BACKGROUND: Patients routinely use Twitter to share feedback about their experience receiving healthcare. Identifying and analysing the content of posts sent to hospitals may provide a novel real-time measure of quality, supplementing traditional, survey-based approaches. OBJECTIVE: To assess the use of Twitter as a supplemental data stream for measuring patient-perceived quality of care in US hospitals and compare patient sentiments about hospitals with established quality measures. DESIGN: 404 065 tweets directed to 2349 US hospitals over a 1-year period were classified as having to do with patient experience using a machine learning approach. Sentiment was calculated for these tweets using natural language processing. 11 602 tweets were manually categorised into patient experience topics. Finally, hospitals with >/=50 patient experience tweets were surveyed to understand how they use Twitter to interact with patients. KEY RESULTS: Roughly half of the hospitals in the US have a presence on Twitter. Of the tweets directed toward these hospitals, 34 725 (9.4%) were related to patient experience and covered diverse topics. Analyses limited to hospitals with >/=50 patient experience tweets revealed that they were more active on Twitter, more likely to be below the national median of Medicare patients (p<0.001) and above the national median for nurse/patient ratio (p=0.006), and to be a non-profit hospital (p<0.001). After adjusting for hospital characteristics, we found that Twitter sentiment was not associated with Hospital Consumer Assessment of Healthcare Providers and Systems (HCAHPS) ratings (but having a Twitter account was), although there was a weak association with 30-day hospital readmission rates (p=0.003). CONCLUSIONS: Tweets describing patient experiences in hospitals cover a wide range of patient care aspects and can be identified using automated approaches. These tweets represent a potentially untapped indicator of quality and may be valuable to patients, researchers, policy makers and hospital administrators.

OBJECTIVE: To characterize the conclusions and production of nonsystematic reviews about neuraminidase inhibitors relative to financial competing interests held by the authors. STUDY DESIGN AND SETTING: We searched for articles about neuraminidase inhibitors and influenza (January 2005 to April 2015), identifying nonsystematic reviews and grading them according to the favorable/nonfavorable presentation of evidence on safety and efficacy. We recorded financial competing interests disclosed in the reviews and from other articles written by their authors. We measured associations between competing interests, author productivity, and conclusions. RESULTS: Among 213 nonsystematic reviews, 138 (65%) presented favorable conclusions. Financial competing interests were identified for 26% (137/532) of authors; 51% (108/213) of reviews were associated with a financial competing interest. Reviews produced exclusively by authors with financial competing interests (33%; 71/213) were more likely to present favorable conclusions than reviews with no competing interests (risk ratio 1.27; 95% confidence interval 1.03-1.55). Authors with financial competing interests published more articles about neuraminidase inhibitors than their counterparts. CONCLUSION: Half of nonsystematic reviews about neuraminidase inhibitors included an author with a financial competing interest. Reviews produced exclusively by these authors were more likely to present favorable conclusions, and authors with financial competing interests published a greater number of reviews.

Conflicts of interest held by researchers remain a focus of attention in clinical research. Biases related to these relationships have the potential to directly impact the quality of healthcare by influencing decision-making, yet conflicts of interest remain under-reported, inconsistently described, and difficult to access. Initiatives aimed at improving the disclosure of researcher conflicts of interest are still in their infancy but represent a vital reform that must be addressed before potential biases associated with conflicts of interest can be mitigated, and trust in the impartiality of clinical evidence restored. In this review, we examine the prevalence of conflicts of interest, evidence of the effects that disclosed and undisclosed conflicts of interest have had on the reporting of clinical evidence, and the emerging approaches for improving the completeness and consistency of disclosures. Through this review of emerging technologies, we recognize a growing interest in publicly-accessible registries for researcher conflicts of interest, and propose five desiderata aimed at maximizing the value of such registries: mandates for ensuring that researchers keep their records up to date; transparent records that are made available to the public; interoperability to allow researchers, bibliographic databases, and institutions to interact with the registry; a consistent taxonomy for describing different classes of conflicts of interest, and the ability to automatically generate conflicts of interest statements for use in published articles.

BACKGROUND: Elderly patients represent the greatest consumers of healthcare per capita but have historically been underrepresented in clinical trials. It is unknown how many trials are designed to focus exclusively on elderly patients. OBJECTIVE: To define the prevalence of interventional trials that study exclusively elderly persons and describe the characteristics of these trials, including their distribution across conditions most prevalent in the elderly. DESIGN: All interventional clinical trials enrolling exclusively elderly patients (>/=65 years), conducted primarily in high-income countries, and initiated between 2006 and 2014, identified through ClincialTrials.gov. MAIN MEASURES: Trials were identified and characterized according to design features and disease categories studied. Across disease categories we examined the burden of disease in the elderly in high-income countries (measured in disability-adjusted life years [DALYs]) and compared to the number of trials conducted exclusively in the elderly. RESULTS: Among 80,965 interventional trials, 1,112 (1.4%) focused on elderly patients. Diverse types of interventions were studied in these trials (medications 33%, behavioral interventions 18%, and dietary supplements 10%) and the majority was funded by non-profit organizations (81%). Studies tended to be small (median sample size 122 participants [IQR 58, 305]), single-center studies (67%). Only 43% of 126 disease categories affecting elderly persons were studied in trials focused on the elderly. Among these disease categories, there was a 5162-fold range in the ratio of DALYs per trial. Across 5 conditions where over 80% of DALYs are in the elderly, there were a total of only 117 trials done exclusively in the elderly. CONCLUSIONS: Very few and mostly small studies are conducted exclusively in elderly persons, even for conditions that affect almost exclusively the elderly.

OBJECTIVES: To examine the use of supervised machine learning to identify biases in evidence selection and determine if citation information can predict favorable conclusions in reviews about neuraminidase inhibitors. STUDY DESIGN AND SETTING: Reviews of neuraminidase inhibitors published during January 2005 to May 2013 were identified by searching PubMed. In a blinded evaluation, the reviews were classified as favorable if investigators agreed that they supported the use of neuraminidase inhibitors for prophylaxis or treatment of influenza. Reference lists were used to identify all unique citations to primary articles. Three classification methods were tested for their ability to predict favorable conclusions using only citation information. RESULTS: Citations to 4,574 articles were identified in 152 reviews of neuraminidase inhibitors, and 93 (61%) of these reviews were graded as favorable. Primary articles describing drug resistance were among the citations that were underrepresented in favorable reviews. The most accurate classifier predicted favorable conclusions with 96.2% accuracy, using citations to only 24 of 4,574 articles. CONCLUSION: Favorable conclusions in reviews about neuraminidase inhibitors can be predicted using only information about the articles they cite. The approach highlights how evidence exclusion shapes conclusions in reviews and provides a method to evaluate citation practices in a corpus of reviews.

BACKGROUND: Industry-sponsored clinical trials, in the past performed almost exclusively in more developed countries, now often recruit participants globally. However, recruitment from outside high-income countries may not represent the ultimate target population for the intervention. Clinical trial registries provide an opportunity to quantify and examine the type of clinical research performed in various geographic regions. We sought to characterize industry-sponsored randomized controlled trials conducted in high-income countries and to compare these trials to those performed outside high-income countries. METHODS: Clinical trial data on all industry-funded randomized controlled trials conducted between 2006 and 2014 were obtained from the registry ClinicalTrials.gov. Trials were classified according to their study sites as conducted in high or non-high income countries, and data on trial characteristics were collected. RESULTS: Of 22,511 relevant trials, a total of 6,085 (27.0 %) trials included study sites outside a high-income country, and 2,045 (9.1 %) were conducted exclusively outside high-income countries. Of country groups, Central Europe had the greatest number of trials (3,127), followed by Eastern Europe (2,075). The percentage of trials with study sites outside high-income countries remained relatively constant over the study period. Studies with sites outside high-income countries tended to recruit more participants (median enrolled participants 265 vs. 71, P <0.001), to be longer (median study duration 20 vs. 13 months, P <0.05), and to study more advanced phase interventions (Phase 3 or 4 trial 58 % vs. 33 %, P <0.001). CONCLUSIONS: More than a quarter of industry-sponsored trials include participants from outside high-income countries and this rate remained stable over the 7-year study period. Trials conducted outside high-income countries tend to be larger, have a longer duration, and study later phase interventions compared to studies performed exclusively in high-income countries.

OBJECTIVES: Emergency department (ED) utilization by children is common and growing more expensive. Tracking trends and variability in ED charges is essential for policymakers who strive to improve the efficiency of the health care system and for payers who prepare health care budget forecasts. Our objective was to examine trends and variability in ED charges for pediatric patients across Massachusetts. METHODS: This was a comprehensive analysis of the statewide database containing all the visits of children aged 0 to 18 years evaluated in any of the state's EDs from 2000 to 2011, excluding patients with chronic medical conditions and those whose visits resulted in hospital admission. A validated system designed to specifically classify pediatric emergency patients into major diagnostic groups was used. Mean charges as well as interhospital variability of charges over time were examined for the most common diagnostic groups. RESULTS: Seventy-six hospitals provided emergency care in Massachusetts during the study period, with 6,249,923 pediatric patients treated and discharged. Statewide charges significantly increased from 2000 until 2007/2008, before plateauing or decreasing through 2011. There was no evidence that interhospital variability changed over time. With the exception of academic teaching status, no hospital-level factors emerged as consistent predictors of charges. CONCLUSIONS: Charges for common pediatric emergency conditions varied widely across Massachusetts EDs, and hospital-level factors by and large could not consistently explain the variability. Although a plateau (and in some cases decrease) of statewide pediatric emergency health care charges was observed after 2007, no evidence was found that interhospital variability decreased. These data may be useful in the ongoing effort to reform the economics of health care delivery systems.

We examined a process for automating the classification of articles in MEDLINE aimed at minimising manual effort without sacrificing accuracy. From 22,808 articles pertaining to 19 antidepressants, 1000 were randomly selected and manually labelled according to article type (including, randomised controlled trials, editorials, etc.). We applied a machine learning approach termed 'active learning', where the learner (machine) selects the order in which the user (human) labels examples. Via simulation, we determined the number of articles a user needed to label to produce a classifier with at least 95% recall and 90% precision in three scenarios related to evidence synthesis. We found that the active learning process reduced the number of training instances required by 70%, 19%, and 14% in the three scenarios. The results show that the active learning method may be used in some scenarios to produce accurate classifiers that meet the needs of evidence synthesis tasks and reduce manual effort.

PURPOSE: Physicians require information on the comparative benefits and harms of medications for optimal treatment decisions. However, this type of data is limited, especially for pediatric patients. OBJECTIVE: Our aim was to use observational data to measure and compare medication utilization patterns in a pediatric patient population. METHODS: Using pharmacy claims data from a large, national-scale insurance program in the USA, we identified all patients with a diagnosis of epilepsy treated with a first-generation antiepileptic drug (carbamazepine, ethosuximide, phenobarbital, phenytoin, or valproate) or a second-generation antiepileptic drug [carbamazepine extended release (XR), gabapentin, lamotrigine, levetiracetam, oxcarbazepine, tiagabine, topiramate, valproate XR, or zonisamide]. Treatment periods were defined on the basis of prescription fill dates and medication days supplied. Medication use was measured for individual antiepileptic drugs and for first-generation and second-generation drugs as groups. RESULTS: There were 2527 patients (54 %) who initiated therapy with first-generation antiepileptics and 2139 patients (46 %) who initiated therapy with second-generation antiepileptics. First- and second-generation drugs had the same 1-year retention rates [26 % (95 % confidence interval (CI) 24-28) and 26 % (95 % CI 25-28), respectively], and 26 % of patients (95 % CI 25-28) and 29 % of patients (95 % CI 27-31) who started on a first- or second-generation antiepileptic medication, respectively, resumed treatment with the initial drug after discontinuation. Overall, 73 % of patients (95 % CI 71-74) were treated with only one antiepileptic drug, with similar rates for patients started on first- and second-generation drugs [71 % (95 % CI 69-73) versus 74 % (95 % CI 72-76)]. CONCLUSION: Comparing drug utilization patterns in a pediatric population using observational data, we found similar rates of retention and therapeutic changes. These findings are consistent with the available comparative data and demonstrate an approach that could be extended to other drug classes and conditions in pediatric populations to examine drug effectiveness.

OBJECTIVES: To examine the use of supervised machine learning to identify biases in evidence selection and determine if citation information can predict favorable conclusions in reviews about neuraminidase inhibitors. STUDY DESIGN AND SETTING: Reviews of neuraminidase inhibitors published during January 2005 to May 2013 were identified by searching PubMed. In a blinded evaluation, the reviews were classified as favorable if investigators agreed that they supported the use of neuraminidase inhibitors for prophylaxis or treatment of influenza. Reference lists were used to identify all unique citations to primary articles. Three classification methods were tested for their ability to predict favorable conclusions using only citation information. RESULTS: Citations to 4,574 articles were identified in 152 reviews of neuraminidase inhibitors, and 93 (61%) of these reviews were graded as favorable. Primary articles describing drug resistance were among the citations that were underrepresented in favorable reviews. The most accurate classifier predicted favorable conclusions with 96.2% accuracy, using citations to only 24 of 4,574 articles. CONCLUSION: Favorable conclusions in reviews about neuraminidase inhibitors can be predicted using only information about the articles they cite. The approach highlights how evidence exclusion shapes conclusions in reviews and provides a method to evaluate citation practices in a corpus of reviews.