Current Antibiotic Therapy for Isolated Urinary Tract Infections in Women

ORIGINAL INVESTIGATION Current Antibiotic Therapy for Isolated Urinary Tract Infections in Women

Alexander J. Kallen, MD; H. Gilbert Welch, MD, MPH; Brenda E. Sirovich, MD, MS

Background: Sulfa antibiotics, such as a combination recent year of data, quinolones were prescribed in 48% product of trimethoprim and sulfamethoxazole, have tra- and sulfas in 33% of UTI visits (PϽ.04). Quinolones were ditionally been the drugs of choice for urinary tract in- significantly more likely to be prescribed to older pa- fections (UTIs) and remained the most common treat- tients and in visits occurring in the Northeast; however, ment as recently as a decade ago. However, increasing no difference in quinolone prescribing was seen when sulfa resistance among Escherichia coli may have led to evaluating insurance status, setting, race, ethnicity, health changes in prescribing practices. care provider type, and year. Approximately one third of the quinolones used were broader-spectrum agents. Methods: We used the 2000-2002 National Ambula- tory Medical Care Survey and National Hospital Ambu- Conclusions: Quinolones have surpassed sulfas as the latory Medical Care Survey to obtain nationally repre- most common class of antibiotic prescribed for isolated sentative data on antibiotics prescribed for women with outpatient UTI in women. Few significant predictors of isolated outpatient UTIs following visits to physicians’ quinolone use exist, suggesting that the increase is not con- offices, hospital clinics, and emergency departments (n=2638). Logistic regression was used to determine pre- fined to a certain subset of patients. This pervasive growth dictors of quinolone use. in quinolone use raises concerns about increases in resistance to this important class of antibiotics. Results: Quinolones were more commonly prescribed than sulfa antibiotics in each year evaluated. In the most Arch Intern Med. 2006;166:635-639

RINARY TRACT INFECTIONS treating UTIs when resistance becomes a po- (UTIs) are a commonly tential problem. Nalidixic acid, released in encountered problem 1963, was the original quinolone; how- among women treated in ever, it has been replaced by a large num- ambulatory care settings. It ber of agents with greater spectrums of ac- isU estimated that one third of women will tivity. Narrower-spectrum first-generation have been treated for at least one UTI by the agents, such as norfloxacin, ofloxacin, and age of 26 years, a figure that increases to ap- ciprofloxacin, are most active against gram- proximately 60% during a woman’s life- negative organisms, whereas broader- time.1,2 Escherichia coli has been shown to spectrum second- and third-generation cause approximately 80% of outpatient UTIs agents, such as levofloxacin, moxifloxacin in otherwise healthy people.3 National es- hydrochloride, and gatifloxacin, add greater timates showed that sulfa antibiotics, such gram-positive and anaerobic coverage.7 as a combination product of trimethoprim No nationwide review has been performed of antibiotic therapy for UTIs since CME course available at the late 1990s.4,8 Because of concerns about increasing inappropriate quinolone use,9,10 www.archinternmed.com we sought to describe the current use of antibiotics for UTIs, the proportion of UTI and sulfamethoxazole, were the dominant visits resulting in a quinolone prescrip- UTI treatment as late as 1995-1996.4 How- tion, predictors of quinolone use, and the ever, because the prevalence of resistance proportion of quinolone use that is made to trimethoprim-sulfamethoxazole among up of broader-spectrum agents. E coli has been increasing, some experts now suggest that trimethoprim-sulfamethoxa- METHODS zole should not be considered a first-line Author Affiliations: VA Outcomes Group, VA Medical agent in areas where the prevalence of re- 5,6 DATA SOURCE Center, White River Junction, sistant E coli exceeds 15% to 20%. Vt, and Dartmouth Medical Quinolones represent a natural alterna- We combined data from the 3 most recent years School, Hanover, NH. tive to trimethoprim-sulfamethoxazole for (2000-2002) of the National Ambulatory Medi-

(REPRINTED) ARCH INTERN MED/ VOL 166, MAR 27, 2006 WWW.ARCHINTERNMED.COM 635

Downloaded From: https://jamanetwork.com/ on 10/02/2021 midwife) or house staff (information on house staff available only from NHAMCS, not from NAMCS). We also determined Women (Age >17 years) With UTIs whether a urinalysis or urine culture was performed. Data for (N = 3030) urine cultures were available only for the years 2001 and 2002. Admitted to the Hospital (n = 268) ANTIBIOTICS Women With UTIs Not Admitted (n = 2762) Data on antibiotics were included in 1914 of the UTI visits (the denominator used for our analyses of antibiotic use). Each visit Coinfection Present was associated with up to 6 medications and their correspond- (n = 72) ing drug classes. We determined antibiotic class using a vari- Women With Outpatient UTIs able included in the data set that is based on the 1995 Na- Without Coinfection tional Drug Code Directory Drug Classes, except in the case of (n = 2690) nitrofurantoin, which was identified using the actual medica- Upper UTIs Present tion code rather than the class. This was because nitrofuran- (n = 52) toin was categorized in a class that included drugs not used for Women With Isolated UTI treatment (urinary tract antiseptics, class 0354), such as Outpatient UTIs phenazopyridine. Specific quinolones were identified by name (n = 2638) using their brand and generic name codes. These antibiotics were further categorized into narrower-spectrum (norfloxacin, ofloxacin, ciprofloxacin) and broader-spectrum groups (le- Figure 1. Selection process used to determine isolated outpatient urinary vofloxacin, gatifloxacin, trovafloxacin).14,15 tract infection (UTI) visits. DATA ANALYSIS cal Care Survey (NAMCS) and the National Hospital Ambula- tory Medical Care Survey (NHAMCS). These surveys, admin- Descriptive analyses were performed on the general character- istered by the National Center for Health Statistics of the Centers istics of UTI visits and the type of antibiotic used in each visit. for Disease Control and Prevention, gather data on a nation- Multiple logistic regression was performed to determine pre- ally representative sample of outpatient visits to nonfederally dictors of quinolone use. Because this was a common out- funded office-based physicians (NAMCS), nonfederal short- come, all odds ratios were converted to relative risks using the stay hospital emergency departments (NHAMCS-ED), and hos- method of Zhang and Yu.16 All analysis incorporated appro- pital outpatient departments (NHAMCS-OPD) in the United priate sampling weights and accounted for a multistage clus- States. Using multistage probability sampling, these surveys are tered sampling strategy using the survey commands in STATA designed to give national estimates of the characteristics of out- 8.2 (StataCorp, College Station, Tex). Our weighted data rep- patient visits in each of these 3 settings. In each survey a ran- resent approximately 27 million visits. domly selected health care provider is asked to complete data collection forms on a random sample of patient visits during a RESULTS prespecified period, supplying data such as reason for visit, di- agnoses, and medications prescribed.11-13 CHARACTERISTICS OF UTI VISITS IDENTIFICATION OF UTI VISITS The UTI visits were most likely to take place in physicians’ offices (75%) rather than the emergency depart- Figure 1 shows our method for identifying visits for isolated outpatient UTIs in women (subsequently referred to as a UTI ment (18%) or hospital clinics (8%) (Table). The typi- visit). We included visits in our analysis if they involved adult cal patient was a white woman approximately 50 years (18 years or older) women who had any of 3 International Clas- old with private insurance. Although urinalysis was com- sification of Diseases, Ninth Revision (ICD-9) codes for UTI listed monly part of the workup (76%), an order for urine cul- as a diagnosis: acute cystitis (595.0), unspecified cystitis (595.9), ture was relatively rare (23%). or site-unspecified UTI (599.0). To best capture uncomplicated isolated outpatient UTIs, we started by excluding pa- ANTIBIOTIC USE tients whose disposition was hospital admission. We then excluded those who had a coexisting infection (eg, upper Throughout the study period (2000-2002), quinolones respiratory tract infections, ICD-9 codes 460-466 and 473; pneu- were the most common antibiotic class prescribed (44% monia, ICD-9 codes 480-487; otitis media, ICD-9 code 382; or of visits), followed by sulfa antibiotics (30%) and nitro- cellulitis and abscess, ICD-9 codes 680-686). Last, patients who furantoin (18%). An apparent nonsignificant trend oc- had an upper UTI (ICD-9 code 590) were also excluded. This yielded 2638 sample visits, the denominator used in our analy- curred toward increasing quinolone use with each suc- sis of the general characteristics of UTI visits. cessive year of data: 38% in 2000, 46% in 2001, and 48% in 2002 (P=.11). As shown in Figure 2, in the most re- UTI VISIT CHARACTERISTICS cent year of data (2002), quinolones were prescribed in 48% of UTI visits, and antibiotics and sulfas were pre- Ͻ For each eligible visit, we determined patient age and race and scribed in 33% (P for difference .04). visit payer, practice setting, region, and health care provider type. If more than one health care provider type was listed (eg, PREDICTORS OF QUINOLONE USE nurse practitioner and attending physician or house staff and attending physician), visits were assigned to the midlevel health Figure 3 shows that age and region were the only sig- care provider (nurse practitioner, physician assistant, or nurse nificant predictors of quinolone use. Quinolones were sig-

(REPRINTED) ARCH INTERN MED/ VOL 166, MAR 27, 2006 WWW.ARCHINTERNMED.COM 636

Downloaded From: https://jamanetwork.com/ on 10/02/2021 nificantly more likely to be used if the patient was older than 30 years. Women 30 to 49 years old (45%) and 50 Table. Characteristics of All Visits for Isolated years or older (49%) were significantly more likely to re- Outpatient Urinary Tract Infection in Women ceive a quinolone than those younger than 30 years (31%). Quinolones were significantly less likely to be prescribed Characteristic Episodes, % in visits occurring in the Midwest (41%) or West (35%) Age range, y compared with the Northeast (55%). The prescribing of 18-29 21 30-49 32 quinolones did not vary significantly by race, ethnicity, Ն50 47 health care provider type, setting type, or year of visit. Race/ethnicity We performed additional subanalyses on 2 variables White 87 that were not consistently available in our data: physi- African American 10 cian specialty and whether an encounter was an initial Asian/Pacific Islander 3 visit or a follow-up visit. When specialty was included Hispanic 11 Ͻ in a model, obstetrician-gynecologist (n=357) was a sig- Other 1 Insurance status nificant negative predictor of quinolone use (relative risk, Private 52 0.31; 95% confidence interval, 0.11-0.73). When initial Medicare or Medicaid 35 and follow-up visit were included in a model (n=1358), Other 13 the variable had little impact (relative risk for follow- Site of care up, 0.96; 95% confidence interval, 0.67-1.18). Physician’s office 75 Emergency department 18 Hospital clinic 8 SPECTRUM OF QUINOLONE USE Region Northeast 21 Ciprofloxacin (61%) and levofloxacin (32%) were the Midwest 23 most commonly used quinolones in 2002. Narrower- South 37 West 19 spectrum quinolones were more commonly used than Health care provider broader-spectrum quinolones. Nevertheless, broader- Physician 86 spectrum quinolones were used in approximately one House staff* 2 third of UTI visits in 2002 (37%). Midlevel† 8 Other 4 Diagnostic evaluation COMMENT Urinalysis performed 76 Urine culture performed‡ 23 Quinolones have become the dominant treatment for UTI. *Data available only from National Hospital Ambulatory Medical Care Their use has steadily increased; between 1989-1990 and Survey; denominator is all urinary tract infection visits (N = 2638). 1997-1998, reported use of quinolones for the treat- †Midlevel category includes nurse practitioner, physician assistant, and ment of UTIs increased from 19% to 29% in one study.4 nurse midwife. ‡Data available from 2001 and 2002; denominator is only 2001 and 2002 Our analysis demonstrates that this trend has continued visits for urinary tract infection treatment (N = 1837). through 2002, with quinolones now firmly established as the most common class of antibiotics used for treatment of UTI. Few patient and visit characteristics pre- dicted quinolone use, suggesting that the increase in the use of quinolones is not limited to a few subgroups of 50 P for Difference patients or to certain settings. Finally, when quinolones .04 are prescribed, broader-spectrum agents from this class 40 are too often used. Although it is tempting to ascribe increased quino- 30 lone use to rising sulfa resistance, patterns of resistance across both geography and time do not support the ob- 20 served quinolone use. Sulfa resistance rates vary greatly Prescriptions, % by area.17 If the increasing use of quinolones was solely 10 a response to increasing resistance, one would expect higher rates of quinolone use in areas of higher resis- 0 Quinolones Sulfa Nitrofurantoin Cephalosporins Other tance. However, although at the time these data were col- Classes lected rates of sulfa-resistant E coli were lowest in the Drug Class Northeast,17,18 we found that these were the areas where quinolones were most likely to be prescribed. Addition- Figure 2. Antibiotic class prescribed during visits for treatment of urinary ally, the West, which had among the highest sulfa resis- tract infections (2002). These categories sum to more than 100 because 7% tance,17,18 was the area with the lowest proportion of qui- of visits involved more than 1 antibiotic. nolone use. Further, patterns of quinolone use do not track with sulfa resistance rates over time. Overall nation- leading to speculation that rates of trimethoprim- wide resistance rates remained at approximately 16% be- sulfamethoxazole resistance are plateauing.17-19 Despite tween 2 national studies performed in 1998 and 2001, stable resistance patterns, we found that use of quino-

(REPRINTED) ARCH INTERN MED/ VOL 166, MAR 27, 2006 WWW.ARCHINTERNMED.COM 637

Downloaded From: https://jamanetwork.com/ on 10/02/2021 Receiving Adjusted Relative Rate More Likely to Receive More Likely to Receive Variable Quinolone, % (95% CI) Nonquinolone Quinolone Age range, y 18-29 31 Referent 30-49 45 1.52 (1.10-1.96) ∗ ≥50 49 1.64 (1.15-2.14) Insurance status Private insurance 45 Referent Medicaid or Medicare 47 1.03 (0.77-1.29) Setting Physician’s office 45 Referent Hospital clinic 34 0.81 (0.58-1.07) Emergency department 42 1.08 (0.86-1.30) Race White 44 Referent Not white 40 0.95 (0.66-1.28) Ethnicity Not Hispanic 45 Referent ∗ Hispanic 38 0.86 (0.48-1.33) Health care provider Attending physician 43 Referent House staff 35 0.92 (0.64-1.22) ∗ Midlevel 48 1.14 (0.71-1.58) Region Northeast 55 Referent Midwest 41 0.73 (0.50-0.97) ∗ South 44 0.81 (0.55-1.08) ∗ West 35 0.66 (0.41-0.97) Year 2000 38 Referent 2001 46 1.19 (0.82-1.58) 2002 48 1.23 (0.95-1.52)

0.5 0.7 1.5 2.0

Figure 3. Percentage of patients receiving and adjusted relative rate of receiving a quinolone among women with isolated outpatient urinary tract infections (2000-2002). The relative rate was adjusted for age, insurance status, setting, race, ethnicity, health care provider type, region, and year. Midlevel health care provider includes nurse practitioner, physician assistant, and nurse. Asterisk indicates that relative standard error for these estimates was greater than 30% and therefore may be less reliable (all estimates were based on a sample size greater than 30). CI indicates confidence interval.

lones has continued to increase. These observations across Our analysis is subject to several limitations. First, be- geography and time demonstrate that patterns of sulfa cause this analysis is based on a survey, any conclusions resistance are not the primary explanation of patterns of drawn depend on the completeness and accuracy of sur- quinolone use for UTI. vey responses. Failure to correctly document the actual an- Furthermore, it is not clear to what extent measured re- tibiotics prescribed could lead to inaccurate estimates of sistance rates provide useful guidance for quinolone use. their use. It is unlikely, however, that any inaccuracy would First, resistance rates are typically measured in vitro (eg, systematically lead to an overestimation of quinolone use. on a culture plate) and likely overestimate clinically rel- Furthermore, survey responses in the NAMCS have been evant resistance. Clinical cure rates with trimethoprim- shown to compare favorably with measurements by di- sulfamethoxazole are estimated to be close to 85% even rect observation for the provision of discrete services such when sulfa resistance rates among E coli are 30% (well above as procedures and tests.22 Second, some care for symptom- those currently seen in the United States). This finding re- atic UTI takes place without a face-to-face encounter and flects both the activity of the drug on sensitive species and thus would not be included among our data. It may be dif- spontaneous resolution of a subset of resistant infections.2 ficult to extrapolate our results to this subset of patients. Second, current methods of measuring in vitro sulfa Third, some detail that would have been useful in in- resistance may overestimate resistance rates at the popu- terpreting our results was not available. Clinical detail lation level. Our finding that most visits for UTI (77%) was lacking for medication allergies and pregnancy sta- do not result in a urine culture is in keeping with rec- tus, two factors that could potentially influence the use ommendations by experts that patients with typical symp- of quinolones for UTI treatment. It would, however, seem toms and without risk factors for complicated UTIs may unlikely that the increasing use of quinolones during the be treated empirically.20,21 However, the unintended effect last decade would be explained by either an increase in of selectively performing urine cultures may be to in- the rate of sulfa allergy or a decrease in the proportion flate estimates of antibiotic resistance by oversampling of pregnant patients. Geographic detail was lacking at a cases where resistant organisms are more likely. This level finer than the 4 census regions; thus, our finding would overestimate the true prevalence of resistant or- of a mismatch between quinolone use and resistance pat- ganisms, especially in studies that base their estimates terns could differ in an analysis of smaller areas (ie, state, on cultures obtained in clinical practice. Prevalence es- county, or hospital service area). Nevertheless, this level timates of resistant organisms that are obtained from these of aggregation is similar to that frequently reported in types of studies, such as the antibiograms produced by US population–based sulfa resistance data.17,18 many institutions, may therefore mislead clinicians at- Finally, it is not clear why 27% of those visits with a tempting to choose appropriate antibiotics. diagnosis of UTI did not involve the prescription of an-

(REPRINTED) ARCH INTERN MED/ VOL 166, MAR 27, 2006 WWW.ARCHINTERNMED.COM 638

Downloaded From: https://jamanetwork.com/ on 10/02/2021 tibiotics. Similar results have been reported in another Enhancement Award 03-098 from the Department of Vet- NAMCS-based study.4 At least 3 possible explanations erans Affairs to investigate the harms from excessive medi- exist for this observation. First, some of these UTI visits cal care. may represent follow-up visits and therefore not result Disclaimer: The views expressed herein do not neces- in a new antibiotic prescription. Second, some visits may sarily represent the views of the Department of Veterans reflect physician visits where antibiotics are withheld Affairs or the US government. pending results of a culture or other tests. Third, incor- Previous Presentation: This study was presented as a rect diagnosis codes or drug class codes could have been poster at the 28th Annual Meeting of the Society of Gen- used for some of these visits. Regardless of the explana- eral Internal Medicine; May 14, 2005; New Orleans, La. tion, exclusion of these patients from our analysis of an- Acknowledgment: We thank the members of the VA Out- tibiotic use is unlikely to have systematically biased our comes Group and Jennifer Shu, MD, for their thought- estimates of quinolone use. ful review of this project. In conclusion, quinolones have now replaced sulfa antibiotics as the most common antibiotic prescribed during visits for uncomplicated UTIs. The increase in quino- REFERENCES lone use may be a reaction to potentially inflated measures of sulfa resistance among E coli or exaggerated percep- 1. Foxman B, Barlow R, D’Arcy H, Gillespie B, Sobel JD. Urinary tract infection: self- tions of resistance among physicians or it may be due to reported incidence and associated costs. Ann Epidemiol. 2000;10:509-515. 2. Hooton TM, Besser R, Foxman B, Fritsche TR, Nicolle LE. Acute uncomplicated other factors, including clinicians’ increasing familiarity cystitis in an era of increasing antibiotic resistance: a proposed approach to em- and comfort with quinolones. Regardless, much of this in- pirical therapy. Clin Infect Dis. 2004;39:75-80. creased use is likely unnecessary and may be associated 3. Ronald A. The etiology of urinary tract infection: traditional and emerging pathogens. with increased cost and the further development of resis- Am J Med. 2002;113(suppl 1A):14S-19S. 4. Huang EC, Stafford RS. National patterns in the treatment of urinary tract infec- tance to this important class of antimicrobials. tions in women by ambulatory care physicians. Arch Intern Med. 2002;162: Given these concerns, we believe it is time to reassess 41-47. antibiotic treatment strategies for UTIs. In particular, it is 5. Gupta K, Scholes D, Stamm W. Increasing prevalence of antimicrobial resis- important to persuade clinicians that narrower-spectrum tance among uropathogens causing uncomplicated cystitis in women. JAMA. agents are preferable to broader-spectrum agents to 1999;281:736-738. 6. Warren JW, Abrutyn E, Hebel J, Johnson J, Schaffer A, Stamm W. Guidelines reduce the selective pressure for resistance to the latter. for the treatment of uncomplicated acute bacterial cystitis and acute pyelone- Using sulfas (or nitrofurantoin) instead of a quinolone or phritis in women. Clin Infect Dis. 1999;29:745-758. using a narrower-spectrum quinolone rather than a 7. Bakken JS. The fluoroquinolones: how long will their utility last? Scand J Infect broader-spectrum quinolone will help preserve the clini- Dis. 2004;36:85-92. 8. Steinman MA, Gonzales R, Linder JA, Landefeld CS. Changing use of antibiotics cal activity of these important antimicrobials. Addition- in community-based outpatient practice, 1991-1999. Ann Intern Med. 2003; ally, to better inform treatment decisions, researchers need 138:525-534. to develop better methods to measure clinically relevant 9. Linder JA, Huang ES, Steinman MA, Gonzales R, Stafford RS. Fluoroquinolone (in vivo) rates of sulfa resistance in representative popu- prescribing in the United States 1995 to 2002. Am J Med. 2005;118:259-268. lations of patients treated (not just those cultured) for UTIs, 10. Lautenbach E, Larosa L, Kasbekar N, Peng H, Maniglia R, Fishman N. Fluoro- quinolone utilization in the emergency departments of academic medical centers. as well as specific clinical predictors of such resistance. Arch Intern Med. 2003;163:601-605. 11. Hing E, Middleton K. National Hospital Ambulatory Medical Survey: 2001 Out- Accepted for Publication: October 30, 2005. patient Department Survey. Hyattsville, Md: National Center for Health Statis- Correspondence: Alexander J. Kallen, MD, VA Out- tics; August 5, 2003. No. 338. 12. McCaig L, Burt C. National Hospital Ambulatory Medical Care Survey: 2001 Emer- comes Group, 111B, 215 N Main St, VA Medical Cen- gency Department Summary. Hyattsville, Md: National Center for Health Statis- ter, White River Junction, VT 05009 (alexander.j.kallen tics; June 4, 2003. No. 335. @dartmouth.edu). 13. CherryD,BurtC,WoodwellD.NationalAmbulatoryMedicalCareSurvey:2001Summary. Author Contributions: Dr Kallen had full access to all Hyattsville, Md: National Center for Health Statistics; August 11, 2003. No. 337. the data in the study and takes responsibility for the in- 14. Schaeffer AJ. The expanding role of quinolones. AmJMed. 2002;113(suppl 1A):45s-54s. tegrity of the data and the accuracy of the data analysis. 15. Norrby SR. Levofloxacin. Expert Opin Pharmacother. 1999;1:109-119. Financial Disclosure: During the period 2000-2002 (be- 16. Zhang J, Yu K. What’s the relative risk? a method of correcting the odds ratio in fore the initiation of this project), Dr Kallen served as a cohort studies of common outcomes. JAMA. 1998;280:1690-1691. member of the Speakers’ Bureau for Johnson & Johnson 17. Karlowsky JA, Kelly LJ, Thornsberry C, Jones ME, Sahm DF. Trends in antimicrobial resistance among urinary tract isolates of Escherichia coli from female (makers of levofloxacin, a drug included in this analy- outpatients in the United States. Antimicrob Agents Chemother. 2002;46:2540- sis). He received several (3) honoraria for lectures he gave 2545. during this time. All these lectures were on general in- 18. Gupta K, Sahm D, Mayfield D, Stamm W. Antimicrobial resistance among uro- fectious disease topics (not drug specific). In this ar- pathogens that cause community-acquired urinary tract infections in women: a ticle, levofloxacin is not singled out specifically with re- nationwide analysis. Clin Infect Dis. 2001;33:89-94. 19. Gupta K. Emerging antibiotic resistance in urinary tract pathogens. Infect Dis Clin spect to other members of its class. Furthermore, the North Am. 2003;17:243-259. article argues against increased use of this class of drugs. 20. Bent S, Saint S. The optimal use of diagnostic testing in women with acute un- Funding/Support: Dr Kallen is supported by a Veterans complicated cystitis. Am J Med. 2002;113(suppl 1A):20S-28S. Administration MD Ambulatory Care Postdoctoral Fel- 21. Fihn SD. Acute uncomplicated urinary tract infection in women. N Engl J Med. 2003;349:259-266. lowship. Dr Sirovich is supported by a Veterans Affairs 22. Gilchrest VJ, Stange KC, Flocke SA, McCord G, Bourguet CC. A comparison of Career Development Award in Health Services Research the National Ambulatory Medical Care Survey (NAMCS) measurement approach & Development. This study was supported by Research with direct observation of outpatient visits. Med Care. 2004;42:276-280.

(REPRINTED) ARCH INTERN MED/ VOL 166, MAR 27, 2006 WWW.ARCHINTERNMED.COM 639

Downloaded From: https://jamanetwork.com/ on 10/02/2021