Classification and regression tree analysis vs. multivariable linear and logistic regression methods as statistical tools for studying haemophilia

NMR metabolomics of cerebrospinal fluid differentiates inflammatory diseases of the central nervous system

BACKGROUND

Myriad infectious and noninfectious causes of encephalomyelitis (EM) have similar clinical manifestations, presenting serious challenges to diagnosis and treatment. Metabolomics of cerebrospinal fluid (CSF) was explored as a method of differentiating among neurological diseases causing EM using a single CSF sample.

METHODOLOGY/PRINCIPAL FINDINGS

1H NMR metabolomics was applied to CSF samples from 27 patients with a laboratory-confirmed disease, including Lyme disease or West Nile Virus meningoencephalitis, multiple sclerosis, rabies, or Histoplasma meningitis, and 25 controls. Cluster analyses distinguished samples by infection status and moderately by pathogen, with shared and differentiating metabolite patterns observed among diseases. CART analysis predicted infection status with 100% sensitivity and 93% specificity.

CONCLUSIONS/SIGNIFICANCE

These preliminary results suggest the potential utility of CSF metabolomics as a rapid screening test to enhance diagnostic accuracies and improve patient outcomes.

Patient scheduling based on a service-time prediction model: a data-driven study for a radiotherapy center

With the growth of the population, access to medical care is in high demand, and queues are becoming longer. The situation is more critical when it concerns serious diseases such as cancer. The primary problem is inefficient management of patients rather than a lack of resources. In this work, we collaborate with the Centre Intégré de Cancérologie de Laval (CICL). We present a data-driven study based on a nonblock approach to patient appointment scheduling. We use data mining and regression methods to develop a prediction model for radiotherapy treatment duration. The best model is constructed by a classification and regression tree; its accuracy is 84%. Based on the predicted duration, we design new workday divisions, which are evaluated with various patient sequencing rules. The results show that with our approach, 40 additional patients are treated daily in the cancer center, and a considerable improvement is noticed in patient waiting times and technologist overtime.

Evaluation of potential predictor variables for PCR assay diagnosis of Anaplasma phagocytophilum infection in equids in Northern California

OBJECTIVE To identify clinical or clinicopathologic variables that can be used to predict a positive PCR assay result for Anaplasma phagocytophilum infection in equids. ANIMALS 162 equids. PROCEDURES Medical records were reviewed to identify equids that underwent testing for evidence of A phagocytophilum infection by PCR assay between June 1, 2007, and December 31, 2015. For each equid that tested positive (case equid), 2 time-matched equids that tested negative for the organism (control equids) were identified. Data collected included age, sex, breed, geographic location (residence at the time of testing), physical examination findings, and CBC and plasma biochemical analysis results. Potential predictor variables were analyzed by stepwise logistic regression followed by classification and regression tree analysis. Generalized additive models were used to evaluate identified predictors of a positive test result for A phagocytophilum. RESULTS Total lymphocyte count, plasma total bilirubin concentration, plasma sodium concentration, and geographic latitude were linear predictors of a positive PCR assay result for A phagocytophilum. Plasma creatine kinase activity was a nonlinear predictor of a positive result. CONCLUSIONS AND CLINICAL RELEVANCE Assessment of predictors identified in this study may help veterinarians identify equids that could benefit from early treatment for anaplasmosis while definitive test results are pending. This information may also help to prevent unnecessary administration of oxytetracycline to equids that are unlikely to test positive for the disease.

Risk factors can classify individuals who develop accelerated knee osteoarthritis: Data from the osteoarthritis initiative

We assessed which combinations of risk factors can classify adults who develop accelerated knee osteoarthritis (KOA) or not and which factors are most important. We conducted a case-control study using data from baseline and the first four annual visits of the Osteoarthritis Initiative. Participants had no radiographic KOA at baseline (Kellgren-Lawrence [KL]<2). We classified three groups (matched on sex): (i) accelerated KOA: >1 knee developed advance-stage KOA (KL = 3 or 4) within 48 months; (ii) typical KOA: >1 knee increased in radiographic scoring (excluding those with accelerated KOA); and (iii) No KOA: no change in KL grade by 48 months. We selected eight predictors: Serum concentrations for C-reactive protein, glycated serum protein (GSP), and glucose; age; sex; body mass index; coronal tibial slope, and femorotibial alignment. We performed a classification and regression tree (CART) analysis to determine rules for classifying individuals as accelerated KOA or not (no KOA and typical KOA). The most important baseline variables for classifying individuals with incident accelerated KOA (in order of importance) were age, glucose concentrations, BMI, and static alignment. Individuals <63.5 years were likely not to develop accelerated KOA, except when overweight. Individuals >63.5 years were more likely to develop accelerated KOA except when their glucose levels were >81.98 mg/dl and they did not have varus malalignment. The unexplained variance of the CART = 69%. These analyses highlight the complex interactions among four risk factors that may classify individuals who will develop accelerated KOA but more research is needed to uncover novel risk factors. © 2017 Orthopaedic Research Society. Published by Wiley Periodicals, Inc. J Orthop Res 36:876-880, 2018.

Advances in the genome-wide association study of chronic hepatitis B susceptibility in Asian population

Chronic hepatitis B (CHB) is the most common chronic liver disease resulting from viral infection and has become a serious threat to human health. Each year, about 1.2 million people in the world die from diseases caused by chronic infection of hepatitis B virus. The genetic polymorphism is significantly associated with the susceptibility to chronic hepatitis B. Genome-wide association study was recently developed and has become an important tool to detect susceptibility genes of CHB. To date, a number of CHB-associated susceptibility loci and regions have been identified by scientists over the world. To clearly understand the role of susceptibility loci in the occurrence of CHB is important for the early diagnosis and prevention of CHB.

Conditional Inference Tree for Multiple Gene-Environment Interactions on Myocardial Infarction

BACKGROUND AND AIMS

Identifying gene-environment interaction in the context of multiple environmental factors has been a challenging task. We aimed to use conditional inference tree (CTREE) to strata myocardial infarction (MI) risk synthesizing information from both genetic and environmental factors.

METHODS

We conducted a case-control study including 1440 Chinese men (730 MI patients and 710 controls). We first calculated a weighted genetic risk score (GRS) by combining 25 single nucleotide polymorphisms (SNPs) that had been identified to be associated with coronary artery diseases in previous genome wide association studies. We then developed a CTREE model to interpret the gene-environment interaction network in predicting MI.

RESULTS

We detected high-order interactions between dyslipidemia, GRS, smoking status, age and diabetes. Of all the variables examined, high density lipoprotein cholesterol (HDL-C) of 1.25 mmlo/L was identified as the key discriminator. The subsequent splits of MI were low density lipoprotein cholesterol (LDL-C) of 4.01 mmol/L and GRS of 20.9. We found that individuals with HDL-C ≤1.25 mmol/L, GRS >20.9 and lipoprotein (a) > 0.09 g/L had a higher risk of MI than those who at the lowest risk group (OR: 5.89, 95% CI: 3.99-8.69). This magnitude of MI risk was similar to the combination of HDL-C ≤1.25 mmol/L, GRS ≤20.9, smoking and lipoprotein (a) > 0.15 g/L (OR: 5.49, 95% CI: 3.51-8.58).

CONCLUSIONS

The multiple interactions between genetic and environmental factors can be visually present via the CTREE approach. The tree diagram also simplifies the decision making procedure by answering a sequence of questions along the branches.

Clinical signs, profound acidemia, hypoglycemia, and hypernatremia are predictive of mortality in 1,400 critically ill neonatal calves with diarrhea

Profound acidemia impairs cellular and organ function and consequently should be associated with an increased risk of mortality in critically ill humans and animals. Neonatal diarrhea in calves can result in potentially serious metabolic derangements including profound acidemia due to strong ion (metabolic) acidosis, hyper-D-lactatemia, hyper-L-lactatemia, azotemia, hypoglycemia, hyperkalemia and hyponatremia. The aim of this retrospective study was to assess the prognostic relevance of clinical and laboratory findings in 1,400 critically ill neonatal calves with diarrhea admitted to a veterinary teaching hospital. The mortality rate was 22%. Classification tree analysis indicated that mortality was associated with clinical signs of neurologic disease, abdominal emergencies, cachexia, orthopedic problems such as septic arthritis, and profound acidemia (jugular venous blood pH < 6.85). When exclusively considering laboratory parameters, classification tree analysis identified plasma glucose concentrations < 3.2 mmol/L, plasma sodium concentrations ≥ 151 mmol/L, serum GGT activity < 31 U/L and a thrombocyte count < 535 G/L as predictors of mortality. However, multivariable logistic regression models based on these laboratory parameters did not have a sufficiently high enough sensitivity (59%) and specificity (79%) to reliably predict treatment outcome. The sensitivity and specificity of jugular venous blood pH < 6.85 were 11% and 97%, respectively, for predicting non-survival in this study population. We conclude that laboratory values (except jugular venous blood pH < 6.85) are of limited value for predicting outcome in critically ill neonatal calves with diarrhea. In contrast, the presence of specific clinical abnormalities provides valuable prognostic information.

A cross-sectional study for predicting tail biting risk in pig farms using classification and regression tree analysis

Tail biting in pigs has been an identified behavioural, welfare and economic problem for decades, and requires appropriate but sometimes difficult on-farm interventions. The aim of the paper is to introduce the Classification and Regression Tree (CRT) methodologies to develop a tool for prevention of acute tail biting lesions in pigs on-farm. A sample of 60 commercial farms rearing heavy pigs were involved; an on-farm visit and an interview with the farmer collected data on general management, herd health, disease prevention, climate control, feeding and production traits. Results suggest a value for the CRT analysis in managing the risk factors behind tail biting on a farm-specific level, showing 86.7% sensitivity for the Classification Tree and a correlation of 0.7 between observed and predicted prevalence of tail biting obtained with the Regression Tree. CRT analysis showed five main variables (stocking density, ammonia levels, number of pigs per stockman, type of floor and timeliness in feed supply) as critical predictors of acute tail biting lesions, which demonstrate different importance in different farms subgroups. The model might have reliable and practical applications for the support and implementation of tail biting prevention interventions, especially in case of subgroups of pigs with higher risk, helping farmers and veterinarians to assess the risk in their own farm and to manage their predisposing variables in order to reduce acute tail biting lesions.

Intuitive pharmacogenetic dosing of risperidone according to CYP2D6 phenotype extrapolated from genotype in a cohort of first episode psychosis patients

Risperidone (R) is the most prescribed antipsychotic drug for patients with a first episode of psychosis (FEP). In a naturalistic cohort of chronic psychiatric inpatients, we demonstrated that clinicians adjust R dosage by CYP2D6 activity, despite being blinded to the genotype, which we described as an "intuitive pharmacogenetic" process. The aim of the present study is to replicate our previous findings of intuitive pharmacogenetic in a cohort of FEP patients using CYP2D6 phenotype extrapolated from genotypes. 70 FEP patients, under baseline treatment with R monotherapy were genotyped using the iPLEX® ADME PGx multiplex panel and TaqMan® Genotyping and Copy Number Assays. Plasma concentrations of R and its metabolite, 9-hydroxyrisperidone (9-OH), were determined. The predictive properties of those variables associated with R dosage were tested using a multiple linear regression model as well as regression trees. Significant differences in the mean daily dosage of R among CYP2D6 phenotypes were observed (Kruskal-Wallis test p=0.02): PM (4.00±2.3mg/mL), IM (4.56±2.44), EM (6.22±4.0mg/day) and UM (10.20±4.91mg/day). However, non-significant differences were observed in the R/9-OH ratio or in the Concentration/Dose ratio. Regression tree provided better estimations of R dosage than the multiple linear regression model (MAE=0.958 and R²=0.871). We confirm the "intuitive pharmacogenetic" dosing of R according to the CYP2D6 phenotype in a FEP cohort. The results presented provides a rationale for the clinical use of CYP2D6 genotyping in personalized medicine.

Improved clinical trial enrollment criterion to identify patients with diabetes at risk of end-stage renal disease

Design of Phase III trials for diabetic nephropathy currently requires patients at a high risk of progression defined as within three years of a hard end point (end-stage renal disease, 40% loss of estimated glomerular filtration rate, or death). To improve the design of these trials, we used natural history data from the Joslin Kidney Studies of chronic kidney disease in patients with diabetes to develop an improved criterion to identify such patients. This included a training cohort of 279 patients with type 1 diabetes and 134 end points within three years, and a validation cohort of 221 patients with type 2 diabetes and 88 end points. Previous trials selected patients using clinical criteria for baseline urinary albumin-to-creatinine ratio and estimated glomerular filtration rate. Application of these criteria to our cohort data yielded sensitivities (detection of patients at risk) of 70-80% and prognostic values of only 52-63%. We applied classification and regression trees analysis to select from among all clinical characteristics and markers the optimal prognostic criterion that divided patients with type 1 diabetes according to risk. The optimal criterion was a serum tumor necrosis factor receptor 1 level over 4.3 ng/ml alone or 2.9-4.3 ng/ml with an albumin-to-creatinine ratio over 1900 mg/g. Remarkably, this criterion produced similar results in both type 1 and type 2 diabetic patients. Overall, sensitivity and prognostic value were high (72% and 81%, respectively). Thus, application of this criterion to enrollment in future clinical trials could reduce the sample size required to achieve adequate statistical power for detection of treatment benefits.

Using Baidu Search Index to Predict Dengue Outbreak in China

This study identified the possible threshold to predict dengue fever (DF) outbreaks using Baidu Search Index (BSI). Time-series classification and regression tree models based on BSI were used to develop a predictive model for DF outbreak in Guangzhou and Zhongshan, China. In the regression tree models, the mean autochthonous DF incidence rate increased approximately 30-fold in Guangzhou when the weekly BSI for DF at the lagged moving average of 1-3 weeks was more than 382. When the weekly BSI for DF at the lagged moving average of 1-5 weeks was more than 91.8, there was approximately 9-fold increase of the mean autochthonous DF incidence rate in Zhongshan. In the classification tree models, the results showed that when the weekly BSI for DF at the lagged moving average of 1-3 weeks was more than 99.3, there was 89.28% chance of DF outbreak in Guangzhou, while, in Zhongshan, when the weekly BSI for DF at the lagged moving average of 1-5 weeks was more than 68.1, the chance of DF outbreak rose up to 100%. The study indicated that less cost internet-based surveillance systems can be the valuable complement to traditional DF surveillance in China.

The interaction effects of risk factors for hypertension in adults: a cross-sectional survey in Guilin, China

BACKGROUND

The prevalence of hypertension in adults is increasing each year and has become a main public health issue worldwide. We must consider the impact of both individual factors and interactions among these factors on hypertension in adults. This study was designed to elucidate the clinical and metabolic characteristics of the prevalence of hypertension in adults and to explore the risk factors and interactions among these factors in adults with hypertension.

METHODS

We used overall random sampling to conduct a cross-sectional survey of 6660 individuals undergoing a health check from July to November 2012, the subjects were aged 20 to 89 years, including 3480 men and 3180 women. The survey content included a questionnaire, anthropometry, laboratory measurements, and liver Doppler ultrasonography. The clinical and metabolic characteristics were compared between the cases (adult hypertensive patients) and the controls (normotensives). The classification tree model and the non-conditional logistic regression were used to analyze the interactions of risk factors for hypertension in adults.

RESULTS

In total, 1623 adult hypertensive patients (940 men and 683 women) were detected. The results showed that adult hypertensive patients were older and had higher levels of systolic blood pressure, diastolic blood pressure, body mass index, fasting plasma glucose, uric acid, triglycerides, total cholesterol, low-density lipoprotein cholesterol, and prevalence of non-alcoholic fatty liver disease (P < 0.001). The classification tree model comprising 5 layers, 39 nodes, and 20 terminal nodes showed that two variables, age and BMI, were closely related to hypertension in adults. The area under the receiver operating characteristic curve for classification tree model was 81.6 % (95 % CI: 80.6 % ~ 82.5 %). Both univariate and multivariate logistic regression analyses revealed that advanced age and high BMI had a significant positive interaction in terms of hypertension in adults. After controlling for confounding factors, the percentage of attributed interaction was 47.62 %.

CONCLUSIONS

This study showed that age, BMI, UA, TG, and TC were closely associated with the risk of hypertension in adults, and the positive interaction effect between advanced age and high BMI was an important risk factor for the prevalence of hypertension in adults.

Utility of Novel Plasma Metabolic Markers in the Diagnosis of Pediatric Tuberculosis: A Classification and Regression Tree Analysis Approach

Although tuberculosis (TB) has been the greatest killer due to a single infectious disease, pediatric TB is still hard to diagnose because of the lack of sensitive biomarkers. Metabolomics is increasingly being applied in infectious diseases. But little is known regarding metabolic biomarkers in children with TB. A combination of a NMR-based plasma metabolic method and classification and regression tree (CART) analysis was used to provide a broader range of applications in TB diagnosis in our study. Plasma samples obtained from 28 active TB children and 37 non-TB controls (including 21 RTIs and 16 healthy children) were analyzed by an orthogonal partial least-squares discriminant analysis (OPLS-DA) model, and 17 metabolites were identified that can separate children with TB from non-TB controls. CART analysis was then used to choose 3 of the markers, l-valine, pyruvic acid, and betaine, with the least error. The sensitivity, specificity, and area under the curve (AUC) of the 3 metabolites is 85.7% (24/28, 95% CI, 66.4%, 95.3%), 94.6% (35/37, 95% CI, 80.5%, 99.1%), and 0.984(95% CI, 0.917, 1.000), respectively. The 3 metabolites demonstrated sensitivity of 82.4% (14/17, 95% CI, 55.8%, 95.3%) and specificity of 83.9% (26/31, 95% CI, 65.5%, 93.9%), respectively, in 48 blinded subjects in an independent cohort. Taken together, the novel plasma metabolites are potentially useful for diagnosis of pediatric TB and would provide insights into the disease mechanism.

Analysis of risk factors and their interactions in type 2 diabetes mellitus: A cross-sectional survey in Guilin, China

Aims/Introduction

Type 2 diabetes is a metabolic disease characterized by insulin resistance, and is associated with the effects of genetic and environmental factors. The present study aimed to not only analyze the influence of a single factor for type 2 diabetes, but also to investigate the interaction effects between risk factors.

Materials and Methods

A total of 6,660 individuals selected by the method of cluster random sampling accepted a cross‐sectional survey (questionnaire investigation, physical measurement, laboratory examination and liver ultrasound examination). The classification tree was used to analyze the risk factors and their interactions in type 2 diabetes. The clinical and metabolic characteristics were compared between type 2 diabetes patients and controls, and the non‐conditional logistic regression model was used to quantitatively analyze the interactions.

Results

A total of 338 participants were classified as type 2 diabetes (217 men and 121 women), the classification tree model showed three variables with close associations with type 2 diabetes: age, triglycerides (TG) and non‐alcoholic fatty liver disease (NAFLD). Type 2 diabetes patients had higher age and incidences of high TG, NAFLD, hypertension, high body mass index, high uric acid, high total cholesterol, high low‐density lipoprotein cholesterol and low high‐density lipoprotein cholesterol. The multivariate logistic regression analysis showed that the following factors had interactions in type2 diabetes: high TG × advanced age (odds ratio 2.499, 95% confidence interval 1.868–3.344, P = 0.000), NAFLD × advanced age (odds ratio 1.250, 95% confidence interval 1.048–1.491, P = 0.013) and NAFLD × high TG (odds ratio 1.349, 95% confidence interval 1.144–1.590, P = 0.000).

Conclusions

The present study showed that type 2 diabetes resulted from the interactions of many factors; the interactions among age, TG and NAFLD are important risk factors for type 2 diabetes.

Impact of being overweight on factor VIII dosing in children with haemophilia A

INTRODUCTION

Treatment of haemophilia A (HA) requires infusions of factor VIII (FVIII) concentrates. The number of FVIII units infused to obtain a specific circulating FVIII level is calculated with the formula: [body weight (BW) (kg) × desired FVIII increase (%)]/2, with the assumption that each unit of FVIII infused per kg of BW increases the circulating FVIII level by 2%.

AIM

The aim of this study was to evaluate the impact of several morphometric parameters (BW, body mass index (BMI)-for-age, height), age and type of FVIII concentrate on FVIII recovery in children with HA.

METHODS

A total of 66 children aged between 10 and 18 with severe HA selected from six pharmacokinetic (PK) clinical trials using two recombinant FVIII concentrates were included in the analysis. Regression tree (RT) was used to identify predictors of FVIII recovery.

RESULTS

The median age was 14.5 years with a median FVIII recovery of 2.09 for all children. The median FVIII recovery was not significantly different between age groups. Two groups were created by RT: children with a BMI-for-age percentile <P95 (Median FVIII recovery: 1.94) and obese children with a BMI-for-age percentile ≥P95 (Median FVIII recovery: 2.65). The FVIII recovery was significantly different between these two groups (P < 0.001).

CONCLUSION

These results are consistent with previous studies conducted in adults with HA and confirm that the long-held and current practice of applying an arbitrary and universal recovery of two to the calculations of FVIII dosage should be abolished in both children and adults.