중환자실 간호사의 환자안전문화인식, 비판적 사고성향, 투약안전역량과 투약오류 보고의도의 관계

이정원 중앙대학교 대학원 2024 국내석사

In intensive care units (ICU), the risk of medication errors is high, and the consequences of medication errors can be critical. Thus, improving the intention to report medication errors is important because it can prevent medication-related patient safety accidents and prevent recurrence by promoting reporting behavior. Therefore, this study aimed not only to provide a foundation for increasing the intention to report medication errors in the intensive care unit but also to assist in providing foundational data for developing a program aimed at enhancing future intentions to report medication errors in the ICU. This study aimed to investigate the relationships between ICU nurses' perception of patient safety culture, critical thinking disposition, medication safety competency, and intention to report medication errors. We conducted a descriptive survey study to investigate the potential differences in these factors among ICU nurses. Participants included clinical nurses with a minimum of six months' experience directly handling medication tasks in the ICU. Data collection occurred through an online self-report survey from September 16th and 17th, 2023. A total of 187 responses were analyzed using SPSS 26.0. The key findings are summarized as follows: Participants' perception of patient safety culture averaged 3.560.44 out of 5, critical thinking disposition averaged 3.660.33 out of 5, medication safety competency averaged 3.980.45 out of 5, and intention to report medication errors averaged 70.0720.40 out of 100. Differences in intention to report medication errors were noted based on participants' higher educational level(t=5.24, p<.001), longer total work experience(F=4.69, p=.004), elevated job position(t=2.81, p=.005), experience with medication safety education(t=3.05, p=.005), and absence of previous experience with medication errors(t=2.11, p=.036). Intention to report medication errors showed statistically significant positive correlations with total work experience(r=.22, p=.002), perception of patient safety culture(r=.45, p<.001), critical thinking disposition(r=.46, p<.001), and medication safety competency(r=.49, p<.001). Factors associated with the intention to report medication errors included previous experience with medication errors(β=-.16, p=.013), perception of patient safety culture(β=.19, p=.022), and critical thinking disposition(β=.19, p=.023), which explained 31.1% of the variance. This study comprehensively examined ICU nurses' perception of patient safety culture, critical thinking disposition, medication safety competency, and intention to report medication errors. Based on these findings, efforts to increase ICU nurses' intention to report medication errors should focus on fostering a positive and receptive patient safety culture for post-medication error experiences. Additionally, it emphasizes the need for structured educational programs to enhance patient safety culture and critical thinking disposition, necessitating regular program implementation. 중환자실은 투약오류발생위험과 발생 시 초래되는 결과의 심각성이 크다. 투약오류 보고의도를 향상시키는 것은 투약오류보고행위를 증진시켜 투약과 관련된 환자안전사고를 예방하고 재발을 방지할 수 있어서 중요하다. 따라서 본 연구는 중환자실의 투약오류 보고의도를 높이기 위한 기반을 제공할 뿐 아니라 향후 중환자실 투약오류 보고의도 증진을 위한 프로그램을 개발하는데 기초자료를 제공하는 데에 도움이 되고자 한다. 본 연구는 중환자실 간호사의 환자안전문화인식, 비판적 사고성향, 투약안전역량 및 투약오류 보고의도의 수준을 알아보고, 환자안전문화인식, 비판적 사고성향, 투약안전역량과 투약오류 보고의도의 관계를 파악하기 위해 시행된 서술적 조사연구이다. 연구 대상자는 중환자실에 근무 중인 임상간호사로서 투약업무를 직접 수행하는 경력 6개월 이상 간호사이며, 자료 수집은 2023 년 9월 16일부터 2023년 9월 17일까지 온라인 설문조사를 통하여 자가보고 형식으로 진행하였으며, 최종 187부의 자료를 SPSS 26.0을 이용하여 분석하였다. 본 연구 결과는 다음과 같다. 첫째, 대상자의 환자안전문화인식은 5점 만점에 평균 3.560.44점, 비판적 사고성향은 5점 만점에 평균 3.660.33점, 투약안전역량은 5점 만점에 평균 3.980.45점, 투약오류 보고의도는 100점 만점에 평균 70.0720.40점이었다. 둘째, 대상자의 투약오류 보고의도는 최종학력이 높을수록(t=5.24, p<.001), 총근무경력이 길수록(F=4.69, p=.004)), 직위가 높을수록(t=2.81, p=.005), 투약안전 교육경험이 있는 경우(t=3.05, p=.005), 투약오류 경험이 없는 경우(t=2.11, p=.036)에 따라 유의한 차이가 있었다. 셋째, 대상자의 투약오류 보고의도는 총근무경력(r=.22, p=.002), 환자안전문화인식(r=.45, p<.001), 비판적 사고성향(r=.46, p<.001), 투약안전역량(r=.49, p<.001)과 모두 통계적으로 유의한 양의 상관관계가 있었다. 넷째, 대상자의 투약오류 보고의도 관련요인으로는 투약오류 경험(β=-.16, p=.013), 환자안전문화인식(β=.19, p=.022), 비판적 사고성향(β=.19, p=.023)이었으며, 설명력은 31.1%였다. 본 연구에서는 중환자실 간호사의 환자안전문화인식, 비판적 사고성향, 투약안전역량, 투약오류 보고의도를 파악하였고, 투약오류 보고의도의 관련요인을 확인하였다. 이 연구 결과를 바탕으로 중환자실 간호사의 투약오류 보고의도를 높이기 위해서는 투약오류 경험 후 오류보고 시 부정적인 경험이 없도록 긍정적이고 수용적인 환자안전문화를 조성하려는 노력이 필요하고, 비판적 사고성향을 향상시킬 체계적인 교육프로그램 구축과 함께 주기적인 프로그램 운영의 필요성이 있다.

간호관리자의 진성리더십과 팀 내 심리적 안전감, 간호사의 투약오류 보고의도

문선영 연세대학교 대학원 2020 국내석사

The medication error report is one of the improvement activities to reduce medication error, and improving the intention of reporting medication error that can predict medication error rate is one of the measures to prevent patient safety accidents. Thus, in this study, a descriptive investigation study was conducted to identify the relationship between the authentic leadership of the nurse's perceived nursing manager, the psychological safety in work team, and the intent to report the medication error. This study was conducted on 167 nurses using structured questionnaires. The collected data were analyzed by the SPSS WIN 26.0 and SPSS process macro ver 3.4.1 software program and descriptive statistics, such as independent t-test, ANOVA, Pearson/s correlation coedfficient, multiple linear regressionanlysis, Scheff􌝸 test, and mediation analysis. Mediation alalysis was conducted using SPSS macro process model 4 to examine mediationi effect of psychological safety in work team on the relationship between authentic leadership of nursing managers and intention to medication error reporting of nurses. The study found that authentic leadership was 3.41±.56 points, psychological safety was 3.50±.52 points, and intention to medication error reporting was 71.06±19.13 points, and factors affecting the intention to medication error reporting of nurses were participation in education related to medication error (β =.28, p=.003), psychological safety in work team( =.21, p=.β 004), experience of near miss related medication error(β=-.17 and p=.041). The specific indirect effect through psychological safety in work team was β=.2042(CI=.0025-4892). Between the authentic leadership of nursing manager and the intention of the nurse to report an error in medication, psychological safety in the team has been verified to play a mediating role. These results showed that it is necessary to establish a systematic and repetitive medication error-related education program that includes the definition of medication error, cases, medication error reporting system, and the importance of error reporting, and to overcome the negative experiences of nurses who have experienced medication error reporting, and to make efforts to develop authentic leadership of nursing managers and to enhance psychological safety in work team. 투약오류보고는 투약오류를 감소시키기 위한 개선 활동 중 하나로 투약오류보고율을 예측할 수 있는 투약오류 보고의도를 향상시키는 것은 환자안전사고를 예방하고 재발을 방지하기 위한 방안 중 하나이다. 이에 본 연구에서는 간호사가 인식하는 간호관리자의 진성리더십, 팀 내 심리적 안전감과 투약오류 보고의도와의 관계를 파악하기 위한 서술적 조사연구를 시행하였다. 본 연구는 간호사 167명을 대상으로 구조화된 설문지를 이용하여 연구를 진행하였으며, 자료 분석은 SPSS statistics 26.0과 SPSS Process macro ver 3.4.1 프로그램을 이용하여 빈도분석, 기술통계, independent t-test, ANOVA, Pearson’s correlation coefficient, 다중회귀분석, 부트스트랩을 이용한 Process macro model 4로 분석하였다. 연구결과 진성리더십 3.41±.56점, 심리적 안전감 3.50±.52점, 투약오류 보고의도 71.06±19.13점으로 나타났고, 투약오류 보고의도에 영향을 미치는 요인으로는 3회 이상의 투약오류 관련 교육참여(β=.28, p=.003), 팀 내 심리적 안전감(β=.21, p=.004), 투약 근접오류경험(β=-.17, p=.041)으로 나타났다. 진성리더십이 투약오류 보고의도에 미치는 영향에서 심리적 안전감을 통한 간접효과가 검증되어(β=.2042, CI=.0025~.4892) 간호관리자의 진성리더십과 간호사의 투약오류 보고의도 사이에서 팀 내 심리적 안전감은 매개 역할을 하는 것으로 검증되었다. 이상의 연구결과를 종합하면, 투약오류 보고의도를 높이기 위해서는 투약오류에 관한 정의, 사례, 투약오류보고 체계, 오류 보고의 중요성이 포함된 체계적이고 반복적인 투약오류 관련 교육 프로그램을 구축하는 것이 필요하고, 투약오류보고로 겪은 간호사의 부정적인 경험을 극복하고, 심리적 안전감을 증진시키기 위한 조직적 접근 전략과 간호관리자의 진성리더십 개발을 위한 노력이 필요하다.

의료 종사자의 의료오류 공개에 대한 인식과 환자안전문화와의 관계

이덕주 경희대학교 공공대학원 2019 국내석사

This study aims at providing the basic data for establishing a safe medical environment by investigating and identifying the perception of doctors and nurses, working at general hospitals, about the public disclosure of medical errors and the patient safety culture as well as their relationships. Participants of this study were 267 doctors and nurses working in the general hospital in Geyonggi-do, and data were collected using the self-report questionnaire from July 1 to August 31, 2018. It was found that the responsibility of the health care practitioner for the public disclosure of medical errors is high, but the perception about positive and negative effects of the disclosure is somewhat low. In addition, the health care practitioners with direct experiences of medical errors, their perception about the effect of the public disclosure of medical errors was found to be more negative as well as their perception about the report on medical errors. Moreover, it was found that there is a direct relationship between the perception about the public disclosure of medical errors and the patient safety culture, and verified that the perception of the health care practitioner for the public disclosure of medical errors is one element of the patient safety culture. In conclusion, in order to raise the awareness about the patient safety culture through the public disclosure of medical errors, it would be necessary to prepare the guideline and systematic education on the public disclosure of medical errors as well as the social and institutional environment suitable for the public disclosure of medical errors and the programs for supporting of the health care practitioners experiencing medical errors.

간호사의 투약오류 분석 및 개선방안

이순영 을지대학교 임상간호대학원 2008 국내석사

본 연구는 간호사의 투약오류에 대한 분석을 통하여 간호사의 업무 중 간호수행 빈도가 높은 투약행위를 중심으로 오류를 예방하기 위한 다양한 개선방안을 제시하며, 안전한 투약간호를 위한 기초 자료를 제공하고자 하였다. 이를 위해 투약오류와 관련된 간호사의 인식과 경험을 분석하고 투약오류를 개선하기위한 다양한 중재방법을 파악하였다. 본 연구의 대상자는 4개 의료기관에서 임상간호사를 대상으로 자료를 수집하였으며, 직접 투약업무를 담당하는 418명의 자료를 분석하였다. 연구의 설문지는 김은경 등(1998)의 투약오류의 경험과 인식의 도구를 이용하여 재구성한 26개 투약오류 사례이며, 투약오류의 원인과 대처방안에 관한 문항은 기계적/물리적 원인, 시스템 관리의 원인, 간호의 원인, 환자의 원인으로 15개의 문항으로 구성하였다. 자료수집기간은 2007년 9월 17일부터 10월 15일까지 이었으며, 자료의 분석은 빈도와 백분율로 비교?분석하였다. 본 연구의 결과, 간호사가 투약오류라고 인식하는 26가지 사례 중 높은 빈도의 오류는 ‘과다용량을 투약하는 경우(96.7%)’, ‘투여경로가 잘못되어 투약하는 경우(96.4%)’, ‘다른 환자에게 투약하는 경우(95.5%)’의 순이었다. 투약오류 경험은 최근 3개월 동안 경험한 투약오류로 경험빈도가 높은 투약오류의 사례는 ‘투약 전에 미리 의무기록에 서명을 하는 경우(59.6%)’, ‘환자 보호자에게 약을 건네주고 복용여부를 확인 하지 않은 경우(59.1%)’, ‘분할하여 먹는 물약을 한꺼번에 환자에게 주고 일정량씩 먹도록 교육(57.2%)’의 순이었다. 투약오류가 일어나는 원인에 대하여 ‘투약 시 5right를 지키지 않아서(83.3%)’,‘부정확한 의사소통(81.8%)’, ‘제품의 모양이 비슷해서(79.9%)’라고 인식했으며, ‘교육 또는 훈련이 안된 상황에서의 투약으로(73.9%)’순으로 응답하였다. 간호사가 제시하는 투약오류의 개선방안은 ‘정확한 의사소통의 도구 마련이 필요(99.0%)’, ‘표준화된 지침의 필요(98.3%)’, ‘약품에 대한 신속한 정보의 제공(98.3%)’, ‘비슷한 약품의 별도 보관(98.1%)’, ‘투약 시 5right 강화(97.8%)’의 순이었다. 심각한 투약오류의 경험을 알아보기 위해 환자에게 해가 큰 투약오류에 대한 간호사의 인식을 조사하였는데, ‘다른 환자에게 투약을 함(98.7%)’, ‘처방된 약과 다른 약을 투약(92.6%)’, ‘용량을 과다용량으로 투약함(90.3%)’, ‘투여경로가 잘못 되는 경우(67.6%)’, ‘유효기간이 지난 약물을 투여함(44.7%)’의 순이었다. 투약오 류의 상황에서 근무시간은 낮번(day) 61.7%, 투약오류 약물은 주사제가 65.4%이 었으며, 투약오류 후 환자손상의 정도는 환자에게 해가 없는 경우가 65.4%로 조 사 되었다. 투약오류 후 보고유무는 49.4%가 보고를 하지 않은 것으로 나타났 고, 그 이유에 대하여 ‘환자에게 문제가 생기지 않았으므로 보고하지 않았다 (29.2%)’ 라고 응답하였다. 투약오류 후 병원의 조치는 부분적으로 ‘개별면담’이 이루어진 경우가 29.6%가 많았으며, ‘징계 또는 반성문’을 쓴 경우도 21.0%로 나 타났다. 투약오류를 경험 한 뒤 간호사들이 대부분 ‘다음 투약업무에서 더 확인 한다(67.9%)’는 의견이 많았지만, ‘죄책감으로 우울이 지속된다’고 응답한 경우도 18.5%이었다. 이상의 연구결과를 통해 투약오류의 인식과 경험은 인식이 낮았던 문항에서 투약오류 경험이 높은 빈도로 나타남을 알 수 있었다. 따라서 투약오류에 대한 인식을 향상시킬 수 있는 간호 교육프로그램 개발 및 효과에 대한 연구가 필요 하고, 투약오류의 원인을 분석하여 개선하고자 하는 대학과 의료기관의 노력과 임상에 근접한 교육 프로그램 개발과 표준화된 지침의 마련이 필요하다. The purposes of this study are to 1)describe nurses' perception of medication error, 2)examine the causes of medication errors and 3) provide basic data for care quality improvement. A descriptive, survey design was used. A approval for study was obtained from institutions and university ethics committee. Subjects were informed about the study and that their participation was voluntary and assured of anonymity. Written consents were obtained from the participants at enrollment. A pilot test was conducted with a convenient sample of 10 nurses from a general hospital. Subjects were asked to complete a survey consists of 26-item and describe adequacy of the survey and research procedure.Length of time for completion was measured.Based on subject's comments procedures and scales were modified. Researcher contacted the department head obtained agreement. Researcher met potential subjects and explained the purposes, procedures, benefits and no risks involved in the study. Subjects were given the chance to ask questions about the study. After written consents were obtained, and the survey was distributed. The return rate was 83.6%. 26 item on medication error perception and experience, 15 item on cause of error and prevention strategies. Data were coded, entered, and analyzed using SPSS version 14.0. Data were collected from nurses(N=418) working in 3 general hospital and a university hospital. Data were collected between 17 to October 15 in 2007. Data were entered and analyzed by descriptive statistic; range, frequency and percentage. The most frequently perceived medication error was ‘wrong dose error'(96.7%), followed by wrong route error'(96.4%) and 'wrong patient error(95.5%)'. The most frequent types of error experienced by nurses during the 3 months was ‘in advance recording in patient's chart before administration of the medication'(59.6%), followed by ‘not checking whether the patient took the medication'(59.1%). As for the cause of medication error, the response was highest in ‘non-compliance with 5-right when administering medicine'(83.3%). followed by 'incorrect communication'(81.8%), and 'due to similar shape of medicine(79.9%)'. The error improvement strategies nurse suggested where 'to have accurate communication channel'(99.0%), 'to have standardized medication guideline'(98.3%), 'to provide information on new medicine or any changes of medication'(98.3%), 'to store similar shaped medicine in separate places'(98.1%), 'to educated nurses to keep 5-rights'(97.8%). In conclusion it is necessary to develop an education program that will prevent medication error and for quality improvement.

3차 병원 입원환자의 지참약 불일치 검토 및 중재활동 결과 분석 연구

김혜미 조선대학교 임상약학대학원 2024 국내석사

Background : The problem of drug discrepancies (DD) occurs during the transition from personal medications to hospital medications, leading to potential medication errors. Such errors are preventable and can extend hospital stays and increase readmission risks. Despite existing reports on the types and intervention rates of DDs, there is a lack of in-depth analysis of intervention types, methods, and influencing factors. This study aims to provide a more detailed understanding to improve the efficiency of medication reconciliation (MR). Methods : This retrospective study reviewed Electronic Medical Record of patients admitted to Chosun University Hospital between January and March 2023, who had their personal medications switched to hospital medications within five days of admission. The study focused on analyzing the types and frequencies of DDs, the interventions made, and the factors influencing these discrepancies. For the basic characteristics of the study subjects, categorical variables were expressed as frequencies (N) and percentages (%), while continuous variables were expressed as means ± standard deviation. Fisher’s exact test was used to test the significance of intervention outcomes, and multivariable logistic regression analysis was employed to analyze the impact of patient characteristics and drug characteristics on DD. Results : Among the reviewed cases, 79 instances of DDs were identified and addressed. The most common intervention was correcting dosage errors in the prescribed hospital medications (68.35%). Acceptance rate of these interventions by medical staff was high (73.24%), with dosage errors being the most frequently accepted intervention (70.69%). Similar patterns were observed for non-accepted interventions, though specific reasons for non-acceptance were not ascertainable. Conclusion: MR significantly reduces medication errors related to unintended discrepancies, proving to be an effective method in preventing adverse drug reactions. However, the process is time-consuming and requires thorough comparison and communication of medication details between personal and hospital medications. The study highlights the need for prioritizing MR activities for patients at higher risk of DDs to optimize medication safety and healthcare outcomes. Keywords: Drug Discrepancies, Medication Reconciliation, Medication Error

임상간호사의 의약품 전환 경험 : 질적 연구

한가인 중앙대학교 심리서비스대학원 2023 국내석사

본 연구는 임상간호사들에게서 고의성이 없이 발생하는 의약품 전환의 의미에 대한 이해를 목표로 하였다. 의료 환경 내에서 간호사는 특수한 집단으로서 의약품의 보관, 관리, 사용에 대한 전문성을 갖고 있다고 인식되지만, 그럼에도 불구하고 간호사들에 의해 고의성 없이 행해지는 의약품 전환에 대해 심층적인 측면에서 이해가 필요하다. 따라서 본 연구는 기존에 보고되지 않은 간호사의 의약품 전환에 대하여 간호사의 경험을 밝히고자 표적 집단 심층 좌담(FGD)을 통한 질적연구 방법을 시행하였다. 2023년 5월 23일부터 5월 30일까지 상급종합병원에서 근무하는 22명의 임상간호사를 무작위로 선출하여 경력에 따라 4개의 그룹으로 분류하였다. 그중 3~5년 차 사이의 그룹은 임상에서 의약품을 가장 많이 다루는 집단이기에 2그룹으로 구성하였다. 집단토론 시작 전 긴장감 완화와 솔직한 답변을 위해 라포를 형성할 수 있는 시간을 마련하였고, 1시간에서 1시간 40분 정도의 토론을 통해 자료를 수집하고, Van Kaam (1969)의 현상학적 방법론을 사용하여 자료를 분석하였다. 연구 결과, 간호사의 의약품 전환에 대한 이해는 "의약품 전환의 의미", "의약품 전환의 이유", "의약품 전환의 영향과 결과", "의약품 전환 개선"이라는 네 가지 범주로 나타났다. 간호사는 의약품 전환의 개념을 투약 오류와 혼동하였고, 문제가 있는 행위에 대해 전혀 인지하지 못했다. 의약품 전환의 형태는 다양할 것이라 예상했으며 그것은 환자의 안전과 관련하여 발생한다고 하였다. 근무 경력과 연관성에서는 경력이 낮을수록 전환에 대한 두려움이 있는 것으로 나타났고, 의약품 전환은 시스템의 개선이나 추가가 아닌 불필요한 절차를 단순화하거나 직무의 스트레스 상황에서 벗어나고 전환의 범주에 대한 교육이 시행되면 감소할 것이라 하였다. 토론을 통해 간호사는 의약품 전환의 의미를 깨닫고, 문제 해결 방법을 스스로 탐색하며, 앞으로의 의약품 전환을 예방하기 위해 노력하였다. 본 연구는 고의성 없는 의약품 전환으로 인해 보고되지 않는 오류들이 환자 안전, 병원의 손실, 간호사 개인의 법적 및 윤리적 문제로 발생할 수 있다는 점을 강조하며, 조직문화를 개선하고 올바른 전략을 개발하기 위한 데이터를 제공하였다는 데 의의가 있다. The purpose of this study was to understand the significance of unintentional medication diversion among clinical nurses. In the healthcare environment, nurses are recognized as a specialized group with expertise in the storage, management, and use of medications. However, despite this recognition, there is a need for a profound understanding of non-intentional medication diversion carried out by nurses. Therefore, this study employed a qualitative research method using focus group discussions (FGDs) to uncover nurses' experiences and perspectives regarding medication diversion, which has not been extensively reported before. We collected the data for this study through group discussions conducted from May 23 to May 30, 2023. A total of 22 clinical nurses working at a tertiary referral hospital were randomly selected and divided into four groups based on their years of experience. Specifically, the group consisting of nurses with 3 to 5 years of experience was divided into two sub-groups, as they handle a significant number of drugs in clinical practice. Prior to the initiation of the focus group discussions, sufficient time was allocated to establish rapport, alleviate tension, and encourage participants to provide candid responses. The discussions lasted between 1 hour and 1 hour 40 minutes, during which data were collected. Subsequently, the collected data were analyzed using Van Kaam (1969) phenomenological methodology analytical methods. The research findings revealed four categories that represented nurses' understanding of medication diversion: "meaning of medication diversion", "reasons for medication diversion", "impact and outcomes of medication diversion", and "improvement of medication diversion". Nurses tended to confuse the concept of medication diversion with medication errors and lacked awareness of problematic practices. They anticipated diverse forms of medication diversion that could potentially compromise patient safety. The findings also highlighted that nurses with fewer years of experience expressed greater concerns about medication diversion. They suggested that addressing medication diversion should involve streamlining unnecessary procedures, reducing stress in their work environment, and implementing educational interventions targeting the categories of medication diversion. Nurses realized the significance of medication diversion, explored problem-solving approaches, and made efforts to prevent future medication diversion. These results emphasizes the potential consequences of unintentional medication diversion, which can lead to unreported errors with implications for patient safety, hospital losses, and legal and ethical issues for individual nurses. It holds significance in providing data to enhance organizational culture and develop effective strategies.

GPT와 BERTopic을 이용한 MAUDE (Manufacturer and User Facility Device Experience) 데이터베이스 내 의료기기 사용 오류 분석

김시은 국립부경대학교 대학원 2024 국내석사

Medical device use errors are crucial to note because they directly impact patient safety and the quality of medical services. Previous studies have primarily relied on manual reviews by experts to identify and analyze medical device use errors from the MAUDE database, which collects adverse event reports. To overcome these limitations, this study utilized the large language model GPT-3.5-turbo and advanced prompting techniques to automatically extract use errors from unstructured narrative data in the MAUDE database and performed topic analysis using BERTopic. This approach aimed to minimize subjectivity and inefficiency that might arise in previous studies and to more clearly identify patterns and major causes of medical device use errors. Applying this method to adverse event data for the 'Automatic Delivery Peritoneal System,' the main causes of use errors were found to be device setup errors and user interface issues, which were mainly attributed to insufficient user training. This study provides important foundational data that can contribute to medical device safety by systematically analyzing medical device use errors, offering a structured framework for design and manufacturing improvements, accident prevention and management, and enhancing the consistency and reproducibility of data analysis.

The Development of a Next-Generation Human Reliability Analysis: Systems Analysis for Formal Pharmaceutical Human Reliability (SAFPH℞ )

Zheng, Xi State University of New York at Buffalo ProQuest D 2020 해외박사(DDOD)

Medication errors originating in community pharmacies are serious patient safety hazards. However, due to the complexity of the community pharmacy environment, current experimental and observational studies are insufficient to address these problems. This research aims to create a novel, formal, proof-based approach to human reliability analysis (HRA) that will provide pharmacies with the ability to accurately predict error rates, understand why errors are occurring, and better engineer their system to mitigate the errors.Traditional HRA can accurately predict human error rates in a number of environments. However, they are limited in that they are static and thus not able to handle the dynamic environmental elements that can impact human performance. To address this and allow analysts to accurately predict medication error rates, we have developed a next-generation HRA called the Systems Analysis for Formal Pharmaceutical Human Reliability (SAFPHR). This method addresses the limits of previous HRAs by combining concepts from the Cognitive Reliability and Error Analysis Method (CREAM) HRA with probabilistic model checking, a computational tool for automatically proving properties about complex, stochastic systems.By using different estimation methods and cognitive assumptions, we have fully developed the modeling and predictive capabilities of three versions of SAFPHR: basic SAFPHR, CPC-effect extended SAFPHR, and mode-effect extended SAFPHR. These three versions can collectively produce six different methods of computing error rates. To determine which of these estimates were the most accurate and valid, we formally modeled a full, generic, pharmacy dispensing procedure.We then used SAFPHR to make predictions about overall error rates as well as error rates originating from different stages of the dispensing process. These values were computed for all six estimation approaches and then compared with real, comprehensive error rates published in the literature. One method consistently produced accurate predictions both for the overall error rate and the individual stages. The results have important implications for pharmacy because they show that SAFPHR could be used to reduce medication error rates and thus significantly improve patient health and safety. Given its success in pharmacy and the generic nature of its underlying theory, SAFPHR could be used as a general HRA to improve safety and reliability in other critical domains. More avenues of future research are explored in the end.

선택 실수로 인한 KCl 처방 오류 감지 모델 개발 및 검증 : 딥러닝 기반의 KCl 처방 필요군 예측 모델 활용

민지영 성균관대학교 일반대학원 2021 국내석사

약물 오류(medication error)는 환자에게 위해를 줄 수 있는 중요한 문제로 의료 현장에서 빈번하게 발생하고 있다. 약물 오류는 의료인이 약물을 처방하는 단계부터 환자에게 투약하는 단계에 이르기까지 약물과 관련된 모든 단계에서 발생할 수 있다. 약물 처방 단계에서 발생하는 약물 오류를 처방 오류(prescribing error)라고 하며, 예방 가능한 약물 오류의 대부분이 약물 처방(physician ordering) 단계에서 발생한다. 이러한 처방 오류를 극복하기 위해 Computerized physician order entry (CPOE) 시스템을 도입하였다. 하지만, CPOE 시스템 도입으로 인해 처방 오류 예방 가능해진 반면 잘못된 환자 선택, 잘못된 약물 선택 등의 예기치 않게 발생한 새로운 처방 오류들이 부각되고 있다. 이러한 선택 실수에 의한 약물 처방 오류를 예방하기 위한 시도가 이루어지고 있지만 많지 않으며, 완전히 예방하기에 충분하지 않다. 본 연구에서는 약물 처방 단계에서 환자를 잘못 선택하거나, 처방할 약물 종류를 실수로 잘못 선택하여 발생하는 약물 처방 오류를 “부적절한 처방”이라고 정의하였으며, 이러한 “부적절한 처방”을 감지하는 머신 러닝 기반의 모델을 개발하고 검증하는 것을 목표로 하였다. 이 연구는 삼성서울병원의 약물 처방 데이터를 활용한 후향적 연구이다. 2015 년 1 월 1 일부터 2019 년 12 월 31 일 기간 동안 삼성서울병원에 입원하거나 응급실에 내원한 성인 환자에게 처방된 주사제 약물 처방 케이스가 포함되었다. 처방 단계부터 각별한주의가 필요한 고위험 약물인 KCl 정맥 주사약을 대상으로 연구를 진행하였다. 연구는 “모델 개발 및 내부 검증 단계”와 차트 리뷰를 통한 “모델의 임상적인 검증 단계” 두 단계로 수행되었다. 일반 KCl 처방군은 총 146,364 건 이었으며, 부적절한 KCl 처방군에 629 종류로 구성된 430,842 건의 약물 처방이 포함되었다. 첫 번째 단계에서 모델이 스크리닝 툴로 사용될 수 있도록 모델의 임계 값을 선정하였다. 선정된 임계 값에서 sensitivity 가 0.98 이었으며, AUROC (Area under the receiver operator characteristics curve), PPV, NPV, accuracy 가 0.9 보다 높아 좋은 성능을 보여주었다. 두 번째 단계에서 부적절한 처방 가능성이 높은 KCl 처방 그룹에 대한 모델 결과를 차트 리뷰를 통해 0.95 보다 높은 sensitivity 와 NPV 를 가지는 것을 확인 할 수 있었다. 이는 우리가 개발한 모델이 실제 임상 환경에 적용되었을 때 스크리닝 툴로써 부적절한 처방을 잘 감지할 수 있는 잠재력이 있음을 보여준다. 모델이 전자 건강 기록(Electronic Health Record, EHR)을 기반으로 개발되어 의료인의 추가적인 업무 없이 자동적으로 부적절한 처방의 위험도를 계산할 수 있어 임상 현장에 적용되었을 때 업무 부담을 최소화하여 워크플로우에 통합 적용될 수 있을 것으로 기대된다. Medication errors, which are the potential international harm for patients, can occur in the whole process from the physician ordering phase to the administration to the patient phase. Among them, prescribing errors are common but preventable. Computerized prescriber order entry (CPOE) is a representative of the prevention effort of prescribing errors. However, unintended errors, such as wrong patient and wrong medication selection, derived from CPOE occurred against the intended purpose of adopting it. There are several studies to prevent these errors, but not enough to fully overcome them. The prescribing errors resulting from the wrong patient and wrong medication selection are defined as “unintended prescribing errors” in this study. This study aims to develop and validate a machine learning-based unintended prescribing error detection model for inpatients and emergency department (ED) patients. The study was conducted retrospectively based on electronic health records (EHRs) data of Samsung Medical Center. The study included injectable drug prescriptions for adult inpatients or ED patients between January 2015 and December 2019. We focused on prescriptions for KCl (Potassium chloride for injection), a highalert medication that requires special precautions. The study was conducted in two stages: “Model development and internal validation” and “Clinical validation for the model” through chart review analysis. There was a total of 146,364 cases for “general KCl prescription” and 430,842 cases for “unintended KCl prescription” consisting of 629 medications, not KCl. In the first stage, AUROC, sensitivity, specificity, PPV, NPV, F1-score, and accuracy were 0.952, 0.976, 0.808, 0.937, 0.919, 0.860, 0.933, respectively at the selected threshold for screening tool. In the second stage, the model performances in the group, which have a high potential for unintended KCl prescription, were 0.968 for sensitivity and 0.959 for NPV. The developed model is expected to detect potential unintended prescribing errors through effective screening when applied to a clinical setting.

Reducing Dosing Errors and Increasing Clinical Efficiency in Guatemala: First Report of a Novel mHealth Medication Dosing App in a Developing Country

Arevalo, John Ben Harvard University ProQuest Dissertations & Theses 2018 해외박사(DDOD)

Background: Medication errors are known to be a widespread problem affecting patient safety and treatment efficacy. We were approached by a Guatemalan clinic interested in piloting an app to aid their junior clinicians in correctly calculating medication dosages.Methods: We programmed a medication-dosing app using CommCare. In a prospective study among junior physicians at a small high-throughput clinic, we primarily assessed the app's dosing accuracy and efficiency. Secondarily, we measured the app's usability and effect on patient centredness.Results: Six clinicians aged 21–24 tested the app. Among 366 prescriptions, dosages were 40% more likely to be correct when calculated using the app (relative risk: 1.39; 95% CI 1.16 to 1.68; p=0.0005). Accuracy improved from 64.7% (N=156) to 92.4% (N=210). Using the app in a time-constrained context improved clinician efficiency by over 20% with a decrease in average consultation time of 1.5 min (p<0.0001) to 5.23 min on average (N=178). However, questionnaires revealed most clinicians did not believe the app improved efficiency, and none thought its recommendations were ‘always accurate’. No change in patient-centredness was observed (N=167).Conclusions: The app was shown to be safe and efficient. Making this app available to junior physicians may significantly improve patient safety by enhancing dosing accuracy. This study demonstrates that dosing apps can be an efficacious means of decreasing medication errors in developing countries. We found that different strategies to introduce novel apps to providers might improve providers’ trust in the technologies and thereby make apps more efficacious.