대한방사선종양학회지 게재 논문의 통계적 오류 현황

박희철(Hee Chul Park),최두호(Doo Ho Choi),안성복(Song-Vogue Ahn),강진오(Jin Oh Kang),김은석(Eun-Seog Kim),박원(Won Park),안승도(Seung Do Ahn),양대식(Dae Sik Yang),윤형근(Hyong Geun Yun),정은지(Eun Ji Chung),지의규(Eui Kyu Chie),표홍렬( 대한방사선종양학회 2008 Radiation Oncology Journal Vol.26 No.4

목 적: 본 연구는 대한방사선종양학회지 게재 논문의 통계 오류 현황을 파악하고 이에 근거한 문제 제기를 통해 학회지의 학술적 발전에 기여하고자 하였다. 대상 및 방법: 2006∼2007년 사이에 대한방사선종양학회지에 게재된 총 77편의 논문을 연구 대상으로 하였다. 각각의 논문에 적용된 통계 방법론의 적정성 평가는 통계점검표를 활용하였다. 통계점검표에는 연구의 종류, 통계기법의 사용 범위, 각 논문에 적용된 통계 기법의 종류, 통계적 기법 적용의 타당성 항목이 포함되었다. 통계 오류는 ‘생략의 잘못’과 ‘시행의 잘못’ 항목으로 나누었다. 한 논문에서 서로 다른 항목이 여러 가지 관찰된 경우 각각 횟수로 측정하였다. 같은 항목이 2회 이상 측정된 경우 1회로 횟수를 측정하였다. 통계 전문가가 개별 논문을 대상으로 통계점검표를 작성하였다. 일차 평가자가 방사선종양학 전문가가 아닌 것에서 올 수 있는 평가 오류가 있을 수 있으므로 개별 통계점검표는 간행위원회에서 한 차례 더 세부 점검을 하였다. 작성된 통계점검표의 통계 분석은 SAS (version 9.0, SAS Institute, NC, USA) 소프트웨어를 이용하였고 빈도분석을 시행하여 각 항목의 빈도와 백분율을 산출하였다. 결 과: 총 77편 중 원문을 다운로드 할 수 없었던 4편을 제외한 73편의 게재 논문을 대상으로 평가하였다. 증례보고는 5편, 원저 논문은 68편이었다. 대상 논문 중 46편의 논문에서 통계적 추론을 사용하였고, 16편의 논문에서 단순한 기술통계를 사용했으며, 11편의 논문에서는 통계적 기법을 사용하지 않았다. 추론통계를 사용한 46편의논문에서 사용된 추론통계의 횟수는 각각 분할표분석 17회(37.0%), 비교통계분석 23회(50.0%), 회귀분석 7회(15.2%), 상관분석 5회(10.9%), 생존분석이 27회(58.7%)였다. 통계 기법을 활용하여 연구 결과를 분석할 때 통계적용의 오류가 없는 논문은 19%였다. ‘생략의 잘못’은 34편(50.0%)의 논문에서 총 50회 관찰되었다. ‘시행의 잘못’은 35편(51.5%)의 논문에서 총 47회 관찰되었다. ‘생략의 잘못’과 ‘시행의 잘못’이 모두 발견된 논문은 21편 (30.9%)이었다.결 론: 대한방사선종양학회지에 게재된 논문에서 통계 분석 과정의 다양한 영역에 걸쳐 크고 작은 통계적 오류가 있음을 확인하였다. 향후 대한방사선종양학회지 투고 논문의 심사 과정에서 통계 오류에 관한 적절한 심사를 추가하는 것이 필요하다고 판단된다. Purpose: To improve the quality of the statistical analysis of papers published in the Journal of the Korean Society for Therapeutic Radiology and Oncology (JKOSTRO) by evaluating commonly encountered errors. Materials and Methods: Papers published in the JKOSTRO from January 2006 to December 2007 were reviewed for methodological and statistical validity using a modified version of Ahn’s checklist. A statistician reviewed individual papers and evaluated the list items in the checklist for each paper. To avoid the potential assessment error by the statistician who lacks expertise in the field of radiation oncology; the editorial board of the JKOSTRO reviewed each checklist for individual articles. A frequency analysis of the list items was performed using SAS (version 9.0, SAS Institute, NC, USA) software. Results: A total of 73 papers including 5 case reports and 68 original articles were reviewed. Inferential statistics was used in 46 papers. The most commonly adopted statistical methodology was a survival analysis (58.7%). Only 19% of papers were free of statistical errors. Errors of omission were encountered in 34 (50.0%) papers. Errors of commission were encountered in 35 (51.5%) papers. Twenty-one papers (30.9%) had both errors of omission and commission. Conclusion: A variety of statistical errors were encountered in papers published in the JKOSTRO. The current study suggests that a more thorough review of the statistical analysis is needed for manuscripts submitted in the JKOSTRO.

Sex difference in the effect of the fasting serum glucose level on the risk of coronary heart disease

Ahn, Song Vogue,Kim, Hyeon Chang,Nam, Chung Mo,Suh, Il Elsevier 2018 Journal of cardiology Vol.71 No.2

<P><B>Abstract</B></P> <P><B>Objective</B></P> <P>Diabetic women have a greater relative risk of coronary heart disease than diabetic men. However, the sex difference in the effect of fasting serum glucose levels below the diabetic range on the risk of coronary heart disease is unclear. We investigated whether the association between nondiabetic blood glucose levels and the incident risk of coronary heart disease is different between men and women.</P> <P><B>Methods</B></P> <P>The fasting serum glucose levels and other cardiovascular risk factors at baseline were measured in 159,702 subjects (100,144 men and 59,558 women). Primary outcomes were hospital admission and death due to coronary heart disease during the 11-year follow-up.</P> <P><B>Results</B></P> <P>The risk for coronary heart disease in women significantly increased with impaired fasting glucose levels (≥110mg/dL) compared to normal glucose levels (<100mg/dL), whereas the risk for coronary heart disease in men was significantly increased at a diabetic glucose range (≥126mg/dL). Women had a higher hazard ratio of coronary heart disease associated with the fasting serum glucose level than men (<I>p</I> for interaction with sex=0.021).</P> <P><B>Conclusions</B></P> <P>The stronger effect of the fasting serum glucose levels on the risk of coronary heart disease in women than in men was significant from a prediabetic range (≥110mg/dL).</P> <P><B>Highlights</B></P> <P> <UL> <LI> The risk of coronary heart disease increased from a prediabetic range (≥110mg/dL) in women. </LI> <LI> Women have a higher relative risk of coronary heart disease associated with the fasting serum glucose level than men do. </LI> <LI> More careful glycemic control in women with hyperglycemia is needed to prevent coronary heart disease. </LI> </UL> </P>

우리나라 일부 의학 학술지에 게재된 의약품 광고의 과학적 근거에 대한 평가

안성복,최원,김철준,최성준,이강희,하경수,김현창 한국의료QA학회 2005 한국의료질향상학회지 Vol.12 No.1

Background : The promotion and advertisement of pharmaceuticals should be based on evidence from clinical trials. We conducted this study to assess whether the pharmaceutical advertisement claims in Korean medical journals had relevant references, and whether the claims were supported by the references. Methods : We reviewed pharmaceutical advertisements in five Korean medical journals issued during the first half of 1999 and during the first half of 2004. Three investigators independently reviewed the advertisements to see whether the studies quoted to endorse the advertising messages supported the corresponding claims. Using multiple logistic regression analyses, we investigated which factors were associated with the quality of the advertisement claims. Results : From the 550 advertisements in the five journals, we identified 157 different advertisements and 475 different promotional claims. Only 149 claims had at least one reference, and 105 claims had references of published article. We could find supporting evidences in the 90 claims. The factors which were associated with the quality of advertisement claims were category of drugs, category of claims, and the manufacturer characteristics. Claims for cardiovascular and endocrine drugs, and claims on efficacy, and claims of multinational company were more evidence-based. Conclusion : Majority of the pharmaceutical advertisement claims in Korea did not have appropriate references. Drug category, claim category, and the manufacturer characteristics were associated with the quality of advertisement claims, and the manufacturer characteristics was the most important determinants.

Development of the Korean Standardized Antimicrobial Administration Ratio as a Tool for Benchmarking Antimicrobial Use in Each Hospital

Kim Bongyoung,Ahn Song Vogue,Kim Dong-Sook,Chae Jungmi,정수진,어영,Kim Hong Bin,Kim Hyung-Sook,Park Sun Hee,Park Yoon Soo,최준용 대한의학회 2022 Journal of Korean medical science Vol.37 No.24

Background: The Korea National Antimicrobial Use Analysis System (KONAS), a benchmarking system for antimicrobial use in hospitals, provides Korean Standardized Antimicrobial Administration Ratio (K-SAAR) for benchmarking. This article describes K-SAAR predictive models to enhance the understanding of K-SAAR, an important benchmarking strategy for antimicrobial usage in KONAS. Methods: We obtained medical insurance claims data for all hospitalized patients aged ≥ 28 days in all secondary and tertiary care hospitals in South Korea (n = 347) from January 2019 to December 2019 from the Health Insurance Review & Assessment Service. Modeling was performed to derive a prediction value for antimicrobial use in each institution, which corresponded to the denominator value for calculating K-SAAR. The prediction values of antimicrobial use were modeled separately for each category, for all inpatients and adult patients (aged ≥ 15 years), using stepwise negative binomial regression. Results: The final models for each antimicrobial category were adjusted for different significant risk factors. In the K-SAAR models of all aged patients as well as adult patients, most antimicrobial categories included the number of hospital beds and the number of operations as significant factors, while some antimicrobial categories included mean age for inpatients, hospital type, and the number of patients transferred from other hospitals as significant factors. Conclusion: We developed a model to predict antimicrobial use rates in Korean hospitals, and the model was used as the denominator of the K-SAAR.

코로나19와 간호사의 직무스트레스 및 소진과의 관련성

윤세영,안성복 건강보험심사평가원 2022 HIRA RESEARCH Vol.2 No.2

Poster Session:PS 0528 ; Nephrology : Risk of Major Cardiovascular Events in the Incident Chronic Dialysis Patients: A Population-Based Study in Korea

( Hyun Wook Kim ),( Song Vogue Ahn ),( Shi Na Lee ),( Seung Jung Kim ),( Duk Hee Kang ),( Kyu Bok Choi ),( Dong Ryeol Ryu ) 대한내과학회 2014 대한내과학회 추계학술대회 Vol.2014 No.1

Background: To date, there has been no data on the detailed analyses for the incidences of major cardiovascular events in large-scale, Asian population-based cohort. Therefore, in this study, we approached these issues using the Korean Health Insurance Review and Assessment Service (HIRA) database. Methods: We analyzed eligible 30279 [22892 hemodialysis (HD) patients and 7387 peritoneal dialysis (HD) patients] incident chronic dialysis patients who were defi ned as starting dialysis therapy and remaining on that modality at least 3 months. The events were assessed by “major adverse cardiac (MACE) and cerebrovascular events” (“MACCE”). To estimate and to compare the incidence rate of each event, Poisson regression models and Cox proportional hazards models with the propensity score matching were used. Results: During median follow-up of 24.5 months, crude incidence rates of MACCE, MACE, all-cause mortality, non-fatal AMI, TVR, and non-fatal stroke were 182 (95% CI, 178-185), 138 (95% CI, 135-141), 116 (95% CI, 113-118), 18 (95% CI, 17-19), 17 (95% CI, 16-18), and 60 (95% CI, 58-62) per 1000 patient-years, respectively. When comparing all baseline covariate-adjusted relative risks of each end point between HD and PD patients, MACCE, MACE, all-cause mortality, non-fatal AMI, and TVR occurred signifi cantly more frequently in the patients on PD than in those on HD, whereas hemorrhagic subtype of non-fatal stroke was more frequent in the patients on HD than in those on PD. Conclusions: The results suggest that HD is overall superior to PD in terms of MACCE, as a composite end point, but analyzed by the separate end points, the comparison results are rather complex. The results of our study can help to give some basis for risk stratifi cation and customized care for major cardiovascular events in ESRD patients.

Clinical outcomes, when matched at presentation, do not vary between adult-onset Henöch-Schönlein purpura nephritis and IgA nephropathy

Oh, Hyung Jung,Ahn, Song Vogue,Yoo, Dong Eun,Kim, Seung Jun,Shin, Dong Ho,Lee, Mi Jung,Kim, Hyoung Rae,Park, Jung Tak,Yoo, Tae-Hyun,Kang, Shin-Wook,Choi, Kyu Hun,Han, Seung Hyeok International Society of Nephrology 2012 Kidney international Vol.82 No.12

Henöch-Schönlein purpura nephritis (HSPN) is considered a systemic form of immunoglobulin A nephropathy (IgAN). Although these are different pictures of a single disease, there are no studies directly comparing long-term outcomes of these two clinical entities. To clarify this, we studied 120 patients with biopsy-proven HSPN and 1070 patients with IgAN. The primary outcome was the composite of a doubling of baseline serum creatinine, end-stage renal disease, or death. Secondary outcomes included the individual renal outcomes or the rate of decline in estimated glomerular filtration rate. In the unmatched cohort, patients with HSPN had more vasculitic symptoms, more favorable histologic features, and were more commonly treated with steroids than patients with IgAN. The risk of reaching the primary outcome was significantly lower in HSPN patients than patients with IgAN (hazard ratio, 0.67). The 1:2 propensity score matching gave matched pairs of 89 patients with HSPN and 178 patients with IgAN, resulting in no differences in baseline conditions. In this matched cohort, there were no significant differences in reaching the primary and secondary outcomes between the two groups. Thus, after adjustment by propensity score matching, clinical outcomes did not differ between HSPN and IgAN, suggesting the two forms of the same disease have a similar prognosis.

Impact of Multiple Cardiovascular Risk Factors on the Carotid Intima-media Thickness in Young Adults: The Kangwha Study

Hoo Sun Chang,Hyeon Chang Kim,Song Vogue Ahn,허남욱,Il Suh 대한예방의학회 2007 예방의학회지 Vol.40 No.5

우리나라 일부 의학 학술지에 게재된 의약품 광고의 과학적 근거에 대한 평가

안성복,최원,김철준,최성준,이강희,하경수,김현창,Ahn, Song Vogue,Choi, Won,Kim, Chul Joon,Choe, Seong Choon,Lee, Kang Hee,Ha, Kyoungsoo,Kim, Hyeon Chang 한국의료질향상학회 2006 한국의료질향상학회지 Vol.12 No.1

Background : The promotion and advertisement of pharmaceuticals should be based on evidence from clinical trials. We conducted this study to assess whether the pharmaceutical advertisement claims in Korean medical journals had relevant references, and whether the claims were supported by the references. Methods : We reviewed pharmaceutical advertisements in five Korean medical journals issued during the first half of 1999 and during the first half of 2004. Three investigators independently reviewed the advertisements to see whether the studies quoted to endorse the advertising messages supported the corresponding claims. Using multiple logistic regression analyses, we investigated which factors were associated with the quality of the advertisement claims. Results : From the 550 advertisements in the five journals, we identified 157 different advertisements and 475 different promotional claims. Only 149 claims had at least one reference, and 105 claims had references of published article. We could find supporting evidences in the 90 claims. The factors which were associated with the quality of advertisement claims were category of drugs, category of claims, and the manufacturer characteristics. Claims for cardiovascular and endocrine drugs, and claims on efficacy, and claims of multinational company were more evidence-based. Conclusion : Majority of the pharmaceutical advertisement claims in Korea did not have appropriate references. Drug category, claim category, and the manufacturer characteristics were associated with the quality of advertisement claims, and the manufacturer characteristics was the most important determinants.

Improving Outcome of Capd: Twenty-Five Years’ Experience in a Single Korean Center

Han, Seung Hyeok,Lee, Sang Choel,Ahn, Song Vogue,Lee, Jung Eun,Choi, Hoon Young,Kim, Beom Seok,Kang, Shin-Wook,Choi, Kyu Hun,Han, Dae Suk,Lee, Ho Yung SAGE Publications 2007 Peritoneal dialysis international Vol.27 No.4

<B>Background</B><P> Continuous ambulatory peritoneal dialysis (CAPD) is an established treatment for end-stage renal disease (ESRD). We investigated the outcome of CAPD over a period of 25 years at our institution. </P><B>Methods</B><P> CAPD has been performed in 2301 patients in 25 years. After excluding patients with less than 3 months of follow-up and missing data, we evaluated 1656 patients who started peritoneal dialysis between November 1981 and December 2005. Data for sex, age, primary disease, co-morbidities, follow-up duration, cause of death, and cause of technique failure were collected. We also examined data for urea kinetic modeling (UKM), beginning in 1990, and peritonitis episodes, including causative organisms, starting in 1992. </P><B>Results</B><P> Compared to incident patients from 1981 - 1992, mean age and incidence of ESRD caused by diabetic nephropathy increased in patients from 1993 to 2005. Technique survival after 5 and 10 years was 71.9% and 48.1% respectively. Technique survival was significantly higher in patients who started CAPD after 1992 than in those who started before 1992. Peritonitis was the main reason for technique failure. Overall peritonitis rate was 0.38 episodes per patient-year, with a significant downward trend to 0.29 per patient-year over 10 years, corresponding to a decrease in gram-positive peritonitis. Patient survival after 5 and 10 years was 69.8% and 51.8% respectively. Patient survival improved significantly during 1992 - 2005 compared to 1981 - 1992 after adjustment for age, gender, diabetes, and cardiovascular comorbidities [hazard ratio (HR) 0.68, p @@<@@ 0.01]. Subgroup analysis based on UKM revealed that dialysis adequacy did not affect patient survival. However, diabetes (HR 2.78, p @@<@@ 0.001), older age (per 1 year: HR 1.06; p @@<@@ 0.001), serum albumin level (per 1 g/dL: increase, HR 0.52; p @@<@@ 0.05), and cardiovascular comorbidities (HR 2.32, p @@<@@ 0.01) were identified as significant risk factors. </P><B>Conclusion</B><P> Technique survival has improved due partly to a decrease in peritonitis, which was attributed to a decrease in gram-positive peritonitis. Patient survival has also improved considering increases in aged patients and ESRD caused by diabetes. The mortality rate of CAPD is still high in older, diabetic, malnourished, and cardiovascular diseased patients. A more careful management of higher risk groups will be needed to improve the outcome of CAPD patients in the future. </P>