Community-acquired pneumonia (CAP) remains a. Comparison of 3 Techniques for Implementing a Guideline for Community-Acquired Pneumonia

Transcription

1 Comparison of 3 Techniques for Implementing a Guideline for Community-Acquired Pneumonia Gregory A. Volturo, MD, Virginia B. Mangolds, MSN, FNP-C, Jennifer L. Mazzola, PharmD, Ronald J. DeBellis, PharmD, and Oren P. Schaefer, MD Abstract Objective: To compare 3 techniques used in implementing a community-acquired pneumonia (CAP) guideline: traditional educational sessions, academic detailing, and preprinted order forms. Setting and participants: Patients with CAP receiving care at an academic medical center in central Massachusetts. Methods: A retrospective analysis of medical records was performed to assess guideline adherence. Results: 587 patients with CAP were identified; 71 were excluded because their primary diagnosis was not CAP. After traditional guideline education, 29% of patients were treated according to the guideline, with 75% treated with preferred formulary antibiotics. Following academic detailing, 24% of patients were treated according to the guideline, with 71% treated with formulary antibiotics. Following implementation of preprinted order forms, 38% of patients were treated according to the guideline, with 77% treated with formulary antibiotics. Conclusion: Overall, guideline adherence was suboptimal for all interventions at both sites. Formulary adherence was better, with nearly three quarters of patients receiving formulary-designated antibiotics. Implementation of preprinted orders was associated with high utilization of appropriate antibiotic therapy. Community-acquired pneumonia (CAP) remains a significant cause of morbidity, mortality, and economic burden. In the United States, CAP is the sixth leading cause of death overall and the number one cause of death from infectious diseases in adults [1]. The direct health care cost for treating patients with CAP exceeds $9.7 billion annually [2]. Guidelines for management of CAP were first published in 1993 by the American Thoracic Society (ATS) [3], the British Thoracic Society [4], and the Canadian Infectious Disease Society [5], and in 1998 by the Infectious Diseases Society of America (IDSA) [6]. Both the ATS and IDSAguidelines have recently been updated [3,7]. Studies have demonstrated benefits of guideline use in the management of CAP, such as reduced mortality, shorter length of hospital stay, and reduced overall costs [8 12]. Despite this, guidelines often are not followed [13 18]. We used 3 techniques to implement a CAP guideline in a health care system: (1) traditional direct mailing and departmental educational sessions, (2) targeted small-group lectures to physicians caring for a high volume of CAP patients who were identified as noncompliant with the CAP guideline (academic detailing), and (3) utilization of preprinted order forms. Each intervention was independent from the others, separated by several months in time. We then examined the impact that each technique had on the ultimate utilization of the guideline. Setting UMass Memorial Medical Center is a 765-bed, tertiary care academic medical center in central Massachusetts consisting of 2 inpatient campuses. The University campus is a traditional academic setting with virtually all of the patients being managed by house officers with attending supervision. The supervising attending may not necessarily be the patient s usual physician. The Memorial campus is a community teaching hospital where some of the patients are cared for by house officers; however, the patient s usual physician has a much more active role in inpatient management. Additionally, some patients are cared for only by their personal physician and do not have house officers involved in their care. Physician relationships with UMass Memorial Medical Center vary from employed providers to private practice physicians with courtesy admitting privileges. There are no financial or administrative incentives for practitioners to use the CAP guideline we developed. From the University of Massachusetts Medical School (Drs. Volturo and Schaefer and Ms. Mangolds) and the Massachusetts College of Pharmacy and Health Services (Drs. Mazzola and DeBellis). Vol. 12,. 4 April 2005 JCOM 199

2 COMMUNITY-ACQUIRED PNEUMONIA Guideline Development A multispecialty team consisting of 12 physicians from internal medicine, infectious diseases, pulmonary medicine, primary care, and emergency medicine along with nurses and pharmacists within our hospital system developed a set of CAP treatment and disposition guidelines (Figure 1). The team met on several occasions, reviewed data regarding current CAP treatment practices at our hospital, determined the best method for dissemination of the guidelines based on prior institutional experience, and reviewed the current literature regarding the treatment of CAP. Ultimately our guidelines were based on ATS and IDSAguidelines, evidence-based risk factors suggesting the need for admission (Pneumonia Severity Index [19]), local clinical practice, local resistance rates, and organism prevalence. The final guideline reflected the consensus of this multispecialty group. Implementation The guidelines were first implemented in vember 2001 through broad distribution of a pocket card to medical staff and house officers, which was followed by educational sessions in key departments caring for patients with CAP (ie, medicine, family medicine, and emergency medicine). All educational sessions were delivered by 2 of the authors (GAV, OPS) using a standard slide set. The sessions were provided as departmental grand rounds and as part of the house officer s core curriculum lecture series. In addition to distribution of the guideline card, CAP guidelines were posted in all patient care areas and worksheets were available in all emergency departments, clinics, and wards where patients were treated. Four months after initial implementation, physicians who admitted a high volume of patients with pneumonia (more than 5 patients within the past 6 months) yet had low guideline utilization (< 80%) were provided with additional education. These physicians received personal written and phone invitations from the authors. Three small-group educational sessions (10 to 15 physicians each) were held as dinner meetings at convenient locations. Physicians were presented with practice site specific guideline utilization data, hospital-wide guideline utilization and mortality data, and education regarding the evidence supporting the CAP guideline. Ten months after initial implementation, without further education, preprinted CAP order forms consistent with the guidelines were implemented in all clinics, emergency departments, and patient care wards. A letter was sent to all medical staff and house officers providing notice that the CAP standard order forms were now available in all patient care areas and that they could be printed from any hospital computer. It was recommended that these should be utilized for all patients being treated for CAP. Assessment After each intervention phase, patients were retrospectively identified from a central database of all patients admitted to the hospital or treated in the outpatient clinics or emergency departments. All patients over 18 years of age with a clinical and/or radiographic diagnosis of pneumonia, identified by ICD-9 code, who were evaluated at either hospital campus were included. Patients were excluded if they had been admitted to the hospital within the preceding 14 days, had a history of HIV/AIDS, were suspected of aspiration, or had presented from a nursing home in which they had resided for more than 14 days. A 7-person team consisting of physicians, nurses, and pharmacists reviewed the charts of eligible patients using a standardized data collection tool. Chart reviewers completed inter-reviewer reliability testing prior to data collection, demonstrating a 90.5% agreement among the 7 reviewers calculation of the pneumonia severity risk category. The following data were collected: patient demographics, initial vital signs, place of service, CAP-pertinent comorbid conditions, previous antibiotic use, pneumonia severity score, time of medication administration, admission status (ward or intensive care unit), length of hospital stay for admitted patients, and mortality. The diagnosis of pneumonia was determined based on retrospective review of the discharge diagnosis. Questions regarding individual irregularities of patient presentations arising during chart review were brought to the study group for a consensus determination. We defined guideline adherence as administration of the appropriate antibiotic within 4 hours and subsequent admission or discharge to the appropriate treatment setting. As appropriate antibiotic utilization is an important aspect of CAP treatment, we also looked specifically at use of guideline formulary antibiotics independent of other guideline compliance parameters. The study was approved by the institutional review board for research in human subjects of the University of Massachusetts Medical School. Financial support was provided in the form of a research grant from Roche Pharmaceuticals. Results Overall, 587 patients were identified with a discharge diagnosis of CAP. Of the patients identified, 71 (12.1%) were excluded due to CAP being a secondary discharge diagnosis or because patients were determined to have exclusionary criteria upon chart review. The medical records of 288 patients were reviewed after a period of traditional education; 95 patient records were reviewed after academic detailing, and 133 patient records were reviewed after implementation of prewritten CAP orders. Medical records were reviewed 200 JCOM April 2005 Vol. 12,. 4

3 UMass Memorial Community-Acquired Pneumonia Community-acquired pneumonia (CAP) is a major cause of morbidity worldwide. In the US, it accounts for 10 million physician visits and more than 500,000 hospitalizations annually. Despite its high prevalence and its extensive study, many aspects of CAP management remain controversial. This guideline reflects newer information available to aid in the clinician s management of the patient with CAP. Whereas, the hospital management of such patients remains paramount, equal consideration is given to the decisions to both admit and discharge these patients. Oren Schaefer, MD References American Thoracic Society. Guidelines for the initial management of adults with community-acquired pneumonia: diagnosis, assessment of severity, initial antimicrobial therapy. Am J Respir Crit Care Med 2001;163:1730. Bartlett JG, et al. Practice guidelines for the management of community-acquired pneumonia in adults. Clin Infect Dis 2000;31:347. Fine MJ, et al. A prediction rule to identify low-risk patients with community-acquired pneumonia. N Engl J Med 1997;336:243. Fine MJ, et al. The hospital discharge decision for patients with community-acquired pneumonia. Results from the pneumonia patient outcomes research team cohort study. Arch Intern Med 1997;157:47. Halm EA, et al. Time to clinical stability in patients hospitalized with community-acquired pneumonia. Implications for practice guidelines. JAMA 1998;279: Heffelfinger JC, et al. Management of communityacquired pneumonia in the era of pneumococcal resistance. A report from the drug-resistant Streptococcuspneumonia therapeutic working group. Arch Intern Med 2000;160: This guideline is not intended to represent the sole acceptable or best practice for any patient. After assessing the individual features of a specific case, the patient s physician is in the best position to decide on the most appropriate therapeutic strategy. This guideline is not meant for use in the immunocompromised host. Antibiotic Guidelines for CAP Antibiotic choice will be dependent upon diagnosis Antibiotic choices are consistent with IDSA/ATS guidelines as well as being cost effective at UMMHC Rapid delivery of antibiotics is essential Risk Category I or II Primary: Doxycycline 100 mg PO bid Secondary: Azithromycin 500 mg PO then 250 mg/day Risk Category III Primary: Azithromycin 500 mg IV qd Secondary: Levofloxacin 500 mg IV qd Risk Category IV Primary: Ceftriaxone 1 g IV q 24 hr; plus Azithromycin 500 mg IV q 24 hr Secondary: Levofloxacin 500 mg IV q 24 hr Risk Category V Primary: Ceftriaxone 1 g IV q 24 hr; plus Azithromycin 500 mg IV q 24 hr Secondary: Levofloxacin 500 mg IV q 24 hr Severe CAP-ICU Admission Primary: Ceftriaxone 1 g IV q 24 hr; plus Azithromycin 500 mg IV q 24 hr Secondary: Levofloxacin 500 mg IV q 24 hr; plus Ceftriaxone 1 gm IV q 24 hr Ceftriaxone and azithromycin combination should be utilized first in an effort to preserve the spectrum of fluoroquinalone antibiotics. If a patient has an allergy to β-lactam antibiotics, failed initial therapy, or has a known drug-resistant organism, levofloxacin alone would serve as an alternative regimen in Class III IV CAP (but not severe CAP). Oral levofloxacin has excellent bioavailability and lung concentrations and is preferred if the patient is able to take PO. For severe CAP, consider anti-pseudomonal therapy if risk factors are present: immunosuppression, structural lung disease, chronic steroids, history of pseudomonal isolate. Duration of Therapy for Simple CAP Total duration of therapy: cephalosporin or levofloxacin for 10 days; azithromycin for 5 days Antibiotic delivery within 4 8 hours is essential Intravenous to Oral Step-Down Therapy Ceftriaxone alone to cefpodoxime 200 mg PO bid Ceftriaxone plus azithromycin to cefpodoxime 200 mg PO bid plus azithromycin 250 mg PO qd Levofloxacin IV to levofloxacin 500 mg PO q day Figure 1. Community-acquired pneumonia pocket guideline. Vol. 12,. 4 April 2005 JCOM 201

4 COMMUNITY-ACQUIRED PNEUMONIA Evaluation and Management of Community-Acquired Pneumonia?Pneumonia? Patient > 50 years Risk Class I & II 70 Consider empiric treatment as an outpatient Consult casemanagement as needed Risk Class III Risk Class IV (91 130) & V (> 130) 1. Consider sputum for GS, C&S 2. Blood culture prior to antibiotic delivery 3. Administer initial dose of antibiotics (see antibiotic guidelines) Does the patient have any of the following? Neoplastic disease Congestive heart failure Cerebrovascular disease Renal disease Liver disease Does the patient have any of the following abnormalities on physical examination? Altered mental status Pulse 125/min Respiratory rate 30/min Systolic BP < 90 mm Hg Temperature 35 C or 40 C Assign to Risk Class I further diagnostic testing necessary Consider empiric treatment as an outpatient* *Consider observational admission for hypoxemia, vomiting, poor social situation. All Class I, II patients admitted under observation 1. If condition not present/ not measured, score 0 points 2. Calculate patient total score 3. Assign pneumonia risk class Score Class Mortality ne I 0.1% 70 II 0.6% III 2.8% IV 9.3% > 130 V 27.0%0 Diagnostic testing to include: Initial blood cultures, CBC, electrolytes, BUN, creatinine, glucose, ABG for low Pox or for suspected acidosis COMPLETE PSI WORKSHEET Pneumonia Severity Index (PSI) Scoring Worksheet Demographic factors Age: Males Age (yrs) Females Age (yrs) 10 Admitted from + 10 nursing home Comorbid illness Neoplastic disease + 30 Liver disease + 20 CHF + 10 Cerebrovascular disease + 10 Renal disease + 10 Physical examination Respiratory rate + 20 ( 30/min) Systolic BP ( 90 mm Hg) + 20 Altered mental status + 20 Temperature (< 35 C + 15 or > 40 C) Pulse ( 125/min) + 10 Laboratory findings ph (< 7.35) + 30 BUN ( 30 mg/dl) + 20 Sodium ( 130 meq/l) + 20 Glucose ( 250 mg/dl) + 10 Hematocrit (< 30%) + 10 PO 2 (< 60 mm Hg) + 10 Pleural effusion + 10 PATIENT TOTAL SCORE Admit? 1. Consider 23-hour observation admission, or discharge with early (24 48 hr) follow-up 2. Consult case-management as needed Criteria for stability 1. Temperature 37.8 C 2. RR 24 bpm 3. SpO 2 92% 4. WBC at or approaching normal 5. Able to take PO 6. Mental status is at baseline Criteria for stability should be assessed every day Once stable, a patient should be switched to oral antibiotics Discharge should occur at time of switch There does not appear to be any advantage to observing a patient in-hospital for 24 hrs after the switch has occurred Patients with high-risk pneumonia such as S. aureus or P. aeruginosa or post obstructive pneumonia may require a larger duration of IV antibiotic. In such cases consider home IV therapy Consider minimizing lab draws There is no need for follow-up CXR if patient is clinically improved There is no need for follow-up CBC if patient is improved and WBC is returning towards normal Influenza vaccine or Pneumovax where appropriate Counsel on smoking cessation Admit to hospital Consider ICU Admission for Severe CAP Definite 1. Needs mechanical ventilation 2. Respiration failure: PaO 2 /FiO 2 ratio < 250 mm Hg 3. Vasopressors > 4 hrs Consider 1. RR > 30 bpm at admit 2. Multilobar involvement 3. SBP < 90 or DBP < 60 mm Hg 4. Urine output < 20 ml/hr, or acute renal failure requiring hemodialysis 5. Acute respiratory acidosis Case-management assessment and discharge planning within 24 hours Call attending MD for discharge order Criteria for stability met? If stability is not reached within 72 hrs, consider pulmonary or infectious disease consultation Figure 1. (cont d) 202 JCOM April 2005 Vol. 12,. 4

5 for the 3-month period that immediately followed each individual intervention. The Table shows findings for the 3 groups. Patient characteristics based on pneumonia severity score did not differ significantly between groups. There was no difference in mortality between groups. Guideline adherence was 29%, 24%, and 38% for the 3 groups, respectively. Formulary adherence was 75%, 71%, and 77%, respectively (Table). Different responses to implementation techniques were noted at the practice sites (Figure 2), with 210 patients being treated at the university site and 309 patients being treated at the community teaching hospital site. Overall mortality and length of stay across the 3 intervention periods at the university site was 2.5% and 5.7 days, while at the community site it was 1.9% and 4.2 days. Mortality and length of stay were compared between intervention groups using Fisher exact test and no statistical differences regardless of practice site were seen. Discussion We used 3 methods of guideline implementation across a 2-campus health care system, with each campus having somewhat different missions, and assessed physician adherence to guidelines. Overall, guideline adherence was subobtimal for all interventions at both sites, with academic detailing associated with the lowest adherence (24%). Preprinted order forms were associated with the highest adherence (38%). Formulary adherence alone was higher overall as compared with guideline adherence (> 70% for all interventions), with formulary adherence reaching 87% in the community setting after the implementation of prewritten orders. In comparing the 2 campuses, implementation of the guideline through traditional medical education methods was associated with greater guideline adherence in the university hospital, where the majority of patient orders are written by resident physicians. Academic detailing appeared to have an impact on formulary adherence in the community teaching hospital setting but not in the academic setting. Implementation of a preprinted order form had a greater impact on formulary adherence in the community setting. There are several limitations to our study. First, patients were identified by ICD-9 codes uploaded into the primary care health information system from the hospital billing system. It is possible that patients had pneumonia listed as a secondary diagnosis and were thus excluded from our database even though pneumonia may have been the primary reason for hospitalization, or there could have been errors in coding. Second, the number of charts reviewed after each intervention had considerable variability. This was a function of the number of patients with CAP who were evaluated at the hospital during those time periods. Third, since each of the interventions was applied to the same group of Table. Patient Group Findings Traditional Academic Preprinted Education Detailing Orders Risk class 2.8 ± ± ± 1.1 Evaluated in ED, % Admitted from ED, % LOS 4.4 ± ± ± 3.5 Guideline adherence, % Formulary adherence, % ED = emergency department; LOS = length of stay. physicians, each built upon the success of the preceding intervention. While relative changes in guideline adherence can be commented upon, the final impact of a single intervention cannot be fully appreciated due to influence from preceding interventions. Lastly, while there were different core groups of physicians at each practice site, there was a small number of physicians who practiced across both sites. This could contribute to some of the intercampus difference. Interestingly, attending physicians who were adherent to guidelines when working with residents in an academic setting were often found to be noncompliant with guideline use when caring for patients without residents in a community setting. Dissemination of literature-based practice guidelines through opinion leaders has been demonstrated to have a sustaining effect for at least 2 years [20], and academic detailing also has been shown to be effective [21 24]. Administrative interventions, such as diagnosis-specific order forms to reflect the preferred formulary and dosing intervals for antibiotics or eliminating certain diagnostic tests from order forms have also been successful [8,25]. This study supports previous studies demonstrating that a preprinted order form can improve utilization of appropriate antibiotic therapy and that academic detailing can have a positive impact in changing physician behavior. A traditional educational approach was less successful in the community teaching hospital. This approach may be more successful in an academic teaching setting, where education is focused on residents in training, rather than in a community teaching hospital, where the educational efforts are aimed at changing attending physician behaviors. A variety of techniques can be used to facilitate implementation of clinical guidelines. Success may be dependent on the type of practice environment and use of a variety of implementation techniques. Implementation techniques may have to vary to achieve optimal guideline utilization. Vol. 12,. 4 April 2005 JCOM 203

6 COMMUNITY-ACQUIRED PNEUMONIA 100 Guideline adherence Formulary adherence % 40 Figure 2. Response to guideline implementation efforts. CH = community hospital; UH = university hospital UH CH UH CH UH CH Traditional education Dec 01 Feb 02 Academic detailing Mar 02 May 02 Preprinted orders Oct 02 Dec 02 Corresponding author: Gregory A. Volturo, MD, Univ. of Mass. Medical School, 55 Lake Ave. N., Worcester, MA 01655, volturog@ummhc.org. References 1. Adams PF, Marano MA. Current estimates from the National Health Interview Survey, Vital Health Stat 1995;10: Garibaldi RA. Epidemiology of community-acquired respiratory tract infections in adults. Incidence, etiology, and impact. Am J Med 1985;78: Niederman MS, Bass JB Jr, Campbell GD, et al. Guidelines for the initial management of adults with communityacquired pneumonia: diagnosis, assessment of severity, and initial antimicrobial therapy. American Thoracic Society. Medical Section of the American Lung Association. Am Rev Respir Dis 1993;148: Guidelines for the management of community-acquired pneumonia in adults admitted to hospital. British Thoracic Society. Br J Hosp Med 1993;49: Mandell L, Niederman M, Marrie TJ. Initial antimicrobial treatment of community acquired pneumonia in adults: a conference report, Canadian Community Acquired Pneumonia Consensus Conference Group. Can J Infect Dis 1993;4: Bartlett JG, Breiman RF, Mandell LA, File TM Jr. Communityacquired pneumonia in adults: guidelines for management. The Infectious Diseases Society of America. Clin Infect Dis 1998;26: Mandell LA, Bartlett JG, Dowell SF, et al. Update of practice guidelines for the management of community-acquired pneumonia in immunocompetent adults. Clin Infect Dis 2003;37: Avorn J, Soumerai SB, Taylor W, et al. Reduction of incorrect antibiotic dosing through a structured educational order form. Arch Intern Med 1988;148: Gleason PP, Meehan TP, Fine JM, et al. Associations between initial antimicrobial therapy and medical outcomes for hospitalized elderly patients with pneumonia. Arch Intern Med 1999;159: Battleman DS, Callahan M, Thaler HT. Rapid antibiotic delivery and appropriate antibiotic selection reduce length of hospital stay of patients with community-acquired pneumonia: link between quality of care and resource utilization. Arch Intern Med 2002;162: Menendez R, Ferrando D, Valles JM, Vallterra J. Influence of deviation from guidelines on the outcome of communityacquired pneumonia. Chest 2002;122: Dean NC, Silver MP, Bateman KA, et al. Decrease mortality after implementation of a treatment guideline for community acquired pneumonia. Am J Med 2001;110: Davis DA, Thomson MA, Oxman AD, Haynes RB. Evidence for the effectiveness of CME. A review of 50 randomized controlled trials. JAMA 1992;268: Lomas J, Anderson GM, Domnick-Pierre K, et al. Do practice guidelines guide practice? The effect of a consensus statement on the practice of physicians. N Engl J Med 1989; 321: Gleicher N. Cesarean section rates in the United States. The short-term failure of the National Consensus Development Conference in JAMA 1984;252: Greco PJ, Eisenberg JM. Changing physicians practices. N Engl J Med 1993;329: Mittman BS, Tonesk X, Jacobson PD. Implementing clinical practice guidelines: social influence strategies and practitioner behavior change. QRB Qual Rev Bull 1992;18: Tunis SR, Hayward RS, Wilson MC, et al. Internists attitudes about clinical practice guidelines. Ann Intern Med 1994;120: Fine MJ, et al. A prediction rule to identify low-risk patients with community-acquired pneumonia. N Engl J Med 1997; 336: JCOM April 2005 Vol. 12,. 4

7 20. Everitt DE, Soumerai SB, Avorn J, et al. Changing surgical antimicrobial prophylaxis practices through education targeted at senior department leaders. Infect Control Hosp Epidemiol 1990;11: Avorn J, Soumerai SB. A new approach to reducing suboptimal drug use [editorial]. JAMA 1983;250: Schaffner W, Ray WA, Federspiel CF, Miller WO. Improving antibiotic prescribing in office practice. A controlled trial of three educational methods. JAMA 1983;250: Ray WA, Blazer DG 2nd, Schaffner W, et al. Reducing longterm diazepam prescribing in office practice. A controlled trial of educational visits. JAMA 1986;256: Ray WA, Schaffner W, Federspiel CF. Persistence of improvement in antibiotic prescribing in office practice. JAMA 1985; 253: Zaat JO, van Eijk JT, Bonte HA. Laboratory test form design influences test ordering by general practitioners in The Netherlands. Med Care 1992;30: Copyright 2005 by Turner White Communications Inc., Wayne, PA. All rights reserved. Vol. 12,. 4 April 2005 JCOM 205