In the United States, community‐acquired pneumonia (CAP) leads to nearly 1 million hospitalizations annually, with aggregate costs of hospitalization approaching $9 billion.1, 2 In an effort to improve the appropriate, cost‐effective care for patients with CAP, several professional societies have developed clinical practice guidelines and pathways for pneumonia.37 Although the guidelines address all aspects of care, they devote substantial attention to antibiotic recommendations. Most U.S. guidelines recommend treatment of hospitalized patients with an intravenous beta‐lactam combined with a macrolide, or a fluoroquinolone with activity against Streptococcus pneumoniae. Of the major U.S. CAP practice guidelines, only one6 recommends doxycycline as an alternative to a macrolide for inpatients.

Doxycycline is an attractive alternative to macrolides. Similar to macrolides, doxycycline is active against a wide variety of organisms including atypical bacteria (Chlamydia pneumoniae, Legionella pneumophilus, and Mycoplasma pneumoniae) and is well tolerated.810 In addition, it is inexpensive (cost of $1.00/day [awp] for 100 mg p.o. bid), and rates of tetracycline/doxycycline resistance among S. pneumoniae isolates have remained low, in contrast to the increasing rates of resistance to macrolides and fluoroquinolones.11, 12 The most recent guidelines from the Infectious Diseases Society of America cited limited published clinical data on the effectiveness of doxycycline in CAP as a barrier to increased use.7 Only one study of hospitalized patients has been published in the era of penicillin‐resistant pneumococcus, and this study included only 43 low‐risk patients treated with doxycycline.13 At the university hospital affiliated with the University of California, San Francisco, ceftriaxone plus doxycycline is generally recommended as initial empiric antibiotic therapy for patients hospitalized with CAP, but significant variability in prescribing exists, allowing for comparisons between patients treated with different initial empiric antibiotic regimens. We compared outcomes of hospitalized patients with CAP treated with ceftriaxone plus doxycycline to those of patients treated with alternative initial empiric therapy at an academic medical center.

METHODS

RESULTS

DISCUSSION

In our hospital setting, the use of ceftriaxone plus doxycycline as the initial empiric antibiotic therapy for patients hospitalized with community‐acquired pneumonia was associated with significantly lower inpatient and 30‐day mortality, even after adjusting for clinical differences between groups. We did not find a difference between regimens in hospital length of stay or 30‐day readmission rate. In case the multivariable model was insufficient to account for the clinical differences (i.e., selection bias) between groups, we also performed an analysis of a subgroup of less severely ill patients by excluding those admitted from nursing homes, those admitted to the intensive care unit, and those with aspiration pneumonia. In this subset, use of ceftriaxone plus doxycycline remained associated with lower inpatient mortality but not with lower 30‐day mortality. Although, as an observational study, the results of our findings could still be a result of residual confounding, we believe the results provide valuable information regarding doxycycline.

Combination therapy with a macrolide, but not doxycycline, is advocated by the practice guidelines of several major U.S. professional societies,3, 4, 7 apparently because of a lack of data on the effectiveness of combination therapy with doxycycline.7 Only one randomized, unblinded study, in 87 low‐risk patients hospitalized with CAP, that compared monotherapy with IV doxycycline versus physician‐determined therapy has been conducted.13 This study found no differences between treatment groups in clinical outcomes but did find that use of doxycycline was associated with shorter hospital stays and reduced costs. Our results, achieved in a real‐world setting in relatively ill hospitalized patients (58% were in PSI risk class IV or V), provide further support for the use of combination therapy with doxycycline.

Hospitalized patients treated with a beta‐lactam in combination with a macrolide are often discharged on macrolide monotherapy. In our population most patients treated with ceftriaxone plus doxycycline were discharged on doxycycline if they required continued therapy (data not shown). In the current era of resistance of Streptococcus pneumoniae to antibiotics, there is good reason to believe doxycycline may perform as well, if not better, than macrolides when hospitalized patients with CAP are discharged on oral monotherapy. Macrolide resistance rates among invasive pneumococcal isolates in the United States doubled from 10% to 20% during a period in which prescriptions for macrolides increased by 13%.12 In addition, a large surveillance study of more than 1500 isolates collected in 1999 and 2000 found that 26% of the isolates were resistant to macrolides, whereas only 16% were resistant to tetracycline.16 In vitro testing against S. Pneumoniae has also suggested that tetracycline resistance overestimates doxycycline resistance.17, 18 More recently, Streptococcus pneumoniae susceptibility data from the SENTRY Antimicrobial Surveillance program reaffirmed doxycycline's in vitro superiority over macrolides.17

Our study had several limitations. The study design adopted precluded determining whether favorable results with the use of ceftriaxone plus doxycycline resulted from an effect unique to this combination of antibiotics, the possible anti‐inflammatory properties of doxycycline alone,19, 20 or unmeasured confounders. For example, processes of care that affect clinical outcomes for patients hospitalized with CAP, such as the timing of antibiotic delivery, the timing of blood cultures, and stability assessment on discharge were not measured in this study. To affect outcomes, these processes of care would need to be differentially distributed between our comparison groups. However, because this study was performed in a single institution during a single interval, it is likely that the performance of these processes of care would be similar for all patients.

In conclusion, ceftriaxone plus doxycycline appears to be an effective, and possibly superior, therapy for patients hospitalized with CAP. Randomized controlled trials of doxycycline‐containing regimens versus other regimens are warranted.

In the United States, community‐acquired pneumonia (CAP) leads to nearly 1 million hospitalizations annually, with aggregate costs of hospitalization approaching $9 billion.1, 2 In an effort to improve the appropriate, cost‐effective care for patients with CAP, several professional societies have developed clinical practice guidelines and pathways for pneumonia.37 Although the guidelines address all aspects of care, they devote substantial attention to antibiotic recommendations. Most U.S. guidelines recommend treatment of hospitalized patients with an intravenous beta‐lactam combined with a macrolide, or a fluoroquinolone with activity against Streptococcus pneumoniae. Of the major U.S. CAP practice guidelines, only one6 recommends doxycycline as an alternative to a macrolide for inpatients.

Doxycycline is an attractive alternative to macrolides. Similar to macrolides, doxycycline is active against a wide variety of organisms including atypical bacteria (Chlamydia pneumoniae, Legionella pneumophilus, and Mycoplasma pneumoniae) and is well tolerated.810 In addition, it is inexpensive (cost of $1.00/day [awp] for 100 mg p.o. bid), and rates of tetracycline/doxycycline resistance among S. pneumoniae isolates have remained low, in contrast to the increasing rates of resistance to macrolides and fluoroquinolones.11, 12 The most recent guidelines from the Infectious Diseases Society of America cited limited published clinical data on the effectiveness of doxycycline in CAP as a barrier to increased use.7 Only one study of hospitalized patients has been published in the era of penicillin‐resistant pneumococcus, and this study included only 43 low‐risk patients treated with doxycycline.13 At the university hospital affiliated with the University of California, San Francisco, ceftriaxone plus doxycycline is generally recommended as initial empiric antibiotic therapy for patients hospitalized with CAP, but significant variability in prescribing exists, allowing for comparisons between patients treated with different initial empiric antibiotic regimens. We compared outcomes of hospitalized patients with CAP treated with ceftriaxone plus doxycycline to those of patients treated with alternative initial empiric therapy at an academic medical center.

METHODS

RESULTS

DISCUSSION

In our hospital setting, the use of ceftriaxone plus doxycycline as the initial empiric antibiotic therapy for patients hospitalized with community‐acquired pneumonia was associated with significantly lower inpatient and 30‐day mortality, even after adjusting for clinical differences between groups. We did not find a difference between regimens in hospital length of stay or 30‐day readmission rate. In case the multivariable model was insufficient to account for the clinical differences (i.e., selection bias) between groups, we also performed an analysis of a subgroup of less severely ill patients by excluding those admitted from nursing homes, those admitted to the intensive care unit, and those with aspiration pneumonia. In this subset, use of ceftriaxone plus doxycycline remained associated with lower inpatient mortality but not with lower 30‐day mortality. Although, as an observational study, the results of our findings could still be a result of residual confounding, we believe the results provide valuable information regarding doxycycline.

Combination therapy with a macrolide, but not doxycycline, is advocated by the practice guidelines of several major U.S. professional societies,3, 4, 7 apparently because of a lack of data on the effectiveness of combination therapy with doxycycline.7 Only one randomized, unblinded study, in 87 low‐risk patients hospitalized with CAP, that compared monotherapy with IV doxycycline versus physician‐determined therapy has been conducted.13 This study found no differences between treatment groups in clinical outcomes but did find that use of doxycycline was associated with shorter hospital stays and reduced costs. Our results, achieved in a real‐world setting in relatively ill hospitalized patients (58% were in PSI risk class IV or V), provide further support for the use of combination therapy with doxycycline.

Hospitalized patients treated with a beta‐lactam in combination with a macrolide are often discharged on macrolide monotherapy. In our population most patients treated with ceftriaxone plus doxycycline were discharged on doxycycline if they required continued therapy (data not shown). In the current era of resistance of Streptococcus pneumoniae to antibiotics, there is good reason to believe doxycycline may perform as well, if not better, than macrolides when hospitalized patients with CAP are discharged on oral monotherapy. Macrolide resistance rates among invasive pneumococcal isolates in the United States doubled from 10% to 20% during a period in which prescriptions for macrolides increased by 13%.12 In addition, a large surveillance study of more than 1500 isolates collected in 1999 and 2000 found that 26% of the isolates were resistant to macrolides, whereas only 16% were resistant to tetracycline.16 In vitro testing against S. Pneumoniae has also suggested that tetracycline resistance overestimates doxycycline resistance.17, 18 More recently, Streptococcus pneumoniae susceptibility data from the SENTRY Antimicrobial Surveillance program reaffirmed doxycycline's in vitro superiority over macrolides.17

Our study had several limitations. The study design adopted precluded determining whether favorable results with the use of ceftriaxone plus doxycycline resulted from an effect unique to this combination of antibiotics, the possible anti‐inflammatory properties of doxycycline alone,19, 20 or unmeasured confounders. For example, processes of care that affect clinical outcomes for patients hospitalized with CAP, such as the timing of antibiotic delivery, the timing of blood cultures, and stability assessment on discharge were not measured in this study. To affect outcomes, these processes of care would need to be differentially distributed between our comparison groups. However, because this study was performed in a single institution during a single interval, it is likely that the performance of these processes of care would be similar for all patients.

In conclusion, ceftriaxone plus doxycycline appears to be an effective, and possibly superior, therapy for patients hospitalized with CAP. Randomized controlled trials of doxycycline‐containing regimens versus other regimens are warranted.

In the United States, community‐acquired pneumonia (CAP) leads to nearly 1 million hospitalizations annually, with aggregate costs of hospitalization approaching $9 billion.1, 2 In an effort to improve the appropriate, cost‐effective care for patients with CAP, several professional societies have developed clinical practice guidelines and pathways for pneumonia.37 Although the guidelines address all aspects of care, they devote substantial attention to antibiotic recommendations. Most U.S. guidelines recommend treatment of hospitalized patients with an intravenous beta‐lactam combined with a macrolide, or a fluoroquinolone with activity against Streptococcus pneumoniae. Of the major U.S. CAP practice guidelines, only one6 recommends doxycycline as an alternative to a macrolide for inpatients.

Doxycycline is an attractive alternative to macrolides. Similar to macrolides, doxycycline is active against a wide variety of organisms including atypical bacteria (Chlamydia pneumoniae, Legionella pneumophilus, and Mycoplasma pneumoniae) and is well tolerated.810 In addition, it is inexpensive (cost of $1.00/day [awp] for 100 mg p.o. bid), and rates of tetracycline/doxycycline resistance among S. pneumoniae isolates have remained low, in contrast to the increasing rates of resistance to macrolides and fluoroquinolones.11, 12 The most recent guidelines from the Infectious Diseases Society of America cited limited published clinical data on the effectiveness of doxycycline in CAP as a barrier to increased use.7 Only one study of hospitalized patients has been published in the era of penicillin‐resistant pneumococcus, and this study included only 43 low‐risk patients treated with doxycycline.13 At the university hospital affiliated with the University of California, San Francisco, ceftriaxone plus doxycycline is generally recommended as initial empiric antibiotic therapy for patients hospitalized with CAP, but significant variability in prescribing exists, allowing for comparisons between patients treated with different initial empiric antibiotic regimens. We compared outcomes of hospitalized patients with CAP treated with ceftriaxone plus doxycycline to those of patients treated with alternative initial empiric therapy at an academic medical center.

METHODS

RESULTS

DISCUSSION

In our hospital setting, the use of ceftriaxone plus doxycycline as the initial empiric antibiotic therapy for patients hospitalized with community‐acquired pneumonia was associated with significantly lower inpatient and 30‐day mortality, even after adjusting for clinical differences between groups. We did not find a difference between regimens in hospital length of stay or 30‐day readmission rate. In case the multivariable model was insufficient to account for the clinical differences (i.e., selection bias) between groups, we also performed an analysis of a subgroup of less severely ill patients by excluding those admitted from nursing homes, those admitted to the intensive care unit, and those with aspiration pneumonia. In this subset, use of ceftriaxone plus doxycycline remained associated with lower inpatient mortality but not with lower 30‐day mortality. Although, as an observational study, the results of our findings could still be a result of residual confounding, we believe the results provide valuable information regarding doxycycline.

Combination therapy with a macrolide, but not doxycycline, is advocated by the practice guidelines of several major U.S. professional societies,3, 4, 7 apparently because of a lack of data on the effectiveness of combination therapy with doxycycline.7 Only one randomized, unblinded study, in 87 low‐risk patients hospitalized with CAP, that compared monotherapy with IV doxycycline versus physician‐determined therapy has been conducted.13 This study found no differences between treatment groups in clinical outcomes but did find that use of doxycycline was associated with shorter hospital stays and reduced costs. Our results, achieved in a real‐world setting in relatively ill hospitalized patients (58% were in PSI risk class IV or V), provide further support for the use of combination therapy with doxycycline.

Hospitalized patients treated with a beta‐lactam in combination with a macrolide are often discharged on macrolide monotherapy. In our population most patients treated with ceftriaxone plus doxycycline were discharged on doxycycline if they required continued therapy (data not shown). In the current era of resistance of Streptococcus pneumoniae to antibiotics, there is good reason to believe doxycycline may perform as well, if not better, than macrolides when hospitalized patients with CAP are discharged on oral monotherapy. Macrolide resistance rates among invasive pneumococcal isolates in the United States doubled from 10% to 20% during a period in which prescriptions for macrolides increased by 13%.12 In addition, a large surveillance study of more than 1500 isolates collected in 1999 and 2000 found that 26% of the isolates were resistant to macrolides, whereas only 16% were resistant to tetracycline.16 In vitro testing against S. Pneumoniae has also suggested that tetracycline resistance overestimates doxycycline resistance.17, 18 More recently, Streptococcus pneumoniae susceptibility data from the SENTRY Antimicrobial Surveillance program reaffirmed doxycycline's in vitro superiority over macrolides.17

Our study had several limitations. The study design adopted precluded determining whether favorable results with the use of ceftriaxone plus doxycycline resulted from an effect unique to this combination of antibiotics, the possible anti‐inflammatory properties of doxycycline alone,19, 20 or unmeasured confounders. For example, processes of care that affect clinical outcomes for patients hospitalized with CAP, such as the timing of antibiotic delivery, the timing of blood cultures, and stability assessment on discharge were not measured in this study. To affect outcomes, these processes of care would need to be differentially distributed between our comparison groups. However, because this study was performed in a single institution during a single interval, it is likely that the performance of these processes of care would be similar for all patients.

In conclusion, ceftriaxone plus doxycycline appears to be an effective, and possibly superior, therapy for patients hospitalized with CAP. Randomized controlled trials of doxycycline‐containing regimens versus other regimens are warranted.

You hold in your hands the inaugural issue of the Journal of Hospital Medicine (JHM). Our goal is for JHM to become the premier forum for peer‐reviewed research articles and evidence‐based reviews in the specialty of hospital medicine.

Yes, the specialty of hospital medicine. This official publication of the Society of Hospital Medicine signifies another step forward in the evolution of this specialty. With the publication of JHM the Society of Hospital Medicine continues its pivotal educational and leadership role in shaping the practice of hospital medicine. The Society is dedicated to promoting the highest‐quality care for all hospitalized patients and excellence in hospital medicine through education, advocacy, and research. As part of the Society's effort to improve care and standards, it is providing JHM to all members as part of their membership. We hope that our readership will grow to include individuals involved in all aspects of hospital care.

Packed with the results of new studies and state‐of‐the‐art reviews, JHM is not aimed solely at academicians and voracious readers of the medical literature. Rather, we hope that it fills a practical need to promote lifelong learning in both hospitalists and their hospital colleagues. For example, in this issue, national experts in palliative care and geriatrics summarize the pertinent literature and the important role of such care for hospitalized patients. JHM will also serve as a key venue for hospital medicine researchers to disseminate their findings and for educators to share their knowledge and techniques.

Why bother to create yet another journal? Given the stacks of journals that adorn many of our desks (and some of our chairs and windowsills), do we really need another to get lost among the mail that inundates us? We believe the field of hospital medicine involves a growing body of knowledge deserving of a journal focused solely on it. Hospital medicine evolved from efforts to fill a need identified by overstretched primary care physicians in the late 1980s. Physicians like the cofounders of SHM, John Nelson in Florida and Win Whitcomb in Massachusetts, began careers in a field that today numbers more than 12,000 physicians. Labeled with the moniker hospitalist given us by Bob Wachter and Lee Goldman,1 we now make up the fastest‐growing medical specialty in the United States.2 Yet, until now, no journal was devoted solely to this specialty.

The Journal of Hospital Medicine aims to provide physicians and other health care professionals with continuing insight into the basic and clinical sciences to support informed clinical decision making in the hospital. As hospitalists increasingly take an active role in the successful delivery of bench research discoveries to the bedside and become vigorous participants in the translational and clinical research sought by the National Institutes of Health,3 JHM will disseminate their findings. In addition, we hope to foster balanced debates on medical issues and health care trends that affect hospital medicine and patient care. Nonclinical aspects of hospital medicine also will be featured, including public health and the political, philosophic, ethical, legal, environmental, economic, historical, and cultural issues surrounding hospital care. We especially want to encourage submissions that evaluate projects involving the entire hospital care team: physicians and our colleagues in the hospitalnurses, pharmacists, administrators, physical and occupational therapists, social workers, and case managers.

Two articles (see pages 48 and 57) highlight this inaugural issue. One describes the development of The Core Competencies in Hospital Medicine: A Framework for Curriculum Development (a supplement to this issue), and the other demonstrates how this document can be applied to curriculum development.4, 5 This milestone in the evolution of hospital medicine provides an initial structural framework to guide medical educators in developing curricula that incorporate these competencies into the training and evaluation of students, clinicians‐in‐training, and practicing hospitalists.4 The president and CEO of the American Board of Internal Medicine (ABIM), Christine Cassel, offers her perspective on this landmark document.6 Its timeliness is reflected by the current efforts of the American College of Physicians, the ABIM, and others to redesign the training and the certification requirements of internists. As this supplement demonstrates, the Society of Hospital Medicine will be intimately involved in this process.

After this auspicious start, subsequent issues will include articles in the following categories. Original research articles will report results of randomized controlled trials, evaluations of diagnostic tests, prospective cohort studies, casecontrol studies, and high‐quality observational studies. We are interested in publishing both quantitative and qualitative research. Review articles, especially those targeting the hospital medicine core competencies, are eagerly sought. We also seek descriptions of interventions that transform hospital care delivery in the hospital. For example, accounts of the implementation of quality‐improvement projects and outcomes, including barriers that were overcome or that blocked implementation, would be invaluable to hospitalists throughout the country. Clinical conundrums should describe clinical cases that present diagnostic dilemmas or involve issues of medical errors. To facilitate the professional development of hospitalists, we seek articles focused on their professional development in community, academic, and administrative settings. Examples of leadership topics are managing physician performance, leading and managing change, and self‐evaluation. Teaching tips or descriptions of educational programs or curricula also are desired. For researchers, potential topics include descriptions of specific techniques used for surveys, meta‐analyses, economic evaluations, and statistical analyses.Penetrating point manuscripts, those that go beyond the cutting edge to explain the next potential breakthrough or intervention in the developing field of hospital medicine, may be authored by thought leaders inside and outside the health care field as well as by hospitalists with novel ideas. Equally vital, I want to share the illuminating perspectives of physicians, patients, and families of patients as they reflect on the experience of being in the hospitalhospitalists can enlighten us through their handoffs, and patients and their families can inform us about their view from the hospital bed.

Finally, never forget that this is your journal. Let me know what you like and what changes you think can make it better. Please e‐mail your suggestions, comments, criticisms, and ideas to us at JHMeditor@ hospitalmedicine.org. This is your chance to help shape the practice of hospital medicine and the future of hospital care. I look forward to your guidance. Together we can expand our knowledge and continue to grow in our careers.

You hold in your hands the inaugural issue of the Journal of Hospital Medicine (JHM). Our goal is for JHM to become the premier forum for peer‐reviewed research articles and evidence‐based reviews in the specialty of hospital medicine.

Yes, the specialty of hospital medicine. This official publication of the Society of Hospital Medicine signifies another step forward in the evolution of this specialty. With the publication of JHM the Society of Hospital Medicine continues its pivotal educational and leadership role in shaping the practice of hospital medicine. The Society is dedicated to promoting the highest‐quality care for all hospitalized patients and excellence in hospital medicine through education, advocacy, and research. As part of the Society's effort to improve care and standards, it is providing JHM to all members as part of their membership. We hope that our readership will grow to include individuals involved in all aspects of hospital care.

Packed with the results of new studies and state‐of‐the‐art reviews, JHM is not aimed solely at academicians and voracious readers of the medical literature. Rather, we hope that it fills a practical need to promote lifelong learning in both hospitalists and their hospital colleagues. For example, in this issue, national experts in palliative care and geriatrics summarize the pertinent literature and the important role of such care for hospitalized patients. JHM will also serve as a key venue for hospital medicine researchers to disseminate their findings and for educators to share their knowledge and techniques.

Why bother to create yet another journal? Given the stacks of journals that adorn many of our desks (and some of our chairs and windowsills), do we really need another to get lost among the mail that inundates us? We believe the field of hospital medicine involves a growing body of knowledge deserving of a journal focused solely on it. Hospital medicine evolved from efforts to fill a need identified by overstretched primary care physicians in the late 1980s. Physicians like the cofounders of SHM, John Nelson in Florida and Win Whitcomb in Massachusetts, began careers in a field that today numbers more than 12,000 physicians. Labeled with the moniker hospitalist given us by Bob Wachter and Lee Goldman,1 we now make up the fastest‐growing medical specialty in the United States.2 Yet, until now, no journal was devoted solely to this specialty.

The Journal of Hospital Medicine aims to provide physicians and other health care professionals with continuing insight into the basic and clinical sciences to support informed clinical decision making in the hospital. As hospitalists increasingly take an active role in the successful delivery of bench research discoveries to the bedside and become vigorous participants in the translational and clinical research sought by the National Institutes of Health,3 JHM will disseminate their findings. In addition, we hope to foster balanced debates on medical issues and health care trends that affect hospital medicine and patient care. Nonclinical aspects of hospital medicine also will be featured, including public health and the political, philosophic, ethical, legal, environmental, economic, historical, and cultural issues surrounding hospital care. We especially want to encourage submissions that evaluate projects involving the entire hospital care team: physicians and our colleagues in the hospitalnurses, pharmacists, administrators, physical and occupational therapists, social workers, and case managers.

Two articles (see pages 48 and 57) highlight this inaugural issue. One describes the development of The Core Competencies in Hospital Medicine: A Framework for Curriculum Development (a supplement to this issue), and the other demonstrates how this document can be applied to curriculum development.4, 5 This milestone in the evolution of hospital medicine provides an initial structural framework to guide medical educators in developing curricula that incorporate these competencies into the training and evaluation of students, clinicians‐in‐training, and practicing hospitalists.4 The president and CEO of the American Board of Internal Medicine (ABIM), Christine Cassel, offers her perspective on this landmark document.6 Its timeliness is reflected by the current efforts of the American College of Physicians, the ABIM, and others to redesign the training and the certification requirements of internists. As this supplement demonstrates, the Society of Hospital Medicine will be intimately involved in this process.

After this auspicious start, subsequent issues will include articles in the following categories. Original research articles will report results of randomized controlled trials, evaluations of diagnostic tests, prospective cohort studies, casecontrol studies, and high‐quality observational studies. We are interested in publishing both quantitative and qualitative research. Review articles, especially those targeting the hospital medicine core competencies, are eagerly sought. We also seek descriptions of interventions that transform hospital care delivery in the hospital. For example, accounts of the implementation of quality‐improvement projects and outcomes, including barriers that were overcome or that blocked implementation, would be invaluable to hospitalists throughout the country. Clinical conundrums should describe clinical cases that present diagnostic dilemmas or involve issues of medical errors. To facilitate the professional development of hospitalists, we seek articles focused on their professional development in community, academic, and administrative settings. Examples of leadership topics are managing physician performance, leading and managing change, and self‐evaluation. Teaching tips or descriptions of educational programs or curricula also are desired. For researchers, potential topics include descriptions of specific techniques used for surveys, meta‐analyses, economic evaluations, and statistical analyses.Penetrating point manuscripts, those that go beyond the cutting edge to explain the next potential breakthrough or intervention in the developing field of hospital medicine, may be authored by thought leaders inside and outside the health care field as well as by hospitalists with novel ideas. Equally vital, I want to share the illuminating perspectives of physicians, patients, and families of patients as they reflect on the experience of being in the hospitalhospitalists can enlighten us through their handoffs, and patients and their families can inform us about their view from the hospital bed.

Finally, never forget that this is your journal. Let me know what you like and what changes you think can make it better. Please e‐mail your suggestions, comments, criticisms, and ideas to us at JHMeditor@ hospitalmedicine.org. This is your chance to help shape the practice of hospital medicine and the future of hospital care. I look forward to your guidance. Together we can expand our knowledge and continue to grow in our careers.

You hold in your hands the inaugural issue of the Journal of Hospital Medicine (JHM). Our goal is for JHM to become the premier forum for peer‐reviewed research articles and evidence‐based reviews in the specialty of hospital medicine.

Yes, the specialty of hospital medicine. This official publication of the Society of Hospital Medicine signifies another step forward in the evolution of this specialty. With the publication of JHM the Society of Hospital Medicine continues its pivotal educational and leadership role in shaping the practice of hospital medicine. The Society is dedicated to promoting the highest‐quality care for all hospitalized patients and excellence in hospital medicine through education, advocacy, and research. As part of the Society's effort to improve care and standards, it is providing JHM to all members as part of their membership. We hope that our readership will grow to include individuals involved in all aspects of hospital care.

Packed with the results of new studies and state‐of‐the‐art reviews, JHM is not aimed solely at academicians and voracious readers of the medical literature. Rather, we hope that it fills a practical need to promote lifelong learning in both hospitalists and their hospital colleagues. For example, in this issue, national experts in palliative care and geriatrics summarize the pertinent literature and the important role of such care for hospitalized patients. JHM will also serve as a key venue for hospital medicine researchers to disseminate their findings and for educators to share their knowledge and techniques.

Why bother to create yet another journal? Given the stacks of journals that adorn many of our desks (and some of our chairs and windowsills), do we really need another to get lost among the mail that inundates us? We believe the field of hospital medicine involves a growing body of knowledge deserving of a journal focused solely on it. Hospital medicine evolved from efforts to fill a need identified by overstretched primary care physicians in the late 1980s. Physicians like the cofounders of SHM, John Nelson in Florida and Win Whitcomb in Massachusetts, began careers in a field that today numbers more than 12,000 physicians. Labeled with the moniker hospitalist given us by Bob Wachter and Lee Goldman,1 we now make up the fastest‐growing medical specialty in the United States.2 Yet, until now, no journal was devoted solely to this specialty.

The Journal of Hospital Medicine aims to provide physicians and other health care professionals with continuing insight into the basic and clinical sciences to support informed clinical decision making in the hospital. As hospitalists increasingly take an active role in the successful delivery of bench research discoveries to the bedside and become vigorous participants in the translational and clinical research sought by the National Institutes of Health,3 JHM will disseminate their findings. In addition, we hope to foster balanced debates on medical issues and health care trends that affect hospital medicine and patient care. Nonclinical aspects of hospital medicine also will be featured, including public health and the political, philosophic, ethical, legal, environmental, economic, historical, and cultural issues surrounding hospital care. We especially want to encourage submissions that evaluate projects involving the entire hospital care team: physicians and our colleagues in the hospitalnurses, pharmacists, administrators, physical and occupational therapists, social workers, and case managers.

Two articles (see pages 48 and 57) highlight this inaugural issue. One describes the development of The Core Competencies in Hospital Medicine: A Framework for Curriculum Development (a supplement to this issue), and the other demonstrates how this document can be applied to curriculum development.4, 5 This milestone in the evolution of hospital medicine provides an initial structural framework to guide medical educators in developing curricula that incorporate these competencies into the training and evaluation of students, clinicians‐in‐training, and practicing hospitalists.4 The president and CEO of the American Board of Internal Medicine (ABIM), Christine Cassel, offers her perspective on this landmark document.6 Its timeliness is reflected by the current efforts of the American College of Physicians, the ABIM, and others to redesign the training and the certification requirements of internists. As this supplement demonstrates, the Society of Hospital Medicine will be intimately involved in this process.

After this auspicious start, subsequent issues will include articles in the following categories. Original research articles will report results of randomized controlled trials, evaluations of diagnostic tests, prospective cohort studies, casecontrol studies, and high‐quality observational studies. We are interested in publishing both quantitative and qualitative research. Review articles, especially those targeting the hospital medicine core competencies, are eagerly sought. We also seek descriptions of interventions that transform hospital care delivery in the hospital. For example, accounts of the implementation of quality‐improvement projects and outcomes, including barriers that were overcome or that blocked implementation, would be invaluable to hospitalists throughout the country. Clinical conundrums should describe clinical cases that present diagnostic dilemmas or involve issues of medical errors. To facilitate the professional development of hospitalists, we seek articles focused on their professional development in community, academic, and administrative settings. Examples of leadership topics are managing physician performance, leading and managing change, and self‐evaluation. Teaching tips or descriptions of educational programs or curricula also are desired. For researchers, potential topics include descriptions of specific techniques used for surveys, meta‐analyses, economic evaluations, and statistical analyses.Penetrating point manuscripts, those that go beyond the cutting edge to explain the next potential breakthrough or intervention in the developing field of hospital medicine, may be authored by thought leaders inside and outside the health care field as well as by hospitalists with novel ideas. Equally vital, I want to share the illuminating perspectives of physicians, patients, and families of patients as they reflect on the experience of being in the hospitalhospitalists can enlighten us through their handoffs, and patients and their families can inform us about their view from the hospital bed.

Finally, never forget that this is your journal. Let me know what you like and what changes you think can make it better. Please e‐mail your suggestions, comments, criticisms, and ideas to us at JHMeditor@ hospitalmedicine.org. This is your chance to help shape the practice of hospital medicine and the future of hospital care. I look forward to your guidance. Together we can expand our knowledge and continue to grow in our careers.

An emergency room resident once was instructing a medical student in how to place a nasogastric tube in order to evaluate a patient with melena and postural hypotension. When the tube came to a stop, the student connected a syringe to the tube and aspirated. Then, to the consternation of the resident, the student yanked out the tube as soon as he saw blood flowing into the syringe. Why'd you do that? the surprised resident asked. There's blood down there! came the quick reply.

Like that medical student, hospital‐based physicianshospitalists, geriatricians, and othersmay miss the boat when caring for hospitalized older patients. Hospitals are full, and they're filled largely with older patients. These patients, like those who are younger, generally want to be treated and sent home. Older patients, however, frequently pose specific challenges. They may talk and move more slowly, stay longer, and be more likely to die. They more often need help in caring for themselves, and many have lost the support necessary to remain at home, making it difficult for them to return there. In short, older patients often need more care and more time.

It may be tempting to ignore the challenges that arise in caring for older patients. An avoidance strategy is expedient, at least in the short term. Ultimately, however, ignoring the challenges of caring for older patients will prove no wiser than yanking the nasogastric tube. Instead, we can recognize the challenges and use this opportunity to learn to improve their care.

This article describes the state of the science in hospital care for older patients, opportunities awaiting those who care for these patients, and barriers to seizing those opportunities. I conclude with five recommendations for physicians who care for hospitalized older patients.

STATE OF THE SCIENCE

Older patients shape hospital medicine and will determine its future. In 2002 the 12% of the population age 65 years or older accounted for roughly 50% of all hospitalizations unrelated to childbirth.1, 2 Hospital admissions of older persons will balloon as the number of persons older than age 65 rises by a million a year, increasing from 13% of the population today to 21% by 2030.2

Older persons in hospitals pose substantial clinical challenges. Many have multiple comorbid diseases and virtually all have complex medical regimens.1, 35 Many have cognitive impairment or dementia, often accompanied by delirium, which hinder communication and can lead to behaviors that require extra attention and impede diagnostic tests and treatment.611 Some have difficulty walking and caring for themselves, and a third leave the hospital without having recovered to their baseline level of function, with those age 85 years or older at highest risk for this decline independent of the reason for admission.1215 These characteristics increase the care, resources, and staff time older patients need, prolong their stays, and increase their hospital costs beyond those expected for their diagnosis.16 They also are at higher risk for iatrogenic complications, death, and rehospitalization,1720 and the risk of errors may be increased by frequent transitions in providers and sites of care.2125 Older persons require greater assistance at home, and yet they have often lost much of the support needed to live at home.10, 13, 20

Despite the magnitude of these challenges, we know surprisingly little about how best to care for hospitalized older persons, especially those older than age 75. The evidence base for treatment of specific common diseases is inadequate. The very old are underrepresented in clinical trials,26, 27 and the majority of older patients with common conditions such as heart failure may not meet the enrollment criteria for clinical trials.28 Thus, what is known about treating diseases in younger patients may be extrapolated to determine treatments in older persons based only on a leap of faith, which may be misguided.29, 30 In fact, the efficacy of conventional treatments for common conditions (e.g., acute myocardial infarction and hypertension) may diminish with age,31, 32 indicating that clinical trials targeted specifically to older patients may be necessary.

Despite the dearth of evidence about the management of common diseases in hospitalized older patients, hospital‐based geriatricians have developed substantial high‐grade evidence about the prevention of two geriatric syndromes, functional disability and delirium. The incidence of both syndromes can be reduced (without increasing hospital or health care costs) by multicomponent interventions that include comprehensive assessment, targeted treatment, and environmental modification to promote independence and safety.3, 3335 Moreover, the randomized trials that evaluated these interventions have provided models for how other innovations by hospital‐based physicians can be evaluated. Despite the evidence that these approaches are effective and either cost saving or cost neutral, these models have not been widely adopted.36

Many challenges in the prevention and management of geriatric syndromes in the hospital remain. For example, sophisticated approaches to the management of delirium are disappointingonce delirium has developed, intensive state‐of‐the‐art approaches to its management are no more effective than standard care in shortening its duration or ameliorating its sequelae.37, 38 The indiscriminate use of indwelling urinary catheters is decried, but there is no evidence that their use is declining, even in patients without an indication for catheterization.3942 Malnutrition and falls can be prevented and depression treated in patients outside the hospital,4345 but it is unclear whether these maladies can be prevented or treated effectively in hospitalized elders. Finally, intriguing evidence suggests that geriatric syndromes and their sequelae may be prevented and outcomes improved by caring for patients at home whenever possible, bringing intensive nursing and physician care into the home without some of the adverse effects of hospitalization.46

The physician workforce is not prepared to provide optimal care to hospitalized older persons. Few hospitalists or other hospital‐based physicians have received more than minimal training in geriatric medicine, and few geriatricians practice extensively in the hospital. At the same time that the ranks of physicians who consider themselves hospitalists have been expanding by 1000 or more a year in the United States, the number of certified geriatricians has been decreasing as hundreds decide each year not to renew their certificates.47, 48 Fewer than 300 geriatricians complete training each year and enter the workforce, and most new geriatricians practice in ambulatory or long‐term‐care settings. Wald's study in this issue indicates the paucity of geriatricians in hospital medicine (with the apparently single exception of the Mayo Clinic's Hospital Internal Medicine Group) and a relative lack of interest among hospitalists in developing knowledge about the effective and efficient treatment of older persons, in particular.49

OPPORTUNITIES

Opportunities to improve the care of hospitalized older patients arise from the state of the science in their care and from the common ground that hospitalists and geriatricians share. The older patients of both hospitalists and geriatricians are seriously ill, with annual mortality rates of 20%30% for patients with common conditions such as myocardial infarction or colon cancer and mortality rates of 50% or higher for patients with dementia or severe disability.5, 5053 We should view the care of our patients in the context of their prognoses,5, 54 recognizing that patients' preferences for the goals, style, and site of care vary widely.55, 56 The substantial association of mortality with geriatric syndromes such as disability, dementia, delirium, and depressionan association that is independent of pathophysiologic indicators of disease severitysuggests that substantial benefits may accrue by targeting interventions to the prevention or amelioration of these syndromes.5, 9, 10, 53, 57, 58

Hospitalists and geriatricians also share the perspective of working in complex systems in which the effectiveness, efficiency, and safety of care depend on system functions as well as on their technical expertise as individuals.5961 Together, and with colleagues in other disciplines, they may redesign how hospitals and the systems around them work to reduce errors, increase attention to aspects of care that are easily overlooked, and improve patient outcomes.

BARRIERS

Hospitalists and geriatricians face barriers to improving care for hospitalized older patients. First, gaps in knowledge limit the capacity to provide the care and achieve the outcomes desired. Fundamental discoveries in clinical science are needed to prevent or treat geriatric syndromes, to treat common diseases in the very old, and to put into practice what is known. Addressing these gaps in knowledge will require a sustained effort that spans methods and disciplines.

Second, the dominant reductionist paradigm values discovery of the mechanism of disease over discovery of ways to manage illness effectively and efficiently.6267 Similarly, diagnostic tests and therapies based on beliefs about the mechanism of diseasefor example, PET scans in persons with memory disorders and chemotherapy in persons with refractory cancersare pursued aggressively and paid handsomely, whereas efforts to reduce errors or improve continuity of care receive little attention or reward. The challenges of caring for hospitalized older patients will require advances on both fronts: in our knowledge of the pathogenesis of disorders that have proven resistant to current therapies (such as delirium) and in our knowledge of how to structure clinical care that engages patients and families and achieves desired outcomes effectively, consistently, and efficiently.

The structure and styles of our practices provide the third challenge. Hospitalists pride themselves on their efficient management of patients while maintaining or improving patient outcomes. A focus on efficient management can, however, lead to an assembly‐line approach, turning each patient into a series of do‐order‐call‐check tasks to get the patient out of the hospital as quickly as possible. This approach has advantages but may also blind physicians to the scope and complexity of issues that arise in caring for the very old through the course of an illness that often extends beyond hospitalization.25 Geriatricians pride themselves on their comprehensive management of patients, gathering clinical information from many sources (especially in the many patients with cognitive impairment), exploring and articulating goals of care, and assessing self‐care and neurologic, psychological, and social domains in addition to conventional pathophysiology. Yet too often, geriatricians are not available in hospitals, and as Wald found, they have rarely been integrated into hospitalist groups.

FIVE RECOMMENDATIONS FOR HOSPITALISTS AND GERIATRICIANS

I conclude with five recommendations for hospital‐based physicians who care for older patients and for geriatricians. First, step back, look at your patients, and note their predicament in its full complexity. Once hospitalists start looking for cognitive impairment, weakness, and difficulty walking and the difficulty of finding a good situation after leaving the hospital, it will be easy to see these problems. And once geriatricians start looking at why their patients are going into the hospital and what happens to them, it will be easy to see the need to become engaged. Seeing the full range of patients' problems won't address them, but we certainly won't address them if we don't look.

Second, learn what is known about how best to care for the aged and integrate this learning into your hospital practice. For hospitalists, learning how to identify each patient's goals of care, what works to prevent delirium and promote mobility, which drugs to avoid and which doses to modify, and how to access resources to help patients and families achieve their goals after they leave the hospital will benefit older patients. Pocket and PDA resources to extend learning are readily available.68 For geriatricians, learning how to avoid hospitalization (especially when resources can be mobilized to provide a hospital at home), how to work within the timeframe of hospitalization, and what current disease‐specific management strategies have been shown to be effective in the aged will benefit their patients. Maintaining the distinction between what is believed and what is known on the basis of high‐quality evidence will enhance learning and decrease the risk of stubbornly pursuing harmful or wasteful practices. This is especially important in situations where the evidence is weak and opinions are strong.

Third, to provide the best care for our older patients, we must embrace aging, not deny it. Most hospitalized older patients, and most patients of geriatricians, will decline and die in a few years. The inevitability of these outcomes may tempt us either to abandon our incurable patients or to focus single‐mindedly on treatable problems one at a time, rather than on the interplay of multiple problems in an individual person. Either choice is mistaken. Although we are powerless to prevent decline and death in the long run, we have a tremendous capacity to delay and ameliorate decline, to enhance comfort and joy, to protect from harm, and, often, to delay death.

Fourth, ask questions about what you do not know or understand. The risk, of course, is that your curiosity will be sparked, possibly slowing you in completing the myriad tasks to be donea risk worth taking. Will ACE inhibitors and beta‐blockers benefit patients with heart failure without systolic dysfunction? Why do so many older patients become delirious, and are features of hospitalization catalyzing the effects of disease in causing delirium? Why do we continue to send cognitively impaired patients home without scheduled follow‐up and with instruction sheets they cannot read, and how can we change the system to prevent this? If you cannot find answers to your questions grounded in strong evidence, maintain your skepticism about answers given easily.

Finally, consider discovering the answers to some of your questions. Part of the excitement of caring for the very old is that we have so much to learn and that what we do learn can be so powerful.

An emergency room resident once was instructing a medical student in how to place a nasogastric tube in order to evaluate a patient with melena and postural hypotension. When the tube came to a stop, the student connected a syringe to the tube and aspirated. Then, to the consternation of the resident, the student yanked out the tube as soon as he saw blood flowing into the syringe. Why'd you do that? the surprised resident asked. There's blood down there! came the quick reply.

Like that medical student, hospital‐based physicianshospitalists, geriatricians, and othersmay miss the boat when caring for hospitalized older patients. Hospitals are full, and they're filled largely with older patients. These patients, like those who are younger, generally want to be treated and sent home. Older patients, however, frequently pose specific challenges. They may talk and move more slowly, stay longer, and be more likely to die. They more often need help in caring for themselves, and many have lost the support necessary to remain at home, making it difficult for them to return there. In short, older patients often need more care and more time.

It may be tempting to ignore the challenges that arise in caring for older patients. An avoidance strategy is expedient, at least in the short term. Ultimately, however, ignoring the challenges of caring for older patients will prove no wiser than yanking the nasogastric tube. Instead, we can recognize the challenges and use this opportunity to learn to improve their care.

This article describes the state of the science in hospital care for older patients, opportunities awaiting those who care for these patients, and barriers to seizing those opportunities. I conclude with five recommendations for physicians who care for hospitalized older patients.

STATE OF THE SCIENCE

Older patients shape hospital medicine and will determine its future. In 2002 the 12% of the population age 65 years or older accounted for roughly 50% of all hospitalizations unrelated to childbirth.1, 2 Hospital admissions of older persons will balloon as the number of persons older than age 65 rises by a million a year, increasing from 13% of the population today to 21% by 2030.2

Older persons in hospitals pose substantial clinical challenges. Many have multiple comorbid diseases and virtually all have complex medical regimens.1, 35 Many have cognitive impairment or dementia, often accompanied by delirium, which hinder communication and can lead to behaviors that require extra attention and impede diagnostic tests and treatment.611 Some have difficulty walking and caring for themselves, and a third leave the hospital without having recovered to their baseline level of function, with those age 85 years or older at highest risk for this decline independent of the reason for admission.1215 These characteristics increase the care, resources, and staff time older patients need, prolong their stays, and increase their hospital costs beyond those expected for their diagnosis.16 They also are at higher risk for iatrogenic complications, death, and rehospitalization,1720 and the risk of errors may be increased by frequent transitions in providers and sites of care.2125 Older persons require greater assistance at home, and yet they have often lost much of the support needed to live at home.10, 13, 20

Despite the magnitude of these challenges, we know surprisingly little about how best to care for hospitalized older persons, especially those older than age 75. The evidence base for treatment of specific common diseases is inadequate. The very old are underrepresented in clinical trials,26, 27 and the majority of older patients with common conditions such as heart failure may not meet the enrollment criteria for clinical trials.28 Thus, what is known about treating diseases in younger patients may be extrapolated to determine treatments in older persons based only on a leap of faith, which may be misguided.29, 30 In fact, the efficacy of conventional treatments for common conditions (e.g., acute myocardial infarction and hypertension) may diminish with age,31, 32 indicating that clinical trials targeted specifically to older patients may be necessary.

Despite the dearth of evidence about the management of common diseases in hospitalized older patients, hospital‐based geriatricians have developed substantial high‐grade evidence about the prevention of two geriatric syndromes, functional disability and delirium. The incidence of both syndromes can be reduced (without increasing hospital or health care costs) by multicomponent interventions that include comprehensive assessment, targeted treatment, and environmental modification to promote independence and safety.3, 3335 Moreover, the randomized trials that evaluated these interventions have provided models for how other innovations by hospital‐based physicians can be evaluated. Despite the evidence that these approaches are effective and either cost saving or cost neutral, these models have not been widely adopted.36

Many challenges in the prevention and management of geriatric syndromes in the hospital remain. For example, sophisticated approaches to the management of delirium are disappointingonce delirium has developed, intensive state‐of‐the‐art approaches to its management are no more effective than standard care in shortening its duration or ameliorating its sequelae.37, 38 The indiscriminate use of indwelling urinary catheters is decried, but there is no evidence that their use is declining, even in patients without an indication for catheterization.3942 Malnutrition and falls can be prevented and depression treated in patients outside the hospital,4345 but it is unclear whether these maladies can be prevented or treated effectively in hospitalized elders. Finally, intriguing evidence suggests that geriatric syndromes and their sequelae may be prevented and outcomes improved by caring for patients at home whenever possible, bringing intensive nursing and physician care into the home without some of the adverse effects of hospitalization.46

The physician workforce is not prepared to provide optimal care to hospitalized older persons. Few hospitalists or other hospital‐based physicians have received more than minimal training in geriatric medicine, and few geriatricians practice extensively in the hospital. At the same time that the ranks of physicians who consider themselves hospitalists have been expanding by 1000 or more a year in the United States, the number of certified geriatricians has been decreasing as hundreds decide each year not to renew their certificates.47, 48 Fewer than 300 geriatricians complete training each year and enter the workforce, and most new geriatricians practice in ambulatory or long‐term‐care settings. Wald's study in this issue indicates the paucity of geriatricians in hospital medicine (with the apparently single exception of the Mayo Clinic's Hospital Internal Medicine Group) and a relative lack of interest among hospitalists in developing knowledge about the effective and efficient treatment of older persons, in particular.49

OPPORTUNITIES

Opportunities to improve the care of hospitalized older patients arise from the state of the science in their care and from the common ground that hospitalists and geriatricians share. The older patients of both hospitalists and geriatricians are seriously ill, with annual mortality rates of 20%30% for patients with common conditions such as myocardial infarction or colon cancer and mortality rates of 50% or higher for patients with dementia or severe disability.5, 5053 We should view the care of our patients in the context of their prognoses,5, 54 recognizing that patients' preferences for the goals, style, and site of care vary widely.55, 56 The substantial association of mortality with geriatric syndromes such as disability, dementia, delirium, and depressionan association that is independent of pathophysiologic indicators of disease severitysuggests that substantial benefits may accrue by targeting interventions to the prevention or amelioration of these syndromes.5, 9, 10, 53, 57, 58

Hospitalists and geriatricians also share the perspective of working in complex systems in which the effectiveness, efficiency, and safety of care depend on system functions as well as on their technical expertise as individuals.5961 Together, and with colleagues in other disciplines, they may redesign how hospitals and the systems around them work to reduce errors, increase attention to aspects of care that are easily overlooked, and improve patient outcomes.

BARRIERS

Hospitalists and geriatricians face barriers to improving care for hospitalized older patients. First, gaps in knowledge limit the capacity to provide the care and achieve the outcomes desired. Fundamental discoveries in clinical science are needed to prevent or treat geriatric syndromes, to treat common diseases in the very old, and to put into practice what is known. Addressing these gaps in knowledge will require a sustained effort that spans methods and disciplines.

Second, the dominant reductionist paradigm values discovery of the mechanism of disease over discovery of ways to manage illness effectively and efficiently.6267 Similarly, diagnostic tests and therapies based on beliefs about the mechanism of diseasefor example, PET scans in persons with memory disorders and chemotherapy in persons with refractory cancersare pursued aggressively and paid handsomely, whereas efforts to reduce errors or improve continuity of care receive little attention or reward. The challenges of caring for hospitalized older patients will require advances on both fronts: in our knowledge of the pathogenesis of disorders that have proven resistant to current therapies (such as delirium) and in our knowledge of how to structure clinical care that engages patients and families and achieves desired outcomes effectively, consistently, and efficiently.

The structure and styles of our practices provide the third challenge. Hospitalists pride themselves on their efficient management of patients while maintaining or improving patient outcomes. A focus on efficient management can, however, lead to an assembly‐line approach, turning each patient into a series of do‐order‐call‐check tasks to get the patient out of the hospital as quickly as possible. This approach has advantages but may also blind physicians to the scope and complexity of issues that arise in caring for the very old through the course of an illness that often extends beyond hospitalization.25 Geriatricians pride themselves on their comprehensive management of patients, gathering clinical information from many sources (especially in the many patients with cognitive impairment), exploring and articulating goals of care, and assessing self‐care and neurologic, psychological, and social domains in addition to conventional pathophysiology. Yet too often, geriatricians are not available in hospitals, and as Wald found, they have rarely been integrated into hospitalist groups.

FIVE RECOMMENDATIONS FOR HOSPITALISTS AND GERIATRICIANS

I conclude with five recommendations for hospital‐based physicians who care for older patients and for geriatricians. First, step back, look at your patients, and note their predicament in its full complexity. Once hospitalists start looking for cognitive impairment, weakness, and difficulty walking and the difficulty of finding a good situation after leaving the hospital, it will be easy to see these problems. And once geriatricians start looking at why their patients are going into the hospital and what happens to them, it will be easy to see the need to become engaged. Seeing the full range of patients' problems won't address them, but we certainly won't address them if we don't look.

Second, learn what is known about how best to care for the aged and integrate this learning into your hospital practice. For hospitalists, learning how to identify each patient's goals of care, what works to prevent delirium and promote mobility, which drugs to avoid and which doses to modify, and how to access resources to help patients and families achieve their goals after they leave the hospital will benefit older patients. Pocket and PDA resources to extend learning are readily available.68 For geriatricians, learning how to avoid hospitalization (especially when resources can be mobilized to provide a hospital at home), how to work within the timeframe of hospitalization, and what current disease‐specific management strategies have been shown to be effective in the aged will benefit their patients. Maintaining the distinction between what is believed and what is known on the basis of high‐quality evidence will enhance learning and decrease the risk of stubbornly pursuing harmful or wasteful practices. This is especially important in situations where the evidence is weak and opinions are strong.

Third, to provide the best care for our older patients, we must embrace aging, not deny it. Most hospitalized older patients, and most patients of geriatricians, will decline and die in a few years. The inevitability of these outcomes may tempt us either to abandon our incurable patients or to focus single‐mindedly on treatable problems one at a time, rather than on the interplay of multiple problems in an individual person. Either choice is mistaken. Although we are powerless to prevent decline and death in the long run, we have a tremendous capacity to delay and ameliorate decline, to enhance comfort and joy, to protect from harm, and, often, to delay death.

Fourth, ask questions about what you do not know or understand. The risk, of course, is that your curiosity will be sparked, possibly slowing you in completing the myriad tasks to be donea risk worth taking. Will ACE inhibitors and beta‐blockers benefit patients with heart failure without systolic dysfunction? Why do so many older patients become delirious, and are features of hospitalization catalyzing the effects of disease in causing delirium? Why do we continue to send cognitively impaired patients home without scheduled follow‐up and with instruction sheets they cannot read, and how can we change the system to prevent this? If you cannot find answers to your questions grounded in strong evidence, maintain your skepticism about answers given easily.

Finally, consider discovering the answers to some of your questions. Part of the excitement of caring for the very old is that we have so much to learn and that what we do learn can be so powerful.

An emergency room resident once was instructing a medical student in how to place a nasogastric tube in order to evaluate a patient with melena and postural hypotension. When the tube came to a stop, the student connected a syringe to the tube and aspirated. Then, to the consternation of the resident, the student yanked out the tube as soon as he saw blood flowing into the syringe. Why'd you do that? the surprised resident asked. There's blood down there! came the quick reply.

Like that medical student, hospital‐based physicianshospitalists, geriatricians, and othersmay miss the boat when caring for hospitalized older patients. Hospitals are full, and they're filled largely with older patients. These patients, like those who are younger, generally want to be treated and sent home. Older patients, however, frequently pose specific challenges. They may talk and move more slowly, stay longer, and be more likely to die. They more often need help in caring for themselves, and many have lost the support necessary to remain at home, making it difficult for them to return there. In short, older patients often need more care and more time.

It may be tempting to ignore the challenges that arise in caring for older patients. An avoidance strategy is expedient, at least in the short term. Ultimately, however, ignoring the challenges of caring for older patients will prove no wiser than yanking the nasogastric tube. Instead, we can recognize the challenges and use this opportunity to learn to improve their care.

This article describes the state of the science in hospital care for older patients, opportunities awaiting those who care for these patients, and barriers to seizing those opportunities. I conclude with five recommendations for physicians who care for hospitalized older patients.

STATE OF THE SCIENCE

Older patients shape hospital medicine and will determine its future. In 2002 the 12% of the population age 65 years or older accounted for roughly 50% of all hospitalizations unrelated to childbirth.1, 2 Hospital admissions of older persons will balloon as the number of persons older than age 65 rises by a million a year, increasing from 13% of the population today to 21% by 2030.2

Older persons in hospitals pose substantial clinical challenges. Many have multiple comorbid diseases and virtually all have complex medical regimens.1, 35 Many have cognitive impairment or dementia, often accompanied by delirium, which hinder communication and can lead to behaviors that require extra attention and impede diagnostic tests and treatment.611 Some have difficulty walking and caring for themselves, and a third leave the hospital without having recovered to their baseline level of function, with those age 85 years or older at highest risk for this decline independent of the reason for admission.1215 These characteristics increase the care, resources, and staff time older patients need, prolong their stays, and increase their hospital costs beyond those expected for their diagnosis.16 They also are at higher risk for iatrogenic complications, death, and rehospitalization,1720 and the risk of errors may be increased by frequent transitions in providers and sites of care.2125 Older persons require greater assistance at home, and yet they have often lost much of the support needed to live at home.10, 13, 20

Despite the magnitude of these challenges, we know surprisingly little about how best to care for hospitalized older persons, especially those older than age 75. The evidence base for treatment of specific common diseases is inadequate. The very old are underrepresented in clinical trials,26, 27 and the majority of older patients with common conditions such as heart failure may not meet the enrollment criteria for clinical trials.28 Thus, what is known about treating diseases in younger patients may be extrapolated to determine treatments in older persons based only on a leap of faith, which may be misguided.29, 30 In fact, the efficacy of conventional treatments for common conditions (e.g., acute myocardial infarction and hypertension) may diminish with age,31, 32 indicating that clinical trials targeted specifically to older patients may be necessary.

Despite the dearth of evidence about the management of common diseases in hospitalized older patients, hospital‐based geriatricians have developed substantial high‐grade evidence about the prevention of two geriatric syndromes, functional disability and delirium. The incidence of both syndromes can be reduced (without increasing hospital or health care costs) by multicomponent interventions that include comprehensive assessment, targeted treatment, and environmental modification to promote independence and safety.3, 3335 Moreover, the randomized trials that evaluated these interventions have provided models for how other innovations by hospital‐based physicians can be evaluated. Despite the evidence that these approaches are effective and either cost saving or cost neutral, these models have not been widely adopted.36

Many challenges in the prevention and management of geriatric syndromes in the hospital remain. For example, sophisticated approaches to the management of delirium are disappointingonce delirium has developed, intensive state‐of‐the‐art approaches to its management are no more effective than standard care in shortening its duration or ameliorating its sequelae.37, 38 The indiscriminate use of indwelling urinary catheters is decried, but there is no evidence that their use is declining, even in patients without an indication for catheterization.3942 Malnutrition and falls can be prevented and depression treated in patients outside the hospital,4345 but it is unclear whether these maladies can be prevented or treated effectively in hospitalized elders. Finally, intriguing evidence suggests that geriatric syndromes and their sequelae may be prevented and outcomes improved by caring for patients at home whenever possible, bringing intensive nursing and physician care into the home without some of the adverse effects of hospitalization.46

The physician workforce is not prepared to provide optimal care to hospitalized older persons. Few hospitalists or other hospital‐based physicians have received more than minimal training in geriatric medicine, and few geriatricians practice extensively in the hospital. At the same time that the ranks of physicians who consider themselves hospitalists have been expanding by 1000 or more a year in the United States, the number of certified geriatricians has been decreasing as hundreds decide each year not to renew their certificates.47, 48 Fewer than 300 geriatricians complete training each year and enter the workforce, and most new geriatricians practice in ambulatory or long‐term‐care settings. Wald's study in this issue indicates the paucity of geriatricians in hospital medicine (with the apparently single exception of the Mayo Clinic's Hospital Internal Medicine Group) and a relative lack of interest among hospitalists in developing knowledge about the effective and efficient treatment of older persons, in particular.49

OPPORTUNITIES

Opportunities to improve the care of hospitalized older patients arise from the state of the science in their care and from the common ground that hospitalists and geriatricians share. The older patients of both hospitalists and geriatricians are seriously ill, with annual mortality rates of 20%30% for patients with common conditions such as myocardial infarction or colon cancer and mortality rates of 50% or higher for patients with dementia or severe disability.5, 5053 We should view the care of our patients in the context of their prognoses,5, 54 recognizing that patients' preferences for the goals, style, and site of care vary widely.55, 56 The substantial association of mortality with geriatric syndromes such as disability, dementia, delirium, and depressionan association that is independent of pathophysiologic indicators of disease severitysuggests that substantial benefits may accrue by targeting interventions to the prevention or amelioration of these syndromes.5, 9, 10, 53, 57, 58

Hospitalists and geriatricians also share the perspective of working in complex systems in which the effectiveness, efficiency, and safety of care depend on system functions as well as on their technical expertise as individuals.5961 Together, and with colleagues in other disciplines, they may redesign how hospitals and the systems around them work to reduce errors, increase attention to aspects of care that are easily overlooked, and improve patient outcomes.

BARRIERS

Hospitalists and geriatricians face barriers to improving care for hospitalized older patients. First, gaps in knowledge limit the capacity to provide the care and achieve the outcomes desired. Fundamental discoveries in clinical science are needed to prevent or treat geriatric syndromes, to treat common diseases in the very old, and to put into practice what is known. Addressing these gaps in knowledge will require a sustained effort that spans methods and disciplines.

Second, the dominant reductionist paradigm values discovery of the mechanism of disease over discovery of ways to manage illness effectively and efficiently.6267 Similarly, diagnostic tests and therapies based on beliefs about the mechanism of diseasefor example, PET scans in persons with memory disorders and chemotherapy in persons with refractory cancersare pursued aggressively and paid handsomely, whereas efforts to reduce errors or improve continuity of care receive little attention or reward. The challenges of caring for hospitalized older patients will require advances on both fronts: in our knowledge of the pathogenesis of disorders that have proven resistant to current therapies (such as delirium) and in our knowledge of how to structure clinical care that engages patients and families and achieves desired outcomes effectively, consistently, and efficiently.

The structure and styles of our practices provide the third challenge. Hospitalists pride themselves on their efficient management of patients while maintaining or improving patient outcomes. A focus on efficient management can, however, lead to an assembly‐line approach, turning each patient into a series of do‐order‐call‐check tasks to get the patient out of the hospital as quickly as possible. This approach has advantages but may also blind physicians to the scope and complexity of issues that arise in caring for the very old through the course of an illness that often extends beyond hospitalization.25 Geriatricians pride themselves on their comprehensive management of patients, gathering clinical information from many sources (especially in the many patients with cognitive impairment), exploring and articulating goals of care, and assessing self‐care and neurologic, psychological, and social domains in addition to conventional pathophysiology. Yet too often, geriatricians are not available in hospitals, and as Wald found, they have rarely been integrated into hospitalist groups.

FIVE RECOMMENDATIONS FOR HOSPITALISTS AND GERIATRICIANS

I conclude with five recommendations for hospital‐based physicians who care for older patients and for geriatricians. First, step back, look at your patients, and note their predicament in its full complexity. Once hospitalists start looking for cognitive impairment, weakness, and difficulty walking and the difficulty of finding a good situation after leaving the hospital, it will be easy to see these problems. And once geriatricians start looking at why their patients are going into the hospital and what happens to them, it will be easy to see the need to become engaged. Seeing the full range of patients' problems won't address them, but we certainly won't address them if we don't look.

Second, learn what is known about how best to care for the aged and integrate this learning into your hospital practice. For hospitalists, learning how to identify each patient's goals of care, what works to prevent delirium and promote mobility, which drugs to avoid and which doses to modify, and how to access resources to help patients and families achieve their goals after they leave the hospital will benefit older patients. Pocket and PDA resources to extend learning are readily available.68 For geriatricians, learning how to avoid hospitalization (especially when resources can be mobilized to provide a hospital at home), how to work within the timeframe of hospitalization, and what current disease‐specific management strategies have been shown to be effective in the aged will benefit their patients. Maintaining the distinction between what is believed and what is known on the basis of high‐quality evidence will enhance learning and decrease the risk of stubbornly pursuing harmful or wasteful practices. This is especially important in situations where the evidence is weak and opinions are strong.

Third, to provide the best care for our older patients, we must embrace aging, not deny it. Most hospitalized older patients, and most patients of geriatricians, will decline and die in a few years. The inevitability of these outcomes may tempt us either to abandon our incurable patients or to focus single‐mindedly on treatable problems one at a time, rather than on the interplay of multiple problems in an individual person. Either choice is mistaken. Although we are powerless to prevent decline and death in the long run, we have a tremendous capacity to delay and ameliorate decline, to enhance comfort and joy, to protect from harm, and, often, to delay death.

Fourth, ask questions about what you do not know or understand. The risk, of course, is that your curiosity will be sparked, possibly slowing you in completing the myriad tasks to be donea risk worth taking. Will ACE inhibitors and beta‐blockers benefit patients with heart failure without systolic dysfunction? Why do so many older patients become delirious, and are features of hospitalization catalyzing the effects of disease in causing delirium? Why do we continue to send cognitively impaired patients home without scheduled follow‐up and with instruction sheets they cannot read, and how can we change the system to prevent this? If you cannot find answers to your questions grounded in strong evidence, maintain your skepticism about answers given easily.

Finally, consider discovering the answers to some of your questions. Part of the excitement of caring for the very old is that we have so much to learn and that what we do learn can be so powerful.

For many of you, the hospital has by now become your home away from home. You spend a great deal of time there. You know your way from the ER to the ICU and the morgue. You've learned the hierarchy of who can tell whom what to do, and when it's your turn to do so, you can rattle off an impressive string of acronyms and medspeak in rapid fire. And then, there's that white coat that gives you added stature and authority.

We, the patients, family members, and other concerned visitors, are babes in the woods in this setting. If we've been lucky in life heretofore, we find that visiting a hospital can be like stepping off a hijacked plane into a foreign country we never planned to visit. Few things look familiar. We don't fully understand what we're being made to do. We don't speak the language well enough to communicate with those surrounding us, and we're not certain how friendly they are or what might provoke them into hurting us or our loved ones.

The signs we see don't mean much to usthe few words we recognize tend to scare us more than anything else. We don't know how to interpret the various uniforms people are wearing, other than the white coats we're all familiar with. Quite a few of the busy people moving with confidence around us have the aura of authority figures, no matter what they're wearing. When it's our turn, they focus on us or our loved ones, freely taking samples of blood, attaching instruments, probing private body parts, and asking intensely personal questions. But they don't really say much at all directly to us, and they don't seem interested in a lot of what we try to tell them. We're left feeling confused, humiliated, and somewhat stupid. Obviously, we're not astute enough to figure out what they want us to tell them. Why couldn't we remember everything we'd eaten or taken in the last 24 hours? We failed the test, and the consequences could be life threatening.

I am not exaggerating the situation. Last winter, my husbandwho had some chronic health issues but was still able to hold down a demanding job and carry on a fairly normal lifewas taken down by a nasty anonymous virus that attacked several major organ systems. We thought it was a bad case of the flu. I only took him to the emergency room because his weakness failed to pass in a few days and listening to him struggle to breathe scared me badly. It was the doctor's answering service (not the doctor on call, who never returned my call) that advised me to take him to the ER.

Within the first 24 hours at the hospital, I lost all ability to communicate with my husband when he was sedated and hooked up to a ventilator. For the next three and a half weeks, he lay unconscious, struggling for survival, and I lived on what his doctors and nurses chose to tell me. And they weren't saying a lot that I could make sense of.

For one thing, as his condition worsened he was in the care of eight specialists. Each would tell me something different, and I wasn't equipped to put all these pieces of information together in a meaningful way. For example, early on, I heard the following statements from different doctors all in the same day: He has a virus. His lungs are in really bad shape, and he could die. There are some indications he had a heart attack. His kidneys are failing, and he's going to need dialysis.

I knew that viruses can kill people. I knew that my husband was considered a high risk for heart attacks and that his lungs had been compromised by years of smoking. But I could not fathom how all of these things were related, let alone what was causing the kidneys to fail at this particular time. It took a very long timedays, maybe a week or morebefore I had both the courage and opportunity to pester my husband's infectious disease specialist with so many questions that he explained that the virus had attacked the heart, lungs, and kidneys and that most of what we were seeing was the aftermath of that battle.

This same doctor also helped me by shedding light on why the brief reports I was receiving from different specialists often seemed, to my layman's ears, to contradict each other. (One day the cardiologist told me, His heart's basically in pretty good shape. Yet, a scant hour later, the pulmonary specialist informed me, He's getting worse. He could die.) Each specialist, he told me, tends to focus on the area of the body he or she specializes in, not the patient's overall condition, and their comments reflect that narrow focus. How long would it have taken me to figure that out, if not for this man?

A lot of the frustration I felt as I chased after valuable tidbits of information about my husband's condition could have been alleviated with just a few words. Some days it would have been enough just to have had my anguish acknowledged with something as simple as I know you're frustrated and tired of hearing that we have no clear answers. Believe me, we want a better understanding of what's happening to your husband, too.

Aside from having to assemble the comments my husband's many doctors gave me into a comprehensive picture, I grew very weary of trying to catch these men and women as they came through on their rounds. Some routinely came through the ICU before the start of visiting hours each day. A few had no discernable routine at allI was as likely to encounter them at 8:30 p.m. as at 3 p.m. or 10 a.m. Yet if I was not at my husband's bedside or in the ICU waiting room, I'd get no report from them that day. This system forces the family to forgo any semblance of a normal life. In my case, there were no other family members with which to trade off this vigil, so I missed doctors whenever I went home or out of the room for a bite to eat.

I do not believe that our local hospital is unusual in any of these respects. In talking to people in other parts of the country who've been through a hospital experience, I have heard similar complaints voiced over and over.

And I do not question the quality of the medical care my husband received. As far as I know, everyone involved did their best to save him. Sadly, they did not succeed. Doug died on March 16, 2005, after five and a half weeks in the ICU. Many very good‐hearted, caring people worked on him during that time.

But some, though pleasant, didn't go out of their way to help me one iota. The day my husband started waking up after 25 days in a comalike state, I was in a funk and had found excuses to stay home until midafternoon, figuring there'd be no change in his condition. When I finally dragged myself into his room and spoke to him, I was astonished to see him react with a very slight head movement. Overjoyed, I immediately informed his nurse that he had responded to me, and she replied with a smile, Yes, I know. He's been responsive all day!

Now, these people knew that I'd been hovering at his bedside for 25 days, anxiously waiting for him to wake up, pestering them about why he wasn't and asking what was wrong. Did no one think this development worth a phone call to me?

Since my husband's death, I have heard that some hospitals have patient advocates and hospitalists, but my impression is that a fairly small percentage of hospitals have invested in these types of positions. And I question how well one or two such people in a hospital full of sick patients can help everyone who needs their services. It was hard enough for me to connect with the one woman in our hospital who, during the short time my husband seemed to be on the way to recovering, could help to arrange his transfer to a long‐term care facility.

My point is, the system that is routinely followed in most hospitalsthe system that determines doctors' routines, the system that causes health care workers to tend to treat patients and their families more like objects than human beingsdoes not do service to those it was set up to serve.

My point is, when it's everyone's responsibility to communicate with a patient's family, it winds up being no one's responsibility. Hospitals need to assign this responsibility to a specific person when a critically ill patient is in the care of a team of specialists.

My point is, please think of us wandering lost and scared in that foreign land you call home the next time you encounter one of us there.

For many of you, the hospital has by now become your home away from home. You spend a great deal of time there. You know your way from the ER to the ICU and the morgue. You've learned the hierarchy of who can tell whom what to do, and when it's your turn to do so, you can rattle off an impressive string of acronyms and medspeak in rapid fire. And then, there's that white coat that gives you added stature and authority.

We, the patients, family members, and other concerned visitors, are babes in the woods in this setting. If we've been lucky in life heretofore, we find that visiting a hospital can be like stepping off a hijacked plane into a foreign country we never planned to visit. Few things look familiar. We don't fully understand what we're being made to do. We don't speak the language well enough to communicate with those surrounding us, and we're not certain how friendly they are or what might provoke them into hurting us or our loved ones.

The signs we see don't mean much to usthe few words we recognize tend to scare us more than anything else. We don't know how to interpret the various uniforms people are wearing, other than the white coats we're all familiar with. Quite a few of the busy people moving with confidence around us have the aura of authority figures, no matter what they're wearing. When it's our turn, they focus on us or our loved ones, freely taking samples of blood, attaching instruments, probing private body parts, and asking intensely personal questions. But they don't really say much at all directly to us, and they don't seem interested in a lot of what we try to tell them. We're left feeling confused, humiliated, and somewhat stupid. Obviously, we're not astute enough to figure out what they want us to tell them. Why couldn't we remember everything we'd eaten or taken in the last 24 hours? We failed the test, and the consequences could be life threatening.

I am not exaggerating the situation. Last winter, my husbandwho had some chronic health issues but was still able to hold down a demanding job and carry on a fairly normal lifewas taken down by a nasty anonymous virus that attacked several major organ systems. We thought it was a bad case of the flu. I only took him to the emergency room because his weakness failed to pass in a few days and listening to him struggle to breathe scared me badly. It was the doctor's answering service (not the doctor on call, who never returned my call) that advised me to take him to the ER.

Within the first 24 hours at the hospital, I lost all ability to communicate with my husband when he was sedated and hooked up to a ventilator. For the next three and a half weeks, he lay unconscious, struggling for survival, and I lived on what his doctors and nurses chose to tell me. And they weren't saying a lot that I could make sense of.

For one thing, as his condition worsened he was in the care of eight specialists. Each would tell me something different, and I wasn't equipped to put all these pieces of information together in a meaningful way. For example, early on, I heard the following statements from different doctors all in the same day: He has a virus. His lungs are in really bad shape, and he could die. There are some indications he had a heart attack. His kidneys are failing, and he's going to need dialysis.

I knew that viruses can kill people. I knew that my husband was considered a high risk for heart attacks and that his lungs had been compromised by years of smoking. But I could not fathom how all of these things were related, let alone what was causing the kidneys to fail at this particular time. It took a very long timedays, maybe a week or morebefore I had both the courage and opportunity to pester my husband's infectious disease specialist with so many questions that he explained that the virus had attacked the heart, lungs, and kidneys and that most of what we were seeing was the aftermath of that battle.

This same doctor also helped me by shedding light on why the brief reports I was receiving from different specialists often seemed, to my layman's ears, to contradict each other. (One day the cardiologist told me, His heart's basically in pretty good shape. Yet, a scant hour later, the pulmonary specialist informed me, He's getting worse. He could die.) Each specialist, he told me, tends to focus on the area of the body he or she specializes in, not the patient's overall condition, and their comments reflect that narrow focus. How long would it have taken me to figure that out, if not for this man?

A lot of the frustration I felt as I chased after valuable tidbits of information about my husband's condition could have been alleviated with just a few words. Some days it would have been enough just to have had my anguish acknowledged with something as simple as I know you're frustrated and tired of hearing that we have no clear answers. Believe me, we want a better understanding of what's happening to your husband, too.

Aside from having to assemble the comments my husband's many doctors gave me into a comprehensive picture, I grew very weary of trying to catch these men and women as they came through on their rounds. Some routinely came through the ICU before the start of visiting hours each day. A few had no discernable routine at allI was as likely to encounter them at 8:30 p.m. as at 3 p.m. or 10 a.m. Yet if I was not at my husband's bedside or in the ICU waiting room, I'd get no report from them that day. This system forces the family to forgo any semblance of a normal life. In my case, there were no other family members with which to trade off this vigil, so I missed doctors whenever I went home or out of the room for a bite to eat.

I do not believe that our local hospital is unusual in any of these respects. In talking to people in other parts of the country who've been through a hospital experience, I have heard similar complaints voiced over and over.

And I do not question the quality of the medical care my husband received. As far as I know, everyone involved did their best to save him. Sadly, they did not succeed. Doug died on March 16, 2005, after five and a half weeks in the ICU. Many very good‐hearted, caring people worked on him during that time.

But some, though pleasant, didn't go out of their way to help me one iota. The day my husband started waking up after 25 days in a comalike state, I was in a funk and had found excuses to stay home until midafternoon, figuring there'd be no change in his condition. When I finally dragged myself into his room and spoke to him, I was astonished to see him react with a very slight head movement. Overjoyed, I immediately informed his nurse that he had responded to me, and she replied with a smile, Yes, I know. He's been responsive all day!

Now, these people knew that I'd been hovering at his bedside for 25 days, anxiously waiting for him to wake up, pestering them about why he wasn't and asking what was wrong. Did no one think this development worth a phone call to me?

Since my husband's death, I have heard that some hospitals have patient advocates and hospitalists, but my impression is that a fairly small percentage of hospitals have invested in these types of positions. And I question how well one or two such people in a hospital full of sick patients can help everyone who needs their services. It was hard enough for me to connect with the one woman in our hospital who, during the short time my husband seemed to be on the way to recovering, could help to arrange his transfer to a long‐term care facility.

My point is, the system that is routinely followed in most hospitalsthe system that determines doctors' routines, the system that causes health care workers to tend to treat patients and their families more like objects than human beingsdoes not do service to those it was set up to serve.

My point is, when it's everyone's responsibility to communicate with a patient's family, it winds up being no one's responsibility. Hospitals need to assign this responsibility to a specific person when a critically ill patient is in the care of a team of specialists.

My point is, please think of us wandering lost and scared in that foreign land you call home the next time you encounter one of us there.

For many of you, the hospital has by now become your home away from home. You spend a great deal of time there. You know your way from the ER to the ICU and the morgue. You've learned the hierarchy of who can tell whom what to do, and when it's your turn to do so, you can rattle off an impressive string of acronyms and medspeak in rapid fire. And then, there's that white coat that gives you added stature and authority.

We, the patients, family members, and other concerned visitors, are babes in the woods in this setting. If we've been lucky in life heretofore, we find that visiting a hospital can be like stepping off a hijacked plane into a foreign country we never planned to visit. Few things look familiar. We don't fully understand what we're being made to do. We don't speak the language well enough to communicate with those surrounding us, and we're not certain how friendly they are or what might provoke them into hurting us or our loved ones.

The signs we see don't mean much to usthe few words we recognize tend to scare us more than anything else. We don't know how to interpret the various uniforms people are wearing, other than the white coats we're all familiar with. Quite a few of the busy people moving with confidence around us have the aura of authority figures, no matter what they're wearing. When it's our turn, they focus on us or our loved ones, freely taking samples of blood, attaching instruments, probing private body parts, and asking intensely personal questions. But they don't really say much at all directly to us, and they don't seem interested in a lot of what we try to tell them. We're left feeling confused, humiliated, and somewhat stupid. Obviously, we're not astute enough to figure out what they want us to tell them. Why couldn't we remember everything we'd eaten or taken in the last 24 hours? We failed the test, and the consequences could be life threatening.

I am not exaggerating the situation. Last winter, my husbandwho had some chronic health issues but was still able to hold down a demanding job and carry on a fairly normal lifewas taken down by a nasty anonymous virus that attacked several major organ systems. We thought it was a bad case of the flu. I only took him to the emergency room because his weakness failed to pass in a few days and listening to him struggle to breathe scared me badly. It was the doctor's answering service (not the doctor on call, who never returned my call) that advised me to take him to the ER.

Within the first 24 hours at the hospital, I lost all ability to communicate with my husband when he was sedated and hooked up to a ventilator. For the next three and a half weeks, he lay unconscious, struggling for survival, and I lived on what his doctors and nurses chose to tell me. And they weren't saying a lot that I could make sense of.

For one thing, as his condition worsened he was in the care of eight specialists. Each would tell me something different, and I wasn't equipped to put all these pieces of information together in a meaningful way. For example, early on, I heard the following statements from different doctors all in the same day: He has a virus. His lungs are in really bad shape, and he could die. There are some indications he had a heart attack. His kidneys are failing, and he's going to need dialysis.

I knew that viruses can kill people. I knew that my husband was considered a high risk for heart attacks and that his lungs had been compromised by years of smoking. But I could not fathom how all of these things were related, let alone what was causing the kidneys to fail at this particular time. It took a very long timedays, maybe a week or morebefore I had both the courage and opportunity to pester my husband's infectious disease specialist with so many questions that he explained that the virus had attacked the heart, lungs, and kidneys and that most of what we were seeing was the aftermath of that battle.

This same doctor also helped me by shedding light on why the brief reports I was receiving from different specialists often seemed, to my layman's ears, to contradict each other. (One day the cardiologist told me, His heart's basically in pretty good shape. Yet, a scant hour later, the pulmonary specialist informed me, He's getting worse. He could die.) Each specialist, he told me, tends to focus on the area of the body he or she specializes in, not the patient's overall condition, and their comments reflect that narrow focus. How long would it have taken me to figure that out, if not for this man?

A lot of the frustration I felt as I chased after valuable tidbits of information about my husband's condition could have been alleviated with just a few words. Some days it would have been enough just to have had my anguish acknowledged with something as simple as I know you're frustrated and tired of hearing that we have no clear answers. Believe me, we want a better understanding of what's happening to your husband, too.

Aside from having to assemble the comments my husband's many doctors gave me into a comprehensive picture, I grew very weary of trying to catch these men and women as they came through on their rounds. Some routinely came through the ICU before the start of visiting hours each day. A few had no discernable routine at allI was as likely to encounter them at 8:30 p.m. as at 3 p.m. or 10 a.m. Yet if I was not at my husband's bedside or in the ICU waiting room, I'd get no report from them that day. This system forces the family to forgo any semblance of a normal life. In my case, there were no other family members with which to trade off this vigil, so I missed doctors whenever I went home or out of the room for a bite to eat.

I do not believe that our local hospital is unusual in any of these respects. In talking to people in other parts of the country who've been through a hospital experience, I have heard similar complaints voiced over and over.

And I do not question the quality of the medical care my husband received. As far as I know, everyone involved did their best to save him. Sadly, they did not succeed. Doug died on March 16, 2005, after five and a half weeks in the ICU. Many very good‐hearted, caring people worked on him during that time.

But some, though pleasant, didn't go out of their way to help me one iota. The day my husband started waking up after 25 days in a comalike state, I was in a funk and had found excuses to stay home until midafternoon, figuring there'd be no change in his condition. When I finally dragged myself into his room and spoke to him, I was astonished to see him react with a very slight head movement. Overjoyed, I immediately informed his nurse that he had responded to me, and she replied with a smile, Yes, I know. He's been responsive all day!

Now, these people knew that I'd been hovering at his bedside for 25 days, anxiously waiting for him to wake up, pestering them about why he wasn't and asking what was wrong. Did no one think this development worth a phone call to me?

Since my husband's death, I have heard that some hospitals have patient advocates and hospitalists, but my impression is that a fairly small percentage of hospitals have invested in these types of positions. And I question how well one or two such people in a hospital full of sick patients can help everyone who needs their services. It was hard enough for me to connect with the one woman in our hospital who, during the short time my husband seemed to be on the way to recovering, could help to arrange his transfer to a long‐term care facility.

My point is, the system that is routinely followed in most hospitalsthe system that determines doctors' routines, the system that causes health care workers to tend to treat patients and their families more like objects than human beingsdoes not do service to those it was set up to serve.

My point is, when it's everyone's responsibility to communicate with a patient's family, it winds up being no one's responsibility. Hospitals need to assign this responsibility to a specific person when a critically ill patient is in the care of a team of specialists.

My point is, please think of us wandering lost and scared in that foreign land you call home the next time you encounter one of us there.

A 49‐year‐old man presented with 2 days of chills, fever, anorexia, and increased cough and dyspnea. The patient had a history of chronic obstructive pulmonary disease (COPD) and noted that his cough and dyspnea had increased above normal for several days. He was now dyspneic with minimal activity and had slept at a 45‐degree incline the night prior to evaluation due to dyspnea. He noted less improvement than usual with the use of his metered dose inhaler. His cough was occasionally productive of small amounts of white phlegm. He had vomited once. During a coughing episode the patient experienced a sudden onset of sharp right upper quadrant abdominal pain that worsened with coughing and sudden position changes. The patient denied a prior history of abdominal pain or surgery. The patient's last bowel movement was 2 days prior to admission. He denied melena or bright red blood per rectum.

My initial differential diagnosis for this patient's dyspnea and cough is pneumonia, acute exacerbation of COPD, or congestive heart failure. The presence of fever and anorexia increases the likelihood of infectious etiologies, whereas the presence of orthopnea points toward congestive heart failure. Noncardiac processessuch as a large pleural effusion or apical lung diseasecould also cause orthopnea. His abdominal pain could be a result of pneumonia alone (perhaps in the right lower lobe with diaphragmatic irritation), but I am also considering complications of pneumonia such as empyema. Although his abdominal pain, dyspnea, and cough could also be a result of hepatobiliary disease, a perforated viscus, or pancreatitis, we currently have little reason to suspect a direct abdominal etiology. My top diagnosis is community‐acquired pneumonia, perhaps accompanied by pleural effusion.

His medical history was significant for dilated cardiomyopathy and heavy alcohol use. His medications included various meter‐dosed inhalers, bupropion, digoxin, spironolactone, lisinopril, and metoprolol. He had never received corticosteroid therapy and had not previously been hospitalized for COPD‐related problems. He had smoked one pack of cigarettes daily for 40 years.

Heavy alcohol use is associated with an increased risk of several pulmonary infections such as gram‐negative necrotizing pneumonia (classically, Klebsiella pneumoniae), pneumococcal pneumonia, aspiration pneumonia, anaerobic lung abscesses, and tuberculosis. Given his right upper quadrant pain, acute alcoholic hepatitis and alcohol‐related pancreatitis enter the differential. His history of cardiomyopathy makes me consider congestive heart failure as more likely than before, and perhaps his abdominal pain is a result of hepatic congestion from right heart failure. His fever, however, cannot be attributed to cardiac failure. Less likely diagnoses include ischemic conditions related to his cardiomyopathy such as mesenteric ischemia from low perfusion or embolism from a cardiac thrombus. A pulmonary infection remains the most likely diagnosis.

He was an ill‐appearing man in moderate respiratory distress, looking older than his stated age. His temperature was 38.4C, heart rate 129 beats/minute, blood pressure 85/56 mm Hg, respiratory rate 24 breaths/minute, and oxygen saturation 92% on room air. A cardiovascular exam revealed no murmur, gallop, or rub. The jugular venous pulse was not elevated. His lungs were clear to auscultation. Abdominal exam revealed right‐sided abdominal tenderness that appeared to localize to the rectus sheath. Otherwise, the abdomen was soft, with normal bowel sounds and no organomegaly. Rectal examination revealed guaiac negative stool and no focal tenderness. His extremities were normal.

His vital signs are worrisome for impending cardiovascular collapse and shock, possibly due to sepsis. The relatively nonfocal cardiopulmonary exam is surprising given his initial symptoms and makes me wonder if his dyspnea is primarily related to an abdominal process leading to diaphragmatic irritation rather than to a thoracic process. Congestive heart failure seems unlikely given the lack of supportive physical examination findings. His abdominal exam findings are puzzling. Although his abdominal wall tenderness could be benignperhaps from muscular strain or a tear from coughingit could represent a more worrisome process such as infection or a hematoma in the abdominal wall muscles. Mesenteric ischemia is still possible, as the exam is often unimpressive. A hepatic abscess or subphrenic abscess should be considered, as physical exam findings in these conditions can be subtle.

My differential remains relatively unchanged, but I have now put consideration of a hepatic or subphrenic abscess higher on my list. Early empiric broad‐spectrum antibiotics seem necessary.

He had a white blood cell count of 26,700/mL with 92% neutrophils, a hemoglobin of 14.6 g/dL, and a platelet count of 312,000/mL. Sodium was 134 mmol/L, potassium was 4.3 mmol/L, chloride was 94 mmol/L, bicarbonate was 23 mmol/L, blood urea nitrogen was 23 mg/dL, and creatinine was 2.1 mg/dL. The results of the calcium, protein, albumin, and liver function tests were normal. Urinalysis was negative for protein and red blood cells. An electrocardiogram revealed sinus tachycardia. A chest radiograph at admission revealed mild opacities in both lower lobes and the right middle lobe consistent with either atelectasis or pneumonia (Fig. 1). A very small left effusion was also identified.

Figure 1

The additional data reinforce my clinical impression that this process is likely to be infectious. The chest radiograph is consistent with community‐acquired pneumonia, possibly from an atypical pathogen. Given his elevated creatinine, I am also considering a pulmonary‐renal syndrome such as vasculitis, though hematuria was not present. A subphrenic abscess, mesenteric ischemia, or an abdominal wall process (because his abdominal tenderness on exam still needs an explanation) remain possibilities; my suspicion would increase if he does not respond appropriately to therapy for community‐acquired pneumonia.

The clinical team's working diagnosis also was community‐acquired pneumonia. Blood and sputum cultures were obtained, and the patient was treated with intravenous ceftriaxone, azithromycin, and intravenous fluid. By the second day, his creatinine had normalized; however, his hypoxemia had worsened, and he now required supplemental oxygen. His temperature was 39.3C, and his heart rate was 150 beats/minute. The findings of an abdominal ultrasound of the kidneys, spleen, and right upper quadrant were normal.

It is too early to say the patient has failed therapy because a patient can get worse before getting better during the course of antibiotic therapy for community‐acquired pneumonia. Fever, for example, may take up to 7 days to resolve, depending on host factors and the pathogen. Though I typically wait about 72 hours before assuming a patient is not appropriately responding to therapy, the additional information has made me concerned. The degree of tachycardia is significant and warrants an EKG to exclude an arrthymia. I would also repeat the chest radiograph to evaluate for worsening infiltrates or increased pleural effusion.

On the third hospital day, the patient's abdominal pain had decreased with analgesia, but his fever, cough, and dyspnea remained largely unchanged. Antibiotics were changed to intravenous levofloxacin. A repeat chest radiograph revealed elevation of the right hemidiaphragm and bilateral effusions (Fig. 2). An electrocardiogram revealed sinus tachycardia. Blood cultures revealed no growth, and sputum cultures grew oral flora.

Figure 2

A significantly elevated right hemidiaphragm makes me reconsider the diagnosis of simple community‐acquired pneumonia. The differential diagnosis for an elevated hemidiaphragm is best considered by location in relation to the diaphragm. Causes above the diaphragm include rib fracture, atelectasis, pleural thickening, and volume loss of the lung for another reason (e.g., surgery, bronchial obstruction due to tumor or mucus plugging), as well as mimics such as a densely consolidated pneumonia, pulmonary infarction, or a subpulmonary effusion. Diaphragmatic causes include eventration, rupture, phrenic nerve weakness, and intrinsic weakness because of neuromuscular disease (usually bilateral). Causes below the diaphragm that must be considered are subphrenic or liver abscess, liver (and other abdominal) malignancy, pancreatic pseudocyst, and distended bowel. Given the clinical picture, I am focusing below the diaphragmespecially on a possible hepatic or subphrenic abscess (which could be missed on ultrasound) and mimics of it such as dense consolidation or a subpulmonary effusion. Given the lack of response to antibiotics, I need to consider an infection that is not being treated, either because of location (abscess, effusion) or microbiology (tuberculosis, a parasite, a fungus, resistant bacteria). After confirming that the patient has a substantive pleural effusion, he needs a thoracentesis.

On the fourth hospital day, his temperature was 38.8C, and his white blood cell count was 21,000/mL. A right‐sided thoracentesis was performed; approximately 250 cc of fluid was obtained. Pleural fluid analysis revealed bloody fluid, with a white blood cell count of 16,750/mL with 94% neutrophils, 40,000 red blood cells/mL, lactate dehydrogenase of 278 U/L (normal serum value 80200 U/L), protein of 3.7 g/dL, and glucose of 81 mg/dL. A pleural fluid pH was not obtained. A gram stain revealed many white blood cells with no organisms noted. Serum protein was 7.4 g/dL. These results were thought to represent an exudative parapneumonic effusion; levofloxacin and supplemental oxygen were continued.

The pleural fluid appears exudative, but I am not sure this man has a parapneumonic effusion because, despite clinical deterioration, an obvious infiltrate is not seen on interval chest radiography. We must look closely at the fluid because this is a bloody effusion and somewhat atypical for a parapneumonic effusion. Also, the effusion does not appear large enough to explain why he has not improved on the current antibiotics. We should thus reconsider our diagnosis and management. I would obtain additional imaging (such as an abdominal and chest computed tomography [CT]) and perhaps obtain a consultation from the pulmonary team regarding the postulated initial diagnosis of pneumonia with effusion.

On the fifth day of hospitalization, the patient's dyspnea and cough persisted but were improved. His abdominal pain was minimal and felt improved with flatus. Fever continued to 38.8C, and the white blood cell count was 20,000/mL. On examination the patient had decreased breath sounds at the right base and bibasilar crackles. His abdomen was soft, with tenderness in his right upper quadrant only with deep palpation; bowel sounds remained. An ultrasound of the chest was performed to look for a loculated effusion; however, no fluid was identified. The pulmonary consultant thought it likely that the patient had a subpulmonic effusion and recommended CT of the abdomen and chest.

His right upper quadrant tenderness is still unexplained. I would agree with the CT, primarily to evaluate other causes of his elevated diaphragm such as subphrenic or hepatic abscess. For now, I would make no change in antibiotic therapy.

On the sixth hospital day, the patient had an episode of bilious emesis. Chest and abdominal CT revealed collapse of the right middle and lower lobes with a small adjacent effusion, and a 6 6 16 cm abscess intimately opposed to the right lobe of the liver. Extending from the inferior extent of the abscess was a tubular thick‐walled structure connecting to the cecum that was suspicious of a thickened inflamed appendix. There was periappendiceal stranding suggesting inflammation. The small bowel was diffusely dilated up to 4.5 cm, suggesting a small bowel obstruction.

I suspect that his abscess is related to a perforated appendix and that the dilated small bowel is most likely a result of localized irritation of the bowel by the abscess and appendicitis. The collapsed lung is most likely due to local inflammation from the subdiaphragmatic abscess. Treatment should now be changed substantially. I would ask a surgeon to evaluate the patient because the most likely diagnosis is perforated appendicitis with abscess formation.

When the periappendiceal abscess was drained percutaneously, 190 mL of purulent fluid was removed. The cultures were positive for Klebsiella pneumonia, Enterococcus faecalis,and Streptococcus milleri. The patient was given 6 weeks of intravenous antibiotics with improvement in his clinical symptoms. During the interval the findings on his chest radiograph resolved completely. A laproscopic appendectomy 3 months later revealed significant right lower quadrant adhesions. The pathology specimen identified a distorted appendix with regeneration consistent with prior appendicitis. The patient was contacted 4 months after his surgery, and he reported that he was doing well, with no cardiopulmonary or gastrointestinal symptoms.

COMMENTARY

Community‐acquired pneumonia (CAP) is a common cause of acute illness and accounts for nearly 1 million admissions per year in the United States.1 The diagnosis of CAP is made when symptoms including dyspnea, fever, cough, or leukocytosis are present, with confirmation provided by a chest radiograph. Often the diagnosis is clear; however, there is no pathognomonic constellation of signs or symptoms that establish the diagnosis with certainty.2 Many physicians learn that pneumoniaespecially lower‐lobe pneumoniacan lead to abdominal findings such as upper quadrant pain, vomiting, and tenderness to palpation. Conversely, the patient discussed above illustrates that a primary abdominal process can also result in a symptom complex that mimics pneumonia.

The prevalence of CAP coupled with the inherent uncertainty of a clinical diagnosis of CAP leads to an important question: How long is too long before questioning the diagnosis? An analysis of the pneumonia Patient Outcomes Research Trial (PORT) limited to inpatients with CAP examined time to clinical stability. For the majority of patients, abnormal vital signs resolved within 23 days.3 In this study, 29% of patients had severe disease, and not surprisingly, these patients took longer to improve. Using the pneumonia severity index score, which accounts for age, comorbidity, abnormal vital signs, and laboratory data, the patient described in this article would be considered at high risk for death and complication with an estimated mortality of 9%.4 Using a combination of defervescence, resolution of tachycardia, tachypnea, and hypoxemia as markers of clinical stability, a patient like ours should respond within 4 days (with a range of 27 days). On the basis of these dataand the discrepancy between the patient's severe illness and relatively minor pulmonary infiltratesit seems reasonable to have considered this patient as failing CAP therapy as early as the fourth day of hospitalization.

In approximately 10% of hospitalized patients with CAP, the clinical course is protracted.5 When patients do not improve as quickly as expected, the reasons that could explain this should be investigated. In a cohort of 49 patients with CAP who failed therapy the most common reasons for failure to improve were severity of the pneumonia and drug resistance.6 A multicenter study found that the incidence of resistance to penicillin by Streptococcus pneumoniae, the most common bacterial pathogen in CAP, was 30%, with a 4% in vitro resistance rate to ceftriaxone.7 How well in vitro resistance predicts clinical response, however, is unclear. Risk factors for antibiotic resistance include close exposure to children, recent antibiotic use, and recent hospitalization. Immunosuppressive conditions should also be considered in patients who fail to improve. Suppurative complications of pneumoniasuch as empyema, parapneumonic effusion, and lung abscessalso delay recovery.

Another consideration in a patient with what appears to be a nonresolving pneumonia with pleural effusion is that the initial diagnosis is incorrect and the cause is extrathoracic. Pulmonary and cardiac diseases account for more than 90% of effusions, whereas less than 5% of pleural effusions result from intraabdominal causes.8 When should intraabdominal diseases be sought in patients with an effusion, fever, dyspnea, and cough? Light suggests that intraabdominal pathology should be investigated in patients who have pleural effusions without significant parenchymal disease.8 This point is underscored by the experience of our patient, whose chest radiographs showed, despite clinical decline, minimal airspace disease.

Several abdominal entities cause pleural effusion. Pancreatitis, either acute or chronic, with pseudocyst formation is the most common abdominal cause of exudative pleural effusions. Approximately 10% of patients with pancreatic disease will develop effusions, usually left‐sided.9 These left‐sided effusions are also seen in splenic abscesses, usually as a result of endocarditis. Intrahepatic abscess is associated with effusions in 20% of patients.10 A subphrenic abscess, as seen in our patient, is an uncommon cause of exudative pleural effusions. Historically, subphrenic abscesses resulted from a perforated viscus, with ruptured appendicitis the most common cause,11 followed by perforated peptic ulcers and biliary tract disease. With the advent of antibiotics, the causes of subphrenic abscess changed considerably, with the majority of current cases resulting from postsurgical complications.12 The findings of a chest radiograph are abnormal in 80% of patients with subphrenic abscess;1214 an elevated hemidiaphragm and pleural effusion are found in the majority of cases. The symptoms of a subphrenic abscess are nonspecific, and patient's complaints are equally split between predominantly thoracic and predomninantly abdominal complaints.15

Appendicitis, a common disease predominantly of the young, may lead to atypical presentations in older individuals. In a retrospective analysis of 113 patients older than 60 years with appendicitis, 70% presented in an atypical fashion.16 Typical symptoms include right lower quadrant pain, fever, anorexia and a white blood cell count greater than 10,000/mL. Fever was the most frequently absent symptom, seen in only 37% of older patients. In this cohort, approximately one third of older patients waited more than 48 hours prior to presentation. The time between symptom onset and clinical presentation is a strong predictor of perforation risk.17 As in this case, roughly 2% of patients with acute appendicitis will present with perforation and abscess formation.18 In such patients the management is initially conservative. Percutaneous drainage and broad spectrum antibiotics are the treatment of choice, followed by an interval appendectomy in 612 weeks.19 The rationale for delayed surgery is that earlier surgery may disseminate a localized inflammatory process.20

Community‐acquired pneumonia is a more frequent cause of hospital admission than is intraabdominal abscess. Physicians often face the dilemma of when to pursue alternative diagnoses after a patient who is thought to have an atypical presentation of a common disease (ie, CAP) fails to respond to conventional therapy. Although clinicians learn that right upper quadrant pain may be a symptom of pneumonia, our patient revealed that abdominal causes may mimic pneumonia and produce a pleural effusion. Determining whether the primary disease originates above or below the diaphragm is critical to guiding therapy. When patients fail to respond adequately to therapy, clinicians should set a low threshold for deciding to image the abdomen in a patient with modest pulmonary infiltrates, pleural effusion, and abdominal pain.

A 49‐year‐old man presented with 2 days of chills, fever, anorexia, and increased cough and dyspnea. The patient had a history of chronic obstructive pulmonary disease (COPD) and noted that his cough and dyspnea had increased above normal for several days. He was now dyspneic with minimal activity and had slept at a 45‐degree incline the night prior to evaluation due to dyspnea. He noted less improvement than usual with the use of his metered dose inhaler. His cough was occasionally productive of small amounts of white phlegm. He had vomited once. During a coughing episode the patient experienced a sudden onset of sharp right upper quadrant abdominal pain that worsened with coughing and sudden position changes. The patient denied a prior history of abdominal pain or surgery. The patient's last bowel movement was 2 days prior to admission. He denied melena or bright red blood per rectum.

My initial differential diagnosis for this patient's dyspnea and cough is pneumonia, acute exacerbation of COPD, or congestive heart failure. The presence of fever and anorexia increases the likelihood of infectious etiologies, whereas the presence of orthopnea points toward congestive heart failure. Noncardiac processessuch as a large pleural effusion or apical lung diseasecould also cause orthopnea. His abdominal pain could be a result of pneumonia alone (perhaps in the right lower lobe with diaphragmatic irritation), but I am also considering complications of pneumonia such as empyema. Although his abdominal pain, dyspnea, and cough could also be a result of hepatobiliary disease, a perforated viscus, or pancreatitis, we currently have little reason to suspect a direct abdominal etiology. My top diagnosis is community‐acquired pneumonia, perhaps accompanied by pleural effusion.

His medical history was significant for dilated cardiomyopathy and heavy alcohol use. His medications included various meter‐dosed inhalers, bupropion, digoxin, spironolactone, lisinopril, and metoprolol. He had never received corticosteroid therapy and had not previously been hospitalized for COPD‐related problems. He had smoked one pack of cigarettes daily for 40 years.

Heavy alcohol use is associated with an increased risk of several pulmonary infections such as gram‐negative necrotizing pneumonia (classically, Klebsiella pneumoniae), pneumococcal pneumonia, aspiration pneumonia, anaerobic lung abscesses, and tuberculosis. Given his right upper quadrant pain, acute alcoholic hepatitis and alcohol‐related pancreatitis enter the differential. His history of cardiomyopathy makes me consider congestive heart failure as more likely than before, and perhaps his abdominal pain is a result of hepatic congestion from right heart failure. His fever, however, cannot be attributed to cardiac failure. Less likely diagnoses include ischemic conditions related to his cardiomyopathy such as mesenteric ischemia from low perfusion or embolism from a cardiac thrombus. A pulmonary infection remains the most likely diagnosis.

He was an ill‐appearing man in moderate respiratory distress, looking older than his stated age. His temperature was 38.4C, heart rate 129 beats/minute, blood pressure 85/56 mm Hg, respiratory rate 24 breaths/minute, and oxygen saturation 92% on room air. A cardiovascular exam revealed no murmur, gallop, or rub. The jugular venous pulse was not elevated. His lungs were clear to auscultation. Abdominal exam revealed right‐sided abdominal tenderness that appeared to localize to the rectus sheath. Otherwise, the abdomen was soft, with normal bowel sounds and no organomegaly. Rectal examination revealed guaiac negative stool and no focal tenderness. His extremities were normal.

His vital signs are worrisome for impending cardiovascular collapse and shock, possibly due to sepsis. The relatively nonfocal cardiopulmonary exam is surprising given his initial symptoms and makes me wonder if his dyspnea is primarily related to an abdominal process leading to diaphragmatic irritation rather than to a thoracic process. Congestive heart failure seems unlikely given the lack of supportive physical examination findings. His abdominal exam findings are puzzling. Although his abdominal wall tenderness could be benignperhaps from muscular strain or a tear from coughingit could represent a more worrisome process such as infection or a hematoma in the abdominal wall muscles. Mesenteric ischemia is still possible, as the exam is often unimpressive. A hepatic abscess or subphrenic abscess should be considered, as physical exam findings in these conditions can be subtle.

My differential remains relatively unchanged, but I have now put consideration of a hepatic or subphrenic abscess higher on my list. Early empiric broad‐spectrum antibiotics seem necessary.

He had a white blood cell count of 26,700/mL with 92% neutrophils, a hemoglobin of 14.6 g/dL, and a platelet count of 312,000/mL. Sodium was 134 mmol/L, potassium was 4.3 mmol/L, chloride was 94 mmol/L, bicarbonate was 23 mmol/L, blood urea nitrogen was 23 mg/dL, and creatinine was 2.1 mg/dL. The results of the calcium, protein, albumin, and liver function tests were normal. Urinalysis was negative for protein and red blood cells. An electrocardiogram revealed sinus tachycardia. A chest radiograph at admission revealed mild opacities in both lower lobes and the right middle lobe consistent with either atelectasis or pneumonia (Fig. 1). A very small left effusion was also identified.

Figure 1

The additional data reinforce my clinical impression that this process is likely to be infectious. The chest radiograph is consistent with community‐acquired pneumonia, possibly from an atypical pathogen. Given his elevated creatinine, I am also considering a pulmonary‐renal syndrome such as vasculitis, though hematuria was not present. A subphrenic abscess, mesenteric ischemia, or an abdominal wall process (because his abdominal tenderness on exam still needs an explanation) remain possibilities; my suspicion would increase if he does not respond appropriately to therapy for community‐acquired pneumonia.

The clinical team's working diagnosis also was community‐acquired pneumonia. Blood and sputum cultures were obtained, and the patient was treated with intravenous ceftriaxone, azithromycin, and intravenous fluid. By the second day, his creatinine had normalized; however, his hypoxemia had worsened, and he now required supplemental oxygen. His temperature was 39.3C, and his heart rate was 150 beats/minute. The findings of an abdominal ultrasound of the kidneys, spleen, and right upper quadrant were normal.

It is too early to say the patient has failed therapy because a patient can get worse before getting better during the course of antibiotic therapy for community‐acquired pneumonia. Fever, for example, may take up to 7 days to resolve, depending on host factors and the pathogen. Though I typically wait about 72 hours before assuming a patient is not appropriately responding to therapy, the additional information has made me concerned. The degree of tachycardia is significant and warrants an EKG to exclude an arrthymia. I would also repeat the chest radiograph to evaluate for worsening infiltrates or increased pleural effusion.

On the third hospital day, the patient's abdominal pain had decreased with analgesia, but his fever, cough, and dyspnea remained largely unchanged. Antibiotics were changed to intravenous levofloxacin. A repeat chest radiograph revealed elevation of the right hemidiaphragm and bilateral effusions (Fig. 2). An electrocardiogram revealed sinus tachycardia. Blood cultures revealed no growth, and sputum cultures grew oral flora.

Figure 2

A significantly elevated right hemidiaphragm makes me reconsider the diagnosis of simple community‐acquired pneumonia. The differential diagnosis for an elevated hemidiaphragm is best considered by location in relation to the diaphragm. Causes above the diaphragm include rib fracture, atelectasis, pleural thickening, and volume loss of the lung for another reason (e.g., surgery, bronchial obstruction due to tumor or mucus plugging), as well as mimics such as a densely consolidated pneumonia, pulmonary infarction, or a subpulmonary effusion. Diaphragmatic causes include eventration, rupture, phrenic nerve weakness, and intrinsic weakness because of neuromuscular disease (usually bilateral). Causes below the diaphragm that must be considered are subphrenic or liver abscess, liver (and other abdominal) malignancy, pancreatic pseudocyst, and distended bowel. Given the clinical picture, I am focusing below the diaphragmespecially on a possible hepatic or subphrenic abscess (which could be missed on ultrasound) and mimics of it such as dense consolidation or a subpulmonary effusion. Given the lack of response to antibiotics, I need to consider an infection that is not being treated, either because of location (abscess, effusion) or microbiology (tuberculosis, a parasite, a fungus, resistant bacteria). After confirming that the patient has a substantive pleural effusion, he needs a thoracentesis.

On the fourth hospital day, his temperature was 38.8C, and his white blood cell count was 21,000/mL. A right‐sided thoracentesis was performed; approximately 250 cc of fluid was obtained. Pleural fluid analysis revealed bloody fluid, with a white blood cell count of 16,750/mL with 94% neutrophils, 40,000 red blood cells/mL, lactate dehydrogenase of 278 U/L (normal serum value 80200 U/L), protein of 3.7 g/dL, and glucose of 81 mg/dL. A pleural fluid pH was not obtained. A gram stain revealed many white blood cells with no organisms noted. Serum protein was 7.4 g/dL. These results were thought to represent an exudative parapneumonic effusion; levofloxacin and supplemental oxygen were continued.

The pleural fluid appears exudative, but I am not sure this man has a parapneumonic effusion because, despite clinical deterioration, an obvious infiltrate is not seen on interval chest radiography. We must look closely at the fluid because this is a bloody effusion and somewhat atypical for a parapneumonic effusion. Also, the effusion does not appear large enough to explain why he has not improved on the current antibiotics. We should thus reconsider our diagnosis and management. I would obtain additional imaging (such as an abdominal and chest computed tomography [CT]) and perhaps obtain a consultation from the pulmonary team regarding the postulated initial diagnosis of pneumonia with effusion.

On the fifth day of hospitalization, the patient's dyspnea and cough persisted but were improved. His abdominal pain was minimal and felt improved with flatus. Fever continued to 38.8C, and the white blood cell count was 20,000/mL. On examination the patient had decreased breath sounds at the right base and bibasilar crackles. His abdomen was soft, with tenderness in his right upper quadrant only with deep palpation; bowel sounds remained. An ultrasound of the chest was performed to look for a loculated effusion; however, no fluid was identified. The pulmonary consultant thought it likely that the patient had a subpulmonic effusion and recommended CT of the abdomen and chest.

His right upper quadrant tenderness is still unexplained. I would agree with the CT, primarily to evaluate other causes of his elevated diaphragm such as subphrenic or hepatic abscess. For now, I would make no change in antibiotic therapy.

On the sixth hospital day, the patient had an episode of bilious emesis. Chest and abdominal CT revealed collapse of the right middle and lower lobes with a small adjacent effusion, and a 6 6 16 cm abscess intimately opposed to the right lobe of the liver. Extending from the inferior extent of the abscess was a tubular thick‐walled structure connecting to the cecum that was suspicious of a thickened inflamed appendix. There was periappendiceal stranding suggesting inflammation. The small bowel was diffusely dilated up to 4.5 cm, suggesting a small bowel obstruction.

I suspect that his abscess is related to a perforated appendix and that the dilated small bowel is most likely a result of localized irritation of the bowel by the abscess and appendicitis. The collapsed lung is most likely due to local inflammation from the subdiaphragmatic abscess. Treatment should now be changed substantially. I would ask a surgeon to evaluate the patient because the most likely diagnosis is perforated appendicitis with abscess formation.

When the periappendiceal abscess was drained percutaneously, 190 mL of purulent fluid was removed. The cultures were positive for Klebsiella pneumonia, Enterococcus faecalis,and Streptococcus milleri. The patient was given 6 weeks of intravenous antibiotics with improvement in his clinical symptoms. During the interval the findings on his chest radiograph resolved completely. A laproscopic appendectomy 3 months later revealed significant right lower quadrant adhesions. The pathology specimen identified a distorted appendix with regeneration consistent with prior appendicitis. The patient was contacted 4 months after his surgery, and he reported that he was doing well, with no cardiopulmonary or gastrointestinal symptoms.

COMMENTARY

Community‐acquired pneumonia (CAP) is a common cause of acute illness and accounts for nearly 1 million admissions per year in the United States.1 The diagnosis of CAP is made when symptoms including dyspnea, fever, cough, or leukocytosis are present, with confirmation provided by a chest radiograph. Often the diagnosis is clear; however, there is no pathognomonic constellation of signs or symptoms that establish the diagnosis with certainty.2 Many physicians learn that pneumoniaespecially lower‐lobe pneumoniacan lead to abdominal findings such as upper quadrant pain, vomiting, and tenderness to palpation. Conversely, the patient discussed above illustrates that a primary abdominal process can also result in a symptom complex that mimics pneumonia.

The prevalence of CAP coupled with the inherent uncertainty of a clinical diagnosis of CAP leads to an important question: How long is too long before questioning the diagnosis? An analysis of the pneumonia Patient Outcomes Research Trial (PORT) limited to inpatients with CAP examined time to clinical stability. For the majority of patients, abnormal vital signs resolved within 23 days.3 In this study, 29% of patients had severe disease, and not surprisingly, these patients took longer to improve. Using the pneumonia severity index score, which accounts for age, comorbidity, abnormal vital signs, and laboratory data, the patient described in this article would be considered at high risk for death and complication with an estimated mortality of 9%.4 Using a combination of defervescence, resolution of tachycardia, tachypnea, and hypoxemia as markers of clinical stability, a patient like ours should respond within 4 days (with a range of 27 days). On the basis of these dataand the discrepancy between the patient's severe illness and relatively minor pulmonary infiltratesit seems reasonable to have considered this patient as failing CAP therapy as early as the fourth day of hospitalization.

In approximately 10% of hospitalized patients with CAP, the clinical course is protracted.5 When patients do not improve as quickly as expected, the reasons that could explain this should be investigated. In a cohort of 49 patients with CAP who failed therapy the most common reasons for failure to improve were severity of the pneumonia and drug resistance.6 A multicenter study found that the incidence of resistance to penicillin by Streptococcus pneumoniae, the most common bacterial pathogen in CAP, was 30%, with a 4% in vitro resistance rate to ceftriaxone.7 How well in vitro resistance predicts clinical response, however, is unclear. Risk factors for antibiotic resistance include close exposure to children, recent antibiotic use, and recent hospitalization. Immunosuppressive conditions should also be considered in patients who fail to improve. Suppurative complications of pneumoniasuch as empyema, parapneumonic effusion, and lung abscessalso delay recovery.

Another consideration in a patient with what appears to be a nonresolving pneumonia with pleural effusion is that the initial diagnosis is incorrect and the cause is extrathoracic. Pulmonary and cardiac diseases account for more than 90% of effusions, whereas less than 5% of pleural effusions result from intraabdominal causes.8 When should intraabdominal diseases be sought in patients with an effusion, fever, dyspnea, and cough? Light suggests that intraabdominal pathology should be investigated in patients who have pleural effusions without significant parenchymal disease.8 This point is underscored by the experience of our patient, whose chest radiographs showed, despite clinical decline, minimal airspace disease.

Several abdominal entities cause pleural effusion. Pancreatitis, either acute or chronic, with pseudocyst formation is the most common abdominal cause of exudative pleural effusions. Approximately 10% of patients with pancreatic disease will develop effusions, usually left‐sided.9 These left‐sided effusions are also seen in splenic abscesses, usually as a result of endocarditis. Intrahepatic abscess is associated with effusions in 20% of patients.10 A subphrenic abscess, as seen in our patient, is an uncommon cause of exudative pleural effusions. Historically, subphrenic abscesses resulted from a perforated viscus, with ruptured appendicitis the most common cause,11 followed by perforated peptic ulcers and biliary tract disease. With the advent of antibiotics, the causes of subphrenic abscess changed considerably, with the majority of current cases resulting from postsurgical complications.12 The findings of a chest radiograph are abnormal in 80% of patients with subphrenic abscess;1214 an elevated hemidiaphragm and pleural effusion are found in the majority of cases. The symptoms of a subphrenic abscess are nonspecific, and patient's complaints are equally split between predominantly thoracic and predomninantly abdominal complaints.15

Appendicitis, a common disease predominantly of the young, may lead to atypical presentations in older individuals. In a retrospective analysis of 113 patients older than 60 years with appendicitis, 70% presented in an atypical fashion.16 Typical symptoms include right lower quadrant pain, fever, anorexia and a white blood cell count greater than 10,000/mL. Fever was the most frequently absent symptom, seen in only 37% of older patients. In this cohort, approximately one third of older patients waited more than 48 hours prior to presentation. The time between symptom onset and clinical presentation is a strong predictor of perforation risk.17 As in this case, roughly 2% of patients with acute appendicitis will present with perforation and abscess formation.18 In such patients the management is initially conservative. Percutaneous drainage and broad spectrum antibiotics are the treatment of choice, followed by an interval appendectomy in 612 weeks.19 The rationale for delayed surgery is that earlier surgery may disseminate a localized inflammatory process.20

Community‐acquired pneumonia is a more frequent cause of hospital admission than is intraabdominal abscess. Physicians often face the dilemma of when to pursue alternative diagnoses after a patient who is thought to have an atypical presentation of a common disease (ie, CAP) fails to respond to conventional therapy. Although clinicians learn that right upper quadrant pain may be a symptom of pneumonia, our patient revealed that abdominal causes may mimic pneumonia and produce a pleural effusion. Determining whether the primary disease originates above or below the diaphragm is critical to guiding therapy. When patients fail to respond adequately to therapy, clinicians should set a low threshold for deciding to image the abdomen in a patient with modest pulmonary infiltrates, pleural effusion, and abdominal pain.

A 49‐year‐old man presented with 2 days of chills, fever, anorexia, and increased cough and dyspnea. The patient had a history of chronic obstructive pulmonary disease (COPD) and noted that his cough and dyspnea had increased above normal for several days. He was now dyspneic with minimal activity and had slept at a 45‐degree incline the night prior to evaluation due to dyspnea. He noted less improvement than usual with the use of his metered dose inhaler. His cough was occasionally productive of small amounts of white phlegm. He had vomited once. During a coughing episode the patient experienced a sudden onset of sharp right upper quadrant abdominal pain that worsened with coughing and sudden position changes. The patient denied a prior history of abdominal pain or surgery. The patient's last bowel movement was 2 days prior to admission. He denied melena or bright red blood per rectum.

My initial differential diagnosis for this patient's dyspnea and cough is pneumonia, acute exacerbation of COPD, or congestive heart failure. The presence of fever and anorexia increases the likelihood of infectious etiologies, whereas the presence of orthopnea points toward congestive heart failure. Noncardiac processessuch as a large pleural effusion or apical lung diseasecould also cause orthopnea. His abdominal pain could be a result of pneumonia alone (perhaps in the right lower lobe with diaphragmatic irritation), but I am also considering complications of pneumonia such as empyema. Although his abdominal pain, dyspnea, and cough could also be a result of hepatobiliary disease, a perforated viscus, or pancreatitis, we currently have little reason to suspect a direct abdominal etiology. My top diagnosis is community‐acquired pneumonia, perhaps accompanied by pleural effusion.

His medical history was significant for dilated cardiomyopathy and heavy alcohol use. His medications included various meter‐dosed inhalers, bupropion, digoxin, spironolactone, lisinopril, and metoprolol. He had never received corticosteroid therapy and had not previously been hospitalized for COPD‐related problems. He had smoked one pack of cigarettes daily for 40 years.

Heavy alcohol use is associated with an increased risk of several pulmonary infections such as gram‐negative necrotizing pneumonia (classically, Klebsiella pneumoniae), pneumococcal pneumonia, aspiration pneumonia, anaerobic lung abscesses, and tuberculosis. Given his right upper quadrant pain, acute alcoholic hepatitis and alcohol‐related pancreatitis enter the differential. His history of cardiomyopathy makes me consider congestive heart failure as more likely than before, and perhaps his abdominal pain is a result of hepatic congestion from right heart failure. His fever, however, cannot be attributed to cardiac failure. Less likely diagnoses include ischemic conditions related to his cardiomyopathy such as mesenteric ischemia from low perfusion or embolism from a cardiac thrombus. A pulmonary infection remains the most likely diagnosis.

He was an ill‐appearing man in moderate respiratory distress, looking older than his stated age. His temperature was 38.4C, heart rate 129 beats/minute, blood pressure 85/56 mm Hg, respiratory rate 24 breaths/minute, and oxygen saturation 92% on room air. A cardiovascular exam revealed no murmur, gallop, or rub. The jugular venous pulse was not elevated. His lungs were clear to auscultation. Abdominal exam revealed right‐sided abdominal tenderness that appeared to localize to the rectus sheath. Otherwise, the abdomen was soft, with normal bowel sounds and no organomegaly. Rectal examination revealed guaiac negative stool and no focal tenderness. His extremities were normal.

His vital signs are worrisome for impending cardiovascular collapse and shock, possibly due to sepsis. The relatively nonfocal cardiopulmonary exam is surprising given his initial symptoms and makes me wonder if his dyspnea is primarily related to an abdominal process leading to diaphragmatic irritation rather than to a thoracic process. Congestive heart failure seems unlikely given the lack of supportive physical examination findings. His abdominal exam findings are puzzling. Although his abdominal wall tenderness could be benignperhaps from muscular strain or a tear from coughingit could represent a more worrisome process such as infection or a hematoma in the abdominal wall muscles. Mesenteric ischemia is still possible, as the exam is often unimpressive. A hepatic abscess or subphrenic abscess should be considered, as physical exam findings in these conditions can be subtle.

My differential remains relatively unchanged, but I have now put consideration of a hepatic or subphrenic abscess higher on my list. Early empiric broad‐spectrum antibiotics seem necessary.

He had a white blood cell count of 26,700/mL with 92% neutrophils, a hemoglobin of 14.6 g/dL, and a platelet count of 312,000/mL. Sodium was 134 mmol/L, potassium was 4.3 mmol/L, chloride was 94 mmol/L, bicarbonate was 23 mmol/L, blood urea nitrogen was 23 mg/dL, and creatinine was 2.1 mg/dL. The results of the calcium, protein, albumin, and liver function tests were normal. Urinalysis was negative for protein and red blood cells. An electrocardiogram revealed sinus tachycardia. A chest radiograph at admission revealed mild opacities in both lower lobes and the right middle lobe consistent with either atelectasis or pneumonia (Fig. 1). A very small left effusion was also identified.

Figure 1

The additional data reinforce my clinical impression that this process is likely to be infectious. The chest radiograph is consistent with community‐acquired pneumonia, possibly from an atypical pathogen. Given his elevated creatinine, I am also considering a pulmonary‐renal syndrome such as vasculitis, though hematuria was not present. A subphrenic abscess, mesenteric ischemia, or an abdominal wall process (because his abdominal tenderness on exam still needs an explanation) remain possibilities; my suspicion would increase if he does not respond appropriately to therapy for community‐acquired pneumonia.

The clinical team's working diagnosis also was community‐acquired pneumonia. Blood and sputum cultures were obtained, and the patient was treated with intravenous ceftriaxone, azithromycin, and intravenous fluid. By the second day, his creatinine had normalized; however, his hypoxemia had worsened, and he now required supplemental oxygen. His temperature was 39.3C, and his heart rate was 150 beats/minute. The findings of an abdominal ultrasound of the kidneys, spleen, and right upper quadrant were normal.

It is too early to say the patient has failed therapy because a patient can get worse before getting better during the course of antibiotic therapy for community‐acquired pneumonia. Fever, for example, may take up to 7 days to resolve, depending on host factors and the pathogen. Though I typically wait about 72 hours before assuming a patient is not appropriately responding to therapy, the additional information has made me concerned. The degree of tachycardia is significant and warrants an EKG to exclude an arrthymia. I would also repeat the chest radiograph to evaluate for worsening infiltrates or increased pleural effusion.

On the third hospital day, the patient's abdominal pain had decreased with analgesia, but his fever, cough, and dyspnea remained largely unchanged. Antibiotics were changed to intravenous levofloxacin. A repeat chest radiograph revealed elevation of the right hemidiaphragm and bilateral effusions (Fig. 2). An electrocardiogram revealed sinus tachycardia. Blood cultures revealed no growth, and sputum cultures grew oral flora.

Figure 2

A significantly elevated right hemidiaphragm makes me reconsider the diagnosis of simple community‐acquired pneumonia. The differential diagnosis for an elevated hemidiaphragm is best considered by location in relation to the diaphragm. Causes above the diaphragm include rib fracture, atelectasis, pleural thickening, and volume loss of the lung for another reason (e.g., surgery, bronchial obstruction due to tumor or mucus plugging), as well as mimics such as a densely consolidated pneumonia, pulmonary infarction, or a subpulmonary effusion. Diaphragmatic causes include eventration, rupture, phrenic nerve weakness, and intrinsic weakness because of neuromuscular disease (usually bilateral). Causes below the diaphragm that must be considered are subphrenic or liver abscess, liver (and other abdominal) malignancy, pancreatic pseudocyst, and distended bowel. Given the clinical picture, I am focusing below the diaphragmespecially on a possible hepatic or subphrenic abscess (which could be missed on ultrasound) and mimics of it such as dense consolidation or a subpulmonary effusion. Given the lack of response to antibiotics, I need to consider an infection that is not being treated, either because of location (abscess, effusion) or microbiology (tuberculosis, a parasite, a fungus, resistant bacteria). After confirming that the patient has a substantive pleural effusion, he needs a thoracentesis.

On the fourth hospital day, his temperature was 38.8C, and his white blood cell count was 21,000/mL. A right‐sided thoracentesis was performed; approximately 250 cc of fluid was obtained. Pleural fluid analysis revealed bloody fluid, with a white blood cell count of 16,750/mL with 94% neutrophils, 40,000 red blood cells/mL, lactate dehydrogenase of 278 U/L (normal serum value 80200 U/L), protein of 3.7 g/dL, and glucose of 81 mg/dL. A pleural fluid pH was not obtained. A gram stain revealed many white blood cells with no organisms noted. Serum protein was 7.4 g/dL. These results were thought to represent an exudative parapneumonic effusion; levofloxacin and supplemental oxygen were continued.

The pleural fluid appears exudative, but I am not sure this man has a parapneumonic effusion because, despite clinical deterioration, an obvious infiltrate is not seen on interval chest radiography. We must look closely at the fluid because this is a bloody effusion and somewhat atypical for a parapneumonic effusion. Also, the effusion does not appear large enough to explain why he has not improved on the current antibiotics. We should thus reconsider our diagnosis and management. I would obtain additional imaging (such as an abdominal and chest computed tomography [CT]) and perhaps obtain a consultation from the pulmonary team regarding the postulated initial diagnosis of pneumonia with effusion.

On the fifth day of hospitalization, the patient's dyspnea and cough persisted but were improved. His abdominal pain was minimal and felt improved with flatus. Fever continued to 38.8C, and the white blood cell count was 20,000/mL. On examination the patient had decreased breath sounds at the right base and bibasilar crackles. His abdomen was soft, with tenderness in his right upper quadrant only with deep palpation; bowel sounds remained. An ultrasound of the chest was performed to look for a loculated effusion; however, no fluid was identified. The pulmonary consultant thought it likely that the patient had a subpulmonic effusion and recommended CT of the abdomen and chest.

His right upper quadrant tenderness is still unexplained. I would agree with the CT, primarily to evaluate other causes of his elevated diaphragm such as subphrenic or hepatic abscess. For now, I would make no change in antibiotic therapy.

On the sixth hospital day, the patient had an episode of bilious emesis. Chest and abdominal CT revealed collapse of the right middle and lower lobes with a small adjacent effusion, and a 6 6 16 cm abscess intimately opposed to the right lobe of the liver. Extending from the inferior extent of the abscess was a tubular thick‐walled structure connecting to the cecum that was suspicious of a thickened inflamed appendix. There was periappendiceal stranding suggesting inflammation. The small bowel was diffusely dilated up to 4.5 cm, suggesting a small bowel obstruction.

I suspect that his abscess is related to a perforated appendix and that the dilated small bowel is most likely a result of localized irritation of the bowel by the abscess and appendicitis. The collapsed lung is most likely due to local inflammation from the subdiaphragmatic abscess. Treatment should now be changed substantially. I would ask a surgeon to evaluate the patient because the most likely diagnosis is perforated appendicitis with abscess formation.

When the periappendiceal abscess was drained percutaneously, 190 mL of purulent fluid was removed. The cultures were positive for Klebsiella pneumonia, Enterococcus faecalis,and Streptococcus milleri. The patient was given 6 weeks of intravenous antibiotics with improvement in his clinical symptoms. During the interval the findings on his chest radiograph resolved completely. A laproscopic appendectomy 3 months later revealed significant right lower quadrant adhesions. The pathology specimen identified a distorted appendix with regeneration consistent with prior appendicitis. The patient was contacted 4 months after his surgery, and he reported that he was doing well, with no cardiopulmonary or gastrointestinal symptoms.

COMMENTARY

Community‐acquired pneumonia (CAP) is a common cause of acute illness and accounts for nearly 1 million admissions per year in the United States.1 The diagnosis of CAP is made when symptoms including dyspnea, fever, cough, or leukocytosis are present, with confirmation provided by a chest radiograph. Often the diagnosis is clear; however, there is no pathognomonic constellation of signs or symptoms that establish the diagnosis with certainty.2 Many physicians learn that pneumoniaespecially lower‐lobe pneumoniacan lead to abdominal findings such as upper quadrant pain, vomiting, and tenderness to palpation. Conversely, the patient discussed above illustrates that a primary abdominal process can also result in a symptom complex that mimics pneumonia.

The prevalence of CAP coupled with the inherent uncertainty of a clinical diagnosis of CAP leads to an important question: How long is too long before questioning the diagnosis? An analysis of the pneumonia Patient Outcomes Research Trial (PORT) limited to inpatients with CAP examined time to clinical stability. For the majority of patients, abnormal vital signs resolved within 23 days.3 In this study, 29% of patients had severe disease, and not surprisingly, these patients took longer to improve. Using the pneumonia severity index score, which accounts for age, comorbidity, abnormal vital signs, and laboratory data, the patient described in this article would be considered at high risk for death and complication with an estimated mortality of 9%.4 Using a combination of defervescence, resolution of tachycardia, tachypnea, and hypoxemia as markers of clinical stability, a patient like ours should respond within 4 days (with a range of 27 days). On the basis of these dataand the discrepancy between the patient's severe illness and relatively minor pulmonary infiltratesit seems reasonable to have considered this patient as failing CAP therapy as early as the fourth day of hospitalization.

In approximately 10% of hospitalized patients with CAP, the clinical course is protracted.5 When patients do not improve as quickly as expected, the reasons that could explain this should be investigated. In a cohort of 49 patients with CAP who failed therapy the most common reasons for failure to improve were severity of the pneumonia and drug resistance.6 A multicenter study found that the incidence of resistance to penicillin by Streptococcus pneumoniae, the most common bacterial pathogen in CAP, was 30%, with a 4% in vitro resistance rate to ceftriaxone.7 How well in vitro resistance predicts clinical response, however, is unclear. Risk factors for antibiotic resistance include close exposure to children, recent antibiotic use, and recent hospitalization. Immunosuppressive conditions should also be considered in patients who fail to improve. Suppurative complications of pneumoniasuch as empyema, parapneumonic effusion, and lung abscessalso delay recovery.

Another consideration in a patient with what appears to be a nonresolving pneumonia with pleural effusion is that the initial diagnosis is incorrect and the cause is extrathoracic. Pulmonary and cardiac diseases account for more than 90% of effusions, whereas less than 5% of pleural effusions result from intraabdominal causes.8 When should intraabdominal diseases be sought in patients with an effusion, fever, dyspnea, and cough? Light suggests that intraabdominal pathology should be investigated in patients who have pleural effusions without significant parenchymal disease.8 This point is underscored by the experience of our patient, whose chest radiographs showed, despite clinical decline, minimal airspace disease.

Several abdominal entities cause pleural effusion. Pancreatitis, either acute or chronic, with pseudocyst formation is the most common abdominal cause of exudative pleural effusions. Approximately 10% of patients with pancreatic disease will develop effusions, usually left‐sided.9 These left‐sided effusions are also seen in splenic abscesses, usually as a result of endocarditis. Intrahepatic abscess is associated with effusions in 20% of patients.10 A subphrenic abscess, as seen in our patient, is an uncommon cause of exudative pleural effusions. Historically, subphrenic abscesses resulted from a perforated viscus, with ruptured appendicitis the most common cause,11 followed by perforated peptic ulcers and biliary tract disease. With the advent of antibiotics, the causes of subphrenic abscess changed considerably, with the majority of current cases resulting from postsurgical complications.12 The findings of a chest radiograph are abnormal in 80% of patients with subphrenic abscess;1214 an elevated hemidiaphragm and pleural effusion are found in the majority of cases. The symptoms of a subphrenic abscess are nonspecific, and patient's complaints are equally split between predominantly thoracic and predomninantly abdominal complaints.15

Appendicitis, a common disease predominantly of the young, may lead to atypical presentations in older individuals. In a retrospective analysis of 113 patients older than 60 years with appendicitis, 70% presented in an atypical fashion.16 Typical symptoms include right lower quadrant pain, fever, anorexia and a white blood cell count greater than 10,000/mL. Fever was the most frequently absent symptom, seen in only 37% of older patients. In this cohort, approximately one third of older patients waited more than 48 hours prior to presentation. The time between symptom onset and clinical presentation is a strong predictor of perforation risk.17 As in this case, roughly 2% of patients with acute appendicitis will present with perforation and abscess formation.18 In such patients the management is initially conservative. Percutaneous drainage and broad spectrum antibiotics are the treatment of choice, followed by an interval appendectomy in 612 weeks.19 The rationale for delayed surgery is that earlier surgery may disseminate a localized inflammatory process.20

Community‐acquired pneumonia is a more frequent cause of hospital admission than is intraabdominal abscess. Physicians often face the dilemma of when to pursue alternative diagnoses after a patient who is thought to have an atypical presentation of a common disease (ie, CAP) fails to respond to conventional therapy. Although clinicians learn that right upper quadrant pain may be a symptom of pneumonia, our patient revealed that abdominal causes may mimic pneumonia and produce a pleural effusion. Determining whether the primary disease originates above or below the diaphragm is critical to guiding therapy. When patients fail to respond adequately to therapy, clinicians should set a low threshold for deciding to image the abdomen in a patient with modest pulmonary infiltrates, pleural effusion, and abdominal pain.

Gastrointestinal bleeding occurring in hospitalized patients admitted for nongastrointestinal disorders has been extensively studied in intensive care unit patients. However, a systematic study in noncritically ill medical patients has not yet been done. In critically ill patients the incidence of hospital‐acquired gastrointestinal bleeding (GIB) varies from 0.17% to 5%, depending on its definition.16 These bleeding events significantly increase the morbidity and duration of hospitalization.1, 5, 79

Risk factors for bleeding in the intensive care unit include mechanical ventilation, coagulopathy, burns, chronic renal failure, and neurological insults.15 Several studies have found that stress ulcer prophylaxis with histamine‐2 (H2) receptor antagonists, sucralfate, or proton pump inhibitors (PPIs) decreases bleeding in this group of patients, with a relative risk reduction of 29%61%.10, 11 However, use of these drugs outside this high‐risk group has been questioned because of the low overall risk of bleeding.1, 11, 12 Despite their being an unproven benefit in the noncritically ill population, prophylactic H2 antagonists or PPIs are prescribed in an indiscriminant fashion to up to 30%50% of patients admitted to the hospital,13, 14 suggesting that physician preference dictates this practice. To shed light on this issue in noncritically ill patients, we conducted a retrospective casecontrol study in order to identify risk factors that predict hospital‐acquired gastrointestinal bleeding in this group of patients and to assess whether treatment with prophylactic acid suppression was associated with fewer bleeding events. We also sought to characterize the endoscopic lesions in these patients.

MATERIALS AND METHODS

Study Patients

The institutional review board of the Cleveland Clinic Foundation (Cleveland, OH) approved this study. All patients admitted to the General Internal Medicine service between January 1, 1999, and December 31, 2002, were eligible for inclusion. Two types of cases were included: 1) patients admitted for nongastrointestinal illnesses who developed bleeding at least 24 hours after admission and required esophagogastroduodenoscopy (EGD) during hospitalization (designated in‐hospital bleeding), and 2) patients admitted with gastrointestinal bleeding (requiring EGD) who had been hospitalized on the General Medical service during the preceding 4 weeks for a nongastrointestinal illness (designated out‐of‐hospital bleeding). This second group was included to identify risk factors for delayed bleeding that might not be obvious during hospitalization.

Medical records of all General Medicine patients who underwent EGD were reviewed in a standardized fashion (Fig. 1). We excluded patients with documented gastrointestinal complaints (including bleeding) at the time of the index admission or within 24 hours of admission, bleeding in the intensive care unit (ICU) or in another hospital prior to transfer to the General Medicine service, or a history of gastrointestinal bleeding during the month prior to admission. ICU stay prior to General Medicine admission, if not associated with GI bleeding, was not an exclusion criterion for our study.

Figure 1

Controls, also without any acute gastrointestinal symptoms at admission, were randomly matched to cases in a 1:1 ratio by date of admission. We used this liberal matching strategy because any factors matched for would no longer be eligible to be risk factors for bleeding. If more than one control was admitted on the same day as a case, then a random number was used to select the control.

DISCUSSION

Our data suggest that the incidence of hospital‐acquired gastrointestinal bleeding in noncritically ill medical patients is low (approximately 0.4%) and that treatment with anticoagulants or clopidogrel predisposes to this complication. Anticoagulation is a well‐known risk factor for gastrointestinal bleeding, with an estimated odds ratio of 2416; our study confirmed this risk.

Although some studies have questioned the utility of prophylactic acid blockade in the intensive care unit,15 the weight of current evidence supports prophylaxis in selected critically ill patients. In a randomized double‐blind study of 1200 mechanically ventilated patients, the relative risk of gastrointestinal bleeding in patients treated with ranitidine was 0.44 (95% CI 0.210.92 P = .02).11 Many experts discourage indiscriminant use of prophylaxes, even by patients in intensive care units, recommending that it be used only in patients with established risk factors for bleeding.1, 12

Despite the absence of evidence of any benefit of the use of prophylactic acid blockade outside the intensive care unit, this practice is common. In our study, 27.5% of patients who were not on outpatient acid suppression medications (PPIs or H2 antagonists) were started on them on admission to the hospital, presumably as prophylaxes, as we excluded patients admitted for acute gastrointestinal complaints. Other studies have reported prophylaxis rates of 30%50%.13, 14 Many patients started on this prophylaxis during hospitalization go on to take these drugs following discharge, creating an unnecessary economic burden.13, 14 In our study, GI prophylaxis did not appear to prevent hospital‐acquired gastrointestinal bleeding. However, the odds ratio associating the use of prophylactic acid suppression with gastrointestinal bleeding (1.0) was associated with a wide 95% confidence interval (0.42.4), so we cannot exclude the possibility that these medications might provide a relative risk reduction that we were unable to detect. Finally, although gastrointestinal bleeding in the intensive care unit is associated with significant morbidity and mortality,8, 9 we found no evidence to suggest that gastrointestinal bleeding in our patients was associated with poor outcomes.

In interpreting the data from this study, it is important to note that the definition of hospital‐acquired gastrointestinal bleeding in the literature has been inconsistent. Some studies have required that bleeding be hemodynamically significant1, 2, 5, 11a stringent criterion that may be present in only 10%15% of patients with bleeding16whereas other studies defined gastrointestinal bleeding on the basis of occult‐blood‐positive nasogastric aspirates or positive endoscopic findings.7, 15 Because the definition used in the present study required a hard clinical event (EGD), it excluded bleeding events that were considered clinically insignificant by treating physicians. We justified this definition on our belief that any bleeding that warrants invasive evaluation is clinically relevant because it is expensive and puts the patient at some physical risk. Even though some of our patients were diagnosed with GIB without obvious melena or hematemesis (ie, based on stool positivity for occult blood), many of these patients had significant drops in hemoglobin during hospitalization, which, accompanied by occult blood positivity, justified inpatient EGD. We do not believe our definition of GI bleeding was too restrictive, at least for our institution, as physicians at the Cleveland Clinic generally pursue inpatient EGD with clinically apparent gastrointestinal bleeding; we maintain that bleeding that is minor enough not to change management is of limited clinical relevance. Nevertheless, the threshold for EGD at a given institution could affect the rate of EGD for soft indications and the overall prevalence of nosocomial GI bleeding based on our definition.

It also is worth noting that our definition of nosocomial bleeding encompassed some patients with recent hospitalization on the medical service who bled following discharge (15% of cases in this study). This inclusion criterion was chosen because of our concern that the stress of hospitalization might lead to complications even after discharge. We chose an arbitrary postdischarge cutoff of 4 weeks. When we excluded these patients from analysis, the results were similar (data not shown). Although it is possible that we missed some patient who presented to other institutions with GI bleeding following discharge from the Cleveland Clinic, we suspect that the number of such patients was very small based on current referral patterns.

We do not have complete information to determine exactly why patients were on acid‐suppressive therapy prior to admission, but the available data suggest that many had gastroesophageal reflux disease (GERD), PUD, or prior GI bleeding. For this reason, we focused the investigation of the potential efficacy of prophylactic initiation of acid blockade among patients who at presentation were not taking these medications, as prior GERD (or undocumented GIB) leading to chronic use of acid blockade may predispose to subsequent GIB. Although we analyzed only those patients who had newly started taking acid‐suppressive medications, we acknowledge that a few of them may have been started on these medications for other reasons, like chest pain or GERD. However, the evidence suggests that an overwhelming number are started on these medications for the sole purpose of GI prophylaxis.13, 14

Our study was limited by its retrospective casecontrol design. However, because of the low prevalence of hospital‐acquired gastrointestinal bleeding outside the critical care unit, a prospective study would have to enroll thousands of patients in order to generate statistically meaningful results.

In summary, hospital‐acquired gastrointestinal bleeding outside the intensive care unit is uncommon, with an incidence of about 0.4% according to our definition of bleeding. We found no evidence that these bleeding episodes are associated with increased mortality or with occult malignancy. Furthermore, we found no evidence that prophylactic gastric acid suppression prevents these events, and only 41 patients (0.2% of the total cohort) had lesions that might be preventable with gastric acid blockade. We discourage the indiscriminant use of prophylactic acid suppressants in general medical patients.

Gastrointestinal bleeding occurring in hospitalized patients admitted for nongastrointestinal disorders has been extensively studied in intensive care unit patients. However, a systematic study in noncritically ill medical patients has not yet been done. In critically ill patients the incidence of hospital‐acquired gastrointestinal bleeding (GIB) varies from 0.17% to 5%, depending on its definition.16 These bleeding events significantly increase the morbidity and duration of hospitalization.1, 5, 79

Risk factors for bleeding in the intensive care unit include mechanical ventilation, coagulopathy, burns, chronic renal failure, and neurological insults.15 Several studies have found that stress ulcer prophylaxis with histamine‐2 (H2) receptor antagonists, sucralfate, or proton pump inhibitors (PPIs) decreases bleeding in this group of patients, with a relative risk reduction of 29%61%.10, 11 However, use of these drugs outside this high‐risk group has been questioned because of the low overall risk of bleeding.1, 11, 12 Despite their being an unproven benefit in the noncritically ill population, prophylactic H2 antagonists or PPIs are prescribed in an indiscriminant fashion to up to 30%50% of patients admitted to the hospital,13, 14 suggesting that physician preference dictates this practice. To shed light on this issue in noncritically ill patients, we conducted a retrospective casecontrol study in order to identify risk factors that predict hospital‐acquired gastrointestinal bleeding in this group of patients and to assess whether treatment with prophylactic acid suppression was associated with fewer bleeding events. We also sought to characterize the endoscopic lesions in these patients.

MATERIALS AND METHODS

Study Patients

The institutional review board of the Cleveland Clinic Foundation (Cleveland, OH) approved this study. All patients admitted to the General Internal Medicine service between January 1, 1999, and December 31, 2002, were eligible for inclusion. Two types of cases were included: 1) patients admitted for nongastrointestinal illnesses who developed bleeding at least 24 hours after admission and required esophagogastroduodenoscopy (EGD) during hospitalization (designated in‐hospital bleeding), and 2) patients admitted with gastrointestinal bleeding (requiring EGD) who had been hospitalized on the General Medical service during the preceding 4 weeks for a nongastrointestinal illness (designated out‐of‐hospital bleeding). This second group was included to identify risk factors for delayed bleeding that might not be obvious during hospitalization.

Medical records of all General Medicine patients who underwent EGD were reviewed in a standardized fashion (Fig. 1). We excluded patients with documented gastrointestinal complaints (including bleeding) at the time of the index admission or within 24 hours of admission, bleeding in the intensive care unit (ICU) or in another hospital prior to transfer to the General Medicine service, or a history of gastrointestinal bleeding during the month prior to admission. ICU stay prior to General Medicine admission, if not associated with GI bleeding, was not an exclusion criterion for our study.

Figure 1

Controls, also without any acute gastrointestinal symptoms at admission, were randomly matched to cases in a 1:1 ratio by date of admission. We used this liberal matching strategy because any factors matched for would no longer be eligible to be risk factors for bleeding. If more than one control was admitted on the same day as a case, then a random number was used to select the control.