Search

Injection-related pain and fear are common adverse reactions in children undergoing vaccination and influence vaccine acceptance. Despite the large body of literature on sources of vaccine non-compliance, there is no estimate of the prevalence of pain and fear as contributing factors. The objective was to estimate the prevalence of injection pain or fear of needles as barriers to childhood (i.e., 0-18 years) vaccination.Four databases were searched from inception for relevant English and French articles until August 2021. In addition, the references of recent systematic reviews and all articles included in the review were hand searched. Article screening and data extractions were performed in duplicate. Studies were included if they reported on injection-related pain or fear of needles in children (0-18 years) using a checklist/closed-ended question(s). Results were stratified by respondent (parents or children), type of pediatric population (general or under-vaccinated), and relative importance of barrier (pain or needle fear as primary reason or any reason for under-vaccination). Prevalence rates of pain or needle fear were combined using a random effects model. Quality of included studies was assessed using the Joanna Briggs Institute critical appraisal checklist for prevalence data. Quality across studies was assessed using GRADE.There were 26 studies with 45 prevalence estimates published between 1995 and 2021. For parent reports (of children) and children self-reported reasons for non-compliance, prevalence rates of pain or needle fear ranged from 5 to 13% in a general population and 8 to 28% in an under-vaccinated population, with a substantial variation in the prevalence estimates. There was no difference between category of respondent or relative importance on pain or needle fear prevalence rate. A regression model demonstrated an overall prevalence rate of pain or needle fear as an obstacle to vaccination of 8% in the general population and 18.3% in the under-vaccinated population. All evidence was very low in quality.This is the first review to systematically quantify the prevalence and therefore, importance, of pain and needle fear as obstacles to vaccination in children around the world. Pain from injection or fear of needles were demonstrated to be sufficiently prevalent as barriers to vaccination in children to warrant attention. Addressing pain and fear has the potential to significantly improve vaccination acceptance.

Older adults are at high risk of influenza-related complications or hospitalization. The purpose of this systematic review is to assess the relative cost-effectiveness of all influenza vaccine options for older adults.This systematic review identified economic evaluation studies assessing the cost-effectiveness of influenza vaccines in adults ≥65 years of age from 5 literature databases. Two reviewers independently selected, extracted, and appraised relevant studies using the JBI Critical Appraisal Checklist for Economic Evaluations and Heyland's generalizability checklist. Costs were converted to 2019 Canadian dollars and adjusted for inflation and purchasing power parity.A total of 27 studies were included. There were 18 comparisons of quadrivalent inactivated vaccine (QIV) versus trivalent inactivated vaccine (TIV): 5 showed QIV dominated TIV (ie, lower costs and higher health benefit), and 13 showed the results depended on willingness to pay (WTP). There were 9 comparisons of high-dose TIV (TIV-HD) versus TIV: 5 showed TIV-HD dominated TIV, and 4 showed the results depended on WTP. There were 8 comparisons of adjuvanted TIV (TIV-ADJ) versus TIV: 4 showed TIV-ADJ dominated TIV, and 4 showed the results depended on WTP. There were few pairwise comparisons among QIV, TIV-HD, and TIV-ADJ.The evidence suggests QIV, TIV-HD, and TIV-ADJ are cost-effective against TIV for a WTP threshold of $50 000 per quality-adjusted life-year. Future studies should include new and existing vaccine options for broad age ranges and use more robust methodologies-such as real-world evaluations or modeling studies accounting for methodological, structural, and parameter uncertainty.

The incidence of type 1 diabetes mellitus is increasing every year requiring substantial expenditure on treatment and complications. A systematic review was conducted on the cost-effectiveness of insulin formulations, including ultralong-, long-, or intermediate-acting insulin, and their biosimilar insulin equivalents.MEDLINE, EMBASE, the Cochrane Central Register of Controlled Trials, HTA, and NHS EED were searched from inception to June 11, 2021. Cost-effectiveness and cost-utility analyses were included if insulin formulations in adults (≥ 16 years) with type 1 diabetes mellitus were evaluated. Two reviewers independently screened titles, abstracts, and full-text articles, extracted study data, and appraised their quality using the Drummond 10-item checklist. Costs were converted to 2020 US dollars adjusting for inflation and purchasing power parity across currencies.A total of 27 studies were included. Incremental cost-effectiveness ratios ranged widely across the studies. All pairwise comparisons (11 of 11, 100%) found that ultralong-acting insulin was cost-effective compared with other long-acting insulins, including a long-acting biosimilar. Most pairwise comparisons (24 of 27, 89%) concluded that long-acting insulin was cost-effective compared with intermediate-acting insulin. Few studies compared long-acting insulins with one another.Long-acting insulin may be cost-effective compared with intermediate-acting insulin. Future studies should directly compare biosimilar options and long-acting insulin options and evaluate the long-term consequences of ultralong-acting insulins.

Randomized controlled trials remain the reference standard for healthcare research on effects of interventions, and the need to report both benefits and harms is essential. The Consolidated Standards of Reporting Trials (the main CONSORT) statement includes one item on reporting harms (i.e., all important harms or unintended effects in each group). In 2004, the CONSORT group developed the CONSORT Harms extension; however, it has not been consistently applied and needs to be updated. Here, we describe CONSORT Harms 2022, which replaces the CONSORT Harms 2004 checklist, and shows how CONSORT Harms 2022 items could be incorporated into the main CONSORT checklist. Thirteen items from the main CONSORT were modified to improve harms reporting. Three new items were added. In this article, we describe CONSORT Harms 2022 and how it was integrated into the main CONSORT checklist and elaborate on each item relevant to complete reporting of harms in randomized controlled trials. Until future work from the CONSORT group produces an updated checklist, authors, journal reviewers, and editors of randomized controlled trials should use the integrated checklist presented in this paper.

Background Systematic reviews (SRs) are being published at an accelerated rate. Decision makers are often faced with the challenge of comparing and choosing between multiple SRs on the same topic. We surveyed individuals in the healthcare field to understand what criteria they use to compare and select one or more SRs from multiple on the same topic. Methods We developed a survey with 21 open and closed questions. We disseminated it through social media and professional networks. Results Of the 684 respondents, 25% were health practitioners, 9% policymakers, 39% researchers, and 13% were students. Policymakers, practitioners and researchers frequently sought out SRs (98.1%) as a source of evidence to inform decision making. They frequently (97.7%) found more than one SR on a given topic of interest to them. Half (50%) struggled to choose the most valid and trustworthy SR amongst multiple. These difficulties related to lack of time (55.2%), insufficient skills/ experience in quality appraisal (27.7%), and difficulty comparing different SRs (54.3%). Respondents compared SRs based on: relevance to their question of interest; its methodological quality, and recency of SR search date. Conclusions The exponential growth in the number of SRs leads to duplication of research on similar questions and compounds the problem of identifying which evidence is of the highest quality for decision makers. Failure to critically appraise and choose the highest quality SRs means that practice and policy decisions may not reflect the best evidence, the implementation of better intervention strategies is delayed, and patients may unduly suffer.

Objective To analyze a case of heavy uterine bleeding in adolescence caused by an arteriovenous malformation (AVM) in Hanoi Obstetrics and Gynecology Hospital, Vietnam. Methods We observed a case of a 14-year-old girl, without vaginal sexual intercourse experience, having heavy uterine bleeding caused by AVM. She underwent a laparoscopic operation at the hospital for vascular lesions in the uterine anterior wall, which caused an internal hemorrhage of 1500 ml blood loss into the peritoneal cavity. Three years later, this patient was hospitalized twice for massive vaginal bleeding. Results Results of ultrasound and magnetic resonance imaging indicated a uterine intramural mass with enlarged vessels connecting to the endometrial cavity. A pelvic digital subtraction angiography was performed and showed profuse bleeding from a ruptured branch of the left uterine artery. This artery was embolized at Bach Mai hospital and the bleeding was stopped. The patient had stable health and normal menstrual periods after 4 months of follow up. Conclusion Abnormal mass with dilated vessels in the myometrium in a patient experiencing heavy uterine bleeding is an exclusive sign of uterine vascular malformation and can be treated by angioembolization.

2D Ballbot is an actual under-actuated system with second-order nonholonomic velocity constraints and input coupling case where only control input is employed to control two outputs of the system. Controlling such a system is not easy because it faces many changelings including nonlinearities, external disturbances, and uncertainties. This study proposed a robust control system for a Ballbot mobile robot. The proposed control scheme is constructed using the hierarchical sliding mode control technique. The kinematics and dynamics of the Ballbot are derived. A Lyapunov function is used to prove the stability of the closed-loop control system. The stabilizing and transferring problems are investigated through several simulations and experiments by using the actual Ballbot platform.

BACKGROUND Clear, accurate and transparent risk communication is critical to providing policy makers and the public with direction to reduce transmission and implement effective preparedness strategies during a health emergency. OBJECTIVE For those living in Canada, we aimed to explore preferred sources of obtaining COVID-19 information, perceptions on the prevalence and drivers of misinformation during the pandemic, and suggestions to optimize health communications during future public health emergencies. METHODS We administered an online survey to residents of Ontario, Canada and aimed to recruit a sample that reflected population diversity with respect to age and gender. Data were collected between June 10, 2020 – December 31, 2020. RESULTS A total of 1823 individuals participated in the survey (54% women, 39% men; 54% aged 18-40; 28% aged 41-60, 12% aged 61+). Participants most commonly obtained COVID-19 information from local television news (61%) followed by social media (51%), national/international television news (49%) and friends and family (46%). Approximately 55% of participants encountered misinformation. Participants reported high levels of trust in health authority websites and in healthcare providers; less trustworthy sources included friends and family, talk radio, social media, and blogs/opinion websites. Men were more likely to report encountering misinformation and to trust friends or family and blogs/opinion websites, compared to women. Compared to those aged 18-40 years, participants aged 41+ were more likely to trust all assessed information sources, with the exception of online media sources, and less likely to report encountering misinformation. Of those surveyed, 58% had challenges identifying or appraising COVID-19 information. CONCLUSIONS Over half of our participants encountered misinformation while seeking COVID-19 information and 58% had challenges identifying or appraising COVID-19 information. Gender and age differences in perceptions of misinformation and trust in information sources were observed. Future research on patterns of information seeking by age and gender may provide useful insights on optimizing health communication during public health emergencies.

The sliding mode control is well-known as a useful control technique that can be applied in several real-world applications. However, a straightforward and efficient process of tuning the gains of a sliding mode controller remains a challenging but interesting topic. In this paper, a mathematical method for the sliding mode controller gain tuning is investigated. Firstly, we obtain relations between the gains and the natural and damping ratio of the closed-loop system. Secondly, the time constant of the system's actuators and the system response performance criteria, including settling time and delay time, are taken into consideration to determine the gains. In addition, appropriate gain ranges, which ensure that the system performance is met and the actuators work properly, are derived. Finally, the proposed method is applied to the gain tuning process of a sliding mode altitude controller for an actual quadcopter unmanned aerial vehicle. Simulation and experimental results demonstrate the applicability and effectiveness of this method.

BACKGROUND: Increasing availability of competing biosimilar alternatives makes it challenging to make treatment decisions. The purpose of this review is to evaluate the comparative efficacy and safety of ultra-long-/long-/intermediate-acting insulin products and biosimilar insulin compared to human/animal insulin in adults with type 1 diabetes mellitus (T1DM). METHODS: MEDLINE, EMBASE, CENTRAL, and grey literature were searched from inception to March 27, 2019. Randomized controlled trials (RCTs), quasi-experimental studies, and cohort studies of adults with T1DM receiving ultra-long-/long-/intermediate-acting insulin, compared to each other, as well as biosimilar insulin compared to human/animal insulin were eligible for inclusion. Two reviewers independently screened studies, abstracted data, and appraised risk-of-bias. Pairwise meta-analyses and network meta-analyses (NMA) were conducted. Summary effect measures were mean differences (MD) and odds ratios (OR). RESULTS: We included 65 unique studies examining 14,200 patients with T1DM. Both ultra-long-acting and long-acting insulin were superior to intermediate-acting insulin in reducing A1c, FPG, weight gain, and the incidence of major, serious, or nocturnal hypoglycemia. For fasting blood glucose, long-acting once a day (od) was superior to long-acting twice a day (bid) (MD - 0.44, 95% CI: - 0.81 to - 0.06) and ultra-long-acting od was superior to long-acting bid (MD - 0.73, 95% CI - 1.36 to - 0.11). For weight change, long-acting od was inferior to long-acting bid (MD 0.58, 95% CI: 0.05 to 1.10) and long-acting bid was superior to long-action biosimilar od (MD - 0.90, 95% CI: - 1.67 to - 0.12). CONCLUSIONS: Our results can be used to tailor insulin treatment according to the desired results of patients and clinicians and inform strategies to establish a competitive clinical market, address systemic barriers, expand the pool of potential suppliers, and favor insulin price reduction. PROSPERO REGISTRATION: CRD42017077051 SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1007/s11606-021-06642-7.

ObjectivesThe COVID-19 pandemic has stimulated growing research on treatment options. We aim to provide an overview of the characteristics of studies evaluating COVID-19 treatment.DesignRapid scoping reviewData sourcesMedline, Embase and biorxiv/medrxiv from inception to 15 May 2021.SettingHospital and community care.ParticipantsCOVID-19 patients of all ages.InterventionsCOVID-19 treatment.ResultsThe literature search identified 616 relevant primary studies of which 188 were randomised controlled trials and 299 relevant evidence syntheses. The studies and evidence syntheses were conducted in 51 and 39 countries, respectively.Most studies enrolled patients admitted to acute care hospitals (84%), included on average 169 participants, with an average age of 60 years, study duration of 28 days, number of effect outcomes of four and number of harm outcomes of one. The most common primary outcome was death (32%).The included studies evaluated 214 treatment options. The most common treatments were tocilizumab (11%), hydroxychloroquine (9%) and convalescent plasma (7%). The most common therapeutic categories were non-steroidal immunosuppressants (18%), steroids (15%) and antivirals (14%). The most common therapeutic categories involving multiple drugs were antimalarials/antibiotics (16%), steroids/non-steroidal immunosuppressants (9%) and antimalarials/antivirals/antivirals (7%). The most common treatments evaluated in systematic reviews were hydroxychloroquine (11%), remdesivir (8%), tocilizumab (7%) and steroids (7%).The evaluated treatment was in favour 50% and 36% of the evaluations, according to the conclusion of the authors of primary studies and evidence syntheses, respectively.ConclusionsThis rapid scoping review characterised a growing body of comparative-effectiveness primary studies and evidence syntheses. The results suggest future studies should focus on children, elderly ≥65 years of age, patients with mild symptoms, outpatient treatment, multimechanism therapies, harms and active comparators. The results also suggest that future living evidence synthesis and network meta-analysis would provide additional information for decision-makers on managing COVID-19.

Automotive companies continue to develop integrated safety, sustainability, and reliability features that can help mitigate some of the most common driving risks associated with autonomous vehicles (AVs). Hybrid electric vehicles (HEVs) offer practical solutions to use control strategies to cut down fuel usage and emissions. AVs and HEVs are combined to take the advantages of each kind to solve the problem of wasting energy. This paper presents an intelligent driver assistance system, including adaptive cruise control (ACC) and an energy management system (EMS), for HEVs. Our proposed ACC determines the desired acceleration and safe distance with the lead car through a switched model predictive control (MPC) and a neuro-fuzzy (NF) system. The performance criteria of the switched MPC toggles between speed and distance control appropriately and its stability is mathematically proven. The EMS intelligently control the energy consumption based on ACC commands. The results show that the driving risk is extremely reduced by using ACC-MPC and ACC-NF, and the vehicle energy consumption by driver assistance system based on ACC-NF is improved by 2.6%.

Background: Hand hygiene is a great way to ensure safety for health staff and prevent infections in hospital. Objective: The study aimed to determine the rate of compliance with routine hand hygiene and to analyze some factors affecting hand hygiene compliance routine of medical staff. Method: A study that describes a cross-sectional study, a study that combines both quantitative and qualitative methods through the observation by a checklist of 92 health-care workers who perform a procedure on 368 hand-hygiene opportunities and gather information through burns. interviewed 92 medical staff, conducted 04 in-depth interviews and 02 group discussions, and collected from March to the end of June 2020. Research Using Epidata 3.1 software to input data and manage data; Stata 14.0 software for data analysis. Results show that the percentage of health staffs who complied with routine hand hygiene was 14.13%, and the knowledge and attitudes of hospital staffs were related to routine hand hygiene compliance, with p

International audience; Recently, treatment advances in direct-acting antivirals have radically changed the management of HCV patients. However, in resource-limited countries, identification of patients with active HCV infection is still challenging in remote settings due to the limited access to laboratories able to measure HCV viral load. This study evaluated whether dried blood spots (DBS) transferred to a central laboratory could overcome this challenge. A total of 315 HCV-infected patients, naïve to anti-HCV treatment, provided each three type of samples: plasma, DBS with calibrated quantities of venous blood and DBS with uncalibrated quantities of capillary blood. Qualitative comparison was conducted in terms of detection of HCV viral load on DBS as opposed to plasma to estimate sensitivity and specificity. Quantitative comparisons were conducted by means of correlation estimation. Of the 250 patients with detected plasma HCV viral load, 245 also had detectable DBS HCV viral load (capillary or venous) leading to a sensitivity of 98.0% (95% confidence interval (CI): 95.4%-99.3%); importantly, all measurements with a plasma HCV viral load >118 IU/mL were also detected in DBS. When HCV was not detected in plasma, it was also not detected in DBS resulting in 100% specificity (95% CI: 94.5%-100%). Quantitative HCV viral load results were very similar when utilizing plasma or DBS sample types as illustrated by correlations >0.99. In conclusion, DBS sample types, with either uncalibrated capillary blood or calibrated venous blood, performed well to distinguish patients with active HCV infection, and who therefore need treatment, from other patients.

To develop a web-based decision support tool that guides users through a series of simple questions for recommending knowledge synthesis methods suitable for their research question.We used findings from previous work to structure a set of questions along key dimensions of different knowledge synthesis methods. We developed the tool using four steps: (1) designing the tool, (2) conducting usability testing, (3) disseminating the tool, and (4) evaluating its real-world use. Steps 1-3 were conducted iteratively, and the tool was evaluated using the RE-AIM framework.The "Right Review" tool separates quantitative reviews and qualitative evidence synthesis (QES). Five questions are asked to select from among 26 methods for quantitative reviews, and 10 questions to select methods from among 15 QES. Conduct/reporting guidance and open-access examples are provided for each recommended method. The tool was disseminated to4,600 users worldwide within 12 months. Evaluation results showed that the tool was fit-for-purpose and easy to use.The proliferation of knowledge synthesis methods makes it challenging for reviewers to select the "right" method. "Right Review" is a free, practical decision support tool that helps reviewers choose an appropriate method from 41 alternatives.

A sustainable circular economy involves designing and promoting new products with the least environmental impact through increasing efficiency. The emergence of autonomous vehicles (AVs) has been a revolution in the automobile industry and a breakthrough opportunity to create more sustainable transportation in the future. Autonomous vehicles are supposed to provide a safe, easy-to-use and environmentally friendly means of transport. To this end, improving AVs’ safety and energy efficiency by using advanced control and optimization algorithms has become an active research topic to deliver on new commitments: carbon reduction and responsible innovation. The focus of this study is to improve the energy consumption of an AV in a vehicle-following process while safe driving is satisfied. We propose a cascade control system in which an autonomous cruise controller (ACC) is integrated with an energy management system (EMS) to reduce energy consumption. An adaptive model predictive control (AMPC) is proposed as the ACC to control the acceleration of the ego vehicle (the following vehicle) in a vehicle-following scenario, such that it can safely follow the lead vehicle in the same lane on a highway. The proposed ACC appropriately switches between speed and distance control systems to follow the lead vehicle safely and precisely. The computed acceleration is then used in the EMS component to find the optimal engine torque that minimizes the fuel consumption of the ego vehicle. EMS is designed based on two methods: type 1 fuzzy logic system (T1FLS) and interval type 2 fuzzy logic system (IT2FLS). Results show that the combination of AMPC and IT2FLS significantly reduces fuel consumption while the ego vehicle follows the lead vehicle safely and with a minimum spacing error. The proposed controller facilitates smarter energy use in AVs and supports safer transportation.

Complications of type 2 diabetes mellitus (T2DM) adversely influence patients’ health-related quality of life (HRQOL). This study is aimed at examining HRQOL of T2DM patients, as well as the effects of diabetic complications and comorbidities on HRQOL in this population. This was a hospital-based cross-sectional study on 214 T2DM patients in Hanoi, Vietnam. Short-form 12 version 2 (SF-12v2) and EuroQOL-5 Dimensions-5 Levels (EQ-5D-5L) were employed to measure the HRQOL. The median physical component summary score (PCS), mental component summary score (MCS), and EQ-5D index were 45.6, 56.3, and 0.94, respectively. Having at least one diabetic complication was associated with the reduction of SF-12 scores in social functioning (Diff.=−5.69, 95%CI=−9.24; -2.13), role emotional (Diff.=−1.81, 95%CI=−3.12; -0.51), and MCS (Diff.=−2.55, 95%CI=−5.01; -0.1). Significant decrement of physical functioning, role physical, social functioning, role emotional, and MCS was found in patients having diabetic heart diseases compared to those without diabetic complications. The study revealed that HRQOL of Vietnamese patients with diabetic complications was moderately low, especially in social and mental health perspectives. Strategies to prevent the onset of diabetic complications should be developed as a priority in diabetes management.

A ridable ballbot has unstable underactuated dynamics with nonholonomic velocity constraints and input coupling case. Such a robot can carry people and move in any direction on the floor. In this study, a pseudo-2D dynamic model of the ridable ballbot is derived. An improved nonlinear double-loop control system combined with the partial feedback linearization (PFL) technique is proposed for the ridable ballbot, which allows it to balance and transfer. Feedforward compensation is added to the controller to consider uncertainties, friction, and disturbances. Numerical simulations and experiments show the validity of the PFL double-loop control based on the pseudo-2D dynamic model.

A ridable ballbot, a personal robot, is an underactuated system. The four state variables of the ball motion and body motion are controlled by two input signals acting on the ball. Euler–Lagrange equation is applied to obtain the dynamic model of the ridable ballbot. On the basis of this dynamic model, a nonlinear controller is analyzed and designed to control balancing and transferring of the robot. The nonlinear control scheme is proposed based on the combination of two control design techniques: (i) partial feedback linearization, which is designed to maintain the body in the upright position; and (ii) sliding mode control, which provides robust control in ball motion on the floor against model imprecision, uncertainty of system parameters and friction, and external disturbances. These two control mechanisms are successfully merged into a combined controller. Numerical and experimental results indicate the effectiveness of the combined controller and proposed a dynamic model. The control algorithm asymptotically stabilizes all system responses.

In this study, we propose a nonlinear double-loop control system for a ridable ballbot. An improved nonlinear double-loop control technique based on double-loop control scheme and sliding mode technology is used, and the inner-loop consists of proportional–integral feedback plus feedforward control. A sliding mode control enables the ridable ballbot to balance and transfer on the floor. Feedforward compensation has been added for the stability of the ballbot. As a result, the control system can stabilize all state variables of the ballbot system despite the uncertainties of the system parameters such as model, friction, and external disturbances, and relatively large body inertia. Experiments demonstrate the effectiveness of the proposed control system and the performance validation of the nonlinear double-loop control.

Background. The demand for peer reviewers is disproportionate to the supply and availability of reviewers. Identifying the factors associated with peer review behaviour can allow for the development of solutions to manage the growing demand for peer reviewers. The objective of this research was to identify factors associated with completing a large number of peer reviews in a given year. Methods. A case-control study design was used to examine factors associated with individuals completing at least 100 peer reviews (‘mega peer reviewers’) from January 2018 to December 2018 as compared to a control group of peer reviewers completing between 1 and 18 peer reviews over the same time period. Data was provided by Publons, which offers a repository of peer reviewer activities in addition to tracking peer reviewer publications and research metrics. A series of independent sample t-tests and chi-square tests were conducted comparing characteristics (e.g., number of publications, number of citations, word count of peer review) of mega peer reviewers to the control group of reviewers. Results. A total of 1596 peer reviewers had data provided by Publons. A total of 396 mega peer reviewers and a random sample of 1200 control group reviewers were included. Both groups were comprised of a greater number of males than females (mega peer reviews = 92.4% male, control reviewers = 70.0% male). Mega peer reviewers demonstrated a significantly greater average number of total publications, citations, receipt of Publons awards, and a higher average h index as compared to the control group of reviewers (all p Conclusions. There is a sub-set of highly active peer reviewers that complete a large proportion of peer review activities. These individuals demonstrate significantly different characteristics than reviewers completing a more typical number of peer reviews over a one year period. Additional research that considers motivations associated with peer review behaviour should be conducted to help inform peer reviewing activity. Registration. https://osf.io/vxdhf/?view_only=313fd05399664b94bc7a9042aa225be3

Additional File 1. Supplementary Appendices

Background Current text mining tools supporting abstract screening in systematic reviews are not widely used, in part because they lack sensitivity and precision. We set out to develop an accessible, semi-automated “workflow” to conduct abstract screening for systematic reviews and other knowledge synthesis methods. Methods We adopt widely recommended text-mining and machine-learning methods to (1) process title-abstracts into numerical training data; and (2) train a classification model to predict eligible abstracts. The predicted abstracts are screened by human reviewers for (“true”) eligibility, and the newly eligible abstracts are used to identify similar abstracts, using near-neighbor methods, which are also screened. These abstracts, as well as their eligibility results, are used to update the classification model, and the above steps are iterated until no new eligible abstracts are identified. The workflow was implemented in R and evaluated using a systematic review of insulin formulations for type-1 diabetes (14,314 abstracts) and a scoping review of knowledge-synthesis methods (17,200 abstracts). Workflow performance was evaluated against the recommended practice of screening abstracts by 2 reviewers, independently. Standard measures were examined: sensitivity (inclusion of all truly eligible abstracts), specificity (exclusion of all truly ineligible abstracts), precision (inclusion of all truly eligible abstracts among all abstracts screened as eligible), F1-score (harmonic average of sensitivity and precision), and accuracy (correctly predicted eligible or ineligible abstracts). Workload reduction was measured as the hours the workflow saved, given only a subset of abstracts needed human screening. Results With respect to the systematic and scoping reviews respectively, the workflow attained 88%/89% sensitivity, 99%/99% specificity, 71%/72% precision, an F1-score of 79%/79%, 98%/97% accuracy, 63%/55% workload reduction, with 12%/11% fewer abstracts for full-text retrieval and screening, and 0%/1.5% missed studies in the completed reviews. Conclusion The workflow was a sensitive, precise, and efficient alternative to the recommended practice of screening abstracts with 2 reviewers. All eligible studies were identified in the first case, while 6 studies (1.5%) were missed in the second that would likely not impact the review’s conclusions. We have described the workflow in language accessible to reviewers with limited exposure to natural language processing and machine learning, and have made the code available to reviewers.