How to Draw Blood for Vanco Trough
Abstract
Objectives
Blood samples for vancomycin levels are often drawn too early, leading to potential misinterpretation of results. However, only a few studies describe interventions to reduce mistimed vancomycin levels.
Methods
We implemented an information technology (IT)–based intervention that provided educational instructions to nurses and determined the percentage of levels drawn too early for 27 months before (n = 6,291) and 14 months after (n = 3,608) the intervention. In addition, we conducted nurse interviews (n = 40) and dataset analysis to assess the root causes of mistimed levels.
Results
The percentage of vancomycin timing errors decreased from 39% (2,438/6,291) to 32% (1,137/3,608), though in a time series analysis this decrease was not statistically significant (P = .64). Four common causes of mistimed levels were found: (1) unclear provider orders, (2) scheduling levels to be drawn with morning laboratory tests, (3) lack of communication between providers, and (4) failure to adjust the blood draw in relation to the previous dose.
Conclusions
A real-time, IT-based intervention that links the timing of levels with medication administration might have a more substantial impact.
Therapeutic drug monitoring (TDM) is routinely used for assessing drug efficacy and potential toxicity. Precise timing of specimen collection is paramount for ensuring accurate interpretation of test results by clinicians. 1–5 Numerous studies have shown that a high percentage of TDM tests may provide inaccurate information because of improper collection timing, 2,3 labeling or documentation errors, 2 or unnecessary routine daily monitoring. 1,4 Furthermore, the mistimed TDM tests may lead to inappropriate dosage adjustments 3,6 or over-ordering of tests. 1,4
In a recent 13-month study at our institution 3 we observed that 41% of specimens for measuring vancomycin levels were drawn too early, yielding significantly increased plasma vancomycin concentrations. Consequently, clinicians who received these early level results were more likely to decrease, discontinue, or hold the patient's vancomycin dose, thus potentially increasing the risk of therapeutic failure or the development of vancomycin-resistant species. We postulated that an information gap on the part of clinicians as to appropriate timing of the draw might be the primary cause of these mistimed levels and subsequent treatment changes.
Several studies have suggested ways to reduce specimen collection timing errors. 2,6,7 For example, at our institution, we successfully reduced errors associated with ordering antiepileptic drug levels 8 by using computer-based interventions. However, relatively few studies have reported interventions to improve timing of vancomycin levels. 5,9 Furthermore, to our knowledge, no studies have assessed root causes of collection timing errors for vancomycin. In this study, we implemented and evaluated the effectiveness of an educational information technology (IT)–based intervention directed at nursing staff to reduce vancomycin level timing errors. We also used our 14-month postintervention dataset to determine the root causes for vancomycin timing errors and to identify targets for further process improvement.
Materials and Methods
Study Site
Brigham and Women's Hospital (BWH) is a 779-bed academic medical center in Boston, MA, with approximately 46,000 admissions per year. The BWH clinical chemistry laboratory performs approximately 11,000 vancomycin assays annually. Our institution's therapeutic range for vancomycin is 15 to 20 mg/L. The Partners Human Research Committee approved this study.
Workflow
At BWH, a multistep process is undertaken to obtain a specimen for vancomycin testing. First, the majority of vancomycin levels are ordered through our computerized provider order entry system. Next, the patients' nurse receives the electronic order and either the nurse or unit coordinator fills out a laboratory requisition. The nurse schedules the person who is to draw the specimen (ie, nurse or phlebotomist) and at what time (eg, requisitions for the phlebotomist are placed in a folder corresponding to the round time when the specimen should be drawn). Finally, the specimen is collected by either a nurse for central line draws (approximately 50% of cases) or a phlebotomist for peripheral draws (the remainder).
Study Design and Intervention
Our study was conducted in 3 phases: (1) implementation of an IT intervention aimed to reduce vancomycin timing errors, (2) assessment of the impact of the intervention, and (3) assessment of root causes of vancomycin timing errors, using both quantitative and qualitative methods.
Our institution previously implemented an IT solution in August 2008, which displayed an electronic reminder to the provider ordering the level that trough vancomycin levels should be collected 60 minutes prior to dose. In the current study, we implemented an intervention that provided these instructions to nurses Figure 1, who are responsible for scheduling the blood draws for the majority of vancomycin levels. Nurses were required to acknowledge that they reviewed the instructions by checking a box. The intervention was implemented hospital-wide on October 3, 2011.
Figure 1
Information technology–based intervention. When reviewing providers' orders (green arrow), nurses were shown instructions on when to schedule vancomycin levels (red arrow). Nurses were also required to check a box acknowledging that they had seen the instructions in the electronic medication administration record (not shown).
Figure 1
Information technology–based intervention. When reviewing providers' orders (green arrow), nurses were shown instructions on when to schedule vancomycin levels (red arrow). Nurses were also required to check a box acknowledging that they had seen the instructions in the electronic medication administration record (not shown).
To assess the impact of the second intervention, we compared the frequency of early vancomycin levels before the intervention (April 1, 2009, to September 30, 2011, n = 6,291, excluding levels from April 13, 2010 through July 31, 2010, for which data were not available) vs after the intervention (November 1, 2011, through December 31, 2012, n = 3,608). Of note, the preintervention data in this study were expanded to include both the dataset from our previous study 3 (April 1, 2009-April 30, 2010) as well as some more recent data (August 1, 2010-September 30, 2011).
Assessment of Level Timing
To obtain the number of mistimed vancomycin levels before and after intervention, we obtained collection times from our laboratory information system (LIS) and vancomycin administration times and dosing schedules from our electronic medication administration record (eMAR). The sample collection time is manually entered into the LIS at the time of laboratory receipt, either from specimen labels printed electronically at the time of collection by the positive patient identification device used by phlebotomists or from the paper requisition accompanying the sample if not obtained by a phlebotomist. The collection times are the actual time of specimen collection (not the time the specimen was sent to the laboratory or the time the specimen was received in the laboratory).
Our inclusion criteria, which were identical to those used in our previous study, 3 were as follows: (1) an available collection time for the level in our LIS, (2) patients on a 12-hour vancomycin dosing schedule (Q12H), and (3) a vancomycin level that was drawn within 14 hours of the previous vancomycin dose. We measured collection timing relative to the previous dose and considered timing errors of ±2 hours as acceptable. Thus, for patients on a Q12H schedule, we considered all levels drawn less than 10 hours after the last dose as "too early," and those drawn at least 10 hours after the dose as "correctly timed."
Nursing Interviews
We interviewed a representative number of nurses after the intervention (October 15, 2012, through February 26, 2013), including nurses who scheduled levels that were timed correctly (n = 20) and incorrectly (n = 20). Our 10-question survey was designed to determine the effectiveness of the intervention as well to assess the level of nurses' understanding of the appropriate timing and interpretation of vancomycin levels.
Root Cause Analysis
We analyzed our postintervention dataset (November 1, 2011, through December 31, 2012 [n = 3,608]) to determine the causes of mistimed vancomycin levels as outlined herein. To investigate the possibility that many vancomycin levels were drawn along with routine morning blood draws, we compared the frequency of early collection errors during morning rounds (ie, 4 AM to 10 AM) vs the rest of the day.
To determine if late administration of vancomycin contributed to early collection of levels, we compared the actual vs scheduled dose administration time using our eMAR system for all levels in our postintervention dataset. Individual vancomycin levels were grouped in different categories based on the difference between the administered and scheduled time (eg, 30–60 minutes before, 30 minutes to 1 hour later, and more than 3 hours later than scheduled) and the frequency of mistimed levels was calculated for each category.
When ordering a vancomycin level, the provider must choose instructions for specimen timing from a drop-down menu (eg, "Next Available," "Trough/Predose") or enter specific instructions in a free-text field. To investigate whether the collection instructions provided affected the frequency of mistimed levels, we downloaded electronic orders for 7 months after the intervention (November 1, 2011, to May 31, 2012) and manually reviewed a subset of those orders (n = 723). All cases (153 [21%] of 723) without electronic orders were eliminated from the analysis. We divided the remaining orders (n = 570) into 3 categories (order states a specific time and date [I], order states "prior to dose," "predose," "before dose," and/or "trough" [II], or order is not specific [III]) and determined the number of mistimed levels in each category. The frequency of mistimed levels in category III was compared with each of the other categories.
To investigate whether having the level drawn by a phlebotomist or nurse affected the rate of timing errors we compared the frequency of early levels in each group using the collection information in our LIS.
Statistics
All categorical variables were analyzed using the χ2 test with 2-tailed P values (Prism 2012, GraphPad Software, La Jolla, CA). The Fisher exact test was used to analyze the provider orders (Prism 2012). The frequency of TDM errors for preintervention vs postintervention data was compared using PROC GENMOD time series analysis (SAS Institute, Cary, NC). P values < .05 were considered significant.
Results
Effect of Intervention
The percentage of vancomycin level timing errors decreased from 39% (2,438/6,291) in the preintervention period to 32% (1,137/3,608) in the postintervention period. However, because of a modest decreasing trend throughout the entire study period Figure 2, the observed effect was not statistically significant using time series analysis (P = .64).
Figure 2
Monthly frequency of vancomycin timing errors before and after intervention. The percentage of early vancomycin levels by month is shown for preintervention (n = 6,291) and postintervention (n = 3,608) periods (solid line). The bars represent the number of vancomycin levels included each month. No data were available from May 1, 2010, to July 31, 2010.
Figure 2
Monthly frequency of vancomycin timing errors before and after intervention. The percentage of early vancomycin levels by month is shown for preintervention (n = 6,291) and postintervention (n = 3,608) periods (solid line). The bars represent the number of vancomycin levels included each month. No data were available from May 1, 2010, to July 31, 2010.
Nursing Interviews
The nursing interviews (n = 40) showed that only 33% (13/40) of nurses used the instructions we provided despite the requirement to acknowledge the receipt of instructions in eMAR. Some nursing staff members did not find the instructions helpful and some did not notice them at all. These results were similar for the mistimed and correctly timed levels.
Of the 20 nurses who scheduled vancomycin levels to be drawn too early, 50% (10/20) described the timing as "close enough—a usable level," and 10% (2/20) described the timing as "exactly right." Of the 20 nurses who correctly timed the levels, 50% (10/20) described the timing as "close enough—a usable level," and 40% (8/20) described the timing as "exactly right."
We also observed that 75% (15/20) of nurses scheduling early levels and 35% (7/20) of nurses scheduling correctly timed levels aimed to have the results back before administering the next dose. To achieve this, some nurses would time the levels early or delay the next vancomycin dose until the levels were reported.
Root Cause Analysis
A significantly higher number of mistimed levels were drawn between 4 AM and 10 AM (48% [578/1214]), compared with the rest of the day (23% [559/2,394]; P < .0001). For 32% (1,164/3,608) of vancomycin levels, the previous dose was given at least 30 minutes later than scheduled. Vancomycin levels were more frequently mistimed in cases when the dose was given late (44% [513/1,164]) compared with when doses were administered at (or before) the scheduled time (26% [624/2,444]; P < .0001). Further, the later the dose was given, the more likely the vancomycin level was to be mistimed.
Providers' orders that did not specify a collection time (type III) or stated "predose/trough" (type II) had the highest error rate, with 36% and 32%, respectively, of vancomycin levels being drawn too early Table 1. The lowest error rate, 19%, was observed with orders that specified the exact date and time (type I) (Table 1). The error rate with type I orders was significantly lower than the error rate with type III orders (Table 1; P = .01).
Table 1
Effect of Provider Order on Vancomycin Timing Error
Table 1
Effect of Provider Order on Vancomycin Timing Error
In the postintervention dataset, nurses drew levels too early 26% (375/1,445) of the time and phlebotomists drew levels too early 35% (762/2,163) of the time (P < .0001).
Discussion
Ensuring accurate timing of therapeutic drug levels has been a challenge for many institutions, including our own. In our previous study, 3 we observed that a significant proportion of vancomycin levels were drawn too early, leading to potential misinterpretation of drug levels. Our main goal in the present study was to reduce this problem by implementing a simple IT intervention to improve the timing of vancomycin levels, and a secondary goal was to better understand the reasons for mistimed levels. Although the timing errors decreased from 39% before the intervention to 32% after the intervention, the results did not achieve statistical significance. However, we were able to establish several factors contributing to the lack of a major impact of our intervention Table 2.
Table 2
Root Causes and Solutions for Mistimed Vancomycin Levels
Table 2
Root Causes and Solutions for Mistimed Vancomycin Levels
Most of the nurses we interviewed did not use the instructions provided in our intervention. This suggested that an educational reminder alone was insufficient to ensure proper scheduling of vancomycin levels. Although one study showed that educational interventions can be successful in improving vancomycin timing, 9 they are labor-intensive and the long-term effectiveness is questionable. 6,7
Our nursing interviews also revealed 2 common gaps in knowledge regarding the timing of vancomycin levels (Table 2). In many cases, nurses stated that levels drawn more than 2 hours before the dose were appropriately timed, thus illustrating a lack of understanding of the definition of a true trough level. Furthermore, several nurses believed that results of vancomycin levels were needed before administering the next vancomycin dose. As a result, they would draw the vancomycin earlier than scheduled to have the level back or delay administration of the next dose until the result was obtained.
Another common issue was that vancomycin levels were frequently scheduled to be drawn as part of the morning phlebotomy rounds, probably because of (1) unclear providers' orders, (2) a lack of understanding on the nurses' part regarding appropriate scheduling of blood draws, or (3) patient convenience (Table 2). In fact, our results confirmed that the first of these factors, lack of specific providers' orders did contribute to levels being drawn early.
Doses were frequently given later than scheduled at our institution (Table 2). In our current workflow, eMAR prompts the nurse to administer vancomycin at the originally scheduled time and does not adjust the timing if the previous dose was given late, which may lead to mistiming errors. In addition, our intervention specified that the trough specimen should be drawn less than 60 minutes before the dose, but did not address appropriate timing if the previous level was given late. Future interventions to improve specimen timing in TDM should safeguard against scenarios in which levels are drawn too early because of late administration of the previous dose. An IT solution that integrates information from the eMAR and specimen collection system and adjusts level draw in real time would be especially promising.
Phlebotomy draws were associated with a higher rate of early vancomycin levels, though this contributing factor was not as significant as the other causes identified herein. This revealed the importance of communication of appropriate level timing with care team members not ordering or scheduling vancomycin levels. An electronic system that notifies the phlebotomy team of a timed draw several hours earlier would be ideal to optimize workflow and promote efficient staffing.
The limitations of this study are similar to those identified in our previous study. 3 We did not establish the proportion of vancomycin levels ordered after a steady state had been reached. Furthermore, we only evaluated patients on Q12H dosing schedules. We did not evaluate factors such as age, sex, renal function, or previous vancomycin levels, because our previous study found no association of these factors with level timing errors. 3
Our experience in the present study suggests that merely informative, educational IT interventions aimed at reducing vancomycin timing errors are ineffective because of low compliance of clinical staff. Thus, we believe that an integrated, IT-driven solution providing a single channel for ordering and collecting timed levels linked to dosing times is necessary to produce a lasting effect (Table 2). Such a solution requires integration of various IT subsystems, including computerized provider order entry, medication administration, specimen collection, and the LIS. A quality improvement initiative is currently under way to devise a solution that we hope will result in more substantial and durable improvement.
CME/SAM
Upon completion of this activity you will be able to:
-
describe how vancomycin collection timing errors can affect clinical decision making.
-
list the common causes of vancomycin collection timing errors.
-
discuss interventions (eg, educational, electronic) that may improve timing of vancomycin levels and their benefits and limitations.
The ASCP is accredited by the Accreditation Council for Continuing Medical Education to provide continuing medical education for physicians. The ASCP designates this journal-based CME activity for a maximum of 1 AMA PRA Category 1 Credit TM per article. Physicians should claim only the credit commensurate with the extent of their participation in the activity. This activity qualifies as an American Board of Pathology Maintenance of Certification Part II Self-Assessment Module.
The authors of this article and the planning committee members and staff have no relevant financial relationships with commercial interests to disclose.
Questions appear on p 913. Exam is located at www.ascp.org/ajcpcme.
References
1.
et al. .
Evaluating the appropriateness of digoxin level monitoring
.
Arch Intern Med
.
1999
;
159
:
363
–
368
.
2.
.
Appropriateness of sampling times for therapeutic drug monitoring
.
Am J Hosp Pharm
.
1984
;
41
:
1796
–
1801
.
3.
et al. .
What proportion of vancomycin trough levels are drawn too early? frequency and impact on clinical actions
.
Am J Clin Pathol
.
2012
;
137
:
472
–
478
.
4.
et al. .
Appropriateness of antiepileptic drug level monitoring
.
JAMA
.
1995
;
274
:
1622
–
1626
.
5.
et al. .
Effects of therapeutic drug monitoring criteria in a computerized prescriber-order-entry system on the appropriateness of vancomycin level orders
.
Am J Health Syst Pharm
.
2011
;
68
:
347
–
352
.
6.
.
Improving the use of therapeutic drug monitoring
.
Ther Drug Monit
.
1998
;
20
:
550
–
555
.
7.
et al. .
Strategies for physician education in therapeutic drug monitoring
.
Clin Chem
.
1998
;
44
:
401
–
407
.
8.
et al. .
A computer-based intervention for improving the appropriateness of antiepileptic drug level monitoring
.
Am J Clin Pathol
.
2003
;
119
:
432
–
438
.
9.
et al. .
Short term impact of guidelines on vancomycin dosing and therapeutic drug monitoring
.
Int J Clin Pharm
.
2012
;
34
:
282
–
285
.
© American Society for Clinical Pathology
How to Draw Blood for Vanco Trough
Source: https://academic.oup.com/ajcp/article/140/6/801/1760783
0 Response to "How to Draw Blood for Vanco Trough"
Post a Comment