Previous Submissions Research Articles

Research agenda for ending preventable maternal deaths from postpartum haemorrhage: a WHO research prioritisation exercise

IntroductionPostpartum haemorrhage (PPH) remains the leading cause of maternal death. Yet there is a lack of clarity around what research is needed to determine what works and how best to...

Searching for a recipe for success: environmental citizen petitions under free trade agreements

ABSTRACT Submissions on enforcement matters (SEMs) allow civil society members to assert that one party to a free trade agreement (FTA) is failing to enforce its domestic environmental laws. Submissions that survive every step of the SEM process result in a ‘factual record’ prepared by a secretariat on enforcement matters. Relying on qualitative comparative analysis and an original dataset compiling the 158 SEMs submitted under the framework of US FTAs, I investigate what conditions make submissions more likely to succeed in leading to a factual record. I find that the existence of previous submissions dealing with the same facts, submissions against developing countries, and the presence of at least one non-governmental organization as a submitter contribute to a submission’s success, but only in conjunction with other conditions and for a limited number of cases. This result provides overall evidence that secretariats’ decisions are unbiased by the factors examined in this paper.

Zombie cheminformatics: extraction and conversion of Wiswesser Line Notation (WLN) from chemical documents

PurposeWiswesser Line Notation (WLN) is a old line notation for encoding chemical compounds for storage and processing by computers. Whilst the notation itself has long since been surpassed by SMILES and InChI, distribution of WLN during its active years was extensive. In the context of modernising chemical data, we present a comprehensive WLN parser developed using the OpenBabel toolkit, capable of translating WLN strings into various formats supported by the library. Furthermore, we have devised a specialised Finite State Machine l, constructed from the rules of WLN, enabling the recognition and extraction of chemical strings out of large bodies of text. Available open-access WLN data with corresponding SMILES or InChI notation is rare, however ChEMBL, ChemSpider and PubChem all contain WLN records which were used for conversion scoring. Our investigation revealed a notable proportion of inaccuracies within the database entries, and we have taken steps to rectify these errors whenever feasible.Scientific contributionTools for both the extraction and conversion of WLN from chemical documents have been successfully developed. Both the Deterministic Finite Automaton (DFA) and parser handle the majority of WLN rules officially endorsed in the three major WLN manuals, with the parser showing a clear jump in accuracy and chemical coverage over previous submissions. The GitHub repository can be found here: https://github.com/Mblakey/wiswesser.

AI for conceptual architecture: Reflections on designing with text-to-text, text-to-image, and image-to-image generators

In this paper we present a research-through-design study where we employed text-to-text, text-to-image, and image-to-image generative tools for a conceptual architecture project for the eVolo skyscraper competition. We trained these algorithms on a dataset that we collected and curated, consisting of texts about and images of architecture. We describe our design process, present the final proposal, reflect on the usefulness of such tools for early-stage design, and discuss implications for future research and practice. By analysing the results from training the text-to-text generators we could establish a specific design brief that informed the final concept. The results from the image-to-image generator gave an overview of the shape grammars of previous submissions. All results were intriguing and can assist creativity and in this way, the tools were useful for gaining insight into historical architectural data, helped shape a specific design brief, and provoked new ideas. By reflecting on our design process, we argue that the use of language when employing such tools takes a new role and that three layers of language intertwined in our work: architectural discourse, programming languages, and annotations. We present a map that unfolds how these layers came together as a contribution to making machine learning more explainable for creatives.

SAMPL9 blind predictions using nonequilibrium alchemical approaches.

We present our blind predictions for the Statistical Assessment of the Modeling of Proteins and Ligands (SAMPL), ninth challenge, focusing on the binding of WP6 (carboxy-pillar[6]arene) with ammonium/diammonium cationic guests. Host-guest binding free energies have been calculated using the recently developed virtual double system single box approach, based on the enhanced sampling of the bound and unbound end-states followed by fast switching nonequilibrium alchemical simulations [M. Macchiagodena et al., J. Chem. Theory Comput. 16, 7160 (2020)]. As far as Pearson and Kendall coefficients are concerned, performances were acceptable and, in general, better than those we submitted for calixarenes, cucurbituril-like open cavitand, and beta-cyclodextrines in previous SAMPL host-guest challenges, confirming the reliability of nonequilibrium approaches for absolute binding free energy calculations. In comparison with previous submissions, we found a rather large mean signed error that we attribute to the way the finite charge correction was addressed through the assumption of a neutralizing background plasma.

SAMPL9 blind predictions using nonequilibrium alchemical approaches

We present our blind predictions for the Statistical Assessment of the Modeling of Proteins and Ligands (SAMPL), 9th challenge, focusing on binding of WP6 (carboxy-pillar[6]arene) with ammonium/diammonium cationic guests. Host-guest binding free energies have been calculated using the recently developed virtual double system single box approach, based on the enhanced sampling of the bound and unbound end-states followed by fast switching nonequilibrium alchemical simulations [M Macchiagodena, M Pagliai, M Karrenbrock, G Guarnieri, F Iannone, Procacci, J Chem Theory Comput, 16, 7260, 2020]. As far as Pearson and Kendall coefficients are concerned, performances were acceptable and, in general, better than those we submitted for calixarenes, cucurbituril-like open cavitand, and beta-cyclodextrines in previous SAMPL host-guest challenges, confirming the reliability of nonequilibrium approaches for absolute binding free energy calculations. In comparison with previous submissions, we found, nonetheless, a rather large mean signed error that we attribute to the way the finite charge correction was addressed through the assumption of a neutralizing background plasma.

The Role of Feedback in Dynamic Crowdsourcing Contests: A Structural Empirical Analysis

In this paper, we empirically examine the impact of performance feedback on the outcome of crowdsourcing contests. We develop a dynamic structural model to capture the economic processes that drive contest participants’ behavior and estimate the model using a detailed data set about real online logo design contests. Our rich model captures key features of the crowdsourcing context, including a large participant pool; entries by new participants throughout the contest; exploitation (revision of previous submissions) and exploration (radically novel submissions) behaviors by contest incumbents; and the participants’ strategic choice among these entry, exploration, and exploitation decisions in a dynamic game. Using counterfactual simulations, we compare the outcome of crowdsourcing contests under alternative feedback disclosure policies and award levels. Our simulation results suggest that, despite its prevalence on many platforms, the full feedback policy (providing feedback throughout the contest) may not be optimal. The late feedback policy (providing feedback only in the second half of the contest) leads to a better overall contest outcome. This paper was accepted by Gabriel Weintraub, revenue management and market analytics department.

Sequential school choice: Theory and evidence from the field and lab

We analyze sequential preference submission in centralized matching problems. Our motivation comes from school districts and colleges that use an application website where students submit their preferences over schools sequentially, after learning information about previous submissions. Comparing the widely used Boston Mechanism (BM) to the celebrated Deferred Acceptance (DA) mechanism, we show that a sequential implementation of BM can achieve superior equilibrium outcomes relative to the student-optimal stable matching. Our empirical tests use data from the field and from a laboratory experiment. We find that sequential preference submission may allow students to overcome the coordination problem in school choice. Our findings may have important policy implications in numerous places across the world where BM is currently in use.

Playing the innovation subsidy game: Experience, clusters, consultancy, and networking in regional innovation support

Government support to promote firm-level innovation is seen as a crucial factor for economic growth. This support is frequently channeled through firm-level subsidies. Despite their relevance within the policy portfolio, there is an open academic debate on whether subsidies are effective for innovation. This is by no means related to a potential inadequacy of subsidies, but because the mechanisms of assignment may be unsatisfactory. We argue that this may be the case when subsidies are awarded to larger firms with a solid international and innovative trajectory or to those that know how toplay the system,” rather than to the most deserving firms and projects. To test whether this is the case, we use data from 17,866 applicants for innovation subsidies managed by the Valencian Institute of Competitiveness. We find that firms with specific knowledge accrued through previous submissions, public funding and grant consultancy or cluster location, are the main beneficiaries of public innovation support, generally at the expense of more promising candidates that lack the know-how to navigate a complex and often flawed process. This inertia gets policy-makers stuck in a sub-optimal assignment system that should be deeply reconsidered.

Data‐Driven Medium‐Range Weather Prediction With a Resnet Pretrained on Climate Simulations: A New Model for WeatherBench

AbstractNumerical weather prediction has traditionally been based on the models that discretize the dynamical and physical equations of the atmosphere. Recently, however, the rise of deep learning has created increased interest in purely data‐driven medium‐range weather forecasting with first studies exploring the feasibility of such an approach. To accelerate progress in this area, the WeatherBench benchmark challenge was defined. Here, we train a deep residual convolutional neural network (Resnet) to predict geopotential, temperature and precipitation at 5.625° resolution up to 5 days ahead. To avoid overfitting and improve forecast skill, we pretrain the model using historical climate model output before fine‐tuning on reanalysis data. The resulting forecasts outperform previous submissions to WeatherBench and are comparable in skill to a physical baseline at similar resolution. We also analyze how the neural network creates its predictions and find that, for the case studies analyzed, the model has learned physically reasonable correlations. Finally, we perform scaling experiments to estimate the potential skill of data‐driven approaches at higher resolutions.

Seedling Morphology and Molecular Characterisation of some Ocimum L. species (Lamiaceae) in Southwest states of Nigeria.

The genus Ocimum L. (Lamiaceae) is composed of aromatic plants known for their peculiar strong smell added to their close similarities in morphological attributes, increase attention and applications as ethnobotanicals of culinary and medicinal use. There are reports of difficulty in Ocimum taxonomy. Seedling morphology and molecular characterization of some Ocimum accessions collected from Southwest states of Nigeria were carried out to delimit members of the genus. Morphological measurements were carried out on the seedlings characters and recorded. Molecular characterization was carried out on their fresh leaf‐samples. Statistical analyses of quantitative seedling data were performed using SAS/PC1 version 9.01. Seedling morphology reveals phanerocotylar epigeal with foliaceous cotyledon type; reniform and flabellate cotyledon shape, obcordate and obtuse cotyledon apex, cordate and truncate cotyledon base. There is a significant mean difference in all the quantitative seedling characters at p<0.05. Blast query reveals the accessions as five species exhibiting 99 % homologous identities with previous submissions in the GenBank database.

RW1 A REVIEW OF REAL-WORLD EVIDENCE USE IN UNITED STATES REGULATORY SUBMISSIONS

The United States Food and Drug Administration’s (FDA) real-world evidence (RWE) framework outlines the agency’s perspective for potential uses of RWE as supported by the 21st Centuries Cures Act. Historically, the use of RWE to support product approval has been limited. However, given increased opportunities for manufacturers to utilize RWE under the FDA framework, there is a need to characterize these instances to inform planning for future regulatory submissions. This review aims to identify and describe RWE use in previous FDA regulatory submissions.

Re: Comments on the Data Protection Bill, 2019

Barbados is pushing to advance a digital development agenda. Central to this effort is an appropriate consumer protection framework. Passage of comprehensive data protection legislation is seen as critical to this effort. Enter the Data Protection Bill, 2019. Prior drafts were floated as far back as 2005 and as recently as first quarter 2018. Following public comments, the 2019 edition of the Bill was crafted to reflect heavy inspiration from both the European General Data Protection Regulation and the United Kingdom's Data Protection Act, 1998. While the dual sources of inspiration are clear, the application to the draft resulted in a proposed act with some unworkable elements. This paper represents expanded commentary, following previous oral submissions to a Joint Select Committee of the Barbados Parliament by the author on June 26, 2019.

Dynamically generating T32 training documents using structured data.

BackgroundThe US National Institutes of Health (NIH) funds academic institutions for training doctoral (PhD) students and postdoctoral fellows. These training grants, known as T32 grants, require schools to create, in a particular format, seven or eight Word documents describing the program and its participants. Weill Cornell Medicine aimed to use structured name and citation data to dynamically generate tables, thus saving administrators time.Case PresentationThe author’s team collected identity and publication metadata from existing systems of record, including our student information system and previous T32 submissions. These data were fed into our ReCiter author disambiguation engine. Well-structured bibliographic metadata, including the rank of the target author, were output and stored in a MySQL database. We then ran a database query that output a Word extensible markup (XML) document according to NIH’s specifications. We generated the T32 training document using a query that ties faculty listed on a grant submission with publications that they and their mentees authored, bolding author names as required. Because our source data are well-structured and well-defined, the only parameter needed in the query is a single identifier for the grant itself. The open source code for producing this document is at http://dx.doi.org/10.5281/zenodo.2593545.ConclusionsManually writing a table for T32 grant submissions is a substantial administrative burden; some documents generated in this manner exceed 150 pages. Provided they have a source for structured identity and publication data, administrators can use the T32 Table Generator to readily output a table.

Relaunching teaching occupation

ABSTRACTAlmost 3 years ago, Hocking (2016) introduced a special section on Teaching Occupation in the Journal of Occupational Science. As she noted, two of the main driving forces to include this section were Florence Clark’s (2006) advocacy, as a founder of occupational science in the USA, to actually teach occupation if we hope to fulfill the promise of this science and guard against its fragmentation and Ann Wilcock’s educational vision as founder of the journal. As 2019 unfolds, the journal is reinforcing the invitation for submissions to a renamed section, Learning and Knowing Occupation. Submissions may focus on one or a combination of topics related to learning occupation, teaching occupation, and cultural, gendered, or other ways of knowing occupation based on particular ontologies and epistemic traditions. Citations from previous submissions to this section of the journal are used to stimulate submissions on Learning and Knowing Occupation. Inquiries to the Executive Editor are welcome.

Fact or Fiction

Subjective reporting polarizes competing viewpoints. However, helping readers to recognize subjective content leads to more impartial discussions. Towards this end, we develop machine learning models that classify sentence objectivity. We contribute a set of linguistic rules for determining sentence objectivity collated from previous work. We also develop a labeled dataset with over 5000 sentences retrieved from various news sources. Further, we evaluate traditional machine learning classification models and artificial neural networks on our dataset. The best performing model, a convolutional neural network, achieved an accuracy of 85% and an AUC of 0.933. Using our subjective-objective sentence classification model, we implement Fact-or-Fiction, an end-to-end web system that highlights objective sentences in user text. Fact-or-Fiction provides additional information, such as links to related web pages and related previous submissions.

PRM67 - UTILITIES IN ONCOLOGY SUBMISSIONS SUBMITTED TO NICE —APPROACH OF THE MANUFACTURERS

To provide an overview of the collection and use of utility values in health economic models for oncology submissions to the National Institute for Health and Clinical Excellence (NICE). Oncology submissions to NICE from January 2012 to February 2018 were reviewed using publicly available evidence on the NICE website. Appraisal committee guidance and manufacturer submissions were examined. When not available, other sections of the committee papers were consulted. Data regarding the assessment, utility data collection, and approach to the incorporation of utilities were extracted from the most recent version of a submission. Data were extracted for each regimen. A total of 95 regimens were included in the study, after the exclusion of 10 regimens (2 were replaced and 8 lacked cost-effectiveness evidence). The most common source of utility values was the pivotal clinical trial of the submission, and the EQ-5D was overwhelmingly the most frequently employed instrument. When utility values were not available in the trial or the trial utilities were deemed implausible by the manufacturer, literature or previous submissions to NICE were referenced or were combined with trial data. The most often used type of utility values were health-state specific values corresponding to disease progression. Differentiation between treatment arm and being on or off treatment was also common, with time-dependent utilities also observed. Only 18 submissions made use of end-of-life utility values. Disutilities associated with adverse events were more likely to be derived from literature than health-state utilities, with a common source frequently used within several submissions for the same indication. When treatment-specific utility values were used, adverse events were often assumed to already be accounted for. There is increasing consistency in utility modelling and adherence to the NICE guidance on health-related quality of life, with a clear preference for using trial data.

Lessons learnt from the Victorian Building Authority’s cladding audit in 2015

This paper provides a brief history of the first audit of cladding conducted in Australia, commencing in 2015, and provides details on the lessons learnt from the process. This includes the need for regulators to set clear expectations when addressing non-compliant issues, the standard of certification by practitioners and greater clarity in the construction codes. Some material from this paper has been drawn from previous submissions by the VBA to other bodies, such as the Australian Federal Senate Parliamentary Inquiry into Non- Conforming Building Products (Senate Inquiry).

PCN46 - Retrospective Validation of Predicted Long-Term Survival Estimates from an Early Data Cut of the Phase III Checkmate 025 Study in Advanced Renal Cell Carcinoma

Manufacturers often seek reimbursement of new treatments based on early clinical trial results. Extrapolation of overall survival (OS) trial data to a lifetime is often required to estimate cost-effectiveness. In oncology, this involves fitting parametric models to trial Kaplan-Meier data, however, different models will lead to different long-term survival estimates. For renal cell carcinoma, previous reimbursement submissions for nivolumab used OS data from a 1-year data cut of the CheckMate 025 (CM-025) trial. We assessed predicted survival from parametric models fitted to these data against a 3-year survival update from CM-025. We used a previous reimbursement dossier to fit a spline-based survival model to 1-year OS data for nivolumab and everolimus from CM-025. Function selection was based on criteria including visual inspection, statistical model fit, and validation versus longer-term data from the phase I CA209-003 study. A number of other parametric survival models were also evaluated. Our base-case spline function fitted to early Kaplan-Meier data from CM-025 predicted 3-year OS rates of 36.1% and 27.3% for nivolumab and everolimus, respectively. Analysis of the updated 3-year data shows observed 3-year OS rates of 39.1% and 29.5%, respectively. A lognormal distribution, which gave closer predictions to the long-term data from CA209-003, estimated 3-year OS rates of 41.6% and 32.0 %, and the best fitting non-spline-model, log-logistic, provided rates of 39.1% and 29.5% for nivolumab and everolimus, respectively. Three-year OS rate predictions using base-case parametric models originally fitted to 1-year survival data from CM-025 were conservative and underestimated the actual, observed 3-year survival rates. The observed incremental difference in 3-year OS between nivolumab and everolimus was also greater than the predicted estimates, with implications for cost-effectiveness analyses. These results show the importance of scenario testing with different OS curves in economic analyses and retrospective validation of the original results.

LO12: Implementation of an editorial internship at the Canadian Journal of Emergency Medicine to foster education and participation in academic emergency medicine

Introduction: Medical journals are an essential venue for knowledge translation. Skilled reviewers and editors are required to ensure quality standards in research publications and yet postgraduate programs rarely include this training in their curricula. Imparting appropriate skills and developing capacity in journalship has thus proved challenging. The Canadian Journal of Emergency Medicine (CJEM) is the national journal for Emergency Medicine (EM) in Canada. The CJEM editorial board recently decided to provide longitudinal mentorship for junior academic faculty members and trainees through an editorial internship. The internship had three goals for participants: (1) introduce and develop the responsibilities and skills of a good editor; (2) enhance a career in academic EM; and, (3) galvanize future participation as a reviewer or editor in scientific publications. Methods: The senior editorial board of CJEM and the inaugural intern developed a one-year Editorial Internship that was launched in June 2017. The curricular framework was designed by current and prior CJEM senior editors from four Canadian universities, and was informed by similar programs in the United States. The curriculum was refined iteratively based on feedback and discussion between the senior editors and intern. The internship was designed for a single individual in the Canadian EM community, including residents, pediatric fellows and practicing emergency physicians. Results: To develop the responsibilities and skills of being a good editor, the intern performed six mentored reviews of manuscripts either under current review at CJEM or previous submissions identified as difficult peer review decisions. In addition, the intern learned about CJEM values and norms by participating in monthly videoconference meetings and quarterly editorial board meetings. To enhance an academic career, the intern was assigned two writing projects under the guidance of senior editors for publication in CJEM, and completed an online critical appraisal course. Conclusion: The inaugural editorial intern gained experience as an editor and produced scholarly work. We feel the internship met its first two goals, and CJEM has committed to continue the internship annually. The ultimate determination of whether the internship achieved its third goal will only be known after longitudinal tracking of participants career involvement in academic publishing and editing.