Spectral clustering identifies patterns of chiropractic care in a national longitudinal cohort.

<h3>Context:</h3> The COVID-19 pandemic has catalyzed the use of video visits in primary care. It is estimated that 73% of primary care visits can be effectively completed via video. However, there are no studies that demonstrate clinical effectiveness of video visits for specific chief complaints. <h3>Objective:</h3> To evaluate the clinical effectiveness of video visits compared to in-person visits for 2 common primary care chief complaints. <h3>Study design:</h3> Retrospective chart review. <h3>Dataset:</h3> Manual chart review of in-person visits from August-October 2019 and video visits from August-October 2020 from our institution’s outpatient urgent care clinic (Stanford Express Care), restricted to 2 of the most common presenting chief complaints (CCs): low back pain and headache. <h3>Population studied:</h3> Patients who presented to a Stanford Express Care clinic with one of the aforementioned CCs. <h3>Outcome measures:</h3> Frequency of clinician recommendation for an urgent office or ED visit after the initial patient visit and frequency of follow-up visits within a 3-week period were used to assess clinical effectiveness of the visit. A visit is considered clinically effective when a clinician does not recommend an urgent office or ED visit after the initial patient visit and the patient does not have in-person follow-up visits within 3 weeks of the initial visit. Frequency of referrals placed and diagnostic imaging studies ordered during the initial patient visit were also measured. <h3>Results:</h3> Video visits for low back pain were less likely to be effectively assessed compared to in-person visits [74% (37/50) vs 82% (54/66), chi-square p=0.3]. During video visits for low back pain clinicians placed fewer referrals [24% (12/50) vs 36% (24/66), chi-square p=0.2] and ordered fewer diagnostic imaging studies [12% (6/50) vs 21% (14/66), chi-square p=0.2]. Video visits for headache were more likely to be effectively assessed compared to in-person visits [86% (43/50) vs 74% (37/50), chi-square p=0.1]. During video visits for headache, clinicians placed fewer referrals [14% (7/50) vs 22% (11/50), (chi-square p=0.3 ) and ordered fewer diagnostic imaging studies [2% (1/50) vs 18% (9/50), chi-square p=0.007]. <h3>Conclusions:</h3> For low back pain and headache, video visits were not significantly less likely than in-person visits to be effective. There was a statistically significant decrease in diagnostic imaging studies ordered during video visits for headaches.

To describe the frequency of primary care seeking for neck or back-related conditions among people with chronic neck and low back pain and to develop prediction models of primary care seeking and frequent visits. We included participants of the Trøndelag Health Study (HUNT4, 2017-19) in Norway who self-reported chronic neck and/or low back pain in the preceding year, and extracted data of primary care visits from the Norwegian primary healthcare registry. We investigated a total of 23 potential predictors and used multivariable logistic regression models to predict primary care seeking for neck or back-related conditions and frequent visits by healthcare provider (i.e., the highest quartile of number of visits). Among the 15,352 HUNT4 participants with chronic neck and/or low back pain, 6231 participants (40.6%) sought primary care for neck or back-related conditions (median = 5 visits, IQR 2-15) within 2 years after the study. Participants who consulted physical therapists sought care the most frequently (median = 10 visits, IQR 3-26). Discrimination of the best-fit prediction model of primary care seeking and frequent visits by healthcare provider, assessed by C-statistic, ranged from 0.66-0.76. Participants who made frequent primary care visits in the preceding year were highly likely to continue frequent care seeking in the following 2 years. Around 40% of people seek primary care for chronic neck and low back pain, and frequent care seeking may continue for years. Future studies should investigate strategies to reduce recurrent primary care visits, especially seeking physical therapist care, and promote self-management of chronic pain. People with chronic neck and low back pain who seek physical therapist care had the highest frequency of care seeking, underscoring the significant burden on healthcare systems. The high frequency of visits and associated healthcare expenditures highlight the critical need for effective and valuable primary care for chronic pain management. To mitigate recurrent visits and reduce costs, it is essential to provide patients with evidence-based treatments and self-management interventions.

Letter to the Editor-in-Chief of JOSPT as follows: "Promoting Roles and Services Within Physical Therapy Not Supported by Evidence" with Authors' Response J Orthop Sports Phys Ther 2018;48(8):669-671. doi:10.2519/jospt.2018.0204.

This study is an analysis of national survey data from 5 sample years. The authors characterized the frequency of office visits for low back pain, the content of ambulatory care, and how these vary by physician specialty. Few recent data are available regarding ambulatory care for low back pain or how case mix and patient management vary by physician specialty. Data from the National Ambulatory Medical Care Survey were grouped into three time periods (1980-81, 1985, 1989-90). Frequency of visits for low back pain, referral status, tests, and treatments were tabulated by physician specialty. There were almost 15 million office visits for "mechanical" low back pain in 1990, ranking this problem fifth as a reason for all physician visits. Low back pain accounted for 2.8 percent of office visits in all three time periods. Nonspecific diagnostic labels were most common, and 56 percent of visits were to primary care physicians. Specialty variations were observed in caseload, diagnostic mix, and management. Back pain remains a major reason for all physician office visits. This study describes visit, referral, and management patterns among specialties providing the most care.

<h3>Context:</h3> Tapering of long-term opioid therapy (LTOT) increased after publication of the 2016 CDC opioid guidelines, followed by anecdotal reports of exacerbated pain among tapered patients. However, two systematic reviews of limited evidence from multidisciplinary pain control programs found similar or better pain ratings after tapering versus baseline. <h3>Objective:</h3> To evaluate the association between opioid dose tapering and subsequent emergency (ED) visits, outpatient primary care visits and hospitalizations for pain among patients prescribed LTOT. <h3>Study Design:</h3> Retrospective cohort study. <h3>Dataset:</h3> 2015-2019 de-identified administrative data from the Optum Labs Data Warehouse, including medical and pharmacy claims and eligibility information for commercial and Medicare Advantage enrollees, representing a mixture of ages and geographical regions. <h3>Population:</h3> Adults ≥18 years old who were prescribed stable doses of LTOT ≥50 morphine milligram equivalents per day during a 12-month baseline period <h3>Outcome Measures:</h3> Monthly counts of ED visits, primary care visits, and hospitalizations for pain up to 12 months after cohort entry. Pain visits were defined by diagnostic codes for musculoskeletal or other specific chronic pain in the primary position on ED and hospitalization claims, or in any position for primary care visit claims. <h3>Analysis:</h3> Monthly counts were modeled using negative binomial regression as a function of tapering (≥15% relative dose reduction during 6 overlapping 60-day periods after cohort entry), total baseline pain and non-pain ED visits, primary care visits and hospitalizations, and adjusted for patient level-covariates (sociodemographics, comorbidities). <h3>Results:</h3> Among 51,361 patients prescribed stable LTOT, 14,331 patients (27.9%) tapered after cohort entry. Tapering was associated with more subsequent ED visits (adjusted incidence rate ratio [aIRR] 1.18, 95% CI: 1.11-1.27) and fewer subsequent primary care visits (aIRR 0.95, CI: 0.92-0.99) for pain. Hospitalizations for pain control were unchanged (aIRR 1.04, CI: 0.95-1.15). <h3>Conclusions:</h3> Among patients prescribed LTOT, opioid tapering was associated with subsequently more ED visits yet fewer primary care visits for pain, suggesting a shift in pain care from outpatient to the higher acuity emergency setting post-taper. The findings suggest tapering may have led to increased pain, disruption of primary care relationships, or some combination of these effects.

Usual Care for Low Back Pain at United States Emergency Departments, 2016-2022.

Predicting 7-day and 3-month functional outcomes after an ED visit for acute nontraumatic low back pain

Tapering of chronic opioids has increased, with subsequent reports of exacerbated pain among patients who tapered. We aimed to evaluate the association between opioid dose tapering and subsequent pain-related healthcare utilization (emergency department [ED] visits, hospitalizations and primary care visits). We conducted a retrospective cohort study from years 2015-2019 using data from the Optum Labs Data Warehouse that contains de-identified retrospective administrative claims data for commercial and Medicare Advantage enrollees in the United States. Adults aged ≥18 years who were prescribed stable doses of opioids, ≥50 morphine milligram equivalents (MME)/day, during a 12-month baseline period. Tapering was defined as ≥15% relative reduction in mean daily opioid dose during one of 6 overlapping 60-day periods. Tapered patient-periods were subclassified as tapered-and-continued (MME > 0) vs tapered-and-discontinued (MME = 0). We modeled monthly counts of visits for pain diagnoses up to 12 months after cohort entry using negative binomial regression as a function of tapering, baseline utilization, and patient level-covariates. Among 47 033 patients, 13 793 patients tapered. Compared to no taper, any taper was associated with more ED visits for pain (adjusted incidence rate ratio [aIRR] 1.21, 95% confidence interval [CI]: 1.11-1.30), tapered then continued status was associated with more ED visits (aIRR 1.23, CI: 1.14-1.32) and hospitalizations (aIRR 1.14, CI: 1.03-1.27) f-or pain, and tapered-and-discontinued was associated with fewer primary care visits for pain (aIRR 0.68, CI: 0.61-0.76). These associations suggest that opioid tapering may lead to increased emergency and hospital utilization for acute pain and possibly a decreased perceived need for primary care for those whose opioids were discontinued.

BackgroundHigh dimensionality in electronic health records (EHR) causes a significant computational problem for any systematic search for predictive, diagnostic, or prognostic patterns. Feature selection (FS) methods have been indicated to be effective in feature reduction as well as in identifying risk factors related to prediction of clinical disorders. This paper examines the prediction of patients with alcohol use disorder (AUD) using machine learning (ML) and attempts to identify risk factors related to the diagnosis of AUD.MethodsA FS framework consisting of two operational levels, base selectors and ensemble selectors. The first level consists of five FS methods: three filter methods, one wrapper method, and one embedded method. Base selector outputs are aggregated to develop four ensemble FS methods. The outputs of FS method were then fed into three ML algorithms: support vector machine (SVM), K-nearest neighbor (KNN), and random forest (RF) to compare and identify the best feature subset for the prediction of AUD from EHRs.ResultsIn terms of feature reduction, the embedded FS method could significantly reduce the number of features from 361 to 131. In terms of classification performance, RF based on 272 features selected by our proposed ensemble method (Union FS) with the highest accuracy in predicting patients with AUD, 96%, outperformed all other models in terms of AUROC, AUPRC, Precision, Recall, and F1-Score. Considering the limitations of embedded and wrapper methods, the best overall performance was achieved by our proposed Union Filter FS, which reduced the number of features to 223 and improved Precision, Recall, and F1-Score in RF from 0.77, 0.65, and 0.71 to 0.87, 0.81, and 0.84, respectively. Our findings indicate that, besides gender, age, and length of stay at the hospital, diagnosis related to digestive organs, bones, muscles and connective tissue, and the nervous systems are important clinical factors related to the prediction of patients with AUD.ConclusionOur proposed FS method could improve the classification performance significantly. It could identify clinical factors related to prediction of AUD from EHRs, thereby effectively helping clinical staff to identify and treat AUD patients and improving medical knowledge of the AUD condition. Moreover, the diversity of features among female and male patients as well as gender disparity were investigated using FS methods and ML techniques.

We examined fluctuations in emergency department (ED) visits for nonspecific back and neck pain during the COVID-19 pandemic and explored potential contributing factors. This retrospective cohort study included patients who presented to the ED with nonspecific back and neck pain between January 2019 and December 2021. Demographic data, visit frequencies, and clinical outcomes were analyzed to assess the impact of the pandemic on visit patterns. A total of 1245 ED visits were recorded. Visits decreased by 30% during the peak of the pandemic, with a gradual return to baseline by mid-2021. No significant changes in patient demographics or clinical outcomes were noted during the pandemic. However, a proportional increase in neck pain visits was observed. The observed decline in visits may be linked to pandemic-related concerns, such as fear of exposure in the hospital and reduced activities. The increased neck pain visits highlights the potential influence of pandemic-related stress and lifestyle changes. Visit patterns rebounded as the pandemic eased, indicating a temporary decrease unrelated to condition severity. The COVID-19 pandemic temporarily affected ED visits for nonspecific back and neck pain. Further research is needed to explore the long-term effects of the pandemic on health care utilization.

This paper delves into the critical need for enhanced security measures within the Internet of Things (IoT) landscape due to inherent vulnerabilities in IoT devices, rendering them susceptible to various forms of cyber-attacks. The study emphasizes the importance of Intrusion Detection Systems (IDS) for continuous threat monitoring. The objective of this study was to conduct a comprehensive evaluation of feature selection (FS) methods using various machine learning (ML) techniques for classifying traffic flows within datasets containing intrusions in IoT environments. An extensive benchmark analysis of ML techniques and FS methods was performed, assessing feature selection under different approaches including Filter Feature Ranking (FFR), Filter-Feature Subset Selection (FSS), and Wrapper-based Feature Selection (WFS). FS becomes pivotal in handling vast IoT data by reducing irrelevant attributes, addressing the curse of dimensionality, enhancing model interpretability, and optimizing resources in devices with limited capacity. Key findings indicate the outperformance for traffic flows classification of certain tree-based algorithms, such as J48 or PART, against other machine learning techniques (naive Bayes, multi-layer perceptron, logistic, adaptive boosting or k-Nearest Neighbors), showcasing a good balance between performance and execution time. FS methods' advantages and drawbacks are discussed, highlighting the main differences in results obtained among different FS approaches. Filter-feature Subset Selection (FSS) approaches such as CFS could be more suitable than Filter Feature Ranking (FFR), which may select correlated attributes, or than Wrapper-based Feature Selection (WFS) methods, which may tailor attribute subsets for specific ML techniques and have lengthy execution times. In any case, reducing attributes via FS has allowed optimization of classification without compromising accuracy. In this study, F1 score classification results above 0.99, along with a reduction of over 60% in the number of attributes, have been achieved in most experiments conducted across four datasets, both in binary and multiclass modes. This work emphasizes the importance of a balanced attribute selection process, taking into account threat detection capabilities and computational complexity.

We wished to determine whether patient satisfaction was related to physicians' confidence in their abilities to effectively manage low back pain, and to examine their attitudes about patients with back pain. The confidence and attitudes of primary care providers were determined using self-administered questionnaires. Patient satisfaction with care was assessed during telephone interviews conducted 3 weeks after a clinic visit for low back pain. The study was conducted in a primary care clinic of a large health maintenance organization. Completed surveys were obtained from 21 primary care providers (18 physicians and three physician assistants) and 270 of their patients with low back pain. Three satisfaction scales specific to low back pain were used to measure patient satisfaction with regard to information received from provider, caring, and effectiveness of treatment. The results showed that the providers' attitudes about patients with low back pain were not associated with any of the patient satisfaction measures. However, patients of more confident providers were significantly more satisfied with the information they received than were patients of less confident providers. These differences could not be explained by years in practice, length of visit, patient demographics, or the severity and duration of low back pain. These findings suggest that providers who have more confidence in their abilities to effectively manage low back pain may in fact be more effective patient educators.

Heart disease is the abnormal health condition that influences parts of the heart and all its parts. World Health Organization (WHO) is assured that the disease is one of the leading killer disease of the worldwide population. The prevalence of the disease is also increasing through developing countries like Ethiopia. Machine Learning (ML) is one of the key technique in the management and processing of a huge number of health data’s and it supports in diagnosis and prediction of disease at early stages. The main objective of this study is developing an early detection of Heart Disease (HD) enhancing prediction through ML technique; such as Random forest (RF), K Nearest Neighbor (KNN), Support vector Machine (SVM), Gradient Boosting (GB) and Voting Classifier with two Feature Selection (FS) methods, of Chi-Square (CFS) and Sequential Forward Feature Selection (SFFS) methods. The data used for the experimentation purpose was collected from Local Hospitals. Before FS methods are performed, all the ML algorithms are applied for the imbalanced and balanced HD dataset. Then after, the two FS methods are applied with ML techniques on these imbalanced and balanced datasets. Models are evaluated through different model evaluation metrics with two data splitting technique namely Percentage Splitting (PS) and 10-Fold-Cross Validation (10-F-CV) techniques and finally different results are registered. Thus, before FS methods are applied on the full balanced datasets, SVM and GB achieved a good accuracy score of 99.2% using PS and similarly after FS technique is applied, Both RF with CFS and VC with CFS achieved a better accuracy score of 99.4% using PS for the combined dataset, so this will helps users and experts to detect and appropriate prevention of the disease at an early stage.

To determine whether emotional distress reported at the initial consultation affects subsequent healthcare use either directly or indirectly via moderating the influence of symptoms. Longitudinal observational study of 2891 participants consulting primary care for low back pain. Negative binomial regression models were constructed to estimate independent effects of emotional distress on healthcare use.Potential confounders were identified using directed acyclic graphs. After the initial consultation, participants had a mean (SD) of one (1.2) visit for back pain over 3months, and nine (14) visits for back pain over 12months. Higher reports of anxiety during the initial consultation led to increased short-term healthcare use (IRR 1.06, 95% CI 1.01-1.11) and higher reports of depression led to increased long-term healthcare use (IRR 1.04, 95% CI 1.02-1.07). The effect sizes suggest that a patient with a high anxiety score (8/10) would consult 50% more frequently over 3months, and a person with a high depression score (8/10) would consult 30% more frequently over 12months, compared to a patient with equivalent pain and disability and no reported anxiety or depression. Emotional distress in the acute stage of low back pain increased subsequent consultation rates. Interventions that target emotional distress during the initial consultation are likely to reduce costly and potentially inappropriate future healthcare use for patients with non-specific low back pain.

Leaf area index (LAI) is an essential vegetation parameter that represents the light energy utilization and vegetation canopy structure. As the only in-operation hyperspectral satellite launched by China, GF-5 is potentially useful for accurate LAI estimation. However, there is no research focus on evaluating GF-5 data for LAI estimation. Hyperspectral remote sensing data contains abundant information about the reflective characteristics of vegetation canopies, but these abound data also easily result in a dimensionality curse. Therefore, feature selection (FS) is necessary to reduce data redundancy to achieve more reliable estimations. Currently, machine learning (ML) algorithms have been widely used for FS. Moreover, the same ML algorithm is usually conducted for both FS and regression in LAI estimation. However, no evidence suggests that this is the optimal solution. Therefore, this study focuses on evaluating the capacity of GF-5 spectral reflectance for estimating LAI and the performances of different combination of FS and ML algorithms. Firstly, the PROSAIL model, which coupled leaf optical properties model PROSPECT and the scattering by arbitrarily inclined leaves (SAIL) model, was used to generate simulated GF-5 reflectance data under different vegetation and soil conditions, and then three FS methods, including random forest (RF), K-means clustering (K-means) and mean impact value (MIV), and three ML algorithms, including random forest regression (RFR), back propagation neural network (BPNN) and K-nearest neighbor (KNN) were used to develop nine LAI estimation models. The FS process was conducted twice using different strategies: Firstly, three FS methods were conducted to search the lowest dimension number, which maintained the estimation accuracy of all bands. Then, the sequential backward selection (SBS) method was used to eliminate the bands having minimal impact on LAI estimation accuracy. Finally, three best estimation models were selected and evaluated using reference LAI. The results showed that although the RF_RFR model (RF used for feature selection and RFR used for regression) achieved reliable LAI estimates (coefficient of determination (R2) = 0.828, root mean square error (RMSE) = 0.839), the poor performance (R2 = 0.763, RMSE = 0.987) of the MIV_BPNN model (MIV used for feature selection and BPNN used for regression) suggested using feature selection and regression conducted by the same ML algorithm could not always ensure an optimal estimation. Moreover, RF selection preserved the most informative bands for LAI estimation so that each ML regression method could achieve satisfactory estimation results. Finally, the results indicated that the RF_KNN model (RF used as feature selection and KNN used for regression) with seven GF-5 spectral band reflectance achieved the better estimation results than others when validated by simulated data (R2 = 0.834, RMSE = 0.824) and actual reference LAI (R2 = 0.659, RMSE = 0.697).