Running head: A PROCESS IMPROVEMENT 1 A Process Improvement for the Management of Hypoglycemia and Hyperglycemia Valerie D. Jones University of Florida
A PROCESS IMPROVEMENT 2 Abstract Background: A Diabetic Management Work Group (DMWG) was established at the North Florida/South Georgia Veterans Affairs (VA) Medical Center in February 2009 to assess documentation and administration of sliding scale insulin (SSI) and hypoglycemic agents. Purpose: T he purpose was to see the effects of the process change on documentation of SSI administration, as well as the effects of the process change on usage of oral glucose gel and intravenous (IV) glucose. Participants: This is a non human subject research study Methods: administration of diabetic medications. The resulting process change involved modifying the available was then implemented. Data collected in June, July, and August 2009 (before the process change) and May, June, and July 2010 (after the process chang e) were compared. Results: In 2009, 194 of 833 glucose measurements had SSI indicated but not administered. A range of 9 39% error was seen among the seven units that were reviewed. In 2010, 102 of 951 glucose measurements had SSI indicated but not admin istered. Thus, the percentage range decreased to between 3 18% on each unit. Conclusion: The data shows that in May, June, and July 2010 there was an improvement in the administration and documentation of SSI. A decrease in the use of IV Dextrose and an increase in oral glucose were seen as well.
A PROCESS IMPROVEMENT 3 A Process Improvement for the Management of Hypoglycemia and Hyperglycemia Introduction Glycemic control plays an important role in decreasing the long term complications of diabetes including cardiovascular disease, organ damage, and neuropathies (Wright, 2009). Avoiding errors in the administration of insulin is a key point in managing dia betes since proper treatment of hyperglycemia in patients will reduce morbidity as well as mortality (Shogban & Levy, 2010). One way to prevent these errors would be to document the blood glucose level. Additionally, the amount of insulin administered or not administered should be documented. According to Helmons, Wargel, and Daniels (2009), Bar code assisted medication administration (BCMA) was developed as an additional safety barrier between the nurse and the patient (p. 1202). With the BCMA system (Helmons, Wargel, & Daniels, 2009). Background In February 2009, a mock Joint Commission on the Accreditation of Healthcare Organizations (JCAHO) survey took place at the North Florida/South Georgia Malcom Randall Veterans Affairs (VA) Medical Center. During this survey, a VA pharmacist assisted the JCAHO representative in reviewing the records of ten patients. The VA pharmacist and JCAHO representative found a high error rate with at least one error of omission in regards to the administration of sliding scale insulin (SSI). Due to this finding, the multidisciplinary Diabetic Management Workgroup (DMWG) was cre ated and included physicians, pharmacists, nurses, and nurse educators. The DMWG was formed to oversee the improvement in the administration and documentation of sliding scale insulin as well as hypoglycemia agents. The process change
A PROCESS IMPROVEMENT 4 was implemented soo n after in February 2010, and data from May, June, and July 2010 was compared with the months of June, July, and August 2009. Purpose After the process change was initiated, it was important to assess the effects of the process change on the administratio n and documentation of SSI. Furthermore, the effects of the process change on the administration of oral glucose and intravenous (IV) Dextrose were assessed as well. The aim of this quantitative retrospective study was to identify a decrease in the numbe r of medication errors while also seeing an improvement in the documentation of SSI. Evaluating the outcomes of a process change are important in determining whether or not these led to an improvement in patient care. Since the mock JCAHO revealed that t here was a safety issue with the administration of insulin, it would be necessary to ensure that the resulting hospital wide process change did improve patient safety. Hypothesis After developing and implementing the process change at the VA hospitals, it was important to set goals to be attained in order to assess whether or not the process change did in fact improve patient safety regarding the administration of insulin and hypoglycemia agents. There were two hypotheses that were formulated after the pr ocess change took place. The first hypothesis was that the errors of omission of SSI orders would decrease after the process change. The second hypothesis was that the process change would lead to an increase in oral and decrease in intravenous modes of administering glucose. Sample This study is a non human subject research study. According to the University of Florida Institutional Review Board (IRB) questionnaire, to be considered a non human subject research
A PROCESS IMPROVEMENT 5 study it must meet the following stipulat ions: there must be no interventions or interactions with a living person; no identifiable private data or information obtained in a form associable with the individual; include no individuals who become participants as healthy individuals, control subject s, patients, or recipients of a test article; and it does not involve human subjects on whose specimens an investigational device is used (Institutional Review Board [IRB], 2010). Methods After the process change, the physicians were able to write orders to have nurses administer SSI before meals and at bedtime since SSI was now reflected as a scheduled medication. With this change, the nurses and providers had to be educated on the new process. In addition to changing SSI to a scheduled medication, the physicians were able to order glucagon and glucose gel as options for treating hypoglycemia. These hypoglycemia agents were then added to the Pyxis machine, which holds all of the medications on the VA hospital units. After all this took place, previousl y collected data was reviewed, and this data included: numerical data of the frequency of administration of oral glucose and IV Dextrose, percentage of sliding scale insulin coverage indicated and administered, and percentage of sliding scale insulin cover age indicated and not administered. The data reviewed was from the months June, July, and August of 2009 and the months May, June, and July of 2010. Analysis The information that was used for the study was de identified, and was kept locked at the Malcom Randall VA Hospital in the office of the primary investigator for the study. The data evaluated was for the months of June, July, and August of 2009 and the months of May, June, and July of 2010; the administration and documentation of sliding scale insu lin, as well as hypoglycemia agents before and after the process change, were compared. The analysis was of
A PROCESS IMPROVEMENT 6 the following information: total number of charts that were reviewed, total number of chemsticks reviewed, total number of chemsticks within the no rmal range, total number of insulin coverage indicated and administered, and total number of insulin coverage indicated and not administered. This data was reviewed on seven different units of the North Florida/South Georgia Veterans Affairs Medical Cente r. Findings In 2009, there were 194 of 833 glucose measurements where SSI was indicated but not administered. There was a range of 9 39% error among the seven different units involved in the study. In 2010, the data revealed that there were 102 of 951 g lucose measurements where SSI was indicated but not administered. This shows the decrease percentage range of error to be from 3 18% on each unit. There was also a decrease in the frequency of IV Dextrose being administered. In June, July, and August of 2009, IV Dextrose was administered 169 times across the seven units, while in May, June, and July of 2010 it was only given 61 times. An increase in the frequency of administration of oral glucose gel from 2 in 2009 to 74 in 2010 was also reflected in th e reviewed data. Conclusions Based on the findings of the study, it is apparent that the process change initiated in February 2010 prompted the improvement in the administration and documentation of sliding scale insulin. Additionally, the use of IV Dext rose decreased while oral glucose gel was administered more frequently to treat hypoglycemia. Although there was an improvement after the process change, there is still some error in the administration and documentation of SSI. This finding can infer tha t more nursing education may be needed. This initial study has encouraged the researchers to continue to look for ways to improve patient safety. The next step is to
A PROCESS IMPROVEMENT 7 compare the variance between the time the blood glucose measurements were taken and the time SSI was administered. The data previously collected will again be used to compare the variances before and after the process change. Functions of the Student The student initially discussed with her faculty member, Dr. Sandra Citty, PhD about her di fferent research options. Once they decided on a research study, the student started to review the literature to read about the protocols and management of diabetes and the administration of n the safety issues regarding the improper administration of SSI, as well as, using oral glucose gel versus IV Dextrose to treat hypoglycemia. Before doing the literature review, the student first had to understand why the researchers were hoping for cert ain outcomes from the study. For example, the student learned that IV dextrose 50% is very irritating to the veins (Kee, Hayes, & McCuistion, 2009). Furthermore, the student had to remember that when glucose levels are not well controlled in diabetic pat ients, various microvascular and macrovascular complications could occur (Shogban documentation system. At the VA, the medications are located in a machine called t he Pyxis machine, and the nurses document the medication administration in the Computerized Patient Record System (CPRS). The scheduled medications and the PRN medications are listed separately in the BCMA; this is what the DMWG found to be one of the mai n causes for the insulin administration and documentation errors. The student attended weekly meetings with Kendall Williamson, her partner in the study, Dr. Citty and Lisa Zumberg, PharmD. Lisa Zumb erg is the primary investigator of the study,
A PROCESS IMPROVEMENT 8 also directed at discussing the aspects of the first phase of the study, and how the next phase was going to be focusing on comparing pre and post process change data. The student was able to learn about the IRB process and also the process of getting the study approved by the VA IRB. The DMWG met monthly, and the student attended these meetings as well. During the meetings, the group reviewed the data that was collected in 2009 and 2010 and discussed the different outcomes that were seen. In addition, the group met to converse about any further issues they were having on the different units concerning diabetic patients. The meetings usually consisted of several nurses, pharmacists, informatics, physicians, and nursing education. Before the student could participate in the study she had to complete several hours of computer training and one in person trainin g. The primary investigator had to sign some documents and submit all of the training certificates so that the student would be able to participate in the research study. Once all of the paperwork was completed, the student and her partner wrote an abstr act to attend the Southern Nursing Research Society (SNRS) conference. The abstract was accepted in the Top Student Poster Session, and the student and her partner presented their research study at the conference. During the discussion session the studen t and her partner were asked to sit in front of an audience and answer questions regarding the study and any significant outcomes. One of the problems that arose during the course of the study was that some of the mandatory training sessions for researche rs at the VA hospital were held when the student had class. The student decided that missing class to attend the mandatory training session was a necessary compromise. This study also needed approval from UF and VA IRB. Since the VA IRB only meets once a month, this created some issues with the student and her partner being able to create a poster for the SNRS conference. The study did not receive VA approval until
A PROCESS IMPROVEMENT 9 February 7, 2011, and the poster was to be submitted that same day. The student and her partner only had a week to create a poster so that it could be printed in time for the conference. Being that the abstract was accepted into the Top Student Poster Session, an electronic version of the poster was due by January 19, 2011. Since the VA IRB did not yet approve the study, the student and her partner were unable to complete the electronic poster in time. The last issue that came about was when the VA Medical Center required all researchers to have computer access training. Without this, the research study was going to be shut down. This was brought to the along with the primary investigator, had to contact the person who was in charge of the comput er training in order to set up a meeting. The briefing had to take place with very short notice, and fingerprints were needed as well for computer access. This minor set back could have potentially led to the student being unable to continue with the stu dy. In order to deal with the issues that the students and researchers encountered, deadlines were established. The student was also able to work with the VA medical media staff in order to get the poster printed before leaving for the conference. The st udent learned that time management and patience were key aspects in doing any type of research. Being flexible and handling some unforeseen problems was to look up the deadline for the IRB approval process to know when certain documents have to be submitted. When doing a study that requires approval from both UF and the VA, it is necessary to be prepared to complete all the necessary paperwork in a timely manner.
A PROCESS IMPROVEMENT 10 Recommendations for Future Students For the student who decides to continue with this research study, the recommendations would be to start early and plan ahead. There are many issues that may happen later on in the study, and the next student should be prepared to deal with those problems. Since research is highly unpredictable, the student must be aware of this from the beginning. When doing research, the team members must acknowledge that the results may be different from what was hoped for, and/ or the process may take longer than expected. Also, the student should be aware of the various training sessions that are mandatory in order to participate in research at the VA hospital. As soon as the student receives confirmation of acceptance into the H she should meet with Dr. Citty to discuss all the computer training that needs to be completed, as well as to identify the dates for the training sessions at the VA hospital. If the training is completed early enough, the student can start reviewing the data from the previous study and begin analyzing the data for the next step in the study. In order to be successful with this study, the junior student should first do a literature review to gain knowledge and understanding about why this r esearch topic is important, and what interventions can improve patient safety. Another recommendation is to only get involved in this study if the student is interested in insulin, diabetes, or simply wants to be involved in learning about hospital wide p olicy/process changes. If the next student involved in the study has a busy schedule, then he or she needs to meet with Dr. Citty and Lisa Zumberg to formulate a schedule that will work for everyone. Having a positive attitude and willingness to compromi se are two essential characteristics that the junior student should strive towards.
A PROCESS IMPROVEMENT 11 The mock JACHO survey performed at the VA Medical Center in February 2009 revealed a high rate of error regarding the administration and documentation of sliding scale i nsulin. With the development of the DMWG, a process change was initiated in February 2010, medication as oppose to a PRN medication. Oral glucose gel also bec ame an option in the Pyxis machine for treating hypoglycemia. A quantitative retrospective study was implemented to determine the effects of the process change on documentation of SSI as well as the effects of the change on usage of oral glucose gel and I V Dextrose. Data was reviewed in the months June, July, and August of 2009 (pre process change) and compared to that of the months May, June, and July of 2010 (post process change) to see whether or not the process change had improved the administration a nd documentation of SSI.
A PROCESS IMPROVEMENT 12 References America Diabetes Association. (2009). Standards of medical care in diabetes 2009. Diabetes Care, 1 (32), S13 S61. doi:10.2337/dc09 S013 Helmons, P. J., Wargel, L. N., & Daniels, C. E. (2009). Effect of bar code assisted medication administration on medication administration errors and accuracy in multiple patients care areas. American Journal of Health System Pharmacy, 66 (13), 1202 1210. doi: 10.2146/ajhp080357 Institutional Review Board. (2010). Introductory Questionnaire: Non human, exempt, training grant, & center projects. Unpublished manuscript, Department of Research, University of Florida, Gainesville, Florida. Jerreat, L. (2009) Treatment of hyperglycaemia in patients with type 2 diabetes. Nursing Standard, 24 (1), 50 57. Kee, J. L., Hayes, E. R., & McCuistion, L. E. (2009). Pharmacology: A nursing process approach (6th ed.). St. Louis, MO: Saunders. Matheny, M. E., Shubina, M., Kimmel, Z. M., Pendergrass. M. L., & Turchin, A. (2008). Treatment intensification and blood glucose control among hospitalized diabetic patients. Journal of General Internal Medicine, 23 (2), 184 189. doi:10.1007/s11606 007 0468 x Shogbon, A. O., & Levy, S. B. (2010). Intensive glucose control in the management of diabetes mellitus and inpatient hyperglycemia. American Journal of Health System Pharmacy, 67 (10), 798 805. doi:10.2146/ajhp090211 Wright, E. E., Jr., (2009). Overview of insulin replacement therapy. Journal of Family Practice, 58 (8), S3 S9.