Reliability and validity of the Turkish version of Scleroderma Skin Patient-Reported Outcome in patients with systemic sclerosis.

Objectives: This study aimed to translate the Scleroderma Skin Patient-Reported Outcome (SSPRO) questionnaire to the Turkish (SSPRO-T) language and to assess its validity and reliability. Patients and methods: Fifty-four systemic sclerosis (SSc) patients (51 females, 3 males; mean age: 49.8±10.4 years; range, 22 to 65 years) participated in the reliability and validity analysis between October 2022 and December 2022. The translation and cross-cultural adaptation of the SSPRO-T was applied in accordance with the procedure described by the Beaton guidelines. The SSPRO-T, the Scleroderma Health Assessment Questionnaire (SHAQ), the Health Assessment Questionnaire Disability Index (HAQ-DI), Skindex-29, and patient global skin severity were conducted in all participants for construct validity. The SSPRO-T was retested to assess its reliability after seven days. Results: The SSPRO-T had a four-factor structure. The total SSPRO-T score and its subgroups correlated positively with SHAQ, HAQ-DI, Skindex-29, and patient global skin severity. The internal consistency and reliability were excellent in overall SSPRO-T and in the subgroups: physical effect, emotional effect, physical limitation, and social effect (Cronbach's α=0.94, 0.80, 0.95, 0.93, and 0.84, respectively). The SSPRO-T had excellent test-retest reliability (r=0.91, p<0.001). In addition, no floor effect or ceiling effect was observed. Conclusion: The SSPRO-T questionnaire is a reliable and valid tool and can be used in research and clinical practice in Turkish patients with SSc.

Sodium-Glucose Co-transporter Type-2 Inhibitors' Effect on Quality of Life in Older Adult Population.

Introduction The global elderly population is expanding, with chronic conditions like diabetes diminishing their quality of life. Sodium-glucose co-transporter type 2 (SGLT-2) inhibitors hold promise in improving quality of life by addressing hypervolemia, obesity, and lipid irregularities. However, these drugs can lead to adverse effects, such as polyuria, dehydration, and weight loss, which may detrimentally impact older patients. We aimed to investigate the association between SGLT-2 inhibitors and quality of life in older adults with diabetes. Methods The research included 100 type II diabetes mellitus patients over 65, without active infections, malignancies, immunodeficiencies, and hematological disorders. Fifty patients were using empagliflozin or dapagliflozin and 50 patients were using other oral antidiabetics for at least six months. Patient demographics, laboratory studies, drug usage and side effects, additional diseases, Geriatric Depression Scale scores, and World Health Organization Quality of Life OLD (WHOQoL-OLD) module scores were noted. Results No significant difference between gender distribution, SGLT usage, chronic disease existence, chronic disease count, depression scores, or incidents of chronic diseases other than hyperlipidemia was observed. Hyperlipidemia incidence was significantly higher in the SGLT group, while other laboratory parameters were not statistically significantly different between groups. There were no significant differences in autonomy, past-present-future activities, social skills, death, intimacy, and total WHOQoL-OLD scores between the two groups. However, there were statistically significantly worse outcomes in patients with at least one SGLT adverse effect in terms of sensory quality of life scores. Dehydration existence was negatively correlated with lower autonomy, PPF activities, and total quality of life scores. Multivariate linear regression analysis showed no significant differences in the total WHOQoL-OLD score after adjusting for confounding factors. Conclusion Age and depression remained the main factors affecting the quality of life in diabetic patients. SGLT-2 inhibitor side effects did not decrease the quality of life in older individuals, who are more prone to unfavorable consequences.

Semaphorin 3A levels in vascular and nonvascular phenotypes in systemic sclerosis.

OBJECTIVE: Semaphorin 3A (Sema3A) plays a regulatory role in immune responses. The aim of this study was to evaluate Sema3A levels in patients with systemic sclerosis (SSc), especially in major vascular involvements such as digital ulcer (DU), scleroderma renal crisis (SRC), pulmonary arterial hypertension (PAH), and to compare Sema3A level with SSc disease activity. METHODS: In SSc patients, patients with DU, SRC, or PAH were grouped as major vascular involvements and those without as nonvascular, and Sema3A levels were compared between the groups and with a healthy control group. The Sema3A levels and acute phase reactants in SSc patients, as well as their association with the Valentini disease activity index and modified Rodnan skin score, were evaluated. RESULTS: The Sema3A values (mean ±â€…SD) were 57.60 ±â€…19.81 ng/mL in the control group (n = 31), 44.32 ±â€…5.87 ng/mL in patients with major vascular involvement SSc (n = 21), and 49.96 ±â€…14.00 ng/mL in the nonvascular SSc group (n = 35). When all SSc patients were examined as a single group, the mean Sema3A value was significantly lower than controls (P = .016). The SSc with major vascular involvement group had significantly lower Sema3A levels than SSc with nonmajor vascular involvement group (P = .04). No correlation was found between Sema3A, acute phase reactants, and disease activity scores. Also, no relationship was observed between Sema3A levels and diffuse (48.36 ±â€…11.47 ng/mL) or limited (47.43 ±â€…12.38 ng/mL) SSc types (P = .775). CONCLUSION: Our study suggests that Sema3A may play a significant role in the pathogenesis of vasculopathy and can be used as a biomarker in SSc patients with vascular complications such as DU and PAH.

A Context-Enhanced De-identification System.

Many modern entity recognition systems, including the current state-of-the-art de-identification systems, are based on bidirectional long short-term memory (biLSTM) units augmented by a conditional random field (CRF) sequence optimizer. These systems process the input sentence by sentence. This approach prevents the systems from capturing dependencies over sentence boundaries and makes accurate sentence boundary detection a prerequisite. Since sentence boundary detection can be problematic especially in clinical reports, where dependencies and co-references across sentence boundaries are abundant, these systems have clear limitations. In this study, we built a new system on the framework of one of the current state-of-the-art de-identification systems, NeuroNER, to overcome these limitations. This new system incorporates context embeddings through forward and backward n -grams without using sentence boundaries. Our context-enhanced de-identification (CEDI) system captures dependencies over sentence boundaries and bypasses the sentence boundary detection problem altogether. We enhanced this system with deep affix features and an attention mechanism to capture the pertinent parts of the input. The CEDI system outperforms NeuroNER on the 2006 i2b2 de-identification challenge dataset, the 2014 i2b2 shared task de-identification dataset, and the 2016 CEGS N-GRID de-identification dataset (p < 0.01). All datasets comprise narrative clinical reports in English but contain different note types varying from discharge summaries to psychiatric notes. Enhancing CEDI with deep affix features and the attention mechanism further increased performance.

Comparing Attention-Deficit/Hyperactivity Disorder and Oppositional Defiant Disorder Singly and Together in Terms of Behavioral Problems, Family Conflict, and Cognitive Functions.

OBJECTIVES: The purpose of this study is to profile three groups of children with attention-deficit/hyperactivity disorder (ADHD), oppositional defiant disorder (ODD), and both ADHD and ODD, through analyzing their cognitive abilities, personality traits, and family characteristics. METHODS: The study included 60 patients, with 20 patients in each group. Patients were selected according to the DSM IV criteria. They completed the Wechsler Intelligence Scale for Children-Revised (WISC-R) and the Bender Visual Motor Gestalt Test, and their mothers filled out the Child Behavior Checklist (CBCL) and Marital Conflict Questionnaire. RESULTS: There were no significant differences in picture completion, block design, and coding, which are the WISC-R subtests, between the three groups. In addition, there was no significant difference in verbal, performance, and WISC-R scores. Finally, there was no significant difference when the subdivisions of the CBCL externalizing and internalizing behaviors were analyzed individually. The Frequency of Marital Conflict Score and Conflict Expansion Score were analyzed, and there were no significant differences found between the three groups. The highest average of the Bender Visual Motor Gestalt Test was in the ODD group, whereas the lowest average was in the ADHD group. CONCLUSION: When comparing ADHD and ODD in terms of cognitive abilities, the observed differences may be because ODD has no genetic or organic component, and ADHD has an organic basis. In ODD, cognitive abilities are intact, which should underline the environmental and family factors.

Patient Privacy in the Era of Big Data.

Privacy was defined as a fundamental human right in the Universal Declaration of Human Rights at the 1948 United Nations General Assembly. However, there is still no consensus on what constitutes privacy. In this review, we look at the evolution of privacy as a concept from the era of Hippocrates to the era of social media and big data. To appreciate the modern measures of patient privacy protection and correctly interpret the current regulatory framework in the United States, we need to analyze and understand the concepts of individually identifiable information, individually identifiable health information, protected health information, and de-identification. The Privacy Rule of the Health Insurance Portability and Accountability Act defines the regulatory framework and casts a balance between protective measures and access to health information for secondary (scientific) use. The rule defines the conditions when health information is protected by law and how protected health information can be de-identified for secondary use. With the advents of artificial intelligence and computational linguistics, computational text de-identification algorithms produce de-identified results nearly as well as those produced by human experts, but much faster, more consistently and basically for free. Modern clinical text de-identification systems now pave the road to big data and enable scientists to access de-identified clinical information while firmly protecting patient privacy. However, clinical text de-identification is not a perfect process. In order to maximize the protection of patient privacy and to free clinical and scientific information from the confines of electronic healthcare systems, all stakeholders, including patients, health institutions and institutional review boards, scientists and the scientific communities, as well as regulatory and law enforcement agencies must collaborate closely. On the one hand, public health laws and privacy regulations define rules and responsibilities such as requesting and granting only the amount of health information that is necessary for the scientific study. On the other hand, developers of de-identification systems provide guidelines to use different modes of operations to maximize the effectiveness of their tools and the success of de-identification. Institutions with clinical repositories need to follow these rules and guidelines closely to successfully protect patient privacy. To open the gates of big data to scientific communities, healthcare institutions need to be supported in their de-identification and data sharing efforts by the public, scientific communities, and local, state, and federal legislators and government agencies.

Modes of De-identification.

De-identification of protected health information is an essential method for protecting patient privacy. Most institutes require de-identification of patient data prior to conducting scientific studies; therefore, it is important for clinical scientists to be cognizant of all modes of de-identification and all services provided by their de-identification tools. In this article, we discuss eight different modes of de-identification that yield de-identified data at different levels of quality. Most of these modes can be used in combination to achieve the best performance.

Challenges and Insights in Using HIPAA Privacy Rule for Clinical Text Annotation.

The Privacy Rule of Health Insurance Portability and Accountability Act (HIPAA) requires that clinical documents be stripped of personally identifying information before they can be released to researchers and others. We have been manually annotating clinical text since 2008 in order to test and evaluate an algorithmic clinical text de-identification tool, NLM Scrubber, which we have been developing in parallel. Although HIPAA provides some guidance about what must be de-identified, translating those guidelines into practice is not as straightforward, especially when one deals with free text. As a result we have changed our manual annotation labels and methods six times. This paper explains why we have made those annotation choices, which have been evolved throughout seven years of practice on this field. The aim of this paper is to start a community discussion towards developing standards for clinical text annotation with the end goal of studying and comparing clinical text de-identification systems more accurately.

The Challenges of Creating a Gold Standard for De-identification Research.

We created a Gold Standard corpus comprised over 20,000 records of annotated narrative clinical reports for use in the training and evaluation of NLM Scrubber, a de-identification software system for medical records. Our experience with designing the corpus demonstrated the conceptual complexity of the task.

Piloting a deceased subject integrated data repository and protecting privacy of relatives.

Use of deceased subject Electronic Health Records can be an important piloting platform for informatics or biomedical research. Existing legal framework allows such research under less strict de-identification criteria; however, privacy of non-decedent must be protected. We report on creation of the decease subject Integrated Data Repository (dsIDR) at National Institutes of Health, Clinical Center and a pilot methodology to remove secondary protected health information or identifiable information (secondary PxI; information about persons other than the primary patient). We characterize available structured coded data in dsIDR and report the estimated frequencies of secondary PxI, ranging from 12.9% (sensitive token presence) to 1.1% (using stricter criteria). Federating decedent EHR data from multiple institutions can address sample size limitations and our pilot study provides lessons learned and methodology that can be adopted by other institutions.

De-identification of Address, Date, and Alphanumeric Identifiers in Narrative Clinical Reports.

INTRODUCTION: The Privacy Rule of Health Insurance Portability and Accountability Act requires that clinical documents be stripped of personally identifying information before they can be released to researchers and others. We have been developing a software application, NLM Scrubber, to de-identify narrative clinical reports. METHODS: We compared NLM Scrubber with MIT's and MITRE's de-identification systems on 3,093 clinical reports about 1,636 patients. The performance of each system was analyzed on address, date, and alphanumeric identifier recognition separately. Their overall performance on de-identification and on conservation of the remaining clinical text was analyzed as well. RESULTS: NLM Scrubber's sensitivity on de-identifying these identifiers was 99%. It's specificity on conserving the text with no personal identifiers was 99% as well. CONCLUSION: The current version of the system recognizes and redacts patient names, alphanumeric identifiers, addresses and dates. We plan to make the system available prior to the AMIA Annual Symposium in 2014.

Visualization of patient prescription history data in emergency care.

Interpreting patient's medication history from long textual data can be unwieldy especially in emergency care. We developed a real-time software application that converts one-year-long patient prescription history data into a visually appealing and information-rich timeline chart. The chart can be digested by healthcare providers quickly; hence, it could be an invaluable clinical tool when the rapid response time is crucial as in stroke or severe trauma cases. Furthermore, the visual clarity of the displayed information may help providers minimize medication errors. The tool has been deployed at the emergency department of a trauma center. Due to its popularity, we developed another version of this tool. It provides more granular drug dispensation information, which clinical pharmacists find very useful in their routine medication-reconciliation efforts.

The pattern of name tokens in narrative clinical text and a comparison of five systems for redacting them.

OBJECTIVE: To understand the factors that influence success in scrubbing personal names from narrative text. MATERIALS AND METHODS: We developed a scrubber, the NLM Name Scrubber (NLM-NS), to redact personal names from narrative clinical reports, hand tagged words in a set of gold standard narrative reports as personal names or not, and measured the scrubbing success of NLM-NS and that of four other scrubbing/name recognition tools (MIST, MITdeid, LingPipe, and ANNIE/GATE) against the gold standard reports. We ran three comparisons which used increasingly larger name lists. RESULTS: The test reports contained more than 1 million words, of which 2388 were patient and 20,160 were provider name tokens. NLM-NS failed to scrub only 2 of the 2388 instances of patient name tokens. Its sensitivity was 0.999 on both patient and provider name tokens and missed fewer instances of patient name tokens in all comparisons with other scrubbers. MIST produced the best all token specificity and F-measure for name instances in our most relevant study (study 2), with values of 0.997 and 0.938, respectively. In that same comparison, NLM-NS was second best, with values of 0.986 and 0.748, respectively, and MITdeid was a close third, with values of 0.985 and 0.796 respectively. With the addition of the Clinical Center name list to their native name lists, Ling Pipe, MITdeid, MIST, and ANNIE/GATE all improved substantially. MITdeid and Ling Pipe gained the most--reaching patient name sensitivity of 0.995 (F-measure=0.705) and 0.989 (F-measure=0.386), respectively. DISCUSSION: The privacy risk due to two name tokens missed by NLM-NS was statistically negligible, since neither individual could be distinguished among more than 150,000 people listed in the US Social Security Registry. CONCLUSIONS: The nature and size of name lists have substantial influences on scrubbing success. The use of very large name lists with frequency statistics accounts for much of NLM-NS scrubbing success.

Extracting laboratory test information from biomedical text.

BACKGROUND: No previous study reported the efficacy of current natural language processing (NLP) methods for extracting laboratory test information from narrative documents. This study investigates the pathology informatics question of how accurately such information can be extracted from text with the current tools and techniques, especially machine learning and symbolic NLP methods. The study data came from a text corpus maintained by the U.S. Food and Drug Administration, containing a rich set of information on laboratory tests and test devices. METHODS: THE AUTHORS DEVELOPED A SYMBOLIC INFORMATION EXTRACTION (SIE) SYSTEM TO EXTRACT DEVICE AND TEST SPECIFIC INFORMATION ABOUT FOUR TYPES OF LABORATORY TEST ENTITIES: Specimens, analytes, units of measures and detection limits. They compared the performance of SIE and three prominent machine learning based NLP systems, LingPipe, GATE and BANNER, each implementing a distinct supervised machine learning method, hidden Markov models, support vector machines and conditional random fields, respectively. RESULTS: Machine learning systems recognized laboratory test entities with moderately high recall, but low precision rates. Their recall rates were relatively higher when the number of distinct entity values (e.g., the spectrum of specimens) was very limited or when lexical morphology of the entity was distinctive (as in units of measures), yet SIE outperformed them with statistically significant margins on extracting specimen, analyte and detection limit information in both precision and F-measure. Its high recall performance was statistically significant on analyte information extraction. CONCLUSIONS: Despite its shortcomings against machine learning methods, a well-tailored symbolic system may better discern relevancy among a pile of information of the same type and may outperform a machine learning system by tapping into lexically non-local contextual information such as the document structure.

Comparison of electronic pharmacy prescription records with manually collected medication histories in an emergency department.

STUDY OBJECTIVE: Medication history is an essential part of patient assessment in emergency care. Patient-reported medication history can be incomplete. We study whether an electronic pharmacy-sourced prescription record can supplement the patient-reported history. METHODS: In a community hospital, we compared the patient-reported history obtained by triage nurses to a proprietary electronic pharmacy record in all emergency department (ED) patients during 3 months. RESULTS: Of 9,426 triaged patients, 5,001 (53%) had at least 1 (mean 7.7) prescription medication in the full-year electronic pharmacy record. Counting only recent prescription medications (supply lasting to at least 7 days before the ED visit), 3,688 patients (39%) had at least 1 (mean 4.0) recent medication. After adjustment for possible false-positive results, recent electronic prescription medication record enriched the patient-reported history by 28% (adding 1.1 drugs per patient). However, only 60% of patients with any active prescription medications from either source had any recent prescription medications in their electronic pharmacy record. CONCLUSION: The electronic pharmacy prescription record augments the manually collected history.

ICU Outcome Predictions using Physiologic Trends in the First Two Days.

AIMS: This study aims to accurately predict patient mortality in the ICU. Given all physiologic measurements in the first 48 hours of the ICU stay, the Bayesian model of the study predicts outcome with a posterior probability. METHODS: This study modeled the outcome as a binary random variable dependent on trends of daily physiologic measures of the patient, where trends were conditionally independent given the outcome. A two-day trend is a sequence of two discrete values, one for each day. Each value (low, medium, high or unmeasured) is a function of the arithmetic mean of that measure on the corresponding day. RESULTS: The prediction performance of the model was measured as the minimum of sensitivity and positive predictive values. The model yielded a score of 0.39 along with a Hosmer-Lemeshow H statistic of 36, which measures calibration. The perfect scores would be 1.0 and 0, respectively. CONCLUSION: The prediction performance of the study was an improvement over the established ICU scoring metric SAPS-I, whose score was 0.32. Calibration of the model outputs was comparable to that of SAPS-I.

Biological entity recognition with conditional random fields.

Due to the rapid evolution of molecular biology and the lack of naming standards, biological entity recognition (BER) remains a challenging task for information extraction and natural language understanding. In this study, we presented a statistical machine learning approach for extracting features, modeling, and predicting biological named entities. Our approach utilizes UMLS semantic types together with MetaMap, SemRep, and ABGene, as well as the conditional random fields (CRF) framework, and learns both the structure and parameters of a statistical model. Results of this study are competitive with the results of the state of the art tools in this field. Unlike competing similar approaches, the presented method is fully automatic, hence more generalizable and directly transferable to other named entity recognition (NER) problems in medical informatics.

Interpretation of data and identification of information are sine qua non for modern digital information services.

Digital information is at the center of the new scientific endeavor and, if managed carefully, it may bridge scientists across disciplines. Scientists, clinicians, and the public need enabling tools to overcome barriers to communication of biomedical information. Biomedical information systems need to (1) interpret queries and the needs of users accurately, (2) identify, evaluate and combine all relevant information among a comprehensive set of sources, and (3) provide users the right information that they seek.

Semi-automatic indexing of full text biomedical articles.

The main application of U.S. National Library of Medicine's Medical Text Indexer (MTI) is to provide indexing recommendations to the Library's indexing staff. The current input to MTI consists of the titles and abstracts of articles to be indexed. This study reports on an extension of MTI to the full text of articles appearing in online medical journals that are indexed for Medline. Using a collection of 17 journal issues containing 500 articles, we report on the effectiveness of the contribution of terms by the whole article and also by each section. We obtain the best results using a model consisting of the sections Results, Results and Discussion, and Conclusions together with the article's title and abstract, the captions of tables and figures, and sections that have no titles. The resulting model provides indexing significantly better (7.4%) than what is currently achieved using only titles and abstracts.