-
Statistics in Medicine Dec 2022Estimating relationships between multiple incomplete patient measurements requires methods to cope with missing values. Multiple imputation is one approach to address... (Review)
Review
Estimating relationships between multiple incomplete patient measurements requires methods to cope with missing values. Multiple imputation is one approach to address missing data by filling in plausible values for those that are missing. Multiple imputation procedures can be classified into two broad types: joint modeling (JM) and fully conditional specification (FCS). JM fits a multivariate distribution for the entire set of variables, but it may be complex to define and implement. FCS imputes missing data variable-by-variable from a set of conditional distributions. In many studies, FCS is easier to define and implement than JM, but it may be based on incompatible conditional models. Imputation methods based on multilevel modeling show improved operating characteristics when imputing longitudinal data, but they can be computationally intensive, especially when imputing multiple variables simultaneously. We review current MI methods for incomplete longitudinal data and their implementation on widely accessible software. Using simulated data from the National Health and Aging Trends Study, we compare their performance for monotone and intermittent missing data patterns. Our simulations demonstrate that in a longitudinal study with a limited number of repeated observations and time-varying variables, FCS-Standard is a computationally efficient imputation method that is accurate and precise for univariate single-level and multilevel regression models. When the analyses comprise multivariate multilevel models, FCS-LMM-latent is a statistically valid procedure with overall more accurate estimates, but it requires more intensive computations. Imputation methods based on generalized linear multilevel models can lead to biased subject-level variance estimates when the statistical analyses involve hierarchical models.
Topics: Humans; Longitudinal Studies; Models, Statistical; Biometry; Research Design; Software; Computer Simulation
PubMed: 36220138
DOI: 10.1002/sim.9592 -
Digestive Diseases and Sciences Jul 2022Endoscopic retrograde cholangiopancreatography (ERCP) in patients with surgically altered anatomy is technically difficult. Extensive training is required to develop the...
BACKGROUND
Endoscopic retrograde cholangiopancreatography (ERCP) in patients with surgically altered anatomy is technically difficult. Extensive training is required to develop the ability to perform this procedure.
AIMS
To investigate the learning curve of single-balloon-assisted enteroscopy ERCP (SBE-ERCP).
METHODS
We conducted a retrospective, observational case series at a single center. We evaluated the SBE-ERCP procedures between April 2011 and February 2021. The main outcomes were the rate of reaching the target site and the success rate of the entire procedure. These parameters were additionally expressed as a learning curve.
RESULTS
A total of 687 SBE-ERCP procedures were analyzed. The learning curve was analyzed in blocks of 10 cases. In this study, seven endoscopists, experts in conventional ERCP, were included. The overall SBE-ERCP procedural success rate was 92.2% (634/687 cases). Combining all data from individual endoscopists' evaluation periods, the insertion and success rates of the SBE-ERCP procedures gradually increased with increased experience performing SBE-ERCP. The insertion success rates for the number of SBE-ERCP cases (< 20, 21-30, > 30) were 82.9%, 92.9%, and 94.3%, respectively; the procedure success rates were 74.3%, 81.4%, and 92.9%, respectively. The endoscopists who had performed > 30 SBE-ERCP cases had a success rate of ≥ 90%.
CONCLUSIONS
Our results suggest that performing > 30 cases is one of the targets for conventional ERCP experts to become competent in performing SBE-ERCP in patients with a surgically altered anatomy.
Topics: Cholangiopancreatography, Endoscopic Retrograde; Double-Balloon Enteroscopy; Humans; Learning Curve; Retrospective Studies; Single-Balloon Enteroscopy
PubMed: 34973148
DOI: 10.1007/s10620-021-07342-2 -
Statistics in Medicine Jul 2022Treatment selection biomarkers are those that can be useful in guiding choice of therapy. Just as new therapies require evaluation in appropriately designed clinical... (Randomized Controlled Trial)
Randomized Controlled Trial
Treatment selection biomarkers are those that can be useful in guiding choice of therapy. Just as new therapies require evaluation in appropriately designed clinical trials to determine their benefit, therapy selection biomarkers require evaluation in appropriately designed studies. These studies may be prospective clinical trials or retrospective studies based on specimens stored from a completed clinical trial. Ideally, patient treatment assignments should be randomized, and consideration should be given to an appropriate sample size-either for prospective planning of a new study or access to a sufficient number of stored specimens. Here, we develop a novel sample size method for estimation of a confidence interval of specified average width, for an intuitively appealing previously proposed parameter that reflects the expected benefit of using biomarker-guided therapy relative to a standard-of-care therapy. The estimation approach combines Monte Carlo and regression to result in a procedure that performs well over a range of scenarios. Although derived under a specific Cox proportional hazards regression model, robustness to model violations is demonstrated by evaluation under accelerated failure time and cure models. The sample size method produces adequate or conservative sample size estimates under a range of scenarios. Computer code in R and C++, and applications for Mac and Windows are made available for implementation of the sample size estimation procedure. The method is applied to a real data setting and results discussed.
Topics: Biomarkers; Humans; Proportional Hazards Models; Prospective Studies; Retrospective Studies; Sample Size
PubMed: 35491401
DOI: 10.1002/sim.9412 -
Journal of Education & Teaching in... Jan 2020This curriculum was developed for emergency medicine (EM) residents at the post-graduate year (PGY) 1-4 level, and attending EM physicians. It may be adapted for...
AUDIENCE
This curriculum was developed for emergency medicine (EM) residents at the post-graduate year (PGY) 1-4 level, and attending EM physicians. It may be adapted for training of any healthcare provider or learner who might be required to perform an emergency cricothyrotomy, including emergency medical technicians, senior medical students, and advanced practice providers (ie, nurse practitioners and physician assistants); however, we did not specifically validate it for these providers.
INTRODUCTION
Emergency cricothyrotomy (EC) is a lifesaving surgical procedure, often the option of last resort, used to secure the airway when other methods of airway control have failed or are not feasible. It is a high-risk procedure since it is infrequently performed, but it is time-sensitive and critical for survival when needed.1,2 Time-sensitive procedural skills such as EC are subject to relatively rapid decay,3,4 and unlike other high-risk procedures, in which just-in-time training (JITT) may improve real time procedural performance, the extreme time sensitivity of cricothryotomy precludes JITT as a feasible educational intervention to improve EC performance.5 As such, clinicians must periodically review the essential concepts and practice the physical actions of the procedure in order to build and maintain familiarity with the steps involved and to develop and maintain the muscle memory necessary to perform it quickly and confidently. Previous studies have shown that simulation-based training improves both confidence and competence in the performance of the simulated procedures,6,7 and that small group learning situations are effective for procedural learning.8,9Commercially produced mannequins are available to simulate cricothyrotomy. However, being made of plastic materials, they suffer from unrealistic "tissue" feel that is radically different from that of biologic tissue.10,11 Additionally, because they are mass-produced, they tend to be fairly homogeneous in their anatomic representations, lacking the variability encountered in the human population.We developed an inexpensive procedure simulator using commercially available porcine byproduct that more closely mimics the feel of cricothyrotomy in real life, and a comprehensive curriculum for instruction in, or review of, EC, intended for implementation in a small-group format. This publication is intended to provide interested educators with a comprehensive suite of materials for teaching EC at their own institution. Included are instructions for how to construct the simulator, an EC case scenario with discussion points, a PowerPoint didactic module covering the fundamental concepts of EC, and sample course evaluation forms that may be implemented directly or adapted to meet institutional requirements.
EDUCATIONAL OBJECTIVES
After completing this activity, the learner will be able to:correctly describe the indications for and contraindications to emergency cricothyrotomycorrectly describe and identify on the simulator the anatomic landmarks involved in emergency cricothyrotomycorrectly list the required equipment and the sequence of the steps for the "standard" and "minimalist" variations of the proceduredemonstrate proper technique when performing a cricothyrotomy on the simulator without prompts or pauses.
EDUCATIONAL METHODS
Small group activity combining didactic learning, case-based learning, and procedural simulation. The didactic component may be delivered in an asynchronous learning or "flipped classroom" format.
RESEARCH METHODS
The cricothyrotomy simulator was initially pilot-tested on a group of emergency medicine attending faculty, who were asked to evaluate the simulator, with results demonstrating that it was felt to be superior to typical plastic mannequin simulators. This simulator was then subsequently integrated into our airway workshops teaching EC, which include hands-on practice, didactic, and discussion components. The content and delivery of these workshops were assessed by the learners via standardized evaluation forms after completion of each workshop, and the overall clinical relevance, appropriateness of content, and satisfaction with the workshop format were highly rated (mean score 4.87 on a 5-point scale, with 5 denoted as "Excellent").
DISCUSSION
The real-tissue model to simulate the procedure was well liked by course participants, and the learning environment was felt to be conducive to asking questions and discussion. Overall, the instructors and the learners felt that the workshops were effective in improving understanding of the procedure and increasing the comfort level and skill of the emergency physician learners in performing the procedure.
TOPICS
Cricothyrotomy, cricothyroidotomy, emergency airway, surgical airway, failed airway, rescue airway, can't intubate can't ventilate, small group activity, simulation.
PubMed: 37465596
DOI: 10.21980/J8JS9W -
NeuroImage Aug 2023Cognitive neuroscientists have been grappling with two related experimental design problems. First, the complexity of neuroimaging data (e.g. often hundreds of thousands...
Cognitive neuroscientists have been grappling with two related experimental design problems. First, the complexity of neuroimaging data (e.g. often hundreds of thousands of correlated measurements) and analysis pipelines demands bespoke, non-parametric statistical tests for valid inference, and these tests often lack an agreed-upon method for performing a priori power analyses. Thus, sample size determination for neuroimaging studies is often arbitrary or inferred from other putatively but questionably similar studies, which can result in underpowered designs - undermining the efficacy of neuroimaging research. Second, when meta-analyses estimate the sample sizes required to obtain reasonable statistical power, estimated sample sizes can be prohibitively large given the resource constraints of many labs. We propose the use of sequential analyses to partially address both of these problems. Sequential study designs - in which the data is analyzed at interim points during data collection and data collection can be stopped if the planned test statistic satisfies a stopping rule specified a priori - are common in the clinical trial literature, due to the efficiency gains they afford over fixed-sample designs. However, the corrections used to control false positive rates in existing approaches to sequential testing rely on parametric assumptions that are often violated in neuroimaging settings. We introduce a general permutation scheme that allows sequential designs to be used with arbitrary test statistics. By simulation, we show that this scheme controls the false positive rate across multiple interim analyses. Then, performing power analyses for seven evoked response effects seen in the EEG literature, we show that this sequential analysis approach can substantially outperform fixed-sample approaches (i.e. require fewer subjects, on average, to detect a true effect) when study designs are sufficiently well-powered. To facilitate the adoption of this methodology, we provide a Python package "niseq" with sequential implementations of common tests used for neuroimaging: cluster-based permutation tests, threshold-free cluster enhancement, t-max, F-max, and the network-based statistic with tutorial examples using EEG and fMRI data.
Topics: Humans; Cognitive Neuroscience; Research Design; Sample Size; Magnetic Resonance Imaging; Neuroimaging
PubMed: 37348624
DOI: 10.1016/j.neuroimage.2023.120232 -
BMJ Open Feb 2022Gastrointestinal stromal tumours (GISTs) are the most common mesenchymal tumours of the digestive system, and complete resection is the only way to provide a radical...
INTRODUCTION
Gastrointestinal stromal tumours (GISTs) are the most common mesenchymal tumours of the digestive system, and complete resection is the only way to provide a radical cure for resectable GISTs. Open surgery and minimally invasive approaches, including laparoscopy, robotic surgery and endoscopy, consist of the mainstream GIST resection. However, there is still a lack of evidence regarding which surgical outcomes and long-term prognosis would be better. Thus, we are planning to conduct a network meta-analysis and systematic review aiming to determine the comparative effectiveness among laparotomy, laparoscopy, endoscopy, robotic surgery, and laparoscopic and endoscopic cooperative surgery in GISTs.
METHOD AND ANALYSIS
PubMed, EMBASE, the Cochrane Library and Web of Science will be searched for published studies to identify the proper literature comparing open resection, laparoscopy, endoscopy, robotic surgery, and laparoscopic and endoscopic cooperative surgery for resecting GISTs from inception to February 2021. Randomised controlled trials (RCTs) and non-randomised studies comparing at least two different interventions for GIST resection will be included. RCTs and non-randomised studies will be synthesised and analysed separately. Bayesian network meta-analysis will be performed to compare the surgical outcomes and long-term prognosis among the resection methods above. The included studies will be divided into several subgroups according to tumour location and size for further analysis. Sensitivity analysis will be performed to identify and explain heterogeneity to make our results robust. Meta-regression will serve as a supplementary method if data are available. The quality of evidence will be evaluated by the Grading of Recommendations, Assessment, Development and Evaluation.
ETHICS AND DISSEMINATION
No ethical approval is required for this network meta-analysis, as it is based on already published data. The findings of the review will be published in a peer-reviewed journal.
PROSPERO REGISTRATION NUMBER
CRD42021237892.
Topics: Gastrointestinal Stromal Tumors; Humans; Laparoscopy; Laparotomy; Meta-Analysis as Topic; Network Meta-Analysis; Robotic Surgical Procedures; Systematic Reviews as Topic
PubMed: 35131818
DOI: 10.1136/bmjopen-2021-050414 -
The British Journal of Surgery Jul 2019RCTs in surgery are challenging owing to well established methodological issues. Well designed pilot and feasibility studies (PFS) may help overcome such issues to... (Review)
Review
BACKGROUND
RCTs in surgery are challenging owing to well established methodological issues. Well designed pilot and feasibility studies (PFS) may help overcome such issues to inform successful main trial design and conduct. This study aimed to analyse protocols of UK-funded studies to explore current use of PFS in surgery and identify areas for practice improvement.
METHODS
PFS of surgical interventions funded by UK National Institute for Health Research programmes from 2005 to 2015 were identified, and original study protocols and associated publications sourced. Data extracted included study design characteristics, reasons for performing the work including perceived uncertainties around conducting a definitive main trial, and whether the studies had been published.
RESULTS
Thirty-five surgical studies were identified, of which 29 were randomized, and over half (15 of 29) included additional methodological components (such as qualitative work examining recruitment, and participant surveys studying current interventions). Most studies focused on uncertainties around recruitment (32 of 35), with far fewer tackling uncertainties specific to surgery, such as intervention stability, implementation or delivery (10 of 35). Only half (19 of 35) had made their results available publicly, to date.
CONCLUSION
The full potential of pretrial work to inform and optimize definitive surgical studies is not being realized.
Topics: Clinical Trials as Topic; Feasibility Studies; Humans; Patient Selection; Pilot Projects; Randomized Controlled Trials as Topic; Research Design; Surgical Procedures, Operative; United Kingdom
PubMed: 31074503
DOI: 10.1002/bjs.11167 -
ALTEX 2023Every test procedure, scientific and non-scientific, has inherent uncertainties, even when performed according to a standard operating procedure (SOP). In addition, it...
Every test procedure, scientific and non-scientific, has inherent uncertainties, even when performed according to a standard operating procedure (SOP). In addition, it is prone to errors, defects, and mistakes introduced by operators, laboratory equipment, or materials used. Adherence to an SOP and comprehensive validation of the test method cannot guarantee that each test run produces data within the acceptable range of variability and with the precision and accuracy determined during the method validation. We illustrate here (part I) why controlling the validity of each test run is an important element of experimental design. The definition and application of acceptance criteria (AC) for the validity of test runs is important for the setup and use of test methods, particularly for the use of new approach methods (NAM) in toxicity testing. AC can be used for decision rules on how to handle data, e.g., to accept the data for further use (AC fulfilled) or to reject the data (AC not fulfilled). The adherence to AC has important requirements and consequences that may seem surprising at first sight: (i) AC depend on a test method's objectives, e.g., on the types/concentrations of chemicals tested, the regulatory context, the desired throughput; (ii) AC are applied and documented at each test run, while validation of a method (including the definition of AC) is only performed once; (iii) if AC are altered, then the set of data produced by a method can change. AC, if missing, are the blind spot of quality assurance: Test results may not be reliable and comparable. The establishment and uses of AC will be further detailed in part II of this series.
Topics: Humans; Toxicity Tests; Research Design; Biological Science Disciplines
PubMed: 37889190
DOI: 10.14573/altex.2310021 -
Clinical Trials (London, England) Aug 2022Factorial designs and multi-arm multi-stage (MAMS) platform designs have many advantages, but the practical advantages and disadvantages of combining the two designs...
BACKGROUND
Factorial designs and multi-arm multi-stage (MAMS) platform designs have many advantages, but the practical advantages and disadvantages of combining the two designs have not been explored.
METHODS
We propose practical methods for a combined design within the platform trial paradigm where some interventions are not expected to interact and could be given together.
RESULTS
We describe the combined design and suggest diagrams that can be used to represent it. Many properties are common both to standard factorial designs, including the need to consider interactions between interventions and the impact of intervention efficacy on power of other comparisons, and to standard multi-arm multi-stage designs, including the need to pre-specify procedures for starting and stopping intervention comparisons. We also identify some specific features of the factorial-MAMS design: timing of interim and final analyses should be determined by calendar time or total observed events; some non-factorial modifications may be useful; eligibility criteria should be broad enough to include any patient eligible for any part of the randomisation; stratified randomisation may conveniently be performed sequentially; and analysis requires special care to use only concurrent controls.
CONCLUSION
A combined factorial-MAMS design can combine the efficiencies of factorial trials and multi-arm multi-stage platform trials. It allows us to address multiple research questions under one protocol and to test multiple new treatment options, which is particularly important when facing a new emergent infection such as COVID-19.
Topics: Clinical Trials as Topic; Humans; Random Allocation; Research Design
PubMed: 35579066
DOI: 10.1177/17407745221093577 -
BMJ Open Oct 2020Coeliac disease (CD) is a systemic immune-mediated disorder triggered by gluten in genetically predisposed individuals. CD is diagnosed using a combination of serology...
INTRODUCTION
Coeliac disease (CD) is a systemic immune-mediated disorder triggered by gluten in genetically predisposed individuals. CD is diagnosed using a combination of serology tests and endoscopic biopsy of the small intestine. However, because of non-specific symptoms and heterogeneous clinical presentation, diagnosing CD is challenging. Early detection of CD through improved case-finding strategies can improve the response to a gluten-free diet, patients' quality of life and potentially reduce the risk of complications. However, there is a lack of consensus in which groups may benefit from active case-finding.
METHODS AND ANALYSIS
We will perform a systematic review to determine the accuracy of diagnostic indicators (such as symptoms and risk factors) for CD in adults and children, and thus can help identify patients who should be offered CD testing. MEDLINE, Embase, Cochrane Library and Web of Science will be searched from 1997 until 2020. Screening will be performed in duplicate. Data extraction will be performed by one and checked by a second reviewer. Disagreements will be resolved through discussion or referral to a third reviewer. We will produce a narrative summary of identified prediction models. Studies, where 2×2 data can be extracted or reconstructed, will be treated as diagnostic accuracy studies, that is, the diagnostic indicators are the index tests and CD serology and/or biopsy is the reference standard. For each diagnostic indicator, we will perform a bivariate random-effects meta-analysis of the sensitivity and specificity.
ETHICS AND DISSEMINATION
Results will be reported in peer-reviewed journals, academic and public presentations and social media. We will convene an implementation panel to advise on the optimum strategy for enhanced dissemination. We will discuss findings with Coeliac UK to help with dissemination to patients. Ethical approval is not applicable, as this is a systematic review and no research participants will be involved.
PROSPERO REGISTRATION NUMBER
CRD42020170766.
Topics: Adult; Celiac Disease; Child; Humans; Mass Screening; Meta-Analysis as Topic; Quality of Life; Research Design; Sensitivity and Specificity; Systematic Reviews as Topic
PubMed: 33020103
DOI: 10.1136/bmjopen-2020-038994