Skip to main content

ICMRA Teleconference Minutes COVID-19 Real-World Evidence observational studies

Friday 20 May 2022 14:00 – 15:30 CET

Chairs: Xavier Kurz (European Medicines Agency - EMA) and Kelly Robinson (Health Canada - HC)

1. Welcome and introduction

Xavier Kurz (EMA) and Kelly Robinson (HC) opened this 7th ICMRA meeting on COVID-19 Real-World Evidence and Observational studies on behalf of Emer Cooke, the EMA Executive Director and Chair of ICMRA. The successful collaborations and achievements throughout the COVID-19 pandemic have led regulators from a simple exchange of good intentions, to the establishment of an active and effective platform that allows easy and rapid sharing of data and experience. This resource may be used beyond the scope of the pandemic and the lessons learned may be transferrable to other topic areas.

Kelly Robinson (HC) outlined the meeting’s objectives, which were to illustrate our successful collaboration to date through updates from the three technical sub-working groups, to take stock on lessons learnt over the last two years by various regulatory agencies, and to brainstorm on what the collective future holds beyond the COVID-19 pandemic.

2. Update on the three technical sub-working groups initiated at the 19 May 2020 ICMRA meeting

a. Vaccine surveillance and vigilance (led by MHRA + TGA)

Patrick Batty (MHRA) provided an update on behalf of the ICMRA COVID-19 Vaccine Pharmacovigilance Network (VPN) and shared thoughts on future collaboration. The presentation started with a reminder on the wide composition (23 members) and the group’s purpose which is to share information on vaccines PhV (before and after authorisation) and emerging benefit/risk profiles.

The group meets regularly to discuss standing topics like emerging safety information as well as topics of special interest not limited to specific PhV issues, like the ICMRA/WHO COVID-19 Vaccines Confidence joint Statement for Health Care Professionals (ICMRA website / WHO website) that has gone through several revisions taking account of paediatric use, booster doses, and virus variants.

A survey was circulated among VPN members to seek their feedback on future ways of working. The majority indicated that the VPN group should remain past December 2022 but that changes should be applied even possibly before the end of the year, such as a reduced frequency / ad-hoc basis meetings while continuing e-mail exchange and with consideration of the potential for a widened scope (e.g. therapeutics, methodologies for post-market surveillance).

b. Pregnancy observational research (led by EMA)

Kelly Plueschke and Corinne de Vries informed the working group that the results of the EMA funded CONSIGN study (Covid-19 infectiON and medicineS In preGNancy) led by Utrecht University/Utrecht Medical Centre University in the Netherlands are expected in Q3 2022. CONSIGN is composed of three studies using different types of data sources: electronic health records from primary care in nine European countries, registry data from antenatal clinics worldwide (COVI-PREG network – 198 countries) and hospital data from 12 European countries (INOSS). The synthesis of all these data will help answer 3 objectives, which are to estimate the prevalence of use of COVID-19 treatments in pregnant women with COVID-19, to describe the severity and clinical outcomes of COVID-19 disease in pregnant women, and to assess and compare the rate of relevant pregnancy/maternal and neonatal outcomes by treatment groups.

CONSIGN Preliminary results show that infected pregnant women are less likely to receive treatment compared to non-pregnant women, and when treated, important variations in treatments’ patterns are visible across countries. Antibiotics and antithrombotics remain the most used medicines in pregnant women with COVID-19. Hydroxychloroquine was used extensively in Italy during the first wave but its prevalence quickly decreased from summer 2020. Maternal ICU admissions and pre-term births have increased but no concerns can be raised on stillbirth, neonatal deaths and congenital malformations based on the currently available data (February to August 2020). At the moment, it is difficult to conclude on any association between COVID-19 treatments and adverse outcomes due to the strong confounding by disease combined with low numbers of exposure and of adverse outcomes.

EMA and the CONSIGN consortium is now working with international partners on a global meta-analysis to explore whether differences in medicine use between regions can help address these confounding factors. Two sub-meta-analyses are being conducted based on the CONSIGN protocols and CONSIGN data:

The results expected early 2023 will hopefully help regulators to better understand the impact of COVID-19 treatments in pregnancy/mother and baby, to develop guidance on the appropriateness of treating/vaccinating pregnant women, and to further develop our ability to appraise proposals for medicines risk management in pregnant women beyond the COVID-19 pandemic. At the end of the project, EMA will prepare an evaluation of the lessons learned on hurdles and benefits of a collaboration based on the CONSIGN meta-analysis.

Xavier Kurz (EMA) asked if we know whether hospitalisations were due to COVID and if the treatments received were intended to treat COVID-19. Corinne de Vries (EMA) clarified that whilst INOSS has collected data on 2166 women who were hospitalised with COVID-19 as of August 2020, 935 women were hospitalised due to and treated specifically for COVID. The final results will cover pregnant women admitted to hospital due to COVID-19 in order to limit potential bias caused by maternal outcomes of pregnant women admitted for other reasons than COVID-19. EMA has however asked the CONSIGN consortium if it would be possible to get information on treatments used in these latter groups of women. Kelly Plueschke (EMA) also indicated that the currently available results cover data until August 2020, but that the final results expected in Q3 2022 will cover all 2020, i.e. different COVID waves across countries and possibly different treatments patterns by the end of the same year.

c. Building international cohorts (led by HC)

Melissa Kampman (HC) presented an overview of the various studies performed in the context of the sub-working group composed of HC, U.S. Food and Drug Administration (US FDA) and EMA initially established to share expertise, increase statistical power, and maximize data quality to meet regulatory requirements and address existing knowledge gaps. The group first identified knowledge gaps and opportunities for collaboration, and then determined criteria for selecting common research questions. Each regulator has been leveraging research teams and different data sources within their distinct jurisdiction to work on three main projects: utilisation of steroids and COVID-19 patients, the natural history of coagulopathy, and the validity and reliability of the COVID-19 case definitions.

  • Steroid Use & COVID-19 (project led by EMA)

    • At the beginning of the pandemic, there was conflicting information on the outcomes of steroids in COVID-19 patients. For example, the recovery trial found reduced mortality for dexamethasone in ventilated patients and patients with supplemental oxygen, but observational studies showed mixed clinical outcomes especially in milder cases;
    • The objectives of this project are to investigate the utilisation of steroids in two different settings (hospitalised and ambulatory settings) within 90 days of COVID-19 diagnosis, to evaluate differences in mortality rates post-steroid administration, to evaluate the safety of steroid use and the association between adverse events and treatment related factors;
    • EMA study (contracted to IQVIA) focused on the utilisation of systemic glucocorticoids for the treatment of COVID-19 in nine countries. Results showed that prednisolone, prednisone, and dexamethasone were the most prescribed glucocorticoids in ambulatory setting, while dexamethasone, methylprednisolone, and hydrocortisone were most prescribed in hospital settings. The study was finalised in September 2021 and a final report was published in the EUPAS Register (EUPAS38759);
    • US FDA published the results from Medicare and Sentinel System on corticosteroid use in non-hospitalised COVID-19 patients in JAMA. The CDC subsequently issued a Health Alert Network Health Advisory citing the work and reminding health care providers and the public that: “Systemic corticosteroids are not recommended to treat patients with mild to moderate COVID-19 who do not require supplemental oxygen;
    • In Canada, CNODES is conducting research aligned with EMA's study protocol, with the exception of the secondary objective for case definitions as it is being addressed in a separate query. The statistical analysis has been approved and descriptive analysis is underway in Ontario (once completed, comparative analyses will begin in other provinces).
  • Coagulopathy & COVID-19 (project led by US FDA)

    • Preliminary evidence indicated development of arterial or venous thromboembolic complications in a potentially significant proportion of COVID-19 patients;
    • The objectives of this project are to determine the incidence of arterial and venous thrombotic complications and subsequent risks of death among COVID-19 patients, to evaluate the risk factors for arterial and venous thrombotic events, to determine the risk of arterial and venous thrombotic events (ATE/VTE) between COVID-19 and influenza patients;
    • US FDA created a protocol for their study in the Sentinel System that they shared with EMA and HC. The statistical analysis plan is complete, as are the feasibility assessment and the results. A manuscript is currently under review for publication - LINK ;
    • At the same time, EMA has been conducting research with Erasmus and Oxford University aligned with the first and second aims using the common data model on data from 5 different countries. The study was finalised in October 2021 with a final report published in the EUPAS Register (EUPAS40414). VTE and ATE risks were up to 1% among COVID-19 cases, increase with age, in males, and in those hospitalized, with higher mortality;
    • In Canada, the CNODES team is also conducting research aligned with aims 1, 2, and 3. The results from the first two objectives will be included in a meta-analysis involving seven countries. The protocol for the international meta-analysis has been approved and the data tables are being finalized are being circulated to all international partners;
    • The protocol and statistical plan have been agreed upon for an international meta-analysis. The table shell population, data sharing & analysis are planned by August 2022 and the manuscript by October 2022.
  • COVID-19 Case Definitions (project led by HC)

    • At the beginning of the pandemic, WHO released emergency ICD 10 codes for COVID-19: U07.1 for cases of COVID-19 that were laboratory confirmed, and U07.2 where laboratory confirmation was not available, but COVID-19 was suspected. HC hypothesised differences in the use of these codes over time. As healthcare systems started to use these codes more, and once they were adopted, HC wanted to look into the validity and the reliability of these codes in Canada and to see if a gold standard could be developed in terms of case definition that could also apply to other jurisdictions;
    • The objectives of this project are to develop a ‘gold standard’ case definition that can be either prospectively implemented or retrospectively identified, to determine whether a single gold standard is possible between countries, to investigate the operating characteristics of the case definitions of both confirmed and suspected cases across individual sites;
    • In Canada, CNODES analysis began in November 2021 and results received in February 2022 are currently being summarised in a final report under review. A second DSEN research team, the Canadian Network for Advanced Interdisciplinary Methods (CAN-AIM), conducted an independent study to provide a different perspective for which the final report is expected in June 2022. Discussions with both the CNODES in the CAN-AIM team have taken place about a high degree of heterogeneity amongst databases in the sense that definitions were not adopted at the same intervals or in the same way across provinces and settings;
    • EMA study (contracted to IQVIA): The definition based on a diagnosis of COVID-19 showed the best sensitivity whilst the specificity was best for the diagnosis of COVID-19. There was a high heterogeneity among databases (9) and not all definitions were present in all databases. The study finalised in September 2021 and final report published in the EUPAS Register (EUPAS38759);
    • US FDA: the results of the Study in the Sentinel System to validate diagnostic codes to identify hospitalised US COVID-19 patients support the use of code U07.1 (Diagnosis code U07.2 is not available for use in the US);
    • The next update meeting of this group will take place in early September 2022. This will include a discussion on joint publications. We have built excellent collaborations and networks amongst the different regulators and their research teams and have leveraged the different datasets. As such, we would be interested in continued collaboration both in the COVID-19 space and beyond.

3. Sharing experiences of work on observational studies, and discussion on the future and path forward for the working group

Prior to the meeting, ICMRA members were invited to share their experience on observational studies related to COVID-19, as well as their views on opportunities for international collaboration. Two regulatory agencies provided responses in writing.

  • Kelly Robinson (HC) reported on Swissmedic’s feedback, indicating that an internal working group on RWE has been established and a position paper is under way to include the use of RWE in the Swiss legislation. According to Swissmedic, there are opportunities to leverage COVID-19 work or to collaborate moving forward, for example on finding regulatory consensus on the expected quality of (GCP-like) RWD data to avoid (diverging) decision making based on inconclusive scientific evidence. The scope of the working group should therefore be extended to develop recommendations on the use of RWD/RDE, including on advanced therapies or personalized / individualized medicines.
  • Aloka Chakravarty (US FDA) presented on the COVID-19 Evidence Accelerator that was launched in May 2020 and aimed at conducting parallel distributed analyses by different research organisations using a common analytic plan. The project is based on three research questions: 1) Natural History/HCQ Use (complete - LINK), Remdesivir in hospitalised patients (manuscript in draft) and natural history of coagulopathy (manuscript in preparation). Work is also ongoing on diagnostics, including on the use of real world test performance of serology for recent infections (submitted for publication). The unprecedented nature of the pandemic brought everyone together to answer questions on the same topics in a timely fashion. We may need a slightly different approach for research questions that do not relate to an urgent public health issue like a pandemic.
  • Xavier Kurz (EMA) re-iterated the success of the coagulopathy studies and meta-analysis performed by EMA, US FDA and HC. When the thromboembolic and thrombocytopenia syndrome (TTS) emerged in 2021 after authorisation of COVID-19 vaccines, background incidence rates on coagulopathy and related complications were needed quickly to allow the conduct of observed to expected (O/E) analyses. It was fortunate that regulatory assessment of such major adverse events could benefit from evidence generated from ongoing studies. This emphasises that state of readiness is crucial, and that we should build on networks established during COVID-19 pandemic. The experience gained has shown that we are able to effectively and quickly create experts groups on specific topics of interest.
  • Melissa Kampman (HC) highlighted that connection with researchers and academia groups is very important, and that any existing gaps should be closed to ensure readiness to rapidly generate suitable evidence in case of a new pandemic or other public health priority.
  • Daniel Lopez indicated that the challenge beyond COVID-19 will be to access real-world data on medicines intended for orphan diseases which might be less straight forward than vaccines in a pandemic situation. Xavier Kurz (EMA) responded that this experience has shown that the collaboration is possible, including the establishment of working groups with many participants in a brief period of time, which can be repeated as needed.

4. ICMRA Workshop on Real World Evidence (29/30 June)

Kelly Plueschke (EMA) updated ICMRA members on the upcoming face-to-face ICMRA Workshop on Real World Evidence scheduled on 29 and 30 June 2022 that will be chaired by EMA, FDA, and HC (with possibility for virtual attendance). The objectives will be to share experiences on the accomplishments and challenges of RWE studies, and build on the successes of our COVID-19 collaborations to identify opportunities for future regulatory collaboration. It is hoped that at the end of these two days, an ICMRA statement on international coordination of activities related to RWE will be issued.

5. Summary and next steps - Xavier Kurz (EMA), Kelly Robinson (HC)

The Chairs concluded the meeting by praising the enormous amount of work conducted by regulatory partners and the three technical working groups to fight back this pandemic.
The VPN presentation has shown that a continuous adaptation of the regulatory systems is key to answer emerging needs and challenges. The pregnancy study across the continents has led not only to the collection of ambulatory data, but also to hospital data on pregnant women. This is the first time where such an exhaustive collaborative collection of data on this special population has been achieved. The international cohorts group has shown that performing studies based on the same protocol with some adaptations to the local databases allows us to conduct extremely strong studies in relatively short time periods. Access to RWE is possible, and as demonstrated by the variability of the results we have shown, it is certainly possible to have international collaborations to access RWD with the same protocols. The discussions also highlighted that a state of readiness is essential, and we may say that we are ready to start further collaborations. New working groups may have to be established with tailored expertise on new topics. This experience has made ICMRA evolve and improve in terms of partnerships that can be created and/or continued through the current groups.

As there is support from members to keep this ICMRA working group beyond COVID-19, the Chairs proposed to develop formal terms of reference that would help us move forward outside of the COVID-19 space. The three technical sub-working groups will take the lead in drafting a clear but flexible mandate to enable adaption to any future emerging situations. ICMRA working group members will be consulted in writing prior to the next meeting. The outcome of the ICMRA workshop on RWE scheduled on 29/30 June 2022 should also help us draw a roadmap for our future collaborative activities.


Invited participants:

  1. Therapeutic Goods Administration (TGA), Australia;
  2. National Administration of Medicines, Food and Medical Technology (ANMAT) Argentina;
  3. National Health Surveillance Agency (ANVISA), Brazil;
  4. Health Products and Food Branch, Health Canada (HPFB-HC), Canada;
  5. National Medical Products Administration (NMPA), China; 
  6. Center for the State Control of Medicines, Equipment and Medical Devices (CECMED) Cuba;
  7. European Medicines Agency (EMA), European Union;
  8. European Commission - Directorate General for Health and Food Safety (DG - SANTE), European Union; 
  9. National Agency for Medicines and Health Products Safety (ANSM), France; 
  10. Paul-Ehrlich-Institute (PEI), Germany; 
  11. Health Product Regulatory Authority (HPRA), Ireland; 
  12. Italian Medicines Agency (AIFA), Italy;
  13. Ministry of Health, Labour and Welfare (MHLW) and Pharmaceuticals and Medical Devices Agency (PMDA), Japan; 
  14. Ministry of Food and Drug Safety (MFDS), Korea; 
  15. Federal Commission for the Protection against Sanitary Risks (COFEPRIS), Mexico; 
  16. Medicines Evaluation Board (MEB), Netherlands; 
  17. Medsafe, Clinical Leadership, Protection & Regulation, Ministry of Health, New Zealand; 
  18. National Agency for Food Drug Administration and Control (NAFDAC), Nigeria; 
  19. Health Sciences Authority (HSA), Singapore; 
  20. South African Health products Regulatory Agency (SAHPRA), South Africa; 
  21. Medical Products Agency, Sweden; 
  22. Swissmedic, Switzerland; 
  23. Medicines and Healthcare Products Regulatory Agency (MHRA), United Kingdom; 
  24. Food and Drug Administration (FDA), United States;
  25. World Health Organization (WHO);
  26. Austrian Agency for Health and Food Safety Ltd. (AGES), Austria;
  27. Danish Medicines Agency, Denmark;
  28. Office of Medical Technology, Health Information and Research (MTHIR), Israel;
  29. Office of Registration of Medicinal Products, Medical Devices and Biocidal Products (URPLWMiPB), Poland;
  30. Spanish Agency of Medicines and Medical Devices (AEMPS), Spain.