제2형 당뇨병 환자의 당뇨망막병증 발생 예측도구 개발
Development of a predictive tool for diabetic retinopathy in patients with type 2 diabetes
- 주제(키워드) 당뇨망막병증 , 위험요인 , 예측도구 , 제2형 당뇨병
- 발행기관 고려대학교 보건대학원
- 지도교수 천병철
- 발행년도 2017
- 학위수여년월 2017. 8
- 학위구분 석사
- 학과 보건대학원 역학및보건정보학과
- 원문페이지 33 p
- 실제URI http://www.dcollection.net/handler/korea/000000076465
- 본문언어 한국어
- 제출원본 000045920357
초록/요약
Objectives : To identify the risk factors for diabetic retinopathy(DR) in type 2 diabetic patients and to develop a tool for predicting DR Methods: A retrospective cohort study was performed. We retrieved data of 279 type 2 diabetes patients who did not have DR at their first visit to a tertiary diabetic center and we assessed them to determine whether they developed DR after their first visit from 2009 to 2010. Age, sex, underlying chronic condition, health behavior, family history of diabetes mellitus(DM), duration of DM, HbA1c and other prognostic factors of the subjects were identified and collected from electronic medical records. Hazard ratios(HRs) for risk factors associated with DR were obtained using the Cox proportional hazards analysis. The predictive tool was scored through the HRs of risk factors. The Area under the curve(AUC) of the receiver operating characteristic(ROC) curve was obtained to determine the predictive value of the DR occurrence tool, and the Youden index was used to determine the cut-off value for the score to predict the DR risk. Results: Of the 279 patients, 73 (26.2%) developed DR. In multivariate Cox proportional hazards analysis, the risk factors for DR were family history of DM (HR: 1.77, 95% CI:1.07-2.95), smoking (HR: 2.39, 95% CI:1.37-4.16), duration of DM (11-20 years HR:1.48, 95% CI:0.65-3.39), (≥21 years, HR:3.71, 95% CI:1.88-11.63), high hemoglobin A1c level (7-7.9% HR:1.55, 95% CI:0.90-2.68), (≥8% HR:2.50, 95% CI:1.37-4.55) and proteinuria (≥urine albumin creatinine ratio 30mg HR: 1.98, 95% CI:1.22-3.20). A predictive tool was developed using significant variables and HRs in multivariate analysis. The score of the tool was weighted by the HR of each variable, and it had a value from 0 to 13. The AUC value of the ROC curve showing the predictive power of the tool was 0.89 (95% CI: 0.79-0.97) and the cut-off point at risk of occurrence was 4.5 points. Conclusion: The predictive tool for the assessment of DR that was developed in this study could be used to select the subjects at risk for DR and to establish the eye observation plan. In addition, it can be a useful tool for customized education considering the characteristics of the individual and the progress of the disease in diabetes management education. Key words: diabetic retinopathy, risk factors, predictive tool, type 2 diabetes
more목차
목 차
Abstract
Ⅰ. 서론 1
Ⅱ. 연구대상 및 방법 3
1. 연구 대상 3
2. 연구 방법 3
3. 자료 분석 방법 3
4. 윤리적 고려 5
Ⅲ. 연구결과 6
1. 당뇨망막병증 발생군과 비발생군의 특성 비교 6
2. 당뇨망막병증 발생 관련 위험요인 분석 9
3. 당뇨망막병증 발생 예측도구 11
Ⅳ. 고찰 14
Ⅴ. 결론 18
참고문헌 19
국문요약
