Assessing COVID-19 Health Information on Google Using the Quality Evaluation Scoring Tool (QUEST): Cross-sectional and Readability Analysis.

BACKGROUND: The COVID-19 pandemic spurred an increase in online information regarding disease spread and symptomatology. OBJECTIVE: Our purpose is to systematically assess the quality and readability of articles resulting from frequently Google-searched COVID-19 terms in the United States. METHODS: We used Google Trends to determine the 25 most commonly searched health-related phrases between February 29 and April 30, 2020. The first 30 search results for each term were collected, and articles were analyzed using the Quality Evaluation Scoring Tool (QUEST). Three raters scored each article in authorship, attribution, conflict of interest, currency, complementarity, and tone. A readability analysis was conducted. RESULTS: Exactly 709 articles were screened, and 195 fulfilled inclusion criteria. The mean article score was 18.4 (SD 2.6) of 28, with 7% (14/189) scoring in the top quartile. National news outlets published the largest share (70/189, 36%) of articles. Peer-reviewed journals attained the highest average QUEST score compared to national/regional news outlets, national/state government sites, and global health organizations (all P<.05). The average reading level was 11.7 (SD 1.9, range 5.4-16.9). Only 3 (1.6%) articles were written at the recommended sixth grade level. CONCLUSIONS: COVID-19-related articles are vastly varied in their attributes and levels of bias, and would benefit from revisions for increased readability.

Current Trends and Predictors of Case Outcomes for Malpractice in Colonoscopy in the United States.

BACKGROUND: Over 14 million colonoscopies are performed annually, and this procedure remains the largest contributor to malpractice claims against gastroenterologists. The aim of this study was to evaluate reasons for litigation and predictors of case outcomes. MATERIALS AND METHODS: Cases related to colonoscopy were reviewed within the Westlaw legal database. Patient demographics, reasons for litigation, case payouts, and verdicts were assessed. Multivariate regression was used to determine predictors of defendant verdicts. RESULTS: A total of 305 cases were included from years 1980 to 2017. Average patient age was 54.9 years (range, 4 to 93) and 52.8% of patients were female. Juries returned defendant and plaintiff verdicts in 51.8% and 25.2% of cases, respectively, and median payout was $995,000. Top reasons for litigation included delay in treatment (65.9%) and diagnosis (65.6%), procedural error (44.3%), and failure to refer (25.6%). Gastroenterologists were defendants in 71% of cases, followed by primary care (32.2%) and surgeons (14.8%). Cases citing informed consent predicted defendant verdict (odds ratio, 4.05; 95% confidence interval, 1.90-9.45) while medication error predicted plaintiff verdict (odds ratio, 0.18; 95% confidence interval, 0.04-0.59). Delay in diagnosis (P=0.060) and failure to refer (P=0.074) trended toward plaintiff verdict but did not reach significance. Most represented states were New York (21.0%), California (13.4%), Pennsylvania (13.1%), Massachusetts (12.5%). CONCLUSIONS: Malpractice related to colonoscopy remains a significant and has geographic variability. Errors related to sedation predicted plaintiff verdict and may represent a target to reduce litigation. Primary care physicians and surgeons were frequently cited codefendants, underscoring the significance of interdisciplinary care for colonoscopy.

#radonc: Growth of the global radiation oncology Twitter network.

INTRODUCTION: Social media connects people globally and may enhance access to radiation oncology information. We characterized the global growth of the radiation oncology Twitter community using the hashtag #radonc. MATERIALS AND METHODS: We analyzed all public tweets bearing the hashtag #radonc from 2014 to 2019 using Symplur Signals. We collected data on #radonc activity and growth, stakeholder distribution, user geolocation, and languages. We obtained global Twitter user data and calculated average annual growth rates for users and tweets. We analyzed growth rates by stakeholder. We conducted thematic analysis on a sample of tweets in each three-year period using frequently occurring two-word combinations. RESULTS: We identified 193,115 tweets including #radonc composed by 16,645 Twitter users. Globally, users wrote in 35 languages and came from 122 countries, with the known highest users from the United States, United Kingdom, and Spain. Use of #radonc expanded from 23 countries in 2014 to 116 in 2019. The average annual growth rate in #radonc users and tweets was 70.5% and 69.2%, respectively. The annual growth rate of #radonc users was significantly higher than for all Twitter users (p = 0.004). While doctors were the source of 46.9% of all tweets, research and government organizations had annual increases in tweet volume of 84.6% and 211.4%, respectively. From 2014 to 2016, promotion of the radiation oncology community was the most active theme, though this dropped to 7th in 2017-2019 as discussion increased regarding aspects of radiation and treated disease sites. CONCLUSION: Use of #radonc has grown rapidly into a global community. Focused discussion related to radiation oncology has outpaced the growth of general Twitter use, both among physicians and non-physicians. #radonc has grown into a self-sustaining community. Further research is necessary to define the risks and benefits of social media in medicine and to determine whether it adds value to oncology practice.

The #PalliativeCare Conversation on Twitter: An Analysis of Trends, Content, and Caregiver Perspectives.

CONTEXT: Palliative care is known to improve patients' quality of life, but oftentimes these conversations occur outside of the health-care setting. OBJECTIVES: To characterize the #PalliativeCare Twitter network and evaluate the caregiver experience within palliative care. METHODS: In this cross-sectional study, a total of 182,661 #PalliativeCare tweets by 26,837 users from June 1, 2015 to June 1, 2019 were analyzed using Symplur Signals. Analysis included activity metrics, content analysis, user characteristics, engagement, and network analysis. Similar metrics were performed on tweets by self-identified caregivers (482), who wrote a total of 3952 tweets. Qualitative analysis was completed on a systematic sample of caregiver tweets. RESULTS: The number of #PalliativeCare tweets, users, and impressions has increased by an annual average of 18.7%, 16.4%, and 32.5%, respectively. Support, access, and patients were among the Trending Terms. About 39.4% of Trending Articles were scientifically valid, and information about palliative care and comorbidities had the greatest number of articles. The majority of users wrote five or less #PalliativeCare tweets. Network analysis revealed central hubs to be palliative care advocacy organizations and physicians. The five main themes from qualitative analysis of caregiver tweets were 1) advocacy and events, 2) care strategies, 3) resources, 4) public health issues, and 5) myths related to palliative care. CONCLUSION: The use of Twitter as a platform for palliative care conversations is growing rapidly. Twitter serves as a platform to facilitate #PalliativeCare conversation among patients, caregivers, physicians, and other healthcare providers.

Evaluating the Readability of Online Patient Education Materials Related to Orthopedic Oncology.

The internet is increasingly used to access patient education materials. The average American reading level has been found to be that of a 7th- to 8th-grade student, prompting the National Institutes of Health (NIH) and the American Medical Association (AMA) to advise that patient education materials be written between the 4th- to 6th-grade reading level. The purpose of this study was to evaluate the reading level of current patient education materials for the most common musculoskeletal oncological tumors. A Google search was performed with all location filters off to account for geographic variability for patient education materials related to 28 orthopedic primary or secondary tumors. All patient education articles from the first 10 website hits for each tumor type were analyzed. Patient education materials from these websites were evaluated using 8 validated readability scales. Patient resources were found to be written at an average grade level nearly double the NIH and AMA recommendation. Patient education materials for soft tissue chondromas were written at the highest level (14.8±1.9), whereas education materials for chordomas (10.1±1.0) most closely approached national recommendations, despite still being written at a readability level nearly 4 grade levels higher than has been recommended. The Flesch Reading Ease assessment provided a mean score of 46.5±7.7, corresponding with a "difficult to read" result. Current patient education materials regarding oncological musculoskeletal-related patient education materials are written significantly above the recommended reading level. Further modification of these resources is warranted to ensure adequate comprehension and informed decision making in the clinical setting. [Orthopedics. 2021;44(1):38-42.].

Evaluating the Readability of Online Patient Education Materials Regarding Shoulder Surgery: How Do Medical Institution Web Sites Rate?

This study evaluated the readability of online patient education materials on shoulder surgery. Medical and nonmedical institution web sites were compared and it was hypothesized that medical institution materials are written at lower grade levels than nonmedical institution materials, because medical institutions understand physician-patient interactions. Eighty-six articles were scored according to 10 readability tests: cumulative combined average grade level was 12.5 ± 2.8 and average Flesch reading score was 43.5 ± 12.6 (college level). The average composite grade level readability for medical institution web sites was 13 ± 2, significantly higher than for nonmedical institution web sites (11.9 ± 2.1; p = .017). Patient education materials available online are written at a higher level than American Medical Association and National Institutes of Health guidelines. Medical institution articles are written at a statistically significant higher grade level than nonmedical institution articles, but the difference is small and both rate poorly compared with current standards. (Journal of Surgical Orthopaedic Advances 28(3):209-214, 2019).

What patients see online: assessing the online identities of Pennsylvania dermatologists.

INTRODUCTION: Patients use the internet to search for health-related information. We sought to characterize the information that patients find when searching for dermatologists on Google. METHODS: The Centers for Medicare and Medicaid Services (CMS) Physician Comparable Downloadable File was utilized to identify all Medicare-participating dermatologists practicing in Pennsylvania (PA). A custom Google-based search engine was used to search each dermatologist. Up to the top 10 results for each physician were then sorted into: (1) physician, hospital, or healthcare system, (2) third-party, (3) social media, (4) academic journal articles, or (5) other. RESULTS: Within the CMS, 519 health care providers (53.9% male, 46.1% female) self-identified as dermatologists practicing in PA. At least one search result was obtained for each physician (4,963 total search results). About 30.6% (1,519) search results were hospital, health system, or physician-controlled websites, and 26.6% (1,318) were third-party websites (1,318; 26.6%). Social media websites accounted for 601 (12.1%) hits whereas peer-reviewed academic journal websites generated 135 (2.7%) results. One-way chi-square analysis showed domains were not randomly distributed across the five categories (P<0.0001). CONCLUSION: Dermatologists should be better aware of their digital presence and the strategies to better control their online identity.

A Qualitative Analysis of Malpractice Litigation in Cardiology Using Case Summaries Through a National Legal Database Analysis.

Introduction Physicians are increasingly practicing defensive medicine as a response to society's litigious climate. This study sought to characterize cardiology malpractice claims and elucidate the allegations underlying the use of defensive medicine. Methods The WestlawNext™ database was queried to obtain state and federal jury verdicts and settlements related to medical malpractice and cardiology that occurred in the United States between 2010 and 2015. Cardiology cases were identified using the search terms "medical malpractice" and "cardiology" and reviewed by two individuals utilizing available case documents. Duplicate and nonpertinent cases were excluded. Binary logistic regression models were created to predict the likelihood of defendant verdict, plaintiff verdict, and settlement based on the various reasons for litigation cited. Results Inclusion criteria were met in 166 cases. The plaintiffs were predominantly male (94 cases; 56.6%), and the average patient age was 53.3±17.5 years. More than half of the cases involved a cardiologist as a defendant. The most common reasons for litigation were: failure to treat (129; 77.7%), failure to diagnose (115; 69.3%), failure to refer/order diagnostic tests (107; 64.5%), and patient death (118; 71.1%). Among cases tried for failure to diagnose, the most commonly missed diagnosis was myocardial infarction. Cases most commonly resulted in a defendant verdict (94; 56.6%). However, odds of a plaintiff verdict were significantly higher when failure to diagnose was alleged with an odds ratio (OR) of 7.60 (95% confidence interval 1.14 - 50.87, p = 0.0365). Conclusions Failure to diagnose remains a commonly alleged base for litigation. In conclusion, our analysis suggests increased training for non-cardiologists in the recognition of the acute coronary syndrome and enhanced awareness of inherent biases among all physicians may facilitate reducing missed diagnoses.

Mobile health applications for atrial fibrillation: A readability and quality assessment.

BACKGROUND: Mobile health applications may improve patient education and self-care for a complex condition such as atrial fibrillation (AF). Little is known about the accessibility of mobile health applications ("apps") and their readability. We evaluated the readability and quality of available apps for AF. METHODS: We searched the Apple and Google Play app stores with the terms "atrial fibrillation" and "afib." We downloaded English-language apps (up to n = 100 for each term) and categorized them by name, App store, cost, content, uploading agency (heath care associated [HCA] versus non-HCA), target audience (health care professional [HCP] versus non-HCP), scientific validity (i.e., citation of peer-reviewed or validated medical information), and user ratings. We analyzed the text of apps intended for a non-HCP target audience for readability with 10 established measures. RESULTS: Of the 206 downloaded apps, 50.5% were excluded as unrelated to AF, inaccessible, or non-English language. The majority of apps contained information about AF (63.2% Apple, 52.2% Google Play) and AF detection (52.6% Apple, 56.5% Google Play). A minority of non-HCP apps contained scientifically validated content (Apple, 15.8%; Google Play, 13.0%; P = NS). App mean readability was grade 12.1 ±â€¯2.6. CONCLUSIONS: Most AF apps lacked scientific validation and were written at excessively high reading-grade levels. Our results suggest caution with mobile health apps, particularly for users with limited health literacy. There is potential opportunity for a multi-disciplinary effort by regulatory agencies, healthcare organizations, and app stores to improve relevance, scientific validity, and readability of AF apps for patients with this complex and morbid disease.

A Readability Analysis of Online Cardiovascular Disease-Related Health Education Materials.

BACKGROUND: Online cardiovascular health materials are easily accessible with an Internet connection, but the readability of its content may limit practical use by patients. OBJECTIVE: The goal of our study was to assess the readability of the most commonly searched Internet health education materials for cardiovascular diseases accessed via Google. METHODS: We selected 20 commonly searched cardiovascular disease terms: aneurysm, angina, atherosclerosis, cardiomyopathy, congenital heart disease, coronary artery disease, deep vein thrombosis, heart attack, heart failure, high blood pressure, pericardial disease, peripheral arterial disease, rheumatic heart disease, stroke, sudden death, valvular heart disease, mini-stroke, lower extremity edema, pulmonary embolism, and exertional dyspnea. Terms were selected on Google and selected up to 10 results in order of presentation in the search results by reviewing a maximum of 15 pages of Google search results specifically providing education toward patients to yield 196 total patient education articles. KEY RESULTS: All readability measures assessing grade level measures found the 196 articles were written at a mean 10.9 (SD = 1.8) grade reading level. Moreover, 99.5% of the articles were written beyond the 5th- to 6th-grade level recommended by the American Medical Association. CONCLUSIONS: Given the prominent use of online patient education material, we consider readability as a quality metric that should be evaluated prior to online publication of any health education materials. Further study of how to improve the readability of online materials may enhance patient education, engagement, and health outcomes.

Medical Malpractice in Orthopedic Surgery: A Westlaw-Based Demographic Analysis.

A recent study that evaluated the risk of facing a malpractice claim by physician specialty found that orthopedic surgeons were at a significantly greater risk of being sued than other medical specialists. To date, no studies have characterized trends in orthopedic surgery malpractice claims. The Westlaw legal database was used to locate state and federal jury verdicts and settlements related to medical malpractice and orthopedic surgery from 2010 to 2016. Eighty-one cases were analyzed. The mean age of the affected patients and/or plaintiffs was 53.4 years. Spine surgery (21 cases; 25.9%), knee surgery (17 cases; 21.0%), and hip surgery (11 cases; 13.6%) were litigated most often. Procedural error (71 cases; 87.7%) and negligence (58 cases; 71.6%) were the 2 most commonly cited reasons for litigation. The jury found in favor of the defendant in most (50 cases; 61.7%) of the cases. The mean plaintiff (17 cases; 21.0%) verdict payout was $3,015,872, and the mean settlement (13 cases; 16.0%) value was $1,570,833. Unnecessary surgery (odds ratio [OR], 12.29; 95% confidence interval [CI], 1.91-108.46; P=.040) and surgery resulting in death (OR, 26.26; 95% CI, 2.55-497.42; P=.040) were significant predictors of a verdict in favor of the plaintiff. Patient death (OR, 0.05; 95% CI, 0.01-0.38; P=.021) and male patient sex (OR, 0.26; 95% CI, 0.09-0.71; P=.033) were significant negative predictors of a verdict in favor of the defendant. The jury found in favor of the defendant orthopedic surgeon in most cases. Procedural error and/or negligence were cited most commonly by the plaintiffs as the bases for the claims. Verdicts in favor of the plaintiffs resulted in payouts nearly double those of settlements. [Orthopedics. 2018; 41(5):e615-e620.].

Abdominal imaging and patient education resources: enhancing the radiologist-patient relationship through improved communication.

INTRODUCTION: The relative ease of Internet access and its seemingly endless amount of information creates opportunities for Americans to research medical diseases, diagnoses, and treatment plans. Our objective is quantitative evaluation of the readability level of patient education websites, written for the lay public, pertaining to common radiologic diagnostic test, and radiologic diagnoses specific to abdominal imaging. METHODS: In October 2015, 10 search terms were entered in the Google search engine, and the top 10 links for each term were collected and independently examined for their readability level using 10 well-validated quantitative readability scales. Search terms included CT abdomen, MRI abdomen, MRI enterography, ultrasound abdomen, X-ray abdomen, cholecystitis, diverticulitis, hepatitis, inflammatory bowel disease, and pancreatitis. Websites not written exclusively for patients were excluded from the analysis. RESULTS: As a group, the 100 articles were assessed at an 11.7 grade level. Only 2% (2/100) were written at the National Institutes of Health (NIH), and American Medical Association (AMA) suggested 3rd to 7th grade level to meet the 8th grade average reading level in the United States. In fact, 49% were written at a level that required a high school education or higher (greater than 12th grade). CONCLUSIONS: With websites like radiologyinfo.org, generating over a million visitors a month, it is that clear there is a public interest in learning about radiology. However, given the discordance between the level of readability of the majority of the Internet articles and the NIH and AMA guidelines noted in this study on abdominal imaging readability, it is likely that many readers do not fully benefit from these resources on abdominal imaging.