Evidence Based Review Article

The Journal of Informed Pharmacotherapy 2001;5:200-202.

Early Switch and Early Discharge Strategies for Community-Acquired Pneumonia: Can we Reduce Costs and Still Achieve the Desired Outcomes?

Download a PDF version of this article

Reviewer: Alfred Gin, Pharm.D.
Reviewer's email address:
Reviewer's profession/specialty:
Clinical Pharmacist/Infectious Diseases

Original Citation

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 Meta-analysis.  Arch Intern Med 2001 Mar 12;161:722-727.  PubMed Cit

Overall Study Question

To evaluate criteria used in studies of community-acquired pneumonia (CAP) for early switch therapy (i.e. parenteral to oral therapy conversion) and to assess the impact of early switch and early discharge on clinical outcomes and length of stay (LOS).  Several past studies have been published have suggested that early switch and early discharge may decrease the length of stay without adverse affecting outcomes.

Are the Results of the Study Valid?

1. Did the overview address a focused clinical question?

Yes.  The authors specifically focused on the impact of early switch therapy and discharge on clinical outcomes and LOS.

2. Were the criteria used to select articles for inclusion appropriate?

Yes.  The researchers searched several databases using several search terms for articles addressing early switch therapy antibiotic therapy and early discharge in CAP patients.  Early switch and early discharge strategy was defined as a method or intervention designed to shorten LOS through early switch and/or early discharge recommendations.  Articles were excluded if following existed less rigorous study design, specific patient population (e.g. unrelated to CAP), non-bacterial cause of CAP, non-clinical evaluation, unrelated study objective and absence of early switch criteria.

3. Is it unlikely that important, relevant studies were missed?

No.  The authors searched for English

language articles published between January 1, 1980 and March 31, 2000.  The researchers searched MEDLINE, HEALTHSTAR, EMBASE, Cochrane Collaboration and Best Evidence databases for prospective and retrospective trials, meta-analyses and systematic reviews.  The extent of search terms used would have captured the majority of relevant articles pertaining to the study question.

4. Was the validity of the included studies appraised?

The methodologies

of articles selected were not evaluated specifically. 

The authors evaluated articles to address four areas of investigation; criteria for switch therapy, criteria for discharge, switch on specific day/length of parenteral therapy and outcomes.

5. Were assessments of studies reproducible?

The authors applied criteria to reviewed articles to compare criteria for early switch and early discharge in CAP. 

Only articles that examined LOS are included in the outcomes evaluation. Chi-square and random effects meta-analysis was conducted on LOS for qualifying studies. 

Evaluation of clinical outcomes between intervention and control groups was not well described. 

Kappa values for agreement were calculated for the article reviewers.

6. Were the results similar from study to study?

Differences in the criteria and methodology used in early switch and early discharge studies assessed by the researchers resulted in variation in results with respect to the LOS and discharge.  This also made it difficult for the researchers to assess the outcome resulting from the early switch and/or discharge.  Although mean age among assessed studies appeared similar, the researchers suggested the patients included in the pooled analysis may not have been similar.  Two of the assessed studies enrolled "low-risk" patients with stays shorter than the interventional guidelines.  Other factors accounting for the variation in LOS included geographic variation and secular trends.

What are the Results? 

1. What are the overall results of the review?

The authors identified

1794 articles, of which 266 abstracts were reviewed. 

Only 10 prospective, interventional CAP studies involved an evaluation of


Criteria for switch were reported in 9 studies. The most common criteria for switch was afebrile (100%), improvement/resolution of signs and symptoms (89%)and able to take oral antibiotics (67%). 

Five studies applied criteria for early discharge criteria with care of comorbid condition the most common. 

There was disparity in other discharge criteria among the 5 studies.

The median recommended day of switch therapy was 3 days (range 2-10 days) in 3 studies. 

Five studies recommended switch therapy after a minimum number of parenteral days (3 days, range 2-10). Chi-square analysis revealed

significant variability (p<0.001) in the LOS among

st the studies.  Of the 10 prospective studies, 6 applied early switch and discharge criteria to an intervention and control group. The remaining 4 studies involved

no control group

s or used historical controls. 

Among the 6 evaluable studies, the LOS  was

quite variable (range, 3.5 to 11 days).  Meta-analysis was performed on 5 studies where standard deviation values for LOS were provided. 

The mean change in LOS between the intervention and control was not significantly reduced (p=0.05) based on a random effects meta-analysis (-1.64 days; 95% confidence interval

-3.30 to 0.02). 

Two studies had an interventional LOS equal or longer than the control. 

Exclusion of these studies increased the mean change of LOS between the intervention and control group to 3 days (-3.04 days; 95% confidence interval -4.90 to -1.19).

2.How precise were the results?

As above.

3. How much does allowance for uncertainty change the results? (i.e. Sensitivity analysis? Analysis of heterogeneity?)

Sensitivity analysis or analysis of heterogeneity was not done.  Characteristics of the 5 prospective interventional studies that may have confounded LOS analysis were presented.

Will the results help me in caring for my patients? 

1. Can the results be applied to my patient care?

Although criteria for switch therapy are generally similar, there are variations in the definitions used among some of the studies.  As noted above, there is considerable variability in the criteria for early discharge and LOS due to study design.  Overall, the comparison of criteria used for early switch therapy provides guidance to clinicians when evaluating a patient for oral antibiotic therapy.  The authors correctly indicate that the baseline LOS affects the impact (change in LOS) as a result of early switch or discharge interventions and must thus be considered if institutions were to implement similar programs.

2. Were all clinically important outcomes considered? 

Due to differences in the methodology of the studies assessed, clinical outcomes (e.g. complications or

therapeutic failures, mortality, readmission, quality of life, etc...) among the 10 prospective, interventional studies were not pooled and adequately examined.

3. Are the benefits worth the harms and costs?

Intuitively, the application of early switch and discharge criteria should decrease the LOS of stay. 

Early switch criteria such as the resolution of fever, clinical improvement and ability to ingest oral antibiotics are reasonable when evaluating antibiotic therapy in patients with CAP. 

Early switch therapy in hospitalized patients may yield cost savings

with respect to antimicrobial costs. The use of early switch and early discharge for CAP may decrease LOS without adversely affecting outcome. 

It should be noted that economic impact of early switch therapy and/or early discharge were not evaluated in this study.


With the increasing armamentarium of antibiotics directed specifically against CAP, there is perhaps increasing confusion with respect to the therapeutic differences among and within antibiotic classes.


differences in CAP studies with respect to discharge criteria, LOS and clinical outcomes make it difficult to assess the impact of a given agent or class. 

Several studies have clearly shown that many patients admitted to hospital with CAP at low risk of mortality may be treated adequately as an outpatient. 

To assist clinicians, prognostic scoring has been used to assess the risk of mortality among patients with CAP and thus reduce unnecessary hospital admissions. Recent Canadian data suggests that the LOS appear longer than studies conducted in the US. (1)

Hence approaches to decrease the cost of CAP-related hospitalization (early switch) and/or decrease the LOS (early discharge) may be of benefit. 

The findings of the study above confirm what clinicians have been intuitively using as a guide to convert patients from parenteral to oral therapy (resolution of temperature, clinical improvement and ability to take oral antibiotics). 

However, the clinical and economic impact of decreasing the LOS in Canada as a result of early switch and discharge interventions require critical evaluation through further research.


  1. 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. JAMA 2000;283:749-55.

Copyright © 2001 by the Journal of Informed Pharmacotherapy. All rights reserved.