Drug Safety Matters

#36 Pregnancy-related pharmacovigilance – Levente Pápai, Lovisa Sandberg & Sara Vidlin

• Uppsala Monitoring Centre

Share episode

There are many reasons why use of medical products during pregnancy requires special attention. First and foremost, we want to be sure that the medicine is as safe as possible for both the pregnant person and the unborn child. Unfortunately, the safety profiles of medicines used in pregnancy are often incomplete, which makes it difficult for patients and healthcare professionals to make informed decisions.

The Research section at Uppsala Monitoring Centre has a team that is currently focussing their efforts on pregnancy-related pharmacovigilance (PV). In this episode, Data scientists Sara Vidlin and Levente Papai, and Senior Pharmacovigilance scientist Lovisa Sandberg from this team, discuss complexities and challenges of pregnancy-related PV, and new solutions for addressing those challenges. 

Tune in to find out

  • Why is the world still behind when it comes to pregnancy-related pharmacovigilance?
  • What are the challenges faced by pharmacovigilance assessors wanting to look at pregnancy cases?
  • How can healthcare professionals, patients and carers help assessors overcome these challenges, when reporting pregnancy-related adverse drug events?
  • How can the VigiBase pregnancy algorithm, and other algorithms, support the identification of pregnancy cases?

How to use the VigiBase pregnancy algorithm

  • Users of VigiLyze and VigiBase Custom Searches can use the VigiBase pregnancy algorithm as a filter when performing searches. 
  • In the qualitative view in VigiLyze, click on “Filter” -> “Patient” -> “Pregnancy” to apply the filter.

Want to know more?

Join the conversation on social media
Follow us on X, LinkedIn, or Facebook and share your thoughts about the show with the hashtag #DrugSafetyMatters.

Got a story to share?
We’re always looking for new content and interesting people to interview. If you have a great idea for a show, get in touch!

About UMC
Read more about Uppsala Monitoring Centre and how we work to advance medicines safety.

Fredrik Brounéus:

Welcome to Drug Safety Matters, a podcast by Uppsala Monitoring Center, where we explore current issues in pharmacovigilance and patient safety. I'm Fredrik Brouneus and today I'm looking forward to learning more about pregnancy-related pharmacovigilance. There are many reasons why use of medicinal products during pregnancy requires special attention. First and foremost, we want to be sure that the medicine is as safe as possible for both the pregnant person and the unborn child. Unfortunately, the safety profiles of medicines used in pregnancy are often incomplete, which makes it difficult for patients and healthcare professionals to make informed decisions. Here at Uppsala Monitoring Center, our research section has a team that is currently focusing their efforts on pregnancy-related pharmacovigilance.

Fredrik Brounéus:

Today we have Data Scientists Sara Vidlin, Levente Papai and Senior Pharmacovigilance Scientist Lovisa Sandberg from this team in the Drug Safety Matters studio to talk about complexities and challenges of pregnancy-related pharmacovigilance and about developing solutions for tackling those challenges.

Fredrik Brounéus:

Hi and welcome to the studio. So with me around this table, or this round table here in the studio, we have, from left to right: Levente

Levente Pápai:

Hello, welcome.

Fredrik Brounéus:

Lovisa

Lovisa Sandberg:

Hello.

Fredrik Brounéus:

and Sara ...

Sara Vidlin:

Hi.

Fredrik Brounéus:

You are all part of a UMC research initiative with a special focus on pregnancy-related pharmacovigilance, or PV as we say for short. So let's start from the beginning here. What do we mean when we say pregnancy-related PV?

Sara Vidlin:

Well, first of all, pharmacovigilance can be defined as the science and activities relating to the detection, the assessment, understanding and prevention of adverse effects or any other medicine- or vaccine- related problem. So with pregnancy related pharmacovigilance we mean those efforts, but when they are pregnancy- related, so focused on safety in the pregnant persons themselves and their fetuses, or individuals exposed to the medicine and vaccine while still in the womb.

Fredrik Brounéus:

I guess it's also safe to say that medicine safety during pregnancy has had a profound impact on the development of pharmacovigilance as a whole, right?

Levente Pápai:

So, the thalidomide tragedy was actually a main reason for the birth of modern pharmacovigilance. In the late 1950s, thalidomide was used to treat morning sickness during pregnancy, especially during the first trimester. However, by the early 1960s it became apparent that the medicine caused severe birth defects, such as underdeveloped limbs, so fo comelia. Consequently, thalidomide was withdrawn. This tragic event highlighted the need for improved processes in both the approval and safety surveillance of medicines. It also called for global collaboration, as it was clear that sharing experiences was key to identify early signs of safety issues. This led to the formation of the WHO Programme for International Drug Monitoring and the establishment of a common global database for adverse event reports, which bears the name VigiBase. Today, over 150 member countries contribute to this database by sharing adverse event reports to enhance global pharmacovigilance. So, despite the significant influence of the thalidomide tragedy on pharmacovigilance, we are still very much behind when it comes to pregnancy-related pharmacovigilance.

Fredrik Brounéus:

We will get back to why we're still behind, because I know there are multiple reasons for that. But before we go down that path, could you perhaps tell us about some of the reasons why special considerations are needed? Why is this such a special patient group?

Lovisa Sandberg:

Yes. So first off, it is important to recognize that pregnant persons also need medication. For instance, they might have a chronic condition that doesn't pause during pregnancy, such as epilepsy, or they could develop a specific issue to the pregnancy, such as gestational diabetes. And of course, they might catch any acute disease also during the pregnancy and that could be an infection, for example. And we know it is common to use medicines during pregnancy. So, some studies have shown that over 80% of pregnant individuals use at least one medication during their pregnancy, and also we know that polypharmacy, or using several medicines, has become more common.

Lovisa Sandberg:

An important consideration here is that the pregnant body may react differently to medicines and vaccines as compared to the non-pregnant, and this may influence both the effect of the medicine and also the risk of adverse reactions. And this is due to physiological changes in the pregnant body that may influence, for example, the uptake, the distribution and the elimination of a medicine and, for example, the blood volume increases and this may affect the distribution of medicines, and an increased renal blood flow, or blood flow through the kidneys, may increase the rate of elimination. Also, there are changes in enzyme activity that may affect the metabolism, and another example is changed levels of certain proteins which may affect protein-binding medicines. So both the pharmacokinetics and the pharmacodynamics are affected and this could influence both, like I said, the effect of the medicine but also the risk of adverse reactions. But of course, the main consideration here is that there are two lives to consider and that serious outcomes in the prenatally exposed child may have lifelong consequences, and not only for one person but for a whole family.

Levente Pápai:

Yeah, once again, at the same time, it's important to emphasize that medication may be crucial for the pregnant individual, so discontinuing an ongoing treatment or refraining from starting the new medication may even be more harmful than the risk of an adverse reaction, both for the fetus and for the pregnant individual. So pregnant individuals and their health professionals should ideally understand the balance between benefits and risks when deciding whether to take a medication or when choosing between different treatment options. This knowledge is essential for making an informed decision.

Fredrik Brounéus:

So could you perhaps give a clinical example of this kind of benefit-risk balance considerations?

Levente Pápai:

Yes, take epilepsy as an example. There are known risks of major congenital malformations in the child due to transplacental exposure of certain anticonvulsants. On the other hand, there are also serious risks for both the pregnant person and also the baby if experiencing severe seizures during the pregnancy. In this case, treatment may be necessary despite risks. But it is still of utmost importance to understand if the safety profiles between different options, in this case anticonvulsants, differ, to be able to choose the most appropriate medicine.

Fredrik Brounéus:

I noticed that you said that pregnant individuals and their healthcare professionals should ideally know the benefit-risk balance of treatment options. And I mean you say ideally, because this is not always the case, is it?

Lovisa Sandberg:

No, and unfortunately making informed decisions on whether to use a medicine in pregnancy is difficult because the safety profiles about use in pregnancy are often very incomplete. And especially at the time of marketing of a new medicine there's very sparse safety data on use in pregnancy, and the limited information available, if any, is generally based on non-clinical data. And this is because pregnant persons are generally excluded from clinical trials due to potential safety concerns, unless the medication is intended to be used specifically in the pregnant population. And also, those who become pregnant during a clinical trial are often withdrawn from the study, although the child is often followed up for safety data. So there might be some data from these accidental exposures, but this is not enough to fill the knowledge gaps.

Fredrik Brounéus:

And sometimes quite wide knowledge gaps, as I understand it, and in practice the informed decisions are not very informed, then

Levente Pápai:

Yes, that's right.

Levente Pápai:

Generally, at the time of marketing and for a considerable period afterwards, there is an insufficient knowledge about the use of most medicines and vaccines during pregnancy. It makes a challenge to truly make informed decisions. Recently, there has been a growing debate about whether it is more unethical to exclude pregnant individuals from certain clinical trials. But still, post-marketing surveillance, such as spontaneous reporting and pregnancy registries, is crucial to complete the safety profiles relating to use in pregnancy.

Fredrik Brounéus:

Well, we can say that there are many layers to this complexity. In every decision about medical treatment during pregnancy, there are two lives involved: the pregnant person and the unborn child. Which adds a quite complicated dimension to the benefit-risk balance to be considered by both healthcare professionals and patients, which, in turn, is further complicated by the lack of safety information available. So, safety-wise, as you said, due to a lack of data from clinical trials, the knowledge on the risks of any given medicine on pregnancy is behind already from the start, and this makes it, of course, even more important for us to extract every little drop of information about the safety of the medicine once it's out on the market and being used by pregnant persons. Could you tell us something about the challenges here, as seen from the pharmacovigilance assessor's point of view?

Sara Vidlin:

If we now focus on spontaneous reporting systems, the first issue for a PV assessor who wants to investigate pregnancy issues is basically finding the relevant cases in the database. And ironically, even though the spontaneous reporting systems were introduced as a consequence of the thalidomide tragedy, unfortunately they are still not optimized for these types of reports. And even though they may capture them, the challenge is to find them. So if we look at VigiBase, the WHO Global Database of Adverse Event Reports, now, this database contains over 40 million reports as of today and only approximately 1% of those relate to pregnancy. So the burning question is how to find them. It would have been quite easy if each report had a flag just saying "is this a report about pregnancy, yes or no, and you could just filter on that flag. But, however, the challenge is that in the electronic transmission standard; so the global transmission standard for individual case safety reports is E2B, which is the most commonly used format to exchange reports between databases. So in this transmission standard there is not one specific field to indicate pregnancy exposure like "yes, this is a pregnancy exposure or no, this is not a report related to pregnancy.

Sara Vidlin:

Sometimes some databases have a local pregnancy flag indicating that this is a case about pregnancy. But these local flags, if they exist, they are usually lost in transmission of the reports when transmitted to other databases. So, for example, they will be lost when transferred to Eudra Vigilance or to VigiBase. And although it's great that these local pregnancy flags exist; I mean it's great because it means that someone somewhere in the process have tried to make it easier to report and find these really valuable cases; there is also a risk that they might lead to a false security that it is enough just to tick this box. But I would say it is not enough to tick a box. I mean especially if the information later on is lost. But even if it would be preserved, more information will be needed if the report is to be useful down the line, like exposure terms describing a type of exposure or information on the timing of exposure and similar.

Lovisa Sandberg:

So we could also add here that the standard transmission format is quite extensive and allows for many different types of pregnancy-related information to be reported in many different ways, and this is really a good thing. But due to different reporting cultures, reporting guidelines and coding practices across the globe, and also over time, this results in that not all pregnancy reports look the same when collecting them in a common database like VigiBase, and this, together with this non-existence of a dedicated pregnancy field, like Sara mentioned, leads to challenges finding the relevant cases.

Fredrik Brounéus:

This is possibly a naive reflection or question, but how difficult would it be to introduce such a flag that would be transferable between databases, I mean, since we obviously have lots of other information that does travel the distance?

Sara Vidlin:

Purely technically it's not difficult, but I guess this will have to be something considered for future versions of the transmission format. In the current version, however, there is a possibility to extend the format with additional fields, and if this is done in a harmonized way, it could potentially be an option. Meanwhile, but important though, it is very crucial to make sure that a pregnancy flag means the same thing across databases, so the scope of the flag must be very clear. Are paternal exposures included? Are exposures during labor included? Are exposures during breastfeeding included? Does it consider reports with adverse events in the pregnant person or only those with adverse events in fetus? And so on, and so on.

Fredrik Brounéus:

It's not just adding a flag. I can see that, yeah.

Levente Pápai:

One solution that we have explored at UMC is to introduce an algorithm to support the identification of pregnancy cases. So this is because what we mentioned already, because pregnancy related information can be reported in multiple fields indicating that the report involves exposure during pregnancy. This makes it non-trivial to retrieve pregnancy cases. Our approach was a rule-based method, that is, it is a type of algorithm that follows a set of predefined rules or conditions to make decisions or perform actions. These rules are created by experts and are based on logical statements that define specific criteria for inclusion or exclusion.

Levente Pápai:

In the context of identifying pregnancy cases in VigiBase, the rule-based algorithm works by checking several fields for pregnancy-related information and flagging those as pregnancy cases or unknown cases. The process involves two main steps ruling out and ruling in. In the exclusion step, it filters out any reports that are considered to be unlikely to be pregnancy cases, so, for example, patient age. In the inclusion step, the algorithm identifies reports that meet the criteria for potential pregnancy exposure. So, for example, it looks for reports where exposure during pregnancy has been reported as an event, the route of administration is transplacental, or a gestational age is mentioned in the report.

Fredrik Brounéus:

So it's a sorting mechanism of sorts, that first sort out all the cases that cannot be related to pregnancy, like you say, and then searching through those that remain to see if they are related to pregnancy. But how do you go about ensuring that a solution like this is picking up the right reports and, so to say, it's specificity and sensitivity?

Sara Vidlin:

This is a very good question. When it comes to finding relevant pregnancy cases, we've seen that many different approaches have been used in different studies, and the approach chosen depends, of course, on the aim of the study. So, for example, if it is more important to capture all potential cases but then risking also getting a few false positives, so irrelevant cases as well, or if it's more important to be very specific, that the cases retrieved really are indeed relevant, at the cost of then potentially missing quite a few. So in order to choose methods, it's good to know how it performs. So for us, it was very important that the VigiBase pregnancy algorithm that we developed was thoroughly evaluated, and this in order to understand how many of the true pregnancy cases in the database that we can expect to find, and also how many of the cases that the algorithm flags that are actually indeed pregnancy cases.

Sara Vidlin:

In many cases, it's also, of course, impossible to know the real truth. If you have a case with a 30-year-old woman, you can never with 100% certainty say that this is not regarding a pregnant patient. So we classify all cases as pregnancy cases or unknown. But coming back to the performance of the VigiBase pregnancy algorithm, we found that 92% of all reports flagged by the algorithm can be actually expected to be true pregnancy cases. So this means that if you perform a search for pregnancy exposures and then end up with, say, 100 reports, around 92 of these will indeed be related to pregnancy.

Fredrik Brounéus:

Yeah.

Sara Vidlin:

And looking at VigiBase as a whole, the algorithm managed to find 75% of all pregnancy cases when looking at the whole database. But worthy to note is that if we narrow down the analysis to only reports submitted in the current transmission format, E2B, the algorithm actually managed to find 91% of all pregnancy cases and I think it's very important that you have this in mind; the performance and the reasons also why certain cases might be missed when you perform a study.

Fredrik Brounéus:

That does sound like quite impressive results from the algorithm. Were you surprised? Was this ... is this good? Were you happy with the results?

Sara Vidlin:

I would say it's quite good because it's a challenging program.

Fredrik Brounéus:

Yeah, speaking of challenges, did you find any specific challenges or obstacles for the algorithm to do its algorithm work, so to say?

Lovisa Sandberg:

So the main challenge for the algorithm in VigiBase, which is a global database with reports dating back to the 1960s even, is that reports are quite inconsistently reported, so using different reporting formats and different coding practices. But besides that, the challenge is more referred to the quality of the reports and mostly relate to the amount or clarity of the information given. So, for example, sometimes the pregnancy- related information is given only in free text, for example in the case narrative, and while a narrative description is crucial to assess a case report, if there's no structured information indicating pregnancy in the report, the algorithm, as of today, does not retrieve it. And also, some reports lack specific pregnancy information and pregnancy exposure can only be inferred from several pieces of information in the report. And these reports are difficult even for a human to interpret.

Lovisa Sandberg:

And we also have examples of ambiguous reports where the pregnancy information has been miscoded as such or where it's unclear if the pregnancy is concurrent or historical; and also reports where it's unclear who the report is about, who the report subject is. So if it's the pregnant person or the child. For instance, reports could have a patient age referring to the pregnant person while the events refer to the child, or there could be a mix of events for the pregnant person and the child in the same report, and these reports are very difficult to handle and interpret. But at the same time we see these as challenges. We also see the opportunities to develop the algorithm further, and one idea could be to explore the potential to use more advanced methods, like large language models, to incorporate also free text information in the algorithm. So that could be something for the future.

Fredrik Brounéus:

Before we got into talking about the algorithm, we were talking about challenges for PV assessors when investigating possible pregnancy issues, and well, finding the cases obviously is the first challenge that you have to overcome, but it doesn't end there, does it?

Levente Pápai:

No, it doesn't. Once the reports are retrieved, the next challenge is assessing them. So, as we have mentioned, according to our findings, these reports often contain too little information or the information is unclear or misleading. Assessing causality in pregnancy reports, especially regarding outcomes in the baby, relies on different information than usual causality assessments. Key elements like time to onset, de-challenge and re-challenge cannot typically be applied to prenatal exposure cases, so, like mentioned, for example, the timing of exposure in relation to the gestational age is crucial, and it's important to understand who the report subject is – the pregnant person or the child. Additionally, it's essential to know whether the exposure was maternal or paternal, along with any comorbidities and medical history. Yet another challenge is how to spot the relevant cases for analysis, such as in signal detection. At UMC, we have started exploring statistical methods to highlight issues that are reported more frequently than expected within the pregnancy population.

Fredrik Brounéus:

Now, so far we have been talking about VigiBase, the WHO Global Database of Adverse Event Reports, as you said, but there are other databases as well, and do we see similar problems and solutions – for example algorithms – with regards to pregnancy- related cases in these databases?

Sara Vidlin:

Yeah, so for the same reason a similar algorithm has, for example, been developed for EudraVigilance, the European database, and another one has been developed for FAERS, the US FDA adverse event reporting system, and actually UMC recently participated in a workshop held at the University of Oslo involving representatives of all these three algorithms and we compared the differences and similarities between them. And although they have the same intention, they have somewhat different scopes, for example, that one of them aims to capture paternal exposures and avoids normal and unintended pregnancies, while the other two do not. But this is still somewhat ongoing work as we speak and we are working on comparing and characterizing these three different algorithms and preparing to share the results in a scientific article.

Fredrik Brounéus:

When you have different algorithms such as these, is it possible to combine them in one way or another to improve their functionality?

Lovisa Sandberg:

I would say these three algorithms that we compared are quite similar in their method. So, yes, we can learn from each other and this is what we also did during this workshop that some of the aspects used in one of the other algorithms could be used to improve also the UMC algorithm, and vice versa. So, yes, that could in one way be done.

Fredrik Brounéus:

So they would sort of cross-fertilize each other, so to say yeah.

Fredrik Brounéus:

So technical innovations seem to be quite important to make the most of the data available. Then, if we consider the PV process a bit more upstream, so to say; let's say I'm a reporter, for instance, a doctor with a pregnant patient who has experienced a suspected adverse drug reaction, what can I, as the doctor, do to help assessors overcome some of the challenges that you described? What can I do when I'm reporting to make it easier for the assessor to find this case?

Lovisa Sandberg:

I'm thinking, considering that under-reporting is a major limitation of spontaneous reporting systems, with only a fraction of adverse reactions being reported, the most important is that you do report this adverse event, and especially when it comes to pregnancy exposures, as the data is so limited and the knowledge gap is so wide. Of note here is that many countries do allow for direct patient reporting, meaning that also you as a pregnant individual could share your own experiences in a report and thereby contribute to increased knowledge. But then, thinking also of the challenges to find your report in the databases we have talked about, of help would be if you clearly indicate that this report is about a pregnancy exposure and of course, depending on the local system, the means to do this may vary, but one example is, for example, that you add the exposure during pregnancy as an adverse event.

Fredrik Brounéus:

So yeah, because you said that the algorithm doesn't pick up information from the free text field, so I need to find other places to put this information.

Lovisa Sandberg:

If you have the possibility. In some systems you may only be able to report free text and then you just clearly have to indicate it there and then the code during the next step hopefully do that for you. But if you have the possibility to add it in some structured way, that is good also. Then also, if you could have the assessor in mind who will analyze the report in the end. So it's important – we have also talked about that – that you clearly indicate who the report is about. So is it about the pregnant person or the child? Or if both the pregnant person and the child have experienced an adverse event, you should create two separate reports. And finally, to also report other information that is relevant to the pregnancy, and here, of course, we can highlight the timing of exposure in relation to the gestational age, as this is really crucial to understand whether the exposure was, for example, during the organogenesis, so in the first trimester or later, to be able to assess the biological possibility.

Fredrik Brounéus:

That was a lot of information, quite a lot to keep in mind. Are there any recommendations or guidelines that can be followed by reporters?

Sara Vidlin:

Guidelines on how to report pregnancy information do exist, but these may not always be tailored to or accessible to the primary reporter. So I think we need more clear and harmonized guidelines, as well as trainings on how to use the guidelines correctly. There have been some great initiatives in this area in recent years, like the IMI ConcePTION Project. IMI ConcePTION have suggested frameworks for data collection to increase the quality of pregnancy data and also how to assess the quality of the pregnancy information in the case reports.

Levente Pápai:

Also, as briefly mentioned, coders at pharmacovigilance centers have an important role in the creation of reports, making sure that the reported information is adequately represented in the report. For example, they may do the coding of medicines and adverse events to standardized terminologies like WHOdrug and MedDRA from free text entries made by the primary reporter. Here, guidelines such as one from the ICH on the E2B format and MedDRA Points to Consider may be helpful, but also the local guidelines.

Fredrik Brounéus:

We'll make sure to add links to that in the show notes that interested listeners can follow to find out more. It just strikes me that putting pieces of this pregnancy PV puzzle together must be such a difficult task. I mean as any adverse pregnancy outcome that may or may not be related to a specific medicine may not be noticeable until long after the medicine was used. And, for instance, when a child is born with some kind of birth defect, any medicine that may have been caused or contributed to this could have been taken by the parent several months ago without noticing any side effects at the time.

Levente Pápai:

Yes, that is true. In these cases we have to consider recall bias, as it can be difficult to remember all these details of the pregnancy. Another challenge with those reports is the time that has passed since the exposure. It makes it hard to recognize other influential factors or alternative explanations.

Lovisa Sandberg:

Some events could appear also much later in life, so, for example, neurodevelopmental disorders, which could become apparent first when you're an adolescent or even an adult, and then it's even more difficult to remember all the details about the medications used, like 15 years ago, and any other relevant details around your pregnancy. But nevertheless it's important to report also these cases and to then clearly indicate that you suspect the prenatal exposure so that they can be recognized as pregnancy reports.

Fredrik Brounéus:

Obviously, individual case safety reports, or ICSRs, and VigiBase are central to pregnancy-related pharmacovigilance, but what other data sources are being used specifically for pregnancy-related pharmacovigilance?

Lovisa Sandberg:

Yeah, well, just to mention a few examples, we have national population-based pregnancy registries. We have also registries specifically designed to follow up specific medications or conditions such as epilepsy. There are pregnancy cohort studies and electronic healthcare records, and each of these systems has its strengths and limitations, but they may also complement each other. So, for example, spontaneous reports are typically used to generate early hypotheses or signals, while other data sources, like the pregnancy registries, can be used to confirm or refute these hypotheses.

Fredrik Brounéus:

Okay, so we're nearing the end of this interview now and I just have two more questions, and they're both possibly huge, so apologies for that. But the first is: what's next in the pipeline for you this research initiative at UMC with regards to pregnancy-related pharmacovigilance? And the other question is: what do you think that we, the pharmacovigilance community, should focus on to get ahead on pregnancy and medicines safety?

Sara Vidlin:

Well, I can start with the first question. So our next step is to try to explore problems and needs that national pharmacovigilance centers have – within this area, of course – and we hope to gain insights into simply where next to focus our attention so that we can have as much impact as possible with any future scientific research we want to do. So depending on where we see most needs, that is where we will continue development. So our aim is to be able to support PV centers in their very important work of pharmacovigilance related to pregnancy.

Fredrik Brounéus:

And how will you identify the needs?

Sara Vidlin:

We will try to talk to several centers.

Fredrik Brounéus:

Great, thank you.

Lovisa Sandberg:

Yes, and when it comes to your second question, so about the PV community, it's really inspiring to see all the current initiatives and activity in this area and it feels like the awareness of the needs of this vulnerable population is increasing. But we are still behind, as we said in the beginning, and we have a lot to do to continue to improve the spontaneous reporting systems to make them even more useful. For example, one of the main opportunities we see is the need for further harmonization of reporting standards and also awareness of, and adherence to, those standards and also guidelines that already exist.

Fredrik Brounéus:

Again, see the show notes. We'll add links and information where you can read more about standards and guidelines. And thank you very much for coming to the show Levente, Sara, Lovisa,

Lovisa Sandberg:

Thank you.

Sara Vidlin:

It was a pleasure.

Fredrik Brounéus:

As always, I learned a lot, and also thanks to my colleague, Alexandra Coutinho, who is sitting behind us here in the studio, for producing this episode.

Fredrik Brounéus:

Thanks a lot, take car e.

Fredrik Brounéus:

producing this episode. Thanks a lot, take care. And if you'd like to learn more about pregnancy-related pharmacovigilance, we've put together some links for you in the episode show notes. Apart from the Drug Uppsala Safety Matters podcast, uppsalareports. org we have our pharmacovigilance magazine, uppsala Reports. Visit upsalareportsorg to stay up to date with news, research and trends in the field, and don't forget to subscribe to the newsletter for the latest articles. If you have any comments or suggestions for the podcast or the magazine LinkedIn please Blues ky free to reach out on Facebook, linkedin, blue Sky or X. You can also visit our website to learn more about what we do to promote safer use of medicines and vaccines for everyone, everywhere. If you like the podcast, episode; please subscribe to make sure you won't miss an episode and spread the word so Matters, listeners can find Brouneus us too. For Drug

Fredrik Brounéus:

Safety Matters. I'm Fredrik Brunius. Thanks for listening.

People on this episode

Head Shot

Alexandra Coutinho

Host
Head Shot

Federica Santoro

Host
Head Shot

Fredrik Brounéus

Host