Comparison of text processing methods in social media-based signal detection.

PURPOSE: Adverse event (AE) identification in social media (SM) can be performed using various types of natural language processing (NLP) and machine learning (ML). These methods can be categorized by complexity and precision level. Co-occurrence-based ML methods are rather basic, as they identify simultaneous appearance of drugs and clinical events in a single post. In contrast, statistical learning methods involve more complex NLP and identify drugs, events, and associations between them. We aimed to compare the ability of co-occurrence and NLP to identify AEs and signals of disproportionate reporting (SDR) in patient-generated SM. We also examined the performance of lift in SM-based signal detection (SD). METHODS: Our examination was performed in a corpus of SM posts crawled from open online patient forums and communities, using the spontaneously reported VigiBase data as reference data set. RESULTS: We found that co-occurrence and NLP produce AEs, which are 57% and 93% consistent with VigiBase AEs, respectively. Among the SDRs identified both in SM and in VigiBase, up to 55.3% were identified earlier in co-occurrence, and up to 32.1% were identified earlier in NLP-processed SM. Using lift in SM SD provided performance similar to frequentist methods, both in co-occurrence and in NLP-processed AEs. CONCLUSION: Our results indicate that using SM as a data source complementary to traditional pharmacovigilance sources should be considered further. Various levels of SM processing may be considered, depending on the preferred policies and tolerance for false-positive to false-negative balance in routine pharmacovigilance processes.

Immunolab: Combining targeted real-world data with advanced analytics to generate evidence at scale in immunology.

Real-world evidence (RWE) has traditionally been used by regulatory or payer authorities to inform disease burden, background risk, or conduct post-launch pharmacovigilance, but in recent years RWE has been increasingly used to inform regulatory decision-making. However, RWE data sources remain fragmented, and datasets are disparate and often collected inconsistently. To this end, we have constructed an RWE-generation platform, Immunolab, to facilitate data-driven insights, hypothesis generation and research in immunological diseases driven by type 2 inflammation. Immunolab leverages a large, anonymized patient cohort from the Optum electronic health record and claims dataset containing over 17 million patient lives. Immunolab is an interactive platform that hosts three analytical modules: the Patient Journey Mapper, to describe the drug treatment patterns over time in patient cohorts; the Switch Modeler, to model treatment switching patterns and identify its drivers; and the Head-to-Head Simulator, to model the comparative effectiveness of treatments based on relevant clinical outcomes. The Immunolab modules utilize various analytic methodologies including machine learning algorithms for result generation which can then be presented in various formats for further analysis and interpretation.

The respective roles of controlled clinical trials and cohort monitoring studies in the pre- and postmarketing assessment of drugs.

The respective roles of controlled clinical trials and observational studies (cohort or case-control studies) in evaluating the efficacy, safety and usefulness of a drug were analysed. A randomised, controlled, double-blind study is the best method of estimating the efficacy of a treatment. It provides the least biased and most robust estimate of the causal relationship. In certain situations and on the basis of certain criteria, observational studies can have a proof-of-efficacy value. Randomised, controlled, pre- and postmarketing authorisation (MA) clinical studies identify the rarer adverse effects and compare them with those resulting from the reference treatment. Before the MA, the pooled safety data from different controlled trials can provide an estimation of relatively frequent adverse events and subjects at risk. However, an observational study is the most appropriate method of evaluating the safety of a drug in the currently used conditions. By definition, a drug influences the health of a population if it directly or indirectly improves its health. A drug would have a major role in public health if it reduced mortality or morbidity related to a particular disease or if it improved the quality of life of patients with this disease. Prior to marketing a product, modelling is the approach of choice to quantify the expected effect. Pragmatic, postmarketing trials and observational studies are the reference methods used to define the population affected, the efficacy and safety of the drug in a real situation and its usefulness for public health. In conclusion, randomised clinical trials remain the reference approach for evaluating efficacy, while observational studies have a confirmatory value. Observational studies are the most appropriate way of evaluating safety in the currently used conditions, as the clinical trial has limited indications. In the interests of public health, modelling is the pre-marketing approach of choice, while pragmatic trials and observational studies are the postmarketing reference approaches.

A multicenter, randomized, double-blind, 8-week comparative trial of low-dose esomeprazole (20 mg) and standard-dose omeprazole (20 mg) in patients with erosive esophagitis.

The objective of this trial was to compare the efficacy of esomeprazole, 20 mg, with that of omeprazole, 20 mg, in patients with erosive esophagitis (EE). In this multicenter, double-blind, parallel-group trial, 1176 patients with EE confirmed by endoscopy (Helicobacter pylori-negative by serology) were randomized to once-daily treatment with 20 mg esomeprazole or 20 mg omeprazole for 8 weeks. The primary outcome was the proportion of patients with healed EE through week 8. Secondary outcomes included diary and investigator assessments of heartburn symptoms. Cumulative life-table healing rates at week 8 were similarly high for 20 mg esomeprazole (90.6%; 95% confidence interval, 88.1%-93%) and 20 mg omeprazole (88.3%; 95% confidence interval, 85.5%-91.0%). The two treatments were comparable for other secondary measures and had similar tolerability profiles.

A multicenter, randomized, double-blind, 8-week comparative trial of standard doses of esomeprazole (40 mg) and omeprazole (20 mg) for the treatment of erosive esophagitis.

To compare esomeprazole with omeprazole for healing erosive esophagitis (EE), 1148 patients with endoscopically confirmed EE were randomized to once-daily esomeprazole, 40 mg, or omeprazole, 20 mg, for 8 weeks in this multicenter, double-blind, parallel-group trial. The primary outcome was the proportion of patients with healed EE at week 8. Secondary outcomes included diary and investigator assessments of heartburn symptoms. At week 8, estimated healing rates were 92.2% (95% CI, 89.9%-94.5%) with esomeprazole and 89.8% (95% CI, 87.2%-92.4%) with omeprazole. Healing rates with esomeprazole were significantly higher than those with omeprazole at weeks 8 (88.4% vs 77.5%; P = 0.007) and 4 (60.8% vs 47.9%; P = 0.02) in patients with moderate to severe (Los Angeles grade C or D) EE at baseline but were not significantly different for patients with mild (Los Angeles grade A or B) EE. Both treatments were comparable for other secondary measures and had similar tolerability profiles.