Pa Patient Saf Advis 2009 Sep;6(3):74-8. 
Medication Monitoring Errors: Inappropriate Levofloxacin Doses
Emergency Medicine; Gerontology; Internal Medicine and Subspecialties; Nursing
Expand All Collapse All

Abstract

Measuring and monitoring a patient’s clinical laboratory values is critical for diagnosing problems, gauging effectiveness of treatment, and selecting appropriate doses of many drugs. Breakdowns can occur when necessary monitoring of laboratory values is not ordered or performed or when abnormal levels occur and action is not taken by healthcare practitioners. More than 2,500 event reports submitted to the Pennsylvania Patient Safety Authority were specifically classified as errors involving drug-related clinical monitoring. Analysis reveals that 48% of the events reached the patient. The medication most frequently involved in these reports was levofloxacin (416 events), a drug that requires dosage adjustments based on renal function. Nearly 99% (345) of the 349 reports that included information regarding the node in which the error occurred identified the prescribing node. Almost 92% (382) of the event reports indicated that patients required dose adjustment due to renal function. Of the 416 events involving levofloxacin, 95.4% (397) occurred with patients who were 65 years of age or older, a population with naturally occurring age-related decreases in renal function. The majority (411) of these events was intercepted by practitioners before reaching the patient. Strategies to address these problems include obtaining baseline renal function information, ensuring that current laboratory and testing information is available to all practitioners, interfacing laboratory computer systems with order entry systems, and building computer-based clinical decision support to help guide medication dosing for patients with renal insufficiency.

Introduction

Patient information helps practitioners select appropriate therapy. One set of critical patient-specific information is the patient’s clinical laboratory values. Laboratory values provide information for a range of activities, including diagnosing problems and gauging effectiveness of treatment. Another area in which laboratory values play an important role is in medication and dose selection, especially for drugs that are dosed based on renal function, liver function, or drug levels. However, breakdowns can occur when necessary monitoring of laboratory values are not ordered and performed or when abnormal levels occur and are not acted upon by healthcare practitioners.

Pennsylvania facilities have submitted reports to the Pennsylvania Patient Safety Authority describing situations such as these, often involving antibiotics or anticoagulants. Clinical analysts reviewed 2,564 event reports submitted to the Authority from June 2004 through November 2008 that categorized the event type as “Medication Error, Monitoring Error, Clinical (lab value, vital sign).” Further breakdown of these events by harm score, which is adapted from the National Coordinating Council for Medication Error Reporting and Prevention harm index,1 shows that 48% (1,227) of the events reached the patient (harm index = C to I), 0.7% (18) of the events resulted in harm significant enough to require additional treatment, and one event contributed to a patient death.

It was anticipated that medications that require intense, regular monitoring to ensure their safe and effective use, such as heparin, enoxaparin, warfarin, vancomycin, gentamicin, and insulin, would be predominant in the data. Some of these medications appear in the top five medications cited in the reports (see Table 1); however, levofloxacin (Levaquin®) is the medication most often cited. Therefore, the analysis focused on levofloxacin to determine why and where in the process these errors are occurring.

Table 1. Top 5 Medications Involved in Therapeutic Monitoring Error Reports (n = 1,294) ​
Medication Total
Levofloxacin416
Heparin387
Vancomycin180
Insulin170
Warfarin141

 

Levofloxacin Use and Dosing

Levofloxacin is classified as a fluoroquinolone antibiotic.2 It has a broad spectrum of activity against gram positive, gram negative, and atypical organisms. A variety of infections, including lung, sinus, skin, and urinary tract infections, can be treated with levofloxacin, including hospital- and community-acquired pneumonia, acute bacterial sinusitis, complicated and uncomplicated skin infections, and acute pyelonephritis. Levofloxacin is available in oral solid, oral solution, and intravenous injectable formulations.

The dosing of levofloxacin is first based on indication. While the duration of treatment may vary, levofloxacin is typically administered in daily doses. For example, oral levofloxacin is administered every 24 hours, and levofloxacin injection is infused over 60 to 90 minutes every 24 hours (see Table 2).

Table 2. Dosage in Adults with Normal Renal Function (Creatinine Clearance Greater Than or Equal to 50 mL/min) ​ ​
Type Of Infection Daily Dose (In Mg) Days Of Treatment
Hospital-acquired pneumonia 7507 to 14
Community-acquired pneumonia500
750
7 to 14
5
Acute bacterial sinusitis 750
500
5
10 to 14
Acute bacterial exacerbation of chronic bronchitis 5007
Complicated skin and skin structure infections7507 to 14
Uncomplicated skin and skin structure infections5007 to 10
Chronic bacterial prostatitis 50028
Complicated urinary tract infection or acute pyelonephritis750
250
5
10
Uncomplicated urinary tract infection 2503
Inhalational anthrax (postexposure)50060
Source: Ortho-McNeil-Janssen Pharmaceuticals, Inc. Levaquin® (levofloxacin) [full prescribing information] [online]. 2008 Sep [cited 2009 Apr 15]. Available from Internet: http://www.levaquin.com/levaquin/shared/pi/levaquin.pdf. ​ ​


The second critical piece of information needed to safely prescribe levofloxacin is the patient’s renal function. Appropriate laboratory studies and clinical observation is indicated before and during therapy.

Table 3. Dosage Adjustment in Adult Patients with Renal Impairment ​ ​ ​
Creatinine Clearance(Crcl) Greater than or equal to 50 Ml/Min Crcl 20 to 49 Ml/Min Crcl 10 to 19 Ml/Min Hemodialysis or Chronic Ambulatory Peritoneal Dialysis
750 mg daily750 mg every 48 hours 750 mg initial dose, then 500 mg every 48 hours 750 mg initial dose, then 500 mg every 48 hours
500 mg daily 500 mg initial dose, then 250 mg daily 500 mg initial dose, then 250 mg every 48 hours 500 mg initial dose, then 250 mg every 48 hours
250 mg daily No dosage adjustment required 250 mg every 48 hours (No dosage adjustment needed for uncomplicated urinary tract infection)No dosing adjustment information available
Source: Ortho-McNeil-Janssen Pharmaceuticals, Inc. Levaquin® (levofloxacin) [full prescribing information] [online]. 2008 Sep [cited 2009 Apr 15]. Available from Internet: http://www.levaquin.com/levaquin/shared/pi/levaquin.pdf. ​ ​ ​

 

Analysis of Levofloxacin Event Reports

The analysis of the levofloxacin-related event reports (416) submitted to the Authority included a review of the nodes—prescribing, transcription/order processing, preparation/dispensing, administration, and monitoring—of the medication-use process in which the event occurred, each report’s description, and the age of the patient. Failure to use critical patient information during the various nodes, or stages, of the medication-use process, particularly during the prescribing phase, can lead to inappropriate dose selection.

Clinical analysts specifically reviewed the nodes in the medication-use process in which the error occurred, as reported by the facilities (see Table 4). Overwhelmingly, facilities indicated that errors involving levofloxacin originated in the prescribing phase. Nearly 99% (345) of the reports that included information regarding the node(s) identified the prescribing node. Even though these reports were classified as errors involving drug-related clinical monitoring, none of the reports indicated an error during the monitoring phase.

Table 4. Nodes of the Medication-Use Process in which Events Occurred (n = 349)* ​
Node Total
Prescribing345 (98.9%)
Transcription1    (0.3%)
Preparation1    (0.3%)
Administration5    (1.4%)
Monitoring0
* Facilities may select more than one node for a given report; therefore, percentages may be greater than 100%.

 

The reports indicate that pharmacists are intercepting a large number of the errors associated with levofloxacin. This may be a result of independent pharmacist review of a patient’s CrCl or the pharmacy computer system alerting pharmacists to abnormal CrCl values. Analysts noted that pharmacists intercepted at least 48% (199) of the 416 levofloxacin-related events. Of the reports that indicated the error occurred in the prescribing node, pharmacists caught roughly 57% (198). Analysts identified these reports according to wording included in the event details, such as “PharmInt” (i.e., pharmacy intervention), “pharmacist recommend to change the dose,” and “Doctor changed order after speaking with pharmacist.”

It is likely that pharmacists intercepted more of the levofloxacin-related events. For example, analysts found an additional 95 reports attributed to the pharmacy that indicated that the dose had been “clarified” or the prescriber was “called,” “contacted,” or “notified,” resulting in a change in dose. If these were actual pharmacist interventions, then the number of events that pharmacists intercepted would be 294 or roughly 70% of the levofloxacin-related events. Of the remaining 122 reports, 117 indicated the error had been caught before reaching the patient, but it was not possible to determine whether the error had been detected by a prescriber, pharmacist, nurse, or patient. This data shows that effective pharmacy and nursing intervention programs can intercept medication errors and generate data that can be analyzed to identify failure modes in the medication-use system.

The Authority does not require facilities to include specific laboratory information for this event type when submitting reports through its reporting system. However, for 62.3% (259) of the 416 levofloxacin-related reports, facilities included specific patient laboratory values in the event description field; all of these patients had below normal renal function (e.g., CrCl below 50 mg/mL) (see Table 5). This laboratory information appears to not have been referenced during the prescribing phase despite the need to adjust the levofloxacin dose when a patient’s CrCl falls below 50 mg/mL to avoid drug accumulation and decrease risk of toxic reactions. Another 29.6% (123) of the event reports documented that the levofloxacin dose was reduced due to renal function but did not provide actual laboratory values.

Table 5. Below Normal Creatinine Clearance Values Documented in Event Reports (n = 259) ​
Creatinine Clearance (Ml/Min)Total
20 to 49215   (83%)
10 to 1932   (12.4%)
Less than 912     (4.6%)


Potential contributing factors to the prescribing and dispensing of inappropriate levofloxacin doses may include the following:

  • Prescribers may not be aware of the need to assess renal function and adjust doses as necessary.
  • Clinical laboratory values may not be readily available to practitioners at the time of prescribing, dispensing, and administering. Even in facilities with electronic systems, laboratory computer systems may not interface with computerized prescriber order-entry (CPOE) and pharmacy computer systems.4 If the systems do interface, CPOE and pharmacy computer systems may not generate alerts to warn practitioners about high doses with respect to serum creatinine levels.5
  • CPOE systems may not have effective computer-based clinical decision support to help guide appropriate dose selection for patients with renal insufficiency or advanced age.4,6
  • The clinical status of the patient may require administration of a first dose of an antibiotic before an accurate assessment of renal function is available.

Another piece of patient information that should at least trigger investigation of the need for dose adjustments is the patient’s age. Decreased muscle mass and a corresponding lower production of creatinine in the elderly patient may give the impression that the patient’s serum creatinine level, and therefore renal function, is “normal.”7 However, it is well known that renal function decreases with age.8 Cross-sectional studies have shown that renal function, specifically glomerular filtration rate as measured by CrCl, decreases as age increases.9,10 Therefore, age-related changes in pharmacokinetics may be expected.

The event reports suggest that many prescribers are not using age as a signal to consider or investigate potential age-related decreases in renal function before prescribing levofloxacin. The age of the patient was disproportionally skewed toward the elderly population. Of the 416 events involving levofloxacin, 95.4% (397) occurred with patients who were 65 years of age or older. In fact, a vast majority (91.9% [365]) was 75 years of age or older (see Table 6).

Table 6. Patients 65 Years of Age or Older (n = 397) ​
Age (Years) Total
65 to 696  (1.5%)
70 to 7426  (6.6%)
75 to 7964  (16.1%)
80 to 8495  (23.9%)
85 or older206  (51.9%)

 

Risk Reduction Strategies

Healthcare facilities should take steps to safeguard the prescribing, dispensing, and administering of levofloxacin, as well as other medications that require routine monitoring of laboratory values to maintain appropriate dosing. Examination of the prescribing process for levofloxacin, in particular, should be reviewed as inappropriate doses are being prescribed to patients, especially elderly patients. Based on the review of reports submitted to the Authority, as well as observations at the Institute for Safe Medication Practices and in the literature, strategies include the following:

  • Determine the microbiological indication for levofloxacin.
  • Obtain baseline patient information such as age and renal function information (e.g., serum creatinine, CrCl). Continue to assess renal function throughout therapy and over time to obtain a true picture of the patient’s renal function. Adjust the dose as necessary for patients with renal impairment.2
  • In the event that CrCl data is not available, consider dosage adjustments based on age. It can be expected that elderly individuals may have lower CrCl values due to age-related reduction in renal function.9,10
  • Use appropriate assessment techniques (and/or patient information such as age or laboratory values) before drug administration.
  • Ensure that current laboratory and testing information is available to all practitioners. Work with physicians’ offices to develop a process for communicating this essential patient information timely and efficiently.
  • Work with vendors to interface the laboratory system with CPOE and pharmacy systems. During order entry, computer systems should warn practitioners when these agents are about to be used for patients with decreased renal function.11 Computer-based clinical decision support to help guide medication dosing for patients with renal insufficiency can result in improved dose and frequency choices.12
  • Expand organizational medication-use policies to include the provision that medications are not prescribed, dispensed, or administered unless relevant clinical information such as critical laboratory values for patients are available and considered by practitioners.13
  • Provide essential patient information, including patient age, height (cm), weight (kg),14,15 allergies with descriptions of the reactions, previous adverse drug reactions with manifestations, diagnosis, and comorbid conditions (e.g., renal impairment) at the top of all preprinted order sets, medication administration records, and pharmacy patient profiles, as well as screens within future CPOE systems.
  • Create competency statements for the medical and pharmacy staff to use laboratory parameters when necessary to check on the proper dosing of drug therapy. Consider specific competency exercises on appropriate patient and dose selection for levofloxacin.
  • Develop efficient methods for collecting and documenting information on prescribing errors through the tracking of pharmacy and nursing interventions. Once collected, refer to the following strategies:
    • Present quarterly aggregate information at the pharmacy and therapeutics committee meetings and to the medical board.
    • Gain insight into why such prescribing errors are occurring, and generate ideas for improvement.
    • Work with the medical staff to take action (e.g., establish prescribing/dosing guidelines, develop drug protocols, design preprinted order forms) on identified problems, and evaluate their effectiveness (e.g., perform medication-use reviews, monitor use of protocols).
    • Provide frontline, leadership, and medical staff with ongoing information on errors prevented by pharmacy and nursing interventions.

Notes

  1. National Coordinating Council for Medication Error Reporting and Prevention. NCC MERP index for categorizing medication errors [online]. 2001 [cited 2009 Apr 15]. Available from Internet: http://www.nccmerp.org/medErrorCatIndex.html.
  2. Ortho-McNeil-Janssen Pharmaceuticals, Inc. Levaquin® (levofloxacin) [full prescribing information] [online]. 2008 Sep [cited 2009 Apr 15]. Available from Internet: http://www.levaquin.com/levaquin/shared/pi/levaquin.pdf.
  3. Mehlhorn AJ, Brown DA. Safety concerns with fluoroquinolones. Ann Pharmacother 2007 Nov;41(11):1859-66.
  4. Grissinger M, Cohen H, Vaida AJ. Using technology to prevent medication errors. In: Cohen MR, ed. Medication errors. 2nd Ed. Washington (DC): American Pharmacists Association; 1999.
  5. Results of the PA-PSRS Workgroup on Pharmacy Computer System Safety. PA PSRS Patient Saf Advis [online] 2007 May 31;4(Suppl. 2). [cited 2009 Jul 10]. Available from Internet: http://patientsafetyauthority.org/ADVISORIES/AdvisoryLibrary/2007/may31_4(suppl2)/Pages/home.aspx.
  6. Nebeker JR, Hoffman JM, Weir CR, et al. High rates of adverse drug events in a highly computerized hospital. Arch Intern Med 2005 May;165(10):1111-6.
  7. Perrone RD, Madias NE, Levey AS. Serum creatinine as an index of renal function: new insights into old concepts. Clin Chem 1992 Nov;38:1933-53.
  8. Yuen GJ. Altered pharmacokinetics in the elderly. Clin Geriatr Med 1990 May;6(2):257–67.
  9. Cockroft DW, Gault MH. Prediction of creatinine clearnance from serum creatinine. Nephron 1976;16(1):31-41.
  10. Lindeman RD. Assessment of renal function in the old. Special considerations. Clin Lab Med 1993 Mar;13(1):269-77.
  11. Institute for Safe Medication Practices. Common floor stock bowel prep drugs may pose problems in renal patients. ISMP Med Saf Alert 2001 May 16;10(6):2.
  12. Chertow GM, Lee J, Kuperman GJ, et al. Guided medication dosing for inpatients with renal insufficiency. JAMA 2001 Dec 12;286(22):2839-44.
  13. Institute for Safe Medication Practices (ISMP). Frequent problems with medication systems noted during ISMP hospital evaluations. ISMP Med Saf Alert 1998 Jun 3;11(3):2.
  14. Hogden LA, Low JK, Knoerlein KD, et al. Preprinted prescription forms decrease incomplete handwritten medication prescriptions in a neonatal intensive care unit. J Patient Saf 2005 Jun;1(2):100-4.
  15. Institute for Safe Medication Practices. Root causes: A roadmap to action. ISMP Med Saf Alert 2004 Aug 26;9(17):2.

Supplemental Material

Self Assessment Questions

The following questions about this article may be useful for internal education and assessment. You may use the following examples or come up with your own.

  1. Which node of the medication-use process is associated with the highest number of levofloxacin-related events?
    1. Administration
    2. Monitoring
    3. Prescribing
    4. Preparation
  2. Potential adverse events with levofloxacin include all of the following EXCEPT:
    1. Prolongation of the QTc interval
    2. Ototoxicity
    3. Clostridium difficile-associated colitis
    4. Tendon rupture
  3. Potential contributing factors to the prescribing of inappropriate levofloxacin doses may include all of the following EXCEPT:
    1. The clinical status of the patient may require the immediate administration of a first dose of an antibiotic.
    2. Computerized prescriber order-entry (CPOE) systems may not have effective computer-based clinical decision support to help guide appropriate dose selection.
    3. Essential patient information is provided at the top of all preprinted order sets and medication administration records, as well as screens within CPOE systems.
    4. Clinical laboratory values may not be readily available to practitioners at the time of prescribing.
  4. All of the following are risk reduction strategies to prevent errors with levofloxacin EXCEPT:
    1. Obtaining baseline patient information such as age and laboratory results
    2. Developing efficient methods for collecting and documenting information on prescribing errors
    3. Working with vendors to interface the laboratory system with CPOE systems
    4. Selecting a dose based on the patient’s liver function
  5. A 77-year-old female is diagnosed with acute bacterial exacerbation of bronchitis due to H. influenzae. The prescriber wants to order levofloxacin, which the patient used successfully about five years ago, to treat the infection.
    What would be the appropriate approach to dosing this patient to minimize the risk of adverse events?

    1. Check the patient’s renal function.
    2. Order a normal dose for the given indication.
    3. Order the same dose the patient received in the past.
    4. Check the patient’s liver function.
PSA LOGO

The Pennsylvania Patient Safety Advisory may be reprinted and distributed without restriction, provided it is printed or distributed in its entirety and without alteration. Individual articles may be reprinted in their entirety and without alteration, provided the source is clearly attributed.

Current and previous issues are available online at http://patientsafety.pa.gov.

©2020 Pennsylvania Patient Safety Authority