Correction: Using Natural Language Processing to Examine the Uptake, Content, and Readability of Media Coverage of a Pan-Canadian Drug Safety Research Project: Cross-Sectional Observational Study.

[This corrects the article DOI: 10.2196/13296.].

Using Natural Language Processing to Examine the Uptake, Content, and Readability of Media Coverage of a Pan-Canadian Drug Safety Research Project: Cross-Sectional Observational Study.

BACKGROUND: Isotretinoin, for treating cystic acne, increases the risk of miscarriage and fetal abnormalities when taken during pregnancy. The Health Canada-approved product monograph for isotretinoin includes pregnancy prevention guidelines. A recent study by the Canadian Network for Observational Drug Effect Studies (CNODES) on the occurrence of pregnancy and pregnancy outcomes during isotretinoin therapy estimated poor adherence to these guidelines. Media uptake of this study was unknown; awareness of this uptake could help improve drug safety communication. OBJECTIVE: The aim of this study was to understand how the media present pharmacoepidemiological research using the CNODES isotretinoin study as a case study. METHODS: Google News was searched (April 25-May 6, 2016), using a predefined set of terms, for mention of the CNODES study. In total, 26 articles and 3 CNODES publications (original article, press release, and podcast) were identified. The article texts were cleaned (eg, advertisements and links removed), and the podcast was transcribed. A dictionary of 1295 unique words was created using natural language processing (NLP) techniques (term frequency-inverse document frequency, Porter stemming, and stop-word filtering) to identify common words and phrases. Similarity between the articles and reference publications was calculated using Euclidian distance; articles were grouped using hierarchical agglomerative clustering. Nine readability scales were applied to measure text readability based on factors such as number of words, difficult words, syllables, sentence counts, and other textual metrics. RESULTS: The top 5 dictionary words were pregnancy (250 appearances), isotretinoin (220), study (209), drug (201), and women (185). Three distinct clusters were identified: Clusters 2 (5 articles) and 3 (4 articles) were from health-related websites and media, respectively; Cluster 1 (18 articles) contained largely media sources; 2 articles fell outside these clusters. Use of the term isotretinoin versus Accutane (a brand name of isotretinoin), discussion of pregnancy complications, and assignment of responsibility for guideline adherence varied between clusters. For example, the term pregnanc appeared most often in Clusters 1 (14.6 average times per article) and 2 (11.4) and relatively infrequently in Cluster 3 (1.8). Average readability for all articles was high (eg, Flesch-Kincaid, 13; Gunning Fog, 15; SMOG Index, 10; Coleman Liau Index, 15; Linsear Write Index, 13; and Text Standard, 13). Readability increased from Cluster 2 (Gunning Fog of 16.9) to 3 (12.2). It varied between clusters (average 13th-15th grade) but exceeded the recommended health information reading level (grade 6th to 8th), overall. CONCLUSIONS: Media interpretation of the CNODES study varied, with differences in synonym usage and areas of focus. All articles were written above the recommended health information reading level. Analyzing media using NLP techniques can help determine drug safety communication effectiveness. This project is important for understanding how drug safety studies are taken up and redistributed in the media.

Development of a clinical decision support tool for diagnostic imaging use in patients with low back pain: a study protocol.

BACKGROUND: Low back pain is one of the most common and disabling health problems in Canada and internationally. In most cases, low back pain is a benign, self-limiting condition that can be managed with little diagnostic investigation or treatment. Yet contrary to clinical practice guideline recommendations, diagnostic imaging (here meaning X-ray, MRI, CT) is commonly used in the assessment of low back pain. Diagnostic imaging is of limited value in most cases, exposing patients to unnecessary radiation and leading to increased health services use and worse patient health outcomes. The Choosing Wisely campaign has highlighted the need to reduce diagnostic imaging for low back pain; however, no clinical decision rules are available. METHODS: This project will develop a clinical decision support tool for appropriate use of diagnostic imaging for patients with low back pain in the emergency department. We will conduct a prospective cohort study at five Canadian emergency departments. The study will follow recommendations for prediction model development and testing. The study population will be 4000 patients presenting to the emergency department with low back pain. We will assess potential clinical indications of emergent-cause (i.e., "red flag" items), including clinical characteristics and past history. Our outcome, emergent-cause for low back pain such as fracture, cancer, infection, or cauda equina syndrome, will be assessed at discharge and at 1-, 3-, and 12-month follow-up periods using information from self-report and health administrative data. We will construct and assess the performance of a multivariable prediction model that has strong measurement properties, presented as a clinical decision support tool acceptable to knowledge users. DISCUSSION: Practice guidelines describe "red flags" for which diagnostic imaging is likely appropriate. However, recommendations across guidelines are discordant, and few studies have evaluated these criteria to determine which characteristics best predict emergent etiology that warrant diagnostic imaging. A clinical decision support tool, that recommends diagnostic imaging where appropriate, has the potential to improve clinical care and patient outcomes and reduce costs associated with managing low back pain patients.

Evaluating quality of life and cost implications of prophylactic radiotherapy in mesothelioma: Health economic analysis of the SMART trial.

BACKGROUND: The SMART trial is a UK-based, multicentre RCT comparing prophylactic radiotherapy and symptom-based (deferred) radiotherapy in 203 patients with Malignant Pleural Mesothelioma who had undergone large bore pleural interventions. Using costs and quality of life data collected alongside the clinical trial, we will estimate the cost-effectiveness of prophylactic radiotherapy compared to deferred radiotherapy over a 1-year period. METHODS: Healthcare utilization and costs were captured during the trial. Utility weights produced by the EQ-5D questionnaire were used to determine quality-adjusted life-years (QALY) gained. The incremental cost-effectiveness ratio was calculated over the one-year trial period. RESULTS: Costs were similar in the immediate and deferred radiotherapy groups: £5480.40 (SD = £7040; n = 102) and £5461.40 (SD = £7770; n = 101) respectively. There was also no difference in QALY: 0.498 (95% CI: [0.45, 0.547]) in the prophylactic radiotherapy group versus 0.525 (95% CI: [0.471, 0.580]) in the deferred group. At a willingness to pay threshold of £30,000/QALY there was only a 24% chance that prophylactic radiotherapy was cost-effective compared to deferred radiotherapy. CONCLUSIONS: There was no significant effect of prophylactic radiotherapy on quality of life in the intervention group, nor was there any discernable decrease in healthcare costs. There is little evidence to suggest that prophylactic radiotherapy is a cost-effective intervention in this population. TRIAL REGISTRATION: ISRCTN72767336 with ISRCTN.

Addressing rural and remote access disparities for patients with inflammatory arthritis through video-conferencing and innovative inter-professional care models.

OBJECTIVE: The aim of the present study was to evaluate whether rheumatoid arthritis (RA) patients followed longitudinally using video-conferencing and inter-professional care support have comparable disease control to those followed in traditional in-person rheumatology clinics. METHODS: This was a randomized controlled trial for 85 RA patients allocated to either traditional in-person rheumatology follow-up or video-conferenced follow-up with urban-based rheumatologists and rural in-person physical therapist examiners. Follow-up was every 3 months for 9 months. Outcome measures included disease activity metrics (disease activity in 28 joints with CRP measure score [DAS28-CRP], and RA disease activity index [RADAI]), modified health assessment questionnaire (mHAQ), quality of life (EuroQOL five dimensions questionnaire [EQ5D]) and patient satisfaction (nine-item visit-specific satisfaction questionnaire [VSQ9]). RESULTS: Of 85 participants, 54 were randomized to the video-conferencing team model and 31 to the traditional clinic (control group). Dropout rates were high, with only 31 (57%) from the video-conferencing and 23 (74%) from the control group completing the study. The mean age for study participants was 56 years; 20% were male. Mean RA disease duration was 13.9 years. There were no significant between-group differences in DAS28-CRP, RADAI, mHAQ or EQ5D scores at baseline or over the study period. Satisfaction rates were high in both groups. CONCLUSIONS: We found no evidence of a difference in effectiveness between inter-professional video-conferencing and traditional rheumatology clinic for both the provision of effective follow-up care and patient satisfaction for established RA patients. High dropout rates reinforce the need for consultation with patients' needs and preferences in developing models of care. While use of video-conferencing/telehealth technologies may be a distinct advantage for some patients, there may be loss of travel-related auxiliary benefits for others.

Investigating Cost Implications of Incorporating Level III At-Home Testing into a Polysomnography Based Sleep Medicine Program Using Administrative Data.

OBJECTIVE: Obstructive sleep apnea is a common problem, requiring expensive in-lab polysomnography for proper diagnosis. Home monitoring can provide an alternative to in-lab testing for a subset of OSA patients. The objective of this project was to investigate the effect of incorporating home testing into an OSA program at a large, tertiary sleep disorders centre. METHODS: The Sleep Disorders Centre in Saskatoon, Canada, has been incorporating at-home testing into their diagnostic pathways since 2006. Administrative data from 2007 to 2013 were extracted (10030 patients) and the flow of patients through the program was followed from diagnosis to treatment. Costs were estimated using 2014 pricing and were stratified by disease attributes and sensitivity analysis was applied. RESULTS: The overall costs per patient were $627.40, with $419.20 for at-home testing and $746.20 for in-lab testing. The cost of home management would rise to $515 if all negative tests were required to be confirmed by an in-lab PSG. DISCUSSION: Our review suggests that at-home testing can be cost-effective alternative to in-lab testing when applied to the correct population, specifically, those with a high pretest probability of obstructive sleep apnea and an absence of significant comorbidities.

Leveraging medical taxonomies to improve knowledge management within online communities of practice: The knowledge maps system.

BACKGROUND AND OBJECTIVE: Online communities of practice contain a wealth of information, stored in the free text of shared communications between community members. The Knowledge Maps (KMaps) system is designed to facilitate Knowledge Translation in online communities through multi-level analyses of the shared messages of these communications. METHODS: Using state-of-the-art semantic mapping technologies (Metamap) the contents of the messages shared within an online community are mapped to terms from the MeSH medical lexicon, providing a multi-level topic-specific summary of the knowledge being shared within the community. Using the inherent hierarchical structure of the lexicon important insights can be found within the community. RESULTS: The KMaps system was applied to two medical mailing lists, the PPML (archives from 2009-02 to 2013-02) and SURGINET (archives from 2012-01 to 2013-04), identifying 27,924 and 50,597 medical terms respectively. KMaps identified content areas where both communities found interest, specifically around Diseases, 22% and 24% of the total terms, while also identifying field-specific areas that were more popular: SURGINET expressed an interest in Anatomy (14% vs 4%) while the PPML was more interested in Drugs (19% vs 9%). At the level of the individual KMaps identified 6 PPML users and 9 SURGINET users that had noticeably more contributions to the community than their peers, and investigated their personal areas of interest. CONCLUSION: The KMaps system provides valuable insights into the structure of both communities, identifying topics of interest/shared content areas and defining content-profiles for individual community members. The system provides a valuable addition to the online KT process.

The role of radiology in the quantification of digital ulnar deviation in rheumatoid arthritis patients.

BACKGROUND: Rheumatoid arthritis (RA) is a common inflammatory polyarthritis, which causes functional digital ulnar deviation (UD). Radiographic and magnetic resonance imaging (MRI) assessment of the hands is essential in RA, but its role in the quantification of UD remains unclear. PURPOSE: To compare UD measurements in RA patients between clinical goniometric assessments versus standardized radiographs and MRI. METHODS: Fifteen RA patients with clinically apparent UD and 11 RA patients without UD underwent a rheumatological examination prior to recruitment to this study. Goniometric measurements for UD at the metacarpophalangeal (MCP) joints were performed by an occupational therapist (OT). Standardized hand radiographs, and MRI studies of the dominant hand using 3T MRI scanner with 16 channel hand/wrist coil were evaluated. Angulation measurements for radiographs and MRI were performed independently by two experienced musculoskeletal radiologists who were blinded to the rheumatologist's, occupational therapist's and each other's assessments. RESULTS: Inter-observer correlation between radiologists was >0.97 for both radiographic and MRI measurements. Correlation between OT goniometric measurements and the imaging-based measurements was limited at 0.496 for radiographs and 0.317 for MRI. Correlation between imaging modalities was 0.513. Compared to OT measurements, radiographic and MRI study measurements significantly underestimate the angulation in RA patients with UD (P < 0.001). CONCLUSIONS: The results of this study demonstrated discordance between radiological and goniometric measurements of digital ulnar angulation at the MCP joints in RA patients. Although imaging plays a key role in understanding structural damage and disease activity in RA, it should be emphasized that radiological measurements underrate joint malalignment.

Evaluating a Telehealth Follow-up Program for Cardiology Patients Using Administrative Data.

This project evaluated the effect of a telehealth follow-up intervention on readmission rates for patients discharged from the Saskatoon Health Region cardiology units with ACS or HF. 1-year readmission rates for ACS patients were 27.4% (95% CI: [26%, 29%]) before telehealth implementation and 25.2% (95% CI: [24%, 27%]) after, producing an insignificant hazard ratio of 1.07 (95% CI: [0.97, 1.18], P=0.1899). For HF 50.3% (95% CI: [44%, 56%]) of pre-implementation and 47.9% (95% CI: [43%, 53%]) of post-implementation patients had a 1-year readmission, producing a HR of 1.04 (95% CI: [0.83, 1.26], P=0.6882). This analysis found no significant effect of telehealth on readmission rates for ACS or HF patients.

Improvement in obstructive sleep apnea diagnosis and management wait times: A retrospective analysis of home management pathway for obstructive sleep apnea.

Obstructive sleep apnea is a common condition within the Canadian population. The current gold standard for diagnosis and management of patients is in-laboratory (in-lab) polysomnography; however, the limited availability of testing options for patients has led to long wait times and increased disease burden within the population. The Sleep Research Laboratory in Saskatoon (Saskatchewan) implemented a home management program to run in parallel with the in-lab system several years ago in an effort to increase their capacity and reduce wait times. The present study was a retrospective analysis of all patients referred to the program between 2009 and 2012. The home management system has improved wait times by diagnosing and managing up to one-half of the referred patient population, reducing the wait for in-lab treatment from a median of 152 days in 2009 to 92 days in 2012 (P<0.0001). Moving forward, home management can provide a viable alternative to in-lab testing for patients who meet strict entry criteria, reducing the in-lab workload and, ultimately, reducing wait times.

Clinical guideline-driven personalized self-management diary for paediatric cancer survivors.

Shared Decision Making (SDM) is the process of patients and clinicians working together to manage medical treatment using the existing knowledge base. This paper presents the YouCan framework, a system for summarizing and presenting the necessary knowledge for SDM related to pediatric cancer follow-up management. Knowledge modelling of a Clinical Practice Guideline produced a customized ontology, which was then passed through a pellet reasoned to produce a customized patient diary that summarizes a patient's oncological history as well as the potential issues they may face in follow-up.

Validity of cardiovascular risk prediction models in kidney transplant recipients.

BACKGROUND: Predicting cardiovascular risk is of great interest in renal transplant recipients since cardiovascular disease is the leading cause of mortality. OBJECTIVE: To conduct a systematic review to assess the validity of cardiovascular risk prediction models in this population. METHODS: Five databases were searched (MEDLINE, EMBASE, SCOPUS, CINAHL, and Web of Science) and cohort studies with at least one year of follow-up were included. Variables that described population characteristics, study design, and prognostic performance were extracted. The Quality in Prognostic Studies (QUIPS) tool was used to evaluate bias. RESULTS: Seven studies met the criteria for inclusion, of which, five investigated the Framingham risk score and three used a transplant-specific model. Sample sizes ranged from 344 to 23,575, and three studies lacked sufficient event rates to confidently reach conclusion. Four studies reported discrimination (as measured by c-statistic), which ranged from 0.701 to 0.75, while only one risk model was both internally and externally validated. CONCLUSION: The Framingham has underestimated cardiovascular events in renal transplant recipients, but these studies have not been robust. A risk prediction model has been externally validated at least on one occasion, but comprehensive validation in multiple cohorts and impact analysis are recommended before widespread clinical application is advocated.

Applying social network analysis to understand the knowledge sharing behaviour of practitioners in a clinical online discussion forum.

BACKGROUND: Knowledge Translation (KT) plays a vital role in the modern health care community, facilitating the incorporation of new evidence into practice. Web 2.0 tools provide a useful mechanism for establishing an online KT environment in which health practitioners share their practice-related knowledge and experiences with an online community of practice. We have implemented a Web 2.0 based KT environment--an online discussion forum--for pediatric pain practitioners across seven different hospitals in Thailand. The online discussion forum enabled the pediatric pain practitioners to share and translate their experiential knowledge to help improve the management of pediatric pain in hospitals. OBJECTIVE: The goal of this research is to investigate the knowledge sharing dynamics of a community of practice through an online discussion forum. We evaluated the communication patterns of the community members using statistical and social network analysis methods in order to better understand how the online community engages to share experiential knowledge. METHODS: Statistical analyses and visualizations provide a broad overview of the communication patterns within the discussion forum. Social network analysis provides the tools to delve deeper into the social network, identifying the most active members of the community, reporting the overall health of the social network, isolating the potential core members of the social network, and exploring the inter-group relationships that exist across institutions and professions. RESULTS: The statistical analyses revealed a network dominated by a single institution and a single profession, and found a varied relationship between reading and posting content to the discussion forum. The social network analysis discovered a healthy network with strong communication patterns, while identifying which users are at the center of the community in terms of facilitating communication. The group-level analysis suggests that there is strong interprofessional and interregional communication, but a dearth of non-nurse participants has been identified as a shortcoming. CONCLUSIONS: The results of the analysis suggest that the discussion forum is active and healthy, and that, though few, the interprofessional and interinstitutional ties are strong.

An infobutton for Web 2.0 clinical discussions: the knowledge linkage framework.

This paper aims to develop an infobutton to automatically retrieve published papers corresponding to a topic-specific online clinical discussion. The knowledge linkages infobutton is designed to supplement online clinical conversations with pertinent medical literature from Pubmed. The project involves three distinct steps: 1) Clinical messages around a specific problem are grouped together into a thread. 2) These threads are processed using Metamap to link the conversations to keywords from the MeSH lexicon. 3) These keywords are used in a novel search strategy to retrieve a set of papers from Pubmed, which are then returned to the user. A pilot study using the messages from 2007 and 2008, was conducted to compare the knowledge linkage search strategy to a vector space model and extended Boolean model. The knowledge linkage model proved to be significantly better in terms of precision ( p = 0.013 and 0.003, respectively) and recall ( p = 0.351 and 0.013). Pertinent papers were returned to over 55% of the threads. This approach has demonstrated how clinicians can supplement their peer communications with evidence based research. Future work should focus on how to improve the threading and keyword-mapping strategies.