A multivariate logistic regression equation to screen for diabetes: development and validation

A Novel Approach for Feature Selection and Classification of Diabetes Mellitus: Machine Learning Methods

An active research area where the experts from the medical field are trying to envisage the problem with more accuracy is diabetes prediction. Surveys conducted by WHO have shown a remarkable increase in the diabetic patients. Diabetes generally remains in dormant mode and it boosts the other diseases if patients are diagnosed with some other disease such as damage to the kidney vessels, problems in retina of the eye, and cardiac problem; if unidentified, it can create metabolic disorders and too many complications in the body. The main objective of our study is to draw a comparative study of different classifiers and feature selection methods to predict the diabetes with greater accuracy. In this paper, we have studied multilayer perceptron, decision trees, K-nearest neighbour, and random forest classifiers and few feature selection techniques were applied on the classifiers to detect the diabetes at an early stage. Raw data is subjected to preprocessing techniques, thus removing outliers and imputing missing values by mean and then in the end hyperparameters optimization. Experiments were conducted on PIMA Indians diabetes dataset using Weka 3.9 and the accuracy achieved for multilayer perceptron is 77.60%, for decision trees is 76.07%, for K-nearest neighbour is 78.58%, and for random forest is 79.8%, which is by far the best accuracy for random forest classifier.

Developing a risk score for undiagnosed prediabetes or type 2 diabetes among Saharawi refugees in Algeria

Aims

To prevent type 2 diabetes mellitus (T2D) and reduce the risk of complications, early identification of people at risk of developing T2D, preferably through simple diabetes risk scores, is essential. The aim of this study was to create a risk score for identifying subjects with undiagnosed prediabetes or T2D among Saharawi refugees in Algeria and compare the performance of this score to the Finnish diabetes risk score (FINDRISC).

Methods

A cross-sectional survey was carried out in five Saharawi refugee camps in Algeria in 2014. A total of 180 women and 175 men were included. HbA1c and cut-offs proposed by the American Diabetes Association (ADA) were used to define cases. Variables to include in the risk score were determined by backwards elimination in logistic regression. Simplified scores were created based on beta coefficients from the multivariable model after internal validation with bootstrapping and shrinkage. The empirical cut-off value for the simplified score and FINDRISC was determined by Area Under the Receiver Operating Curve (AUROC) analysis.

Results

Variables included in the final risk score were age, body mass index (BMI), and waist circumference. The area under the curve (AUC) (C.I) was 0.82 (0.76, 0.88). The sensitivity, specificity, and positive and negative predictive values were 89, 65, 28, and 97%, respectively. AUC and sensitivity were slightly higher and specificity somewhat lower than for FINDRISC.

Conclusions

The risk score developed is a helpful tool to decide who should be screened for prediabetes or T2D by blood sample analysis. The performance of the risk score was adequate based on internal validation with bootstrap analyses, but should be confirmed in external validation studies.

A systematic review of diabetes risk assessment tools in sub-Saharan Africa

Objectives

To systematically review all current studies on diabetes risk assessment tools used in SSA to diagnose diabetes in symptomatic and asymptomatic patients.

Methods

Tools were identified through a systematic search of PubMed, Ovid, Google Scholar, and the Cochrane Library for articles published from January 2010 to January 2020. The search included articles reporting the use of diabetes risk assessment tool to detect individuals with type 2 diabetes in SSA. A standardized protocol was used for data extraction (registry #177726).

Results

Of the 825 articles identified, 39 articles met the inclusion criteria, and three articles reported tools used in SSA population but developed for the Western population. None was validated in SSA population. All but three articles were observational studies (136 and 58,657 study participants aged between the ages of 15 and 85 years). The Finnish Medical Association risk tool, World Health Organization (WHO) STEPS instrument, General Practice Physical Activity Questionnaire (GPPAQ), Rapid Eating and Activity Assessment for Patients (REAP), and an anthropometric tool were the most frequently used non-invasive tools in SSA. The accuracy of the tools was measured using sensitivity, specificity, or area under the receiver operating curve. The anthropometric predictor variables identified included age, body mass index, waist circumference, positive family of diabetes, and activity levels.

Conclusions

This systematic review demonstrated a paucity of validated diabetes risk assessment tools for SSA. There remains a need for the development and validation of a tool for the rapid identification of diabetes for targeted interventions.

Health Data-Driven Machine Learning Algorithms Applied to Risk Indicators Assessment for Chronic Kidney Disease

PURPOSE

As global aging progresses, the health management of chronic diseases has become an important issue of concern to governments. Influenced by the aging of its population and improvements in the medical system and healthcare in general, Taiwan's population of patients with chronic kidney disease (CKD) has tended to grow year by year, including the incidence of high-risk cases that pose major health hazards to the elderly and middle-aged populations.

METHODS

This study analyzed the annual health screening data for 65,394 people from 2010 to 2015 sourced from the MJ Group - a major health screening center in Taiwan - including data for 18 risk indicators. We used five prediction model analysis methods, namely, logistic regression (LR) analysis, C5.0 decision tree (C5.0) analysis, stochastic gradient boosting (SGB) analysis, multivariate adaptive regression splines (MARS), and eXtreme gradient boosting (XGboost), with estimated glomerular filtration rate (e-GFR) data to determine G3a, G3b & G4 stage CKD risk factors.

RESULTS

The LR analysis (AUC=0.848), SGB analysis (AUC=0.855), and XGboost (AUC=0.858) generated similar classification performance levels and all outperformed the C5.0 and MARS methods. The study results showed that in terms of CKD risk factors, blood urea nitrogen (BUN) and uric acid (UA) were identified as the first and second most important indicators in the models of all five analysis methods, and they were also clinically recognized as the major risk factors. The results for systolic blood pressure (SBP), SGPT, SGOT, and LDL were similar to those of a related study. Interestingly, however, socioeconomic status-related education was found to be the third important indicator in all three of the better performing analysis methods, indicating that it is more important than the other risk indicators of this study, which had different levels of importance according to the different methods.

CONCLUSION

The five prediction model methods can provide high and similar classification performance in this study. Based on the results of this study, it is recommended that education as the socioeconomic status should be an important factor for CKD, as high educational level showed a negative and highly significant correlation with CKD. The findings of this study should also be of value for further discussions and follow-up research.

Risk prediction of diabetic nephropathy using machine learning techniques: A pilot study with secondary data

AIMS

This research work presented a comparative study of machine learning (ML), including two objectives: (i) determination of the risk factors of diabetic nephropathy (DN) based on principal component analysis (PCA) via different cutoffs; (ii) prediction of DN patients using ML-based techniques.

METHODS

The combination of PCA and ML-based techniques has been implemented to select the best features at different PCA cutoff values and choose the optimal PCA cutoff in which ML-based techniques give the highest accuracy. These optimum features are fed into six ML-based techniques: linear discriminant analysis, support vector machine (SVM), logistic regression, K-nearest neighborhood, naïve Bayes, and artificial neural network. The leave-one-out cross-validation protocol is executed and compared ML-based techniques performance using accuracy and area under the curve (AUC).

RESULTS

The data utilized in this work consists of 133 respondents having 73 DN patients with an average age of 69.6±10.2 years and 54.2% of DN patients are female. Our findings illustrate that PCA combined with SVM-RBF classifier yields 88.7% accuracy and 0.91 AUC at 0.96 PCA cutoff.

CONCLUSIONS

This study also suggests that PCA combined with SVM-RBF classifier may correctly classify DN patients with the highest accuracy when compared to the models published in the existing research. Prospective studies are warranted to further validate the applicability of our model in clinical settings.

Improved Prediction of Protein-Protein Interaction Mapping on Homo Sapiens by Using Amino Acid Sequence Features in a Supervised Learning Framework

BACKGROUND

Protein-Protein Interaction (PPI) has emerged as a key role in the control of many biological processes including protein function, disease incidence, and therapy design. However, the identification of PPI by wet lab experiment is a challenging task, since it is laborious, time consuming and expensive. Therefore, computational prediction of PPI is now given emphasis before going to the experimental validation, since it is simultaneously less laborious, time saver and cost minimizer.

OBJECTIVE

The objective of this study is to develop an improved computational method for PPI prediction mapping on Homo sapiens by using the amino acid sequence features in a supervised learning framework.

METHODS

The experimentally validated 91 positive-PPI pairs of human protein sequences were collected from IntAct Molecular Interaction Database. Then we constructed three balanced datasets with ratios 1:1, 1:2 and 1:3 of positive and negative PPI samples. Then we partitioned each dataset into training (80%) and independent test (20%) datasets. Again each training dataset was partitioned into four mutually exclusive groups of equal sizes for interchanging each group with independent test group to perform 5-fold cross validation (CV). Then we trained candidate seven classifiers (NN, SVM, LR, NB, KNN, AB and RF) with each ratio case to obtain the better PPI predictor by comparing their performance scores.

RESULTS

The random forest (RF) based predictor that was trained with 1:2 ratio of positive-PPI and negative-PPI samples based on AAC encoding features provided the most accurate PPI prediction by producing the highest average performance scores of accuracy (93.50%), sensitivity (95.0%), MCC (85.2%), AUC (0.941) and pAUC (0.236) with the 5-fold cross-validation. It also achieved the highest average performance scores of accuracy (92.0%), sensitivity (94.0%), MCC (83.6%), AUC (0.922) and pAUC (0.207) with the independent test datasets in a comparison of the other candidate and existing predictors.

CONCLUSION

The final resultant prediction strongly recommend that the RF based predictor is a better prediction model of PPI mapping on Homo sapiens.

A prediction nomogram for the 3-year risk of incident diabetes among Chinese adults

Identifying individuals at high risk for incident diabetes could help achieve targeted delivery of interventional programs. We aimed to develop a personalized diabetes prediction nomogram for the 3-year risk of diabetes among Chinese adults. This retrospective cohort study was among 32,312 participants without diabetes at baseline. All participants were randomly stratified into training cohort (n = 16,219) and validation cohort (n = 16,093). The least absolute shrinkage and selection operator model was used to construct a nomogram and draw a formula for diabetes probability. 500 bootstraps performed the receiver operating characteristic (ROC) curve and decision curve analysis resamples to assess the nomogram's determination and clinical use, respectively. 155 and 141 participants developed diabetes in the training and validation cohort, respectively. The area under curve (AUC) of the nomogram was 0.9125 (95% CI, 0.8887-0.9364) and 0.9030 (95% CI, 0.8747-0.9313) for the training and validation cohort, respectively. We used 12,545 Japanese participants for external validation, its AUC was 0.8488 (95% CI, 0.8126-0.8850). The internal and external validation showed our nomogram had excellent prediction performance. In conclusion, we developed and validated a personalized prediction nomogram for 3-year risk of incident diabetes among Chinese adults, identifying individuals at high risk of developing diabetes.

Comparative analysis of screening models for undiagnosed diabetes in Mexico

BACKGROUND

It is estimated that 37% of Mexican adults have undiagnosed diabetes, and are therefore at high risk of developing the severe and devastating complications associated to it. In recent years, a variety of screening tools based on the characteristics of the adult Mexican population have been proposed in order to reduce the negative effects of the disease.

OBJECTIVES

To assess the performance of screening models to diagnose diabetes in the Mexican adult population and to propose a screening model based on HbA1c measurements.

MATERIALS AND METHODS

Data from the 2016 Halfway National Health and Nutrition Survey (NHNS) were used to assess the screening models and to develop and validate the proposed 2016 NHNS model, built using a multivariate logistic regression model. Explanatory variables included in the 2016 NHNS 2016 model were selected through a stepwise backward procedure, using sensitivity and specificity as performance indicators.

RESULTS

Of the screening models assessed, only the model based on the 2006 NHNS survey showed a performance consistent with previous reports. The proposed 2016 NHNS model included age, waist circumference, and systolic blood pressure as explanatory variables and showed a sensitivity of 0.72 and a specificity of 0.80 in the validation data set.

CONCLUSIONS

Age, waist circumference, and systolic blood pressure are variables of special importance for early detection of undiagnosed diabetes in Mexican adults. Based on the consistent performance of the 2006 NHNS model in different data sets, its use as a screening tool for adults with undiagnosed diabetes in Mexico is recommended.

Sensor-Assisted Weighted Average Ensemble Model for Detecting Major Depressive Disorder

The present methods of diagnosing depression are entirely dependent on self-report ratings or clinical interviews. Those traditional methods are subjective, where the individual may or may not be answering genuinely to questions. In this paper, the data has been collected using self-report ratings and also using electronic smartwatches. This study aims to develop a weighted average ensemble machine learning model to predict major depressive disorder (MDD) with superior accuracy. The data has been pre-processed and the essential features have been selected using a correlation-based feature selection method. With the selected features, machine learning approaches such as Logistic Regression, Random Forest, and the proposed Weighted Average Ensemble Model are applied. Further, for assessing the performance of the proposed model, the Area under the Receiver Optimization Characteristic Curves has been used. The results demonstrate that the proposed Weighted Average Ensemble model performs with better accuracy than the Logistic Regression and the Random Forest approaches.

Building Risk Prediction Models for Type 2 Diabetes Using Machine Learning Techniques

Introduction

As one of the most prevalent chronic diseases in the United States, diabetes, especially type 2 diabetes, affects the health of millions of people and puts an enormous financial burden on the US economy. We aimed to develop predictive models to identify risk factors for type 2 diabetes, which could help facilitate early diagnosis and intervention and also reduce medical costs.

Methods

We analyzed cross-sectional data on 138,146 participants, including 20,467 with type 2 diabetes, from the 2014 Behavioral Risk Factor Surveillance System. We built several machine learning models for predicting type 2 diabetes, including support vector machine, decision tree, logistic regression, random forest, neural network, and Gaussian Naive Bayes classifiers. We used univariable and multivariable weighted logistic regression models to investigate the associations of potential risk factors with type 2 diabetes.

Results

All predictive models for type 2 diabetes achieved a high area under the curve (AUC), ranging from 0.7182 to 0.7949. Although the neural network model had the highest accuracy (82.4%), specificity (90.2%), and AUC (0.7949), the decision tree model had the highest sensitivity (51.6%) for type 2 diabetes. We found that people who slept 9 or more hours per day (adjusted odds ratio [aOR] = 1.13, 95% confidence interval [CI], 1.03–1.25) or had checkup frequency of less than 1 year (aOR = 2.31, 95% CI, 1.86–2.85) had higher risk for type 2 diabetes.

Conclusion

Of the 8 predictive models, the neural network model gave the best model performance with the highest AUC value; however, the decision tree model is preferred for initial screening for type 2 diabetes because it had the highest sensitivity and, therefore, detection rate. We confirmed previously reported risk factors and also identified sleeping time and frequency of checkup as 2 new potential risk factors related to type 2 diabetes.

Statistical characterization and classification of colon microarray gene expression data using multiple machine learning paradigms

OBJECTIVE

A colon microarray data is a repository of thousands of gene expressions with different strengths for each cancer cell. It is necessary to detect which genes are responsible for cancer growth. This study presents an exhaustive comparative study of different machine learning (ML) systems which serves two major purposes: (a) identification of high risk differential genes using statistical tests and (b) development of a ML strategy for predicting cancer genes.

METHODS

Four statistical tests namely: Wilcoxon sign rank sum (WCSRS), t test, Kruskal-Wallis (KW), and F-test were adapted for cancerous gene identification using their p-values. The extracted gene set was used to classify cancer patients using ten classifiers namely: linear discriminant analysis (LDA), quadratic discriminant analysis (QDA), naïve Bayes (NB), Gaussian process classification (GPC), support vector machine (SVM), artificial neural network (ANN), logistic regression (LR), decision tree (DT), Adaboost (AB), and random forest (RF). Performance was then evaluated using cross-validation protocols and standardized metrics viz. accuracy (ACC) and area under the curve (AUC).

RESULTS

The colon cancer dataset consists of 2000 genes from 62 patients (40 cancer vs. 22 control). The overall mean ACC of our ML system using all four statistical tests and all ten classifiers was 90.50%. The ML system showed an ACC of 99.81% using a combination WCSRS test and RF-based classifier. This is an improvement of 8% over previously published values in literature.

CONCLUSIONS

RF-based model with statistical tests for detection of high risk genes showed the best performance for accurate cancer classification in multi-center clinical trials.

Diabetes risk scores for Hispanics living in the United States: A systematic review

AIM

Undiagnosed diabetes is more prevalent among racial/ethnic minorities in the United States (U.S.). Despite the proliferation of risk scores, few have been validated in Hispanics populations. The aim of this study is to systematically review published studies that developed risk scores to identify undiagnosed Type 2 Diabetes Mellitus based on self-reported information that were validated for Hispanics in the U.S.

METHODS

The search included PubMed, EMBASE, Cochrane and CINAHL from inception to 2016 without language restrictions. Risk scores whose main outcome was undiagnosed Type 2 diabetes reporting performance measures for Hispanics were included.

RESULTS

We identified three studies that developed and validated risk scores for undiagnosed diabetes based on questionnaire data. Two studies were conducted in Latin America and one in the U.S. All three studies reported adequate performance (area under the receiving curve (AUC) range between0.68and 0.78). The study conducted in the U.S. reported a higher sensitivity of their risk score for Hispanics than whites. The limited number of studies, small size and heterogeneity of the combined cohorts provide limited evidence of the validity of risk scores for Hispanics.

CONCLUSIONS

Efforts to develop and validate risk prediction models in Hispanic populations in the U.S are needed, particularly given the diversity of thisfast growing population. Healthcare professionals should be aware of the limitations of applying risk scores developed for the general population on Hispanics.

Screening models for undiagnosed diabetes in Mexican adults using clinical and self-reported information

BACKGROUND

Prevalence of diabetes in Mexico has constantly increased since 1993. Since type 2 diabetes may remain undiagnosed for many years, identification of subjects at high risk of diabetes is very important to reduce its impact and to prevent its associated complications.

OBJECTIVE

To develop easily implementable screening models to identify subjects with undiagnosed diabetes based on the characteristics of Mexican adults.

SUBJECTS AND METHODS

Screening models were developed using datasets from the 2006 and 2012 National Health and Nutrition Surveys (NHNS). Variables used to develop the multivariate logistic regression models were selected using a backward stepwise procedure. Final models were validated using data from the 2000 National Health Survey (NHS).

RESULTS

The model based on the 2006 NHNS included age, waist circumference, and systolic blood pressure as explanatory variables, while the model based on the 2012 NHNS included age, waist circumference, height, and family history of diabetes. The sensitivity and specificity values obtained from the external validation procedure were 0.74 and 0.62 (2006 NHNS model) and 0.76 and 0.55 (2012 NHNS model) respectively.

CONCLUSIONS

Both models were equally capable of identifying subjects with undiagnosed diabetes (∼75%), and performed satisfactorily when compared to other models developed for other regions or countries.

Accurate Diabetes Risk Stratification Using Machine Learning: Role of Missing Value and Outliers

Diabetes mellitus is a group of metabolic diseases in which blood sugar levels are too high. About 8.8% of the world was diabetic in 2017. It is projected that this will reach nearly 10% by 2045. The major challenge is that when machine learning-based classifiers are applied to such data sets for risk stratification, leads to lower performance. Thus, our objective is to develop an optimized and robust machine learning (ML) system under the assumption that missing values or outliers if replaced by a median configuration will yield higher risk stratification accuracy. This ML-based risk stratification is designed, optimized and evaluated, where: (i) the features are extracted and optimized from the six feature selection techniques (random forest, logistic regression, mutual information, principal component analysis, analysis of variance, and Fisher discriminant ratio) and combined with ten different types of classifiers (linear discriminant analysis, quadratic discriminant analysis, naïve Bayes, Gaussian process classification, support vector machine, artificial neural network, Adaboost, logistic regression, decision tree, and random forest) under the hypothesis that both missing values and outliers when replaced by computed medians will improve the risk stratification accuracy. Pima Indian diabetic dataset (768 patients: 268 diabetic and 500 controls) was used. Our results demonstrate that on replacing the missing values and outliers by group median and median values, respectively and further using the combination of random forest feature selection and random forest classification technique yields an accuracy, sensitivity, specificity, positive predictive value, negative predictive value and area under the curve as: 92.26%, 95.96%, 79.72%, 91.14%, 91.20%, and 0.93, respectively. This is an improvement of 10% over previously developed techniques published in literature. The system was validated for its stability and reliability. RF-based model showed the best performance when outliers are replaced by median values.

Secure and Efficient Regression Analysis Using a Hybrid Cryptographic Framework: Development and Evaluation

Background

Machine learning is an effective data-driven tool that is being widely used to extract valuable patterns and insights from data. Specifically, predictive machine learning models are very important in health care for clinical data analysis. The machine learning algorithms that generate predictive models often require pooling data from different sources to discover statistical patterns or correlations among different attributes of the input data. The primary challenge is to fulfill one major objective: preserving the privacy of individuals while discovering knowledge from data.

Objective

Our objective was to develop a hybrid cryptographic framework for performing regression analysis over distributed data in a secure and efficient way.

Methods

Existing secure computation schemes are not suitable for processing the large-scale data that are used in cutting-edge machine learning applications. We designed, developed, and evaluated a hybrid cryptographic framework, which can securely perform regression analysis, a fundamental machine learning algorithm using somewhat homomorphic encryption and a newly introduced secure hardware component of Intel Software Guard Extensions (Intel SGX) to ensure both privacy and efficiency at the same time.

Results

Experimental results demonstrate that our proposed method provides a better trade-off in terms of security and efficiency than solely secure hardware-based methods. Besides, there is no approximation error. Computed model parameters are exactly similar to plaintext results.

Conclusions

To the best of our knowledge, this kind of secure computation model using a hybrid cryptographic framework, which leverages both somewhat homomorphic encryption and Intel SGX, is not proposed or evaluated to this date. Our proposed framework ensures data security and computational efficiency at the same time.

The role of autonomic dysfunction in predicting 1-year mortality after liver transplantation

BACKGROUND & AIMS

Model for end-stage liver disease (MELD) score has been extensively used to prioritize patients for liver transplantation and determine their prognosis, but with limited predictive value. Autonomic dysfunction may correlate with increased mortality after liver transplant. In this study, two autonomic biomarkers, complexity and deceleration capacity, were added to the predicting model for 1-year mortality after liver transplantation.

METHODS

In all, 30 patients with end-stage liver diseases awaiting liver transplantation were included. Complexity and deceleration capacity were calculated by multi-scale entropy and phase-rectified signal averaging, respectively. Different combinations of autonomic factors and MELD score were used to predict mortality rate of liver transplant after 1-year follow-up. Receiver-operating characteristics curve analysis was performed to determine clinical predictability. Area under the receiver-operating characteristics curve represents the overall accuracy.

RESULTS

The 1-year mortality rate was 16.7% (5/30). The overall accuracy of MELD score used for predicting mortality after liver transplantation was 0.752. By adding complexity and deceleration capacity into the predicting model, the accuracy increased to 0.912. Notably, the accuracy of the prediction using complexity and deceleration capacity alone was 0.912.

CONCLUSION

Complexity and deceleration capacity, which represent different dynamical properties of a human autonomic system, are critical factors for predicting mortality rate of liver transplantation. We recommend that these pre-operative autonomic factors may be helpful as critical adjuncts to predicting model of mortality rate in prioritizing organ allocation.

Veterans' Voice Through the Lens of Their Medical Records: What It Reveals About Congestive Heart Failure Readmissions

PURPOSE OF STUDY

The medical record is a sea of information that can reveal what patients are trying to tell us about their health condition. It can reveal hints and trends as to why veterans with congestive heart failure (CHF) are being readmitted within 30 days after hospital discharge. These hints and trends lead caregivers to key contributing variables to veterans' readmission. Furthermore, these variables can be used to predict patient outcomes such as readmission and even prognosis. This article looks at readmissions for CHF from documentation within the medical record to see what was driving the 30-day readmissions. Second, it examines whether the driving forces can be used to predict a veteran's increased risk for readmission or other poor prognosis.

PRIMARY PRACTICE SETTING(S)

The study was conducted at a rural 84-bed Veterans Health Administration hospital in the Western United States.

METHODOLOGY AND SAMPLE

A retrospective screen was performed on 1,279 veterans' admissions of which 217 were identified as having CHF as a primary or secondary diagnosis on admission. The descriptive statistics, odds ratio (OR) and multivariate logistic regression were used to examine the data. The multivariate logistic regression equation was p = 1/1 + e, which can be found in the biostatistics textbook by . developed and validated the equation and used it to screen for undiagnosed diabetic patients. The equation was refined by . The variables selected for this study were based on a literature review of 30 articles.

RESULTS

The probability and OR for 30-day readmissions for all ages increased as the age increased. The ORs for 30-day readmissions for the variables selected were as follows: brain natriuretic peptide 6.21 (95% CI [0.36, 108.24]), ejection fraction 1.298 (95% CI [0.68, 2.49]), hypertension 1.795 (95% CI [0.83, 3.85]), comorbid conditions 1.02 (95% CI [0.04, 25.02]), Stage III and below were protective, Stage IV 2.057 (95% CI [0.63, 9.32]), lack of discharge education 0.446 (95% CI [0.19, 6.45]). The impact of these variables on veterans with more than 3 readmissions (N = 66) was examined. In 32% of these admissions, there was insufficient data to compare the values of the variables between readmissions. In almost 26% (N = 17) of the cases as the number of variables increased, the time between admissions decreased. In 23% of the cases (N = 15), the values did not change; of these, 14 died and the one who survived had assistance with his care in the form of home health and telehealth.

IMPLICATIONS FOR CASE MANAGEMENT PRACTICE

Use of this evidence-based tool will help case managers to strategically plan care and prioritize interventions to impact the major variables and risk factors that are impacting veterans' health.

Non-invasive detection of fasting blood glucose level via electrochemical measurement of saliva

Machine learning techniques such as logistic regression (LR), support vector machine (SVM) and artificial neural network (ANN) were used to detect fasting blood glucose levels (FBGL) in a mixed population of healthy and diseased individuals in an Indian population. The occurrence of elevated FBGL was estimated in a non-invasive manner from the status of an individual’s salivary electrochemical parameters such as pH, redox potential, conductivity and concentration of sodium, potassium and calcium ions. The samples were obtained from 175 randomly selected volunteers comprising half healthy and half diabetic patients. The models were trained using 70 % of the total data, and tested upon the remaining set. For each algorithm, data points were cross-validated by randomly shuffling them three times prior to implementing the model. The performance of the machine learning technique was reported in terms of four statistically significant parameters—accuracy, precision, sensitivity and F1 score. SVM using RBF kernel showed the best performance for classifying high FBGLs with approximately 85 % accuracy, 84 % precision, 85 % sensitivity and 85 % F₁ score. This study has been approved by the ethical committee of All India Institute of Medical Sciences, New Delhi, India with the reference number: IEC/NP-278/01-08-2014, RP-29/2014.

Private Data Analytics on Biomedical Sensing Data via Distributed Computation

Advances in biomedical sensors and mobile communication technologies have fostered the rapid growth of mobile health (mHealth) applications in the past years. Users generate a high volume of biomedical data during health monitoring, which can be used by the mHealth server for training predictive models for disease diagnosis and treatment. However, the biomedical sensing data raise serious privacy concerns because they reveal sensitive information such as health status and lifestyles of the sensed subjects. This paper proposes and experimentally studies a scheme that keeps the training samples private while enabling accurate construction of predictive models. We specifically consider logistic regression models which are widely used for predicting dichotomous outcomes in healthcare, and decompose the logistic regression problem into small subproblems over two types of distributed sensing data, i.e., horizontally partitioned data and vertically partitioned data. The subproblems are solved using individual private data, and thus mHealth users can keep their private data locally and only upload (encrypted) intermediate results to the mHealth server for model training. Experimental results based on real datasets show that our scheme is highly efficient and scalable to a large number of mHealth users.

A Review of Emerging Technologies for the Management of Diabetes Mellitus

OBJECTIVE

High prevalence of diabetes mellitus (DM) along with the poor health outcomes and the escalated costs of treatment and care poses the need to focus on prevention, early detection and improved management of the disease. The aim of this paper is to present and discuss the latest accomplishments in sensors for glucose and lifestyle monitoring along with clinical decision support systems (CDSSs) facilitating self-disease management and supporting healthcare professionals in decision making.

METHODS

A critical literature review analysis is conducted focusing on advances in: 1) sensors for physiological and lifestyle monitoring, 2) models and molecular biomarkers for predicting the onset and assessing the progress of DM, and 3) modeling and control methods for regulating glucose levels.

RESULTS

Glucose and lifestyle sensing technologies are continuously evolving with current research focusing on the development of noninvasive sensors for accurate glucose monitoring. A wide range of modeling, classification, clustering, and control approaches have been deployed for the development of the CDSS for diabetes management. Sophisticated multiscale, multilevel modeling frameworks taking into account information from behavioral down to molecular level are necessary to reveal correlations and patterns indicating the onset and evolution of DM.

CONCLUSION

Integration of data originating from sensor-based systems and electronic health records combined with smart data analytics methods and powerful user centered approaches enable the shift toward preventive, predictive, personalized, and participatory diabetes care.

SIGNIFICANCE

The potential of sensing and predictive modeling approaches toward improving diabetes management is highlighted and related challenges are identified.

Design an easy-to-use infection screening system for non-contact monitoring of vital-signs to prevent the spread of pandemic diseases

The outbreak of infectious diseases such as influenza, dengue fever, and severe acute respiratory syndrome (SARS) are threatening the global health. Especially, developing countries in the South-East Asia region have been at serious risk. Rapid and highly reliable screening of infection is urgently needed during the epidemic season at mass gathering places, such as airport quarantine facilities, public health centers, and hospital outpatients units, etc. To meet this need, our research group is currently developing a multiple vital-signs based infection screening system that can perform human medical inspections within 15 seconds. This system remotely monitors facial temperature, heart and respiration rates using a thermopile array and a 24-GHz microwave radar, respectively. In this work, we redesigned our previous system to make a higher performance with a user-friendly interface. Moreover, the system newly included a multivariable logistic regression model (MLRM) to determine the possibility of infection. We tested the system on 34 seasonal influenza patients and 35 normal control subjects at the Japan Self-Defense Forces Central Hospital. The sensitivity and specificity of the screening system using the MLRM were 85.3% and 88.6%, respectively.

Application and applicability of non-invasive risk models for predicting undiagnosed prevalent diabetes in Africa: A systematic literature search

BACKGROUND AND PURPOSE

Prediction algorithms are increasingly advocated in diabetes screening strategies, particularly in developing countries. We conducted a systematic review to assess the application and applicability of existing non-invasive prevalent diabetes risk models to populations within Africa.

DESIGN

systematic review data sources A systematic search of English literatures in Medline via PubMed from 1999 until June, 2014. Study selection Included studies had to report on the development, validation or implementation of a model that was primarily constructed to predict prevalent undiagnosed diabetes using non-laboratory based predictors.

DATA EXTRACTION

Data were extracted on the type of statistical model, type and range of predictors in the model, performance measures in both internal and external validation, and whether the model was developed from, validated or implemented in an African population.

RESULTS

Twenty-three studies reporting on non-invasive prevalent diabetes models were identified. Ten from Europe (some with multiethnic populations), nine models were developed among Asian population, two from the USA and two from the Middle-East. The c-statistics for these models ranged from 0.65 to 0.88 in the development studies, and from 0.63 to 0.80 in the validation studies. Twenty models were validated, and none in Africa. Among predictors commonly included in models, parental/family history of diabetes and personal history of hypertension appear to be more prone to measurement errors in the African context.

CONCLUSION

Existing prevalent diabetes prediction models have not been applied to African populations, and issues with the measurement of key predictors make their applicability likely inaccurate.

Type 2 diabetes mellitus

Type 2 diabetes mellitus (T2DM) is an expanding global health problem, closely linked to the epidemic of obesity. Individuals with T2DM are at high risk for both microvascular complications (including retinopathy, nephropathy and neuropathy) and macrovascular complications (such as cardiovascular comorbidities), owing to hyperglycaemia and individual components of the insulin resistance (metabolic) syndrome. Environmental factors (for example, obesity, an unhealthy diet and physical inactivity) and genetic factors contribute to the multiple pathophysiological disturbances that are responsible for impaired glucose homeostasis in T2DM. Insulin resistance and impaired insulin secretion remain the core defects in T2DM, but at least six other pathophysiological abnormalities contribute to the dysregulation of glucose metabolism. The multiple pathogenetic disturbances present in T2DM dictate that multiple antidiabetic agents, used in combination, will be required to maintain normoglycaemia. The treatment must not only be effective and safe but also improve the quality of life. Several novel medications are in development, but the greatest need is for agents that enhance insulin sensitivity, halt the progressive pancreatic β-cell failure that is characteristic of T2DM and prevent or reverse the microvascular complications. For an illustrated summary of this Primer, visit: http://go.nature.com/V2eGfN.

H2RM: A Hybrid Rough Set Reasoning Model for Prediction and Management of Diabetes Mellitus

Diabetes is a chronic disease characterized by high blood glucose level that results either from a deficiency of insulin produced by the body, or the body’s resistance to the effects of insulin. Accurate and precise reasoning and prediction models greatly help physicians to improve diagnosis, prognosis and treatment procedures of different diseases. Though numerous models have been proposed to solve issues of diagnosis and management of diabetes, they have the following drawbacks: (1) restricted one type of diabetes; (2) lack understandability and explanatory power of the techniques and decision; (3) limited either to prediction purpose or management over the structured contents; and (4) lack competence for dimensionality and vagueness of patient’s data. To overcome these issues, this paper proposes a novel hybrid rough set reasoning model (H2RM) that resolves problems of inaccurate prediction and management of type-1 diabetes mellitus (T1DM) and type-2 diabetes mellitus (T2DM). For verification of the proposed model, experimental data from fifty patients, acquired from a local hospital in semi-structured format, is used. First, the data is transformed into structured format and then used for mining prediction rules. Rough set theory (RST) based techniques and algorithms are used to mine the prediction rules. During the online execution phase of the model, these rules are used to predict T1DM and T2DM for new patients. Furthermore, the proposed model assists physicians to manage diabetes using knowledge extracted from online diabetes guidelines. Correlation-based trend analysis techniques are used to manage diabetic observations. Experimental results demonstrate that the proposed model outperforms the existing methods with 95.9% average and balanced accuracies.

Screening for prediabetes and type 2 diabetes in dental offices

Objectives

Most Americans see dentists at least once a year. Chair‐side screening and referral may improve diagnosis of prediabetes and diabetes. In this study, we developed a multivariate model to screen for dysglycemia (prediabetes and diabetes defined as HbA1c ≥5.7 percent) using information readily available to dentists and assessed the prevalence of dysglycemia in general dental practices.

Methods

We recruited 1,033 adults ≥30 years of age without histories of diabetes from 13 general dental practices. A sample of 181 participants selected on the basis of random capillary glucose levels and periodontal status underwent definitive diagnostic testing with hemoglobin A1c. Logistic models were fit to identify risk factors for dysglycemia, and sample weights were applied to estimate the prevalence of dysglycemia in the population ≥30 years of age.

Results

Individuals at high risk for dysglycemia could be identified using a questionnaire that assessed sex, history of hypertension, history of dyslipidemia, history of lost teeth, and either self‐reported body mass index ≥35 kg/m² (severe obesity) or random capillary glucose ≥110 mg/dl. We estimate that 30 percent of patients ≥30 years of age seen in these general dental practices had dysglycemia.

Conclusions

There is a substantial burden of dysglycemia in patients seen in general dental practices. Simple chair‐side screening for dysglycemia that includes or does not include fingerstick random capillary glucose testing can be used to rapidly identify high‐risk patients.

Practical implications

Further studies are needed to demonstrate the acceptability, feasibility, effectiveness, and cost‐effectiveness of chair‐side screening.

Undiagnosed diabetes from cross-sectional GP practice data: an approach to identify communities with high likelihood of undiagnosed diabetes

OBJECTIVES

To estimate undiagnosed diabetes prevalence from general practitioner (GP) practice data and identify areas with high levels of undiagnosed and diagnosed diabetes.

DESIGN

Data from the North-West Adelaide Health Survey (NWAHS) were used to develop a model which predicts total diabetes at a small area. This model was then applied to cross-sectional data from general practices to predict the total level of expected diabetes. The difference between total expected and already diagnosed diabetes was defined as undiagnosed diabetes prevalence and was estimated for each small area. The patterns of diagnosed and undiagnosed diabetes were mapped to highlight the areas of high prevalence.

SETTING

North-West Adelaide, Australia.

PARTICIPANTS

This study used two population samples-one from the de-identified GP practice data (n=9327 active patients, aged 18 years and over) and another from NWAHS (n=4056, aged 18 years and over).

MAIN OUTCOME MEASURES

Total diabetes prevalence, diagnosed and undiagnosed diabetes prevalence at GP practice and Statistical Area Level 1.

RESULTS

Overall, it was estimated that there was one case of undiagnosed diabetes for every 3-4 diagnosed cases among the 9327 active patients analysed. The highest prevalence of diagnosed diabetes was seen in areas of lower socioeconomic status. However, the prevalence of undiagnosed diabetes was substantially higher in the least disadvantaged areas.

CONCLUSIONS

The method can be used to estimate population prevalence of diabetes from general practices wherever these data are available. This approach both flags the possibility that undiagnosed diabetes may be a problem of less disadvantaged social groups, and provides a tool to identify areas with high levels of unmet need for diabetes care which would enable policy makers to apply geographic targeting of effective interventions.

How diabetes risk assessment tools are implemented in practice: a systematic review

This review aimed to explore the extent of the use of diabetes risk assessment tools and to determine influential variables associated with the implementation of these tools. CINAHL, Google Scholar, ISI Citation Indexes, PubMed, and Scopus were searched from inception to January 2013. Studies that reported the use of diabetes risk assessment tools to identify individuals at risk of diabetes were included. Of the 1719 articles identified, 24 were included. Follow-up of high risk individuals for diagnosis of diabetes was conducted in 5 studies. Barriers to the uptake of diabetes risk assessment tools by healthcare practitioners included (1) attitudes toward the tools; (2) impracticality of using the tools and (3) lack of reimbursement and regulatory support. Individuals were reluctant to undertake self-assessment of diabetes risk due to (1) lack of perceived severity of type 2 diabetes; (2) impracticality of the tools; and (3) concerns related to finding out the results. The current use of non-invasive diabetes risk assessment scores as screening tools appears to be limited. Practical follow up systems as well as strategies to address other barriers to the implementation of diabetes risk assessment tools are essential and need to be developed.

Screening for impaired fasting glucose and diabetes using available health plan data

AIMS

To develop and validate prediction equations to identify individuals at high risk for type 2 diabetes using existing health plan data.

METHODS

Health plan data from 2005 to 2009 from 18,527 members of a Midwestern HMO without diabetes, 6% of whom had fasting plasma glucose (FPG) ≥110mg/dL, and health plan data from 2005 to 2006 from 368,025 members of a West Coast-integrated delivery system without diabetes, 13% of whom had FPG ≥110mg/dL were analyzed. Within each health plan, we used multiple logistic regression to develop equations to predict FPG ≥110mg/dL for half of the population and validated the equations using the other half. We then externally validated the equations in the other health plan.

RESULTS

Areas under the curve for the most parsimonious equations were 0.665 to 0.729 when validated internally. Positive predictive values were 14% to 32% when validated internally and 14% to 29% when validated externally.

CONCLUSION

Multivariate logistic regression equations can be applied to existing health plan data to efficiently identify persons at higher risk for dysglycemia who might benefit from definitive diagnostic testing and interventions to prevent or treat diabetes.

Longitudinal evaluation of CA-125 velocity and prediction of ovarian cancer

OBJECTIVE

To determine whether CA-125 velocity is a statistically significant predictor of ovarian cancer and develop a classification rule to screen for ovarian cancer.

METHODS

In the ovarian component of the PLCO cancer screening trial, 28,038 women aged 55-74 had at least two CA-125 screening tests. Ovarian cancer was diagnosed in 72 (0.26%) women. A multiple logistic regression model was developed to evaluate CA-125 velocity and other related covariates as predictors of ovarian cancer. Predictive accuracy was assessed by the concordance index and measures of discrimination and calibration while the fit of the model was assessed by the Hosmer and Lemeshow's goodness-of-fit χ(2)test.

RESULTS

CA-125 velocity decreased as the number of CA-125 measurements increased but was unaffected by age at baseline screen and family history of ovarian cancer. The average velocity (19.749U/ml per month) of the cancer group was more than 500 times the average velocity (0.035U/ml per month) of the non-cancer group.

CONCLUSION

Among six covariates used in the model, CA-125 velocity and time intervals between baseline and second to last screening test and between last two screening tests were statistically significant predictors of ovarian cancer. The chance of having ovarian cancer increased as velocity increased, and the chance decreased when the time intervals between baseline and the second to last screening test and between last two screening tests of an individual increased.

Developing risk prediction models for type 2 diabetes: a systematic review of methodology and reporting

BACKGROUND

The World Health Organisation estimates that by 2030 there will be approximately 350 million people with type 2 diabetes. Associated with renal complications, heart disease, stroke and peripheral vascular disease, early identification of patients with undiagnosed type 2 diabetes or those at an increased risk of developing type 2 diabetes is an important challenge. We sought to systematically review and critically assess the conduct and reporting of methods used to develop risk prediction models for predicting the risk of having undiagnosed (prevalent) or future risk of developing (incident) type 2 diabetes in adults.

METHODS

We conducted a systematic search of PubMed and EMBASE databases to identify studies published before May 2011 that describe the development of models combining two or more variables to predict the risk of prevalent or incident type 2 diabetes. We extracted key information that describes aspects of developing a prediction model including study design, sample size and number of events, outcome definition, risk predictor selection and coding, missing data, model-building strategies and aspects of performance.

RESULTS

Thirty-nine studies comprising 43 risk prediction models were included. Seventeen studies (44%) reported the development of models to predict incident type 2 diabetes, whilst 15 studies (38%) described the derivation of models to predict prevalent type 2 diabetes. In nine studies (23%), the number of events per variable was less than ten, whilst in fourteen studies there was insufficient information reported for this measure to be calculated. The number of candidate risk predictors ranged from four to sixty-four, and in seven studies it was unclear how many risk predictors were considered. A method, not recommended to select risk predictors for inclusion in the multivariate model, using statistical significance from univariate screening was carried out in eight studies (21%), whilst the selection procedure was unclear in ten studies (26%). Twenty-one risk prediction models (49%) were developed by categorising all continuous risk predictors. The treatment and handling of missing data were not reported in 16 studies (41%).

CONCLUSIONS

We found widespread use of poor methods that could jeopardise model development, including univariate pre-screening of variables, categorisation of continuous risk predictors and poor handling of missing data. The use of poor methods affects the reliability of the prediction model and ultimately compromises the accuracy of the probability estimates of having undiagnosed type 2 diabetes or the predicted risk of developing type 2 diabetes. In addition, many studies were characterised by a generally poor level of reporting, with many key details to objectively judge the usefulness of the models often omitted.

Modulating effect of atorvastatin on paraoxonase 1 activity in type 2 diabetic Egyptian patients with or without nephropathy

The aim of this study was to investigate the modulating effect of atorvastatin on serum paraoxonase 1 enzyme (PON1) activity in type 2 diabetic Egyptian patients with or without nephropathy. The present study was carried out on the following groups: control group, which consisted of 30 healthy persons; Group I, which consisted of 20 type 2 diabetic patients without nephropathy; and Group II, which consisted of 20 type 2 diabetic patients with nephropathy. All the patients selected were under an antidiabetic regimen of insulin, and patients receiving antihypertensive agents were excluded from the follow-up study to avoid drug interaction fallacies. Twenty-two patients (15 without nephropathy and seven with nephropathy) received atorvastatin in individually adjusted oral dosage (range 10-20 mg) once per day for 12 weeks. All cases were subjected to thorough clinical examination and history taking and measurement of serum levels of PON1 activity, malondialdehyde (MDA), glutathione reductase activity, fasting glucose, total cholesterol, triglycerides, high-density lipoprotein (HDL), low-density lipoprotein (LDL), urea, and creatinine. Urine samples were collected for determination of proteinuria. The obtained results showed that PON1 activity and HDL significantly decreased and fasting glucose significantly increased in Group I and Group II when compared to the control group, with significant difference in their levels between Group II and Group I. MDA, total cholesterol, and LDL levels significantly increased and glutathione reductase activity significantly decreased in Group I and Group II when compared to the control group. Urea, creatinine, and proteinuria levels showed significant increase in Group II when compared to the control group and Group I, with nonsignificant difference between control group and Group I. Atorvastatin therapy caused a significant increase in PON1 activity, and serum levels of MDA and glutathione reductase activity were significantly decreased and increased, respectively. Also, total cholesterol, triglyceride and LDL-cholesterol levels were significantly reduced with a significant increase in HDL-cholesterol levels. There was a significant modest reduction in serum urea and creatinine levels as well as in proteinuria level. Fasting glucose level was significantly reduced under the antidiabetic regimen of insulin through the follow-up period. PON1 activity showed a significant negative correlation with glucose and LDL, and a significant positive correlation with HDL in all the studied groups. It could be concluded that atorvastatin with its pleiotropic effects could provide optimal therapeutic intervention to control not only dyslipidemia, but also oxidative stress status with consequent improvement in the course of type 2 diabetes and diabetic nephropathy. More specifically, restoration of PON1 activity by atorvastatin opens a window to investigate other drugs that could provide a new adjuvant therapeutic line for better control of diabetes and diabetic nephropathy. Further research is also recommended to study the distribution of PON1 genetic polymorphism among the Egyptian population to explain the variability in its activity and its relationship with other factors that associate diabetes and its complications.

Validation of a type 2 diabetes screening tool in rural Honduras

OBJECTIVE

To validate a low-cost tool for identifying diabetic patients in rural areas of Latin America.

RESEARCH DESIGN AND METHODS

A regression equation incorporating postprandial time and a random plasma glucose was used to screen 800 adults in Honduras. Patients with a probability of diabetes of > or =20% were asked to return for a fasting plasma glucose (FPG). A random fifth of those with a screener-based probability of diabetes <20% were also asked to return for follow-up. The gold standard was an FPG > or =126 mg/dl.

RESULTS

The screener had very good test characteristics (area under the receiver operating characteristic curve = 0.89). Using the screening criterion of > or =0.42, the equation had a sensitivity of 74.1% and specificity of 97.2%.

CONCLUSIONS

This screener is a valid measure of diabetes risk in Honduras and could be used to identify diabetic patients in poor clinics in Latin America.

The clinical use of hemoglobin A1c

Hemoglobin A1c (HbA1c) has been accepted as an index of glycemic control since the mid-1970s and is the best marker for diabetic microvascular complications. Clinically, it is now used to assess glycemic control in people with diabetes. Assays are most reliable when certified by the National Hemoglobin Standardization Program but are subject to confounders and effect modifiers, particularly in the setting of hematologic abnormalities. Other measures of chronic glycemic control-fructosamine and 1,5-anhydroglucitol-are far less widely used. The relationship of HbA1c to average blood glucose was intensively studied recently, and it has been proposed that this conversion can be used to report an "estimated average glucose, eAG" in milligrams/deciliter or millimolar units rather than as per cent glycated hemoglobin. Finally, HbA1c has been proposed as a useful method of screening for and diagnosing diabetes.

Prediction of periventricular leukomalacia. Part I: Selection of hemodynamic features using logistic regression and decision tree algorithms

OBJECTIVE

Periventricular leukomalacia (PVL) is part of a spectrum of cerebral white matter injury which is associated with adverse neurodevelopmental outcome in preterm infants. While PVL is common in neonates with cardiac disease, both before and after surgery, it is less common in older infants with cardiac disease. Pre-, intra-, and postoperative risk factors for the occurrence of PVL are poorly understood. The main objective of the present work is to identify potential hemodynamic risk factors for PVL occurrence in neonates with complex heart disease using logistic regression analysis and decision tree algorithms.

METHODS

The postoperative hemodynamic and arterial blood gas data (monitoring variables) collected in the cardiac intensive care unit of Children's Hospital of Philadelphia were used for predicting the occurrence of PVL. Three categories of datasets for 103 infants and neonates were used-(1) original data without any preprocessing, (2) partial data keeping the admission, the maximum and the minimum values of the monitoring variables, and (3) extracted dataset of statistical features. The datasets were used as inputs for forward stepwise logistic regression to select the most significant variables as predictors. The selected features were then used as inputs to the decision tree induction algorithm for generating easily interpretable rules for prediction of PVL.

RESULTS

Three sets of data were analyzed in SPSS for identifying statistically significant predictors (p<0.05) of PVL through stepwise logistic regression and their correlations. The classification success of the Case 3 dataset of extracted statistical features was best with sensitivity (SN), specificity (SP) and accuracy (AC) of 87, 88 and 87%, respectively. The identified features, when used with decision tree algorithms, gave SN, SP and AC of 90, 97 and 94% in training and 73, 58 and 65% in test. The identified variables in Case 3 dataset mainly included blood pressure, both systolic and diastolic, partial pressures pO(2) and pCO(2), and their statistical features like average, variance, skewness (a measure of asymmetry) and kurtosis (a measure of abrupt changes). Rules for prediction of PVL were generated automatically through the decision tree algorithms.

CONCLUSIONS

The proposed approach combines the advantages of statistical approach (regression analysis) and data mining techniques (decision tree) for generation of easily interpretable rules for PVL prediction. The present work extends an earlier research [Galli KK, Zimmerman RA, Jarvik GP, Wernovsky G, Kuijpers M, Clancy RR, et al. Periventricular leukomalacia is common after cardiac surgery. J Thorac Cardiovasc Surg 2004;127:692-704] in the form of expanding the feature set, identifying additional prognostic factors (namely pCO(2)) emphasizing the temporal variations in addition to upper or lower values, and generating decision rules. The Case 3 dataset was further investigated in Part II for feature selection through computational intelligence.

A new look at screening and diagnosing diabetes mellitus

OBJECTIVE

Diabetes is underdiagnosed. About one third of people with diabetes do not know they have it, and the average lag between onset and diagnosis is 7 yr. This report reconsiders the criteria for diagnosing diabetes and recommends screening criteria to make case finding easier for clinicians and patients.

PARTICIPANTS

R.M.B. invited experts in the area of diagnosis, monitoring, and management of diabetes to form a panel to review the literature and develop consensus regarding the screening and diagnosis of diabetes with particular reference to the use of hemoglobin A1c (HbA1c). Participants met in open session and by E-mail thereafter. Metrika, Inc. sponsored the meeting.

EVIDENCE

A literature search was performed using standard search engines.

CONSENSUS PROCESS

The panel heard each member's discussion of the issues, reviewing evidence prior to drafting conclusions. Principal conclusions were agreed on, and then specific cut points were discussed in an iterative consensus process.

CONCLUSIONS

The main factors in support of using HbA1c as a screening and diagnostic test include: 1) HbA1c does not require patients to be fasting; 2) HbA1c reflects longer-term glycemia than does plasma glucose; 3) HbA1c laboratory methods are now well standardized and reliable; and 4) errors caused by nonglycemic factors affecting HbA1c such as hemoglobinopathies are infrequent and can be minimized by confirming the diagnosis of diabetes with a plasma glucose (PG)-specific test. Specific recommendations include: 1) screening standards should be established that prompt further testing and closer follow-up, including fasting PG of 100 mg/dl or greater, random PG of 130 mg/dl or greater, or HbA1c greater than 6.0%; 2) HbA1c of 6.5-6.9% or greater, confirmed by a PG-specific test (fasting plasma glucose or oral glucose tolerance test), should establish the diagnosis of diabetes; and 3) HbA1c of 7% or greater, confirmed by another HbA1c- or a PG-specific test (fasting plasma glucose or oral glucose tolerance test) should establish the diagnosis of diabetes. The recommendations are offered for consideration of the clinical community and interested associations and societies.

A simple risk score identifies individuals at high risk of developing Type 2 diabetes: a prospective cohort study

BACKGROUND

Randomized trials have demonstrated that Type 2 diabetes is preventable among high-risk individuals. To date, such individuals have been identified through population screening using the oral glucose tolerance test.

OBJECTIVE

To assess whether a risk score comprising only routinely collected non-biochemical parameters was effective in identifying those at risk of developing Type 2 diabetes.

METHODS

Population-based prospective cohort (European Prospective Investigation of Cancer-Norfolk). Participants aged 40-79 recruited from UK general practices attended a health check between 1993 and 1998 (n = 25 639) and were followed for a mean of 5 years for diabetes incidence. The Cambridge Diabetes Risk Score was computed for 24 495 individuals with baseline data on age, sex, prescription of steroids and anti-hypertensive medication, family history of diabetes, body mass index and smoking status. We examined the incidence of diabetes across quintiles of the risk score and plotted a receiver operating characteristic (ROC) curve to assess discrimination.

RESULTS

There were 323 new cases of diabetes, a cumulative incidence of 2.76/1000 person-years. Those in the top quintile of risk were 22 times more likely to develop diabetes than those in the bottom quintile (odds ratio 22.3; 95% CI: 11.0-45.4). In all, 54% of all clinically incident cases occurred in individuals in the top quintile of risk (risk score > 0.37). The area under the ROC was 74.5%.

CONCLUSION

The risk score is a simple, effective tool for the identification of those at risk of developing Type 2 diabetes. Such methods may be more feasible than mass population screening with biochemical tests in defining target populations for prevention programmes.

Random plasma glucose in serendipitous screening for glucose intolerance: screening for impaired glucose tolerance study 2

BACKGROUND

With positive results from diabetes prevention studies, there is interest in convenient ways to incorporate screening for glucose intolerance into routine care and to limit the need for fasting diagnostic tests.

OBJECTIVE

The aim of this study is to determine whether random plasma glucose (RPG) could be used to screen for glucose intolerance.

DESIGN

This is a cross-sectional study.

PARTICIPANTS

The participants of this study include a voluntary sample of 990 adults not known to have diabetes.

MEASUREMENTS

RPG was measured, and each subject had a 75-g oral glucose tolerance test several weeks later. Glucose intolerance targets included diabetes, impaired glucose tolerance (IGT), and impaired fasting glucose(110) (IFG(110); fasting glucose, 110-125 mg/dl, and 2 h glucose < 140 mg/dl). Screening performance was measured by area under receiver operating characteristic curves (AROC).

RESULTS

Mean age was 48 years, and body mass index (BMI) was 30.4 kg/m(2); 66% were women, and 52% were black; 5.1% had previously unrecognized diabetes, and 24.0% had any "high-risk" glucose intolerance (diabetes or IGT or IFG(110)). The AROC was 0.80 (95% CI 0.74-0.86) for RPG to identify diabetes and 0.72 (0.68-0.75) to identify any glucose intolerance, both highly significant (p < 0.001). Screening performance was generally consistent at different times of the day, regardless of meal status, and across a range of risk factors such as age, BMI, high density lipoprotein cholesterol, triglycerides, and blood pressure.

CONCLUSIONS

RPG values should be considered by health care providers to be an opportunistic initial screening test and used to prompt further evaluation of patients at risk of glucose intolerance. Such "serendipitous screening" could help to identify unrecognized diabetes and prediabetes.

Diabetes in the dental office: using NHANES III to estimate the probability of undiagnosed disease

BACKGROUND AND OBJECTIVE

Recent data have suggested that in the past 15 years there has been a dramatic increase in the incidence of diabetes mellitus in the USA. However, evidence suggests that approximately one-third of diabetes cases remain undiagnosed. Because 60% of Americans see a dentist at least once per year for routine, nonemergent, care, it is reasonable to propose that the dental office can be a healthcare location actively involved in screening for unidentified diabetes.

MATERIAL AND METHODS

This study used NHANES III to develop a predictive equation that can form the basis of a tool to help dentists determine the probability of undiagnosed diabetes by using self-reported data and periodontal clinical parameters routinely assessed in the dental office.

RESULTS

Our analyses reveal that individuals with a self-reported family history of diabetes, hypertension, high cholesterol levels and clinical evidence of periodontal disease bear a probability of 27-53% of having undiagnosed diabetes, with Mexican-American men exhibiting the highest probability and white women the lowest.

CONCLUSION

These findings suggest that the dental office could provide an important opportunity to identify individuals unaware of their diabetic status.

Individualized pharmacokinetic risk assessment for development of diabetes in high risk population

AIM

The objective of this study is to propose a non-parametric pharmacokinetic prediction model that addresses the individualized risk of developing type-2 diabetes in subjects with family history of type-2 diabetes.

METHOD

All selected 191 healthy subjects had both parents as type-2 diabetic. Glucose was administered intravenously (0.5 g/kg body weight) and 13 blood samples taken at specified times were analyzed for plasma insulin and glucose concentrations. All subjects were followed for an average of 13-14 years for diabetic or normal (non-diabetic) outcome.

RESULTS

The new logistic regression model predicts the development of diabetes based on body mass index and only one blood sample at 90 min analyzed for insulin concentration. Our model correctly identified 4.5 times more subjects (54% versus 11.6%) predicted to develop diabetes and more than twice the subjects (99% versus 46.4%) predicted not to develop diabetes compared to current non-pharmacokinetic probability estimates for development of type-2 diabetes.

CONCLUSION

Our model can be useful for individualized prediction of development of type-2 diabetes in subjects with family history of type-2 diabetes. This improved prediction may be an important mediating factor for better perception of risk and may result in an improved intervention.

Sequential screening for diabetes—evaluation of a campaign in Swiss community pharmacies

OBJECTIVE

For early detection of persons at risk for type 2 diabetes, a combination of risk factor assessment and glucose measurement could be a promising approach and an opportunity for health promotion. The object of this study was to develop a sequential screening concept and to evaluate it in a national pharmacy based screening campaign.

METHOD

Community pharmacies of the German speaking part of Switzerland participating in the national Self Care campaign "Stop diabetes--test now" offered a free of charge "sequential screening" with (a) diabetes risk assessment, (b) consecutive capillary blood glucose measurement and (c) assessment of the motivation for lifestyle change based on the Transtheoretical Model (TTM) of behaviour change. A 35 items data sheet served as a structured screening protocol and enabled quick and reliable documentation of all relevant data. Outcomes measures were: age, sex, cigarette smoking, total score of the ADA diabetes risk-factor questionnaire, family history of diabetes, body mass index, insufficient physical activity, blood pressure, capillary blood glucose, motivation for lifestyle change, counselling activities and triage decisions of the pharmacy team.

RESULTS

During the 5 weeks of spring 2002, 530 pharmacies screened a total of 93,258 persons (33.1% male, mean age 60.9 years +/- 14.1 (SD)). Risk profile: family history of diabetes 26.4%; BMI > or = 25 kg/m(2) 49.3%; low physical activity 27.2%; elevated blood pressure 45.7%. Stratification into risk groups: < 2 risk factors 21.6%; > or = 2 risk factors 71.5%; borderline glycaemia (FG 5.3-6.1 mmol/l, confirmed in a second measurement) 2.5% and hyperglycaemia (FG > or = 6.1 mmol/1 or NFG > or = 11.1 mmol/1) 4.4%. Of all persons screened, 6.4% were referred to a physician and 73.7% got targeted advice with respect to physical activity and/or nutrition based on their specific risk profile.

CONCLUSION

The screening campaign attracted an important part of Swiss German speaking adults (2.4%). The sequential screening could successfully be implemented into pharmacy practice. Of the generally elderly persons screened, 6.9% were detected with suspicion for diabetes type 2 and 71.5% had at least two risk factors. This provided an opportunity to initiate targeted counselling regarding therapeutic lifestyle change.

Risk scores for type 2 diabetes can be applied in some populations but not all

OBJECTIVE

Risk scores based on phenotypic characteristics to identify individuals at high risk of having undiagnosed diabetes have been developed in Caucasian populations. The impact of known risk factors on having undiagnosed type 2 diabetes differs between populations from different ethnic origin, and risk scores developed in Caucasians may not be applicable to other ethnic groups. This study evaluated the performance of one risk score in nine populations of diverse ethnic origin.

RESEARCH DESIGN AND METHODS

Data provided by centers from around the world to the DETECT-2 project were used. The database includes population-based surveys with information on at least 500 people without known diabetes having a 75-g oral glucose tolerance test. To date, 52 centers have contributed data on 190,000 individuals from 34 countries. In this analysis, nine cross-sectional studies were selected representing diverse ethnic and regional backgrounds. The risk score assessed uses information on age, sex, blood pressure treatment, and BMI.

RESULTS

This analysis included 29,758 individuals; 1,805 individuals had undiagnosed diabetes. The performance of the risk score varied widely, with sensitivity, specificity, and percentage needing further testing ranging between 12 and 57%, 72 and 93%, and 2 and 25%, respectively, with the worse performance in non-Caucasian populations. This variation in performance was related to differences in the association between prevalence of undiagnosed diabetes and components of the risk score.

CONCLUSIONS

A typical risk score developed in Caucasian populations cannot be applied to other populations of diverse ethnic origins.