1
|
Proust-Lima C, Saulnier T, Philipps V, Traon APL, Péran P, Rascol O, Meissner WG, Foubert-Samier A. Describing complex disease progression using joint latent class models for multivariate longitudinal markers and clinical endpoints. Stat Med 2023; 42:3996-4014. [PMID: 37461227 DOI: 10.1002/sim.9844] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/10/2022] [Revised: 01/31/2023] [Accepted: 06/26/2023] [Indexed: 09/05/2023]
Abstract
Neurodegenerative diseases are characterized by numerous markers of progression and clinical endpoints. For instance, multiple system atrophy (MSA), a rare neurodegenerative synucleinopathy, is characterized by various combinations of progressive autonomic failure and motor dysfunction, and a very poor prognosis. Describing the progression of such complex and multi-dimensional diseases is particularly difficult. One has to simultaneously account for the assessment of multivariate markers over time, the occurrence of clinical endpoints, and a highly suspected heterogeneity between patients. Yet, such description is crucial for understanding the natural history of the disease, staging patients diagnosed with the disease, unravelling subphenotypes, and predicting the prognosis. Through the example of MSA progression, we show how a latent class approach modeling multiple repeated markers and clinical endpoints can help describe complex disease progression and identify subphenotypes for exploring new pathological hypotheses. The proposed joint latent class model includes class-specific multivariate mixed models to handle multivariate repeated biomarkers possibly summarized into latent dimensions and class-and-cause-specific proportional hazard models to handle time-to-event data. Maximum likelihood estimation procedure, validated through simulations is available in the lcmm R package. In the French MSA cohort comprising data of 598 patients during up to 13 years, five subphenotypes of MSA were identified that differ by the sequence and shape of biomarkers degradation, and the associated risk of death. In posterior analyses, the five subphenotypes were used to explore the association between clinical progression and external imaging and fluid biomarkers, while properly accounting for the uncertainty in the subphenotypes membership.
Collapse
Affiliation(s)
- Cécile Proust-Lima
- Univ. Bordeaux, Inserm, Bordeaux Population Health Research Center, U1219, Bordeaux, France
- Inserm, CIC1401-EC, Bordeaux, France
| | - Tiphaine Saulnier
- Univ. Bordeaux, Inserm, Bordeaux Population Health Research Center, U1219, Bordeaux, France
| | - Viviane Philipps
- Univ. Bordeaux, Inserm, Bordeaux Population Health Research Center, U1219, Bordeaux, France
| | - Anne Pavy-Le Traon
- MSA Reference Center and CIC-1436, Department of Clinical Pharmacology and Neurosciences, NeuroToul COEN Center, University of Toulouse 3, CHU of Toulouse, INSERM, Toulouse, France
| | - Patrice Péran
- ToNIC, Toulouse NeuroImaging Center, Univ Toulouse, Inserm, UPS, Toulouse, France
| | - Olivier Rascol
- MSA Reference Center and CIC-1436, Department of Clinical Pharmacology and Neurosciences, NeuroToul COEN Center, University of Toulouse 3, CHU of Toulouse, INSERM, Toulouse, France
| | - Wassilios G Meissner
- Univ. Bordeaux, CNRS, IMN, UMR5293, Bordeaux, France
- Dept. Medicine, University of Otago, Christchurch, and New Zealand Brain Research Institute, Christchurch, New Zealand
| | - Alexandra Foubert-Samier
- Univ. Bordeaux, Inserm, Bordeaux Population Health Research Center, U1219, Bordeaux, France
- Inserm, CIC1401-EC, Bordeaux, France
- Univ. Bordeaux, CNRS, IMN, UMR5293, Bordeaux, France
| |
Collapse
|
2
|
O'Connor S, Blais C, Mésidor M, Talbot D, Poirier P, Leclerc J. Great diversity in the utilization and reporting of latent growth modeling approaches in type 2 diabetes: A literature review. Heliyon 2022; 8:e10493. [PMID: 36164545 PMCID: PMC9508412 DOI: 10.1016/j.heliyon.2022.e10493] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/03/2022] [Revised: 05/09/2022] [Accepted: 08/25/2022] [Indexed: 12/03/2022] Open
Abstract
Introduction The progression of complications of type 2 diabetes (T2D) is unique to each patient and can be depicted through individual temporal trajectories. Latent growth modeling approaches (latent growth mixture models [LGMM] or latent class growth analysis [LCGA]) can be used to classify similar individual trajectories in a priori non-observed groups (latent groups), sharing common characteristics. Although increasingly used in the field of T2D, many questions remain regarding the utilization of these methods. Objective To review the literature of longitudinal studies using latent growth modeling approaches to study T2D. Methods MEDLINE (Ovid), EMBASE, CINAHL and Wb of Science were searched through August 25th, 2021. Data was collected on the type of latent growth modeling approaches (LGMM or LCGA), characteristics of studies and quality of reporting using the GRoLTS-Checklist and presented as frequencies. Results From the 4,694 citations screened, a total of 38 studies were included. The studies were published beetween 2011 and 2021 and the length of follow-up ranged from 8 weeks to 14 years. Six studies used LGMM, while 32 studies used LCGA. The fields of research varied from clinical research, psychological science, healthcare utilization research and drug usage/pharmaco-epidemiology. Data sources included primary data (clinical trials, prospective/retrospective cohorts, surveys), or secondary data (health records/registries, medico-administrative). Fifty percent of studies evaluated trajectory groups as exposures for a subsequent clinical outcome, while 24% used predictive models of group membership and 5% used both. Regarding the quality of reporting, trajectory groups were adequately presented, however many studies failed to report important decisions made for the trajectory group identification. Conclusion Although LCGA were preferred, the contexts of utilization were diverse and unrelated to the type of methods. We recommend future authors to clearly report the decisions made regarding trajectory groups identification.
Collapse
Affiliation(s)
- Sarah O'Connor
- Research Centre, Institut universitaire de Cardiologie et Pneumologie de Québec-Université Laval (IUCPQ-UL), 2725 Ch. Ste-Foy, Quebec City, Quebec, G1V 4G5, Canada
- Faculty of Pharmacy, Université Laval, Ferdinand Vandry Pavillon, 1050 de La Médecine Avenue, Quebec City, Quebec, G1V 0A6, Canada
| | - Claudia Blais
- Faculty of Pharmacy, Université Laval, Ferdinand Vandry Pavillon, 1050 de La Médecine Avenue, Quebec City, Quebec, G1V 0A6, Canada
- Bureau D'information et D’études en Santé des Populations, Institut National de Santé Publique Du Québec, 945, Wolfe Avenue, Quebec City, Quebec, G1V 5B3, Canada
| | - Miceline Mésidor
- Department of Social and Preventive Medicine, Faculty of Medicine, Université Laval, Ferdinand Vandry Pavillon, 1050 de La Médecine Avenue, Quebec City, Quebec, G1V 0A6, Canada
- Research Centre, CHU de Québec – Université Laval, 2400 D'Estimauville Avenue, Québec, QC, G1E 6W2, Canada
| | - Denis Talbot
- Department of Social and Preventive Medicine, Faculty of Medicine, Université Laval, Ferdinand Vandry Pavillon, 1050 de La Médecine Avenue, Quebec City, Quebec, G1V 0A6, Canada
- Research Centre, CHU de Québec – Université Laval, 2400 D'Estimauville Avenue, Québec, QC, G1E 6W2, Canada
| | - Paul Poirier
- Research Centre, Institut universitaire de Cardiologie et Pneumologie de Québec-Université Laval (IUCPQ-UL), 2725 Ch. Ste-Foy, Quebec City, Quebec, G1V 4G5, Canada
- Faculty of Pharmacy, Université Laval, Ferdinand Vandry Pavillon, 1050 de La Médecine Avenue, Quebec City, Quebec, G1V 0A6, Canada
| | - Jacinthe Leclerc
- Research Centre, Institut universitaire de Cardiologie et Pneumologie de Québec-Université Laval (IUCPQ-UL), 2725 Ch. Ste-Foy, Quebec City, Quebec, G1V 4G5, Canada
- Faculty of Pharmacy, Université Laval, Ferdinand Vandry Pavillon, 1050 de La Médecine Avenue, Quebec City, Quebec, G1V 0A6, Canada
- Department of Nursing, Université Du Québec à Trois-Rivières, 3351 des Forges Boulevard, Trois-Rivières, Quebec, G8Z 4M3, Canada
| |
Collapse
|
3
|
Shen F, Li L. Backward joint model and dynamic prediction of survival with multivariate longitudinal data. Stat Med 2021; 40:4395-4409. [PMID: 34018218 DOI: 10.1002/sim.9037] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/21/2020] [Revised: 04/21/2021] [Accepted: 05/01/2021] [Indexed: 11/05/2022]
Abstract
An important approach to dynamic prediction of time-to-event outcomes using longitudinal data is based on modeling the joint distribution of longitudinal and time-to-event data. The widely used joint model for this purpose is the shared random effect model. Presumably, adding more longitudinal predictors improves the predictive accuracy. However, the shared random effect model can be computationally difficult or prohibitive when a large number of longitudinal variables are used. In this paper, we study an alternative way of modeling the joint distribution of longitudinal and time-to-event data. Under this formulation, the log-likelihood involves no more than one-dimensional integration, regardless of the number of longitudinal variables in the model. Therefore, this model is particularly suitable in dynamic prediction problems with large number of longitudinal predictors. The model fitting can be implemented with tractable and stable computation by using a combination of pseudo maximum likelihood estimation, Expectation-Maximization algorithm, and convex optimization. We evaluate the proposed methodology and its predictive accuracy with varying number of longitudinal variables using simulations and data from a primary biliary cirrhosis study.
Collapse
Affiliation(s)
- Fan Shen
- Department of Biostatistics and Data Science, The University of Texas School of Public Health, Dallas, Texas, USA.,Department of Biostatistics, The University of Texas MD Anderson Cancer Center, Houston, Texas, USA
| | - Liang Li
- Department of Biostatistics, The University of Texas MD Anderson Cancer Center, Houston, Texas, USA
| |
Collapse
|
4
|
Robitzsch A. Regularized Latent Class Analysis for Polytomous Item Responses: An Application to SPM-LS Data. J Intell 2020; 8:E30. [PMID: 32823949 PMCID: PMC7555561 DOI: 10.3390/jintelligence8030030] [Citation(s) in RCA: 3] [Impact Index Per Article: 0.8] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/09/2020] [Revised: 07/26/2020] [Accepted: 08/10/2020] [Indexed: 11/28/2022] Open
Abstract
The last series of Raven's standard progressive matrices (SPM-LS) test was studied with respect to its psychometric properties in a series of recent papers. In this paper, the SPM-LS dataset is analyzed with regularized latent class models (RLCMs). For dichotomous item response data, an alternative estimation approach based on fused regularization for RLCMs is proposed. For polytomous item responses, different alternative fused regularization penalties are presented. The usefulness of the proposed methods is demonstrated in a simulated data illustration and for the SPM-LS dataset. For the SPM-LS dataset, it turned out the regularized latent class model resulted in five partially ordered latent classes. In total, three out of five latent classes are ordered for all items. For the remaining two classes, violations for two and three items were found, respectively, which can be interpreted as a kind of latent differential item functioning.
Collapse
Affiliation(s)
- Alexander Robitzsch
- IPN—Leibniz Institute for Science and Mathematics Education, D-24098 Kiel, Germany;
- Centre for International Student Assessment (ZIB), D-24098 Kiel, Germany
| |
Collapse
|
5
|
Raghavan S, Liu WG, Berkowitz SA, Barón AE, Plomondon ME, Maddox TM, Reusch JEB, Ho PM, Caplan L. Association of Glycemic Control Trajectory with Short-Term Mortality in Diabetes Patients with High Cardiovascular Risk: a Joint Latent Class Modeling Study. J Gen Intern Med 2020; 35:2266-2273. [PMID: 32333313 PMCID: PMC7403288 DOI: 10.1007/s11606-020-05848-5] [Citation(s) in RCA: 6] [Impact Index Per Article: 1.5] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 09/03/2019] [Revised: 02/29/2020] [Accepted: 04/08/2020] [Indexed: 10/24/2022]
Abstract
BACKGROUND The relationship between risk factor or biomarker trajectories and contemporaneous short-term clinical outcomes is poorly understood. In diabetes patients, it is unknown whether hemoglobin A1c (HbA1c) trajectories are associated with clinical outcomes and can inform care in scenarios in which a single HbA1c is uninformative, for example, after a diagnosis of coronary artery disease (CAD). OBJECTIVE To compare associations of HbA1c trajectories and single HbA1c values with short-term mortality in diabetes patients evaluated for CAD DESIGN: Retrospective observational cohort study PARTICIPANTS: Diabetes patients (n = 7780) with and without angiographically defined CAD MAIN MEASURES: We used joint latent class mixed models to simultaneously fit HbA1c trajectories and estimate association with 2-year mortality after cardiac catheterization, adjusting for clinical and demographic covariates. KEY RESULTS Three HBA1c trajectory classes were identified: individuals with stable glycemia (class A; n = 6934 [89%]; mean baseline HbA1c 6.9%), with declining HbA1c (class B; n = 364 [4.7%]; mean baseline HbA1c 11.6%), and with increasing HbA1c (class C; n = 482 [6.2%]; mean baseline HbA1c 8.5%). HbA1c trajectory class was associated with adjusted 2-year mortality (3.0% [95% CI 2.8, 3.2] for class A, 3.1% [2.1, 4.2] for class B, and 4.2% [3.4, 4.9] for class C; global P = 0.047, P = 0.03 comparing classes A and C, P > 0.05 for other pairwise comparisons). Baseline HbA1c was not associated with 2-year mortality (P = 0.85; hazard ratios 1.01 [0.96, 1.06] and 1.02 [0.95, 1.10] for HbA1c 7-9% and ≥ 9%, respectively, relative to HbA1c < 7%). The association between HbA1c trajectories and mortality did not differ between those with and without CAD (interaction P = 0.1). CONCLUSIONS In clinical settings where single HbA1c measurements provide limited information, HbA1c trajectories may help stratify risk of complications in diabetes patients. Joint latent class modeling provides a generalizable approach to examining relationships between biomarker trajectories and clinical outcomes in the era of near-universal adoption of electronic health records.
Collapse
Affiliation(s)
- Sridharan Raghavan
- Department of Veterans Affairs, Eastern Colorado Healthcare System, Aurora, CO, USA. .,Division of Hospital Medicine, Department of Medicine, University of Colorado School of Medicine, Aurora, CO, USA. .,Colorado Cardiovascular Outcomes Research Consortium, Aurora, CO, USA. .,Rocky Mountain Regional VA Medical Center Medicine Service (111), 1700 North Wheeling Street, Aurora, CO, 80045, USA.
| | - Wenhui G Liu
- Department of Veterans Affairs, Eastern Colorado Healthcare System, Aurora, CO, USA
| | - Seth A Berkowitz
- Division of General Medicine & Clinical Epidemiology, University of North Carolina School of Medicine, Chapel Hill, NC, USA
| | - Anna E Barón
- Department of Veterans Affairs, Eastern Colorado Healthcare System, Aurora, CO, USA.,Department of Biostatistics and Informatics, Colorado School of Public Health, Aurora, CO, USA
| | - Mary E Plomondon
- Department of Veterans Affairs, Eastern Colorado Healthcare System, Aurora, CO, USA
| | - Thomas M Maddox
- Division of Cardiology, Washington University School of Medicine, St. Louis, MO, USA
| | - Jane E B Reusch
- Department of Veterans Affairs, Eastern Colorado Healthcare System, Aurora, CO, USA.,Division of Endocrinology, Metabolism, and Diabetes, University of Colorado School of Medicine, Aurora, CO, USA
| | - P Michael Ho
- Department of Veterans Affairs, Eastern Colorado Healthcare System, Aurora, CO, USA.,Division of Cardiology, University of Colorado School of Medicine, Aurora, CO, USA
| | - Liron Caplan
- Department of Veterans Affairs, Eastern Colorado Healthcare System, Aurora, CO, USA.,Division of Rheumatology, University of Colorado School of Medicine, Aurora, CO, USA
| |
Collapse
|
6
|
Peng C, Wang J, Asante I, Louie S, Jin R, Chatzi L, Casey G, Thomas DC, Conti DV. A latent unknown clustering integrating multi-omics data (LUCID) with phenotypic traits. Bioinformatics 2019; 36:842-850. [PMID: 31504184 PMCID: PMC7986585 DOI: 10.1093/bioinformatics/btz667] [Citation(s) in RCA: 12] [Impact Index Per Article: 2.4] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/14/2019] [Revised: 08/04/2019] [Accepted: 08/21/2019] [Indexed: 01/31/2023] Open
Abstract
MOTIVATION Epidemiologic, clinical and translational studies are increasingly generating multiplatform omics data. Methods that can integrate across multiple high-dimensional data types while accounting for differential patterns are critical for uncovering novel associations and underlying relevant subgroups. RESULTS We propose an integrative model to estimate latent unknown clusters (LUCID) aiming to both distinguish unique genomic, exposure and informative biomarkers/omic effects while jointly estimating subgroups relevant to the outcome of interest. Simulation studies indicate that we can obtain consistent estimates reflective of the true simulated values, accurately estimate subgroups and recapitulate subgroup-specific effects. We also demonstrate the use of the integrated model for future prediction of risk subgroups and phenotypes. We apply this approach to two real data applications to highlight the integration of genomic, exposure and metabolomic data. AVAILABILITY AND IMPLEMENTATION The LUCID method is implemented through the LUCIDus R package available on CRAN (https://CRAN.R-project.org/package=LUCIDus). SUPPLEMENTARY INFORMATION Supplementary materials are available at Bioinformatics online.
Collapse
Affiliation(s)
- Cheng Peng
- Department of Preventive Medicine, Keck School of Medicine, Los Angeles, CA 90089, USA
| | - Jun Wang
- Department of Preventive Medicine, Keck School of Medicine, Los Angeles, CA 90089, USA
| | - Isaac Asante
- Department of Clinical Pharmacy, School of Pharmacy, University of Southern California, Los Angeles, CA 90089, USA
| | - Stan Louie
- Department of Clinical Pharmacy, School of Pharmacy, University of Southern California, Los Angeles, CA 90089, USA
| | - Ran Jin
- Department of Preventive Medicine, Keck School of Medicine, Los Angeles, CA 90089, USA
| | - Lida Chatzi
- Department of Preventive Medicine, Keck School of Medicine, Los Angeles, CA 90089, USA
| | - Graham Casey
- Center for Public Health Genomics, Department of Public Health Sciences, University of Virginia, Charlottesville, VA 22908, USA
| | - Duncan C Thomas
- Department of Preventive Medicine, Keck School of Medicine, Los Angeles, CA 90089, USA
| | | |
Collapse
|