Free-text Data Research Articles

Bringing organisations together during a pandemic: the case of an intersectoral community support group.

In early 2020, the COVID-19 pandemic emerged, posing multiple challenges to healthcare organisations and communities. The Darling Downs region in Queensland, Australia had its first positive case of COVID-19 confirmed in March 2020, which created understandable anxiety in the community. The Vulnerable Communities Group (VCG) was established to address this anxiety through open lines of communication to strengthen community resilience. This case study reports the evaluation of the VCG, plus lessons learned while establishing and running an intersectoral group, with stakeholders from more than 40 organisations, in response to the COVID-19 pandemic. An anonymous online survey with closed and open-ended questions was administered to participants. Data were subject to descriptive statistical tests and content analysis. Four categories were developed from the free text data for reporting: 'Knowledge is power', 'Beating isolation through partnerships and linkages', 'Sharing is caring', and 'Ripple effects'. Whilst open communication and collaboration are always essential, they can be critically important during times of crisis. The VCG initiative is one example of tackling the challenges of the COVID-19 pandemic through bringing the community together. What is known about the topic? The COVID-19 pandemic has caused immense strain on the global community. It has thrust healthcare professionals and community leaders into an unforeseen situation, with little information being available on how to effectively deal with the pandemic to negate its effects. What does this paper add? This paper reports on the establishment and evaluation of the Vulnerable Communities Group (VCG) in the Toowoomba, Darling Downs region of Queensland, Australia in response to the COVID-19 pandemic, using a community of practice framework. It provides insight into how to establish and retain an intersectoral community of practice group during rapidly changing and challenging times, such as during the COVID-19 pandemic. What are the implications for practitioners? Practitioners can use a community of practice framework to establish and evaluate an intersectoral group, as described in our paper, to enhance community connectedness to reduce isolation and share information and resources to help negate the challenges caused by the COVID-19 pandemic.

Exploring the use of musculoskeletal ultrasound imaging by podiatrists: an international survey

BackgroundPodiatrists, in musculoskeletal services, are demonstrating an expansion of their practice skills through the use of ultrasound imaging. There is an assumption that this practice is beneficial within the context of patient care and health systems. The aim of this research was to further investigate the use of musculoskeletal ultrasound (MSUS) by podiatrists within their clinical setting and gain additional insights into the impact that they perceive use of MSUS has on their approaches to management of musculoskeletal foot and ankle problems.MethodAn international study utilising a cross-sectional design and an internet-based platform was undertaken. The survey was developed and implemented through three phases: 1. survey development, 2. face validity agreement via questionnaire review, and 3. survey distribution and data collection. Twenty-two survey questions were developed and set as a two-step approach collecting quantitative data (part 1) and qualitative free text data (part 2). Data was exported from SurveyMonkey and analysed using Microsoft Excel software. Counts and frequencies were calculated for responses to all twenty closed questions. Responses to the two final open-ended questions were analysed using thematic analysis to search for patterns related to podiatrists’ perceptions of impact.ResultsTwo hundred and thirty-two eligible participants consented to complete the survey. The majority (n = 159) of respondents were from the UK and Spain. Commonly MSUS has been used in practice for (i) diagnosing pathology, (ii) supporting rehabilitation, (iii) supporting interventions or (iv) research purposes. Most frequently, MSUS was used to assist in the diagnosis of injury/pathology (84%). A range of free text comments were received from the participants in response to the question relating to their thoughts on the impact of using MSUS imaging in their practice (n = 109) and on their perceptions of how the use of MSUS has influenced their approaches to management of their patients’ musculoskeletal foot and ankle problems (n = 108). Thematic analysis of the free text comments generated four themes: (i) diagnosis, (ii) delivery and access of care, (iii) patient education and engagement, and (iv) patient empowerment.ConclusionThe perceived benefit podiatrists indicated in using MSUS as part of their practice is the perceived improvement in patient journeys through tighter, focused management plans and reduced waiting times. An additional novel finding was that MSUS provided the capacity for podiatrists to better inform patients of their diagnosis, which they believed led to improved engagement and consequent empowerment of patients in their treatment plans. We propose further investigation of patient experiences as well as testing of the model that embeds podiatrists’ use of MSUS as a key skill in musculoskeletal foot and ankle services.

WAY TO SMART LIFE USING DATA MINING

Information offers numerous prerequisites to the end clients, for example, programming, association and stage go on. In this proposed framework, we concentrate about carefully mining the information on social media. Online media turns out to be well known from the information given by biomedical and other providers. This data is usually shared so that medical service improves, costs declines and utilises the assessment which is created by client. We suggest investigation framework that give attentions on side effects of drugs and also focus on positive and negative response. To improve health care some Clinical documents are mostly useful because these are free text data sources. Clinical reports containing data identified with manifestations and significant meds. Extracting data from enormous dataset became famous in light of the fact that clients get different thoughts from this shifted information. Data Mining and Knowledge mining became well known on the grounds that clients are aware of information and getting data of various region like wellbeing, Social, etc. After information preparing we center on clients positive and negative assessments. We exclude these feelings and discover which prescription is great; to choose this we additionally discover the symptoms of the drugs. Further we center on the manifestations of the sickness of tolerant. By taking the master specialists proposal, we rattle off the medicine of any sickness as per the side effects and we give this medicine or treatment to the client on our gathering. We can grow our examination into Data and Knowledge mining of online media and takes the client’s reviews on different medications of Illness. This day by day refreshed information serves to drug industry, specialists, emergency clinics, and clinical staff, for viable future medicines. KEYWORDS: Information mining, Complex organizations, social figuring, Data mining, semantic Web.

Evaluation of lexical clarification by patients reading their clinical notes: a quasi-experimental interview study

BackgroundPatients benefit from access to their medical records. However, clinical notes and letters are often difficult to comprehend for most lay people. Therefore, functionality was implemented in the patient portal of a Dutch university medical centre (UMC) to clarify medical terms in free-text data. The clarifications consisted of synonyms and definitions from a Dutch medical terminology system. We aimed to evaluate to what extent these lexical clarifications match the information needs of the patients. Secondarily, we evaluated how the clarifications and the functionality could be improved.MethodsWe invited participants from the patient panel of the UMC to read their own clinical notes. They marked terms they found difficult and rated the ease of these terms. After the functionality was activated, participants rated the clarifications provided by the functionality, and the functionality itself regarding ease and usefulness. Ratings were on a scale from 0 (very difficult) to 100 (very easy). We calculated the median number of terms not understood per participant, the number of terms with a clarification, the overlap between these numbers (coverage), and the precision and recall.ResultsWe included 15 participants from the patient panel. They marked a median of 21 (IQR 19.5–31) terms as difficult in their text files, while only a median of 2 (IQR 1–4) of these terms were clarified by the functionality. The median precision was 6.5% (IQR 2.3–14.25%) and the median recall 8.3% (IQR 4.7–13.5%) per participant. However, participants rated the functionality with median ease of 98 (IQR 93.5–99) and a median usefulness of 79 (IQR 52.5–97). Participants found that many easy terms were unnecessarily clarified, that some clarifications were difficult, and that some clarifications contained mistakes.ConclusionsPatients found the functionality easy to use and useful. However, in its current form it only helped patients to understand few terms they did not understand, patients found some clarifications to be difficult, and some to be incorrect. This shows that lexical clarification is feasible even when limited terms are available, but needs further development to fully use its potential.

Natural language processing with deep learning for medical adverse event detection from free-text medical narratives: A case study of detecting total hip replacement dislocation

BackgroundAccurate and timely detection of medical adverse events (AEs) from free-text medical narratives can be challenging. Natural language processing (NLP) with deep learning has already shown great potential for analyzing free-text data, but its application for medical AE detection has been limited. MethodIn this study, we developed deep learning based NLP (DL-NLP) models for efficient and accurate hip dislocation AE detection following primary total hip replacement from standard (radiology notes) and non-standard (follow-up telephone notes) free-text medical narratives. We benchmarked these proposed models with traditional machine learning based NLP (ML-NLP) models, and also assessed the accuracy of International Classification of Diseases (ICD) and Current Procedural Terminology (CPT) codes in capturing these hip dislocation AEs in a multi-center orthopaedic registry. ResultsAll DL-NLP models outperformed all of the ML-NLP models, with a convolutional neural network (CNN) model achieving the best overall performance (Kappa = 0.97 for radiology notes, and Kappa = 1.00 for follow-up telephone notes). On the other hand, the ICD/CPT codes of the patients who sustained a hip dislocation AE were only 75.24% accurate. ConclusionsWe demonstrated that a DL-NLP model can be used in largescale orthopaedic registries for accurate and efficient detection of hip dislocation AEs. The NLP model in this study was developed with data from the most frequently used electronic medical record (EMR) system in the U.S., Epic. This NLP model could potentially be implemented in other Epic-based EMR systems to improve AE detection, and consequently, quality of care and patient outcomes.

Computer-assisted textual analysis of free-text comments in the Swiss Cancer Patient Experiences (SCAPE) survey

BackgroundPatient experience surveys are increasingly conducted in cancer care as they provide important results to consider in future development of cancer care and health policymaking. These surveys usually include closed-ended questions (patient-reported experience measures (PREMs)) and space for free-text comments, but published results are mostly based on PREMs. We aimed to identify the underlying themes of patients’ experiences as shared in their own words in the Swiss Cancer Patient Experiences (SCAPE) survey and compare these themes with those assessed with PREMs to investigate how the textual analysis of free-text comments contributes to the understanding of patients’ experiences of care.MethodsSCAPE is a multicenter cross-sectional survey that was conducted between October 2018 and March 2019 in French-speaking parts of Switzerland. Patients were invited to rate their care in 65 closed-ended questions (PREMs) and to add free-text comments regarding their cancer-related experiences at the end of the survey. We conducted computer-assisted textual analysis using the IRaMuTeQ software on the comments provided by 31% (n = 844) of SCAPE survey respondents (n = 2755).ResultsWe identified five main thematic classes, two of which consisting of a detailed description of ‘cancer care pathways’. The remaining three classes were related to ‘medical care’, ‘gratitude and praise’, and the way patients lived with cancer (‘cancer and me’). Further analysis of this last class showed that patients’ comments related to the following themes: ‘initial shock’, ‘loneliness’, ‘understanding and acceptance’, ‘cancer repercussions’, and ‘information and communication’. While closed-ended questions related mainly to factual aspects of experiences of care, free-text comments related primarily to the personal and emotional experiences and consequences of having cancer and receiving care.ConclusionsA computer-assisted textual analysis of free-text in our patient survey allowed a time-efficient classification of free-text data that provided insights on the personal experience of living with cancer and additional information on patient experiences that had not been collected with the closed-ended questions, underlining the importance of offering space for comments. Such results can be useful to inform questionnaire development, provide feedback to professional teams, and guide patient-centered initiatives to improve the quality and safety of cancer care.

Unsupervised identification of crime problems from police free-text data

We present a novel exploratory application of unsupervised machine-learning methods to identify clusters of specific crime problems from unstructured modus operandi free-text data within a single administrative crime classification. To illustrate our proposed approach, we analyse police recorded free-text narrative descriptions of residential burglaries occurring over a two-year period in a major metropolitan area of the UK. Results of our analyses demonstrate that topic modelling algorithms are capable of clustering substantively different burglary problems without prior knowledge of such groupings. Subsequently, we describe a prototype dashboard that allows replication of our analytical workflow and could be applied to support operational decision making in the identification of specific crime problems. This approach to grouping distinct types of offences within existing offence categories, we argue, has the potential to support crime analysts in proactively analysing large volumes of modus operandi free-text data—with the ultimate aims of developing a greater understanding of crime problems and supporting the design of tailored crime reduction interventions.

Vancomycin Hypersensitivity Reactions Documented in Electronic Health Records

Vancomycin, the most common antimicrobial used in US hospitals, can cause diverse adverse reactions, including hypersensitivity reactions (HSRs). Yet, little is known about vancomycin reactions documented in electronic health records. To describe vancomycin HSR epidemiology from electronic health record allergy data. This was a cross-sectional study of patients with 1 or more encounter from 2017 to 2019 and an electronic health record vancomycin drug allergy label (DAL) in 2 US health care systems. We determined prevalence and trends of vancomycin DALs and assessed active DALs by HSR phenotype determined from structured (coded) and unstructured (free-text) data using natural language processing. We investigated demographic associations with documentation of vancomycin red man syndrome (RMS). Among 4,490,618 patients, 14,426 (0.3%) had a vancomycin DAL with 18,761 documented reactions (2,248 [12.0%] free-text). Quarterly mean vancomycin DALs added were 253 ± 12 and deleted were 12 ± 2. Of 18,761 vancomycin HSRs, 7,903 (42.1%) were immediate phenotypes and 3,881 (20.7%) were delayed phenotypes. Common HSRs were rash (32% of HSRs) and RMS (16% of HSRs). Anaphylaxis was coded in 6% cases of HSRs. Drug reaction eosinophilia and systemic symptoms syndrome was the most common coded vancomycin severe cutaneous adverse reaction. RMS documentation was more likely for males (odds ratio, 1.30; 95% CI, 1.17-1.44) and less likely for blacks (odds ratio, 0.59; 95% CI, 0.47-0.75). Vancomycin causes diverse adverse reactions, including common (eg, RMS) and severe (eg, drug reaction eosinophilia and systemic symptoms syndrome) reactions entered as DAL free-text. Anaphylaxis comprised 6% of documented vancomycin HSRs, although true vancomycin IgE-mediated reactions are exceedingly rare. Improving vancomycin DAL documentation requires more coded entry options, including a coded entry for RMS.

Thoracic thrust joint manipulation: An international survey of current practice and knowledge in IFOMPT member countries

BackgroundThe perceived relative safety of thoracic thrust joint manipulation (TTJM) has contributed to a growth in evidence supporting use in practice. Yet adverse events (AE) have been documented following TTJM. Knowledge of current practice is therefore required to support further research. PurposeTo investigate TTJM knowledge and pre-TTJM examination across IFOMPT Member Organisations (MO) and Registered Interest Groups (RIG). MethodsAn e-survey was designed based on existing evidence and piloted. Eligibility criteria: physiotherapists from member countries of the International Federation of Orthopaedic Manipulative Physical Therapists (IFOMPT) who use TTJM. Recruitment was through IFOMPT networks (May 2018–March 2019). Data analyses included descriptive analyses and content analysis for free text data. ResultsRespondents (n = 363) from 20 countries. Pre-TTJM examination included patient history (22%, n = 81) and physical examination (69%, n = 248). Across presentations (>80% threshold of agreement) contraindications included osteomyelitis, fracture and metastatic disease. Spinal deformity, respiratory disease, serious joint disease and hypermobility achieved >60% agreement as precautions. Consent was obtained by 93% respondents (n = 250). Preferred technique was PA/AP thrust (61%, n = 144). Perception of primary effect was neurophysiological (52%, n = 134), biomechanical (42%, n = 109) and placebo (3%, n = 8). From those who reported AE (n = 100), these included fractures (36%, n = 42) and cord signs/symptoms (6%, n = 7). ConclusionPre-TTJM examination is common, although bias towards physical examination. Differential testing for upper versus lower thoracic spine is limited. Inconsistencies across knowledge of contraindications and precautions, and beliefs for biomechanical effect were found. Findings highlight the importance of high levels of clinical reasoning during patient history for TTJM.

The transition experiences of adolescents with type 1 diabetes from paediatric to adult care providers.

Emerging adults with type 1 diabetes (T1D) are at increased risk for poor health outcomes as they transition from paediatric to adult healthcare providers. This is in part due to the complexities of young adult life as individuals with T1D enter the workforce, leave home or start college while learning to manage the disease on their own. We sought to identify the barriers and facilitators adolescents face during their emerging adult years with T1D. Young adults, aged 24-35, who lived with T1D during their adolescent years were recruited online to complete a survey regarding their experience with care transition. Categorical data were analysed using descriptive statistics. A thematic analysis, guided by the Framework for Emerging Adults with T1D, was used to explore the free-text data. In total, 25 adults (84% female) with mean age of 28 ± 3.2 years participated. Themes that arose from the analysis of the paediatric to adult care transition experiences included (1) importance of support from key players, (2) challenges navigating the healthcare system, (3) mental health needs of emerging adults with T1D, (4) managing day-to-day life with T1D and (5) early independence to ease transition. Individuals with T1D face a variety of challenges as they transition from paediatric to adult care providers. A proactive approach in educating adolescents is needed.

Potential value of electronic health records in capturing data on geriatric frailty for population health

ObjectivesDespite the availability of many frailty measures to identify older adults at risk, frailty instruments are not routinely used for risk assessment in population health management. Here, we assessed the potential value of electronic health records (EHRs) and administrative claims in providing the necessary data for variables used across various frailty instruments. Setting and participantsThe review focused on studies conducted worldwide. Participants included older people aged 50 and older. DesignWe identified frailty instruments published between 2011 and 2018. Frailty variables used in each of the frailty instruments were extracted, grouped, and categorized across health determinants and various clinical factors. MeasuresThe availability of the extracted frailty variables across various data sources (e.g., EHRs, administrative claims, and surveys) was evaluated by experts. ResultsWe identified 135 frailty instruments, which contained 593 unique variables. Clinical determinants of health were the best represented variables across frailty instruments (n = 516; 87 %), unlike social and health services factors (n = 33; ∼5% and n = 32; ∼5%). Most frailty instruments require at least one variable that is not routinely available in EHRs or claims (n = 113; ∼83 %). Only 22 frailty instruments have the potential to completely rely on EHR (structured or free-text data) and/or claims data, and possibly be operationalized on a population-level. Conclusions and implicationsFrailty instruments continue to be highly survey-based. More research is therefore needed to develop EHR-based frailty instruments for population health management. This will permit organizations and societies to stratify risk and better allocate resources among different older adult populations.

Statistical inference for natural language processing algorithms with a demonstration using type 2 diabetes prediction from electronic health record notes.

The pointwise mutual information statistic (PMI), which measures how often two words occur together in a document corpus, is a cornerstone of recently proposed popular natural language processing algorithms such as word2vec. PMI and word2vec reveal semantic relationships between words and can be helpful in a range of applications such as document indexing, topic analysis, or document categorization. We use probability theory to demonstrate the relationship between PMI and word2vec. We use the theoretical results to demonstrate how the PMI can be modeled and estimated in a simple and straight forward manner. We further describe how one can obtain standard error estimates that account for within-patient clustering that arises from patterns of repeated words within a patient's health record due to a unique health history. We then demonstrate the usefulness of PMI on the problem of predictive identification of disease from free text notes of electronic health records. Specifically, we use our methods to distinguish those with and without type 2 diabetes mellitus in electronic health record free text data using over 400 000 clinical notes from an academic medicalcenter.

Free-text patient experience data inform service development.

Understanding patients' experiences of treatment and care is essential to effective service development, and is therefore a common topic of investigation for clinicians and health-service researchers. For service development, different kinds of investigation may be useful for a different purpose. Closed measures of patient satisfaction (e.g. rating scales) are useful in providing reproducible measures in specified domains; however, they are often limited both by the scope of predefined questions and by the level of detail that they provide on processes affecting reported outcomes. Free-text data are therefore often collected alongside such measures, to provide further descriptive detail on patient experience and identify emergent issues that are not covered by closed questions (Bracher et al. BMJ Open 2016;6:e011830). Izzet-Kay et al. present a rigorous and systematic exploration of women's experiences and concerns following mesh-augmented prolapse surgery (BJOG 2021; 128:131–139). The study also provides an exemplar of how free-text analysis can inform service development in more general ways: first, by exploring how patients evaluate treatment success in the context of everyday life; second, by investigating how patients understand costs/risks and benefits of treatment; and third, by identifying issues for further investigation. Findings from the study give us descriptions of experiences as both outcomes (i.e. what happened) and processes (i.e. what factors affected the outcome). This distinction is important, because different types of knowledge can serve distinct but complementary functions in service development: the former identifying areas of concern (e.g. postoperative uncertainty, decision regret), the latter describing mechanisms contributing to the outcome and hence potential points of intervention (e.g. improved information provision, communication and shared decision-making processes) (Elwyn et al. BMJ 2010;341:c5146). Navigating complex healthcare systems (such as treatment pathways involving multidisciplinary teams) involves patients, clinicians and (often) carers in varied kinds of physical, emotional and cognitive work that includes making sense of information, identifying preferences and appraising risk (May et al. BMC Health Serv Res 2014;14(281)). Free-text data can provide important insights into both processes and outcomes that may not be visible through other means such as clinical audit. Findings can be used to complement those from closed measures (as Izzett-Kay et al. have done), inform development of new questions for wider systematic exploration (e.g. through a population-level survey) or identify areas for more detailed qualitative investigation (e.g. through in-depth interviews or longitudinal diaries exploring experiences over time). More detailed qualitative studies may, for example, focus on key points of communication or information provision within treatment pathways, to better understand how presentation and discussion of information affect risk perception and decision-making (Mills et al. J Clin Epidemiol 2011;64:1127–36). Readers of this journal undertaking service evaluation and development activity may wish to consider using free-text data as one additional tool to inform this work, a helpful demonstration of which has been provided by Izzett-Kay et al. None declared. A completed disclosure of interests form is available to view online as supporting information. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.

Highly accurate classification of chest radiographic reports using a deep learning natural language model pre-trained on 3.8 million text reports.

The development of deep, bidirectional transformers such as Bidirectional Encoder Representations from Transformers (BERT) led to an outperformance of several Natural Language Processing (NLP) benchmarks. Especially in radiology, large amounts of free-text data are generated in daily clinical workflow. These report texts could be of particular use for the generation of labels in machine learning, especially for image classification. However, as report texts are mostly unstructured, advanced NLP methods are needed to enable accurate text classification. While neural networks can be used for this purpose, they must first be trained on large amounts of manually labelled data to achieve good results. In contrast, BERT models can be pre-trained on unlabelled data and then only require fine tuning on a small amount of manually labelled data to achieve even better results. Using BERT to identify the most important findings in intensive care chest radiograph reports, we achieve areas under the receiver operation characteristics curve of 0.98 for congestion, 0.97 for effusion, 0.97 for consolidation and 0.99 for pneumothorax, surpassing the accuracy of previous approaches with comparatively little annotation effort. Our approach could therefore help to improve information extraction from free-text medical reports. Availability and implementationWe make the source code for fine-tuning the BERT-models freely available at https://github.com/fast-raidiology/bert-for-radiology. Supplementary data are available at Bioinformatics online.

Evaluating Canadian Radiation Therapists and UK Therapeutic Radiographers’ Experiences and Opinions of a Safety Strap to Secure Patients during Radiotherapy

IntroductionA serious patient safety incident at a cancer centre in Ontario, Canada, saw a patient fall from an elevated treatment couch. A regional investigation recommended the use of a securing safety strap. The authors evaluate the value of the strap through the experiences of the radiation therapists’ who use it. A secondary aim is to explore the potential for using a securing safety strap with UK therapeutic radiographers. MethodsA two-stage design was guided by an evidence-based practice framework. Stage one used a questionnaire to capture treating radiation therapists’ experiences and opinions of the strap at a single cancer centre. Quantitative data were analysed descriptively and free-text data via a content analysis. Stage two used semistructured interviews with thematic analysis to explore views of three UK therapeutic radiographers. ResultsTwenty-five of approximately 130 eligible staff responded to the Canadian questionnaire. Of the respondents, 24% (n = 6) ‘strongly disagreed’, 28% (n = 7) ‘agreed’ and 48% (n = 12) ‘neither agreed nor disagreed’ that they would recommend the strap to other departments. Most of the respondents think strap use should be at the staffs' discretion, with patients with dementia/cognitive impairment ranked as the group benefiting most. Ninety-two percent (n = 23) of respondents confirmed that patients sometimes refuse the strap. Themes arising from stage two interviews are as follows: patient benefit (use for select patients only); patient safety versus control (restraint); and practical implementation issues. ConclusionThe policy of universal use of the strap should be reviewed. Those who use it are equivocal about its value and feel it should be reserved for select patients at the treating professional's discretion. Full evaluation of the effectiveness and acceptability of the device for different patients may promote both staff enthusiasm towards the device and evidence-based practice. Adequate resources are required to evaluate implementation of such safety initiatives.

Extraction of temporal relations from clinical free text: A systematic review of current approaches.

Temporal relations between clinical events play an important role in clinical assessment and decision making. Extracting such relations from free text data is a challenging task because it lies on between medical natural language processing, temporal representation and temporal reasoning. To survey existing methods for extracting temporal relations (TLINKs) between events from clinical free text in English; to establish the state-of-the-art in this field; and to identify outstanding methodological challenges. A systematic search in PubMed and the DBLP computer science bibliography was conducted for studies published between January 2006 and December 2018. The relevant studies were identified by examining the titles and abstracts. Then, the full text of selected studies was analyzed in depth and information were collected on TLINK tasks, TLINK types, data sources, features selection, methods used, and reported performance. A total of 2834 publications were identified for title and abstract screening. Of these publications, 51 studies were selected. Thirty-two studies used machine learning approaches, 15 studies used a hybrid approaches, and only four studies used a rule-based approach. The majority of studies use publicly available corpora: THYME (28 studies) and the i2b2 corpus (17 studies). The performance of TLINK extraction methods ranges widely depending on relation types and events (e.g. from 32% to 87% F-score for identifying relations between clinical events and document creation time). A small set of TLINKs (before, after, overlap and contains) has been widely studied with relatively good performance, whereas other types of TLINK (e.g., started by, finished by, precedes) are rarely studied and remain challenging. Machine learning classifiers (such as Support Vector Machine and Conditional Random Fields) and Deep Neural Networks were among the best performing methods for extracting TLINKs, but nearly all the work has been carried out and tested on two publicly available corpora only. The field would benefit from the availability of more publicly available, high-quality, annotated clinical text corpora.

Use of technology by orientation and mobility professionals in Australia and Malaysia before COVID-19

Purpose Orientation and Mobility (O&M) professionals teach people with low vision or blindness to use specialist assistive technologies to support confident travel, but many O&M clients now prefer a smartphone. This study aimed to investigate what technology O&M professionals in Australia and Malaysia have, use, like, and want to support their client work, to inform the development of O&M technologies and build capacity in the international O&M profession. Materials and Methods A technology survey was completed by professionals (n = 36) attending O&M workshops in Malaysia. A revised survey was completed online by O&M specialists (n = 31) primarily in Australia. Qualitative data about technology use came from conferences, workshops and interviews with O&M professionals. Descriptive statistics were analysed together with free-text data. Results Limited awareness of apps used by clients, unaffordability of devices, and inadequate technology training discouraged many O&M professionals from employing existing technologies in client programmes or for broader professional purposes. Professionals needed to learn smartphone accessibility features and travel-related apps, and ways to use technology during O&M client programmes, initial professional training, ongoing professional development and research. Conclusions Smartphones are now integral to travel with low vision or blindness and early-adopter O&M clients are the travel tech-experts. O&M professionals need better initial training and then regular upskilling in mainstream O&M technologies to expand clients’ travel choices. COVID-19 has created an imperative for technology laggards to upskill for O&M tele-practice. O&M technology could support comprehensive O&M specialist training and practice in Malaysia, to better serve O&M clients with complex needs. Implications for rehabilitation Most orientation and mobility (O&M) clients are travelling with a smartphone, so O&M specialists need to be abreast of mainstream technologies, accessibility features and apps used by clients for orientation, mobility, visual efficiency and social engagement. O&M specialists who are technology laggards need human-guided support to develop confidence in using travel technologies, and O&M clients are the experts. COVID-19 has created an imperative to learn skills for O&M tele-practice. Affordability is a significant barrier to O&M professionals and clients accessing specialist travel technologies in Malaysia, and to O&M professionals upgrading technology in Australia. Comprehensive training for O&M specialists is needed in Malaysia to meet the travel needs of clients with low vision or blindness who also have physical, cognitive, sensory or mental health complications.

Toward the Development of Data Governance Standards for Using Clinical Free-Text Data in Health Research: Position Paper.

BackgroundClinical free-text data (eg, outpatient letters or nursing notes) represent a vast, untapped source of rich information that, if more accessible for research, would clarify and supplement information coded in structured data fields. Data usually need to be deidentified or anonymized before they can be reused for research, but there is a lack of established guidelines to govern effective deidentification and use of free-text information and avoid damaging data utility as a by-product.ObjectiveThis study aimed to develop recommendations for the creation of data governance standards to integrate with existing frameworks for personal data use, to enable free-text data to be used safely for research for patient and public benefit.MethodsWe outlined data protection legislation and regulations relating to the United Kingdom for context and conducted a rapid literature review and UK-based case studies to explore data governance models used in working with free-text data. We also engaged with stakeholders, including text-mining researchers and the general public, to explore perceived barriers and solutions in working with clinical free-text.ResultsWe proposed a set of recommendations, including the need for authoritative guidance on data governance for the reuse of free-text data, to ensure public transparency in data flows and uses, to treat deidentified free-text data as potentially identifiable with use limited to accredited data safe havens, and to commit to a culture of continuous improvement to understand the relationships between the efficacy of deidentification and reidentification risks, so this can be communicated to all stakeholders.ConclusionsBy drawing together the findings of a combination of activities, we present a position paper to contribute to the development of data governance standards for the reuse of clinical free-text data for secondary purposes. While working in accordance with existing data governance frameworks, there is a need for further work to take forward the recommendations we have proposed, with commitment and investment, to assure and expand the safe reuse of clinical free-text data for public benefit.

Experience of Ambulatory Cancer Care: Understanding Patients' Perspectives of Quality Using Sentiment Analysis.

The move of cancer treatment into the outpatient setting can impact patient experience. Understanding how service delivery change impacts different people requires service feedback to inform future delivery development. The use of patient experience questionnaires often generates large amount of free-text data that are difficult to analyze. The aim of this study was to describe a proof-of-concept study exploring the experiences and perceptions of people undergoing cancer treatment, using novel analysis techniques to provide rapid free-text data analysis. This was a mixed-methods qualitative analysis from qualitative questions gathered in Finland using the National Cancer Patient Experience Survey (n = 92 of 208 patients) and supplemented with 7 focus groups (31 people with cancer). Data were analyzed using natural language processing, via an automated sentiment analysis algorithm and supported with focus groups to inform the initial thematic analysis. Participants were on average 65 years of age. Of the 196 free-text comments, 121 (73.6%) were positive about patient experiences and 75 (38.5%) negative with suggestions for improvement. Communication between patients and clinicians was an indicator of quality, and lack of psychological support was a barrier to quality care provision. The methodology of using sentiment analysis for free content to review quality was demonstrated through this study as a novel and feasible method to look at large-scale qualitative data. Using the free content on experience of care questionnaire to review gaps or needs in services is valuable in developing future practice.

Severe and fatal medication errors in hospitals: findings from the Norwegian Incident Reporting System

BackgroundEven with global efforts to prevent medication errors, they still occur and cause patient harm. Little systematic research has been done in Norway to address this issue.ObjectivesTo describe the frequency,...