Bibliographic record
Abstract
KEY POINT: A systematic review summarizes the available evidence on a specific topic by applying a well-defined and rigorous methodology in a structured and reproducible manner.In this issue of Anesthesia & Analgesia, Park et al1 report the results of a systematic review on the efficacy and safety of magnesium for chronic pain treatment. The aim of such a systematic review is to identify the existing relevant literature and to summarize current evidence on a well-defined research question by applying a well-defined and rigorous methodology in a structured and reproducible manner.2 A systematic review involves a series of distinct steps: Define the research question: Analogous to a clear specific aim in a clinical study,3 a well-defined review question is the backbone of the systematic review. A strong review question is clinically relevant, not too narrow yet focused, and typically describes the population, the intervention or exposure, and the outcome(s) of interest. Specify clear inclusion and exclusion criteria: Eligibility criteria result to a large extent directly from the study question. The acronym “PICO” (population, intervention, comparator, outcomes) helps define which patient population, interventions or exposures in the treatment and control groups, and outcome(s) a study must report to be eligible for inclusion. Other characteristics like study design, publication date, or language can also be part of the inclusion and exclusion criteria. Perform a comprehensive literature search: Bibliographic databases such as PubMed, Web of Knowledge, Scopus, and EMBASE are typically the primary resource for the literature search. Sensible search terms that map onto the study selection criteria need to be identified and combined in a meaningful way using Boolean operators. Additional search strategies like screening of reference lists, conference proceedings, or trial registries can identify additional references or unpublished studies. Select studies: First, titles and abstracts are screened against the inclusion and exclusion criteria. For potentially eligible references, full-texts are obtained to assess in more detail which papers to include. Study selection, data extraction, and quality assessment should be independently performed by 2 or more researchers, and results of the search and selection procedure are depicted in a flow diagram (Figure). Extract data: The type and amount of extracted data depend on the aim and scope of the review, but conventionally these data include type(s) of study design, patient characteristics, and outcome data. A well-designed data extraction process and form facilitates consistent and complete data extraction. Assess the quality of included studies: A variety of quality assessment tools are available for various study designs (eg, Cochrane risk of bias tool; Grading of Recommendations Assessment, Development, and Evaluation [GRADE]; Newcastle-Ottawa scale). The results of the quality assessment are vital for readers to judge whether the included studies provide credible and generalizable information. Authors also sometimes chose to exclude completely poor-quality studies to avoid a “garbage in, garbage out” effect. Analyze, present, and interpret the results: A meta-analysis is often, but not always, used to coalesce quantitative data. However, when the included studies are not sufficiently similar to allow a meaningful data synthesis, a meta-analysis should not be performed. For example, Park et al1 observed a substantial clinical and methodological heterogeneity (eg, variable chronic pain conditions, treatment and follow-up periods, and magnesium formulations), and therefore, they appropriately chose to perform a qualitative analysis and to present the data descriptively. Figure.: Figure 1 from Park et al1 depicting the results of the literature search per database, which should ideally also show the total number of citations after elimination of duplicates; the number of full-text articles that were assessed for eligibility; the number of studies excluded with reasons for exclusion; and the number of studies actually included in the systematic review for qualitative and quantitative analyses.All of these steps should be clearly described in a protocol, which is registered with the international prospective register of systematic reviews (PROSPERO) (www.crd.york.ac.uk/prospero/) before commencing data extraction.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.291 | 0.645 |
| Meta-epidemiology (narrow) | 0.005 | 0.005 |
| Meta-epidemiology (broad) | 0.026 | 0.011 |
| Bibliometrics | 0.044 | 0.041 |
| Science and technology studies | 0.004 | 0.017 |
| Scholarly communication | 0.027 | 0.024 |
| Open science | 0.011 | 0.016 |
| Research integrity | 0.019 | 0.017 |
| Insufficient payload (model declined to judge) | 0.071 | 0.024 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; the direct Gemma label and the distilled Codex classifier agree on what is shown here.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".