ABSTRACT
To achieve a single fully harmonised research data set suitable for analysis from data collected at multiple sites requires not only semantic integration of collection concepts and convergence onto single collection units, but harmonisation of data collection processes. We describe our experience of identifying harmonisation challenges in the Precision ALS project, with particular focus on process alignment challenges in a multi-site multi-national research data collection project.
Subject(s)
Data Collection , Humans , Amyotrophic Lateral Sclerosis/therapy , Biomedical ResearchABSTRACT
OBJECTIVES: This study aims to describe the data structure and harmonisation process, explore data quality and define characteristics, treatment, and outcomes of patients across six federated antineutrophil cytoplasmic antibody-associated vasculitis (AAV) registries. METHODS: Through creation of the vasculitis-specific Findable, Accessible, Interoperable, Reusable, VASCulitis ontology, we harmonised the registries and enabled semantic interoperability. We assessed data quality across the domains of uniqueness, consistency, completeness and correctness. Aggregated data were retrieved using the semantic query language SPARQL Protocol and Resource Description Framework Query Language (SPARQL) and outcome rates were assessed through random effects meta-analysis. RESULTS: A total of 5282 cases of AAV were identified. Uniqueness and data-type consistency were 100% across all assessed variables. Completeness and correctness varied from 49%-100% to 60%-100%, respectively. There were 2754 (52.1%) cases classified as granulomatosis with polyangiitis (GPA), 1580 (29.9%) as microscopic polyangiitis and 937 (17.7%) as eosinophilic GPA. The pattern of organ involvement included: lung in 3281 (65.1%), ear-nose-throat in 2860 (56.7%) and kidney in 2534 (50.2%). Intravenous cyclophosphamide was used as remission induction therapy in 982 (50.7%), rituximab in 505 (17.7%) and pulsed intravenous glucocorticoid use was highly variable (11%-91%). Overall mortality and incidence rates of end-stage kidney disease were 28.8 (95% CI 19.7 to 42.2) and 24.8 (95% CI 19.7 to 31.1) per 1000 patient-years, respectively. CONCLUSIONS: In the largest reported AAV cohort-study, we federated patient registries using semantic web technologies and highlighted concerns about data quality. The comparison of patient characteristics, treatment and outcomes was hampered by heterogeneous recruitment settings.
Subject(s)
Anti-Neutrophil Cytoplasmic Antibody-Associated Vasculitis , Granulomatosis with Polyangiitis , Microscopic Polyangiitis , Humans , Granulomatosis with Polyangiitis/drug therapy , Granulomatosis with Polyangiitis/epidemiology , Granulomatosis with Polyangiitis/complications , Data Accuracy , Anti-Neutrophil Cytoplasmic Antibody-Associated Vasculitis/drug therapy , Anti-Neutrophil Cytoplasmic Antibody-Associated Vasculitis/epidemiology , Anti-Neutrophil Cytoplasmic Antibody-Associated Vasculitis/complications , Microscopic Polyangiitis/drug therapy , Microscopic Polyangiitis/epidemiology , Antibodies, Antineutrophil Cytoplasmic , Registries , Information Storage and RetrievalABSTRACT
Objective: Despite the fact that user engagement is critical to the efficacy of mobile health (mHealth) interventions in the Global South, many of these interventions lack user engagement features. This is because sociotechnical aspects of such initiatives are frequently ignored during the design, development, and implementation stages. This research highlighted the importance of considering sociotechnical factors when developing mHealth apps. The intended users for the mHealth technologies in this study are care professionals. Materials and Methods: Five semi-structured interviews and a pilot interview were conducted to identify user engagement facilitators and barriers. The interview data were analysed using NVivo. The Capability, Opportunity, Motivation - Behaviour (COM-B) model is then used to map the facilitators and barriers to mHealth app engagement, allowing researchers to better understand how users engage/disengage with mHealth apps. Results and Discussion: Capability facilitators included features that assist users in learning more about the app (e.g. a user manual and statistical data) as well as features that assist users in developing a routine. The lack of app skills and cognitive overload limit capability. While social connectedness and offline functionality were identified as facilitators of user engagement, non-user-friendly design and cultural dimensions were identified as barriers. Early user engagement and rewards were identified as motivational facilitators that influence user engagement. Furthermore, perceived non-utility and a lack of encouragement were identified as motivational barriers to engagement. Conclusion: Several factors were discovered across all COM-B model components that could be used to develop more engaging mHealth apps. Adopting a techno-centric approach that ignores sociotechnical factors can reduce user engagement. The design process engagement enhancement system (DECENT) framework was proposed based on the findings.
ABSTRACT
Amyotrophic Lateral Sclerosis (ALS) is an incurable neurodegenerative condition. Despite significant advances in pre-clinical models that enhance understanding of disease pathobiology, translation of candidate drugs to effective human therapies has been disappointing. There is increasing recognition of the need for a precision medicine approach toward drug development, as many failures in translation can be attributed in part to disease heterogeneity in humans. PRECISION-ALS is an academic industry collaboration between clinicians, Computer Scientists, Information engineers, technologists, data scientists and industry partners that will address the key clinical, computational, data science and technology associated research questions to generate a sustainable precision medicine based approach toward new drug development. Using extant and prospectively collected population based clinical data across nine European sites, PRECISION-ALS provides a General Data Protection Regulation (GDPR) compliant framework that seamlessly collects, processes and analyses research-quality multimodal and multi-sourced clinical, patient and caregiver journey, digitally acquired data through remote monitoring, imaging, neuro-electric-signaling, genomic and biomarker datasets using machine learning and artificial intelligence. PRECISION-ALS represents a first-in-kind modular transferable pan-European ICT framework for ALS that can be easily adapted to other regions that face similar precision medicine related challenges in multimodal data collection and analysis.
Subject(s)
Amyotrophic Lateral Sclerosis , Humans , Amyotrophic Lateral Sclerosis/diagnosis , Amyotrophic Lateral Sclerosis/epidemiology , Amyotrophic Lateral Sclerosis/genetics , Artificial Intelligence , Biomarkers , Machine LearningABSTRACT
BACKGROUND: The aetiology of ANCA-associated vasculitis (AAV) and triggers of relapse are poorly understood. Vitamin D (vitD) is an important immunomodulator, potentially responsible for the observed latitudinal differences between granulomatous and non-granulomatous AAV phenotypes. A narrow ultraviolet B spectrum induces vitD synthesis (vitD-UVB) via the skin. We hypothesised that prolonged periods of low ambient UVB (and by extension vitD deficiency) are associated with the granulomatous form of the disease and an increased risk of AAV relapse. METHODS: Patients with AAV recruited to the Irish Rare Kidney Disease (RKD) (n = 439) and UKIVAS (n = 1961) registries were studied. Exposure variables comprised latitude and measures of ambient vitD-UVB, including cumulative weighted UVB dose (CW-D-UVB), a well-validated vitD proxy. An n-of-1 study design was used to examine the relapse risk using only the RKD dataset. Multi-level models and logistic regression were used to examine the effect of predictors on AAV relapse risk, phenotype and serotype. RESULTS: Residential latitude was positively correlated (OR 1.41, 95% CI 1.14-1.74, p = 0.002) and average vitD-UVB negatively correlated (0.82, 0.70-0.99, p = 0.04) with relapse risk, with a stronger effect when restricting to winter measurements (0.71, 0.57-0.89, p = 0.002). However, these associations were not restricted to granulomatous phenotypes. We observed no clear relationship between latitude, vitD-UVB or CW-D-UVB and AAV phenotype or serotype. CONCLUSION: Our findings suggest that low winter ambient UVB and prolonged vitD status contribute to AAV relapse risk across all phenotypes. However, the development of a granulomatous phenotype does not appear to be directly vitD-mediated. Further research is needed to determine whether sufficient vitD status would reduce relapse propensity in AAV.
Subject(s)
Anti-Neutrophil Cytoplasmic Antibody-Associated Vasculitis , Vitamin D Deficiency , Anti-Neutrophil Cytoplasmic Antibody-Associated Vasculitis/epidemiology , Chronic Disease , Humans , Recurrence , Ultraviolet Rays/adverse effects , Vitamin DABSTRACT
Rare disease data is often fragmented within multiple heterogeneous siloed regional disease registries, each containing a small number of cases. These data are particularly sensitive, as low subject counts make the identification of patients more likely, meaning registries are not inclined to share subject level data outside their registries. At the same time access to multiple rare disease datasets is important as it will lead to new research opportunities and analysis over larger cohorts. To enable this, two major challenges must therefore be overcome. The first is to integrate data at a semantic level, so that it is possible to query over registries and return results which are comparable. The second is to enable queries which do not take subject level data from the registries. To meet the first challenge, this paper presents the FAIRVASC ontology to manage data related to the rare disease anti-neutrophil cytoplasmic antibody (ANCA) associated vasculitis (AAV), which is based on the harmonisation of terms in seven European data registries. It has been built upon a set of key clinical questions developed by a team of experts in vasculitis selected from the registry sites and makes use of several standard classifications, such as Systematized Nomenclature of Medicine - Clinical Terms (SNOMED-CT) and Orphacode. It also presents the method for adding semantic meaning to AAV data across the registries using the declarative Relational to Resource Description Framework Mapping Language (R2RML). To meet the second challenge a federated querying approach is presented for accessing aggregated and pseudonymized data, and which supports analysis of AAV data in a manner which protects patient privacy. For additional security the federated querying approach is augmented with a method for auditing queries (and the uplift process) using the provenance ontology (PROV-O) to track when queries and changes occur and by whom. The main contribution of this work is the successful application of semantic web technologies and federated queries to provide a novel infrastructure that can readily incorporate additional registries, thus providing access to harmonised data relating to unprecedented numbers of patients with rare disease, while also meeting data privacy and security concerns.
Subject(s)
Semantic Web , Vasculitis , Humans , Rare Diseases , Registries , Systematized Nomenclature of MedicineABSTRACT
We present a knowledge-base to represent collated infectious disease risk (IDR) knowledge. The knowledge is about personal and contextual risk of contracting an infectious disease obtained from declarative sources (e.g. Atlas of Human Infectious Diseases). Automated prediction requires encoding this knowledge in a form that can produce risk probabilities (e.g. Bayesian Network - BN). The knowledge-base presented in this paper feeds an algorithm that can auto-generate the BN. The knowledge from 234 infectious diseases was compiled. From this compilation, we designed an ontology and five rule types for modelling IDR knowledge in general. The evaluation aims to assess whether the knowledge-base structure, and its application to three disease-country contexts, meets the needs of personalized IDR prediction system. From the evaluation results, the knowledge-base conforms to the system's purpose: personalization of infectious disease risk.
Subject(s)
Bayes Theorem , Knowledge Bases , Risk , Algorithms , Humans , ProbabilityABSTRACT
There is an ongoing challenge as to how best manage and understand 'big data' in precision medicine settings. This paper describes the potential for a Linked Data approach, using a Resource Description Framework (RDF) model, to combine multiple datasets with temporal and spatial elements of varying dimensionality. This "AVERT model" provides a framework for converting multiple standalone files of various formats, from both clinical and environmental settings, into a single data source. This data source can thereafter be queried effectively, shared with outside parties, more easily understood by multiple stakeholders using standardized vocabularies, incorporating provenance metadata and supporting temporo-spatial reasoning. The approach has further advantages in terms of data sharing, security and subsequent analysis. We use a case study relating to anti-Glomerular Basement Membrane (GBM) disease, a rare autoimmune condition, to illustrate a technical proof of concept for the AVERT model.
ABSTRACT
Business analytics (BA) is increasingly being utilised by radiology departments to analyse and present data. It encompasses statistical analysis, forecasting and predictive modelling and is used as an umbrella term for decision support and business intelligence systems. The primary aim of this study was to determine whether utilising BA technologies could contribute towards improved decision support and resource management within radiology departments. A set of information technology requirements were identified with key stakeholders, and a prototype BA software tool was designed, developed and implemented. A qualitative evaluation of the tool was carried out through a series of semi-structured interviews with key stakeholders. Feedback was collated, and emergent themes were identified. The results indicated that BA software applications can provide visibility of radiology performance data across all time horizons. The study demonstrated that the tool could potentially assist with improving operational efficiencies and management of radiology resources.
Subject(s)
Decision Support Techniques , Radiology Department, Hospital , Software , Qualitative Research , Radiography , Radiology , Radiology Information SystemsABSTRACT
RATIONALE, AIMS AND OBJECTIVES: Timely access to general practitioner (GP) care is a recognized strategy to address avoidable hospitalization. Little is known about patients seeking planned (decided ahead) and unplanned (decided on day) GP visits. The Patient Journey Record System (PaJR) provides a biopsychosocial real-time monitoring and support service to chronically ill and older people over 65 who may be at risk of an avoidable hospital admission. This study aims to describe reported profiles associated with planned and unplanned GP visits during the week in the PaJR database of regular outbound phone calls made by Care Guides to multi-morbid older patients. METHODS: One hundred fifty consecutive patients with one or more chronic condition (including chronic obstructive pulmonary disease, heart/vascular disease, heart failure and/or diabetes), one or more hospital admission in previous year, and consecutively recruited from hospital discharge, out-of-hour care and GP practices comprised the study sample. Using a semistructured script, Care Guides telephoned the patients approximately every 3 week days, and entered call data into the PaJR database in 2011. The PaJR project identified and prompted unplanned visits according to its algorithms. Logistic regression modelling and descriptive statistics identified significant predictors of planned and unplanned visits and patterns of GP visits on weekdays reported in calls. RESULTS: In 5096 telephone calls, unplanned versus planned GP visits were predicted by change in health state, significant symptom concerns, poor self-rated health, bodily pain and concerns about caregiver or intimates. Calls not reporting visits had significantly fewer of these features. Planned visits were associated with general and medication concerns, reduced social participation and feeling down. Planned visits were highest on Monday and trended downwards to Fridays. Unplanned visits were reported at the same rate each weekday and more frequently when the interval between calls was ≥3 days. The PaJR project Care Guides advised patients to make unplanned visits in 6.3% of calls and advised planned GP visits in 2.5% of calls. CONCLUSION: Unplanned GP visits consistently indicated a significant change to worse health with planned visits presenting less acuity in this study of older multi-morbid patients in general practice, when monitored by regular calls at about every 3 days. The PaJR study actively prompted GP visits according to its algorithms. Assessing and predicting acuity in older multi-morbid patients appears to be a promising strategy to improve access to primary care, and thus to reducing avoidable hospital utilization. Further research is needed to investigate the topic on a wider scale.
Subject(s)
Appointments and Schedules , Comorbidity , Databases, Factual , General Practice , Health Services/statistics & numerical data , Aged , Aged, 80 and over , Chronic Disease , Decision Making , Female , Health Services Accessibility , Humans , Ireland , Male , Middle Aged , Regression AnalysisABSTRACT
BACKGROUND: The Patient Journey Record system (PaJR) is an application of a complex adaptive chronic care model in which early detection of adverse changes in patient biopsychosocial trajectories prompts tailored care, constitute the cornerstone of the model. AIMS: To evaluate the PaJR system's impact on care and the experiences of older people with chronic illness, who were at risk of repeat admissions over 12 months. DESIGN: Community-based cohort study - random assignment into intervention and usual care group, with process and outcome evaluation. STUDY POPULATION: Adult and older patients with multiple morbidity, one or more chronic diseases with one or more overnight hospitalizations, and seven or more general practice visits in the past 6 months. COMPLEX INTERVENTION: PaJR lay care guides/advocates call patients and their caregivers. The care guides summarize their semi-structured conversations about health concerns and well-being. Predictive modelling and rules-based algorithms trigger alerts in relation to online call summaries. Alerts are acted upon according to agreed guidelines. ANALYSIS: Descriptive and comparative statistics. OUTCOMES: Impact on unplanned emergency ambulatory care sensitive admissions (ACSC) with an overnight stay; sensitivity of alerts and predictions; rates of care guides-supported activities. FINDINGS: Five part-time lay care guides and a care manager monitored 153 intervention patients for 500 person months with 5050 phone calls. The 153 patients in the intervention group were comparable to the 61 controls. The intervention group reported in 50% of calls that their health limited their social activities; and one-third of calls reported immediate health concerns. Predictive analytics were highly sensitive to risk of hospitalization. ACSC admissions were reduced by 50% compared to controls across the sites. DISCUSSION: The initial implementation of a complex patient-centred adaptive chronic care model using lay care guides, supported by machine learning, appeared sensitive to risk of hospitalization and capable of stabilizing illness journeys in older patients with multi-morbidity. CONCLUSION: Actions based on alerts produced in this study appeared to significantly reduce hospitalizations. This paves the way for further testing of the model.
Subject(s)
Chronic Disease/therapy , Disease Management , Health Services/statistics & numerical data , Aged , Aged, 80 and over , Algorithms , Artificial Intelligence , Clinical Protocols , Cohort Studies , Emergency Service, Hospital/statistics & numerical data , Environment , Female , Health Promotion , Health Status , Humans , Male , Middle Aged , Social SupportABSTRACT
BACKGROUND: In this paper, we give an overview of methadone treatment in Ireland and outline the rationale for designing an electronic health record (EHR) with extensibility, interoperability and decision support functionality. Incorporating several international standards, a conceptual model applying a problem orientated approach in a hierarchical structure has been proposed for building the EHR. METHODS: A set of archetypes has been designed in line with the current best practice and clinical guidelines which guide the information-gathering process. A web-based data entry system has been implemented, incorporating elements of the paper-based prescription form, while at the same time facilitating the decision support function. RESULTS: The use of archetypes was found to capture the ever changing requirements in the healthcare domain and externalises them in constrained data structures. The solution is extensible enabling the EHR to cover medicine management in general as per the programme of the HRB Centre for Primary Care Research. CONCLUSIONS: The data collected via this Irish system can be aggregated into a larger dataset, if necessary, for analysis and evidence-gathering, since we adopted the openEHR standard. It will be later extended to include the functionalities of prescribing drugs other than methadone along with the research agenda at the HRB Centre for Primary Care Research in Ireland.
Subject(s)
Decision Support Techniques , Drug Therapy, Computer-Assisted , Electronic Health Records/organization & administration , Methadone/therapeutic use , Electronic Health Records/standards , Episode of Care , Humans , Medical Record Linkage/methods , Practice Guidelines as Topic , Semantics , Systematized Nomenclature of Medicine , User-Computer InterfaceABSTRACT
Diabetes Mellitus represents a heterogeneous group of metabolic disorders and its complications are associated with increased morbidity and early mortality. The explosive increase of the number of patients with diabetes, and several recommendations from milestone studies such as the United Kingdom Prospective Study (UKPDS), the Diabetes Prevention Program (DPP) and the Finnish Prevention Study place further constraint on an already overstretched service. This paper examines the care flow process in the current diabetes management and explores the potential role of new technologies and methodologies currently available to support the highly complex clinical requirements. The paper proposes a concept of e-clinic as an electronic triage system with integrated healthcare record, embedded with a decision support system to analyse patients' data and provide a continuous monitoring system according to agreed clinical guidelines.
Subject(s)
Decision Support Systems, Clinical , Diabetes Mellitus, Type 2/therapy , Patient Care Management/methods , Triage/methods , Humans , Internet , Medical Records Systems, Computerized , Patient Care Management/organization & administration , Systems Integration , TelemedicineABSTRACT
Clinical guidelines can be represented using models, such as GLIF, specifically designed for healthcare guidelines. This paper demonstrates that they can also be modelled using a mainstream business modelling language such as UML. The paper presents a guideline in GLIF and as UML activity diagrams, and then presents a mapping of GLIF primitives to UML. The potential benefits of using a mainstream modelling language are outlined. These include availability of advanced modelling tools, transfer between modelling tools, and automation via business workflow technology.