Psychometric validation of diabetes distress scale in Bangladeshi population

Diabetes Distress (DD)—an emotional or affective state arise from challenge of living with diabetes and the burden of self-care—negatively impact diabetes management and quality of life of T2DM patients. Early detection and management of DD is key to efficient T2DM management. The study aimed at developing a valid and reliable instrument for Bangladeshi patients as unavailability such a tool posing challenge in diabetes care. Linguistically adapted, widely used, 17-item Diabetes Distress Scale (DDS), developed through forward–backward translation from English to Bengali, was administered on 1184 T2DM patients, from four diabetes hospitals in Bangladesh. Psychometric assessment of the instrument included, construct validity using principal component factor analysis, internal consistency using Cronbach’s α and discriminative validity through independent t-test and test–retest reliability using intraclass-correlation coefficient (ICC) and Kappa statistics. Factor analysis extracted 4 components similar to original DDS domains, confirms the construct validity. The scale demonstrated satisfactory internal consistency (α = 0.838), stability (test–retest ICC = 0.941) and good agreement across repeated measurements (Kappa = 0.584). Discriminative validity revealed that patients with complication (p < 0.001) and those are on insulin (p < 0.001) had significantly higher distress scores in all domains. Bengali version of DDS is a valid and reliable tool for assessing distress among Bangladeshi T2DM patients.

PERFORMANCE OF MACHINE LEARNING ALGORITHMS FOR MAPPING AND FORECASTING OF FLASH FLOOD SUSCEPTIBILITY IN TETOUAN, MOROCCO

Since the industrial revolution, the world is experiencing a huge change in its climate, which causes many imbalances such as flash floods (FF). The aim of this study is to propose a new approach for detection and forecasting of flash flood susceptibility in the city of Tetouan, Morocco. For this regard, support vector machine (SVM), logistic regression (LR), random forest (RF), Naïve Bayes (NB) and Artificial neural network (ANN) are used based on 1101 points (680 flood points and 421 non-flood points) and 9 flash-flood predictors (Elevation , Slope , Aspect , LU/LC , Stream Power Index , Plan curvature , Profile Curvature , Topographic Position Index and Topographic Wetness Index ) that were extracted from the DEM (10m resolution) and satellite imagery (Sentinel 2B) of the study area . Models were trained on 70% and tested on 30% of this dataset also they were evaluated using several metrics such as the Receiver Operating Characteristic (ROC) Curve, precision, recall, score and kappa index. The result demonstrated that RF (AUC = 0.99, Accuracy = 96%, Kappa statistics = 0.92) has the highest performance, followed by ANN (AUC = 0.98, Accuracy = 95%, Kappa statistics = 0.89) and SVM (AUC = 0.96, Accuracy = 92%, Kappa statistics = 0.80). The proposed approach is an effective tool for forecasting and predicting FF that can help reduce the severity of this disaster.

Accuracy and agreement of national spine register data for 474 patients compared to corresponding electronic patient records

Purpose Data quality is essential for all types of research, including health registers. However, data quality is rarely reported. We aimed to assess the accuracy of data in a national spine register (NORspine) and its agreement with corresponding data in electronic patient records (EPR). Methods We compared data in NORspine registry against data in (EPR) for 474 patients operated for spinal stenosis in 2015 and 2016 at four public hospitals, using EPR as the gold standard. We assessed accuracy using the proportion correctly classified (PCC) and sensitivity. Agreement was quantified using Kappa statistics or interaclass correlation coefficient (ICC). Results The mean age (SD) was 66 (11) years, and 54% were females. Compared to EPR, surgeon-reported perioperative complications displayed weak agreement (kappa (95% CI) = 0.51 (0.33–0.69)), PCC of 96%, and a sensitivity (95% CI) of 40% (23–58%). ASA classification had a moderate agreement (kappa (95%CI) = 0.73 (0.66–0.80)). Comorbidities were underreported in NORspine. Perioperative details had strong to excellent agreements (kappa (95% CI) ranging from 0.76 ( 0.68–0.84) to 0.98 (0.95–1.00)), PCCs between 93% and 99% and sensitivities (95% CI) between 92% (0.84–1.00%) and 99% (0.98–1.00%). Patient-reported variables (height, weight, smoking) had excellent agreements (kappa (95% CI) between 0.93 (0.89–0.97) and 0.99 (0.98–0.99)). Conclusion Compared to electronic patient records, NORspine displayed weak agreement for perioperative complications, moderate agreement for ASA classification, strong agreement for perioperative details, and excellent agreement for height, weight, and smoking. NORspine underreported perioperative complications and comorbidities when compared to EPRs. Patient-recorded data were more accurate and should be preferred when available.

The impact of medication reconciliation and review in patients using oral chemotherapy

Background and aim Verifying and reviewing a patients medication list can detect and reduce drug related problems (DRPs). However little is known about its effects in patients using oral chemotherapy. The aim of this study was to evaluate the impact of these interventions and the adapted Medication Appropriateness Index (aMAI) as a tool to carry out a medication review. Methods A case-control study was carried out. The hospital pharmacist performed a medication reconciliation and medication review, using the aMAI tool, in 54 patients starting oral chemotherapy. Discrepancies, DRP's and associated pharmaceutical interventions were reported via the electronic patient record (EPR). After one month, the acceptance rate was measured and the aMAI score recalculated. Kappa statistics were used to test intra- and interrater reliability. Results The medication list in the EPR was incomplete in 74,1% of patients with an average of 2.4 errors per patient. After medication review, the aMAI score decreased significantly from 7.2 to 5.4 (SD = 4,7; p <0.001), indicating an improvement in the appropriateness of the drugs patients were taking. Acceptance rates were 41,4% and 53,2% for advices resulting from medication reconciliation and medication review respectively. Kappa values of 0.90 and 0.70 respectively indicate good intra- and interrater reliability. Discussion and conclusion The study shows that medication reconciliation can identify and address discrepancies. Furthermore, medication review seems to ensure that drug treatment better meets patient needs. The aMAI was a reliable tool. Future research will have to determine the clinical relevance of these interventions.

Wetland Change Mapping Using Machine Learning Algorithms, and Their Link with Climate Variation and Economic Growth: A Case Study of Guangling County, China

Wetlands are a distinctive terrestrial ecosystem that benefits living things, including people, in various ways. Sustainable wetland ecosystem resources are needed to protect the global environment. Wetlands in China have undergone positive and negative changes in response to several factors, but studies documenting their long-term dynamicity have been few, particularly in Guangling County. This study examines the change of wetlands area based on remotely sensed data while exploring trends associated with climate variations and economic growth in Guangling County, China. Analysis of remotely sensed imagery, mainly in hilly and nonhomogeneous environments is problematic, largely as a result of interference and their high spectral non-homogeneity. We conducted experiments using five classical machine learning algorithms based on the Google Earth Engine (GEE) and obtained the greatest robustness and accuracy using a Support Vector Machine (SVM)—Radial Basis Function (RBF) kernel approach, with overall accuracy and kappa statistics ranging from 86% to 98.1% and from 0.789 to 0.960, respectively. Based on the SVM-RBF model’s outperformance of four other algorithms, we identified spatial distributions of wetland in the study area and associated change trends. We found that 45.71 km2 of wetland area was lost over the past 3.7 decades (January 1984–December 2020), or 81.82% of wetland area coverage. In this paper, we explore how factors such as county economic growth (GDP), humidity, and temperature variations are tightly linked with wetland change.

QUERY EXPANSION RANKING PADA ANALISIS SENTIMEN MENGGUNAKAN KLASIFIKASI MULTINOMIAL NAÏVE BAYES (Studi Kasus : Ulasan Aplikasi Shopee pada Hari Belanja Online Nasional 2020)

Shopee is one of the e-commerce sites that has many users in Indonesia. Shopee provides various attractive promos on special days such as National Online Shopping Day on December 12. Shopee site was a complete error on December 12, 2020. Complaints and opinions of Shopee users were also shared through various media, one of them was Google Play Store. Sentiment analysis was used to see the user's response to the Shopee’s incident. Sentiment analysis results can be extracted to obtain information regarding positive or negative reviews from Shopee users. Sentiment analysis was performed using the Multinomial Naïve Bayes classification. the simplest method of probability classification, but it is sensitive to feature selection so that the amount of data is determined by the results of feature selection Query Expansion Ranking. The algorithm that has the highest accuracy and kappa statistic is the best algorithm in classifying Shopee’s users sentiment. The results showed that the classification performance using Multinomial Naïve Bayes with 80% of the features (terms) which have the highest Query Expansion Ranking value was obtained at the accuracy and kappa statistics values are 89% and 77.62%. This means that Multinomial Nave Bayes has a good performance in classifying reviews and the number of features used affects the performance results obtained.

Quantification of Mucosal Activity from Colonoscopy Reports via the Simplified Endoscopic Mucosal Assessment for Crohn’s Disease

Background Endoscopic mucosal healing is the gold standard for evaluating Crohn’s disease (CD) treatment efficacy. Standard endoscopic indices are not routinely used in clinical practice, limiting the quality of retrospective research. A method for retrospectively quantifying mucosal activity from documentation is needed. We evaluated the simplified endoscopic mucosal assessment for CD (SEMA-CD) to determine if it can accurately quantify mucosal severity recorded in colonoscopy reports. Methods Pediatric patients with CD underwent colonoscopy that was video recorded and evaluated via Simple Endoscopic Score for CD (SES-CD) and SEMA-CD by central readers. Corresponding colonoscopy reports were de-identified. Central readers blinded to clinical history and video scoring were randomly assigned colonoscopy reports with and without images. The SEMA-CD was scored for each report. Correlation with video SES-CD and SEMA-CD were assessed with Spearman rho, inter-rater, and intrarater reliability with kappa statistics. Results Fifty-seven colonoscopy reports were read a total of 347 times. The simplified endoscopic mucosal assessment for CD without images correlated with both SES-CD and SEMA-CD from videos (rho = 0.82, P &lt; .0001 for each). The addition of images provided similar correlation. Inter-rater and intrarater reliability were 0.93 and 0.92, respectively. Conclusions The SEMA-CD applied to retrospective evaluation of colonoscopy reports accurately and reproducibly correlates with SES-CD and SEMA-CD of colonoscopy videos. The SEMA-CD for evaluating colonoscopy reports will enable quantifying mucosal healing in retrospective research. Having objective outcome data will enable higher-quality research to be conducted across multicenter collaboratives and in clinical registries. External validation is needed.

A Food Frequency Questionnaire for Hemodialysis Patients in Bangladesh (BDHD-FFQ): Development and Validation

Diet is a recognized risk factor and cornerstone for chronic kidney disease (CKD) management; however, a tool to assess dietary intake among Bangladeshi dialysis patients is scarce. This study aims to validate a prototype Bangladeshi Hemodialysis Food Frequency Questionnaire (BDHD-FFQ) against 3-day dietary recall (3DDR) and corresponding serum biomarkers. Nutrients of interest were energy, macronutrients, potassium, phosphate, iron, sodium and calcium. The BDHD-FFQ, comprising 132 food items, was developed from 606 24-h recalls and had undergone face and content validation. Comprehensive facets of relative validity were ascertained using six statistical tests (correlation coefficient, percent difference, paired t-test, cross-quartiles classification, weighted kappa, and Bland-Altman analysis). Overall, the BDHD-FFQ showed acceptable to good correlations (p < 0.05) with 3DDR for the concerned nutrients in unadjusted and energy-adjusted models, but this correlation was diminished when adjusted for other covariates (age, gender, and BMI). Phosphate and potassium intake, estimated by the BDHD-FFQ, also correlated well with the corresponding serum biomarkers (p < 0.01) when compared to 3DDR (p > 0.05). Cross-quartile classification indicated that <10% of patients were incorrectly classified. Weighted kappa statistics showed agreement with all but iron. Bland-Altman analysis showed positive mean differences were observed for all nutrients when compared to 3DDR, whilst energy, carbohydrates, fat, iron, sodium, and potassium had percentage data points within the limit of agreement (mean ± 1.96 SD), above 95%. In summary, the BDHD-FFQ demonstrated an acceptable relative validity for most of the nutrients as four out of the six statistical tests fulfilled the cut-off standard in assessing dietary intake of CKD patients in Bangladesh.

Assessing the reliability of automatic sentiment analysis tools on rating the sentiment of reviews of NHS dental practices in England

Background Online reviews may act as a rich source of data to assess the quality of dental practices. Assessing the content and sentiment of reviews on a large scale is time consuming and expensive. Automation of the process of assigning sentiment to big data samples of reviews may allow for reviews to be used as Patient Reported Experience Measures for primary care dentistry. Aim To assess the reliability of three different online sentiment analysis tools (Amazon Comprehend DetectSentiment API (ACDAPI), Google and Monkeylearn) at assessing the sentiment of reviews of dental practices working on National Health Service contracts in the United Kingdom. Methods A Python 3 script was used to mine 15800 reviews from 4803 unique dental practices on the NHS.uk websites between April 2018 –March 2019. A random sample of 270 reviews were rated by the three sentiment analysis tools. These reviews were rated by 3 blinded independent human reviewers and a pooled sentiment score was assigned. Kappa statistics and polychoric evalutaiton were used to assess the level of agreement. Disagreements between the automated and human reviewers were qualitatively assessed. Results There was good agreement between the sentiment assigned to reviews by the human reviews and ACDAPI (k = 0.660). The Google (k = 0.706) and Monkeylearn (k = 0.728) showed slightly better agreement at the expense of usability on a massive dataset. There were 33 disagreements in rating between ACDAPI and human reviewers, of which n = 16 were due to syntax errors, n = 10 were due to misappropriation of the strength of conflicting emotions and n = 7 were due to a lack of overtly emotive language in the text. Conclusions There is good agreement between the sentiment of an online review assigned by a group of humans and by cloud-based sentiment analysis. This may allow the use of automated sentiment analysis for quality assessment of dental service provision in the NHS.

Adherence to Antiseizure vs Other Medications Among US Medicare Beneficiaries With and Without Epilepsy

Objective:To 1) compare adherence to antiseizure medications (ASMs) versus non-ASMs among individuals with epilepsy, 2) assess the degree to which variation in adherence is due to differences between individuals versus between medication classes among individuals with epilepsy, and 3) compare adherence in individuals with versus without epilepsy.Methods:This was a retrospective cohort study using Medicare. We included beneficiaries with epilepsy (≥1 ASM, plus International Classification of Diseases, Ninth Revision, Clinical Modification diagnostic codes), and a 20% random sample without epilepsy. Adherence for each medication class was measured by the proportion of days covered (PDC) in 2013-2015. We used Spearman correlation coefficients, Cohen’s kappa statistics, and multilevel logistic regressions.Results:There were 83,819 beneficiaries with epilepsy. Spearman correlation coefficients between ASM PDCs and each of the 5 non-ASM PDCs ranged 0.44-0.50, Cohen’s kappa ranged 0.33-0.38, and within-person differences between each ASM’s PDC minus each non-ASM’s PDC were all statistically significant (p<0.01) though median differences were all very close to 0. Fifty-four percent of variation in adherence across medications was due to differences between individuals. Adjusted predicted probabilities of adherence were: ASMs 74% (95% confidence interval [CI] 73%-74%), proton pump inhibitors 74% (95% CI 74%-74%), antihypertensives 77% (95% CI 77%-78%), selective serotonin reuptake inhibitors 77% (95% CI 77%-78%), statins 78% (95% CI 78%-79%), and levothyroxine 82% (95% CI 81%-82%). Adjusted predicted probabilities of adherence to non-ASMs were 80% (95% CI 80%-81%) for beneficiaries with epilepsy versus 77% (77%-77%) for beneficiaries without epilepsy.Conclusion:Among individuals with epilepsy, ASM and non-ASM adherence were moderately correlated, half of variation in adherence was due to between-person rather than between-medication differences, adjusted adherence was slightly lower for ASMs than several non-ASMs, and epilepsy was associated with a quite small increase in adherence to non-ASMs. Nonadherence to ASMs may provide an important cue to the clinician to inquire about adherence to other potentially life-prolonging medications as well. Although efforts should focus on improving ASM adherence, patient-level rather than purely medication-specific behaviors are also critical to consider when developing interventions to optimize adherence.