This Article Abstract Full Text (PDF) Free Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Article Archive Download to citation manager Citing Articles Citing Articles via ISI Web of Science (3) Citing Articles via Google Scholar Google Scholar Articles by Labarere, J. Articles by Fine, M. J. Search for Related Content PubMed PubMed Citation Articles by Labarere, J. Articles by Fine, M. J.

Comparison of Outcomes for Low-Risk Outpatients and Inpatients With Pneumonia^*

A Propensity-Adjusted Analysis

^* From the Center for Health Equity Research and Promotion (Drs. Labarere, M.J. Fine, Stone, and Mr. Obrosky), Veterans Affairs Pittsburgh Healthcare System, Pittsburgh, PA; the Department of Emergency Medicine (Dr. Yealy), University of Pittsburgh, Pittsburgh, PA; the Department of Medicine (Dr. Meehan), Yale University School of Medicine, New Haven, CT; and Qualidigm (Drs. J.M. Fine and Graff), Middletown, CT.

Correspondence to: Michael J. Fine, MD, MSc, Veterans Affairs Center for Health Equity Research and Promotion, Veterans Affairs Pittsburgh Healthcare System, University Dr C, Building 28, 1A102, Pittsburgh, PA 15240; e-mail: Michael.Fine{at}med.va.gov

Abstract

Background: Low-risk patients with community-acquired pneumonia are often hospitalized despite guideline recommendations for outpatient treatment.

Methods: Using data from a randomized trial conducted in 32 emergency departments, we performed a propensity-adjusted analysis to compare 30-day mortality rates, time to the return to work and to usual activities, and patient satisfaction with care between 944 outpatients and 549 inpatients in pneumonia severity index risk classes I to III who did not have evidence of arterial oxygen desaturation, or medical or psychosocial contraindications to outpatient treatment.

Results: After adjusting for quintile of propensity score for outpatient treatment, which eliminated all significant differences for baseline characteristics, outpatients were more likely to return to work (odds ratio [OR], 2.0; 95% confidence interval [CI], 1.5 to 2.6) or, for nonworkers, to usual activities (OR, 1.4; 95% CI, 1.1 to 1.8) than were inpatients. Satisfaction with the site-of-treatment decision (OR, 1.1; 95% CI, 0.7 to 1.8), with emergency department care (OR, 1.4; 95% CI, 0.9 to 1.9), and with overall medical care (OR, 1.1; 95% CI, 0.8 to 1.6) was not different between outpatients and inpatients. The overall mortality rate was higher for inpatients than outpatients (2.6% vs 0.1%, respectively; p < 0.01); the mortality rate was not different among the 242 outpatients and 242 inpatients matched by their propensity score (0.4% vs 0.8%, respectively; p = 0.99).

Conclusions: After adjusting for the propensity of site of treatment, outpatient treatment was associated with a more rapid return to usual activities and to work, and with no increased risk of mortality. The higher observed mortality rate among all low-risk inpatients suggests that physician judgment is an important complement to objective risk stratification in the site-of-treatment decision for patients with pneumonia.

Key Words: ambulatory care • community-acquired infections • pneumonia • treatment outcomes

Community-acquired pneumonia (designated pneumonia) causes 4 million episodes of illness and 1 million hospital admissions in the United States each year.¹ An estimated $10 billion is expended annually on pneumonia patients, with inpatient treatment costing approximately 20 times as much as outpatient care.²³ Medical specialty society guidelines⁴⁵⁶ frequently recommend outpatient treatment for patients who are defined as low-risk based on the pneumonia severity index (PSI).

Three previous studies⁷⁸⁹ have demonstrated the effectiveness of using the PSI to reduce the number of hospitalizations of low-risk patients with pneumonia without compromising their safety. However, in these trials, 31 to 62% of low-risk patients were treated as inpatients, despite the implementation of interventions to increase treatment in the outpatient setting.⁷⁸⁹ In the emergency department community-acquired pneumonia (EDCAP) trial,⁷ nearly 40% of low-risk patients were hospitalized, and only one third of low-risk inpatients had medical or psychosocial contraindications to outpatient treatment.¹⁰

Limited data exist on whether or not clinical outcomes differ between low-risk patients treated in an outpatient setting vs those treated in an inpatient setting.¹¹ The aim of the present study was to compare 30-day mortality rates, time to return to work and usual activities, and satisfaction with medical care between low-risk outpatients and low-risk inpatients without contraindications to outpatient treatment.

Materials and Methods

We analyzed the data from low-risk patients without contraindications to outpatient treatment who were enrolled in a cluster randomized trial that was designed to assess the effectiveness and safety of using the PSI to guide the selection of the initial site of treatment for patients with pneumonia. Descriptions of the study design and the primary outcomes of this trial have been reported elsewhere.⁷¹² Institutional review boards at all sites approved the trial, and all enrolled patients provided informed consent for participation.

Physicians from 32 emergency departments in Connecticut and western Pennsylvania recruited patients between January and December 2001. Eligible patients were 18 years of age with a clinical diagnosis of pneumonia and new pulmonary infiltrates seen on a radiograph. Patients were excluded if they were considered to have hospital-acquired pneumonia, pulmonary tuberculosis, immune suppression, positive serology for HIV, alcoholism with evidence of end-organ damage, illicit drug use within the past 30 days, or social problems that were incompatible with outpatient treatment, study enrollment, or follow-up.¹²

The present analysis focused on low-risk patients without contraindications to outpatient treatment. Low-risk patients were defined as those in PSI risk classes I to III who did not have evidence of arterial oxygen desaturation (ie, oxygen saturation of < 90% or oxygen tension of < 60 mm Hg) at presentation.¹² Predefined contraindications to outpatient treatment included clinical and psychosocial factors that may affect compliance with oral antibiotic therapy, frailty or severe neuromuscular disorder, serious concomitant illness, severe abnormalities in vital signs or laboratory values, and suppurative infection (see "Appendix").¹² Trained research nurses collected patient sociodemographics and initial vital signs, and recorded mental status, comorbid conditions, medical treatments received prior to presentation, physical examination findings, pertinent laboratory test results, and chest radiography findings from the time of presentation.

We defined outpatient treatment as discharge from the emergency department to any outpatient setting or discharge from an emergency department observation unit within 24 h of initial presentation. Conversely, we defined inpatient treatment as hospital admission, transfer from an emergency department to an inpatient hospital observation unit, or admission to an emergency department observation unit with discharge to any setting > 24 h after initial presentation.⁷ Although the trial intervention advocated outpatient treatment of low-risk patients without evidence of arterial oxygen desaturation at presentation, all physicians made the initial site-of-treatment decision autonomously.¹²

Outcome measures for this study were 30-day mortality rate, time to return to work and usual activities, and patient satisfaction with care. Research nurses obtained follow-up clinical data from a structured review of medical records covering the first 30 days after the index emergency department presentation, and trained interviewers conducted telephone interviews with patients at least 30 days after the index emergency department presentation. We identified deaths using both medical record review and telephone interviews. The dates of the return to work for workers and the dates of the return to usual activities for workers and nonworkers were collected by telephone interview. Satisfaction with the site of treatment, emergency department care, and overall medical care were assessed by telephone interview.

Statistical Analysis
Because the site of treatment was not randomly assigned in our study population, patients with specific medical and psychosocial characteristics were more likely to be treated as inpatients. Since these characteristics might be related to our study outcomes, a direct comparison between inpatients and outpatients would have been biased. For this reason, we used a propensity score analysis to adjust for potential confounders of site of treatment.¹³¹⁴ Propensity analysis attempts to compare outcomes between outpatients and inpatients who have a similar distribution of measured covariates and, in this way, approximates the conditions of random site-of-treatment assignment.¹⁵¹⁶ For this purpose, we first developed a nonparsimonious logistic regression model that predicted outpatient treatment based on the 34 patient, physician, and emergency department covariates listed in Tables 123 . This logistic regression model yielded a c-statistic of 0.88, indicating a strong ability to discriminate between outpatient and inpatient treatment. This logistic regression model was used to estimate a propensity score for each patient, corresponding to the probability of being treated as an outpatient. Patients were stratified by quintile of increasing propensity score. To validate our propensity score adjustment, we checked for adequate overlap in propensity scores for outpatients and inpatients within each quintile and for the absence of significant residual imbalances in patient characteristics after adjustment for quintile of the propensity score.

We compared mortality rates for all low-risk inpatients and outpatients who were stratified by quintile of propensity score; we also compared mortality rates for the propensity-matched outpatients and inpatients. For these analyses, we used the Fisher exact test due to the sparse number of deaths within the comparison groups. We compared the time to the return to work and to usual activities, and satisfaction with care for outpatients and inpatients after adjusting for the quintile of propensity score in the entire cohort of low-risk patients as well as in the cohort of propensity-matched patients. We used multilevel logistic regression to estimate the odds ratios (ORs) of binary outcomes and associated 95% confidence intervals (CIs) with the three levels defined by patient, physician, and emergency department.¹⁸ The times to the return to work and usual activities were analyzed using multilevel discrete time survival models. ORs for propensity-matched patients were estimated using conditional logistic regression. Two-sided p values of < 0.05 were considered to be statistically significant. Analyses were performed using two statistical software packages (Stata, version 9.0; Stata Corporation; College Station, TX; and MlWiN, version 2.0; Institute of Education; London, UK).

Results

Of the 3,201 patients who were enrolled in the original trial, 1,708 were excluded from this study (higher risk patients, 1,312; low-risk patients with one or more contraindications to outpatient treatment, 350; and low-risk patients with one or more missing baseline characteristics included in the propensity score, 46) [Fig 1 ]. Our analytical sample consisted of 1,493 low-risk patients without a contraindication to outpatient treatment (outpatients, 944 [63%]; and inpatients, 549 [37%]).

View larger version (12K): [in this window] [in a new window] [Download PPT slide]   Figure 1.. Patient enrollment. The analytical sample consisted of low-risk patients (ie, PSI risk classes I to III) without evidence of arterial oxygen saturation at presentation (ie, oxygen saturation of < 90% or oxygen tension of < 60 mm Hg) and without a contraindication to outpatient treatment who had complete baseline data considered in the derivation of the propensity score. This analytical sample consisted of 944 outpatients and 549 inpatients.

Outcome Measures
Although mortality was 0.1% and 2.6% for all low-risk outpatients and inpatients respectively (p < 0.01), we observed no differences in mortality between outpatients and inpatients within each quintile of propensity score or within the cohort of propensity-matched outpatients and inpatients (Table 4 ). Twelve of the 15 deaths (80%) occurred among patients with preexisting treatments, radiographic abnormalities, or comorbid conditions not contained in the PSI

As found in the Pneumonia Patient Outcomes Research Team study,¹⁹ our study showed that low-risk outpatients resumed normal activities 6 to 9 days sooner than low-risk inpatients. This difference persisted after propensity score adjustment for imbalances in baseline characteristics and was not explained by the hospital length of stay among low-risk inpatients. This finding may be explained by the deconditioning that results from hospitalization, whereas patients treated in the outpatient setting remain active.²⁰ Another potential explanation is that inpatients are more likely to experience hospital-related adverse events such as hospital-acquired infection or venous thromboembolic disease.²⁰ Although there was no difference in mortality rate among propensity-matched outpatients and inpatients, the overall mortality rate was higher in low-risk inpatients compared to outpatients (2.6% vs 0.1%, respectively). This finding indicates that even within a homogeneous population of low-risk patients physicians are capable of discerning subtle differences in illness severity and use such clinical judgments to guide the selection of the initial site of treatment.

Two groups of hospitalized low-risk patients could be identified in this study. The first group accounted for 70% of hospitalizations for low-risk patients and consisted of patients with preexisting treatments, radiographic abnormalities, or comorbid conditions not contained in the PSI. None of these factors has independent associations with increased mortality rates after controlling for other demographic and clinical characteristics found in the PSI.¹⁰ Furthermore, many patients with preexisting treatments, radiographic abnormalities, or comorbid conditions not contained in the PSI were treated as outpatients. Nevertheless, 12 of the 15 deaths (80%) occurred among patients with these factors, suggesting that the hospitalization of these patients, based on physician judgment, was appropriate. The second group accounted for 30% of the low-risk inpatients and consisted of patients who had no identifiable risk factors for hospitalization other than those that determine PSI risk class. The reasons for the hospitalization of such patients may have included patient or family preferences, physician risk aversion, or physician judgment that the patient’s severity of illness warranted hospitalization despite their low-risk classification and the absence of medical or psychosocial contraindications to outpatient care. The appropriateness of hospitalization for this group of inpatients with no contraindications to outpatient treatment and no discernable risk factor for hospitalization is less clear.

Most North American guidelines⁴⁵⁶ advocate outpatient treatment for low-risk patients with pneumonia. However, concerns about the safety of using the PSI as a rule for determining the initial site of treatment have been expressed.²¹ The PSI is constructed with dichotomous predictor variables and inevitably oversimplifies the hospital admission decision-making process.²²²³ As a result, several descriptive series of low-risk patients hospitalized for pneumonia have identified the potential limitations of using the PSI alone in guiding the selection of the initial site of treatment, including the presence of arterial hypoxemia,²⁴²⁵²⁶ suspicion of sepsis,²⁵ an inability to maintain oral intake,²⁴²⁵²⁷ and a lack of adequate social support.²⁴²⁵²⁷ To address such concerns in our analysis, we excluded low-risk patients who had medical or psychosocial contraindications to outpatient treatment that were explicitly identified by the EDCAP guideline.¹² Carratala et al¹¹ used similar exclusion criteria in their randomized trial of outpatient treatment for low-risk patients with pneumonia. We believe that the PSI should be viewed as a guide to decisions on hospital admission and should not replace clinical judgment.²²²³ In patients with conditions requiring inpatient treatment or with psychosocial problems that preclude outpatient treatment, physician judgment should supercede the site-of-treatment decision indicated by the PSI.²⁸

To our knowledge, only one study¹¹ has compared clinical outcomes between low-risk patients with pneumonia who were randomly assigned to outpatient treatment vs inpatient treatment. In this trial, Carratala et al¹¹ found that a composite end point composed of 30-day mortality, subsequent hospital admission, changes in initial antibiotic treatment, additional medical visits, medical complications, adverse drug reactions, and lack of cure of pneumonia was not different according to the site of treatment for patients in PSI risk classes II and III. However, this trial was not powered to detect differences in important individual outcomes such as death, given the relatively small sample size. Although Carratala et al¹¹ found that outpatient treatment was associated with greater patient satisfaction, we did not observe differences in this outcome between outpatients and inpatients after adjusting for propensity score.

Our study limitations should be acknowledged. First, the site of treatment was not based on random assignment; therefore, our findings might be explained by the fact that more severely ill patients were more likely to be treated as inpatients. To control for potential confounding, we performed a propensity-adjusted analysis that eliminated significant imbalances between outpatients and inpatients for all measured patient, physician, and emergency department characteristics. Yet, unmeasured confounding factors might explain our findings. Indeed, propensity analysis can only adjust for measured factors, while random assignment balances measured and unmeasured factors.¹³ Second, the associations between the site of treatment and the characteristics examined in our study might have been altered by the design of the EDCAP trial.⁷ Although this trial was designed to test the effectiveness of three interventions of incremental intensity in increasing the proportion of low-risk patients treated as outpatients, we did not observe any significant first-order interactions involving the intervention arm, suggesting that our findings were independent of the intervention arm. Third, our study was conducted in emergency departments in Connecticut and western Pennsylvania, and our findings may not extend to patients who were managed in other clinical settings or geographic locations. Fourth, workers accounted for only 43% of all low-risk patients and only 36% of propensity-matched patients, a fact that may limit the relevance of the observed shorter time to the return to work for outpatients.

In conclusion, outpatient treatment of patients in PSI risk classes I to III was associated with a more rapid return to usual activities and to work, with no decrement in satisfaction with care. Although propensity-matched analyses revealed no difference in mortality rates among low-risk outpatients and inpatients, the increased risk of overall mortality for all low-risk inpatients in this study emphasizes the importance of physician judgment as a complement to objective risk stratification in the selection of the initial site of treatment for patients with pneumonia.

Appendix

Medical and psychosocial contraindications to outpatient treatment included the following new conditions or exacerbations of chronic conditions that were documented as being present during the initial visit in the emergency department: (1) stupor or coma; (2) severe dementia, delirium, psychiatric illness, acute confusion, or disorientation that could affect compliance with prescribed oral antibiotics or other outpatient treatments; (3) intractable vomiting or the inability to take oral antibiotics; (4) frailty or serious neuromuscular disorder with an inability to conduct activities of daily living at home; (5) known or suspected severe concomitant illness requiring hospitalization (ie, unstable angina, acute myocardial infarction, life-threatening ventricular arrhythmia, cardiopulmonary arrest, GI bleeding with volume depletion, acute asthma or exacerbation of chronic pulmonary disease with severe bronchospasm treated with IV steroids, heart failure or pulmonary edema treated with IV diuretics, diabetic ketoacidosis or hyperosmolar coma, acute renal failure, and thromboembolic disease); (6) known or suspected concomitant suppurative infection (ie, empyema, septic arthritis, bacterial meningitis, endocarditis, peritonitis, brain abscess, pulmonary abscess, or osteomyelitis); (7) severe abnormality in one or more vital signs (ie, pulse of > 150 beats/min, respiratory rate of > 40 breaths/min, or systolic BP of < 60 mm Hg) or laboratory values (ie, hematocrit of < 24%, serum sodium level of < 120 mEq/L, serum glucose level of > 500 mg/dL, or arterial pH of < 7.25) documented at any time in the emergency department.¹²

Footnotes

Abbreviations: CI = confidence interval; EDCAP = emergency department community-acquired pneumonia; IQR = interquartile range; OR = odds ratio; PSI = pneumonia severity index

This research was conducted as part of the project entitled "Guideline to Improve Quality of Initial Pneumonia Care," funded by the Agency for Healthcare Research and Quality (grant No. R01 HS10049). Dr. M.J. Fine was supported in part by a K-24 career development award from the National Institute of Allergy and Infectious Diseases (5K24 AI01769). Dr. Labarere was supported in part by a grant from the Egide foundation, Paris (Programme Lavoisier), France, and by a grant from the Délégation Régionale de la Recherche Clinique, Centre Hospitalier Universitaire de Grenoble, France.

The authors have reported to the ACCP that no significant conflicts of interest exist with any companies/organizations whose products or services may be discussed in this article.

Received for publication June 2, 2006. Accepted for publication August 31, 2006.

References

1. DeFrances, CJ, Hall, MJ, Podgornik, MN (2005) 2003 National Hospital Discharge Survey: advance data from Vital and Health Statistics. National Center for Health Statistics. Hyattsville, MD: publication No. 359
2. Lave, JR, Lin, CCJ, Fine, MJ, et al The cost of treating patients with community-acquired pneumonia. Semin Respir Crit Care Med 1999;20,189-197
3. Fine, MJ, Pratt, HM, Obrosky, DS, et al Relation between length of hospital stay and costs of care for patients with community-acquired pneumonia. Am J Med 2000;109,378-385[CrossRef][ISI][Medline]
4. Mandell, LA, Marrie, TJ, Grossman, RF, et al Canadian guidelines for the initial management of community-acquired pneumonia: an evidence-based update by the Canadian Infectious Diseases Society and the Canadian Thoracic Society; the Canadian Community-Acquired Pneumonia Working Group. Clin Infect Dis 2000;31,383-421[CrossRef][Medline]
5. 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;37,1405-1433[CrossRef][ISI][Medline]
6. Bartlett, JG, Dowell, SF, Mandell, LA, et al Practice guidelines for the management of community-acquired pneumonia in adults: Infectious Diseases Society of America. Clin Infect Dis 2000;31,347-382[CrossRef][Medline]
7. Yealy, DM, Auble, TE, Stone, RA, et al Effect of increasing the intensity of implementing pneumonia guidelines: a randomized, controlled trial. Ann Intern Med 2005;143,881-894[Abstract/Free Full Text]
8. Marrie, TJ, Lau, CY, Wheeler, SL, et al 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;283,749-755[Abstract/Free Full Text]
9. Atlas, SJ, Benzer, TI, Borowsky, LH, et al Safely increasing the proportion of patients with community-acquired pneumonia treated as outpatients: an interventional trial. Arch Intern Med 1998;158,1350-1356[Abstract/Free Full Text]
10. Labarere, J, Stone, RA, Obrosky, DS, et al Factors associated with the hospitalization of low-risk patients with community-acquired pneumonia in a cluster-randomized trial. J Gen Intern Med 2006;21,745-752[CrossRef][ISI][Medline]
11. Carratala, J, Fernandez-Sabe, N, Ortega, L, et al Outpatient care compared with hospitalization for community-acquired pneumonia: a randomized trial in low-risk patients. Ann Intern Med 2005;142,165-172[Abstract/Free Full Text]
12. Yealy, DM, Auble, TE, Stone, RA, et al The emergency department community-acquired pneumonia trial: methodology of a quality improvement intervention. Ann Emerg Med 2004;43,770-782[CrossRef][ISI][Medline]
13. Joffe, MM, Rosenbaum, PR Invited commentary: propensity scores. Am J Epidemiol 1999;150,327-333[Abstract/Free Full Text]
14. Stone, RA, Obrosky, DS, Singer, DE, et al Propensity score adjustment for pretreatment differences between hospitalized and ambulatory patients with community-acquired pneumonia: Pneumonia Patient Outcomes Research Team (PORT) Investigators. Med Care 1995;33,AS56-AS66[ISI][Medline]
15. D’Agostino, RB, Jr Propensity score methods for bias reduction in the comparison of a treatment to a non-randomized control group. Stat Med 1998;17,2265-2281[CrossRef][ISI][Medline]
16. Artinian, V, Krayem, H, DiGiovine, B Effects of early enteral feeding on the outcome of critically ill mechanically ventilated medical patients. Chest 2006;129,960-967
17. Leuven E, Sianesi B. PSMATCH2: Stata module to perform full Mahalanobis and propensity score matching, common support graphing, and covariate imbalance testing. Available at: http://ideas.repec.org/c/boc/bocode/s432001.html. Accessed January 9, 2007
18. Rice, N Binomial regression. Leyland, AH Goldstein, H eds. Multilevel modelling of health statistics 2001,27-43 John Wiley & Sons Ltd. Chichester, UK:
19. Fine, MJ, Singer, DE, Marrie, TJ Medical outcomes of ambulatory and hospitalized low risk patients with community-acquired pneumonia [abstract]. J Gen Intern Med 1994;9,29[ISI][Medline]
20. Mandell, L Decisions about treating community-acquired pneumonia. Ann Intern Med 2005;142,215-216[Free Full Text]
21. Niederman, MS What is the prognosis for using the pneumonia severity index to make site-of-care decisions in community-acquired pneumonia? Chest 2003;124,2051-2053
22. 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;163,1730-1754[Free Full Text]
23. 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;336,243-250[Abstract/Free Full Text]
24. Marras, TK, Gutierrez, C, Chan, CK Applying a prediction rule to identify low-risk patients with community-acquired pneumonia. Chest 2000;118,1339-1343
25. Arnold, FW, Ramirez, JA, McDonald, LC, et al Hospitalization for community-acquired pneumonia: the pneumonia severity index vs clinical judgment. Chest 2003;124,121-124
26. Roson, B, Carratala, J, Dorca, J, et al Etiology, reasons for hospitalization, risk classes, and outcomes of community-acquired pneumonia in patients hospitalized on the basis of conventional admission criteria. Clin Infect Dis 2001;33,158-165[CrossRef][ISI][Medline]
27. Goss, CH, Rubenfeld, GD, Park, DR, et al Cost and incidence of social comorbidities in low-risk patients with community-acquired pneumonia admitted to a public hospital. Chest 2003;124,2148-2155
28. Marrie, TJ, Huang, JQ Low-risk patients admitted with community-acquired pneumonia. Am J Med 2005;118,1357-1363[CrossRef][ISI][Medline]

This Article Abstract Full Text (PDF) Free Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Article Archive Download to citation manager Citing Articles Citing Articles via ISI Web of Science (3) Citing Articles via Google Scholar Google Scholar Articles by Labarere, J. Articles by Fine, M. J. Search for Related Content PubMed PubMed Citation Articles by Labarere, J. Articles by Fine, M. J.