Objective:Acknowledging study limitations in a scientific publication is a crucial element in scientific transparency and progress. However, limitation reporting is often inadequate. Natural language processing (NLP) methods could support automated reporting checks, improving research transparency. In this study, our objective was to develop a dataset and NLP methods to detect and categorize self-acknowledged limitations (e.g., sample size, blinding) reported in randomized controlled trial (RCT) publications.Methods:We created a data model of limitation types in RCT studies and annotated a corpus of 200 full-text RCT publications using this data model. We fine-tuned BERT-based sentence classification models to recognize the limitation sentences and their types. To address the small size of the annotated corpus, we experimented with data augmentation approaches, including Easy Data Augmentation (EDA) and Prompt-Based Data Augmentation (PromDA). We applied the best-performing model to a set of about 12K RCT publications to characterize self-acknowledged limitations at larger scale.Results:Our data model consists of 15 categories and 24 sub-categories (e.g., Population and its sub-category DiagnosticCriteria). We annotated 1090 instances of limitation types in 952 sentences (4.8 limitation sentences and 5.5 limitation types per article). A fine-tuned PubMedBERT model for limitation sentence classification improved upon our earlier model by about 1.5 absolute percentage points in F1 score (0.821 vs. 0.8) with statistical significance (). Our best-performing limitation type classification model, PubMedBERT fine-tuning with PromDA (Output View), achieved an F1 score of 0.7, improving upon the vanilla PubMedBERT model by 2.7 percentage points, with statistical significance ().Conclusion:The model could support automated screening tools which can be used by journals to draw the authors’ attention to reporting issues. Automatic extraction of limitations from RCT publications could benefit peer review and evidence synthesis, and support advanced methods to search and aggregate the evidence from the clinical trial literature.
MULTIFILE
Abstract Background: Antipsychotic-induced Weight Gain (AiWG) is a debilitating and common adverse effect of antipsychotics. AiWG negatively impacts life expectancy, quality of life, treatment adherence, likelihood of developing type-2 diabetes and readmission. Treatment of AiWG is currently challenging, and there is no consensus on the optimal management strategy. In this study, we aim to evaluate the use of metformin for the treatment of AiWG by comparing metformin with placebo in those receiving treatment as usual, which includes a lifestyle intervention. Methods: In this randomized, double-blind, multicenter, placebo-controlled, pragmatic trial with a follow-up of 52 weeks, we aim to include 256 overweight participants (Body Mass Index (BMI) > 25 kg/m2) of at least 16years of age. Patients are eligible if they have been diagnosed with schizophrenia spectrum disorder and if they have been using an antipsychotic for at least three months. Participants will be randomized with a 1:1 allocation to placebo or metformin, and will be treated for a total of 26 weeks. Metformin will be started at 500 mg b.i.d. and escalated to 1000 mg b.i.d. 2 weeks thereafter (up to a maximum of 2000mg daily). In addition, all participants will undergo a lifestyle intervention as part of the usual treatment consisting of a combination of an exercise program and dietary consultations. The primary outcome measure is difference in body weight as a continuous trait between the two arms from treatment inception until 26 weeks of treatment, compared to baseline. Secondary outcome measures include: 1) Any element of metabolic syndrome (MetS); 2) Response, defined as ≥5% body weight loss at 26 weeks relative to treatment inception; 3) Quality of life; 4) General mental and physical health; and 5) Cost-effectiveness. Finally, we aim to assess whether genetic liability to BMI and MetS may help estimate the amount of weight reduction following initiation of metformin treatment. Discussion: The pragmatic design of the current trial allows for a comparison of the efficacy and safety of metformin in combination with a lifestyle intervention in the treatment of AiWG, facilitating the development of guidelines on the interventions for this major health problem.
Background: To facilitate adherence to adaptive pain management behaviors after interdisciplinary multimodal pain treatment, we developed a mobile health app (AGRIPPA app) that contains two behavior regulation strategies. Objective: The aims of this project are (1) to test the effectiveness of the AGRIPPA app on pain disability; (2) to determine the cost-effectiveness; and (3) to explore the levels of engagement and usability of app users. Methods: We will perform a multicenter randomized controlled trial with two parallel groups. Within the 12-month inclusion period, we plan to recruit 158 adult patients with chronic pain during the initial stage of their interdisciplinary treatment program in one of the 6 participating centers. Participants will be randomly assigned to the standard treatment condition or to the enhanced treatment condition in which they will receive the AGRIPPA app. Patients will be monitored from the start of the treatment program until 12 months posttreatment. In our primary analysis, we will evaluate the difference over time of pain-related disability between the two conditions. Other outcome measures will include health-related quality of life, illness perceptions, pain self-efficacy, app system usage data, productivity loss, and health care expenses. Results: The study was approved by the local Medical Research Ethics Committee in October 2019. As of March 20, 2020, we have recruited 88 patients. Conclusions: This study will be the first step in systematically evaluating the effectiveness and efficiency of the AGRIPPA app. After 3 years of development and feasibility testing, this formal evaluation will help determine to what extent the app will influence the maintenance of treatment gains over time. The outcomes of this trial will guide future decisions regarding uptake in clinical practice.
LINK
