Data mining approach to policy analysis in a health insurance domain

The application of artificial intelligence in health policy: a scoping review

BACKGROUND

Policymakers require precise and in-time information to make informed decisions in complex environments such as health systems. Artificial intelligence (AI) is a novel approach that makes collecting and analyzing data in complex systems more accessible. This study highlights recent research on AI's application and capabilities in health policymaking.

METHODS

We searched PubMed, Scopus, and the Web of Science databases to find relevant studies from 2000 to 2023, using the keywords "artificial intelligence" and "policymaking." We used Walt and Gilson's policy triangle framework for charting the data.

RESULTS

The results revealed that using AI in health policy paved the way for novel analyses and innovative solutions for intelligent decision-making and data collection, potentially enhancing policymaking capacities, particularly in the evaluation phase. It can also be employed to create innovative agendas with fewer political constraints and greater rationality, resulting in evidence-based policies. By creating new platforms and toolkits, AI also offers the chance to make judgments based on solid facts. The majority of the proposed AI solutions for health policy aim to improve decision-making rather than replace experts.

CONCLUSION

Numerous approaches exist for AI to influence the health policymaking process. Health systems can benefit from AI's potential to foster the meaningful use of evidence-based policymaking.

The application of artificial intelligence in health financing: a scoping review

INTRODUCTION

Artificial Intelligence (AI) represents a significant advancement in technology, and it is crucial for policymakers to incorporate AI thinking into policies and to fully explore, analyze and utilize massive data and conduct AI-related policies. AI has the potential to optimize healthcare financing systems. This study provides an overview of the AI application domains in healthcare financing.

METHOD

We conducted a scoping review in six steps: formulating research questions, identifying relevant studies by conducting a comprehensive literature search using appropriate keywords, screening titles and abstracts for relevance, reviewing full texts of relevant articles, charting extracted data, and compiling and summarizing findings. Specifically, the research question sought to identify the applications of artificial intelligence in health financing supported by the published literature and explore potential future applications. PubMed, Scopus, and Web of Science databases were searched between 2000 and 2023.

RESULTS

We discovered that AI has a significant impact on various aspects of health financing, such as governance, revenue raising, pooling, and strategic purchasing. We provide evidence-based recommendations for establishing and improving the health financing system based on AI.

CONCLUSIONS

To ensure that vulnerable groups face minimum challenges and benefit from improved health financing, we urge national and international institutions worldwide to use and adopt AI tools and applications.

An Efficient Healthcare Data Mining Approach Using Apriori Algorithm: A Case Study of Eye Disorders in Young Adults

In the public health sector and the field of medicine, the popularity of data mining and its usage in knowledge discovery and databases (KDD) are rising. The growing popularity of data mining has discovered innovative healthcare links to support decision making. For this reason, there is a great possibility to better diagnose patient’s diseases and maintain the quality of healthcare services in hospitals. So, there is an urgent need to make disease diagnosis possible by discovering the hidden patterns from the patients’ history information in developing countries. This work is a step towards how to use the extracted knowledge to enhance the quality of healthcare facilities. In this paper, we have proposed a web-centered hospital information management system (HIMS) that identifies frequent patterns from the data with eye disorder patients using the association rule-based Apriori data mining technique. The proposed framework has the capability to overcome all the key issues and problems in the current hospital information management system regarding data analysis and reporting services. For this purpose, data were collected from more than 1000 university students (China citizens) both online and manually (printed questionnaire). After applying the Apriori algorithm on the collected data, we revealed that almost 140 individuals out of 1035 had myopia (near-sighted disorder), at current age of 22 years, and that there were no male patients found with myopia. We concluded that their clinical relevance and utility can generate favorable results from prospective clinical studies by mapping out the habits or lifestyles that potentially lead to fatal diseases. In the future, we plan to extend this work to fully automate HIMS to help practitioners to diagnose the reasons of various diseases by extracting patient lifestyle patterns.

Automated Detection of Poor-Quality Scintigraphic Images Using Machine Learning

Objective  In the present study, we have used machine learning algorithm to accomplish the task of automated detection of poor-quality scintigraphic images. We have validated the accuracy of our machine learning algorithm on 99m Tc-methyl diphosphonate ( 99m Tc-MDP) bone scan images. Materials and Methods  Ninety-nine patients underwent 99mTC-MDP bone scan acquisition twice at two different acquisition speeds, one at low speed and another at double the speed of the first scan, with patient lying in the same position on the scan table. The low-speed acquisition resulted in good-quality images and the high-speed acquisition resulted in poor-quality images. The principal component analysis (PCA) of all the images was performed and the first 32 principal components (PCs) were retained as feature vectors of the image. These 32 feature vectors of each image were used for the classification of images into poor or good quality using machine learning algorithm (multivariate adaptive regression splines [MARS]). The data were split into two sets, that is, training set and test set in the ratio of 60:40. Hyperparameter tuning of the model was done in which five-fold cross-validation was performed. Receiver operator characteristic (ROC) analysis was used to select the optimal model using the largest value of area under the ROC curve. Sensitivity, specificity, and accuracy for the classification of poor- and good-quality images were taken as metrics for the performance of the algorithm. Result  Accuracy, sensitivity, and specificity of the model in classifying poor-quality and good-quality images were 93.22, 93.22, and 93.22%, respectively, for the training dataset and 86.88, 80, and 93.7%, respectively, for the test dataset. Conclusion  Machine learning algorithms can be used to classify poor- and good-quality images with good accuracy (86.88%) using 32 PCs as the feature vector and MARS as the classification model.

Predicting Online Complaining Behavior in the Hospitality Industry: Application of Big Data Analytics to Online Reviews

Purpose: This study aims to enrich the published literature on hospitality and tourism by applying big data analytics and data mining algorithms to predict travelers’ online complaint attributions to significantly different hotel classes (i.e., higher star-rating and lower star-rating). Design/methodology/approach: First, 1992 valid online complaints were manually obtained from over 350 hotels located in the UK. The textual data were converted into structured data by utilizing content analysis. Ten complaint attributes and 52 items were identified. Second, a two-step analysis approach was applied via data-mining algorithms. For this study, sensitivity analysis was conducted to identify the most important online complaint attributes, then decision tree models (i.e., the CHAID algorithm) were implemented to discover potential relationships that might exist between complaint attributes in the online complaining behavior of guests from different hotel classes. Findings: Sensitivity analysis revealed that Hotel Size is the most important online complaint attribute, while Service Encounter and Room Space emerged as the second and third most important factors in each of the four decision tree models. The CHAID analysis findings also revealed that guests at higher-star-rating hotels are most likely to leave online complaints about (i) Service Encounter, when staying at large hotels; (ii) Value for Money and Service Encounter, when staying at medium-sized hotels; (iii) Room Space and Service Encounter, when staying at small hotels. Additionally, the guests of lower-star-rating hotels are most likely to write online complaints about Cleanliness, but not Value for Money, Room Space, or Service Encounter, and to stay at small hotels. Practical implications: By utilizing new data-mining algorithms, more profound findings can be discovered and utilized to reinforce the strengths of hotel operations to meet the expectations and needs of their target guests. Originality/value: The study’s main contribution lies in the utilization of data-mining algorithms to predict online complaining behavior between different classes of hotel guests.

Development of eClaim system for private indemnity health insurance in South Korea: Compatibility and interoperability

South Korea has the mandatory National Health Insurance (NHI) and supplemental Private Indemnity Health Insurance (PIHI). According to the Financial Supervisory Service, the share of the population with PIHI increased to 66% in 2018 due to the financial burden. However, since the traditional PIHI claim workflow is based on the paper attachment method, it is a big burden to every stakeholder and limits the usability and accessibility of the claims data. To improve the traditional PIHI claim workflow, we developed the electronic claim (eClaim) service for the PIHI in Korea. We also applied the HL7^® (Health Level Seven) FHIR^® (Fast Healthcare Interoperability Resources) standard to ensure interoperability of the claims data. The proposed eClaim Service has been launched in 2017. It has been increased from 8155 in the first half of 2018 to 114,087 in the second half of 2020. Currently, 60 healthcare providers and 22 payers participated in this service. In this study, we proposed an eClaim workflow and service to improve the legacy system. The proposed method can be helpful to other entities planning for their own health insurance system and also applied to various practical purposes including value-based care, automated claim review, and clinical research.

Data mining model for predicting the quality level and classification of construction projects

Project managers supervise projects to ensure their smooth completion within a stipulated time frame and budget while guaranteeing construction quality. The relationships of various attributes with quality can be quantified and classified to facilitate such supervision. Therefore, this study used a data mining algorithm to analyze the relationships between defects, quality levels, contract sums, project categories, and progress in 1,015 inspection projects. In the first part, association rule mining (ARM), an unsupervised data mining approach, was used to obtain 11 rules relating two defect types (i.e., quality management system and construction quality) and determine the relationships between the four attributes (i.e., quality level, contract sum, project category, and progress). The resulting association rule may be beneficial for construction management because project managers can use it to determine the correlations between defects and attributes. In the second part, supervised data mining techniques, namely neural network (NN), support vector machine (SVM), and decision tree (C5.0 and QUEST) algorithms, were applied to develop a classification model for quality prediction. The target variable was quality, which was divided into four levels, and the decision variables comprised 499 defects, 3 contract sums, 7 project categories, and 2 progress variables. The results indicated that five defects were important. Finally, the four indicators of gain chart, break-even point (BEP), accuracy, and area under the curve (AUC) were calculated to evaluate the model. For the SVM model, the actual value predicted by the gain chart was 96.04%, the BEP was 0.95, and the AUC was 0.935. The SVM yielded optimal classification efficiency and effectively predicted the quality level. The data mining model developed in this study can serve as a reference for effective construction management.

Application of Cognitive Automation to Structuring Data, Driving Existing Business Models, and Creating Value between Legacy Industries

To build high quality datasets and unlock the value of unstructured data, a systematic approach for data capture is necessary. Cognitive automation (CA), that is, automation of processes with artificial intelligence (AI), enables the information extraction from unstructured data to provide relevant insights and further processing with AI. This study provides an overview of this new technology and shows how it can be used to transform existing business models. Our case studies in the insurance auditing, healthcare, and banking industries show the potential managerial impact of CA, which prepares these legacy industries for their digital future’s challenges and opportunities. We present the novel data extraction pipeline for textual and visual data and demonstrate its efficiency in extracting information from the company’s unstructured data. We show its performance in quality, cost, and time compared with current industry standards and provide management insights for business applications using CA.

Public needs for information disclosure on healthcare performance: Different determinants between Japan and the Netherlands

The accumulated healthcare performance data related to unwarranted practice variations are not necessarily disseminated to patients and citizens. To clarify the needs for public disclosure, we explored Japanese and Dutch citizens' preferences and values towards information disclosure and healthcare disparity.Online opt-in survey was conducted and we asked citizens their preference to know about the healthcare performance indicators of regions and hospitals, and their attitudes towards healthcare equity. After a descriptive statistical analysis, Chi-squared automatic interaction detection tree analysis was performed to explore the socio-demographic determinants which were associated with positive value for information disclosure and healthcare equity. Then, we compared the combination of attributes of the highest and the lowest subgroups of each country and compared within and between countries. Last, logistic regression analysis was performed to further evaluate the impact of each determinant.Significant differences were observed between the 2 countries (Japan [JPN] 1038; Netherlands [NL] 1040). The crucial attributes identified were age, sex, educational background, and living area (JPN), along with age and sex (NL). Japanese comprised multiple subgroups with heterogeneous values, showed relatively low interest in knowing the information, and seemed to accept healthcare inequality, especially among urban males aged 20 to 59 years. Contrarily, Dutch people mostly showed high interest in both items. Female and older respondents valued information disclosure highly across countries.To share healthcare performance knowledge and empowering the public, historical, cultural, and socio-demographic context including health literacy of citizens' subgroups should be considered in making comprehensive public reports.

Non-Traditional Data Mining Applications in Taiwan National Health Insurance (NHI) Databases

This study examines time-sensitive applications of data mining methods to facilitate claims review processing and provide policy information for insurance decision-making vis-à-vis the Taiwan National Health Insurance (NHI) databases. In order to obtain the best payment management, a hybrid mining (HM) approach, which has been grounded on the extant knowledge of data mining projects and health insurance domain knowledge, is proposed. Through the integration of data warehousing, online analytic processing, data mining techniques and traditional data analysis in the healthcare field, an easy-to-use decision support platform, which will assist in directing the health insurance decision-making process, is built. Drawing from lessons learned within a case study setting, results showed that not only is HM approach a reliable, powerful, and user-friendly platform for diversified payment decision support, but that it also has great relevance for the practice and acceptance of evidence-based medicine. Essentially, HM approach can provide a critical boost to health insurance decision support; hence, future researchers should develop and improve the approach combined with their own application systems.

Comprehensible knowledge model creation for cancer treatment decision making

BACKGROUND

A wealth of clinical data exists in clinical documents in the form of electronic health records (EHRs). This data can be used for developing knowledge-based recommendation systems that can assist clinicians in clinical decision making and education. One of the big hurdles in developing such systems is the lack of automated mechanisms for knowledge acquisition to enable and educate clinicians in informed decision making.

MATERIALS AND METHODS

An automated knowledge acquisition methodology with a comprehensible knowledge model for cancer treatment (CKM-CT) is proposed. With the CKM-CT, clinical data are acquired automatically from documents. Quality of data is ensured by correcting errors and transforming various formats into a standard data format. Data preprocessing involves dimensionality reduction and missing value imputation. Predictive algorithm selection is performed on the basis of the ranking score of the weighted sum model. The knowledge builder prepares knowledge for knowledge-based services: clinical decisions and education support.

RESULTS

Data is acquired from 13,788 head and neck cancer (HNC) documents for 3447 patients, including 1526 patients of the oral cavity site. In the data quality task, 160 staging values are corrected. In the preprocessing task, 20 attributes and 106 records are eliminated from the dataset. The Classification and Regression Trees (CRT) algorithm is selected and provides 69.0% classification accuracy in predicting HNC treatment plans, consisting of 11 decision paths that yield 11 decision rules.

CONCLUSION

Our proposed methodology, CKM-CT, is helpful to find hidden knowledge in clinical documents. In CKM-CT, the prediction models are developed to assist and educate clinicians for informed decision making. The proposed methodology is generalizable to apply to data of other domains such as breast cancer with a similar objective to assist clinicians in decision making and education.

Evaluation of Adherence to Nutritional Intervention Through Trajectory Analysis

Classical pre-post intervention studies are often analyzed using traditional statistics. Nevertheless, the nutritional interventions have small effects on the metabolism and traditional statistics are not enough to detect these subtle nutrient effects. Generally, this kind of studies assumes that the participants are adhered to the assigned dietary intervention and directly analyzes its effects over the target parameters. Thus, the evaluation of adherence is generally omitted. Although, sometimes, participants do not effectively adhere to the assigned dietary guidelines. For this reason, the trajectory map is proposed as a visual tool where dietary patterns of individuals can be followed during the intervention and can also be related with nutritional prescriptions. The trajectory analysis is also proposed allowing both analysis: 1) adherence to the intervention and 2) intervention effects. The analysis is made by projecting the differences of the target parameters over the resulting trajectories between states of different time-stamps which might be considered either individually or by groups. The proposal has been applied over a real nutritional study showing that some individuals adhere better than others and some individuals of the control group modify their habits during the intervention. In addition, the intervention effects are different depending on the type of individuals, even some subgroups have opposite response to the same intervention.

Analysis of the factors influencing lung cancer hospitalization expenses using data mining

Background

Hospitalization expenses for the therapy of lung cancer are not only a direct economic burden on patients, but also the focus of medical insurance departments. Therefore, the method for classifying and analyzing lung cancer hospitalization expenses so as to predict reasonable medical cost has become an issue of common interest for both hospitals and insurance institutions.

Methods

A C5.0 algorithm is adopted to analyze factors influencing hospitalization expenses of 731 lung cancer patients. A C5.0 algorithm is a data mining method used to classify calculation.

Results

Increasing the number of input variables leads to variation in the importance of different variables, but length of stay (LOS), major therapy, and medicine cost are the three variables of greater importance. They are important factors that affect the hospitalization cost of lung cancer patients. In all three calculations, the classification accuracy rate of training and testing partition sets reached 84% and above. The classification accuracy rate reached over 95% after addition of the cost variables.

Conclusion

The classification rules are proven to be in accordance with actual clinical practice. The model established by the research can also be applied to other diseases in the screening and analysis of disease hospitalization costs according to selected feature variables.

Classification tree analysis of race-specific subgroups at risk for a central venous catheter-related bloodstream infection

BACKGROUND

Studies of racial disparities in patient safety events often do not use race-specific risk adjustment and do not account for reciprocal covariate interactions. These limitations were addressed by using classification tree analysis separately for black patients and white patients to identify characteristics that segment patients who have increased risks for a venous catheter-related bloodstream infection.

METHODS

A retrospective, cross-sectional analysis of 5,236,045 discharges from 103 Florida acute hospitals in 2005-2009 was conducted. Hospitals were rank ordered on the basis of the black/white Patient Safety Indicator (PSI) 7 rate ratio as follows: Group 1 (white rate higher), Group 2, (equivalent rates), Group 3, (black rate higher), and Group 4, (black rate highest). Predictor variables included 26 comorbidities (Elixhauser Comorbidity Index) and demographic characteristics. Four separate classification tree analyses were completed for each race/hospital group.

RESULTS

Individual characteristics and groups of characteristics associated with increased PSI 7 risk differed for black and white patients. The average age for both races was different across the hospital groups (p < .01). Weight loss was the strongest single delineator and common to both races. The black subgroups with the highest PSI 7 risk were Medicare beneficiaries who were either < or = 25.5 years without hypertension or < or = 39.5 years without hypertension but with an emergency or trauma admission. The white subgroup with the highest PSI 7 risk consisted of patients < or = 45.5 years who had congestive heart failure but did not have either hypertension or weight loss.

DISCUSSION

Identifying subgroups of patients at risk for a rare safety event such as PSI 7 should aid effective clinical decisions and efficient use of resources and help to guide patient safety interventions.

Potential identification of pediatric asthma patients within pediatric research database using low rank matrix decomposition

Asthma is a prevalent disease in pediatric patients and most of the cases begin at very early years of life in children. Early identification of patients at high risk of developing the disease can alert us to provide them the best treatment to manage asthma symptoms. Often evaluating patients with high risk of developing asthma from huge data sets (e.g., electronic medical record) is challenging and very time consuming, and lack of complex analysis of data or proper clinical logic determination might produce invalid results and irrelevant treatments. In this article, we used data from the Pediatric Research Database (PRD) to develop an asthma prediction model from past All Patient Refined Diagnosis Related Groupings (APR-DRGs) coding assignments. The knowledge gleamed in this asthma prediction model, from both routinely use by physicians and experimental findings, will become fused into a knowledge-based database for dissemination to those involved with asthma patients. Success with this model may lead to expansion with other diseases.

Study of the distribution patterns of the constituent herbs in classical Chinese medicine prescriptions treating respiratory disease by data mining methods

OBJECTIVE

To provide the distribution pattern and compatibility laws of the constituent herbs in prescriptions, for doctor's convenience to make decision in choosing correct herbs and prescriptions for treating respiratory disease.

METHODS

Classical prescriptions treating respiratory disease were selected from authoritative prescription books. Data mining methods (frequent itemsets and association rules) were used to analyze the regular patterns and compatibility laws of the constituent herbs in the selected prescriptions.

RESULTS

A total of 562 prescriptions were selected to be studied. The result exhibited that, Radix glycyrrhizae was the most frequently used in 47.2% prescriptions, other frequently used were Semen armeniacae amarum, Fructus schisandrae Chinese, Herba ephedrae, and Radix ginseng. Herbal ephedrae was always coupled with Semen armeniacae amarum with the confidence of 73.3%, and many herbs were always accompanied by Radix glycyrrhizae with high confidence. More over, Fructus schisandrae Chinese, Herba ephedrae and Rhizoma pinelliae was most commonly used to treat cough, dyspnoea and associated sputum respectively besides Radix glycyrrhizae and Semen armeniacae amarum. The prescriptions treating dyspnoea often used double herb group of Herba ephedrae & Radix glycyrrhizae, while prescriptions treating sputum often used double herb group of Rhizoma pinelliae & Radix glycyrrhizae and Rhizoma pinelliae & Semen armeniacae amarum, triple herb groups of Rhizoma pinelliae & Semen armeniacae amarum & Radix glycyrrhizae and Pericarpium citri reticulatae & Rhizoma pinelliae & Radix glycyrrhizae.

CONCLUSIONS

The prescriptions treating respiratory disease showed common compatibility laws in using herbs and special compatibility laws for treating different respiratory symptoms. These principle patterns and special compatibility laws reported here could be useful for doctors to choose correct herbs and prescriptions in treating respiratory disease.

Data mining in healthcare and biomedicine: a survey of the literature

As a new concept that emerged in the middle of 1990's, data mining can help researchers gain both novel and deep insights and can facilitate unprecedented understanding of large biomedical datasets. Data mining can uncover new biomedical and healthcare knowledge for clinical and administrative decision making as well as generate scientific hypotheses from large experimental data, clinical databases, and/or biomedical literature. This review first introduces data mining in general (e.g., the background, definition, and process of data mining), discusses the major differences between statistics and data mining and then speaks to the uniqueness of data mining in the biomedical and healthcare fields. A brief summarization of various data mining algorithms used for classification, clustering, and association as well as their respective advantages and drawbacks is also presented. Suggested guidelines on how to use data mining algorithms in each area of classification, clustering, and association are offered along with three examples of how data mining has been used in the healthcare industry. Given the successful application of data mining by health related organizations that has helped to predict health insurance fraud and under-diagnosed patients, and identify and classify at-risk people in terms of health with the goal of reducing healthcare cost, we introduce how data mining technologies (in each area of classification, clustering, and association) have been used for a multitude of purposes, including research in the biomedical and healthcare fields. A discussion of the technologies available to enable the prediction of healthcare costs (including length of hospital stay), disease diagnosis and prognosis, and the discovery of hidden biomedical and healthcare patterns from related databases is offered along with a discussion of the use of data mining to discover such relationships as those between health conditions and a disease, relationships among diseases, and relationships among drugs. The article concludes with a discussion of the problems that hamper the clinical use of data mining by health professionals.

Implications for Nursing Research and Generation of Evidence

A sound informatics infrastructure is essential to optimise the application of evidence in nursing practice. A comprehensive review of the infrastructure and associated research methods is supported by an extensive resource of references to point the interested reader to further resources for more in depth study. Information and communication technology (ICT) has been recognized as a fundamental component of applying evidence to practice for several decades. Although the role of ICT in generating knowledge from practice was formally identified as a nursing informatics research priority in the early 1990s (NINR Priority Expert Panel on Nursing Informatics, 1993), it has received heightened interest recently. In this chapter, the authors summarize some important trends in research that motivate increased attention to practice-based generation of evidence. These include an increased emphasis on interdisciplinary, translational, and comparative effectiveness research; novel research designs; frameworks and models that inform generation of evidence from practice; and creation of data sets that include not only variables related to biological and genetic measures, but also social and behavioral variables. The chapter also includes an overview of the ICT infrastructure and informatics processes required to facilitate generation of evidence from practice and across research studies: (1) information structures (e.g., re-usable concept representations, tailored templates for data acquisition), (2) processes (e.g., data mining algorithms, natural language processing), and (3) technologies (e.g., data repositories, visualization tools that optimize cognitive support). In addition, the authors identify key knowledge gaps related to informatics support for nursing research and generation of evidence from practice.

Multivariate prediction of upper limb prosthesis acceptance or rejection

OBJECTIVE

To develop a model for prediction of upper limb prosthesis use or rejection.

DESIGN

A questionnaire exploring factors in prosthesis acceptance was distributed internationally to individuals with upper limb absence through community-based support groups and rehabilitation hospitals.

SUBJECTS

A total of 191 participants (59 prosthesis rejecters and 132 prosthesis wearers) were included in this study.

METHODS

A logistic regression model, a C5.0 decision tree, and a radial basis function neural network were developed and compared in terms of sensitivity (prediction of prosthesis rejecters), specificity (prediction of prosthesis wearers), and overall cross-validation accuracy.

RESULTS

The logistic regression and neural network provided comparable overall accuracies of approximately 84 +/- 3%, specificity of 93%, and sensitivity of 61%. Fitting time-frame emerged as the predominant predictor. Individuals fitted within two years of birth (congenital) or six months of amputation (acquired) were 16 times more likely to continue prosthesis use.

CONCLUSIONS

To increase rates of prosthesis acceptance, clinical directives should focus on timely, client-centred fitting strategies and the development of improved prostheses and healthcare for individuals with high-level or bilateral limb absence. Multivariate analyses are useful in determining the relative importance of the many factors involved in prosthesis acceptance and rejection.

Detecting hospital fraud and claim abuse through diabetic outpatient services

Hospitals and health care providers tend to get involved in exaggerated and fraudulent medical claims initiated by national insurance schemes. The present study applies data mining techniques to detect fraudulent or abusive reporting by healthcare providers using their invoices for diabetic outpatient services. This research is pursued in the context of Taiwan's National Health Insurance system. We compare the identification accuracy of three algorithms: logistic regression, neural network, and classification trees. While all three are quite accurate, the classification tree model performs the best with an overall correct identification rate of 99%. It is followed by the neural network (96%) and the logistic regression model (92%).

Analysis of meal patterns with the use of supervised data mining techniques--artificial neural networks and decision trees

BACKGROUND

At present, the analysis of dietary patterns is based on the intake of individual foods. This article demonstrates how a coding system at the meal level might be analyzed by using data mining techniques.

OBJECTIVE

The objective was to evaluate the usability of supervised data mining methods to predict an aspect of dietary quality based on dietary intake with a food-based coding system and a novel meal-based coding system.

DESIGN

Food consumption databases from the North-South Ireland Food Consumption Survey 1997-1999 were used. This was a randomized cross-sectional study of 7-d recorded food and nutrient intakes of a representative sample of 1379 Irish adults. Meal definitions were recorded by the respondent. A healthy eating index (HEI) score was developed. Artificial neural networks (ANNs) and decision trees were used to predict quintiles of the HEI based on combinations of foods consumed at breakfast and main meals.

RESULTS

This study applied both data mining techniques to the food and meal-based coding systems. The ANN had a slightly higher accuracy than did the decision tree in relation to its ability to predict HEI quintiles 1 and 5 based on the food coding system (78.7% compared with 76.9% and 71.9% compared with 70.1%, respectively). However, the decision tree had higher accuracies than did the ANN on the basis of the meal coding system (67.5% compared with 54.6% and 75.1% compared with 72.4%, respectively).

CONCLUSIONS

ANNs and decision trees were successfully used to predict an aspect of dietary quality. However, further exploration of the use of ANNs and decision trees in dietary pattern analysis is warranted.

Navigating the information technology highway: computer solutions to reduce errors and enhance patient safety

Standardized, seamless, integrated information technology in the health-care environment used with other industry tools can markedly decrease preventable errors or adverse events and increase patient safety. According to an Institute of Medicine (IOM) report released in 1999, preventable errors have caused between 44,000 and 98,000 deaths per year. Following the report, President Bill Clinton requested that the Agency of Healthcare Research and Quality, a government agency, look into the issue and fund, at the local or state level, processes that can reduce errors. Funding subsequently was made available for research that utilizes best practice tools in clinical practice to increase patient safety. The Joint Commission on Accreditation of Healthcare Organization has placed a great deal of emphasis on strategies to reduce patient identification errors. Fragmented systems tout the individual as well as enhanced safety applications. These applications, however, are related to prevention in specific conditions and in specific health-care settings. Systems are not integrated with common reference data and common terminology aggregated at a regional or national level to provide access to patient safety risks for timely interventions before errors and adverse events occur. Standardized integrated patient care information systems are not available either on a regional or on a national level. This article examines tangible options to increase patient safety through improved state-of-the-art tools that can be incorporated into the health-care system to prevent errors.

Application of a data-mining technique to analyze coprescription patterns for antacids in Taiwan

BACKGROUND

Although antacids are popular drugs with a long history of use, their true utilization patterns-including over-the-counter use-have rarely been documented. Because all antacids are reimbursed under the National Health Insurance program in Taiwan, it is possible to access and analyze nationwide data for these drugs.

OBJECTIVES

The purposes of this study were to estimate the scale of antacid prescribing in Taiwan using the national insurance claims for outpatient services and to analyze coprescribing patterns of antacids using modern data-mining techniques.

METHODS

The National Health Insurance Research Database in Taiwan supplied the visit-based sampling data sets, which had a sampling ratio of 0.2% for all claims for outpatient medical services in the year 2000. In addition to the plain statistics (ie, data from simple calculations) for antacid prescriptions, we also analyzed relationships between prescriptions for antacids and nonantacid drugs. A data-mining technique-association rule mining-was applied to identify the drugs prescribed in combination with antacids.

RESULTS

Among a total of 409,049 eligible prescriptions for 1,704,595 drug items to be administered orally, antacids were present in 213,494 prescriptions (52.2%). Antacid users were generally older than nonusers (mean [SD] age, 39.9 [23.4] years vs 32.4 [25.7] years). In all, 88.8% of antacid items (189,531/213,494) were prescribed without claims diagnoses of gastrointestinal disorders. Using association rule mining with a 1.0% minimum support factor, there were 36 strong association rules between prescriptions for antacids and other drug subgroups at the third level of Anatomical Therapeutic Chemical classification. Nonsteroidal anti-inflammatory drugs and drugs for treating upper respiratory infections played dominant roles in the associations with antacid prescriptions; vitamin B complex and antivertigo preparations were also strongly associated with antacids.

CONCLUSIONS

Antacid coprescriptions were common in Taiwan in the year 2000. Further study should investigate whether antacid prescribing patterns are influenced by Taiwanese perceptions that Western drugs injure the stomach.