Agreement analysis of between referral note diagnosis and final principal diagnosis

병원행정학과 병원행정전공/석사[한글]이 연구는 진료의뢰기관의 진단명과 의뢰받은 종합전문요양기관의 퇴원 시 주진단명을 비교하여 질환의 유형과 진단명간의 일치도 및 일치도의 영향을 미치는 요인을 분석하여 환자특성 및 의료기관간 특성을 파악하고자 하였다.연구 자료는 2003년 의뢰서를 지참한 부인과 신규 입원환자 291명을 사용하였고, 분석은 통계소프트웨어 SPSS(ver. 13.0)와 SPSS의 데이터 마이닝 툴인 Answer Tree(ver. 3.01)를 사용하였다.주요 결과는 진료정보 특성에서 일치도 점수는 부인과 환자를 대상으로 의료이용특성을 분석하였는데 입원경로는 외래경유가 응급실 입원보다 높았으며 진료의뢰기관의 과목/종별로는 산부인과의원이 일치도 점수가 가장 높았으며 진료의뢰기관 검사에서는 자궁경부 세포진 검사(Pap smear)나 조직병리 검사(Bx) 한 경우가 가장 높았으며 의뢰받은 병원에서의 특수검사, 일반검사 모두 검사안함이 검사함보다 더 높았다.질병정보 특성에서 한국표준사인 분류 4차 개정판의 부인과 질환을 3단 분류 하여 주진단 분포를 보았는데 자궁의 평활근종(D25)이 가장 높았으며 자궁외 임신(O00), 난소의 양성 신생물(D27)의 순이었으며 질병 그룹별로는 여성 생식기관의 양성 신생물(D25-D28)이 가장 높았으며, 여성 생식기의 비염증성 질환(N70-N77)의 순이었다.질병에 따른 일치도 점수를 보면 자궁경의 악성 신생물(C53)이 가장 높았으며 여성 생식기 탈출(N81), 자궁의 평활근종(D25)의 순이었고 질병 그룹별 일치도 점수를 보면 여성생식기관의 악성 신생물(C51-C58)이 가장 높았으며 여성생식기관의 양성 신생물(D25-D28), 상피내의 신생물(D00-D09)의 순이었다.진단명 일치도 점수 예측을 위한 의사결정나무 분석 중 CART algorithm 방법에 의하면 일치도에 영향을 미치는 가장 중요한 요인이 진료의뢰기관 과목/종별이었고 이에 의해 분리가 이루어 졌으며 질병그룹별과 직업그룹에서 일치도를 결정짓는 주요 요인은 각각 신생물과 사무직, 전문직, 주부, 학생이었다. 위의 분석결과, 의뢰서상의 진단명과 의뢰받은 기관의 확진 진단명의 유형과 양상 및 진단명과의 일치 여부를 파악함으로서 의료기관간 연계성 있는 진료가 필요함을 알 수 있었으며 향후 3차 의료기관으로서 중증환자를 유치하여 치료하는 의료기관간 기능분화에 대한 평가가 될 수 있도록 진료의뢰 정보를 이용한 심도있고 다양한 연구가 계속되어야 할 것이다. [영문]The purposes of this study are to compare the diagnosis of the medical institution to which medical treatment was referred with the primary diagnosis in the event of discharging the referred specialized nursing institution and to analyze the types of diseases, degree of agreement, and factors that affect the degree of agreement in order to discover the characteristics of patients and medical institutions.The 291 cases of newly gynecology inpatients who had a referral note for medical treatment in 2003 were used as the targets. SPSS (Ver. 13.0), the statistics software, and Answer Tree (Ver. 3.01), the data mining tool for SPSS, were used as tools of analysis.The main results of this study were as follows:The degree of agreement was analyzed from the gynecology inpatients in terms of the characteristics of information regarding medical treatments. As for methods of hospitalization, the ratio of formal administration was higher than hospitalization through emergency rooms. In terms of the department of medical institutions, the degree of agreement was highest OBGY Clinics.As a result of examinations at the referral hospital, PAP Smears and Bx were highest. Also, the degree of agreement was higher when neither special examinations nor regular ones were performed at the referral hospital than when examinations were performed.In terms of the characteristics of information on diseases, the gynecological diseases based on the 4th edition of the Korean standard Classification of Diseases were classified and the distribution of principal diagnosis was in the following order : Leiomyoma of the uterus (D25), ectopic pregnancy (O00), and benign neoplasms of the ovary (D27). For each disease group, the benign neoplasm of female genital organs (D25-D28) and the inflammatory disease of female genital organs (N70-N77) were high in this order.The degree of agreement for each disease was as follows: The malignant neoplasm of the cervix (C53), female genital prolapse (N81), and leiomyoma of the uterus (D25) were high in this order. The degree of agreement for each disease group was as follows: The malignant neoplasm of female genital organs (C51-C58), the benign neoplasm of female genital organs (D25-D28), and in situ neoplasms (D00-D09) were high in this order.As a result of the multiple regression analysis from the factors that affect the degree of agreement in the diagnosis of inpatients who had a referral note for medical treatment, whether examinations had been performed at the referral hospital or not affected to a higher degree. Also, when a special examination had been performed at the referral hospital, the fact affected to a lower degree. As a result of analyzing the decision tree for estimating the degree of agreement of diagnosis in the CART algorithm method, the first criteria for differentiation were the departments and classification of the referral hospital and the second and third criterion for differentiation were disease groups and occupations, respectively.As a result of the analyses in this study, it turned out that medical treatments that collaborated among medical institutions were needed by figuring out whether the types and patterns of differential diagnosis at the referral hospital matched the diagnosis stated in the referral note for medical treatment. A variety of exhaustive research should continue by using information on the medical refer so that the function specialization of medical institutions that attract and treat serious cases as tertiary medical institutions can be assessed.prohibitio