• 사물인터넷융복합논문지
• /
• 제8권2호
• /
• pp.13-24
• /
• 2022

본 연구는 우리나라에서 매년 증가하고 있는 노인장기요양기관의 부당청구 맥락과 부당청구 예방을 위한 대책들이 어떠한지를 탐색하기 위해서 언론기사를 활용한 텍스트 마이닝 분석을 실시하였다. 기사는 뉴스 빅테이터 분석 시스템인 빅카인즈에서 수집하였고, 수집기간은 노인장기요양보험이 시행된 2008년 7월부터 2022년 2월 28일까지로 약 15년간이다. 이 기간 동안 '노인요양+부당청구', '장기요양+부당청구', 등의 키워드로 총 2,627개의 기사가 수집되었고, 이중 중복된 기사를 제외한 총 946개가 선정되었다. 본 연구의 텍스트마이닝 분석결과로 첫째, 모든 구간(2008.7.1-2022.2.28)에서 가장 높은 빈도로 언급된 상위 10위 키워드는 노인장기요양기관, 부당청구, 국민건강보험공단, 노인장기요양보험, 장기요양급여(비용), 노인요양시설, 보건복지부, 노인, 신고, 포상금(지급)의 순으로 나타났다. 둘째, N-gram 분석결과 장기요양급여(비용)과 부당청구, 부당청구와 노인장기요양기관, 허위와 부당청구, 신고와 포상금(지급), 노인장기요양기관과 신고 등의 순으로 나타났다. 셋째, TF-IDF 분석은 빈도분석의 결과와 유사하게 나타났지만, 신고, 포상금(지급), 증가 등은 순위가 상승하였다. 상기 분석결과를 바탕으로 노인장기요양기관 부당청구 예방을 위한 방향성을 제시하였다.

• Journal of Preventive Medicine and Public Health
• /
• 제38권2호
• /
• pp.132-140
• /
• 2005

Objectives: To propose a risk-adjustment model from insurance claims data, and analyze the changes in cesarean section rates of healthcare organizations after adjusting for risk distribution. Methods: The study sample included delivery claims data from January to September, 2003. A risk-adjustment model was built using the 1st quarter data, and the 2nd and 3rd quarter data were used for a validation test. Patients' risk factors were adjusted using a logistic regression analysis. The c-statistic and Hosmer-Lemeshow test were used to evaluate the performance of the risk-adjustment model. Crude, predicted and risk-adjusted rates were calculated, and compared to analyze the effects of the adjustment. Results: Nine risk factors (malpresentation, eclampsia, malignancy, multiple pregnancies, problems in the placenta, previous Cesarean section, older mothers, bleeding and diabetes) were included in the final risk-adjustment model, and were found to have statistically significant effects on the mode of delivery. The c-statistic (0.78) and Hosmer-Lemeshow test ($x^2$=0.60, p=0.439) indicated a good model performance. After applying the 2nd and 3rd quarter data to the model, there were no differences in the c-statistic and Hosmer-Lemeshow $x^2$. Also, risk factor adjustment led to changes in the ranking of hospital Cesarean section rates, especially in tertiary and general hospitals. Conclusion: This study showed a model performance, using medical record abstracted data, was comparable to the results of previous studies. Insurance claims data can be used for identifying areas where risk factors should be adjusted. The changes in the ranking of hospital Cesarean section rates implied that crude rates can mislead people and therefore, the risk should be adjusted before the rates are released to the public. The proposed risk-adjustment model can be applied for the fair comparisons of the rates between hospitals.

• 한국임상약학회지
• /
• 제24권4호
• /
• pp.255-264
• /
• 2014

Objective: The aims of this study are to investigate the total volume of prescribed medicines against Alzheimer's disease (AD) and the trends of usage by analyzing the claims-data from the Korea National Health Insurance Service. Method: The demographic and claims-data were included the major AD treating medicines such as donepezil, galantamine, rivastigmine and memantine, and analyzed during the period of 2010~2012. The assessing criteria were gender, age, habitation, types of medical institution, code of ingredients, outcomes of treatment, volume and amount of claims, and the numbers of patients with dementias. After trimming the data, it were analyzed by the market size, demographic traits, characteristics of medical service, characteristics of each anti-AD medicine, etc. Results: Among the chosen 4 medicines, donepezil had the top prescription volumes. Most prevalent prescribing preparations of donepezil were conventional types. However, among the non-conventional types, oro-dispersible formulation is the fast increasing one in both volume and growth rate. This specialized preparations to improve both toleration and adherence, tend to being prescribed generally at the tertiary medical institutions. While the younger patients with mild-to-moderate AD mostly treated by expensive medicines in resident at the tertiary hospitals, the rest older patients with severe AD have been treated non-expensive one at long-term care facilities. Conclusion: AD is a chronic illness therefore, long-term use of therapeutic medications are highly important. If an anti-AD treatment was applied steadily in the earlier stages, it would be achieved not only improving the quality of life of patient but also reducing the expenses in the medical and nursing cares. As the socioeconomical impacts of AD is expanding, healthcare professionals need to aware the importance of pharmacotherapy and to improve sociopolitical fundamentals.

• Journal of the Korean Data and Information Science Society
• /
• 제18권4호
• /
• pp.1093-1101
• /
• 2007

Bonus-Malus system in automobile insurance rewards claim-free policyholders by premium discounts and penalizes policyholders with claims by premium surcharges. The purpose of adopting bonus-malus system is to alleviate differences in risk propensity. A well-known side-effect of bonus-malus system is the tendency of policyholders to pay small claims themselves and not report them to their, in order to avoid future premium increases. This phenomenon is called hunger for bonus. In this paper, we introduces an alternative approach to the Bonus-Malus system in automobile insurance - the approach is based on a deductible theory; and then search for a proper way combining both of them. Also, we construct a new algorithm to determine the optimal strategy of the policyholder based on the proposed model.

• 경영정보학연구
• /
• 제5권1호
• /
• pp.49-60
• /
• 2003

본 연구의 목적은 현재 심각한 사회 및 경제 문제로 대두되고 있는 보험사기를 효과적으로 적발하기 위하여, visualization 데이터마이닝 tool을 실제 사례에 적용하여 그 타당성을 검증하는데 있다. 이를 위하여 최근 가장 효과적인 visualization 데이터마이닝 tool로 인정되고 있는 i2사의 Analyst's Notebook을 활용하여 대량의 보험금 청구 자료로부터 보험사기의 혐의가 가는 거래를 찾고, 이를 근거로 보첩사기의 혐의를 입증하는 일련의 과정을 검토하였다. 그 결과 visualization 데이터마이닝 tool이 대량의 보험금 청구 자료에서 혐의가 가는 거래를 찾는 단순한 예측의 수준을 넘어, 관련 범죄를 추적하여 체계적, 계획적으로 기획된 보험사기단을 추적해내는 성과를 올렸다. 따라서 보험사기 둥과 같은 부정거래나 범죄 행위를 적발하는 데는 visualization 데이터마이닝 tool이적합한 것으로 판명되었다.

• 보건행정학회지
• /
• 제24권4호
• /
• pp.322-329
• /
• 2014

Background: As most of people in Korea are covered by National Health Insurance (NHI), the disease information collected in NHI provides high availability for health policy. Nevertheless, the validity of disease codes in NHI data has been controversial till now. So we tried to evaluate the validity of them by comparing the NHI claims data with Korean National Hospital Discharge In-depth Injury Survey (KNHDIIS) data. Methods: We compared the NHI patients sample data (2009) with the KNHDIIS data (2009). We selected the inpatient data of KNHDIIS and NHI patients sample. The weighted number of patients from NHI patients sample was 5,551,210 and the number of patients from KNHDIIS was 5,559,874. We classified the disease codes into principal diagnoses and other diagnoses, and we compared as one, two, three unit level. Also we calculated the agreement rate of each of them. Results: In the comparison of principal diagnoses, NHI claims data had more C code than KNHDIIS data did, whereas KNHDIIS data had more Z code than NHI claims data did. In the comparison of other diagnoses, NHI claims data had 2, 3 more codes than KNHDIIS data did. The overall agreement rate at three unit level was 76.5% in principal diagnoses and 46.8% in other diagnoses. Conclusion: Considering the large difference between the two data, the validity of disease codes in NHI Claims data seems to be low. To increase the validity of them, the definite detail coding indicator, the reinforcement of coding education, and the reform of system are needed.

• Journal of Preventive Medicine and Public Health
• /
• 제33권1호
• /
• pp.76-82
• /
• 2000

Objectives : We attempted to assess He accuracy of ICD codes for cerebrovascular diseases in medical insurance claims (ICMIC) and to investigate the reasons for error. This study was designed as a preliminary study to establish a nationwide surveillance system. Methods : A total of 626 patients with medical insurance claims who indicated a diagnosis of cerebrovascular diseases during the period from 1993 to 1997 was selected from the Korea Medical Insurance Corporation cohort (KMIC cohort: 115,600 persons). The KMIC cohort was 10% of those insured who had taken health examinations in 1990 and 1992 consecutively. The registered medical record administrators were trained in the survey technique and gathered data from March to May 1999. The definition of cerebrovascular diseases in this study included cases which met ore of two criteria (Minnesota, WHO) or 'definite stroke' in CT/MRI finding. We questioned the medical record administrators to explain the error if the final diagnoses were not coded as stroke. Results : The accuracy rate of the ICMIC was 83.0% (425 cases) Medical records were not available for 8.2% (51 cases) due to the closing of hospitals, the absence of a computer system or omission of medical record, etc. Sixty-three cases (10.0%) were classified as impossible to interpret due to insufficient records in 'major clinical symptoms' or 'neurological deficits'. The most common reason was 'to meet review criteria of medical insurance benefits (52.9%)'. The department where errors in the ICMIC occurred most frequently was the department for medical insurance claims in the hospital. Conclusion : The accuracy rate of the ICMIC was 83.0%.

• 보험의학회지
• /
• 제29권2호
• /
• pp.8-17
• /
• 2010

• 디지털융복합연구
• /
• 제11권10호
• /
• pp.593-608
• /
• 2013

상해상병으로 청구되는 건수가 증가함에 따라 조사 대상을 보다 정교하게 선정하여 상해요인 조사 대상을 줄이면서 환수율 및 환수금액을 올릴 수 있는 방안을 마련할 필요가 있다. 이를 위해서 2006~2011년까지의 상해요인 조사자료를 수집하여 의사결정나무 모형을 활용하여 지역가입자 상해상병 진료건에 대한 부당환수 조사대상 선정모형을 개발하였다. 최종 개발된 모형결과에 따르면, 조사대상 유형은 18개로 분류되었고, 이러한 분류결과는 실제 조사가 시행될 시, 모형을 적용하지 않았을 때 보다 최고 12.8배 높은 부당환수결정율을 나타낼 수 있을 것으로 분석되었다. 또한, 본 연구에서 개발된 조사 대상자 선정 모형을 실제 업무에 적용하기 위해서는 조사물량 대비 국민건강보험공단의 조사인력 및 운영 계획을 보다 면밀히 검토해야만 모형 적용의 효과성이 극대화 될 수 있을 것으로 판단된다.

• Journal of Preventive Medicine and Public Health
• /
• 제28권3호
• /
• pp.640-650
• /
• 1995

After the introduction of National Medical Insurance in 1989, the medical demand has rapidly increased. The impact of increased medical demand was followed by an increase in the number of claims in need of review. We studied a new, fair method for reducing the number of claims reviewed. We analysed 90,583 outpatient claims submitted between September and October; claims were made for services given August of 1994. We finally suggested a screening system for claims review using a statistical method of discriminant analysis of the medical costs. The results were as follows. 1. In the cut-off group, age, days of medication, number of hospital or clinic visits, and total charge were significantly high. The cut-off rates according to the hospital-type and existence of accompanied disease were significantly different 2. According to ICD, the cut-off rate was highest in peripheral enthesopathies and allied syndromes(20.76%), lowest in acute sinusitis(0.93%). The mean charges were significantly different according to ICD and existence of cut-off. 3. We build discriminant functions by ICD with such discriminant variables as patient age, sex, existence of accompanied disease, number of hospital or clinic visits, and 9 detailed hospital or clinic charges included in claim. 4. We applied the discriminant function for screening those claims that were expected to be cut-off. The sensitivities comprised from 40% to 70%, and specificities from 70% to 95% by ICD. Acute rhinitis had highest sensitivity(100.00%) and other local infections of skin and subcutaneous tissue had highest specificity(98.45%). The expected number of cut-off was 17,762(19.61%). The total sensitivity was 49.62%, the total specificity was 82.57% and the error rate was 19.66%. We lacked economic analysis such as cost-benefit analysis. But, if the new method of screening claims using discriminant analysis were applied, the number of claims in need of review will reduce considerably.