WP2 - Task 1: Pilot report - Finland | UNITED4Surveillance

UNITED4Surveillance Grant Agreement Number 101102070 WP2 Outbreak detection – Task 1: Improving Laboratory Based-Reporting Milestone 36: Pilot report Finland Integration of microbial characteristics and molecular level information into the Finnish infectious diseases surveillance system using Shiga Toxin-producing E. coli as a model pathogen Disclaimer Copyright message This document was produced under the terms and conditions of Grant Agreement No. 101102070 for the European Commission. Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union or HaDEA. Neither the European Union nor the granting authority can be held responsible for them. © UNITED4Surveillance Consortium, 2025 This deliverable contains original unpublished work except where clearly indicated otherwise. Acknowledgement of previously published material and of the work of others has been made through appropriate citation, quotation or both. Reproduction is authorized provided the source is acknowledged. UNITED4Surveillance Grant Agreement Number 101102070 TABLE OF CONTENT 1. BACKGROUND ...................................................................................................... 2 2. PILOT OBJECTIVES AND TEAM ............................................................................. 2 3. NEEDS AND GAPS ANALYSIS OF THE OLD STEC DATA MODEL .............................. 2 4. THE NEW STEC DATA MODEL AND MODIFICATIONS TO LABORATORY SURVEILLANCE TOOLS ......................................................................................... 4 4.1 New THL LIMS ............................................................................................. 4 4.2 Updated NIDR ............................................................................................. 6 5. TESTING THE ROBUSTNESS OF THE NEW DATA MODEL ........................................ 6 6. LESSONS LEARNT ................................................................................................. 6 7. REFERENCES ........................................................................................................ 7 UNITED4Surveillance Grant Agreement Number 101102070 1. Background In the Finnish Institute for Health and Welfare (THL), needs, gaps and deficiencies were noticed in collecting and storing information for laboratory-based surveillance of Shiga toxin-producing Escherichia coli (STEC). In the beginning of the pilot, which started on Q1/2023, the surveillance systems were THL’s in-house laboratory information management system (LIMS) called MILLA-LIMS and the National Infectious Diseases Register (NIDR). There were deficiences in transfering data between the systems and reporting to clinical microbiology laboratories. These constraints particularly hampered the epidemiological surveillance of STEC. Both, new THL LIMS and technically updated NIDR, were planned to be implemented during 2023-2024, with the aim to improve and broaden the data collection, storage and reporting for STEC. This aim was set as milestone 36 in United4Surveillance. At first, a needs and gaps analysis were performed on the present STEC data model in Q1/2023 among THL microbiologists and epidemiologist both having access to information needed in outbreak investigations and laboratory-based surveillance. Based on the analysis, a new data model was outlined. In the second part on the pilot, which was conducted between Q2/2023 and Q4/2024, the laboratory surveillance tools were set up and modified to meet the requirements of the new STEC data model. These laboratory surveillance tools were newly built THL LIMS and the updated NIDR. THL tested how the tools perform in collecting, storing and reporting genomic data originally obtained by PCR or whole-genomesequencing (WGS) in THL or in clinical laboratories. The robustness of the data model was planned to be tested by using data from an ongoing STEC case or cluster during spring 2025. The data collected by applying the new STEC data model will be available for surveillance by the European Centre for Disease Prevention and Control’s (ECDC) EpiPulse Cases system. 2. Pilot objectives and team The aims of the pilot were 1) To describe a new STEC data model for data collection, storage and reporting 2) To study how it can be integrated in the national surveillance in Finland. Personnel involved in the pilot in THL - Ulla-Maija Nakari, senior researcher, responsible for STEC laboratory methods in THL reference laboratory, MILLA-LIMS admin - Outi Nyholm, senior researcher, THL LIMS admin, co-responsible for STEC laboratory methods in THL reference laboratory - Saara Salmenlinna, leading expert, team lead in bacteriology laboratory in THL - Teemu Möttönen, senior systems analyst, NIDR product owner Stakeholders outside THL i.e. clinical microbiology laboratories, wellbeing services counties, municipal epidemiology units and authorities responsible for food safety, were not involved in the pilot. They will utilize the data obtained with the pilot product. 3. Needs and gaps analysis of the old STEC data model The current gaps in STEC lab-based reporting in THL were assessed. Only some of the laboratory results are reported by THL to NIDR and back to clinical laboratories even though there are more results available. We found out that around 20 % of microbial characteristics is reported from reference laboratory. In addition, gaps in epidemiological data were identified, e.g. information on travel, symptoms, or farm contact are not UNITED4Surveillance Grant Agreement Number 101102070 included in the laboratory referral. Also, EpiPulse Cases reporting (previously TESSy reporting) contain only the number of cases, and serotype, if available. Over the last decade, WGS has become a widely applied typing method in the laboratory, thus providing accurate data that can be utilized in molecular surveillance and epidemiological investigations. It requires modifications to the data model and modern reporting tools to store and report WGS data. The old STEC data model (Figure 1) had to be updated because it did not fully support laboratory-based surveillance with WGS data. The WGS-based methods provide much more data than the previously used methods, and it is not practical to store all of it in MILLA-LIMS. The system does not support instrument interfaces to qPCR machines and to bioinformatics tools. In addition, in MILLA-LIMS it is not possible to combine laboratory data and epidemiological data in a structured form. For this purpose, excel sheets have been used to combine the information provided by the reference laboratory tests and epidemiological investigations in THL. Figure 1. Old STEC data model. Laboratory data and epidemiological data in separate systems. Less data stored in MILLA-LIMS than is available and less data reported back to clinical laboratory and to NIDR. UNITED4Surveillance Grant Agreement Number 101102070 4. The new STEC data model and modifications to laboratory surveillance tools The new STEC data model was described and integrated into the national surveillance tools: new THL LIMS and updated NIDR. Figure 2 describes the new STEC data model with new elements displayed with a bold outline. Figure 2. New STEC data model with laboratory data stored in THL LIMS. The bold outline indicates more data reported and more reporting steps to clinical laboratory and to NIDR compared to the old data model. 4.1 New THL LIMS THL LIMS is a new system in THL that replaces MILLA-LIMS. THL LIMS is based on a commercial solution (LabVantage 8.8) and was implemented between 2022 and 2025. THL LIMS is currently under site acceptance testing and it will be in production as of autumn 2025. MILLA-LIMS will be renounced by the end of 2025. In THL LIMS, three key properties make it possible to collect more data and enhance reporting: 1. To connect laboratory instruments (qPCR machines) and bioinformatics tools/sequence database, InnuendoCLI (Yetukuri, 2025) for automatic data transfer. In InnuendoCLI, the original Innuendoproject (Llarena et al. 2018) was further developed in THL and in Finnish Food Authority to use Innuendo’s tools for genomic typing via command line interface. 2. Electronic reporting with clinical laboratories via Health Level 7 (HL7) v2.3 standard (Health Level 7 Finland, 2025) 3. Separate module called Epidemic view to link laboratory data with epidemiological data (Figure 3). Instrument interfaces in THL LIMS qPCR is used as a method to detect, isolate and type a STEC strain. qPCR machines are connected to THL LIMS via instrument interface to automate storing of the qPCR result in LIMS. InnuendoCLI is a bioinformatics UNITED4Surveillance Grant Agreement Number 101102070 pipeline and a database which performs WGS-data analysis of STEC whole genomes. The results of the pipeline are automatically recorded to THL LIMS via instrument interface. Electronic reporting between THL LIMS and clinical laboratory LIMS HL7-based electronic reporting has been tested with the largest clinical microbiology laboratory in Finland, Helsinki University Hospital Laboratory (HUSLAB). A national laboratory test code (6721 -BKanta) has been applied for STEC sample referral from clinical laboratories to THL reference laboratory. The 6721 -BKanta electronic sample referral has been tested between HUSLAB LIMS and THL LIMS but not yet put into production. For other clinical laboratories in Finland, HL7 is the recommended way for sample referral. However, the capabilities are different. If a clinical laboratory cannot set up HL7 sample referral with THL, it will be paper based. However, the laboratory test reports from THL back to the clinical laboratory will be delivered via secure e-mail automatically from THL LIMS. Epidemic view in THL LIMS In THL LIMS, a separate module called Epidemic view was built for outbreak investigation purposes between the reference laboratory and the epidemiological team in THL (Figure 3). The epidemiologists can access only the Epidemic view and no other parts of THL LIMS. In the Epidemic view, storing of epidemiological variables is done in a structured form. These variables are e.g. travel abroad, symptoms, farm contact, haemolyticuremic syndrome, bloody diarrhea, risk work such as child day care. The epidemiologists will see the laboratory results and the laboraotory personnel can see the epidemiological data. Due to this succesful data integration, the previous ad-hoc Excel sheets can be renounced. In the Epidemic view, it is possible to produce a test report with the data from patients, isolates, cases and outbreak and with a cluster image imported from bioinformatics software. Figure 3. New STEC data model with epidemic view. Epidemiological data is combined with the laboratory data in THL LIMS for surveillance and outbreak investigations. UNITED4Surveillance Grant Agreement Number 101102070 4.2 Updated NIDR The STEC finding and the characteristics of the isolated strain are notified from THL LIMS to NIDR. NIDR was technically updated in 2023. THL LIMS was adapted to be able to communicate machine-to-machine with NIDR. THL LIMS sends laboratory notifications in xml-format to NIDR with more microbial characteristics than with MILLA-LIMS. Some of these new microbial characteristics are ready to use both in THL LIMS and NIDR. For some characteristics, this still requires updates in the NIDR-end for it to be able to receive all the new data. The new data not only includes a comprehensive set of virulence markers obtained from WGS data, but also information on STEC strain isolation from the sample (if a STEC strain cannot be isolated from a sample there will be no typing results). NIDR sends additional data back to THL LIMS after it has received a laboratory notification. These are e.g. municipality of residence of the patient, NIDR case id and NIDR reporting group. 5. Testing the robustness of the new data model The robustness of the new data model was planned to be tested by using data from an ongoing STEC case or cluster. Testing period was between Jan 2025 and April 2025. The budget reductions in the spring 2024 affected STEC surveillance at THL as only samples from severe cases could be typed from the 1st of June 2024 onward. During spring 2025, when the robustness of the data model was tested, there were no severe STEC cases or clusters. In fact, no STEC isolates were received in THL’s laboratory between 1st Jan and 11th April 2025. Therefore, previously received isolates and STEC clusters from the years 2023-2024 were utilized in testing instead. This required migration of the data from MILLA-LIMS to THL LIMS. Testing included the process presented in Figure 2. starting from the data transfer from a clinical laboratory (sample referral) and included reporting steps back to clinical laboratory and notification steps to NIDR. The production of Epidemic view was decided to be postponed after THL LIMS is in production. Epidemic view will be launched by the end of 2025, but it was also tested with the retrospective data from STEC cases and clusters from 2023-2024. Data transfer to TESSy could not be tested from THL LIMS since the system is under site acceptance testing. Currently, data transfer to TESSy/EpiPulse Cases is done directly from NIDR. As EpiPulse Cases is being adapted to use (instead of TESSy), there may be a period when data transfer from NIDR to EpiPulse Cases will be done manually. 6. Lessons learnt It took longer than expected for THL LIMS to go into production, thus hindering the completion of the pilot as planned. Testing could not be done with the new data since there were no STEC isolates available in THL’s reference laboratory. However, this risk was addressed proactively after the budget reduction and it was managed by testing the new THL LIMS and the new data model with retrospective data. For electronic sample referral, THL and HUSLAB have succeeded in testing the messages with the national code for STEC laboratory test. However, this testing has been done to the test THL LIMS, not to the production side of the system yet. This still waits for the HUSLAB LIMS provider MyLab Ltd. to set up the connection to the production. The HL7 related work has required the involvement of four parties: THL, HUSLAB, THL LIMS provider Whitelake Software Point Ltd. and MyLab. The dependancy on multiple external partners has been a challenge that affected the timeline for the pilot. UNITED4Surveillance Grant Agreement Number 101102070 For the sake of time and the fact that THL LIMS production timetable cannot be postponed anymore, certain parts of the system had to be left implemented after the system is in production. The epidemic view was selected to implemented afterwards because it is a new feature and before it is fully implemented, outbreak investigation can be done utilizing Excel sheets. However, epidemic was still tested with retrospective STEC data during spring 2025. With the new STEC data model, we have been able to improve and broaden the microbial characteristic data collection, storage and reporting utilizing new THL LIMS and updated NIDR. To ensure sustainability, development work on data management, transfer, and storage will continue in future Direct Grant project FinSurveillance (THL, 2025). 7. References Health Level 7 Finland, 2025. https://www.hl7.fi Llarena A-K, Ribeiro-Gonçalves BF, Nuno Silva D, Halkilahti J, Machado MP, Da Silva MS, Jaakkonen A, Isidro J, Hämäläinen C, Joenperä J, Borges V, Viera L, Gomes JP, Correia C, Lunden J, Lau kkanen-Ninios R, Fredriksson-Ahomaa M , Bikandi J, San Millan R, MartinezBallesteros I, Laorden L, Mäesaar M, GrantinaIevina L, Hilbert F, Garaizar J, Oleastro M, Nevas M, Salmenlinna S, Hakkinen M, Carriço JA and Rossi M, 2018. I NNUENDO: A cross-sectoral platform for the integration of genomics in the surveillance of food-borne pathogens. EFSA supporting publication 2018: 15(11):EN-1498. 142 pp. doi: 10.2903/sp.efsa.2018.EN-1498 THL, 2025. FinSurveillance. https://thl.fi/en/research-and-development/research-and-projects/finsurveillance2