Diana Shamsutdinova
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Diana Shamsutdinova has a total of 28 co-authors. Here are the top co-authors by number of shared publications:
- Daniel Stahl
Department of Biostatistics and Health Informatics, King’s College London, London, UK
4 shared publications - Hugh Logan Ellis
Department of Biostatistics and Health Informatics, Institute of Psychiatry, Psychology and Neuroscience, King's College London, IoPPN, 16 De Crespigny Park, London, SE5 8AB, United Kingdom, 44 (0)20 7848 000
3 shared publications Angus Roberts
Department of Biostatistics and Health Informatics, Institute of Psychiatry, Psychology and Neuroscience, King’s College London
2 shared publicationsDaniel Stahl
Department of Biostatistics and Health Informatics, Institute of Psychiatry, Psychology and Neuroscience, King’s College London
2 shared publications- Morgan Roberts
Emergency Department, Bristol Royal Infirmary, Bristol, United Kingdom
2 shared publications - Otso Pelkonen2 shared publications
Latest Publications
- 17 September 2026
- Published by The R Foundation
pmsims: Simulation-Based Sample Size Tools for Prediction Models
Provides a flexible, simulation-based toolkit for exploring how much data are needed to develop reliable prediction models. It works by repeatedly generating data, fitting models, and evaluating performance to show how sample size affects predictive accuracy, calibration, and overfitting. The package supports continuous, binary, and time-to-event outcomes and can be used with both regression-based modelling approaches and machine-learning methods. It is designed to help researchers plan studies, assess feasibility, and build more robust and generalisable models. The methods are described in Olaniran et al. (2026) and Shamsutdinova et al. (2026) .
- 17 September 2026
- Published by The R Foundation
pmsims: Simulation-Based Sample Size Tools for Prediction Models
Provides a flexible, simulation-based toolkit for exploring how much data are needed to develop reliable prediction models. It works by repeatedly generating data, fitting models, and evaluating performance to show how sample size affects predictive accuracy, calibration, and overfitting. The package supports continuous, binary, and time-to-event outcomes and can be used with both regression-based modelling approaches and machine-learning methods. It is designed to help researchers plan studies, assess feasibility, and build more robust and generalisable models. The methods are described in Olaniran et al. (2026) and Shamsutdinova et al. (2026) .
- Open Access
- 1 September 2026
- Published by Public Library of Science (PLoS) in PLOS Medicine
- Vol. 23 (9) , e1005004
Evaluation of the FI-Lab, a laboratory-based, automated frailty index for acute care: A multicohort study
Background: Laboratory-based frailty indices (FI-Lab) have shown promise in geriatric medicine research. We aimed, in diverse samples, to determine the optimal construction of an FI-Lab for acute care and evaluate its validity as a measure of latent health status across the adult life span. Methods and findings: Our retrospective multicohort study used emergency department encounters in Boston, USA (MIMIC-IV-ED; 2011–2019) and London, UK (King’s College Hospital (KCH); 2017–2020); and a community-based cohort in the UK (UK Biobank; 2006–2010). We evaluated FI-Lab configurations by varying test selection strategies, number of tests, and minimum test thresholds. Sensitivity analyses examined performance across age, ethnicities, sexes, and model types to identify potential blind spots. The primary outcome was 1-year all-cause mortality. We analysed 227,736 visits (113,032 patients) from MIMIC-IV-ED; 152,305 visits (96,843 patients) from KCH; and 492,703 UK Biobank. An FI-Lab calculated from 25 commonly ordered tests (minimum 15 required) demonstrated hazard ratios for 1-year mortality approaching those of chronological age and exceeding the National Early Warning Score 2 (NEWS2). In combined models, hazard ratios for FI-Lab per standard deviation increase were 2.04 (95 %CI [1.99, 2.09]; p < 0.001) in MIMIC-IV-ED, 2.55 (95 %CI [2.42, 2.69]; p < 0.001) in KCH, and 1.87 (95% CI [1.80, 1.94]; p < 0.001) in UK Biobank. Performance was consistent across subject groupings. Discrimination plateaued at 20–40 tests. This was a retrospective study; future work exploring the impact on everyday use requires prospective evaluation, potentially in a randomised controlled trial. Conclusions: The FI-Lab provides an automated, scalable measure of patient vulnerability that is robust across healthcare settings and populations. A core set of 20–40 commonly ordered tests is sufficient for signal capture. It offers a pragmatic complement to clinical judgement without requiring manual data entry or additional tests.
- 1 September 2026
- Published by Public Library of Science (PLoS) in PLOS Medicine
- Vol. 23 (9) , e1005004
Open AccessEvaluation of the FI-Lab, a laboratory-based, automated frailty index for acute care: A multicohort study
Background: Laboratory-based frailty indices (FI-Lab) have shown promise in geriatric medicine research. We aimed, in diverse samples, to determine the optimal construction of an FI-Lab for acute care and evaluate its validity as a measure of latent health status across the adult life span. Methods and findings: Our retrospective multicohort study used emergency department encounters in Boston, USA (MIMIC-IV-ED; 2011–2019) and London, UK (King’s College Hospital (KCH); 2017–2020); and a community-based cohort in the UK (UK Biobank; 2006–2010). We evaluated FI-Lab configurations by varying test selection strategies, number of tests, and minimum test thresholds. Sensitivity analyses examined performance across age, ethnicities, sexes, and model types to identify potential blind spots. The primary outcome was 1-year all-cause mortality. We analysed 227,736 visits (113,032 patients) from MIMIC-IV-ED; 152,305 visits (96,843 patients) from KCH; and 492,703 UK Biobank. An FI-Lab calculated from 25 commonly ordered tests (minimum 15 required) demonstrated hazard ratios for 1-year mortality approaching those of chronological age and exceeding the National Early Warning Score 2 (NEWS2). In combined models, hazard ratios for FI-Lab per standard deviation increase were 2.04 (95 %CI [1.99, 2.09]; p < 0.001) in MIMIC-IV-ED, 2.55 (95 %CI [2.42, 2.69]; p < 0.001) in KCH, and 1.87 (95% CI [1.80, 1.94]; p < 0.001) in UK Biobank. Performance was consistent across subject groupings. Discrimination plateaued at 20–40 tests. This was a retrospective study; future work exploring the impact on everyday use requires prospective evaluation, potentially in a randomised controlled trial. Conclusions: The FI-Lab provides an automated, scalable measure of patient vulnerability that is robust across healthcare settings and populations. A core set of 20–40 commonly ordered tests is sufficient for signal capture. It offers a pragmatic complement to clinical judgement without requiring manual data entry or additional tests.
- Open Access
- 3 July 2026
- Published by JMIR Publications Inc. in JMIR Research Protocols
- Vol. 15, e85437
Evaluation and Comparison of Latent Health Risk Prediction Models for Clinical Triage: Protocol for a Mixed Methods Study
Background: Clinical triage requires integrating multiple information sources to identify patients at risk of deterioration. Tools capturing global health assessments beyond disease-specific scores are being developed using either bottom-up aggregation of simple indicators or top-down machine learning from large datasets. Their alignment with expert clinical judgment remains poorly characterized. Objective: This study evaluates 2 latent health measurement approaches: Frailty Index-laboratory, a transparent bottom-up tool aggregating laboratory abnormalities via deficit accumulation theory, and ETHOS-ARES (Enhanced Transformer for Health Outcome Simulation-Adaptive Risk Estimation System), a transformer-based foundation model generating multidimensional patient representations from electronic health records. We assess whether each tool’s severity rankings align with clinical consensus and whether they offer utility in triage decisions. Methods: In this 3-phase mixed methods study, at least 30 clinicians across hospital specialties reviewed 20 emergency department presentations derived from Medical Information Mart for Intensive Care IV-Emergency Department. Phase 1 compared unaided clinician severity and urgency judgments against model outputs using Spearman rank correlation, with a Turing-inspired indistinguishability test assessing whether model rankings fell within the distribution of clinician assessments. Phase 2 allocated clinicians to receive Frailty Index-laboratory or ETHOS-ARES outputs, measuring anchoring effects via within-person pre-post comparisons and exploring clinical utility through semistructured interviews analyzed using the Framework Method. Results: Ethics approval was granted in June 2025 (KCL Research Ethics Office; MRSP-24/25‐48707). Recruitment began in October 2025 (32 clinicians recruited as of manuscript submission), with data collection expected to be completed in January 2026 and analysis planned for March or April 2026. Conclusions: This study will quantify model-clinician agreement, measure anchoring effects, and generate qualitative insights on utility, trust, and adoption. The findings will inform the implementation of latent health measurement tools in clinical practice and provide a framework for the early-stage evaluation of artificial intelligence–based clinical decision support systems.
- 3 July 2026
- Published by JMIR Publications Inc. in JMIR Research Protocols
- Vol. 15, e85437
Open AccessEvaluation and Comparison of Latent Health Risk Prediction Models for Clinical Triage: Protocol for a Mixed Methods Study
Background: Clinical triage requires integrating multiple information sources to identify patients at risk of deterioration. Tools capturing global health assessments beyond disease-specific scores are being developed using either bottom-up aggregation of simple indicators or top-down machine learning from large datasets. Their alignment with expert clinical judgment remains poorly characterized. Objective: This study evaluates 2 latent health measurement approaches: Frailty Index-laboratory, a transparent bottom-up tool aggregating laboratory abnormalities via deficit accumulation theory, and ETHOS-ARES (Enhanced Transformer for Health Outcome Simulation-Adaptive Risk Estimation System), a transformer-based foundation model generating multidimensional patient representations from electronic health records. We assess whether each tool’s severity rankings align with clinical consensus and whether they offer utility in triage decisions. Methods: In this 3-phase mixed methods study, at least 30 clinicians across hospital specialties reviewed 20 emergency department presentations derived from Medical Information Mart for Intensive Care IV-Emergency Department. Phase 1 compared unaided clinician severity and urgency judgments against model outputs using Spearman rank correlation, with a Turing-inspired indistinguishability test assessing whether model rankings fell within the distribution of clinician assessments. Phase 2 allocated clinicians to receive Frailty Index-laboratory or ETHOS-ARES outputs, measuring anchoring effects via within-person pre-post comparisons and exploring clinical utility through semistructured interviews analyzed using the Framework Method. Results: Ethics approval was granted in June 2025 (KCL Research Ethics Office; MRSP-24/25‐48707). Recruitment began in October 2025 (32 clinicians recruited as of manuscript submission), with data collection expected to be completed in January 2026 and analysis planned for March or April 2026. Conclusions: This study will quantify model-clinician agreement, measure anchoring effects, and generate qualitative insights on utility, trust, and adoption. The findings will inform the implementation of latent health measurement tools in clinical practice and provide a framework for the early-stage evaluation of artificial intelligence–based clinical decision support systems.
- Open Access
- 22 April 2026
- Published by Springer Nature in Diabetologia
- Vol. 69 (8) , 2240-2252
Inequalities in type 2 diabetes incidence in a multiethnic population: a cohort study investigating the impact of ethnicity, migration and mental health comorbidities
Aims/hypothesis: Ethnic disparities in the incidence of type 2 diabetes mellitus are well documented in multiethnic urban populations, but the contributions of migration status and mental health are less well understood. This study used a large dataset from primary care centres in South London that is unique in that it includes migration-related information together with information on mental and physical health comorbidities. We aimed to assess how migration status and mental health contribute to longitudinal associations of ethnicity and type 2 diabetes risk in a multiethnic urban population. Methods: We conducted a longitudinal cohort study (2012–2019) of approximately 340,000 adults without baseline type 2 diabetes. Cox proportional hazards models were applied with sequential adjustments: first for age and sex; second, adding migration status (country of birth being UK or not); and third, further adding mental health conditions (depression, anxiety, severe mental illness), physical health factors (BMI, hypertension and other macrovascular diseases) and area-level deprivation. This approach allowed us to examine whether ethnic differences in the incidence of type 2 diabetes persist after accounting for additional factors. Results: South Asian, Black African and Black Caribbean groups had 2–3-fold higher type 2 diabetes risks compared with White British individuals, which were only partially explained by socioeconomic and clinical factors. Being born outside the UK increased type 2 diabetes risk by 29% across all ethnic groups. Depression/anxiety and severe mental illness were associated with a higher risk of type 2 diabetes. No statistical evidence of strong interactions between these factors was obtained. Conclusions/interpretation: Ethnicity, migration status and mental health conditions were each independently associated with type 2 diabetes risk, and ethnic disparities persisted after adjustment. The lack of evidence for interactions suggests that migration- and mental health-related mechanisms may operate similarly across ethnic groups rather than amplifying or mitigating existing disparities in type 2 diabetes rates. Efforts to reduce diabetes inequalities will require both support for post‑migration challenges and addressing of the broader structural and environmental determinants underlying persistent ethnic disparities.
- 22 April 2026
- Published by Springer Nature in Diabetologia
- Vol. 69 (8) , 2240-2252
Open AccessInequalities in type 2 diabetes incidence in a multiethnic population: a cohort study investigating the impact of ethnicity, migration and mental health comorbidities
Aims/hypothesis: Ethnic disparities in the incidence of type 2 diabetes mellitus are well documented in multiethnic urban populations, but the contributions of migration status and mental health are less well understood. This study used a large dataset from primary care centres in South London that is unique in that it includes migration-related information together with information on mental and physical health comorbidities. We aimed to assess how migration status and mental health contribute to longitudinal associations of ethnicity and type 2 diabetes risk in a multiethnic urban population. Methods: We conducted a longitudinal cohort study (2012–2019) of approximately 340,000 adults without baseline type 2 diabetes. Cox proportional hazards models were applied with sequential adjustments: first for age and sex; second, adding migration status (country of birth being UK or not); and third, further adding mental health conditions (depression, anxiety, severe mental illness), physical health factors (BMI, hypertension and other macrovascular diseases) and area-level deprivation. This approach allowed us to examine whether ethnic differences in the incidence of type 2 diabetes persist after accounting for additional factors. Results: South Asian, Black African and Black Caribbean groups had 2–3-fold higher type 2 diabetes risks compared with White British individuals, which were only partially explained by socioeconomic and clinical factors. Being born outside the UK increased type 2 diabetes risk by 29% across all ethnic groups. Depression/anxiety and severe mental illness were associated with a higher risk of type 2 diabetes. No statistical evidence of strong interactions between these factors was obtained. Conclusions/interpretation: Ethnicity, migration status and mental health conditions were each independently associated with type 2 diabetes risk, and ethnic disparities persisted after adjustment. The lack of evidence for interactions suggests that migration- and mental health-related mechanisms may operate similarly across ethnic groups rather than amplifying or mitigating existing disparities in type 2 diabetes rates. Efforts to reduce diabetes inequalities will require both support for post‑migration challenges and addressing of the broader structural and environmental determinants underlying persistent ethnic disparities.
- Open Access
- 7 October 2025
- Published by JMIR Publications Inc. in JMIR Preprints
Evaluation of latent health risk prediction models: A protocol for a mixed-methods study using clinical triage as a vehicle for comparison and discussion (Preprint)
BACKGROUND Determining clinical urgency and resource allocation within acute patient populations is complex. Tools are being developed to capture global assessments of patient health beyond disease-specific scores, aiming to provide dynamic assessments incorporating both baseline physiological reserve and immediate illness severity OBJECTIVE This study evaluates two contrasting approaches to latent health measurement: FI-lab, a transparent algorithmic tool using bottom-up aggregation of laboratory abnormalities, and ETHOS-ARES, a transformer-based model using top-down learning from electronic health records. METHODS This two-phase mixed-methods study uses clinical triage scenarios with ≥30 clinicians sampled across hospital roles (physicians, surgeons, critical care nurses, advanced practitioners). Phase 1 compares unaided clinician judgments of severity and clinical urgency against model outputs using Spearman's rank correlation (ρ ≥ 0.70 indicates good agreement as primary outcome). A novel "clinical Turing test" assesses whether model rankings are statistically distinguishable from clinician assessments. Phase 2 allows clinicians to incorporate model outputs into identical tasks, measuring anchoring effects through within-person pre/post comparison. Semi-structured interviews explore clinical utility, trust, and perceived limitations. Case materials derive from MIMIC-IV-ED, presented as slide decks containing emergency department documentation, examination findings, laboratory results, and imaging reports. Qualitative analysis follows the Framework Method with dual independent coding. RESULTS Data collection is planned for October-November 2025, with analysis to follow in December 2025. CONCLUSIONS We anticipate findings will quantify agreement between model outputs and clinician consensus, measure any anchoring effects from model exposure, and generate insights from qualitative data regarding clinical utility, feasibility, and factors influencing clinician trust and adoption of different approaches to latent health measurement. CLINICALTRIAL n/a
- 7 October 2025
- Published by JMIR Publications Inc. in JMIR Preprints
Open AccessEvaluation of latent health risk prediction models: A protocol for a mixed-methods study using clinical triage as a vehicle for comparison and discussion (Preprint)
BACKGROUND Determining clinical urgency and resource allocation within acute patient populations is complex. Tools are being developed to capture global assessments of patient health beyond disease-specific scores, aiming to provide dynamic assessments incorporating both baseline physiological reserve and immediate illness severity OBJECTIVE This study evaluates two contrasting approaches to latent health measurement: FI-lab, a transparent algorithmic tool using bottom-up aggregation of laboratory abnormalities, and ETHOS-ARES, a transformer-based model using top-down learning from electronic health records. METHODS This two-phase mixed-methods study uses clinical triage scenarios with ≥30 clinicians sampled across hospital roles (physicians, surgeons, critical care nurses, advanced practitioners). Phase 1 compares unaided clinician judgments of severity and clinical urgency against model outputs using Spearman's rank correlation (ρ ≥ 0.70 indicates good agreement as primary outcome). A novel "clinical Turing test" assesses whether model rankings are statistically distinguishable from clinician assessments. Phase 2 allows clinicians to incorporate model outputs into identical tasks, measuring anchoring effects through within-person pre/post comparison. Semi-structured interviews explore clinical utility, trust, and perceived limitations. Case materials derive from MIMIC-IV-ED, presented as slide decks containing emergency department documentation, examination findings, laboratory results, and imaging reports. Qualitative analysis follows the Framework Method with dual independent coding. RESULTS Data collection is planned for October-November 2025, with analysis to follow in December 2025. CONCLUSIONS We anticipate findings will quantify agreement between model outputs and clinician consensus, measure any anchoring effects from model exposure, and generate insights from qualitative data regarding clinical utility, feasibility, and factors influencing clinician trust and adoption of different approaches to latent health measurement. CLINICALTRIAL n/a
- 22 January 2024
- Published by The R Foundation
survcompare: Nested Cross-Validation to Compare Cox-PH, Cox-Lasso, Survival Random Forests
- 22 January 2024
- Published by The R Foundation
survcompare: Nested Cross-Validation to Compare Cox-PH, Cox-Lasso, Survival Random Forests
- Open Access
- 4 July 2023
- Published by Wolters Kluwer Health in Psychiatric Genetics
- Vol. 33 (5) , 191-201
Schizophrenia polygenic risk score and type 2 diabetes onset in older adults with no schizophrenia diagnosis
Objectives: An association between type 2 diabetes (T2DM) and schizophrenia has long been observed, and recent research revealed presence of shared genetic factors. However, epidemiological evidence was inconsistent, some reported insignificant contribution of genetic factors to T2DM-schizophrenia comorbidity. Prior works studied people with schizophrenia, particularly, antipsychotic-naive patients, or those during the first psychotic experience to limit schizophrenia-related environmental factors. In contrast, we controlled such factors by utilizing a general population sample of individuals undiagnosed with schizophrenia. We hypothesized that if schizophrenia genetics impact T2DM development and such impact is not fully mediated by schizophrenia-related environment, people with high polygenic schizophrenia risk would exhibit elevated T2DM incidence. Methods: Using a population-representative sample of adults aged ≥50 from English Longitudinal Study of Ageing (n = 5968, 493 T2DM cases, average follow-up 8.7 years), we investigated if schizophrenia polygenic risk score (PGS-SZ) is associated with T2DM onset. A proportional hazards model with interval censoring was adjusted for age and sex (Model 1), and age, sex, BMI, hypertension, cardiovascular diseases, exercise, smoking, depressive symptoms and T2DM polygenic risk score (Model 2). According to the power calculations, hazard rates > 1.14 per standard deviation in PGS-SZ could be detected. Results: We did not observe a significant association between PGS-SZ and T2DM incidence (hazard ratio 1.04; 95% CI 0.93–1.15; and 1.01, 95% CI 0.94–1.09). Conclusion: Our results suggest low contribution of the intrinsic biological mechanisms driven by the polygenic risk of schizophrenia on future T2DM onset. Further research is needed.
- 4 July 2023
- Published by Wolters Kluwer Health in Psychiatric Genetics
- Vol. 33 (5) , 191-201
Open AccessSchizophrenia polygenic risk score and type 2 diabetes onset in older adults with no schizophrenia diagnosis
Objectives: An association between type 2 diabetes (T2DM) and schizophrenia has long been observed, and recent research revealed presence of shared genetic factors. However, epidemiological evidence was inconsistent, some reported insignificant contribution of genetic factors to T2DM-schizophrenia comorbidity. Prior works studied people with schizophrenia, particularly, antipsychotic-naive patients, or those during the first psychotic experience to limit schizophrenia-related environmental factors. In contrast, we controlled such factors by utilizing a general population sample of individuals undiagnosed with schizophrenia. We hypothesized that if schizophrenia genetics impact T2DM development and such impact is not fully mediated by schizophrenia-related environment, people with high polygenic schizophrenia risk would exhibit elevated T2DM incidence. Methods: Using a population-representative sample of adults aged ≥50 from English Longitudinal Study of Ageing (n = 5968, 493 T2DM cases, average follow-up 8.7 years), we investigated if schizophrenia polygenic risk score (PGS-SZ) is associated with T2DM onset. A proportional hazards model with interval censoring was adjusted for age and sex (Model 1), and age, sex, BMI, hypertension, cardiovascular diseases, exercise, smoking, depressive symptoms and T2DM polygenic risk score (Model 2). According to the power calculations, hazard rates > 1.14 per standard deviation in PGS-SZ could be detected. Results: We did not observe a significant association between PGS-SZ and T2DM incidence (hazard ratio 1.04; 95% CI 0.93–1.15; and 1.01, 95% CI 0.94–1.09). Conclusion: Our results suggest low contribution of the intrinsic biological mechanisms driven by the polygenic risk of schizophrenia on future T2DM onset. Further research is needed.
- Open Access
- 7 February 2023
- Published by Elsevier in International Journal of Medical Informatics
- Vol. 172, 105019
Predicting type 2 diabetes prevalence for people with severe mental illness in a multi-ethnic East London population
- 7 February 2023
- Published by Elsevier in International Journal of Medical Informatics
- Vol. 172, 105019
Open AccessPredicting type 2 diabetes prevalence for people with severe mental illness in a multi-ethnic East London population
- 10 June 2022
- Published by Springer Nature
- p. 170-181
Combining Cox Model and Tree-Based Algorithms to Boost Performance and Preserve Interpretability for Health Outcomes
- 10 June 2022
- Published by Springer Nature
- p. 170-181
Combining Cox Model and Tree-Based Algorithms to Boost Performance and Preserve Interpretability for Health Outcomes