JAMA & ARCHIVES
Arch Fam Med
SEARCH
GO TO ADVANCED SEARCH
HOME  PAST ISSUES  TOPIC COLLECTIONS  CME  PHYSICIAN JOBS  CONTACT US  HELP
Institution: STANFORD Univ Med Center  | My Account | E-mail Alerts | Access Rights | Sign In
  Vol. 9 No. 8, August 2000 TABLE OF CONTENTS
  Archives
  •  Online Features
  Original Contribution
 This Article
 •Abstract
 •PDF
 •Send to a friend
 • Save in My Folder
 •Save to citation manager
 •Permissions
 Citing Articles
 •Citing articles on Web of Science (9)
 •Contact me when this article is cited
 Related Content
 •Related article
 •Similar articles in this journal
 Topic Collections
 •Dermatology
 •Immunology
 •Allergy
 •Hypersensitivity
 •Alert me on articles by topic

Costs of {beta}-Lactam Allergies

Selection and Costs of Antibiotics for Patients With a Reported {beta}-Lactam Allergy

Eric J. MacLaughlin, PharmD; Joseph J. Saseen, PharmD, BCPS; Daniel C. Malone, PhD

Arch Fam Med. 2000;9:722-726.

ABSTRACT

Objective  To evaluate antibiotic selection and the cost effect of reported {beta}-lactam allergies.

Design  Retrospective medical records review comparing antimicrobial selection and costs in patients with a reported {beta}-lactam allergy with a group in which no such allergy had been documented.

Setting  University-based family medicine clinic.

Patients  Patients who were prescribed at least 1 antibiotic for an upper respiratory tract infection, otitis media, sinusitis, and/or a urinary tract infection were eligible. One thousand two hundred one patients were identified via ICD-9-CM (International Classification of Diseases, Ninth Revision, Clinical Modification) codes. Four hundred sixty-five patients were initially identified and an additional 195 family members were eligible for inclusion.

Main Outcome Measures  Comparison of antimicrobial selection and costs (by average wholesale price) between patients with and without a reported {beta}-lactam allergy.

Results  Of the 660 patients eligible for inclusion, 99 (15%) had a documented {beta}-lactam allergy. Of the patients with a documented allergy, only 33% had a description of their purported reaction. The mean antibiotic cost for patients with a {beta}-lactam allergy was significantly higher compared with those without a {beta}-lactam allergy ($26.81 vs $16.28, respectively; P = .004). Patients with a {beta}-lactam allergy were more likely to have received a cephalosporin, macrolide, or a miscellaneous agent (eg, quinolone, tetracycline, or nitrofurantoin) (P = .001).

Conclusions  Patients with a {beta}-lactam allergy had higher antibiotic costs and were more likely to receive a broader-spectrum antibiotic. Most patients with a reported allergy did not have a description of their reaction. Skin testing may be of use in detecting true {beta}-lactam allergies; however, further study is needed to determine its cost-effectiveness.



INTRODUCTION
 Jump to Section
 •Top
 •Introduction
 •Patients and methods
 •Results
 •Comment
 •Conclusion
 •Author information
 •References

PENICILLIN antibiotics (eg, penicillin, amoxicillin, and ampicillin) are the most frequently prescribed class of antibiotics.1 They are effective, inexpensive antimicrobials and therefore remain the drugs of choice for many infections commonly seen in clinical practice. Additionally, infections for which penicillin is indicated as the primary antibiotic (eg, syphilis) are frequent and may be increasing in incidence.2

While the precise incidence of penicillin allergy is unknown, it is estimated to be present in 1% to 10% of the general population.3 The presence of an allergy may preclude the use of a {beta}-lactam antibiotic in many patients. In most cases, patients who report a {beta}-lactam allergy will receive a different medication from another antibiotic class. However, the alternative antibiotic chosen may have several limitations. It may be broader in spectrum, have a higher potential for adverse drug events, be more expensive, and have the potential for increased antibiotic resistance.

To our knowledge, to date there are limited or no data that address the economic effect of patient-reported {beta}-lactam allergies. This study examines a family medicine outpatient environment and describes the differences in cost and antimicrobial prescribing between patients who have a documented {beta}-lactam allergy and those who do not.


PATIENTS AND METHODS
 Jump to Section
 •Top
 •Introduction
 •Patients and methods
 •Results
 •Comment
 •Conclusion
 •Author information
 •References

This study was a retrospective medical records review of ambulatory patients who received an antibiotic for either an upper respiratory tract infection (URI), otitis media (OM), sinusitis (acute or chronic), and/or a urinary tract infection (UTI). The purpose was to examine selection and cost of antimicrobial agents for persons with and without a documented {beta}-lactam allergy.

Patients included were from the University of Colorado Health Science Center family medicine clinics in Denver. The University's institutional review board reviewed and approved the study protocol. Patients were initially identified if they had had 1 of the following International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) Codes within the preceding 2 years (July 1996 to June 1998): 465.9 (URI), 382.9 (OM), 461.9/473.9 (acute or chronic sinusitis), and 599.0 (UTI). Patients who did not receive an antimicrobial drug for the aforementioned ICD-9-CM codes were not included. Medical records were organized using family charting. Therefore, after an individual medical record review, other family members were reviewed for possible study inclusion. The rationale for including family members was the time lag between patient encounters and the data being entered into the medical facility's database (approximately 3-4 months). Therefore, family members who were not initially identified by ICD-9-CM codes could be included in the study. Additionally, by including family members, the power ({beta}) of the study was increased. For inclusion in the study, patients must have been prescribed at least 1 antimicrobial medication for a URI, OM, sinusitis (acute or chronic), and/or a UTI. These 4 disease states were selected because of their high incidence in ambulatory patients and the high likelihood of treatment with antimicrobial drugs. Patient records were reviewed between September 1998 and April 1999. The date range of prescriptions was from March 1, 1978, to April 13, 1999, with 95% of the observations occurring after August 23, 1993, and 90% after March 5, 1995. However, only the last antibiotic prescription in the medical record was included for statistical analysis.

Data collected included patient demographics (age, sex), concurrent chronic disease states, presence of a {beta}-lactam allergy, description of antibiotic allergy, and any adverse drug or allergic reactions reported in the medical record. For each encounter in which an antimicrobial drug was prescribed, the drug, dosage, and total dosage units were recorded. Encounters in which no antimicrobial drug was prescribed and those encounters that occurred at a different health care facility, such as a clinic, hospital, or emergency department, were not included in the data collection. All data were collected using standardized data collection forms and then entered into a Microsoft Access (Microsoft Inc, Redmond, Wash) database.

Patients were considered to have a {beta}-lactam allergy if such allergy was written in the medical record (eg, progress notes or medication lists). A {beta}-lactam medication was considered any antibiotic with a {beta}-lactam ring in the chemical structure (eg, penicillins and cephalosporins). Patients were not required to have undergone allergy testing to be considered to have a {beta}-lactam allergy.

Antimicrobial selections and costs were compared between those patients with a {beta}-lactam allergy and those without. Patient age, sex, number of diagnoses, and number of drug allergies were also compared. Patients were categorized into 1 of 6 age groups: infants and young children (<6 years), children and adolescents (6-17 years), young adults (18-30 years), adults (31-50 years), older adults (51-64 years), and elderly adults (>= 65 years). Prescribed medications were grouped into 5 classes based on chemical structure. These included penicillins, cephalosporins, sulfa derivatives, macrolides, and miscellaneous agents (eg, quinolones, tetracyclines, and nitrofurantoin). Using 1997 average wholesale price and total units prescribed, each course of antibiotic therapy was assigned a cost. Because it could not be determined whether the patient received a brand or generic product, the cost assigned to each prescription was the mean average wholesale price from available manufacturers. In situations where the dosage strength was not documented, cost was estimated by averaging all available strengths (5% of the cases).

Data were converted from Microsoft Access to SAS version 6.12 (SAS Institute Inc, Cary, NC) for analyses. Comparisons between groups were conducted using the Student t test for continuous variables, and {chi}2 statistics for noncontinuous data. Logistic regression was also conducted to examine the relationship between patient age and {beta}-lactam allergies. Differences were considered statistically significant at P<.05.


RESULTS
 Jump to Section
 •Top
 •Introduction
 •Patients and methods
 •Results
 •Comment
 •Conclusion
 •Author information
 •References

A total of 1201 patients were initially identified via ICD-9-CM diagnostic codes. Seven hundred thirty-six patients did not receive an antimicrobial medication. Therefore, 465 of the initial patients identified and 195 of their family members met study criteria and were included in the study, for a total of 660 patients. These 660 patients had a total of 2090 treatment encounters documented in the medical records.

Of the 660 patients included in this study, 99 (15%) had a documented {beta}-lactam allergy in their medical records. Differences in patient demographics are presented in Table 1. The mean age for patients with documented {beta}-lactam allergies was higher than the mean age for those without (30.7 vs 24.7 years; P = .006). The distribution of patient diagnoses was as follows: 243 with sinusitis; 95 with a UTI; 46 with a URI; 274 with OM; 1 patient with a UTI and sinusitis; and 1 patient with a URI and a UTI.


View this table:
[in this window]
[in a new window]
Table 1. Patient Demographics*


The percentage of patients with a reported {beta}-lactam allergy varied between each of the 6 age groups (Figure 1). In general, older patients were more likely to have a {beta}-lactam allergy. For each increase in age group, the odds of having a {beta}-lactam allergy increased by 22%. The odds ratio was 1.22 (95% confidence interval, 1.048-1.410).



View larger version (28K):
[in this window]
[in a new window]
The percentage of patients with a reported {beta}-lactam allergy for each age group. Age groups include: infants/young children (<6 years); children/adolescents (6-17 years); young adults (18-30 years); adults (31-50 years); older adults (51-64 years); and elderly adults (>=65 years).


Only 33 (33%) of the 99 patients with a documented {beta}-lactam allergy had a description of their purported allergy. These are described in Table 2.


View this table:
[in this window]
[in a new window]
Table 2. Descriptions of {beta}-Lactam Allergies as Reported in Medical Records


The mean antibiotic cost for all patients and all treatment encounters (ie, cost per illness) was $69.79. However, to control for the number of patient encounters, duration of the study, and length of time a patient may have been a member of the family medicine center, only the last encounter was used to compare costs of antimicrobial therapy. The distribution of antibiotic costs among various drug classes is provided in Table 3. The mean antibiotic cost for patients with a {beta}-lactam allergy was $26.81 compared with $16.28 for patients without an allergy (P = .004). Moreover, patients with a reported {beta}-lactam allergy were more likely to receive a cephalosporin, macrolide, or miscellaneous agent ({chi}24 = 57.19, P = .001).


View this table:
[in this window]
[in a new window]
Table 3. Comparisons of Antibiotic Cost per Episode for Allergic and Nonallergic Patients*


Patients who had a documented {beta}-lactam allergy had more concurrent chronic medical diagnoses such as hypertension or asthma (mean, 1.56 vs 0.90) and had other drug allergies documented more often (mean, 0.57 vs 0.20) than those who did not report a {beta}-lactam allergy (P = .001 and P<.001, respectively). Additionally, of the 99 patients with a {beta}-lactam allergy, 91 were allergic to penicillins, 10 to cephalosporins, and 2 to both penicillins and cephalosporins.


COMMENT
 Jump to Section
 •Top
 •Introduction
 •Patients and methods
 •Results
 •Comment
 •Conclusion
 •Author information
 •References

{beta}-Lactam allergies are common within the American population. As many as 10% of ambulatory and hospitalized patients report being allergic to a medication with a {beta}-lactam structure.4-5 In our study, 15% of ambulatory family medicine patients who received an antibiotic for infections of the upper respiratory tract, sinuses, urinary tract, and/or the middle ear reported a {beta}-lactam allergy. The presence of an allergy appears to significantly affect antimicrobial prescribing and medication costs. This study found that patients with {beta}-lactam allergies noted in the medical record were more likely to have received a cephalosporin, macrolide, or other antibiotic (eg, quinolone, tetracycline, or nitrofurantoin) than those who did not have an allergy documented in the medical record. On average, prescription costs were 37% higher ($26.81 vs $16.28). However, this difference is likely to be an underestimation of the true cost of reported {beta}-lactam allergies. In our study, only the last patient encounter was used. A comparison of the costs throughout a patient's life, with multiple encounters and antimicrobial prescriptions, would result in a much larger difference.

Patients with a penicillin allergy are more likely to have similar allergies to cephalosporins.6 Although this cross-reactivity is relatively low, it often translates into empirically avoiding the use of cephalosporins in patients with a reported {beta}-lactam allergy. However, our data suggest that this may not be true because patients were more likely to receive a cephalosporin if they were allergic to penicillin.

Our data suggest that there is a difference in age between patients who report an allergy and those who do not. On average, patients with a reported {beta}-lactam allergy were 6 years older. As patient age increased, the odds of a patient having a documented allergy increased. This observation is consistent with the natural history of documented drug allergies. Hypersensitivity reactions to penicillin depend on the presence of preformed antibodies.7 The likelihood of prior exposure to a {beta}-lactam increases with age. Therefore, older patients have a greater probability of having a documented allergy. Additionally, once an "allergy" is documented in the medical record, it will likely remain there for the life of the patient.

The date of the "allergic reaction" was rarely noted in the medical record. Therefore, given the available data, the incidence of newly diagnosed {beta}-lactam allergies per age group could not be determined. While penicillin reactions, even severe ones such as anaphylactic reactions, have occurred in both children and older people, young and middle-aged adults are at greatest risk of developing penicillin allergy.8

Inaccuracy in patient reporting of drug allergies is common. More than 50% of allergies reported by patients may not be immunologic in nature.9 While the majority of allergy descriptions documented in this study may have been IgE-mediated (19.2% were dermatological, 3% were anaphylactic), a substantial percentage of reactions appeared to be nonimmunologically mediated adverse effects: 5.1% were gastrointestinal, 5.1% were of the central nervous system. Of interest, 2 patients were documented to have a {beta}-lactam allergy based on "family history" alone. This lack of documented descriptions of {beta}-lactam allergies is similar to that found by Preston et al.5 In their study, the symptoms of the allergic reaction were described in the medical record for only 34% of hospitalized patients with a reported penicillin allergy.

There are several limitations to our study. One is the lack of accurate documentation in outpatient medical records. In many cases, the antibiotic regimen prescribed lacked the dose, frequency, and duration of treatment. Additionally, without access to pharmacy records from where the prescription was processed, it was not possible to determine if the brand or generic (of which there may be multiple manufacturers) product was dispensed or to determine if the patient actually got the prescription filled. To minimize price underestimation/overestimation, an average cost for the antibiotic regimen was used. Another limitation was the inability to use all encounters in which a patient received an antimicrobial agent. Including all visits would have raised the power of the study and shown a larger cost difference between patients who had a {beta}-lactam allergy and those who did not.

This study relied on documentation of the allergy in the medical record and did not assess actual sensitivity to IgE. Reports in the literature indicate that the true incidence of {beta}-lactam allergy is overstated.10-15 Many patients who report a penicillin allergy have negative skin tests and are not at risk for IgE-mediated allergic reactions.7-8,10, 16-17 For patients who report a history consistent with a true drug allergy, skin testing may be a consideration.

In 1980, 1985, 1989, and 1992, antimicrobial agents were the second leading therapeutic category of drugs prescribed by office-based physicians in the United States.1 Specifically, an estimated 110 million oral antimicrobial drugs were prescribed by office-based physicians in the United States in 1992. With the rising problem of antibiotic resistance, proper drug utilization is of paramount importance. Inaccurate reporting of {beta}-lactam allergies may have deleterious consequences. Patients who report a {beta}-lactam allergy often receive an agent that is more costly, potentially toxic, and has a broader spectrum of activity. In a study by Preston et al,5 agents that were substituted for penicillin in those who reported an allergy were potentially more toxic in 70.4% of cases, equally effective in all cases, and more costly in 55.5% of cases.


CONCLUSION
 Jump to Section
 •Top
 •Introduction
 •Patients and methods
 •Results
 •Comment
 •Conclusion
 •Author information
 •References

Reported {beta}-lactam allergies were associated with greater antibiotic costs. This may have been because of patients being more likely to receive a broader-spectrum antibiotic such as a cephalosporin, macrolide, or a miscellaneous agent (eg, quinolone, tetracycline, or nitro-furantoin). Because broad-spectrum antibiotics are often associated with greater costs, resistance, and toxic reactions, an attempt should be made to minimize their use. Further studies are warranted to determine the economic feasibility of confirming hypersensitivity by skin testing patients with a reported allergy. Until such studies are available, documenting an accurate medication history is of paramount importance.


AUTHOR INFORMATION
 Jump to Section
 •Top
 •Introduction
 •Patients and methods
 •Results
 •Comment
 •Conclusion
 •Author information
 •References

Accepted for publication November 24, 1999.

Corresponding author: Eric J. MacLaughlin, PharmD, School of Pharmacy, Texas Tech Health Sciences Center, 1300 S Coulter Dr, Amarillo, TX 79106 (e-mail: ericmac{at}cortex.ama.ttuhsc.edu). Reprints: Joseph J. Saseen, PharmD, BCPS, University of Colorado School of Pharmacy, 4200 E Ninth Ave, Box C238, Denver, CO 80262, (e-mail: joseph.saseen{at}uchsc.edu).

From the School of Pharmacy (Drs MacLaughlin, Saseen, and Malone) and the Department of Family Medicine (Dr Saseen), University of Colorado Health Sciences Center, Denver. Dr MacLaughlin is now with the School of Pharmacy, Texas Tech Health Sciences Center, Amarillo. Dr Malone is now with the College of Pharmacy, University of Arizona, Tucson.


REFERENCES
 Jump to Section
 •Top
 •Introduction
 •Patients and methods
 •Results
 •Comment
 •Conclusion
 •Author information
 •References

1. McCaig LF, Hughes JM. Trends in antimicrobial drug prescribing among office-based physicians in the United States. JAMA. 1995;273:214-219. FREE FULL TEXT
2. Lin RY. A perspective on penicillin allergy. Arch Intern Med. 1992;152:930-937. FREE FULL TEXT
3. Idsoe O, Guthe T, Willcox RR, de Weck AL. Nature and extent of penicillin side-reactions, with particular reference to fatalities from anaphylactic shock. Bull World Health Organ. 1968;38:159-188. ISI | PUBMED
4. Macy E. Elective penicillin skin testing and amoxicillin challenge: effect on outpatient antibiotic use, cost, and clinical outcomes. J Allergy Clin Immunol. 1998;102:281-285. FULL TEXT | ISI | PUBMED
5. Preston SL, Briceland LL, Lesar TS. Accuracy of penicillin allergy reporting. Am J Hosp Pharm. 1994;51:79-84. ABSTRACT
6. Petz LD. Immunologic reactions of humans to cephalosporins. Postgrad Med J. 1971;47(suppl):64-69.
7. Saxon A, Beall GN, Rohr AS, Adelman DC. Immediate hypersensitivity reactions to beta-lactam antibiotics [clinical conference]. Ann Intern Med. 1987;107:204-215.
8. Sogn DD. Penicillin allergy. J Allergy Clin Immunol. 1984;74:589-593. FULL TEXT | ISI | PUBMED
9. Kerr JR. Penicillin allergy: a study of incidence as reported by patients. Br J Clin Pract. 1994;48:5-7. ISI | PUBMED
10. Mendelson LM, Ressler C, Rosen JP, Selcow JE. Routine elective penicillin allergy skin testing in children and adolescents: study of sensitization. J Allergy Clin Immunol. 1984;73:76-81. FULL TEXT | ISI | PUBMED
11. Adkinson NF Jr, Thompson WL, Maddrey WC, Lichtenstein LM. Routine use of penicillin skin testing on an inpatient service. N Engl J Med. 1971;285:22-24.
12. Van Dellen RG, Gleich GJ. Penicillin skin tests as predictive and diagnostic aids in penicillin allergy. Med Clin North Am. 1970;54:997-1007. ISI | PUBMED
13. Levine BB, Zolov DM. Prediction of penicillin allergy by immunological tests. J Allergy. 1969;43:231-244. FULL TEXT | ISI | PUBMED
14. Green GR, Rosenblum AH, Sweet LC. Evaluation of penicillin hypersensitivity: value of clinical history and skin testing with penicilloyl-polylysine and penicillin G: a cooperative prospective study of the penicillin study group of the American Academy of Allergy. J Allergy Clin Immunol. 1977;60:339-345. FULL TEXT | ISI | PUBMED
15. Sullivan TJ, Wedner HJ, Shatz GS, Yecies LD, Parker CW. Skin testing to detect penicillin allergy. J Allergy Clin Immunol. 1981;68:171-180. FULL TEXT | ISI | PUBMED
16. Harris AD, Sauberman L, Kabbash L, Greineder DK, Samore MH. Penicillin skin testing: a way to optimize antibiotic utilization. Am J Med. 1999;107:166-168. FULL TEXT | ISI | PUBMED
17. Pichichero ME, Pichichero DM. Diagnosis of penicillin, amoxicillin, and cephalosporin allergy: reliability of examination assessed by skin testing and oral challenge. J Pediatr. 1998;132:137-143. FULL TEXT | ISI | PUBMED

RELATED ARTICLE

The Archives of Family Medicine Continuing Medical Education Program
Arch Fam Med. 2000;9(8):735.
FULL TEXT  





HOME | CURRENT ISSUE | PAST ISSUES | TOPIC COLLECTIONS | CME | PHYSICIAN JOBS | HELP
CONDITIONS OF USE | PRIVACY POLICY | CONTACT US | SITE MAP
 
© 2000 American Medical Association. All Rights Reserved.