Respiratory Medicine (2008) 102, 1257e1263
available at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/rmed
Efficacy of clinical guideline implementation to improve the appropriateness of chest physiotherapy prescription among inpatients with community-acquired pneumonia Idris Guessous a, Jacques Cornuz a,b, Rebecca Stoianov a, Bernard Burnand c, Jean-William Fitting d, Bertrand Yersin e, Olivier Lamy a,* a
Department of Internal Medicine, University Hospital Centre, CH-1011 Lausanne, Switzerland Department of Community Medicine and Public Health, University Hospital Centre, CH-1011 Lausanne, Switzerland c Institute of Social and Preventive Medicine (IUMSP), University Hospital Centre, CH-1011 Lausanne, Switzerland d Department of Pneumology, University Hospital Centre, CH-1011 Lausanne, Switzerland e Emergency Department, University Hospital Centre, CH-1011 Lausanne, Switzerland b
Received 10 January 2008; accepted 2 April 2008 Available online 10 July 2008
KEYWORDS Community-acquired pneumonia; Chest physiotherapy; Guideline; Implementation
Summary Background: Although there is no strong evidence of benefit, chest physiotherapy (CP) seems to be commonly used in simple pneumonia. CP requires equipment and frequently involves the assistance of a respiratory therapist, engendering a significant medical workload and cost. Aim: To measure and compare the efficacy of two modalities of chest physiotherapy (CP) guideline implementation on the appropriateness of CP prescription among patients hospitalised for community-acquired pneumonia (CAP). Patients and methods: We measured the CP prescription rate and duration in all consecutive CAP inpatients admitted in a division of general internal medicine at an urban teaching community hospital during three consecutive one-year time periods: (1) before any guideline implementation; (2) after a passive implementation by medical grand rounds and guideline diffusion through mailing; (3) after adding a one-page reminder in the CAP patient’s medical chart highlighting our recommendations. Death and recurrent hospitalisation rates within one year after hospitalisation were recorded to assess whether CP prescription reduction, if any, impaired patient outcomes. Results: During the three successive phases, 127, 157, and 147 patients with similar characteristics were included. Among all CAP inpatients, the CP prescription rate decreased from 68%
Abbreviations: CP, chest physiotherapy; COPD, chronic obstructive pulmonary disease; PSI, pneumonia severity index; CAP, communityacquired pneumonia. * Corresponding author. Tel.: þ41 21 314 08 76; fax: þ41 21 314 08 71. E-mail address: firstname.lastname@example.org (O. Lamy). 0954-6111/$ - see front matter ª 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.rmed.2008.04.008
I. Guessous et al. (86/127) to 51% (80/157), and to 48% (71/147), respectively (P for trend <0.01 for trend). A significant reduction in CP duration was observed after the active guideline implementation (12.0, 11.0, 7.0 days, respectively) and persisted after adjustment for length of stay. Reductions in CP prescription rate and duration were also observed among CAP patients with COPD CP prescription rate: 97% (30/31), 67% (24/36), 75% (35/47), respectively (P < 0.01 for trend). The mean cost of CP per patient was reduced by 56%, from $709 to $ 481, and to $309, respectively. Neither the in-hospital deaths, the one-year overall recurrent hospitalisation nor the one-year CAP-specific recurrent hospitalisation significantly differed between the three phases. Conclusion: Both passive and active implementation of guidelines appear to improve the appropriateness of CP prescription among inpatients with CAP without impairing their outcomes. Restricting CP use to patients who benefit from this treatment might be an opportunity to decrease CAP medical cost and workload. ª 2008 Elsevier Ltd. All rights reserved.
Introduction Community-acquired pneumonia (CAP) is a frequent disease often leading to hospitalisation. In community studies, the annual incidence of CAP ranges from 4 to 12 per 1000 adults per year and increases with age.1e3 It is estimated that more than 5 million cases of CAP occur annually in the United States.4 Hospitalisation is required for 15e42% of patients with CAP and accounts for about 90% of the total annual cost of CAP,5,6 which is estimated at $8.4 billion in the United States.4 Therefore, interventions likely to reduce the medical cost of CAP have been proposed.7 The rationale of certain CAP therapies has been questioned. One of them concerns the prescription of chest physiotherapy (CP). Although new techniques for airway clearance have emerged these last decades, the rare evidence for their benefit is weak.8 Reviews and metaanalyses on the efficacy of CP mainly concern patients suffering from cystic fibrosis.9e14 There is no strong evidence to support the use of CP in simple (uncomplicated) pneumonia (i.e., pneumonia without pleural effusion, empyema, lung abscess, necrotising pneumonias, pneumothorax, or ICU status).15,16 Moreover, there is very little data regarding the proportion of CAP inpatients receiving CP prescription. Among the existing data, CP seems clearly overused.15,17,18 We aimed to measure and compare the efficacy of two modalities of CP guideline implementation on the appropriateness of CP prescription.
Patients and methods Study protocol
applied for each recommendation. The guidelines were published in the official journal of the Swiss Society of Internal Medicine.19 The guidelines referred to the management of adult CAP of all ages in the community or in hospital. They contained recommendations on CAP diagnosis (e.g., general and microbiological investigations), severity assessment (e.g., prognostic score), management (e.g., antibiotic choice), and prevention (e.g., pneumococcal vaccination). The panel determined that patients with conditions leading to mucostasisddue to hypersecretion of mucus or to decreased ability to clear secretionsdwere at increased risk of recurrent bronchial infections. Accordingly, the guidelines recommended: (1) to reserve CP prescription for patients with severe pulmonary pathology (COPD), neuromuscular pathology, and important bronchial secretions and (2) to revaluate the prescription every 3 days (Fig. 1).
Chest physiotherapy prescription rate and duration CP prescription rate and duration were measured in all consecutive CAP inpatients admitted during three consecutive one-year time periods: (1) before any guideline implementation (phase I, i.e., baseline); (2) after a passive implementation by medical grand rounds and guideline diffusion through mailing (phase II: passive implementation);
GUIDELINE INPATIENTS WITH CAP Chest physiotherapy prescription recommendations: 1) Indications
An evidence-based guideline for the management of inpatients with CAP was developed at our institution. This clinical guideline was developed by a panel including a hospital practitioner, infectiologist, pneumologist, clinical microbiologist, and clinical epidemiologist working at our institution and was reviewed, commented on and approved by three external experts. Experts in charge of each topic have searched and critically appraised published guidelines and studies identified in a literature search. A hierarchical evaluation of the strength of evidence was
• Severe pulmonary pathology (COPD) • Neuromuscular pathology • Important bronchial secretions 2) Duration • Revaluate the prescription every 3 days
Appropriateness of chest physiotherapy prescription
(3) after adding, on the day of patient admission, a one-page reminder in the CAP patientâ€™s medical chart highlighting our recommendations (phase III: active implementation) (Fig. 2). The CAP guideline was developed at the end of phase I so that it would not affect the clinical management of patients included in phase I. CP sessions in CAP varied from case to case, but generally consisted of positioning the patient to optimise ventilation and perfusion, external vibration, and incentive spirometry. The total cost of CP for each phase was calculated based on a cost estimation of $90.90 per day of CP, which included one intensive session and one soft CP session. In addition, in-hospital mortality and recurrent hospitalisation rates within one year after hospitalisation were recorded to assess whether CP prescription reduction, if any, impaired patient outcomes.
were collected by chart review. Severity of communityacquired pneumonia was assessed by the pneumonia severity index (PSI) proposed by Fine et al.20 The PSI is based on derivation and validation cohorts of more than 50,000 patients with CAP. Briefly, the PSI stratifies patients into 5 mortality risk classes, and its ability to predict mortality has been confirmed in multiple subsequent studies. On the basis of associated mortality rates, it has been suggested that risk class I and II patients should be treated as outpatients, risk class III patients should be treated in an observation unit or with a short hospitalisation, and risk class IV and V patients should be treated as inpatients. A diagnosis of COPD was retained on the basis of ICD-9-CM diagnostic discharge codes. Medical records of all included patients were manually and electronically reviewed using hospital databases to assess the number, date and cause of re-admissions.
Patients and data Statistical analysis This study was conducted from 1997 to 2001 in the Service of Internal Medicine at the University Hospital Centre in Lausanne, Switzerland, a 900-bed teaching hospital serving as a city hospital and tertiary reference centre providing services to an area with more than 300,000 inhabitants. All consecutive adult patients (age 18 years) hospitalised with a primary diagnosis of CAP were eligible for the study. Records of phases I and II (retrospective data) were identified through a sensitive review of ICD-9-CM diagnostic discharge codes including pneumonia with or without identified pathogens. Records of phase III (prospective data) were identified daily in the emergency room by a research study nurse and by the fellow in charge of the study. All adult patients hospitalised with a primary diagnosis of CAP were eligible for the study. The inclusion criteria were the following: age 18 years; acute onset of at least one clinical finding suggestive of pneumonia, including cough, dyspnoea, pleuritic chest pain, sputum production, fever or altered mental status; and the presence of a newly-acquired radiographic pulmonary infiltrate confirmed by a radiologist. Patients with immunodeficiency virus infection, neutropenia related to chemotherapy, organ transplant, cystic fibrosis or congestive heart failure were excluded, as were patients who were transferred to the ICU, admitted from a nursing home, hospitalised within the 10 previous days or discharged home during the first 24 h of hospitalisation. For patients hospitalised with CAP on more than one occasion during the study period, only the initial episode was evaluated. Data on demographic characteristics, comorbid illnesses, physical examination findings, initial investigations, antibiotherapy, ancillary treatments, follow-up and outcome
Baseline Phase I No guideline
Results Patients During the three successive phases, 127, 157 and 147 patients were included, respectively. Patients in the three groups were similar in age, gender, severe CAP according to the pneumonia severity index, and COPD diagnosis. They slightly but significantly differed in admission PaO2, with phase I patients being more frequently hypoxemic (Table 1).
Effect of intervention The number of CP prescriptions was significantly lower during phases II and III compared to phase I (Fig. 3). Among all CAP inpatients, the CP prescription rate decreased from 68% (86/127), to 51% (80/157) and to 48% (71/147), in phase I, II and III, respectively (P for trend <0.01). The rate of CP prescription also significantly decreased among CAP
Guideline diffusion through medical grand rounds/ internet
Descriptive and comparative statistical tests (paired and non-paired t-test, Chi square test or exact Fisher test), and P for linear trend to examine trends across phases were applied using Stata (V 9.1, Stata Corporation, College Station, TX, USA). P values <0.05 were considered significant. The inflation effect on hospital charges were adjusted using the Bureau of Labor Statisticâ€™s Consumer Price Index (http://www.bls.gov/cpi/). Data are given as percentage or mean standard deviation (SD).
Phase II Passive implementation
One-page reminder inpatient medical chart 36 months
Phase III Passive + Active implementation
Phases of implementation.
1260 Table 1
I. Guessous et al. Patient characteristics at time of hospital admissiona
Phase I: baseline (n Z 127)
Phase II: passive implementation (n Z 157)
Phase III: passive Ăž active implementation (n Z 147)
P value for trend
Age, yearsb Male COPD PSI IV/V Respiratory rate >30/min PaO2 <60 mmHg
71 (16) 58 24 61 16 43
71 (17) 59 23 64 17 28
71 (18) 61 31 59 14 34
0.9 0.8 0.2 0.7 0.7 0.04
COPD, chronic obstructive pulmonary disease; PSI, pneumonia severity index. a Data are given as percentage of each group unless otherwise indicated. b Data are given as mean (standard deviation).
inpatients without COPD from 58% (59/96), to 46% (56/121), and to 36% (36/100) (P for trend <0.01). The rate of prescription also decreased for CAP inpatients with COPD from 97% (30/31) in phase I to 67% (24/36) in phase II and 75% (35/47) in phase III (P for trend <0.01). Fig. 4 shows that the reduction in CP duration was particularly impressive after active guideline implementation, decreasing from 12.0 (SD 7.3), to 11.0 (SD 5.6) and to 7.0 (SD 3.9) days, respectively (P for trend <0.01). The duration of CP prescription also significantly decreased after the active guideline implementation among CAP inpatients without COPD [11.2 (SD 7.2), 10.1 (SD 7.7), 6.9 (SD 4.5), (P for trend <0.01)] and CAP inpatients with COPD [12.2 (SD 6.7), 11.2 (SD 6.8), 7.2 (SD 4.3), (P for trend <0.01)]. As the length of hospital stay decreased during the three phases (median length of stay: 9.0, 8.0, and 7.5 days, respectively), a correlation between the CPâ€™s length of prescription and the length of CAP inpatient hospital stay was measured. These correlations decreased considerably after the two modalities of guideline implementations, from a strong correlation (r2 Z 0.78 in phase I) to a weak correlation (r2 Z 0.55 in phase II and r2 Z 0.24 in phase III) after active guideline implementation. Moreover, the CP prescription, when judged necessary, was significantly limited to the beginning of the hospital stay. To better evaluate the mean number of days of CP prescription per patient, we calculated the index of prescription (total
day of prescription divided by total number of patients). The index of prescription significantly decreased between the three phases with greater reduction in phase III. Indeed, the index of prescription decreased from 7.8 to 5.3, and to 3.4 in phases I, II and III, respectively (P for trend <0.01). The total cost of CP decreased from $90,101 (2001 inflation-adjusted $97,895) to $75,654 (2001 inflation-adjusted $80.422) and to $45,460 (2001 inflation-adjusted $46,753) in phases I, II and III, respectively (Fig. 5). The mean cost of CP per one patient hospitalised in a general ward for CAP was reduced by 56%, from $709 (2001 inflation-adjusted $770) to $481 (2001 inflation-adjusted $512), and to $309 (2001 inflation-adjusted $318) during phases I, II and III, respectively.
Secondary outcomes In-hospital death rates among all patients remained low and were similar between the three phases: 4.6%, 6.4%, 5.4%, respectively (P for trend 0.77). Neither the one-year overall recurrent hospitalisation rate (27%, 22%, 28%, P Z 0.65 for trend) nor the one-year CAP-specific recurrent hospitalisation rate (25%, 37% and 36%, P for trend 0.41) significantly differed between the three phases. Additionally, the 3-month, 6-month and 9-month rates of recurrent hospitalisations were similar between the three phases (Table 2).
Figure 3 Chest physiotherapy prescription rate by phases. Phase I (n Z 127), phase II (n Z 157) and phase III (n Z 147). P for trend <.01
Appropriateness of chest physiotherapy prescription
Figure 4 Duration of chest physiotherapy by phases. Phase I (n Z 127), phase II (n Z 157) and phase III (n Z 147). P for trend <.01
Discussion Our study demonstrated that both passive and active implementations of the guideline appeared to improve the appropriateness of CP prescription among inpatients with CAP. Improvement was more pronounced during the active implementation phase with a greater impact on chest physiotherapy duration. Moreover, the reduction in CP prescription and duration seemed disconnected with worse medical outcomes. The total cost of CP was reduced by nearly 50% after the active implementation. In recent years, strategies aimed at reducing the medical workload and cost of patients hospitalised for CAP have been proposed. Since a length of stay reduction has been generally unassociated with an increase in adverse outcomes,21 efforts have focused on interventions inclined to reduce the length of stay for patients with CAP. Therefore, processes including triage decisions, risk stratification for outpatient care and early transition from intravenous to oral antibiotics have been evaluated.19,22e24 in addition to the length of hospital stay, CAP inpatient management includes practices not supported by evidence of effectiveness. Indications regarding CP prescription are clearly lacking in CAP guidelines and review.25e27 The British Thoracic
Figure 5 Index of prescription and cost of chest physiotherapy by phases. Phase I (n Z 127), phase II (n Z 157) and phase III (n Z 147).
Society Guideline and European Study on Community Acquired Pneumonia (ESOCPA) committee are among the few that assessed recommendations regarding CP.28,29 CP in acute pneumonia obtained a grade III recommendation (i.e., one or more retrospective clinical studies that address but do not rigorously answer the question) in the British Thoracic Society Guideline, and the ESOCAP committee stressed that the body of data regarding CP is small but that CP seemed justified for patients with chronic airway disease. Despite the lack of evidence, CP seems to be overused. In 1996, Alexander et al. demonstrated that more than 50% of CP prescriptions were inappropriate.15 This high level of inappropriateness was previously observed.17,30,31 More recently, we reported that 59% of inpatients with CAP received CP although they had no accepted indications.18 Not only has CP been overprescribed, CP prescription duration appeared to be too long, with duration quite similar to the length of hospital stay.18 One explanation might be that physicians do not regularly reassess the rationale for pursuing CP. CP requires equipment and frequently involves the assistance of at least one respiratory therapist, engendering a significant medical workload and cost. When including the personnel and equipment costs, the annual cost of CP delivered in a medical centre has been estimated to be more than $1 million.15 Strategies to reduce the inappropriate utilisation of CP have been evaluated through randomised and prospective observational studies. One of them showed that the number of inappropriate CP prescriptions could be significantly reduced without impairing the patient’s outcome. It was also estimated that the reduction would be associated with annual cost savings of $319,000.15 More recently, three additional studies have investigated different strategies to improve respiratory care utilisation.16,32,33 All compared the impact of respiratory therapist-initiated treatment to physician-directed orders on CP prescription rate. These three studies suggested that the instances of inappropriate respiratory care significantly decrease with respiratory therapist-initiated treatment. Our study has some limitations. First, we have not clearly defined the three CAP indications, i.e., a ‘‘severe
1262 Table 2
I. Guessous et al. Rate and cause of recurrent hospitalisationa
Number of months after hospitalisation
0e3 4e6 7e9 10e12 Overall
Phase I: baseline (n Z 127)
Phase II: passive implementation (n Z 157)
Phase III: passive þ active implementation (n Z 147)
4 2 1 2 9
13 4 4 5 26
8 1 4 0 13
9 5 6 2 22
7 2 3 3 15
6 8 10 3 27
(3.1) (1.5) (0.7) (1.5) (7.0)
(10.2) (3.1) (3.1) (3.9) (20.0)
(5.0) (0.6) (2.5) (0.0) (8.2)
(5.7) (3.1) (3.8) (1.2) (14.0)
(4.7) (1.3) (2.0) (2.0) (10.2)
(4.0) (5.4) (6.8) (2.0) (18.3)
P value for trend
0.1 0.7 0.6 0.4 0.6
CAP, community-acquired pneumonia. a Data are given as number (percentage).
COPD,’’ ‘‘neuromuscular pathology,’’ or ‘‘important bronchial secretions,’’ which might limit the generalisation of our findings. Second, the ‘‘before and after study’’ design might have introduced bias in the following manner. Differences in confounding factors between the three groups might partially explain differences in outcome for reasons unrelated to the CAP guidelines. For example, the lower proportion of hypoxemic patients at admission could have resulted in a decrease in CP prescriptions in groups with less hypoxemic patients. Third, because the physicians who worked in our institution during phases II and III might have been informed of the study aims, they might have changed their practice independently of the guideline implementation (Hawthorne effect). Finally, our guideline reported weak evidence for prescribing CP in CAP. Although guidelines are generally understood as a compression of strong indications aimed to correct the inappropriate underuse of an intervention, we stress that the clinical utility of guidelines should not be limited to this aim. Indeed, the inappropriate overuse of medical interventions has also stimulated the development of recommendations, and guidelines with a limited amount or lack of evidence, appeared to also help doctors in making the right decisions, i.e., not to prescribe.34e39 Although we cannot report the exact mechanism by which providers did eventually decide not to prescribe CP and acknowledge that further work should be done to assess the impact of the level of evidence (i.e., weak) on provider decision, we demonstrated the initial step, that is the reduction of unnecessary CP prescription after guideline implementation. We do think that our findings might add new elements to the experiences assessed from previous studies. First, the method we used to implement our guideline e a one-page reminder in CAP patients’ medical charts e might be easier to reproduce (external validity) and less expensive than strategies previously proposed. For example, Alexander et al. chose to put a physician fellow in charge of explaining by phone to each physician whether CP is indicated for their patients.15 This strategy seems hardly reproducible, as it requires an additional physician, and probably inefficient because of its potential cost. Second, previous studies chose a respiratory therapist-initiated treatment approach (i.e., respiratory therapists were first specifically trained to use algorithms and then ordered or not CP to inpatients).16,32,33 Although these studies showed significant reduction in respiratory care services, their conclusions might be difficult to generalise in other settings with
limited training resources and respiratory therapists. Asking respiratory therapists to evaluate each inpatient and decide whether they should receive CP could further increase their workload and limit their availability for other patients who may require urgent respiratory care. By using medical rounds or a one-page reminder in a patient’s medical chart, our strategy was a physician-based approach that might not only be more reproducible but also better fulfil the final aims of guideline implementation, which are to change and improve physicians’ clinical practice. Finally, our study highlighted the need to further evaluate whether CP is actually useful for inpatients with CAP. The beforeeafter study is the most commonly encountered design in quality improvement research;40 however, to obtain valid assessment of the efficacy of CP in CAP, more robust designs (e.g., randomised controlled trials) are needed. In summary, although it does not use a randomised design, this study indicates that guidelines for the use of CP in patients with CAP might enable more appropriate and cost-saving care by restricting its use and limiting it for absolutely necessary occasions. The external validity of this simple intervention should be evaluated in other settings and eventually at different levels of CAP medical care.
Conflict of interest statement The authors hereby declare that they have no financial or personal interests.
Acknowledgements The Department of Internal Medicine, University Hospital Centre in Lausanne funded the study.
References 1. Bartlett JG, Mundy LM. Community-acquired pneumonia. N Engl J Med 1995;333:1618e24. 2. Marrie TJ. Community-acquired pneumonia. Clin Infect Dis 1994;18:501e13. 3. Jokinen C, Heiskanen L, Juvonen H, et al. Incidence of community-acquired pneumonia in the population of four municipalities in eastern Finland. Am J Epidemiol 1993;137:977e88. 4. Niederman MS. Community-acquired pneumonia: management controversies, part 1; practical recommendations from the latest guidelines. J Respir Dis 2002;23:10e7.
Appropriateness of chest physiotherapy prescription 5. Guest JF, Morris A. Community-acquired pneumonia: the annual cost to the National Health Service in the UK. Eur Respir J 1997;10:1530e4. 6. Niederman MS, McCombs JS, Unger AN, Kumar A, Popovian R. The cost of treating community-acquired pneumonia. Clin Ther 1998;20:820e37. 7. Rhew DC, Tu GS, Ofman J, Henning JM, Richards MS, Weingarten SR. Early switch and early discharge strategies in patients with community-acquired pneumonia: a metaanalysis. Arch Intern Med 2001;161:722e7. 8. Pryor JA. Physiotherapy for airway clearance in adults. Eur Respir J 1999;14:1418e24. 9. Williams MT. Chest physiotherapy and cystic fibrosis. Why is the most effective form of treatment still unclear? Chest 1994; 106:1872e82. 10. Thomas J, Cook DJ, Brooks D. Chest physical therapy management of patients with cystic fibrosis. A meta-analysis. Am J Respir Crit Care Med 1995;151:846e50. 11. Butler SG, Sutherland RJ. Current airway clearance techniques. N Z Med J 1998;111:183e6. 12. Prasad SA, Main E. Finding evidence to support airway clearance techniques in cystic fibrosis. Disabil Rehabil 1998; 20:235e46. 13. Main E, Prasad A, Schans C. Conventional chest physiotherapy compared to other airway clearance techniques for cystic fibrosis. Cochrane Database Syst Rev; 2005:CD002011. 14. van der Schans C, Prasad A, Main E. Chest physiotherapy compared to no chest physiotherapy for cystic fibrosis. Cochrane Database Syst Rev; 2000:CD001401. 15. Alexander E, Weingarten S, Mohsenifar Z. Clinical strategies to reduce utilization of chest physiotherapy without compromising patient care. Chest 1996;110:430e2. 16. Kollef MH, Shapiro SD, Clinkscale D, et al. The effect of respiratory therapist-initiated treatment protocols on patient outcomes and resource utilization. Chest 2000;117:467e75. 17. Kester L, Stoller JK. Ordering respiratory care services for hospitalized patients: practices of overuse and underuse. Cleve Clin J Med 1992;59:581e5. 18. Lamy O, Van Melle G, Cornuz J, Burnand B. Clinical management of immunocompetent hospitalized patients with community-acquired pneumonia. Eur J Intern Med 2004;15:28e34. 19. Lamy O, Zanetti G, Bille J, Aubert JD, Cornuz J, Burnand B. Diagnostic et traitement de la pneumonie acquise a ` domicile de l’adulte. Recommandations pour la pratique clinique. Rev Med Suisse 1999;119:403e27. 20. Fine MJ, Auble TE, Yealy DM, et al. A prediction rule to identify low-risk patients with community-acquired pneumonia. N Engl J Med 1997;336:243e50. 21. McCormick D, Fine MJ, Coley CM, et al. Variation in length of hospital stay in patients with community-acquired pneumonia: are shorter stays associated with worse medical outcomes? Am J Med 1999;107:5e12. 22. Ramirez JA, Srinath L, Ahkee S, Huang A, Raff MJ. Early switch from intravenous to oral cephalosporins in the treatment of hospitalized patients with community-acquired pneumonia. Arch Intern Med 1995;155:1273e6. 23. Marrie TJ, Lau CY, Wheeler SL, Wong CJ, Vandervoort MK, Feagan BG. A controlled trial of a critical pathway for treatment of community-acquired pneumonia. CAPITAL Study
Investigators. Community-Acquired Pneumonia Intervention Trial Assessing Levofloxacin. JAMA 2000;283:749e55. Fine MJ, Stone RA, Lave JR, et al. Implementation of an evidence-based guideline to reduce duration of intravenous antibiotic therapy and length of stay for patients hospitalized with community-acquired pneumonia: a randomized controlled trial. Am J Med 2003;115:343e51. Niederman MS, Mandell LA, Anzueto A, et al. Guidelines for the management of adults with community-acquired pneumonia. Diagnosis, assessment of severity, antimicrobial therapy, and prevention. Am J Respir Crit Care Med 2001;163:1730e54. Mandell LA, Bartlett JG, Dowell SF, File Jr TM, Musher DM, Whitney C. Update of practice guidelines for the management of community-acquired pneumonia in immunocompetent adults. Clin Infect Dis 2003;37:1405e33. Woodhead M, Blasi F, Ewig S, et al. Guidelines for the management of adult lower respiratory tract infections. Eur Respir J 2005;26:1138e80. BTS guidelines for the management of community acquired pneumonia in adults. Thorax 2001;56(Suppl 4):IV1e64. ERS task force report. Guidelines for management of adult community-acquired lower respiratory tract infections. European Respiratory Society. Eur Respir J 1998;11:986e91. La Force FM. Community-acquired lower respiratory tract infections. Prevention and cost-control strategies. Am J Med 1985;78:52e7. Kinlay S, Olson LG. Patterns of use of chest physiotherapy in a teaching hospital. Aust Clin Rev 1991;11:154e8. Stoller JK, Skibinski CI, Giles DK, Kester EL, Haney DJ. Physician-ordered respiratory care vs physician-ordered use of a respiratory therapy consult service. Results of a prospective observational study. Chest 1996;110:422e9. Stoller JK, Mascha EJ, Kester L, Haney D. Randomized controlled trial of physician-directed versus respiratory therapy consult service-directed respiratory care to adult non-ICU inpatients. Am J Respir Crit Care Med 1998;158: 1068e75. Hess JR, Hiippala S. Optimizing the use of blood products in trauma care. Crit Care 2005;9(Suppl 5):S10e4. Mu ¨ller U, Exadaktylos A, Roeder C, et al. Effect of a flow chart on use of blood transfusions in primary total hip and knee replacement: prospective before and after study. BMJ 2004; 328:934e8. Wachtel TJ, O’Sullivan P. Practice guidelines to reduce testing in the hospital. J Gen Intern Med 1990;5:335e41. Verstappen WH, van der Weijden T, Sijbrandij J, et al. Effect of a practice-based strategy on test ordering performance of primary care physicians: a randomized trial. JAMA 2003;289: 2407e12. Wang TJ, Mort EA, Nordberg P, et al. A utilization management intervention to reduce unnecessary testing in the coronary care unit. Arch Intern Med 2002;162:1885e90. Weingarten SR, Riedinger MS, Conner L, et al. Practice guidelines and reminders to reduce duration of hospital stay for patients with chest pain. An interventional trial. Ann Intern Med 1994;120:257e63. Shojania KG, Grimshaw JM. Evidence-based quality improvement: the state of the science. Health Aff (Millwood) 2005; 24:138e50.