Classification And Regression Tree Analysis Research Articles

Quadriceps strength symmetry predicts vertical ground reaction force symmetry during running in patients who have undergone ACL reconstruction

ObjectiveTo determine whether quadriceps strength symmetry can predict peak vertical ground reaction force (vGRF) running force symmetry in patients who have undergone ACL reconstruction (ACLR). We also sought to determine a cutoff for quadriceps strength symmetry to identify patients at risk for vGRF running asymmetry. DesignRetrospective cross-sectional. SettingClinical facility. MethodsBilateral quadriceps strength and vGRF data during running were obtained from 79 patients 26–30 weeks post ACLR. Linear regression was used to determine if quadriceps strength symmetry predicted peak vGRF running force symmetry. Classification and regression tree (CART) analysis was used to determine the cutoff value for quadriceps strength symmetry to identify patients at risk for vGRF running asymmetry. ResultsIncreased quadriceps strength symmetry predicted increased vGRF running symmetry (R2 = 0.20). CART analysis revealed that patients with quadriceps strength symmetry less than or equal to 88% were at highest risk for vGRF running asymmetry (R2 = 26%). ConclusionGreater quadriceps strength symmetry is predictive of greater vGRF running force symmetry in patients who have undergone ACLR. This finding highlights the need for clinicians to consider the degree of quadriceps strength symmetry before initiating a return to running program.

The prevalence of skin lesions and associated factors in hospitalised adult patients with cancer.

To assess the prevalence of skin lesions and evaluate the clinical and sociodemographic factors associated with their presence in hospitalised patients. This descriptive, cross-sectional, correlational study was performed in inpatient units and intensive care units of a cancer hospital in São Paulo, Brazil, after approval by the Institutional Research Ethics Committee. Data from hospitalised adult patients with cancer were collected during physical examinations and from medical records. A Chi-squared test, univariate analysis, a logistic regression model with results expressed as odds ratio (OR) and 95% confidence intervals (CI), and Classification and Regression Tree (CART) analysis were used to evaluate the data. Of 341 patients, 80 had skin lesions, equating to an overall prevalence of 23.5%. The skin lesions included pressure injuries (10%), incontinence-associated dermatitis (6.7%), skin tears (6.5%), malignant wounds (3.8%) and complicated surgical wounds (3.2%). The factors associated with skin lesions in cancer patients were the use of disposable nappies (OR: 4.436) and age (59.1±15.1 years), according to the CART analysis, and the wearing of nappies (OR: 4.466, p<0.001), presence of ecchymosis (OR: 2.532, p<0.001) and infection (OR: 6.449, p=0.040), according to multiple regression analysis. This study contributed to knowledge about prevalence and associated factors of skin lesions in hospitalised patients with cancer, allowing the implementation of preventive measures.

Vitamin d deficiency and metabolic syndrome: The joint effect on cardiovascular and all-cause mortality in the United States adults.

BACKGROUNDThe long-term impact of vitamin D deficiency and metabolic syndrome (MetS) on cardiovascular disease (CVD) and all-cause mortality are still a matter of debate. AIMTo test the hypotheses that lower serum 25 hydroxyvitamin D [25(OH)D] concentrations (a marker of vitamin D level) and MetS have a long-term impact on the risk of CVD and all-cause mortality, and individuals with vitamin D deficiency can be identified by multiple factors. METHODSA sample of 9094 adults, 20 to 90 years of age, who participated in the Third National Health and Nutrition Examination Survey (NHANES III, 1988 to 1994) were followed through December 2015 was analyzed. The associations of serum 25(OH)D concentrations and MetS with CVD and all-cause mortality were analyzed longitudinally using Cox regression models. Classification and regression tree (CART) for machine learning was applied to classify individuals with vitamin D deficiency. RESULTSOf 9094 participants, 30% had serum 25(OH)D concentrations < 20 ng/mL (defined as vitamin D deficiency), 39% had serum 25(OH)D concentrations between 20 to 29 ng/mL (insufficiency), and 31% had serum 25(OH)D concentrations ≥30 ng/mL (sufficiency). Prevalence of MetS was 28.4%. During a mean of 18 years follow-up, vitamin D deficiency and MetS were significantly associated with increased risk of CVD and all-cause mortality. Subjects with both vitamin D deficiency and MetS had the highest risk of CVD mortality (HR = 1.77, 95%CI: 1.22-2.58) and all-cause mortality (HR = 1.62, 95%CI: 1.26-2.09), followed by those with both vitamin D insufficiency and MetS for CVD mortality (HR = 1.59, 95%CI: 1.12-2.24), and all-cause mortality (HR = 1.41, 95%CI: 1.08-1.85). Meanwhile, vitamin D sufficiency significantly decreased the risk of CVD and all-cause mortality for those who even had MetS. Among the total study sample, CART analysis suggests that being non-Hispanic Black, having lower serum folate level, and being female were the first three predictors for those with serum 25(OH)D deficiency. CONCLUSIONVitamin D deficiency and MetS were significantly associated with increased risk of CVD and all-cause mortality. There was a significant joint effect of vitamin D deficiency and MetS on the risk of mortality. Findings of the CART analysis may be useful to identify individuals positioned to benefit from interventions to reduce the risk of CVD and all-cause mortality.

Differentiation of multiple myelomas from osteolytic bone metastases: Diagnostic value of tumor homogeneity on Contrast-Enhanced CT.

To investigate the diagnostic value of tumor homogeneity on contrast-enhanced (CE) computed tomography (CT) to differentiate multiple myeloma (MM) from osteolytic bone metastases (Mets). This retrospective study included patients who were diagnosed with MM or Mets and had multiple (≥2) osteolytic bone tumors on pre-treatment CE-CT. Intratumoral homogeneity was assessed by coefficient of variation (CV, ratio of standard deviation to mean) of the density of a single lesion (CV-lesion). Intertumoral homogeneity was assessed as the CV of the densities of multiple lesions in one patient (CV-patient). A classification model was built from CT parameters using classification and regression tree (CART) analysis. Diagnostic performance of the model was evaluated using C-statistics. A total of 272 lesions (81 MM and 191 Mets) of 105 patients were analyzed. The mean CV-lesion and CV-patient of MM were significantly lower than those of Mets: 0.17 vs 0.26 for CV-lesion (p = 0.005) and 0.16 vs 0.23 for CV-patient (p = 0.013). Thickened struts were more common in MM than in Mets (49.1% vs 12.8%, p ≤ 0.001). In CART analysis, CV-lesion was the first partitioning predictor, followed by thickened struts and by CV patient. The CART model could distinguish MM from Mets in both the model development cohort (C-statistic: 0.843) and the temporal validation cohort (0.721, 0.686, and 0.686 for three reviewers, respectively). MM showed intratumoral and intertumoral homogeneity compared with Mets on CE-CT. The combination of CV-lesion and CV-patient can be helpful to radiologists in differentiation of MM from Mets. Our study showed that MM had intratumoral and intertumoral homogeneity compared with Mets on contrast-enhanced CT.

Using a decision tree approach to determine hearing aid ownership in older adults

Purpose The main clinical intervention for older adults with hearing loss is the provision of hearing aids. However, uptake and usage in this population have historically been reported as low. The aim of this study was to understand the hearing loss characteristics, from measured audiometric hearing loss and self-perceived hearing handicap, that contribute to the decision of hearing aid ownership. Materials and methods A total of 2833 adults aged 50+ years, of which 329 reported hearing aid ownership, were involved with a population-based survey with audiometric hearing assessments. Classification and regression tree (CART) analysis was used to classify hearing aid ownership from audiometric measurements and hearing disability outcomes. Results An overall accuracy of 92.5% was found for the performance of the CART analysis in predicting hearing aid ownership from hearing loss characteristics. By including hearing disability, sensitivity for predicting hearing aid ownership increased by up to 40% compared with just audiometric hearing loss measurements alone. Conclusions A decision tree approach that considers both objectively measured hearing loss and self-perceived hearing disability, could facilitate a more tailored and personalised approach for determining hearing aid needs in the older population. IMPLICATIONS FOR REHABILITATION Without intervention, older adults with hearing loss are at higher risk of cognitive decline and higher rates of depression, anxiety, social isolation. The provision of hearing aids can compensate hearing function, however, uptake and usage have been reported as low. Using a more precise cut-off from audiometric measures and self-perceived hearing disability scores could facilitate a tailored and personalised approach to screen and identify older adults for hearing aid needs.

Selecting the Best Candidates for Cisplatin-based Adjuvant Chemotherapy After Radical Cystectomy Among Patients with pN+ Bladder Cancer

A trend towards greater benefit from adjuvant chemotherapy (ACT) in pN+ bladder cancer (BCa) has been observed in multiple randomized controlled trials. However, it is still unclear which patients might benefit the most from this approach. We retrospectively analyzed a multicenter cohort of 1381 patients with pTany pN1–3 cM0 R0 urothelial BCa treated with radical cystectomy (RC) with or without cisplatin-based ACT. The main endpoint was overall survival (OS) after RC. We performed 1:1 propensity score matching to adjust for baseline characteristics and conducted a classification and regression tree (CART) analysis to assess postoperative risk groups and Cox regression analyses to predict OS. Overall, 391 patients (28%) received cisplatin-based ACT. After matching, two cohorts of 281 patients with pN+ BCa were obtained. CART analysis stratified patients into three risk groups: favorable prognosis (≤pT2 and positive lymph node [PLN] count ≤2; odds ratio [OR] 0.43), intermediate prognosis (≥pT3 and PLN count ≤2; OR 0.92), and poor prognosis (pTany and PLN count ≥3; OR 1.36). Only patients with poor prognosis benefitted from ACT in terms of OS (HR 0.51; p < 0.001). We created the first algorithm that stratifies patients with pN+ BCa into prognostic classes and identified patients with pTany BCa with PLN ≥3 as the most suitable candidates for cisplatin-based ACT. Patient summaryWe found that overall survival among patients with bladder cancer and evidence of lymph node involvement depends on cancer stage and the number of positive lymph nodes. Patients with more than three nodes affected by metastases seem to experience the greatest overall survival benefit from cisplatin-based chemotherapy after bladder removal. Our study suggests that patients with the highest risk should be prioritized for cisplatin-based chemotherapy after bladder removal.

Training Load, Pain Intensity, and Functioning Can Explain Injuries in Dancers: A Classification and Regression Tree (CART) Analysis.

To identify the biopsychosocial factors that explain injuries in dancers through use of a Classification and Regression Tree (CART) analysis. Cross-sectional study performed with dancers over 18 years of age from dance schools in Brazil. Sociodemographic factors, pain intensity (numerical pain rating scale, NPRS), anxiety (Beck anxiety inventory, BAI), and functioning (WHO Disability Assessment Schedule 2.0, WHODAS) were assessed. We used a classification and regression tree (CART) analysis considering the occurrence of injury as a dependent variable, and the independent variables included age, training load, BAI, pain level, and WHODAS 2.0 score, adjusted for body mass index. The sample included 61 dancers (ballet, jazz, contemporary, tap, and urban dance), and they were mainly women (72.1%) with a mean age of 25.6 (± 5.79) years. Thirty-three percent presented dance-related musculoskeletal injuries in the last 12 months. The CART analysis showed that training load was the main associated factor, explaining 52.4% of injuries in dancers with a workload > 11.5 hrs/wk (p=0.05). When NPRS > 6.5 is added to the high load of training, these two variables explain 100% of the cases of injury (p=0.02). Training load < 11.5 hrs/wk and WHODAS 2.0 score < 68.0 were associated factors for no injury in dancers (80%; p=0.03). Our results demonstrated that training load, functioning, and pain intensity were factors associated with injury in dancers, and we established cutoff points. The inclusion of functioning in the algorithm suggests the influence of structure, function, activity, and social participation on injury in dancers and strengthens the patient-centered care.

Voriconazole therapeutic drug monitoring among lung transplant recipients receiving targeted therapy for invasive aspergillosis.

Voriconazole is the first line treatment for invasive aspergillosis (IA) Current guidelines suggest performing regular voriconazole therapeutic drug monitoring (TDM) to optimize treatment efficacy. We aimed to determine if TDM was predictive of clinical outcome in LTRs. Retrospective chart review was performed for all LTRs with probable or proven IA, treated with voriconazole monotherapy and who underwent TDM during therapy. Clinical outcome and toxicity were measured at 12weeks. Classification and regression tree (CART) analysis was used to determine the most predictive voriconazole level thresholds for successful outcome. One hundred and eighteen TDM samples from 30 LTRs with IA were analyzed. Three LTRs were excluded due to early treatment discontinuation. The median TDM level was 1.2 μg/ml (range 0.06-7.3). At 12weeks, 62% (17/27) of patients had a successful outcome, while 37% (10/27) of patients failed therapy. CART analysis determined that the best predictor for successful outcome was a median TDM level >0.72 μg/ml. Seventy percent (14/20) of patients with median TDM above 0.72 μg/ml had a successful outcome, compared to 42.9% (3/7) of patients with a median TDM below 0.72 μg/ml (OR 3.11; 95% CI: 0.53-20.4; P=0.21). CART analysis determined that a TDM level greater than 2.13 μg/ml was predictive of hepatotoxicity. Our data suggests that a voriconazole TDM range between 0.72 μg/ml and 2.13 μg/ml may be associated with improved outcomes. Our study is in line with current recommendations on the use of voriconazole TDM in improving outcome and minimizing toxicity in LTR with IA.

SO-9 Pathological primary tumour and lymph node regression following neoadjuvant chemotherapy in resectable oesophagogastric cancer: Pooled analysis of 1619 patients from two randomised trials

No definitive largescale data exists evaluating the role of pathologically defined regression changes within lymph nodes (LN) of resected oesophagogastric (OG) adenocarcinoma following neoadjuvant chemotherapy. There is also a lack of centralised data assessing the relationship between tumour and LN regression changes and their impact on survival in localised OG cancer. Data and samples from two large prospective randomised trials in resectable OG adenocarcinoma (UK MRC OE05 and ST03 trials) were pooled using a two-stage meta-analysis. H&E stained slides were available for central review from 1619 patients who underwent surgical resection following neoadjuvant chemotherapy. Mandard tumour regression grade (TRG) and LN regression ((LNR) scored as present/absent) were assessed centrally and correlated with survival using a Cox model to obtain hazard ratios. An exploratory analysis to assess whether the available results could define subgroups with distinct prognoses was conducted using a classification and regression tree (CART) analysis. 19,489 H&E slides from 761 OE05 participants and 20,277 slides from 858 ST03 participants were reviewed centrally. Within ST03, there was no difference between treatment arm (ECX ± Bev) in terms of TRG score or LNR. Within OE05, ECX was associated with lower TRG score and higher rates of LNR than CF. Neither trial demonstrated a correlation between TRG score and the presence or absence of LNR. In univariable analysis, lower TRG score, lower pathological N-stage (pN), lower pathological T-stage (pT), the presence of LNR, a well/moderately differentiated tumour, and achieving an R0 resection were all associated with significantly increased survival. However, multivariable analysis found that only pN, pT, differentiation and R0 were independent indicators of prognosis. Importantly, pN stage at resection significantly improved survival (3-year OS for ypN0 74% vs 36% for ypN1+, HR 3.4, p Exploratory CART analysis identified six subgroups based on the prognostic variables above, with 3-year OS ranging from 83% to 22%; with the most important single prognostic variable being pathological N-stage. This study represents the largest central review of primary tumour and LN regression in early OG cancer from patients treated with neoadjuvant chemotherapy within randomised controlled trials. pN and pT stage, tumour differentiation, and resection status in the post-operative specimen are strong independent indicators of survival in multivariate analysis. Exploratory CART analysis can be used to define groups with differing prognoses. Our results suggest that the assessment of LNR and TRG is not necessary to define prognosis further.

Detecting accelerometer non-wear periods using change in acceleration combined with rate-of-change in temperature

BackgroundAccelerometery is commonly used to estimate physical activity, sleep, and sedentary behavior. In free-living conditions, periods of device removal (non-wear) can lead to misclassification of behavior with consequences for research outcomes and clinical decision making. Common methods for non-wear detection are limited by data transformations (e.g., activity counts) or algorithm parameters such as minimum durations or absolute temperature thresholds that risk over- or under-estimating non-wear time. This study aimed to advance non-wear detection methods by integrating a ‘rate-of-change’ criterion for temperature into a combined temperature-acceleration algorithm.MethodsData were from 39 participants with neurodegenerative disease (36% female; age: 45–83 years) who wore a tri-axial accelerometer (GENEActiv) on their wrist 24-h per day for 7-days as part of a multi-sensor protocol. The reference dataset was derived from visual inspection conducted by two expert analysts. Linear regression was used to establish temperature rate-of-change as a criterion for non-wear detection. A classification and regression tree (CART) decision tree classifier determined optimal parameters separately for non-wear start and end detection. Classifiers were trained using data from 15 participants (38.5%). Outputs from the CART analysis were supplemented based on edge cases and published parameters.ResultsThe dataset included 186 non-wear periods (85.5% < 60 min). Temperature rate-of-change over the first five minutes of non-wear was − 0.40 ± 0.17 °C/minute and 0.36 ± 0.21 °C/minute for the first five minutes following device donning. Performance of the DETACH (DEvice Temperature and Accelerometer CHange) algorithm was improved compared to existing algorithms with recall of 0.942 (95% CI 0.883 to 1.0), precision of 0.942 (95% CI 0.844 to 1.0), F1-Score of 0.942 (95% CI 0.880 to 1.0) and accuracy of 0.996 (0.994–1.000).ConclusionThe DETACH algorithm accurately detected non-wear intervals as short as five minutes; improving non-wear classification relative to current interval-based methods. Using temperature rate-of-change combined with acceleration results in a robust algorithm appropriate for use across different temperature ranges and settings. The ability to detect short non-wear periods is particularly relevant to free-living scenarios where brief but frequent removals occur, and for clinical application where misclassification of behavior may have important implications for healthcare decision-making.

Thresholding of prominent biomarkers of breast cancer on overall survival using classification and regression tree.

HER2, ER, PR, and ERBB2 play a vital role in treating breast cancer. These are significant predictive and prognosis biomarkers of breast cancer. We aim to obtain a unique biomarker-specific prediction on overall survival to know their survival and death risk. Survival analysis is performed on classified data using Classification and Regression Tree (CART) analysis. Hazard ratio and Confidence Interval are computed using MLE and the Bayesian approach with the CPH model for univariate and multivariable illustrations. Validation of CART is executed with the Brier score, and accuracy and sensitivity are obtained using the k-nn classifier. Utilizing CART analysis, the cut-off value of continuous-valued biomarkers HER2, ER, PR, and ERBB2 are obtained as 14.707, 8.128, 13.153, and 6.884, respectively. Brier score of CART is 0.16 towards validation of methodology. Survival analysis gives a demonstration of the survival estimates with significant statistical strategies. Patients with breast cancer are at low risk of death, whose HER2 value is below its cut-off value, and ER, PR, and ERBB2 values are greater than their cut-off values. This comparison is with the patient having the opposite side of these cut-off values for the same biomarkers.

Clinical factors associated with antibiotic de-escalation after a positive multiplex molecular respiratory panel

Background: Under ideal circumstances, multiplex molecular respiratory panels can support early all discontinuation of unnecessary antibiotics by facilitating diagnosis of viral infection. Our goal was to identify clinic situations in which a positive respiratory panel was associated with antibiotic de-escalation. We focused on gram-negative antibiotics in recognition of the urgent threat posed by gram-negative resistance. Methods: The sample included hospitalized adults tested by respiratory panel while receiving gram-negative antibiotics at the University of Maryland Medical Center from 2015 to 2020. Only the first respiratory panel performed during hospitalization was included. The primary outcome was the combination of a positive result on respiratory panel indicating detection of a viral pathogen and de-escalation of gram-negative antibiotics. De-escalation was assessed based on antibiotics administered on day 3 after testing and was defined by discontinuation or switch to an agent with a narrower spectrum of activity. Least absolute shrinkage and selection operator (LASSO) regression was used to construct the multivariable logistic regression model. Classification and regression tree (CART) analysis was used to identify subgroups with a higher likelihood of the primary outcome. Results: Of 8,326 patients, 1,462 (17.6%) tested positive by respiratory panel. The most common pathogen was rhinovirus (7.9% of the sample). Gram-negative–targeted antibiotics were de-escalated in 4,456 cases (53.5% of the sample), including 887 patients with a positive result on respiratory panel indicating a viral pathogen (60.7% of patients with a positive viral result). LASSO regression was used to select 12 variables (Table 1). Admitting diagnosis of pneumonia (OR, 1.42), comorbid substance abuse (OR, 1.39), chronic pulmonary disease (OR, 1.39), and admission from home (OR, 1.34) were associated with antibiotic de-escalation in conjunction with a positive respiratory panel. Leukocytosis (OR, 0.59), hematologic malignancy (OR, 0.64), mechanical ventilation at time of testing (OR, 0.68), and hypotension (OR, 0.77) were associated with decreased likelihood of antibiotic de-escalation in conjunction with a positive respiratory panel. CART analysis identified patients tested within 40 hours of admission as having a higher likelihood of a positive result in conjunction with antibiotic de-escalation. Among patients tested within 40 hours of admission, the probability of a positive result followed by antibiotic de-escalation was 11.9% (95% CI, 11.1%–12.8%). For patients tested &gt;40 hours after admission, the probability was 6.0% (95% CI, 4.8%–7.2%). Conclusions: Targeted use of respiratory panel testing may increase the likelihood of an informative result that can drive decision making related to antibiotic use. Our exploratory analysis suggests that respiratory panel testing in the first 2 daysFunding: NoneDisclosures: None

Hippocampal Metastasis Rate Based on Non-Small Lung Cancer TNM Stage and Molecular Markers.

Hippocampal-avoidance whole-brain radiation therapy (HA-WBRT) is justified because of low hippocampal brain metastases (BM) rate and its prevention of cognitive decline. However, we hypothesize that the risk of developing BM in the hippocampal-avoidance region (HAR) may differ depending on the lung-cancer stage and molecular status. We retrospectively reviewed 123 patients with non-small cell lung cancer (NSCLC) at the initial diagnosis of BM. The number of BMs within the HAR (5 mm expansion) was counted. The cohort was divided into patients with and without BMs in the HAR, and their clinical variables, TNM stage, and epidermal growth factor receptor (EGFR) status were compared. The most influential variable predicting BMs in the HAR was determined using multi-variable logistic regression, classification and regression tree (CART) analyses, and gradient boosting method (GBM). The feasibility of HAR expansion was tested using generalized estimating equation marginal model. Patients with BMs in the HAR were more frequently non-smokers, and more likely to have extra-cranial metastases and EGFR mutations (p<0.05). Multi-variable analysis revealed that extra-cranial metastases were independently associated with the presence of BM in the HAR (odds ratio=8.75, p=0.04). CART analysis and GBM revealed that the existence of extra-cranial metastasis was the most influential variable predicting BM occurrence in the HAR (variable importance: 23% and relative influence: 37.38). The estmated BM incidence of patients without extra-cranial metastases in th extended HAR (7.5-mm and 10-mm expansion) did not differ significantly from that in the conventional HAR. In conclusion, NSCLC patients with extra-cranial metastases were more likely to have BMs in the HAR than those without extra-cranial metastases.

A Pilot Study of the Efficacy and Economical Sustainability of Acute Coronavirus Disease 2019 Patient Management in an Outpatient Setting.

ObjectiveTo report a preliminary experience of outpatient management of patients with Coronavirus disease 2019 (COVID-19) through an innovative approach of healthcare delivery.Patients and MethodsPatients evaluated at the Mild-to-Moderate COVID-19 Outpatient clinics (MMCOs) of San Raffaele University Hospital and Luigi Sacco University Hospital in Milan, Italy, from 1 October 2020 to 31 October 2021 were included. Patients were referred by general practitioners (GPs), Emergency Department (ED) physicians or hospital specialists (HS) in case of moderate COVID-19. A classification and regression tree (CART) model predicting ED referral by MMCO physicians was developed to aid GPs identify those deserving immediate ED admission. Cost-effectiveness analysis was also performed.ResultsA total of 660 patients were included. The majority (70%) was referred by GPs, 21% by the ED and 9% by HS. Patients referred by GPs had more severe disease as assessed by peripheral oxygen saturation (SpO2), ratio of arterial oxygen partial pressure to fractional inspired oxygen (PaO2/FiO2), C-reactive protein (CRP) levels and interstitial involvement at lung ultrasound. Among them, 18% were addressed to the ED following MMCO assessment. CART analysis identified three independent predictors, namely home-measured SpO2, age and body mass index (BMI), that robustly divide patients into risk groups of COVID-19 severity. Home-measured SpO2 < 95% and BMI ≥ 33 Kg/m2 defined the high-risk group. The model yielded an accuracy (95% CI) of 83 (77–88)%. Outpatient management of COVID-19 patients allowed the national healthcare system to spare 1,490,422.05 € when compared with inpatient care.ConclusionMild-to-moderate COVID-19 outpatient clinics were effective and sustainable in managing COVID-19 patients and allowed to alleviate pressure on EDs and hospital wards, favoring effort redirection toward non-COVID-19 patients.

The interplay of chronic stress and genetic traits discriminates between patients suffering from multisomatoform disorder with pain as the leading symptom and matched controls

ObjectiveSomatoform disorders and functional somatic syndromes (FSS) with symptoms that are not sufficiently explained by physical or technical examination are among the most challenging underlying causes. Many different somatoform disorders and FSS have overlapping symptoms, often with pain as the most prevalent one, leading to a high burden of disease. The concept of multisomatoform disorder (MSD) has been developed to acknowledge that fact. We analyzed a group of 151 patients and 149 matched controls to identify interactions of genetic and environmental factors with a possible influence on the development of MSD. DesignIn a retrospective case-control study, we performed a statistical analysis on 151 patients and 149 matched controls using logistic regression and a Classification and Regression Tree (CART) analysis. ResultsThe logistic regression analysis of genes and environmental factors demonstrated significant differences in the results of the Trier Inventory of Chronic Stress (TICS) questionnaire, the single nucleotide polymorphism rs1800955 of the dopamine receptor D4 and the single nucleotide polymorphism rs4818 of the enzyme catechol-O-methyltransferase between patients with MSD and healthy controls. The resulting decision tree of the CART analysis determined that the TICS questionnaire was able to differentiate patients and controls most accurately, followed by certain genotypes of the 5-hydroxytryptamine receptor 2A and a single nucleotide polymorphism of the enzyme catechol-O-methyltransferase. ConclusionsThe results of the statistical analysis identified a gene-environmental interaction possibly leading to MSD. The resulting identifiers could be used as a reference to inform diagnostic algorithms to easier identify patients suffering from MSD.

Ranked determinants of telemedicine diabetic retinopathy screening performance in the United States primary care safety-net setting: an exploratory CART analysis

BackgroundDiabetic retinopathy (DR) is a leading cause of blindness worldwide, despite easy detection and effective treatment. Annual screening rates in the USA remain low, especially for the disadvantaged, which telemedicine-based DR screening (TDRS) during routine primary care has been shown to improve. Screening rates from such programs have varied, however, pointing to inconsistent implementation and unaddressed barriers. This work seeks to identify and prioritize modifiable barriers for targeted intervention.MethodsIn this final phase of an exploratory mixed-methods study, we developed, validated, and administered a 62-item survey to multilevel stakeholders involved with TDRS in primary care safety-net clinics. Survey items were aligned with previously identified determinants of clinic-level screening and mapped to the Consolidated Framework for Implementation Research (CFIR). Classification and Regression Tree (CART) analyses were used to identify and rank independent variables predictive of individual-level TDRS screening performance.ResultsOverall, 133 of the 341 invited professionals responded (39%), representing 20 safety-net clinics across 6 clinical systems. Respondents were predominately non-Hispanic White (77%), female (94%), and between 31 and 65 years of age (79%). Satisfaction with TDRS was high despite low self-reported screening rates. The most important screening determinants were: provider reinforcement of TDRS importance; explicit instructions by providers to staff; effective reminders; standing orders; high relative priority among routine diabetic measures; established TDRS workflows; performance feedback; effective TDRS champions; and leadership support.ConclusionsIn this survey of stakeholders involved with TDRS in safety-net clinics, screening was low despite high satisfaction with the intervention. The best predictors of screening performance mapped to the CFIR constructs Leadership Engagement, Compatibility, Goals & Feedback, Relative Priority, Champions, and Available Resources. These findings facilitate the prioritization of implementation strategies targeting determinants of TDRS performance, potentially increasing its public health impact.

Assessing Pre-test Probability Of Transthyretin Genetic Variant Based On Demographic Information: Implications In Genetic Testing Requirements For Tafamidis Coverage

<h3>Background</h3> Tafamidis is an FDA-approved therapy for Transthyretin Amyloid Cardiomyopathy (ATTR-CM) regardless of genotype. Data show treatment decreases hazard of all-cause mortality and cardiovascular-related hospitalizations. Patients also experience a slower decline in exertional capacity and quality of life shortly after receiving the first dose. Current FDA indications and insurance coverage decisions require testing for predisposing genes such as the V122I variant commonly seen in patients of West African descent. Our clinical experience indicates that patients' likelihood of genetic positivity is impacted by their age, sex, and race. We sought to evaluate the likelihood of patients with ATTR-CM having a predisposing gene mutation based on these demographic factors. <h3>Methods</h3> This is a single-center retrospective cohort study of 634 consecutive patients diagnosed with ATTR-CM who underwent genetic testing. Classification and regression trees (CART), a supervised machine learning method, uses pre-specified independent variables in stepwise fashion to maximize accuracy of outcome predictions in a decision tree. Age, sex, and race were used as independent variables to predict gene positivity. <h3>Results</h3> In our study cohort (median age 76 [69, 82], 86% male, 27% African American [AA]), 181 (29%) patients with ATTR-CM had predisposing gene mutations. V122I and Thr60Ala mutations were present in 20% and 2% of the cohort, respectively. CART analysis showed that race served most efficiently as the root node (Figure 1). 73% of AA had a predisposing mutation, compared to 14.5% in non-AA. Patients with ATTR-CM who were not AA and over age 69 had less than a 4% risk of having a predisposing mutation. Our model accurately predicted 89 (sensitivity 80%; specificity 92%) of the entire cohort and 93% (sensitivity 56%; specificity 98%) in patients who were not AA. <h3>Conclusions</h3> In our single-center experience, non-AA patients with ATTR-CM are less likely to have a detectable predisposing genetic mutation than AA patients. Demographic features appear to be predictive of gene positivity, and should inform pre-test probability. In many cases genetic testing requirements may cause unnecessary delays in tafamidis therapy.

Association between self-management behaviour and quality of life in people with heart failure: a retrospective study

BackgroundThe purpose of this study was to investigate the variables that significantly associated with the quality of life in people with heart failure, and particularly, to identify the association between self-management behaviour and the quality of life.MethodsThis retrospective study used data from heart failure outpatient clinics at two large tertiary medical centres in Seoul and Suwon, South Korea. We enrolled 119 participants who completed echocardiography and stress tests and responded to questionnaires on self-management behaviour and quality of life. We collected more data on sociodemographic and clinical characteristics and anthropometric and serum blood test results through electronic medical record review. We analysed data using multiple linear regression and the classification and regression tree (CART) method to explore the associated factors with the quality of life in participants with heart failure.ResultsParticipants’ mean age was 74.61 years, and women represented 52.1% of the sample. It showed that cardiac systolic function (β = 0.26, p = .013) and self-management behaviour (β = 0.20, p = .048) were two major associated factors with the quality of life in participants with heart failure in the multiple linear regression analysis. Also, cardiac systolic function and self-management behaviour were shown to be the primary determinants for the quality of life in those with heart failure in the CART analysis. Therefore, self-management behaviour of the participants with heart failure was a significant modifiable factor that can improve their quality of life.ConclusionsHealthcare providers should be aware of the importance of self-management in people with heart failure and help promote their quality of life by enhancing their self-management behaviour as own efforts to properly maintain and monitor the health status and prevent further worsening of heart failure.

Using Aggregate Vasoactive-Inotrope Scores to Predict Clinical Outcomes in Pediatric Sepsis.

ObjectivesThe heterogeneity of sepsis makes it difficult to predict outcomes using existing severity of illness tools. The vasoactive-inotrope score (VIS) is a quantitative measure of the amount of vasoactive support required by patients. We sought to determine if a higher aggregate VIS over the first 96 h of vasoactive medication initiation is associated with increased resource utilization and worsened clinical outcomes in pediatric patients with severe sepsis.DesignRetrospective cohort study.SettingSingle-center at Children's Wisconsin in Milwaukee, WI.PatientsOne hundred ninety-nine pediatric patients, age less than 18 years old, diagnosed with severe sepsis, receiving vasoactive medications between January 2017 and July 2019.InterventionsRetrospective data obtained from the electronic medical record, calculating VIS at 2 h intervals from 0–12 h and at 4 h intervals from 12–96 h from Time 0.MeasurementsAggregate VIS derived from the hourly VIS area under the curve (AUC) calculation based on the trapezoidal rule. Data were analyzed using Pearson's correlations, Mann-Whitney test, Wilcoxon signed rank test, and classification, and regression tree (CART) analyses.Main ResultsHigher aggregate VIS is associated with longer hospital LOS (p < 0.0001), PICU LOS (p < 0.0001), MV days (p = 0.018), increased in-hospital mortality (p < 0.0001), in-hospital cardiac arrest (p = 0.006), need for ECMO (p < 0.0001), and need for CRRT (p < 0.0001). CART analyses found that aggregate VIS >20 is an independent predictor for in-hospital mortality (p < 0.0001) and aggregate VIS >16 for ECMO use (p < 0.0001).ConclusionsThere is a statistically significant association between aggregate VIS and many clinical outcomes, allowing clinicians to utilize aggregate VIS as a physiologic indicator to more accurately predict disease severity/trajectory in pediatric sepsis.

Cognitive Status Predicts Return to Functional Independence After Minor Stroke: A Decision Tree Analysis.

About two-thirds of patients with minor strokes are discharged home. However, these patients may have difficulties returning to their usual living activities. To investigate the factors associated with successful home discharge, our aim was to provide a decision tree (based on clinical data) that could identify if a patient discharged home could return to pre-stroke activities and to perform an external validation of this decision tree on an independent cohort. Two cohorts of patients with minor strokes gathered from stroke registries at the Hôpital Pitié-Salpêtrière and University Hospital Bern were included in this study (n = 105 for the construction cohort coming from France; n = 100 for the second cohort coming from Switzerland). The decision tree was built using the classification and regression tree (CART) analysis on the construction cohort. It was then applied to the validation cohort. Accuracy, sensitivity, specificity, false positive, and false-negative rates were reported for both cohorts. In the construction cohort, 60 patients (57%) returned to their usual, pre-stroke level of independence. The CART analysis produced a decision tree with the Montreal Cognitive Assessment (MoCA) as the first decision point, followed by discharge NIHSS score or age, and then by the occupational status. The overall prediction accuracy to the favorable outcome was 80% in the construction cohort and reached 72% accuracy in the validation cohort. This decision tree highlighted the role of cognitive function as a crucial factor for patients to return to their usual activities after a minor stroke. The algorithm may help clinicians to tailor planning of patients' discharge.