Non-interventional pharmacovigilance studies - an analysis of real-world data sources in the HMA-EMA catalogs

Non-interventional pharmacovigilance studies - an analysis of real-world data sources in the HMA-EMA catalogs Ines Hababa-Ivanova1, Violeta Getova-Kolarova1, Antonio Ivanov1, Ilko Getov1 1 Department “Organization and Economics of pharmacy”, Faculty of pharmacy, Medical University-Sofia, 2 Dunav str., 1000 Sofia, Bulgaria Corresponding author: Violeta Getova-Kolarova ([email protected]) Received 2 September 2025♦ Accepted 22 September 2025♦ Published 27 October 2025 Citation: Hababa-Ivanova I, Getova-Kolarova V, Ivanov A, Getov I (2025) Non-interventional pharmacovigilance studies - an analysis of real-world data sources in the HMA-EMA catalogs. Pharmacia 72: 1–6. https://doi.org/10.3897/pharmacia.72.e170867 Abstract Real-world data catalogs established by the Heads of Medicines Agencies and the European Medicines Agency (HMA–EMA RWD catalogues) represent a centralized digital approach for broad access to reliable data to ensure patient safety. In this study, RWD catalog analysis was conducted in order to identify best practices and challenges, focusing on finalized non-interventional studies (NISs) with a scope specified as ‘safety’. The results (n = 240) show a predominance of studies required by a pharmaceutical regulator (56.6%) as well as the importance of NIS in pharmacovigilance via inclusion in risk management plans (RMPs) (52%) and broad use of NIS in risk minimization measure (RMM) evaluation. However, the lack of a unified regulatory framework, insufficient published data, and the recognized need for digital tool implementation create the necessity for improvement within the European Union (EU) pharmaceutical legislation. Keywords digitalization, EMA, non-interventional studies, pharmacovigilance, RWD Introduction According to Regulation 536/2014 of the European Commission, a “non-interventional study” means a clinical study that is qualitatively different from a pre-authorization clinical trial. In contrast, for the purposes of its real-world evidence (RWE) program, the US Food and Drug Administration (FDA) uses the term “observational study,” defining it as a clinical study with a design different from that of a clinical trial (EUR-Lex 2014; FDA 2018). NISs differ from clinical trials in several parameters, such as the common absence of randomization and the possibility for retrospective conduct. A difference between the two types of research is also found in one of the most important and rigorously monitored aspects—ethics in their conduct. Randomized, placebo-controlled clinical trials (RCTs) are considered the gold standard for data collection due to targeted allocation to groups (including placebo therapy). However, RCTs are not always applicable due to ethical reasons, such as: • Diseases where treatment can drastically lower mortality; • The capacity to collect sufficient data from observational studies using historical cohorts; • The need for an extremely large number of patients to obtain definitive results, or in the case of rare diseases. A recent example highlighting the significance of real-world data and digitalization in this area is the introduction, on 15 February 2024, of the HMA–EMA RWD Copyright Hababa-Ivanova I et al. This is an open access article distributed under the terms of the Creative Commons Attribution License (CC-BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Pharmacia 72: 1–6 DOI 10.3897/pharmacia.72.e170867 Research Note Hababa-Ivanova I et al.: Non-interventional studies – an analysis of RWD in the HMA-EMA catalogues2 catalogues (EMA 2024). These represent a repository of metadata collected from RWD sources and RWD studies, aimed at helping regulatory authorities, pharmaceutical companies, and researchers identify and utilize such data when examining medicine use, safety, and effectiveness (EMA 2025). Together, these catalogs provide an enhanced and more effective service for data-driven research in healthcare (Fig. 1). While the establishment of registries such as the HMA– EMA catalog forms the basis for research in healthcare, actions are needed in terms of analysis of the available data and identification of best practices and challenges. Hence, efforts can be directed towards strengthening the centralized approach and broad access to reliable data on patient safety. The aim of our study was to assess the potential and significance of non-interventional studies in generating real-world data and their impact on pharmacovigilance. Additionally, possibilities for regulation, digital tool application, and the participation of vulnerable patient groups in NIS were discussed. Materials and methods Research was conducted in the HMA–EMA catalogs database in the section focused on real-world data studies—RWD catalogs, on the following website https:// catalogues.ema.europa.eu/search?f%5B0%5D=content_ type%3Adarwin_study. The specific database was selected as it extends and succeeds the European Network of Centres for Pharmacoepidemiology and Pharmacovigilance (ENCePP) Resource Database and the European post-authorization study (EU PAS) Register as the main information storage system for post-marketing studies in the EU. To limit the total number of results, a set of inclusion criteria, which are available as filters on the RWD catalog website, was used (Table 1). This filtering process excluded any studies that did not include all three core criteria. Upon population of records matching these criteria, they were extracted into an Excel spreadsheet, and all available metadata as of the study date were included for manual processing. Certain main parameters of the NIS were identified and subjected to qualitative and quantitative analysis. For comparison with the full set of RWD studies, a report including all registered records (clinical trials, non-interventional studies, and others) was extracted from the database. Results and discussion The total number of registered records in the RWD catalog as of 27 December 2024 amounts to 2970, 2867 of which are NIS. For comparison, a similar database—clinicaltrials.gov—contained 117,219 observational studies. Applying the set criteria in the RWD catalog resulted in a total of 240 (8%) records, all of which were included in the subsequent analysis. Table 1. Search filters used in the RWD catalog website. Filter Selected value Study type Non-interventional study Scope of the study Safety study (incl. comparative) Status Finalised Figure 1. Advantages of the HMA-EMA RWD Catalogs (EMA 2025). Enhance the discoverability of suitable data sources to support realworld evidence generation for regulatory purposes (e.g., identification of RWD data sources suitable for investigating a specific research question); Improve interoperability between studies and data sources; Enable easy access to study protocol and study report for assessing data source suitability; Improve transparency. Pharmacia 72: 1–6 3 A major characteristic of NIS was the study topic, as shown in Fig. 2. The review demonstrated a clear predominance of studies investigating a disease or condition alongside the effect of a medicinal product for human use. The top three study topics in the current analysis also lead across the complete raw RWD catalog. Topics such as medical devices and medical procedures were not found in the analyzed subgroup of NIS. The analysis showed a 61% cohort design among all results, 3% included case–control elements, and both cluster and cross-sectional designs accounted for 1% each. This confirms the many advantages of cohort design in NIS, which include the study of multiple outcomes related to multiple exposures, investigation of rare exposures, demonstration of causality, and more. The dominance of cohort design can also be explained by the frequent use of cross-sectional and case–control designs as supplementary tools to a cohort framework to enable the collection of more effective data (Thiese 2014; Wang and Kattan 2020). Another primary characteristic complementing NIS design is the method of data collection. A review of the database showed 102 (42.5%) results for secondary use of data, 83 (34.6%) for primary data collection, and 38 (15.8%) for a combination of both. Carlson and Morrison (2009) highlighted several benefits of secondary data use, including relatively faster and more cost-effective data gathering, larger sample sizes, the ability to study wide geographical areas, and assessment of trends regionally, nationally, and internationally. On the downside, disadvantages can involve missing essential variables and difficulty in identifying the means of database formation. Additionally, according to Jungkunz et al. (2021), risks for confidentiality and unauthorized patient identification are noteworthy. For primary data collection, the main risk is bias and validity of results. Problems with forming a representative sample and difficulties in follow-up (loss to follow-up) are significant, especially for pharmacovigilance (Schneider et al. 2023). Notably, 16 (6.7%) NIS were registered with the “Not applicable” data collection method, and 1 (0.4%) did not collect information at the individual level, which can reduce result reliability. Another significant aspect potentially influencing transparency and results is study funding (Table 2). Data show a clear prevalence of NIS sponsored by pharmaceutical companies or the private sector. Such studies are important tools for collecting data for negotiations with regional and national health insurance institutions. In some therapeutic areas (e.g., rare diseases), clinical trials conducted under controlled conditions and patient registries may not be sufficiently reliable for clear evaluation of the economic effect of a medicinal product. NIS can help clarify differences in prescription practices among countries, as well as budget impact, and aid companies in proper valuation and reimbursement decisions, thus improving patients’ access to medicines (Degun et al. 2014). Additionally, the results support the observed need for more involvement of academic communities, nonprofit organizations, and other non-business structures in conducting such research. Pereira et al. (2020) revealed that by informing scientific society members of RWD findings, their organizations can contribute through ensuring independence from industry and obtaining significant clinical results, as well as mediating between the private sector and practitioners. Notably, 3 (1%) NIS declared adherence to the ENCePP Code of Conduct, while 8 (3%) possessed the ENCePP Seal. The ENCePP Code was developed to ensure independence and transparency of the scientific project, thus increasing the confidence of the general public, researchers, and regulators (ENCePP 2025). The ENCePP Seal indicates a study is conducted under the Code, established international guidelines, and the standards described in the “Guide On Methodological Standards In Pharmacoepidemiology”. A low number of ENCePP Seals was also noted previously (a total of 84 since the launch of the ENCePP Seal in 2010), leading to the discontinuation of the use of the Seal from Figure 2. Non-interventional studies by study topic. 1 2 12 17 48 160 020406080100 120140 160180 Herbal medicinal product Other Not applicable Disease/Condition Medicinal product for human use Disease/Condition and Medicinal product for human use Table 2. Non-interventional studies by source of funding. Source of funding Number of results Pharmaceutical company and other private sector 219 Others 11 Pharmaceutical company and other private sector, Others 4 Not applicable 2 ЕМА 1 EU Institutional program 2 No external funding 1 Hababa-Ivanova I et al.: Non-interventional studies – an analysis of RWD in the HMA-EMA catalogues4 January 2025 (ENCePP 2025a). The low level of both ENCePP Seal and Code supports the widely held view in the scientific literature that NIS suffer from a lack of unified international regulation and standards in planning, conducting, and reporting. Acha et al. (2023) noted that randomization is not the only suitable method to reduce bias. There is also increasing consensus regarding the role of methodological standards and good procedural practices for NIS comparing the effectiveness and safety of treatments. Multiple methods are capable of reducing bias such that these studies may support valid causal inferences, for example: • Develop a common set of Good Practice Principles for NIS, building on existing related guidelines in order to establish comprehensive guidance to support the effective conduct of NIS and increase confidence in using these studies for healthcare decision-making, • Ensure that NIS that are used to assess treatment safety and effectiveness (hypothesis evaluating treatment effect—HETE studies) adhere to these Good Science Principles for the benefit of patients and public health, and • Commit to building robust quality data that can enable meaningful NIS. Regarding public data availability, most studies lacked published results, a data analysis plan, or a protocol (72%, 99%, and 80%, respectively; Fig. 3). Leducq et al. (2024), in their scoping review of published papers in leading general medicine journals, also noted that only 12% (24 out of 200) of the observational studies had protocols accessible to the public. Prospective registration and standardization are necessary for guaranteeing transparency. Regarding country scope, 142 (59%) studies used only one country, while 98 (41%) were multinational projects, of which 81 included more than 3 countries. These data emphasize the importance and growth of multinational research. Concerning regulatory requirements, 136 (56.6%) of NIS were required by a pharmaceutical regulatory authority. Categorization according to inclusion in a risk management plan is shown in Table 3. The high proportion of regulator-imposed NIS further highlights their importance in continued pharmacovigilance monitoring. Vreman et al. (2022) studied the use of regulator-imposed post-approval studies in health technology assessments for conditionally approved drugs, showing that when available, such data were used in 88% of initial and 100% of reassessment procedures (Vreman et al. 2022). Our analysis also demonstrated that the scope of studies most frequently included risk minimization measures (RMMs) effectiveness assessment. This confirms the practical use of NIS in RMM evaluation, as shown in the report by the Council for International Organizations of Medical Sciences (CIOMS) Working Group IX on practical approaches to risk minimization for medicinal products. Observational projects can support many RMM evaluation strategies, such as “monitoring,” “trend analysis,” “surveillance,” and more. Moreover, there is growing patient interest in participation in such programs; a survey among 143 participants found greatest willingness for long-term NIS evaluating risk-minimizing strategies (CIOMS 2014). It is also noteworthy that pregnant patients were present in 45 (18.7%) of the analyzed NIS. Due to the specific nature of this group, it is considered vulnerable and is generally significantly underrepresented in clinical trials. In addition, there is a lack of clear understanding of risks Table 3. Non-interventional studies required by a risk management plan (RMP). Category Number of results Category 1 - Imposed mandatory additional pharmacovigilance activities that are conditions of the marketing authorization. 15 Category 2 - Imposed mandatory additional pharmacovigilance activities that are specific obligations in the context of a conditional marketing authorization or a marketing authorization under exceptional circumstances. 9 Category 3 - Required additional pharmacovigilance activities. 82 Non-EU RMP 19 Not applicable 115 Figure 3. Non-interventional studies by data publication. 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Results Publications Data analysis plan StudyreportProtocol Published Unpublished Pharmacia 72: 1–6 5 and appropriate use of new drugs in pregnancy. The WHO confirms that pregnant patients are typically included only in trials of products expected to see widespread use in pregnancy post-approval (WHO 2025). Conducting studies with large numbers of participants and over long periods is crucial for detecting rare adverse events and assessing safety. WHO data show that only in the < 100 participant category do clinical trials dominate over NIS (the ones subject to our analysis); in all other cases, NIS involve more participants (Fig. 4). Chan et al. (2015) explained that along with all the benefits of RCTs, they are less informative for rare or long-term events. Observational studies examine rare events among much larger patient groups in real-world settings and with longer follow-up. For example, the risk of sudden death in children and adolescents treated for attention deficit hyperactivity disorder (ADHD) is estimated at 0.6–6/100,000 per year—a rate not detectable through conventional clinical trials. Thus, for this context, observational research using extensive population databases is more useful and applicable. For the analysis of the extent of digital tool adoption, we selected the number of NIS using the keywords “electronic” and “digital.” A total of 7.5% (18 out of 240) of the studies included digital elements such as the use of electronic health records, electronic informed consent forms, electronic clinical report forms (eCRFs), electronic health insurance data, and digital platforms (apps, web portals, electronic questionnaires). The results show substantial opportunities for expanding digital approaches in such research. Study limitations Despite the successful extraction of significant data from the HMA–EMA RWD catalogues and the resulting conclusions, the world’s largest clinical trial registry, clinicaltrials.gov, was not analyzed in our research. The filters applied on the webpage limited the study scope notably to just 8% of all available studies (240 of 2970 records). For analyzing pregnancy and digital tool adoption, non-standardized entry required broad keyword searches (“pregnant,” “pregnancy,” “pregnancies,” “electronic,” “digital”), raising the risk of missed studies due to lack of structured fields. Conclusion Our analysis shows that NIS have the potential to generate large amounts of credible real-world data, which are challenging to collect in standard clinical trials. The coverage of vulnerable populations, such as pregnant women, is a noteworthy example, as their underrepresentation in clinical trials can be addressed via NIS. The establishment of the RWD catalog as a centralized repository is a major step forward in digitalization. The significant number of studies required by regulatory authorities, as well as those evaluating the effectiveness of risk minimization measures, shows the importance of NIS as an instrument for enhanced data-driven healthcare research. In comparison to clinicaltrials.gov, however, there is a clear need for increased visibility and knowledge of the RWD catalog. There is also a lack of clear regulatory framework and harmonization, which increases the risk of bias. Improving awareness of the catalog and its user guide can enhance both the quality and quantity of generated data. An increase in the availability of registered studies as well as published documentation can potentially lead to broad access to reliable data on patient safety for regulatory authorities, pharmaceutical companies, and researchers. Additionally, development of a well-defined regulatory framework and mandatory requirements within EU legislation is also of critical significance. Additional information Conflict of interest The authors have declared that no competing interests exist. Ethical statements The authors declared that no clinical trials were used in the present study. Figure 4. Number of participants by type of study – clinical trials (WHO 2024) and non-interventional studies. 0.00 10.00 20.00 30.00 40.00 50.00 60.00 70.00 Non-interventional studiesClinical trials Hababa-Ivanova I et al.: Non-interventional studies – an analysis of RWD in the HMA-EMA catalogues6 The authors declared that no experiments on humans or human tissues were performed for the present study. The authors declared that no informed consent was obtained from the humans, donors or donors’ representatives participating in the study. The authors declared that no experiments on animals were performed for the present study. The authors declared that no commercially available immortalised human and animal cell lines were used in the present study. Use of AI No use of AI was reported. Funding This study was financed by the European Union–NextGenerationEU, through the National Recovery and Resilience Plan of the Republic of Bulgaria, project № BG-RRP-2.004-0004-C01. Author contributions Conceptualization: IG, VG-K; Writing—original draft preparation: IH-I, AI, VG-K; Writing—review and editing: VGK, IG; Visualization: IH-I, AI; Supervision: IG; Funding acquisition: IG, VG-K; Methodology: IH-I, AI, VG-K. All authors have read and agreed to the published version of the manuscript. Author ORCIDs Ines Hababa-Ivanova https://orcid.org/0009-0002-8282-1875 Violeta Getova-Kolarova https://orcid.org/0000-0002-7103-3892 Antonio Ivanov https://orcid.org/0009-0003-5691-5410 Ilko Getov https://orcid.org/0000-0001-6513-9327 Data availability All of the data that support the findings of this study are available in the main text. References Acha V, Barefoot B, Juarez Garcia A, Lehner V, Monno R, Sandler S, Spooner A, Verpillat P (2023) Principles for Good Practice in the Conduct of Non-interventional Studies: The View of Industry Researchers. Ther Innov Regul Sci 57(6): 1199–208. https://doi. org/10.1007/s43441-023-00544-y Carlson MDA, Morrison RS (2009) Study design, precision, and validity in observational studies. J Palliat Med 12(1): 77–82. https://doi. org/10.1089/jpm.2008.9690 CIOMS (2014) Practical approaches to risk minimisation for medicinal products: Report of CIOMS Working Group IX. https://cioms.ch/ sd7fdh93gewd882ds/jkdf79ds7dl092dq-purple-book/CIOMS_IX_ Risk_minimisation_SECURED_20140811v8.pdf Chan EW, Liu KQL, Chui CSL, Sing CW, Wong LYL, Wong ICK (2015) Adverse drug reactions – examples of detection of rare events using databases. Br J Clin Pharmacol 80(4): 855–861. https://doi. org/10.1111/bcp.12474 Degun R, Patel D, Marinsek N, Chalmers M (2014) Value of rare disease non-interventional studies to support payer & clinical decision making. Value Health 17(3): A206. https://www.valueinhealthjournal. com/article/S1098-3015(14)01257-1/fulltext FDA (2018) Real-World Evidence Program. U.S. Food & Drug Administration. https://www.fda.gov/media/120060/download?attachment EMA (2024) HMA-EMA Catalogues of real-world data sources and studies. https://www.ema.europa.eu/en/news/launch-new-hmaema-catalogues-real-world-data-sources-studies EMA (2025) Homepage. Launch of new HMA-EMA catalogues of real-world data sources and studies. https://catalogues.ema.europa.eu/ EUR-Lex (2014) Regulation 536/2014 of the European Parliament and of the Council. https://eur-lex.europa.eu/eli/reg/2014/536/oj/eng ENCePP (2025) Code of Conduct. European Union. https://encepp.europa.eu/encepp-toolkit/code-conduct_en ENCePP (2025a) Discontinuation of the ENCePP Seal. European Union. https://encepp.europa.eu/newsroom/news/discontinuation-encepp-seal-2025-02-18_en Jungkunz M, Köngeter A, Mehlis K, Winkler EC, Schickhardt C (2021) Secondary use of clinical data in data-gathering, non-interventional research or learning activities: definition, types, and a framework for risk assessment. J Med Internet Res 23(6):e26631. https://doi. org/10.2196/26631 Leducq S, Zaki F, Hollestein LM, Apfelbacher C, Ponna NP, Mazmudar R, Grana S (2024) The majority of observational studies in leading peer-reviewed medicine journals are not registered and do not have a publicly accessible protocol: a scoping review. J Clin Epidemiol 170: 111341. https://doi.org/10.1016/j.jclinepi.2024.111341 Pereira PL, Helmberger T, Taieb J, Vilgrain V, de Baere T (2020) Scientific societies should start sponsoring observational studies. Cardiovasc Intervent Radiol 43(12): 1865–1867. https://doi.org/10.1007/s00270020-02649-5 Schneider A, Wagenknecht A, Sydow H, Riedlinger D, Holzinger F, Figura A, Deutschbein J, Reinhold T, Pigorsch M, Stasun U, Schenk L, Möckel M (2023) Primary and secondary data in emergency medicine health services research – a comparative analysis in a regional research network on multimorbid patients. BMC Med Res Methodol 23(1): 34. https://doi.org/10.1186/s12874-023-01855-2 Thiese MS (2014) Observational and interventional study design types: an overview. Biochem Med (Zagreb) 24(2): 199–210. https://doi. org/10.11613/BM.2014.022 Vreman RA, Bloem LT, Van Oirschot S, Hoekman J, Van Der Elst ME, Leufkens HG, Klungel OH, Goettsch WG, Mantel-Teeuwisse AK (2022) The role of regulator-imposed post-approval studies in health technology assessments for conditionally approved drugs. Int J Health Policy Manag 11(5): 642–650. https://doi.org/10.34172/ijhpm.2020.198 Wang X, Kattan MW (2020) Cohort studies: design, analysis, and reporting. Chest 158(1): S72–S78. https://doi.org/10.1016/j. chest.2020.03.014 WHO (2024) Number of clinical trials by year, location, disease, phase, age and sex of trial participants. Global Observatory on Health R&D. https://www.who.int/observatories/global-observatory-on-health-research-and-development/monitoring/number-of-trial-registrations-by-year-location-disease-and-phase-of-development WHO (2025) Clinical trials and observational studies. World Health Organization. https://www.who.int/tools/antiretrovirals-in-pregnancy-research-toolkit/clinical-trials-and-observational-studies