Fraud and Misconduct in the Insurance Industry - Literature review Example

Fraud and Misconduct in the Insurance Industry In the insurance industry, fraud occurs when the insured represents false information about the accident to the insurer in order to benefit in terms of money. Therefore, the insurance companies usually put in place some measures to prevent, detect and respond to fraud cases that can result to monetary losses. The following summary reviews of the articles describe how an insurance company can detect fraud and misconduct, prevent fraud and respond to fraud cases. Summary review of the article, “A Model for the Detection of Insurance Fraud” by Belhadji, E.B., Dionne, G. & Tarkhani, F. (2000) The research done in this article involves development of a model to aid the insurance company to detect fraud. The model is only applied to damages that occur in the automobile sector, for example, car damages, passengers’ injuries and destruction of properties in road accidents. (Belhadji, Dionne and Tarkhani 2000, p. 517-518) The research involved finding out the market shares of insurance companies among 20 firms in the Quebec insurance sector, in Canada. Eighteen companies accepted to disclose their closed claim files to the researchers. Therefore, the sample size was eighteen companies that enabled the study of market shares of companies in the Quebec automobile insurance sector (Belhadji, Dionne and Tarkhani 2000, p. 518). Methodology Initially, the participating companies filled some questionnaires assigned to them with the assistance of their chosen adjusters. Then, random selection of the companies’ files was conducted. Then, the representatives of the sample were collected from the companies’ closed claim files. Then, a table was prepared to show the number of questionnaires filled by each company with their corresponding percentages of the total sample (Belhadji, Dionne and Tarkhani 2000, p. 518). Random sampling was conducted where the insurers chose the adjuster who had already reviewed the files in the sample. These adjusters assisted them in filling the questionnaires. There was a control group that ensured procedures were followed to protect the identity of the companies and adjusters. Eighteen companies returned 2,509 as according to the table. Another table was prepared showing the number of files with suspected and established fraud cases. Established fraud means that the file contained fraud whereas suspected file means that the adjusters suspected fraud cases while handling files in the sample. 116 files had suspected fraud cases and 18 files were suspected by the adjusters to contain fraud (Belhadji, Dionne and Tarkhani 2000, p. 518-519)). Reviews of the fraud indicators where appropriate parameters are used allowing the selection of the particular fraud indicators. Then, they were used to predict the probability of presence of fraud in a file. The method was successful as it delivered similar results when using the fraud indicators method and the questionnaire method. Statistics were used in selecting the significant fraud indicators that predicted the probability of fraud in the files (Belhadji, Dionne and Tarkhani 2000, p. 519). The research carried out in 1995 was to evaluate the nature of fraud cases in the automobile industry and to identify the set of indicators that are necessary to identify fraud cases. In the choice of indicators, 50 fraud indicators were presented to the companies’ adjusters. They were to fill the number of fraud indicators in correspondence to the suspected files. The fraud indicators revolved around the damage/accident, the vehicle and the insurer. An example of a fraud indicator is when the vehicle involved in an accident is said to be decent but, a statement in the same file confirms that the vehicle was indeed very old. The fraud indicators are identified using parameters, and they become significant in determining the probability that a file has fraud contents. In addition, other significant parameters, that is, mathematical models, are used to determine whether the suspected and established fraud cases in the files are true or not (Belhadji, Dionne and Tarkhani 2000, p. 519-520). Lack of a comparison method in detecting fraud in the files leaves the discussed method doubted on its effectiveness in detecting fraud. Summary of the article, “Informing the development of the fraud prevention toolset through a situated analysis of fraud investigations expertise; Behavior & Information Technology” by Ormerod, T. C., Ball, L. J., & Morley, N. J. (2012). The projects discussed in this article revolve around early detection of fraud cases and analysis of the fraud cases suspected in the insurance claim files. Additionally, the scope of the research in this article gives the methods applied by insurance companies to prevent fraud (Ormerod, Ball & Morley, 2012 p. 371). Ethnography helps in understanding how the projects allow early fraud detection and investigation of the fraud cases that have already occurred. Insurance companies can periodically deploy their experts in ethnography to check the daily activities of the insured with the support of technological tools discussed in this article. In this way, insurance companies can prevent fraud and misconduct from occurring in the insured domains (Ormerod, Ball & Morley, 2012 p. 372). Procedure In ethnography, an observer works in the system for long periods to study the work practices and document the day to day activities. During the research, four studies were conducted within duration of 2-4 weeks. The claims’ handling offices for four insurance companies were involved where both personal and commercial claims were investigated. The insurers’ commercial and personal claims each ranging between 35,000 and 113,000 per annum was investigated. The handled claims were judged to contain fraudulent cases ranging between 4% and 9%. The loss adjusters and the specialist investigation units handled between 125 and 342 referred claims. The special claims and the loss incurred by the claimants’ data were also collected (Ormerod, Ball & Morley, 2012 p. 372-373)). The researchers were required to take note of the activities carried out in the firms, interview employees, inspect the firms’ documentation and identify the technological tools used for compensating the insured in those firms. There was enough supervision because the researchers adapted an inclusive and open approach when collecting data (Ormerod, Ball & Morley, 2012 p. 373). In collection of data, perceived bias can arise because the method of inspecting and interviewing workers does not specify how fraud can be detected through collection of raw data. After the fieldwork, the findings are summarized in inclusion of the observations made on the workers behavior and their working environments. Those findings that support fraud detection through technology were then identified. A report showing the performance of the fraud investigators and loss adjusters was prepared. In the study, no statistics were used. Then, software prototypes are designed using the ethnographic research. The software prototypes include the Mass Detection Tool (MDT) for handling claims and the Suspicion Building Tool (SBT) for supporting the processes of investigations of fraud in the claims (Ormerod, Ball & Morley, 2012 p. 377) In determining whether the handled claims handled in the different companies contains fraud or not; the MDT and SBT software prototypes are implemented. The MDT helps to determine the whether the current claims in a company are fraudulent or not. Adjusters calculate the probabilities based on other probabilities of other handled claims that have the similar fraud indicators with the current claims. The SBT involves the misconducts capture and analysis, visualization and argumentation. The misconducts in dealing with claims are identified in order to help in the searching process. If the claims in those companies contain fraud, then, the experts deployed in the insured premises are recommended to check such misconducts (Ormerod, Ball & Morley, 2012 p. 378). Summary of the article, “Method for selection of motor insurance fraud management system components based in business performance: Technological and Economic Development of Economy” by Furlan, S., Vasilecas, O., & Bajec, M. (2001). The article discusses the proposed method aimed at improving fraud management in motor vehicle insurance companies. The article focuses on how fraud management system (FMS) components can develop the operations of companies that have already incurred losses due to fraudulent claims. The method developed in the paper is based on data obtained from motor insurance firms in Slovenia (Furlan, Vasilecas & Bajec 2001, p. 536) The Methodology First, a wide-ranging literature review was carried and it covered two areas, that is, broad research and narrow research areas. The three stages of study in the research include the construction of the method, evaluation method and interpretation of the results obtained. The research methods were first described and then, the literature reviews about the system were analyzed. The literature review includes the expert reviews that are aimed at evaluating and discussing the three components of the fraud management system (Furlan, Vasilecas & Bajec 2001, p. 537). Domain experts were also interviewed to explain how they managed fraud in six motor insurance firms. The domain experts provided the evaluation of maturity levels of the processes involving the management of insurance claims in their respective companies. The researchers assessed the maturity of these processes by using Capability Maturity Model Integration (CMMI) where each domain experts were assigned grades 1 to 5 depending on the level of maturity. At this point, a control group lacked to find out whether the researchers were biased in issuing grades to the domain experts. Also, the fact that insurance companies are reluctant in sharing their information, the interviewers might have not obtained enough information from the companies (Furlan, Vasilecas & Bajec 2001, p. 538). The method components, the key detection indicators (KPI) of fraud and the KPI/activity matrix, the FMS component matrix were constructed from the information obtained from the literature interviews. In this study, there were statistics utilized and no control group was assigned to ensure the interviews were conducted properly (Furlan, Vasilecas & Bajec 2001, p. 537). In the follow up, the method was then applied in diverse companies and at the same time implementing the components of the FMS. Each company’s potential improvement was identified, and then the company would select the key detection indicators that lacked the most in the industry. Those FMS components that would influence the performance of companies in that area were identified using the method of evaluation developed. The performance of companies after one year was then assessed and compared to the old performance. The method was successful in detecting fraud because the key problems were identified and rendered to be improved by the FMS components. (Furlan, Vasilecas & Bajec 2001, p. 538). In FMS, insurance companies take corrective actions to manage the fraud that have already occurred through changing their operations. In summary, the FMS components, that is information systems and software programs developed in the study are used to remedy the operations of the companies already affected by fraud and misconduct. References List Belhadji, E.B., Dionne, G. & Tarkhani, F. (2000). A model for the detection of insurance fraud, The Geneva Papers on Risk and Insurance 25 (4), 517-538. Furlan, S., Vasilecas, O., & Bajec, M. (2001). Method for selection of motor insurance fraud management system components based in business performance, Technological and Economic Development of Economy, 17 (3), 535-561. Ormerod, T. C., Ball, L. J., & Morley, N. J. (2012). Informing the development of a fraud prevention toolset through a situated analysis of fraud investigations expertise, Behavior & Information Technology, 31, (4), 371 -381. Read More