Application fields
Neurological Diseases
Technologies
Medical Statistics ,Machine Learning (ML)
Goals
Diagnosis of neurological diseases is a growing concern and one of the most difficult challenges for modern medicine due to high social impact and costs. To study MS epidemiology, it is preferable to use population-based studies. These studies are useful also to describe comorbidities, care pathways, the burden of this disease and to plan the management strategies and resource allocation necessary to cope with it. The presence of comorbid disease is a critical issue for clinicians given the breadth of adverse impacts with which it is associated. Especially in MS, comorbidity is associated with a longer delay between MS symptom onset and diagnosis, more severe disability at diagnosis even after accounting for diagnostic delays, greater disability progression, increased health-care utilization, and higher mortality . Therefore, using routinely collected data to evaluate the comorbidity is a key importance to manage MS. An example is a study conducted by Palladino et al. that showed an association between multiple sclerosis and increased risk of macrovascular disease by using a population-based cohort of 84,823 people with or without multiple sclerosis. Factors affecting the distribution of MS, that may help determine what causes the disease include age, sex, race, genetics, and geographical location. Also in this setting, big data have been used, in fact, a retrospective cohort of more than 9 million person-years has been used to calculate the Incidence of multiple sclerosis in multiple racial and ethnic group . Regarding the geographical location factor, a population-based cohort study about 6,6 million has been conducted to assess the association between residential proximity to major roadways and the incidence of MS in Ontario, Canada . The above demonstrates the usefulness of routinely collected data to investigate about MS in order to improve patient health and quality of life and to optimize health resource use.
Merging of health data-sets Nowadays, extracting information about a certain matter, i.e. merging data-sets belonging to different settings is the new challenge in current research. Especially in a health care setting, integrating clinical data with financial and operational data, is important to get a more complete picture of care and for improvement value of the health system. The paper by Vemulapalli et al. highlights the possibilities offered by merging registries with large administrative databases. The strength of registries is to comprise relevant variables for the considered disease and that are not available when using a clinical database. In fact, administrative databases provide detailed information on hospitalizations and individual health care costs, which are otherwise difficult to collect during patient follow-up . Merging of the different datasets has been used also in multiple sclerosis setting in the Campania region. A linkage between the administrative and clinical datasets has been utilized for MS case-finding algorithm validation . The dataset was created by merging different data sources of the Campania Region. In particular, the cohort comprised three database: Hospital Discharge Record database, Regional Drug Prescription database and Outpatient database. This heterogeneous database, representing 10% of the Italian population, could be utilized to validate healthcare economics, public health actions existing or pending implemented and clinical scales such as the Expanded Disability Status Scale (EDSS) for multiple sclerosis.
Activities
- Explore geographical variations accounting for undetected cases: Understanding the real epidemiology condition is an essential factor for implement public health initiatives focusing on health promotion.
- Validation of clinical scales: Measurement of a condition as variable as MS is notoriously difficult but the need for evidence-based decisions has highlighted the importance of the development of adequate measures.
- Economic evaluation of health resource: An economic evaluation of healthcare resources helps us to assure the efficient use of them.
- Public health policy evaluation : Integrated care models have been proposed to improve outcomes and optimize resource allocation for the care of individuals with complex chronic conditions.
Challenges
In research scientific setting, the re-adaptability of research methods is an important concept. The challenge is to provide elasticity to an analysis method to may contextualize it in several settings optimizing its use. Therefore, developed methodology to answer the about propose aims could be applied to other condition to develop and more comprehensive analysis of the public healthcare.
Bibliography
- L. Brodie, M. Greaves and J.A. Hendler, Databases and AI: The Twain Just Met, 2011 STI Semantic Summit,Riga, Latvia, July 6-8, 2011
- Christian Bizer, Peter A Boncz , Michael L. Brodie , Orri Erling The meaningful use of big data: four perspectives – four challenges, ACM SIGMOD RecordJanuary 2012
- Benke K.K. Uncertainties in Big Data when using Internet Surveillance Tools and Social Media for determination of Patterns in Disease Incidence. JAMA Ophthalmol. 2017;135:402. doi: 10.1001/jamaophthalmol.2017.0138.
- Rubin R.A. Precision Medicine Approach to Clinical Trials. JAMA. 2016;316:1953-1955. doi:10.1001/jama.2016.12137.
- Rubin R. Precision Medicine: The Future or Simply Politics? JAMA. 2015;313:1089-1091. doi:10.1001/jama.2015.0957.
- Ashley E.A. Towards Precision Medicine. Nat. Rev. Genet. 2016;17:507-522. doi: 10.1038/nrg.2016.86.
- Gulshan V., Peng L., Coram M., Stumpe M.C., Wu D., Narayanaswamy A., Venugopalan S., Widner K., Madams T., Cuadros J., et al. Development and validation of a deep learning algorithm for detection of diabetic retinopathy in retinal fundus photographs.
- David W. Bates, Suchi Saria, Lucila Ohno-Machado, Anand Shah, and Gabriel Escobar Big Data In Health Care: Using Analytics To Identify And Manage High-Risk And High-Cost Patients.
- Travis B. Murdoch, MD, MSc; Allan S. Detsky, MD, PhD The Inevitable Application of Big Data to Health Care
- Kalincik, T.; Butzkueven, H. Observational data: Understanding the real MS world. Mult. Scler. J. 2016,22, 1642-1648.
- Birnbaum, H.G.; Cremieux, P.Y.; Greenberg, P.E.; LeLorier, J.; Ostrander, J.A.; Venditti, L. Using healthcare claims data for outcomes research and pharmacoeconomic analyses. Pharmacoeconomics 1999, 16, 1-8.
- Marcello Moccia , Vincenzo Brescia Morra , Roberta Lanzillo , Ilaria Loperto ,Roberta Giordana , Maria Grazia Fumo , Martina Petruzzo , Nicola Capasso , Maria Triassi2,Maria Pia Sormani and Raffaele Palladino Multiple Sclerosis in the Campania Region (South Italy): Algorithm Validation and 2015-2017 Prevalence
- BergamaschiaC.MontomolibE.CandeloroaM.C.MontibR.CioccalebL.BernardinellibP.FratinocV.Cosiathe PREMS Bayesian mapping of multiple sclerosis prevalence in the province of Pavia, northern Italy
- Eleonora Cocco , Claudia Sardu, Rita Massa, Elena Mamusa, Luigina Musu, Paola Ferrigno, Maurizio Melis, Cristina Montomoli, Virginia Ferretti, Giancarlo Coghe, Giuseppe Fenu, Jessica Frau, Lorena Lorefice, Nicola Carboni, Paolo Contu, Maria G Marrosu. Epidemiology of Multiple Sclerosis in South-Western Sardinia
- Martina Petruzzo, Raffaele Palladino, Antonio Nardone, Roberta Lanzillo, Vincenzo Brescia Morra, Marcello Moccia. The impact of diagnostic criteria and treatments on the 20-year costs for treating relapsing-remitting multiple sclerosis.
- Raffaele Palladino, Ruth Ann Marrie, ; Azeem Majeed, ; et al, Evaluating the Risk of Macrovascular Events and Mortality Among People With Multiple Sclerosis in England