, where and when these videos are shared with patients.
I was very happy to see that our suggestion was welcomed by one of the main experts and author of the two Cochrane reviews on TL, Sarah Armstrong. In her talk, opening the session dedicated to “The current state of Time-Lapse technology”, she offered an overview of quantitative and qualitative studies available currently on TL. Although the focus of her talk was discussing current evidence of its clinical benefit, she also mentioned our work on patients’ experiences of receiving and watching embryos, opening up for a discussion on whether, when and how clinics should share embryo videos with their patients.
In this and another presentation, Sarah also shared the highlights of the VALUE study, an international qualitative study exploring patients’, embryologists’, and clinicians’ decisions to use non-evidence-based treatment ‘add-ons’. The early results of this study show some significant overlap with the results of our project. For instance, as we discussed in one of our previous publications, Sarah noticed how the decisions to use add-ons should be seen in the context of hope for both patients and professionals. Applying the current (missing) evidence is challenging for both professionals (as we have also discussed here) and patients (as we discuss in another article currently under review). Finally, she argued that patients don’t want to be denied add-ons but want autonomy in the form of informed consent. I really look forward to reading the detail of the VALUE study results.
The discussion on the lack of evidence on TL efficiency was not limited to these talks, as Dorit Kieslinger presented the results of the last RCT on TL: The SelecTIMO study. This study, including 1,731 women, investigated embryo culture in the Geri+ incubator and the Eeva® Test algorithm. Their results show that the use of TL, with or without algorithm-based selection, does not improve (cumulative) ongoing pregnancy and live birth rates and therefore its widespread use to increase success rate should be questioned. Despite the results, the discussion after the presentations followed the same logics we showed in our previous research. The audience questioned the evidence (results apply only to the incubator and algorithm examined, others could work better) and the need itself for TL to prove its efficacy in terms of increasing success rates (we use it because it’s a great piece of lab equipment). This discussion goes back to the origin of the problem, which for many does not seem to be if there is evidence to support the ongoing use of these tools, but rather that patients are charged for tools that might not increase their chances of having a baby. As a participant from Denmark commented, TL is very widespread among Danish labs but patients are not charged for its use, and nobody talks about add-ons.
It is interesting to see how the use of artificial intelligence (AI) is becoming more widespread in the field, without discussion regarding the evidence supporting it. A discussion in the General Assemby highlighted how all the current publications on the topic of AI, do not offer any information on the data, the data points measured, or the content of the algorithms, justified in terms of the defence of intellectual property. It is perhaps surprising that here, unlike in the debate on add-ons, the question of evidence seems to have been overlooked.