Longitude is delivered on the Nightingale Health platform. The papers below are the validation literature behind it, grouped by disease domain, with the institutions and researchers involved in each. Every entry links to its DOI, which resolves permanently regardless of where a journal later moves the article.
More than 900 peer-reviewed publications have been produced using the platform. This is a selection of the studies most directly relevant to the five risk scores the health check reports.
A. Multi-disease validation: general population
1Buergel et al. · Nature Medicine, September 2022
Original proof-of-concept paper for multi-disease metabolomic prediction. Statistical models trained on 168 circulating metabolic markers in 117,981 UK Biobank participants, externally validated across four European cohorts: Whitehall II Study, Rotterdam Study, Leiden Longevity Study, and PROSPER trial. Independently funded (Charité and the Einstein Foundation Berlin; no Nightingale funding). Cited here as the primary cross-cohort European replication reference; the peer-reviewed foundation of the deployed Nightingale Health Check is the Nightingale Health Biobank Collaborative Group 2024 study (entry 3 below).
- DOI
- 10.1038/s41591-022-01980-3
- Institutions
- Berlin Institute of Health at Charité · University College London · University of Glasgow · MRC Epidemiology Unit, University of Cambridge · Erasmus MC Rotterdam · Leiden University Medical Centre · Netherlands Cancer Institute
- Contributors
- Prof. John Deanfield (UCL, Global Cardiovascular Health Group) – cardiologist and population health scientist, former WHO adviser. Prof. Naveed Sattar (University of Glasgow) – Professor of Metabolic Medicine, Fellow of the Royal Society of Edinburgh. Prof. Claudia Langenberg (MRC Cambridge / Charité Berlin) – Director of Precision Healthcare, epidemiologist specialising in metabolomics and disease prediction. Independently funded; no Nightingale financial interest declared.
2Julkunen et al. · Nature Communications, February 2023
Co-authored by Dr Peter Würtz. NMR biomarker data from 118,461 UK Biobank participants, generating an atlas of associations across 249 biomarkers and over 700 disease endpoints. Results revealed associations well beyond cardiometabolic diseases, including infectious diseases, cancers, joint disorders and mental health. Establishes the breadth of the Nightingale Health Check.
- DOI
- 10.1038/s41467-023-36231-7
- Institutions
- Nightingale Health (Helsinki) · University of Helsinki · University of Oulu · Estonian Genome Centre, University of Tartu · UK Biobank, University of Oxford
- Contributors
- Dr Peter Würtz (Nightingale Health) – Scientific co-founder, Associate Professor, University of Helsinki. Dr Heli Julkunen – Senior Scientist, Nightingale Health. Dr Jeffrey Barrett – former Chief Scientific Officer, Nightingale Health, previously Director of the Wellcome Sanger Institute Human Genetics programme.
3Nightingale Health Biobank Collaborative Group (2024) · Nature Communications, November 2024
The largest validation study to date. 700,217 participants across three national biobanks. Metabolomic scores more strongly associated with future disease onset than polygenic scores for most diseases. In 18,709 individuals with biomarkers measured at two time points, those whose scores changed had different future disease risk – the scientific basis for serial metabolomic profiling as a monitoring and intervention-tracking tool.
- DOI
- 10.1038/s41467-024-54357-0
- Institutions
- Wellcome Sanger Institute · University of Helsinki · Estonian Genome Centre, University of Tartu · UK Biobank / University of Oxford · Nightingale Health (Helsinki) · Broad Institute of MIT and Harvard
- Contributors
- Dr Jeffrey Barrett (Nightingale Health / Wellcome Sanger Institute) – former Director of Human Genetics, Wellcome Sanger Institute. Prof. Lili Milani (Estonian Genome Centre) – Director of the Estonian Genome Centre. Dr Peter Würtz (Nightingale Health) – Scientific co-founder.
B. Cardiovascular disease
4Ritchie et al. · European Heart Journal, 2026
297,463 UK Biobank participants, 8,919 incident CVD cases. Adding NMR scores to SCORE2 gave a net case reclassification rate of 8.85% (95% CI 7.90 to 9.80), more than twice the improvement obtained from polygenic scores. With NMR scores, eleven clinical biomarkers and polygenic scores combined, net case reclassification reached 16.66% and modelled CVD events prevented per 100,000 screened rose from 229 to 413, with the number of statins prescribed per event prevented essentially unchanged. Note: the prevention figures are modelled estimates grounded in trial-validated treatment effects, not directly observed outcomes.
- Citation
- Eur Heart J. 2026;47(15):1861–1873.
- DOI
- 10.1093/eurheartj/ehaf947
- Institutions
- University of Cambridge (MRC Epidemiology Unit / Department of Public Health and Primary Care) · Baker Heart and Diabetes Institute, Melbourne · University of Melbourne
- Contributors
- Prof. Michael Inouye (University of Cambridge) – Professor of Systems Genomics and Population Health, MRC Investigator. Dr Scott Ritchie – Research Fellow, MRC Epidemiology Unit, Cambridge. Nightingale Health has no authors on this paper; it is acknowledged in it for early access to the UK Biobank NMR biomarker data.
5Xie et al. (General Population) · European Journal of Preventive Cardiology, April 2025
Independent of Nightingale. 187,039 UK Biobank participants and 5,578 from the German ESTHER cohort. Integration of NMR metabolomic biomarkers into SCORE2 markedly improved 10-year cardiovascular risk prediction.
- DOI
- 10.1093/eurjpc/zwaf254
- Institutions
- German Cancer Research Centre (DKFZ), Heidelberg · University of Heidelberg · UK Biobank / University of Oxford · Heidelberg University Hospital
- Contributors
- Prof. Hermann Brenner (DKFZ Heidelberg) – Head of the Division of Clinical Epidemiology and Ageing Research, Fellow of the Leopoldina. Dr Ben Schöttker – Senior Research Scientist, DKFZ. Independent of Nightingale Health; no commercial interest declared.
6Xie et al. (T2DM Population) · Cardiovascular Diabetology, January 2025
Companion paper focused on established type 2 diabetes patients – directly relevant to Bermuda and Cayman. Key finding: replacement of clinical chemistry-based biomarkers in the SCORE2-Diabetes model by NMR metabolomics did not lead to worse MACE prediction. NMR can substitute for conventional tests with equivalent predictive performance and lower cost.
- DOI
- 10.1186/s12933-025-02581-3
- Institutions
- German Cancer Research Centre (DKFZ), Heidelberg · University of Heidelberg · Heidelberg University Hospital
- Contributors
- Prof. Hermann Brenner (DKFZ Heidelberg) – Head of the Division of Clinical Epidemiology and Ageing Research, Fellow of the Leopoldina. Dr Ben Schöttker – Senior Research Scientist, DKFZ. Dr Ruijie Xie – postdoctoral epidemiologist. Independent of Nightingale; no commercial interest declared.
7Oexner et al. · European Journal of Heart Failure, 2024
Dedicated heart failure prediction paper. Conclusion: serum metabolomics improves incident heart failure risk prediction over the Pooled Cohort Equations to Prevent HF. Scores based on age, sex and metabolomics exhibit similar predictive power to clinically-based models, potentially offering a cost-effective, standardisable, and scalable single-domain alternative. Provides the specific evidence for heart failure among the additional disease signals.
- DOI
- 10.1002/ejhf.3226
- Institutions
- King’s College London (School of Cardiovascular Medicine and Sciences) · Brigham and Women’s Hospital, Harvard Medical School · UK Biobank / University of Oxford
- Contributors
- Prof. Ajay Shah (King’s College London) – British Heart Foundation Professor of Cardiology and Head of the School of Cardiovascular Medicine, Fellow of the Royal Society. Prof. Ravi Shah (Brigham and Women’s / Harvard) – Associate Professor, heart failure. Dr Rafael Oexner – Postdoctoral Research Fellow, King’s College London.
D. Chronic kidney disease
9Geng et al. · American Journal of Kidney Diseases, January 2024
Primary population-scale validation paper for NMR in CKD risk prediction. 91,532 UK Biobank participants without CKD. 90 of 142 lipid biomarkers significantly associated with incident CKD. Directly relevant to the small-island health economics case: dialysis is among the highest-cost recurring care categories in any island health system.
- DOI
- 10.1053/j.ajkd.2023.05.014
- Institutions
- Tulane University School of Public Health and Tropical Medicine, New Orleans · UK Biobank / University of Oxford · Nightingale Health (Helsinki)
- Contributors
- Prof. Jiang He (Tulane University) – Professor and Chair of Epidemiology. Dr Tao Geng – Research Scientist, Tulane University. Dr Peter Würtz (Nightingale Health) – Scientific co-founder.
10Julkunen et al. · Nephrology Dialysis Transplantation, 2023
275,000 UK Biobank individuals. Adding metabolic biomarkers to standard risk factors improved CKD prediction AUC from 0.74 to 0.82. Particularly significant in type 2 diabetics with mildly to moderately decreased kidney function (eGFR 60–90), where AUC improved from 0.60 to 0.70 – a clinically meaningful gain in a high-risk population.
- DOI
- 10.1093/ndt/gfad063c_3811
- Institutions
- Nightingale Health (Helsinki) · University of Helsinki · UK Biobank / University of Oxford
- Contributors
- Dr Heli Julkunen (Nightingale Health) – Senior Scientist. Dr Peter Würtz (Nightingale Health) – Scientific co-founder. Presented at the European Renal Association Congress 2023.
11Jin et al. · Diabetologia, 2024
1,991 adults with type 2 diabetes from the Hong Kong Diabetes Biobank. NMR metabolites correlated with reduced eGFR and albuminuria, and were associated with incident CVD over 5.2 years. A prediction model comprising age, sex and three selected metabolites performed comparably to established risk models. Noteworthy as an Asian population dataset – extends the cross-ethnic evidence base.
- DOI
- 10.1007/s00125-024-06098-4
- Institutions
- Hong Kong Institute of Diabetes and Obesity, The Chinese University of Hong Kong · Department of Medicine and Therapeutics, Prince of Wales Hospital, Hong Kong · Li Ka Shing Institute of Health Sciences
- Contributors
- Prof. Ronald Ma (The Chinese University of Hong Kong) – Professor of Medicine and Deputy Director of the Hong Kong Institute of Diabetes and Obesity. Prof. Juliana Chan – Emeritus Professor of Medicine, CUHK, Fellow of the Royal College of Physicians. Dr Heung Man Lee – Research Assistant Professor.
E. Liver disease: MASLD and cirrhosis
12Huang et al. · Journal of Hepatology, November 2024
Strongest available paper for NMR in MASLD/MASH prediction. AUROCs of 0.87 (Chinese cohort) and 0.81 (Finnish cohort). Participants with high or intermediate MASH risk had markedly higher risk of MASLD-related mortality – hazard ratio 23.19 in Chinese individuals and 20.15 in European individuals. The score was superior to FIB-4 and NAFLD Fibrosis Score in predicting MASLD-related death. Inter-ethnic validation across Chinese and European populations is noteworthy.
- DOI
- 10.1016/j.jhep.2024.10.015
- Institutions
- National Institute for Health Research (NIHR) Biomedical Research Centre, Nottingham University Hospitals NHS Trust · University of Nottingham · THL Finnish Institute for Health and Welfare (FINRISK cohorts) · University of Helsinki · Nightingale Health (Helsinki)
- Contributors
- Prof. Guruprasad Aithal (University of Nottingham / NIHR Nottingham BRC) – Professor of Hepatology and Director of the NIHR Nottingham Biomedical Research Centre. Dr Xiaoning Huang – Research Fellow, hepatology metabolomics. Dr Peter Würtz (Nightingale Health) – Scientific co-founder.
13Guo et al. · Hepatology, 2025
64,005 UK Biobank individuals with NMR metabolomics at baseline. A metabolomic state-integrated model predicted 10-year risk of liver cirrhosis complications, performing better than APRI and FIB-4. Supports the use of metabolomic profiling for precise prevention of liver cirrhosis complications.
- DOI
- 10.1097/HEP.0000000000001291
- Institutions
- Oxford University Hospitals NHS Foundation Trust · University of Oxford · Cardiff University · University of Glasgow · University of Edinburgh · Hounslow and Richmond Community Healthcare NHS Trust · University of Birmingham · UK Biobank
- Contributors
- Prof. Roger Williams CBE (Institute of Hepatology, London) – founding Chair of the Lancet Commission on Liver Disease in the UK. The collaborative reflects the six UK Biobank recruitment centres involved: Oxford, Cardiff, Glasgow, Edinburgh, Hounslow and Birmingham.
14Williams R et al. · The Lancet, 2021
Commissioned review confirming the current liver disease pathway is diagnostic, not predictive. Key finding: when liver enzyme concentrations are used as the diagnostic entry point, approximately 39% of those with advanced liver disease go undetected. Endorsed by Dr Charles Alessi as the primary reference for the MASLD comparator discussion. Confirms that Fibroscan is used selectively downstream, not as a population screening tool.
- DOI
- 10.1016/S0140-6736(20)32396-5
- Institutions
- Institute of Hepatology, Foundation for Liver Research, London · King’s College London · University of Edinburgh · University of Birmingham · NHS England
- Contributors
- Prof. Roger Williams CBE (Institute of Hepatology) – Chair of the Commission. Prof. Sarah Sherwood (NHS England) – National Clinical Director for Liver Disease. Prof. Peter Hayes (University of Edinburgh) – Professor of Hepatology. The Commission was commissioned by the UK government and NHS England.
14aMusso et al. · BMJ, November 2025
The most current BMJ clinical practice update on MASLD diagnosis and management (November 2025). Directly confirms that no international guideline body – AASLD (US), EASL-EASD-EASO (Europe), ADA, INASL (India), CSH-CMA (China) – recommends population-based screening, on the grounds that the current diagnostic pathway is not cost-effective at population scale. Establishes MASLD prevalence at 31.3% of European, Asian and American adults. Confirms all-cause mortality hazard ratio of 1.23 and CVD hazard ratio of 1.45 for MASLD patients. The diagnostic algorithm presented is entirely reactive – triggered by elevated liver enzymes or known risk factors – confirming the gap that the NHC fills.
- DOI
- 10.1136/bmj-2025-084950
- Institutions
- Department of Emergency Medicine, San Luigi Gonzaga University Hospital, University of Turin, Italy · University of California, San Francisco · Laboratory of Diabetology and Metabolism, Città della Salute e della Scienza, Turin
- Contributors
- Dr Giovanni Musso (University of Turin) – hepatologist and emergency physician, member of European Association for the Study of the Liver guideline panels. Prof. Jacquelyn Maher (University of California, San Francisco) – Professor of Medicine, UCSF School of Medicine. Dr Roberto Gambino (University of Turin) – Professor, metabolic liver disease and diabetology.
F. Dementia and neurological disease
15Harshfield & Markus · Neurology, August 2023
118,021 UK Biobank participants. Baseline metabolomic profiles associated with incident stroke and dementia, and with MRI markers of cerebral small vessel disease (white matter hyperintensities and lacunes). Multiple metabolites identified with longitudinal associations with future dementia risk, using Mendelian randomisation to assess causality. This is the most significant population-level evidence currently available linking NMR metabolomics to dementia risk. There is currently no validated population-level predictive blood test for dementia in routine clinical practice – existing cognitive tools (MoCA, MMSE) detect impairment already present. The NMR signal for dementia is emerging and not yet a validated clinical score, but represents the most promising population-level approach available.
- DOI
- 10.1212/WNL.0000000000207458
- Institutions
- Stroke Research Group, Department of Clinical Neurosciences, University of Cambridge · UK Biobank / University of Oxford
- Contributors
- Prof. Hugh Markus (University of Cambridge) – Professor of Stroke Medicine, Director of the Stroke Research Group, Fellow of the Academy of Medical Sciences. Dr Eric Harshfield – Research Associate, Cambridge Stroke Research Group.
G. Ageing and metabolic age
16Zhang et al. · Nature Communications, 2024
250,341 UK Biobank individuals. A metabolomic ageing score was derived and validated, demonstrating that biological age acceleration measured by NMR metabolomics carries independent mortality risk information. The score demonstrated optimal predictive capability for short-term (1 to 5 year) mortality, surpassing chronological age. Supports the clinical credibility of the metabolic age domain in the Nightingale report – a patient whose metabolic age is significantly above their chronological age is not receiving a cosmetic number but a biologically grounded mortality risk signal.
- DOI
- 10.1038/s41467-024-45638-9
- Institutions
- Xiangya Hospital, Central South University, Changsha, China · UK Biobank / University of Oxford · Nightingale Health (Helsinki)
- Contributors
- Prof. Xiang Chen (Central South University) – Professor of Dermatology. Dr Shiyu Zhang – PhD candidate, Clinical Medicine, Xiangya Hospital, lead author. Dr Peter Würtz (Nightingale Health) – Scientific co-founder. Presented at the UK Biobank Scientific Conference 2026.
17Deelen et al. · Nature Communications, 2019
44,168 individuals across 12 cohorts. NMR-based mortality risk prediction accuracy was high across all cohorts. The authors suggest this type of score could guide treatment decisions – for example, when deciding whether an elderly person is too fragile for an invasive operation. One year of biological age acceleration is associated with a 17% increase in mortality hazard for men and 12% for women. Foundational paper for the mortality risk application of metabolomic profiling.
- DOI
- 10.1038/s41467-019-11311-9
- Institutions
- Max Planck Institute for Biology of Ageing, Cologne · Leiden University Medical Centre · University of Groningen · Erasmus MC Rotterdam · University of Helsinki · Estonian Genome Centre, University of Tartu · University of Oulu · University of Southern Denmark · Copenhagen University Hospital · University of Cambridge · Nightingale Health (Helsinki)
- Contributors
- Prof. P. Eline Slagboom (Leiden University Medical Centre) – Professor of Molecular Epidemiology of Ageing, Head of the Molecular Epidemiology section. Dr Joris Deelen (Max Planck Institute for Biology of Ageing) – Group Leader. Dr Peter Würtz (Nightingale Health) – Scientific co-founder. The study spans 12 cohorts across 6 countries.
H. Health economics
18Martikainen et al. · medRxiv pre-print, 2025
Pre-print, currently under journal review. The study evaluated a use case in which the Nightingale test would replace the currently used standard clinical risk calculators FINRISK (cardiovascular risk, similar to SCORE2) and FINDRISC (diabetes risk, similar to QDiabetes) in Finnish public healthcare. The study concluded that the Nightingale-based approach (1) saved healthcare professionals’ time, (2) increased healthy years of life, and (3) reduced costs, compared with the standard clinical calculators. In health-economics terms a result that is “dominant” across all three axes is unusual and is the strongest current piece of population-scale health-economics evidence for metabolomic risk prediction. Cited with the pre-print caveat.
- DOI
- 10.1101/2025.09.11.25335561
- Institutions
- University of Eastern Finland · Finnish Institute for Health and Welfare (THL) · Nightingale Health (Helsinki)
- Contributors
- Prof. Janne Martikainen (University of Eastern Finland) – Professor of Health Economics, specialising in cost-effectiveness analysis of preventive interventions.
