A risk identification model for detection of patients at risk of antidepressant discontinuation.

Purpose: Between 30 and 68% of patients prematurely discontinue their antidepressant treatment, posing significant risks to patient safety and healthcare outcomes. Online healthcare forums have the potential to offer a rich and unique source of data, revealing dimensions of antidepressant discontinuation that may not be captured by conventional data sources. Methods: We analyzed 891 patient narratives from the online healthcare forum, "askapatient.com," utilizing content analysis to create PsyRisk-a corpus highlighting the risk factors associated with antidepressant discontinuation. Leveraging PsyRisk, alongside PsyTAR [a publicly available corpus of adverse drug reactions (ADRs) related to antidepressants], we developed a machine learning-driven algorithm for proactive identification of patients at risk of abrupt antidepressant discontinuation. Results: From the analyzed 891 patients, 232 reported antidepressant discontinuation. Among these patients, 92% experienced ADRs, and 72% found these reactions distressful, negatively affecting their daily activities. Approximately 26% of patients perceived the antidepressants as ineffective. Most reported ADRs were physiological (61%, 411/673), followed by cognitive (30%, 197/673), and psychological (28%, 188/673) ADRs. In our study, we employed a nested cross-validation strategy with an outer 5-fold cross-validation for model selection, and an inner 5-fold cross-validation for hyperparameter tuning. The performance of our risk identification algorithm, as assessed through this robust validation technique, yielded an AUC-ROC of 90.77 and an F1-score of 83.33. The most significant contributors to abrupt discontinuation were high perceived distress from ADRs and perceived ineffectiveness of the antidepressants. Conclusion: The risk factors identified and the risk identification algorithm developed in this study have substantial potential for clinical application. They could assist healthcare professionals in identifying and managing patients with depression who are at risk of prematurely discontinuing their antidepressant treatment.

The influence of auditory rhythms on the speed of inferred motion.

The present research explored the influence of isochronous auditory rhythms on the timing of movement-related prediction in two experiments. In both experiments, participants observed a moving disc that was visible for a predetermined period before disappearing behind a small, medium, or large occluded area for the remainder of its movement. In Experiment 1, the disc was visible for 1 s. During this period, participants were exposed to either a fast or slow auditory rhythm, or they heard nothing. They were instructed to press a key to indicate when they believed the moving disc had reached a specified location on the other side of the occluded area. The procedure measured the (signed) error in participants' estimate of the time it would take for a moving object to contact a stationary one. The principal results of Experiment 1 were main effects of the rate of the auditory rhythm and of the size of the occlusion on participants' judgments. In Experiment 2, the period of visibility was varied with size of the occlusion area to keep the total movement time constant for all three levels of occlusion. The results replicated the main effect of rhythm found in Experiment 1 and showed a small, significant interaction, but indicated no main effect of occlusion size. Overall, the results indicate that exposure to fast isochronous auditory rhythms during an interval of inferred motion can influence the imagined rate of such motion and suggest a possible role of an internal rhythmicity in the maintenance of temporally accurate dynamic mental representations.

The PsyTAR dataset: From patients generated narratives to a corpus of adverse drug events and effectiveness of psychiatric medications.

The "Psychiatric Treatment Adverse Reactions" (PsyTAR) dataset contains patients' expression of effectiveness and adverse drug events associated with psychiatric medications. The PsyTAR was generated in four phases. In the first phase, a sample of 891 drugs reviews posted by patients on an online healthcare forum, "askapatient.com", was collected for four psychiatric drugs: Zoloft, Lexapro, Cymbalta, and Effexor XR. For each drug review, patient demographic information, duration of treatment, and satisfaction with the drugs were reported. In the second phase, sentence classification, drug reviews were split to 6009 sentences, and each sentence was labeled for the presence of Adverse Drug Reaction (ADR), Withdrawal Symptoms (WDs), Sign/Symptoms/Illness (SSIs), Drug Indications (DIs), Drug Effectiveness (EF), Drug Infectiveness (INF), and Others (not applicable). In the third phases, entities including ADRs (4813 mentions), WDs (590 mentions), SSIs (1219 mentions), and DIs (792 mentions) were identified and extracted from the sentences. In the four phases, all the identified entities were mapped to the corresponding UMLS Metathesaurus concepts (916) and SNOMED CT concepts (755). In this phase, qualifiers representing severity and persistency of ADRs, WDs, SSIs, and DIs (e.g., mild, short term) were identified. All sentences and identified entities were linked to the original post using IDs (e.g., Zoloft.1, Effexor.29, Cymbalta.31). The PsyTAR dataset can be accessed via Online Supplement #1 under the CC BY 4.0 Data license. The updated versions of the dataset would also be accessible in https://sites.google.com/view/pharmacovigilanceinpsychiatry/home.

A systematic approach for developing a corpus of patient reported adverse drug events: A case study for SSRI and SNRI medications.

"Psychiatric Treatment Adverse Reactions" (PsyTAR) corpus is an annotated corpus that has been developed using patients narrative data for psychiatric medications, particularly SSRIs (Selective Serotonin Reuptake Inhibitor) and SNRIs (Serotonin Norepinephrine Reuptake Inhibitor) medications. This corpus consists of three main components: sentence classification, entity identification, and entity normalization. We split the review posts into sentences and labeled them for presence of adverse drug reactions (ADRs) (2168 sentences), withdrawal symptoms (WDs) (438 sentences), sign/symptoms/illness (SSIs) (789 sentences), drug indications (517), drug effectiveness (EF) (1087 sentences), and drug infectiveness (INF) (337 sentences). In the entity identification phase, we identified and extracted ADRs (4813 mentions), WDs (590 mentions), SSIs (1219 mentions), and DIs (792). In the entity normalization phase, we mapped the identified entities to the corresponding concepts in both UMLS (918 unique concepts) and SNOMED CT (755 unique concepts). Four annotators double coded the sentences and the span of identified entities by strictly following guidelines rules developed for this study. We used the PsyTAR sentence classification component to automatically train a range of supervised machine learning classifiers to identifying text segments with the mentions of ADRs, WDs, DIs, SSIs, EF, and INF. SVMs classifiers had the highest performance with F-Score 0.90. We also measured performance of the cTAKES (clinical Text Analysis and Knowledge Extraction System) in identifying patients' expressions of ADRs and WDs with and without adding PsyTAR dictionary to the core dictionary of cTAKES. Augmenting cTAKES dictionary with PsyTAR improved the F-score cTAKES by 25%. The findings imply that PsyTAR has significant implications for text mining algorithms aimed to identify information about adverse drug events and drug effectiveness from patients' narratives data, by linking the patients' expressions of adverse drug events to medical standard vocabularies. The corpus is publicly available at Zolnoori et al. [30].

Semantic-Enhanced Query Expansion System for Retrieving Medical Image Notes.

Most current image retrieval methods require constructing semantic metadata for representing image content. To manually create semantic metadata for medical images is time-consuming, yet it is a crucial component for query expansion. We proposed a new method for searching medical image notes that uses semantic metadata to improve query expansion and leverages a knowledge model developed specifically for the medical image domain to create relevant metadata. We used a syntactic parser and the Unified Medical Language System to analyze the corpus and store text information as semantic metadata in a knowledge model. Our new method has an interactive interface that allows users to provide relevance feedback and construct new queries more efficiently. Sixteen medical professionals evaluated the query expansion module, and each evaluator had prior experience searching for medical images. When using the initial query as the baseline standard, expanded queries achieved a performance boost of 22.6% in terms of the relevance score on first ten results (P-value<0.05). When using Google as another baseline, our system performed 24.6% better in terms of relevance score on the first ten results (P-value<0.05). Overall, 75% of the evaluators said the semantic-enhanced query expansion workflow is logical, easy to follow, and comfortable to use. In addition, 62% of the evaluators preferred using our system instead of Google. Evaluators who were positive about our system found the knowledge map-based visualization of candidate medical search terms helpful in refining cases from the initial search results.

Influence of socioeconomic factors and race on birth outcomes in urban Milwaukee.

PURPOSE: A national study found that infants born in low socioeconomic areas had the worst infant mortality rates (IMRs) and the highest racial disparity. Racial disparities in birth outcomes are also evident in the city of Milwaukee, with African American infants at 3 times greater the risk than white infants. This study was conducted to examine the influence of socioeconomic status (SES) and race on birth outcomes in the city of Milwaukee. METHODS: Milwaukee ZIP codes were stratified into lower, middle, and upper SES groups. IMR, low birth weight, and preterm birth rates by race were analyzed by SES group for the years 2003 to 2007. RESULTS: The overall IMR for the lower, middle, and upper SES groups were 12.4, 10.7, and 7.7, respectively. The largest racial disparity in IMR (3.1) was in the middle SES group, versus lower (1.6) and upper (1.8) SES groups. The overall percent of low birth weight infants for the lower, middle, and upper SES groups was 10.9%, 9.5%, and 7.5%, respectively. Racial disparity ratios in low birth weight were 2.0, 1.9, and 1.9 for lower, middle and upper SES groups. The overall percent of preterm birth was 15.4%, 13.2%, and 10.6% of births within the lower, middle, and upper SES groups, respectively, with a disparity ratio of 1.6 across all SES groups. CONCLUSIONS: For all outcomes, African American infants born in the upper SES group fared the same or worse than white infants born in the lower SES group. Although higher SES appeared to have a protective effect for whites in Milwaukee, it did not have the same protective effect for African Americans.

SNOMED CT coding variation and grouping for "other findings" in a longitudinal study on urea cycle disorders.

Semantic interoperability requires consistent use of controlled terminologies. However, non-terminology experts (although perhaps experts in a particular domain) are prone to produce variant coding. We examine this problem by investigating SNOMED CT coding variation for other findings reported on case report forms from a clinical research study on urea cycle disorders. The natural language findings from the forms were normalized, and the associated SNOMED CT concept descriptions were compared. The subset of normalized strings associated with two different concept descriptions were further compared to determine the relationship among the associated SNOMED CT concepts. We found 45% of the concept description pairs were associated with two hierarchically related concepts or with the same concept, while 55% were associated with two unrelated concepts. Clearer guidelines for use of SNOMED CT in particular contexts, or structured data entry tools tailored to the needs of non-expert coders, are needed to better manage coding variation.

Mapping nursing concepts to ontologies for evidence-based nursing.

In this paper we describe a project to map natural language expressions of nursing-related concepts to standard ontologies in support of an evidence-based nursing initiative. Natural language concept expressions are identified from syntheses of nursing knowledge and mapped to ICNP and SNOMED-CT.

Comparing heterogeneous SNOMED CT coding of clinical research concepts by examining normalized expressions.

OBJECTIVE: A continual problem confronting the implementation of standardized vocabularies such as SNOMED CT is that their expressive flexibility and power provide more than one way to represent a given concept. The goal of this study was to investigate how the CliniClue Expression Transformer tool could be used to help in discerning similarities and differences among three separate sets of clinical research concepts coded in SNOMED CT by three different paid expert coding companies. METHODS: Initial editing of the companies' coded datasets was required to enable accurate input into CliniClue Version: 2006.2.0030 Expression Transformer tool. The normal forms of the company codings for the 319 clinical research question/answer sets were compared to determine whether they were equivalent or otherwise related (e.g., if one was subsumed by the other). Basic frequencies were computed for (957) pairwise comparisons of each of 319 concepts each coded by the three expert coders, and the implications of the results discussed. RESULTS: The primary finding from this study was that, for each of the paired comparisons, approximately half of the time the companies' codings could be related, primarily via subsumption. The greatest percentage of equivalent concepts between any two companies was 33%. These same two companies also agreed most often on the core clinical concept measure from an earlier study by the authors. CONCLUSION: Heterogeneity among coders using the same controlled terminology appears inescapable despite the extensive efforts of terminological standards developers and implementers. In our study, the computable determination of equivalence of discordantly coded concepts still failed to yield acceptably comparable data. A clearer articulation, and perhaps a simplification, of rules for the consistent use for terminologies such as SNOMED CT is needed.

Medical identity theft: an emerging problem for informatics.

This poster reports a preliminary review of medical identity theft. Financial identity theft has received a great deal of media attention. Medical identity theft is a particular kind of identity theft that has received little attention. There are two main subtypes of medical identity theft. In the first type the stolen medical identity is used to receive medical services, and in the second type the stolen medical identity is used to commit healthcare fraud.

ECIDS: standardizing the data for Birth to 3 programs.

There is a compelling need for standardized coding schemes to represent data collected for children receiving services through Milwaukee County early intervention (Birth to 3) programs. We are standardizing Birth to 3 data in our ongoing development of the Early Childhood Integrated Database System. Our efforts balance the need for individualized service plans with the need for aggregate analysis.

The MED-AUDIT Ontology Version 1: an ontology for auditing the accessibility of medical devices.

The MED-AUDIT tool uses an electronic questionnaire designed to assess and quantify the accessibility of medical devices. We created the MED-AUDIT ontology to facilitate the auditing of medical devices by applying standard thesaurus construction techniques to concepts and vocabulary extracted from the questionnaire. A key feature of this ontology is that it links concepts between the MED-AUDIT and the International Classification of Functioning, Disability and Health (ICF).

An ethical review of euthanasia web sites.

Euthanasia is an act of mercifully killing or letting die a sick or injured individual, and is an alternative to palliative care. As a healthcare related web site, a euthanasia site should satisfy standard ethical criteria such as the HONcode. In this paper, we report our preliminary review of euthanasia web sites for compliance with HONcode criteria.

Assessing patient safety awareness and needs in rural hospitals in one US state.

Several initiatives have addressed patient safety by enabling electronic voluntary reporting of adverse events within academic medical centres in urban settings. Such initiatives are lacking in the rural context, and it remains unknown whether the same challenges and solutions apply to rural hospitals. The purpose of this study is to provide insight into the organisational culture and level of readiness to adopt patient safety strategies in a rural setting, as well as to identify critical issues pertaining to the rural context that need to inform the design of such strategies. We conducted telephone interviews with administrators and healthcare providers of rural hospitals in one US state. Questions referred to the respondents' current reporting mechanism, its advantages and disadvantages, and organisational patient safety culture. A total of 16 administrators and 14 providers of eight rural hospitals in the state of Missouri were interviewed. Findings indicate that very few administrators felt that there was a timely response to adverse event reports. Half of the administrators stated that the current mechanism is an appropriate and adequate outlet to ensure patient safety. None of the healthcare providers found errors and adverse events to be over-reported; the majority believe that they are being under-reported. Only 36% of the care providers interviewed have themselves reported an error or adverse drug event during their tenure with their organisation. The study findings demonstrate a definite need for improvement of the current infrastructure of rural hospitals in order to enable an effective outlet for ensuring patient safety.

To telemedically err is human.

BACKGROUND: As telemedicine alters the process of health care and introduces new technology, the extent to which it introduces new errors or allows for the occurrence of familiar errors needs to be examined. TELEMEDICINE'S IMPACT ON PATIENT SAFETY FEATURES OF TRADITIONAL CARE: The accuracy of diagnostic decisions reached via telemedicine can be directly affected by the technology's limitations and the care providers' lack of training. Telemedicine could increase the risk of familiar types of patient-provider communication failure and introduce the possibility of cumulative errors. TELEMEDICINE'S IMPACT ON NEW CARE FEATURES AND CONCEPTS: Telemedical applications that use the Internet to enhance disease management and detection and monitoring of symptoms may place the privacy and confidentiality of individual health information at risk, which imposes a possible barrier to communication. In addition, home care patients' functional limitations need to be addressed by human factors engineering. RECOMMENDATIONS: Patient safety should be integrated in organizational readiness and budget planning for telemedical interventions in hospitals, academic settings, nursing homes, home care agencies, and other health care settings. Specific recommendations are proposed for the development and diffusion of standards in telemedical care, risk management and reduction, and continuous quality improvement. SUMMARY AND CONCLUSIONS: To address patient safety and provide high-quality care, a framework for addressing and examining telemedical errors needs to be established.

Differences in the effects of filters on health information retrieval from the Internet in three languages from three countries: a comparative study.

We selected twenty search terms on woman's health from various sources and tested them on Google, once with strict filter on and once with filter off. Searches were specified to three countries (Mainland China, Germany, USA), in three languages (Simplified Chinese, German, US English). We found that the proportion of relevant women's health web sites that were blocked was quite high. For the Chinese language web sites originated in China, 72.6% of the blocked web sites were relevant. For the German language web sites originated in Germany, nearly half (49.4%) were relevant. For the US English web sites originated in the US, 95% were relevant. We concluded that people might unknowingly miss potentially important health information due to information filtering.

Evidence-based retrieval in evidence-based medicine.

OBJECTIVE: Clinical decisions based on a meta-analysis that is based on an ineffective retrieval strategy may have serious negative consequences for patients. The study objective was to investigate the extent to which meta-analyses report proof of their retrieval strategies' effectiveness. METHODS: The authors examined a random sample (n = 100) of articles in the 1996 to 2002 full-text subset of Ovid MEDLINE indexed as "meta-analysis." We classified the articles in three ways: the article (A) reported both a retrieval strategy in sufficient detail (such that it could be repeated) and with evidence of the strategy's effectiveness, (B) reported a retrieval strategy in sufficient detail but not with evidence of the strategy's effectiveness, or (C) neither reported a strategy in detail nor evidence of the strategy's effectiveness. Articles classified as (A) were further classified according to the level of evidence reported. RESULTS: Of the eighty-nine articles in our final analysis, six (6.7%) were classified as category (A), fifty-seven (64%) as (B), and twenty-six (29%) as (C). Articles in category (A) reported a previously validated search, a published strategy, or strategy based on expert opinion. CONCLUSION: Peer-review standards must be developed that require authors of meta-analyses to report evidence for the effectiveness of their retrieval strategies.

Evidence-based retrieval in E-health.

In this chapter we address the issue of standards for information retrieval to support decision making in e-health. Specifically, we consider the issue of evidence-based retrieval in the e-health domains of the consumer, healthcare practitioner, healthcare researcher, and genomics researcher. We present the results of a preliminary study to assess the current state of evidence-based retrieval in e-health. Within this study, we reviewed articles in e-health and telemedicine to determine the extent to which authors provide details of the information retrieval strategies used, as well as evidence of the effectiveness of those strategies. We also examined the extent to which the associated journals require authors of reviews to explicitly provide details of information retrieval strategies that they used, as well as reporting any evidence for the effectiveness of those strategies.

Marginalization and health geomatics.

Marginalized groups have been defined as groups that have been peripheralized from the center of society. Increasing nursing knowledge of marginalized groups and the dynamics of population diversity will enable nurses to better recognize shifting health patterns, plan for utilization of health services, and determine ethnic and cultural differences that exist in marginalized populations. The authors of this article review theoretical models responsible for defining the concept marginalization, describe geographical information systems as a recommended tool to evaluate marginalized groups, and provide a case study utilizing tools and maps as a means of assessing marginal situations.