scholarly journals Learning from data: A recurring feature on the science and practice of data‐driven learning health systems

2021 ◽  
Author(s):  
Peter J. Embi ◽  
Philip R. O. Payne ◽  
Charles P. Friedman
Keyword(s):  
Health Systems ◽  
Data Driven ◽  
Learning From Data

scholarly journals The Heimdall framework for supporting characterisation of learning health systems

2018 ◽  
Vol 25 (2) ◽  
pp. 77-87 ◽  
Author(s):  
Scott McLachlan ◽  
Henry W. W. Potts ◽  
Kudakwashe Dube ◽  
Derek Buchanan ◽  
Stephen Lean ◽  
...  
Keyword(s):  
Health Systems ◽  
Patient Outcomes ◽  
Critical Mass ◽  
Data Driven ◽  
Barriers To Adoption ◽  
Ongoing Research ◽  
Type Method ◽  
Narrative Study ◽  
New Framework

BackgroundThere are many proposed benefits of using learning health systems (LHSs), including improved patient outcomes. There has been little adoption of LHS in practice due to challenges and barriers that limit adoption of new data-driven technologies in healthcare. We have identified a more fundamental explanation: the majority of developments in LHS are not identified as LHS. The absence of a unifying namespace and framework brings a lack of consistency in how LHS is identified and classified. As a result, the LHS ‘community’ is fragmented, with groups working on similar systems being unaware of each other’s work. This leads to duplication and the lack of a critical mass of researchers necessary to address barriers to adoption.ObjectiveTo find a way to support easy identification and classification of research works within the domain of LHS.MethodA qualitative meta-narrative study focusing on works that self-identified as LHS was used for two purposes. First, to find existing standard definitions and frameworks using these to create a new unifying framework. Second, seeking whether it was possible to classify those LHS solutions within the new framework.ResultsThe study found that with apparently limited awareness, all current LHS works fall within nine primary archetypes. These findings were used to develop a unifying framework for LHS to classify works as LHS, and reduce diversity and fragmentation within the domain.ConclusionsOur finding brings clarification where there has been limited awareness for LHS among researchers. We believe our framework is simple and may help researchers to classify works in the LHS domain. This framework may enable realisation of the critical mass necessary to bring more substantial collaboration and funding to LHS. Ongoing research will seek to establish the framework’s effect on the LHS domain.

scholarly journals Improving district facility readiness: a 12-month evaluation of a data-driven health systems strengthening intervention in rural Rwanda

2015 ◽  
Vol 8 (1) ◽  
pp. 28365 ◽  
Author(s):  
Hari S. Iyer ◽  
Emmanuel Kamanzi ◽  
Jean Claude Mugunga ◽  
Karen Finnegan ◽  
Alice Uwingabiye ◽  
...  
Keyword(s):  
Health Systems ◽  
Health Systems Strengthening ◽  
Data Driven ◽  
Facility Readiness

scholarly journals The need for adaptable global guidance in health systems strengthening for musculoskeletal health: a qualitative study of international key informants

2021 ◽  
Vol 6 (1) ◽  
Author(s):  
Andrew M. Briggs ◽  
Joanne E. Jordan ◽  
Deborah Kopansky-Giles ◽  
Saurab Sharma ◽  
Lyn March ◽  
...  
Keyword(s):  
Qualitative Study ◽  
Health Systems ◽  
Logic Model ◽  
Health Systems Strengthening ◽  
Data Driven ◽  
Key Informants ◽  
Country Level ◽  
Global Guidance ◽  
Guiding Principles ◽  
Contemporary Context

Abstract Background Musculoskeletal (MSK) conditions, MSK pain and MSK injury/trauma are the largest contributors to the global burden of disability, yet global guidance to arrest the rising disability burden is lacking. We aimed to explore contemporary context, challenges and opportunities at a global level and relevant to health systems strengthening for MSK health, as identified by international key informants (KIs) to inform a global MSK health strategic response. Methods An in-depth qualitative study was undertaken with international KIs, purposively sampled across high-income and low and middle-income countries (LMICs). KIs identified as representatives of peak global and international organisations (clinical/professional, advocacy, national government and the World Health Organization), thought leaders, and people with lived experience in advocacy roles. Verbatim transcripts of individual semi-structured interviews were analysed inductively using a grounded theory method. Data were organised into categories describing 1) contemporary context; 2) goals; 3) guiding principles; 4) accelerators for action; and 5) strategic priority areas (pillars), to build a data-driven logic model. Here, we report on categories 1–4 of the logic model. Results Thirty-one KIs from 20 countries (40% LMICs) affiliated with 25 organisations participated. Six themes described contemporary context (category 1): 1) MSK health is afforded relatively lower priority status compared with other health conditions and is poorly legitimised; 2) improving MSK health is more than just healthcare; 3) global guidance for country-level system strengthening is needed; 4) impact of COVID-19 on MSK health; 5) multiple inequities associated with MSK health; and 6) complexity in health service delivery for MSK health. Five guiding principles (category 3) focussed on adaptability; inclusiveness through co-design; prevention and reducing disability; a lifecourse approach; and equity and value-based care. Goals (category 2) and seven accelerators for action (category 4) were also derived. Conclusion KIs strongly supported the creation of an adaptable global strategy to catalyse and steward country-level health systems strengthening responses for MSK health. The data-driven logic model provides a blueprint for global agencies and countries to initiate appropriate whole-of-health system reforms to improve population-level prevention and management of MSK health. Contextual considerations about MSK health and accelerators for action should be considered in reform activities.

scholarly journals Infusing theory into deep learning for interpretable reactivity prediction

2021 ◽  
Vol 12 (1) ◽  
Author(s):  
Shih-Han Wang ◽  
Hemanth Somarajan Pillai ◽  
Siwen Wang ◽  
Luke E. K. Achenie ◽  
Hongliang Xin
Keyword(s):  
Deep Learning ◽  
Active Site ◽  
Catalytic Properties ◽  
Prediction Performance ◽  
Catalyst Design ◽  
Data Driven ◽  
Band Theory ◽  
Learning From Data ◽  
Physical Interactions ◽  
Limited Generalizability

AbstractDespite recent advances of data acquisition and algorithms development, machine learning (ML) faces tremendous challenges to being adopted in practical catalyst design, largely due to its limited generalizability and poor explainability. Herein, we develop a theory-infused neural network (TinNet) approach that integrates deep learning algorithms with the well-established d-band theory of chemisorption for reactivity prediction of transition-metal surfaces. With simple adsorbates (e.g., *OH, *O, and *N) at active site ensembles as representative descriptor species, we demonstrate that the TinNet is on par with purely data-driven ML methods in prediction performance while being inherently interpretable. Incorporation of scientific knowledge of physical interactions into learning from data sheds further light on the nature of chemical bonding and opens up new avenues for ML discovery of novel motifs with desired catalytic properties.

scholarly journals Medicare billing law and practice: Complex, incomprehensible and beginning to unravel

2019 ◽  
Author(s):  
Margaret Faux ◽  
Jonathan Wardle ◽  
Jon Adams
Keyword(s):  
High Risk ◽  
Health System ◽  
Health Systems ◽  
Health Reform ◽  
Payment System ◽  
Data Driven ◽  
Operating Environment ◽  
Medical Practitioners ◽  
Medical Billing

Australia’s Medicare is still widely considered one of the world’s best health systems. However, continual political tinkering for forty years has led to a medical billing and payment system that has become labyrinthine in its complexity and is more vulnerable to abuse now, from all stakeholders, than when first introduced. Continuing to make alterations to Medicare without addressing underlying structural issues, may compound Australia’s health reform challenges, increase the incidence of non-compliance and expenditure and thwart necessary reforms to develop a modern, data driven, digitally informed health system. For the medical practitioners who are required to navigate the increasing complexity and relentless change, they will remain at high risk of investigation and prosecution in what has become an anarchic operating environment that they cannot avoid, but do not understand.

scholarly journals Statistical Inference for Data Adaptive Target Parameters

2016 ◽  
Vol 12 (1) ◽  
pp. 3-19 ◽  
Author(s):  
Alan E. Hubbard ◽  
Sara Kherad-Pajouh ◽  
Mark J. van der Laan
Keyword(s):  
Statistical Inference ◽  
Ad Hoc ◽  
Random Variable ◽  
Data Driven ◽  
Target Parameter ◽  
Learning From Data ◽  
Confirmatory Analysis ◽  
Data Adaptive ◽  
Parameter Approach ◽  
New Framework

Abstract Consider one observes n i.i.d. copies of a random variable with a probability distribution that is known to be an element of a particular statistical model. In order to define our statistical target we partition the sample in V equal size sub-samples, and use this partitioning to define V splits in an estimation sample (one of the V subsamples) and corresponding complementary parameter-generating sample. For each of the V parameter-generating samples, we apply an algorithm that maps the sample to a statistical target parameter. We define our sample-split data adaptive statistical target parameter as the average of these V-sample specific target parameters. We present an estimator (and corresponding central limit theorem) of this type of data adaptive target parameter. This general methodology for generating data adaptive target parameters is demonstrated with a number of practical examples that highlight new opportunities for statistical learning from data. This new framework provides a rigorous statistical methodology for both exploratory and confirmatory analysis within the same data. Given that more research is becoming “data-driven”, the theory developed within this paper provides a new impetus for a greater involvement of statistical inference into problems that are being increasingly addressed by clever, yet ad hoc pattern finding methods. To suggest such potential, and to verify the predictions of the theory, extensive simulation studies, along with a data analysis based on adaptively determined intervention rules are shown and give insight into how to structure such an approach. The results show that the data adaptive target parameter approach provides a general framework and resulting methodology for data-driven science.

scholarly journals Learning health systems, embedded research, and data standards—recommendations for healthcare system leaders

JAMIA Open ◽  
2020 ◽  
Author(s):  
Rachel L Richesson
Keyword(s):  
Health Systems ◽  
Data Driven ◽  
Clinical Settings ◽  
Healthcare Organizations ◽  
Critical Research ◽  
Data Standards ◽  
Quality Healthcare ◽  
System Leaders ◽  
Research Questions

Abstract Learning health systems that conduct embedded research require infrastructure for the seamless adoption of clinical interventions; this infrastructure should integrate with electronic health record (EHR) systems and enable the use of existing data. As purchasers of EHR systems, and as critical partners, sponsors, and consumers of embedded research, healthcare organizations should advocate for EHR system functionality and data standards that will increase the capacity for embedded research in clinical settings. As stakeholders and proponents for EHR data standards, healthcare leaders should support standards development and promote local adoption to support quality healthcare, continuous improvement, innovative data-driven interventions, and the generation of new knowledge. “Standards-enabled” health systems will be positioned to address emergent and critical research questions, including those related to coronavirus disease 2019 (COVID-19) and future public health threats. The role of a data standards officer or champion could enable health systems to realize this goal.

scholarly journals Monitoring Health Systems by Estimating Excess Mortality

2020 ◽  
Author(s):  
Rolando J. Acosta ◽  
Rafael A. Irizarry
Keyword(s):  
Puerto Rico ◽  
New Jersey ◽  
Natural Disasters ◽  
Health Systems ◽  
Excess Mortality ◽  
Data Driven ◽  
Mortality Data ◽  
List Type ◽  
Individual Level ◽  
The Us

AbstractImportance: Monitoring health systems during and after natural disasters, epidemics, or outbreaks is critical for guiding policy decisions and interventions. When the effects of an event are long-lasting and difficult to detect in the short term, the accumulated effects can be devastating.Objective: We aim to leverage the improved access to mortality data to develop data-driven approaches that can help monitor health systems and quantify the effects on mortality of natural disasters, epidemics, or outbreaks.Design, Setting, and Participants: To demonstrate the utility of our approach we conducted several retrospective time-series analyses of mortality assessment after natural disasters. We obtained individual-level mortality records from the Department of Health of Puerto Rico from January 1985 to May 2020 to study the effects of hurricanes Hugo, Georges, and María in September 1989, September 1998, and September 2017, respectively. Further, we obtained daily mortality counts from Florida, New Jersey, and Louisiana’s Vital Statistic systems from January 2015 to December 2018, January 2007 to December 2015, and January 2003 to December 2006, respectively, to study the effects of hurricanes Irma in 2017, Sandy in 2013, and Katrina in 2005. Finally, we obtained individual-level mortality data from the Cook county, IL, medical examiners office, and state-specific weekly mortality counts from the Center for Disease Control and Prevention to assess the effect of the COVID-19 pandemic on the US health system.Exposures: Hurricanes María, Georges, and Hugo in Puerto Rico, Irma in Florida, Sandy in New Jersey, and Katrina in Louisiana, the Chikungunya outbreak in Puerto Rico, and the COVID-19 pandemic in the United States.Main Outcomes: We estimate and provide uncertainty assessments for percent increase from expected mortality, estimated excess deaths, and difference across groups.Results: We found that the death rate increase in Puerto Rico after María and Georges was substantially higher than the other hurricanes we examined. Further, we find that excess mortality in the US was already above 100,000 on May 9, 2020, with over 58% of these occurred in New York, New Jersey, Massachusetts, and Pennsylvania, and that effects of this pandemic were worse for elderly black individuals compared to whites of the same age.Conclusions and Relevance: Our approach can be used to monitor or assess health systems by estimating increased mortality rates and excess deaths from mortality records.Key PointsQuestion: Can we estimate excess mortality and provide accurate uncertainty assessments from vital statistics data?Findings: We developed statistical methodology that accounts for key sources of variability and provides accurate estimates and their standard errors for excess mortality. We applied the approach to datasets from several US states including periods affected by hurricanes and epidemics. We found an elevated and persistent increase in mortality after hurricanes in Puerto Rico that was substantially higher than in other US states. We also found that excess mortality in the US during the COVID-19 pandemic reached 100,000 by May 9, 2020. Finally, we found significant differences in the effects of this pandemic across racial groups in the US.Meaning: Data-driven approaches can help monitor health systems and quantify the effects on mortality of natural disasters, epidemics, or outbreaks.

Sign in / Sign up

Export Citation Format

Share Document