NUR 707 Wk 6 IT and Data Driven Decision Making Discussion
NUR 707 Wk 6 IT and Data Driven Decision Making Discussion
4. The readings for this week are examples of published Quality Improvement studies. (Sometimes, reading works for structure and mechanics is helpful because we are distanced from the knowledge content that is directly related to our project.) Discuss insights obtained on how the authors navigated the problem, intervention, chosen outcomes, and benchmark setting.
Quality in Action Electronic Medical Record–Based Radiation Oncology Toxicity Recording Instrument Aids Benchmarking and Quality Improvement in the Clinic Kevin Albuquerque, Kellie Rodgers, Ann Spangler, Asal Rahimi, and DuWayne Willett University of Texas Southwestern Medical Center, Dallas, TX ASSOCIATED CONTENT Appendix and Data Supplement available online Abstract Purpose The on-treatment visit (OTV) for radiation oncology is essential for patient management. Radiation toxicities recorded during the OTV may be inconsistent because of the use of free text and the lack of treatment site–speciﬁc templates. We developed a radiation oncology toxicity recording instrument (ROTOX) in a health system electronic medical record (EMR). Our aims were to assess improvement in documentation of toxicities and to develop clinic toxicity benchmarks. Methods A ROTOX that was based on National Cancer Institute Common Terminology Criteria for Adverse Events (version 4.0) with ﬂow-sheet functionality was developed in the EMR. Improvement in documentation was assessed at various time intervals. High-grade toxicities (ie, grade $ 3 by CTCAE) by site were audited to develop benchmarks and to track nursing and physician actions taken in response to these. Results A random sample of OTV notes from each clinic physician before ROTOX implementation was reviewed and assigned a numerical document quality score (DQS) that was based on completeness and comprehensiveness of toxicity grading. The mean DQS improved from an initial level of 41% to 99% (of the maximum possible DQS) when resampled at 6 months post-ROTOX. This high-level DQS was maintained 3 years after ROTOX implementation at 96% of the maximum. NUR 707 Wk 6 IT and Data Driven Decision Making Discussion
For months 7 to 9 after implementation (during a 3-month period), toxicity grading was recorded in 4,443 OTVs for 698 unique patients; 107 episodes of high-grade toxicity were identiﬁed during this period, and toxicity-speciﬁc intervention was documented in 95%. Conclusion An EMR-based ROTOX enables consistent recording of treatment toxicity. In a uniform sample of patients, local population toxicity benchmarks can be developed, and clinic response can be tracked. DOI: https://doi.org/10.1200/JOP. 2017.025163; published online ahead of print at jop.ascopubs.org on February 13, 2018. e186 INTRODUCTION Modern radiation therapy, by virtue of better modes of radiation delivery1 has achieved much success in control of tumors at different organ sites. Data about reduction in toxicity from this modern Volume 14 / Issue 3 / March 2018 n Journal of Oncology Practice Copyright © 2018 by American Society of Clinical Oncology Downloaded from ascopubs.org by 18.104.22.168 on December 12, 2019 from 066.207.141.241 Copyright © 2019 American Society of Clinical Oncology. All rights reserved. EMR-Based Radiation Oncology Toxicity Recording Instrument radiation therapy is available only for few disease sites, such as breast and prostate cancer.2-4 Most of the data for acute radiation toxicity come from small trials or large co-operative studies in which careful patient selection does not translate to the general patient population.5 In a local clinic practice, evaluation of toxicities is limited to case discussions of unusual severe toxicities during morbidity reviews.6 Because comprehensive standardized documentation is lacking, incidences and patterns of important toxicities cannot be verified. As part of the usual radiation therapy course, each patient has a weekly physician visit, labeled the on-treatment visit (OTV), which typically is documented in the electronic medical record (EMR) with free text. NUR 707 Wk 6 IT and Data Driven Decision Making Discussion
Difficulty with data extraction and reporting pose challenges in the use of EMRs for population analytics.7 Because of our lack of systematic documentation, we implemented a comprehensive prospective radiation oncology toxicity recording instrument (ROTOX) in the health system EMR, and we associated cumulative reporting for individual patients or specific populations directly from the EMR. The aims of our quality improvement project with ROTOX were to demonstrate consistent and durable improvement in toxicity documentation quality and to enable auditing of clinic response in relation to high-grade (ie, grade 3 or greater) toxicity. METHODS The Department of Radiation Oncology is part of a National Cancer Institute–designated cancer center. With 17 radiation oncologists, the department treats an average of 150 to 160 patients per day at its two main clinics and dedicated radiosurgery suite on the university campus. These patients include a large number of uninsured patients from the local county hospital and a significant pediatric population. Thus, there is a unique mix of patient population with diversity of a real-world situation. The university health system uses a single EMR (Epic; Epic Systems Corporation, Verona, WI) across all inpatient and ambulatory practice sites. Before ROTOX was implemented, recording of radiation therapy toxicity was inconsistent and was in free-text format. By creating this toxicity template, which is based on National Cancer Institute Common Toxicity Criteria for Adverse Events, with the auditing tools added, we hoped to improve consistency and accuracy of this documentation. Creation of custom EMR-based radiation toxicity recording template The radiation oncology toxicity assessment (ie, ROTOX) was designed as an easy-to-use checklist in the EMR, presented within the physician’s normal workflow, to be completed during each patient’s weekly physician OTV. Because each patient is unique, this assessment provided the ability to select only the site (s) currently being treated with radiation. The physicians were guided to provide only toxicity documentation that pertained to the site(s) treated with radiation (Fig 1; Appendix Figs A1 and A2, online only). These lists of toxicities were grouped by regional anatomic site (generally not more than five per site), were chosen by each physician consensus within a disease-oriented team, and focused on common and important toxicities. NUR 707 Wk 6 IT and Data Driven Decision Making Discussion
To achieve comprehensiveness, special stops (hard stops) were incorporated that made completion of ROTOX a required step by physicians before each OTV could be closed. A hard stop was a prompt that did not allow an operation to proceed without insertion of particular data in a standard and correct format in the EMR.8 This requirement ensured that toxicity data were entered in a predetermined templated fashion, as shown in Figure 1 and in Appendix Figs A1 and A2. Audit of radiation toxicity documentation quality score before and after introduction of ROTOX For each physician in our 17-physician practice, three patient charts were audited at baseline and at 6 months (for the similar kind of patient with the identical anatomic site) after introduction of the EMR toxicity tool (N = 6 charts total per physician). We selected common cancers that typically required 5 to 7 weeks of radiation courses. Each patient chart was audited for all five to seven OTVs through the course of radiation for quality of documentation in three aspects: (1) completeness score: We evaluated the number of OTVs that recorded radiation therapy toxicity (regardless of grade or subscale). Toxicity documented for all of the encounters received a score of 3; for two to four encounters, 2; for a single encounter, 1; and for no encounter, 0. (2) comprehensiveness score (multi-symptom recording): Description of three of the possible toxicity subscales in a given site in three or more encounters received a score of 3; descriptions of two, one, or zero subscales received scores of 2, 1, or 0, respectively. (3) grading score: This score described the extent to which grade of toxicity was documented. Toxicity grade clearly mentioned in three or more, two, one, or no encounters received a score of 3, 2, 1, or 0, respectively. An additional, similar, assessment of documentation quality was conducted at 36 months after initiation of ROTOX. Volume 14 / Issue 3 / March 2018 n jop.ascopubs.org Downloaded from ascopubs.org by 22.214.171.124 on December 12, 2019 from 066.207.141.241 Copyright © 2019 American Society of Clinical Oncology. All rights reserved. Copyright © 2018 by American Society of Clinical Oncology e187 Albuquerque et al Fig 1. Example of part of the radiation oncology toxicity recording tool (ROTOX) for female pelvic radiation therapy. Quarterly trends of site-specific high-grade toxicity were collected and monitored to develop benchmarks for anatomic sites. RESULTS The introduction of anatomic site–specific ROTOX in our EMR resulted in marked improvement in documentation of toxicity. A review of 51 patient charts before ROTOX yielded a mean total toxicity reporting score of 189 from a possible 459, or only 41% of the maximum possible total score (Table 1). The average comprehensiveness of toxicity recording was only 13% of maximum possible comprehensiveness score, which indicated that most physicians did not address (or document) all possible toxicity from specific radiation treatments when they relied on free text.NUR 707 Wk 6 IT and Data Driven Decision Making Discussion
The subsequent audit at 6 months after ROTOX implementation showed a dramatic 58% absolute improvement in the document quality score (DQS) to 99% of the maximum total recordable score (t test P , .003). When measured 3 years after ROTOX implementation, this high-level DQS was maintained at 96% of the maximum, and there was no significant fall-off compared with the 6-month score (t test P = .08) e188 Toxicity Monitoring and Trends In the6 monthsafterROTOXimplementation,between months 7 and 9 (ie, a 3-month period), toxicity grading was recorded in 4,443 OTVs for 698 unique patients. A total of 107 episodes of high-grade toxicity were identified during this period, and toxicity-specific intervention was documented in 95%, which indicated appropriate intervention by clinic medical professionals. Appendix Figure A3 (online only) shows the Table 1. Baseline Audit of Toxicity Documentation in Patient Charts Before Introduction of Electronic Toxicity Recording Tool, ROTOX Score Aspect Comprehensiveness Completeness Grading Overall score* Total Score for Radiation Oncologist Physicians Total Possible Maximum % of Score Maximum 20 153 13 121 153 79 48 153 31 189 459 41 Abbreviation: ROTOX, radiation oncology toxicity recording tool. *Reflects data from 51 patients (n = 3 for each of 17 physicians). Volume 14 / Issue 3 / March 2018 n Journal of Oncology Practice Copyright © 2018 by American Society of Clinical Oncology Downloaded from ascopubs.org by 126.96.36.199 on December 12, 2019 from 066.207.141.241 Copyright © 2019 American Society of Clinical Oncology. All rights reserved. EMR-Based Radiation Oncology Toxicity Recording Instrument incidence of physician-recorded toxicity in our practice recorded for 6 months after initiation of ROTOX. Grade 2 and grade 3 or greater toxicity data are shown for pelvic radiation for gynecologic cancers (Fig A3A) and chest radiation for breast cancers (Fig A3B); toxicity rates for both were quite low. NUR 707 Wk 6 IT and Data Driven Decision Making Discussion