Short Title

The United States spends more on healthcare than any country in the world, and it is widely believed that the Nation could spend less while achieving comparable or better outcomes. The recent debate over healthcare reform in the United States, the large Federal budget deficit in the context of the current economic recession, and the prospect of widening gaps in Medicare funding with the increasing entry of baby boomers into old age suggest that the issue of healthcare cost will remain intense for many years to come. What roles hospitalists will play in the nation's struggle to control health care costs remain to be seen. Six papers in this issue of the Journal of Hospital Medicine discuss issues related to costs, and reflect several of the ways in which hospital medicine can contribute to understanding, and ultimately, controlling healthcare costs.

Two papers, one by Whelan et al.1 examining the costs associated with upper vs. lower GI bleeding and one by Lorch et al.2 examining the costs associated with herpex simplex virus (HSV) infections among neonates with and without congenital abnormalities, focus on epidemiologic determinants of healthcare costs. Such studies can identify subgroups of patients with high costs who may be logical targets for efforts to control costs. One tension in the use of such analyses to control cost is that total cost for any patient group is the product of both the cost per patient and the number of patients falling into each group. In the case of gastrointestinal (GI) bleeding, the surprise compared to past reports is that lower GI bleeding is about as common among hospitalized patients as upper GI bleeding. This may be because pharmacotherapy for conditions that cause upper GI bleeding is reducing the rate at which disease progresses to the point where hospitalization is required. The importance of prevalence is reinforced even in the findings about HSV infection, where despite 2‐ to 3‐fold higher average costs among babies with HSV who have congenital abnormalities, the fact that 90% of babies hospitalized with HSV lack congenital abnormalities implies that the clear majority of costs are due to babies without congenital abnormalities. In seeking strategies to control costs, it is important to pay attention to both the prevalence and cost per case of specific conditions. Because hospitalists are generalist physicians who typically care for few patients with any given diagnosis, the importance of prevalence implies that disease‐specific efforts to control costs may produce smaller total gains than those that cross diseases, such as efforts to improve communication between inpatient and outpatient physicians.

Moreover, the presence of high costs for some condition does not, of course, imply that effective interventions exist to reduce those costs. Two other papers, one by Mudge et al.3 examining a disease management program for heart failure, and one by Go et al.4 examining the effects of hospitalists on the costs of hospitalization for GI bleeding, reinforce the idea that interventions to reduce hospital costs are not always as effective as hoped. Even worse, efforts to control costs can have unintended effects, such as the delays in antimicrobial administration with antimicrobial approval policies that are reported by Winters et al.5 These studies also illustrate that analyses of the effectiveness of interventions can be performed using a variety of experimental designs (eg, the before/after comparison used by Mudge et al,3 and the natural experiments based on assignment of patients to physicians based on day of admission used by Go et al.4 or based on time of day used by Winters et al.)5 The role of hospitalists as clinical leaders in hospitals often places them in positions to design and execute experiments, but the role of hospitalists as astute clinicians who can recognize the presence of natural experiments in their clinical environment can be every bit as powerful in producing valid research designs.

As society seeks strategies to control healthcare costs in the years ahead, it will almost certainly turn to the same general strategies that have been used in the past: bundling services into fixed payments for a prospectively defined episode of care, asking patients to pay more of the costs of care, and simply not paying for, or paying less for, any given type of care. Hospitalists already have dealt with many of these approaches in one form or another. Medicare's prospective payment system and the payment of fixed annual fees for the care of patients in health maintenance organizations have given all hospitalists some exposure to the pressure for lower hospital resources use under prospective payment systems. Proposals for demonstration projects within healthcare reform to study the effects of bundling inpatient and outpatient care or even hospital and professional fees suggest that hospitalists may need to be open to new incentive structures in the years to come. For example, reduced incentives for rapid discharge if costs pushed into the outpatient setting are borne by the hospital, there may be co‐management models if professional and hospital fees are bundled. Increases in patient copayments may also play some role in healthcare reform, and the paper by Ross et al.6 should be a reminder to hospitalists that we may do our patients a great disservice if we fail to recognize the effects of our decisions on their out‐of‐pocket costs. Indeed, while doctors and patient both recognize the importance of discussing out‐of‐pocket costs, they both agree that these discussions rarely occur.7 That such discussions are not reimbursed explicitly suggests one of the many challenges of controlling healthcare costs; if physician payments are decreased to control costs and physicians respond by attempting to see even more patients in any given time period, discussions of important but less urgent issues such as out‐of‐pocket costs seem likely to be reduced. Such dilemmas arise frequently as the healthcare system devises increasingly complex approaches to the control of costs and suggest to many that fundamental reform of the payment and delivery system with greater reliance on integrated health systems paid through full capitation will eventually need to become the nation's approach to healthcare cost containment.8

The United States spends more on healthcare than any country in the world, and it is widely believed that the Nation could spend less while achieving comparable or better outcomes. The recent debate over healthcare reform in the United States, the large Federal budget deficit in the context of the current economic recession, and the prospect of widening gaps in Medicare funding with the increasing entry of baby boomers into old age suggest that the issue of healthcare cost will remain intense for many years to come. What roles hospitalists will play in the nation's struggle to control health care costs remain to be seen. Six papers in this issue of the Journal of Hospital Medicine discuss issues related to costs, and reflect several of the ways in which hospital medicine can contribute to understanding, and ultimately, controlling healthcare costs.

Two papers, one by Whelan et al.1 examining the costs associated with upper vs. lower GI bleeding and one by Lorch et al.2 examining the costs associated with herpex simplex virus (HSV) infections among neonates with and without congenital abnormalities, focus on epidemiologic determinants of healthcare costs. Such studies can identify subgroups of patients with high costs who may be logical targets for efforts to control costs. One tension in the use of such analyses to control cost is that total cost for any patient group is the product of both the cost per patient and the number of patients falling into each group. In the case of gastrointestinal (GI) bleeding, the surprise compared to past reports is that lower GI bleeding is about as common among hospitalized patients as upper GI bleeding. This may be because pharmacotherapy for conditions that cause upper GI bleeding is reducing the rate at which disease progresses to the point where hospitalization is required. The importance of prevalence is reinforced even in the findings about HSV infection, where despite 2‐ to 3‐fold higher average costs among babies with HSV who have congenital abnormalities, the fact that 90% of babies hospitalized with HSV lack congenital abnormalities implies that the clear majority of costs are due to babies without congenital abnormalities. In seeking strategies to control costs, it is important to pay attention to both the prevalence and cost per case of specific conditions. Because hospitalists are generalist physicians who typically care for few patients with any given diagnosis, the importance of prevalence implies that disease‐specific efforts to control costs may produce smaller total gains than those that cross diseases, such as efforts to improve communication between inpatient and outpatient physicians.

Moreover, the presence of high costs for some condition does not, of course, imply that effective interventions exist to reduce those costs. Two other papers, one by Mudge et al.3 examining a disease management program for heart failure, and one by Go et al.4 examining the effects of hospitalists on the costs of hospitalization for GI bleeding, reinforce the idea that interventions to reduce hospital costs are not always as effective as hoped. Even worse, efforts to control costs can have unintended effects, such as the delays in antimicrobial administration with antimicrobial approval policies that are reported by Winters et al.5 These studies also illustrate that analyses of the effectiveness of interventions can be performed using a variety of experimental designs (eg, the before/after comparison used by Mudge et al,3 and the natural experiments based on assignment of patients to physicians based on day of admission used by Go et al.4 or based on time of day used by Winters et al.)5 The role of hospitalists as clinical leaders in hospitals often places them in positions to design and execute experiments, but the role of hospitalists as astute clinicians who can recognize the presence of natural experiments in their clinical environment can be every bit as powerful in producing valid research designs.

As society seeks strategies to control healthcare costs in the years ahead, it will almost certainly turn to the same general strategies that have been used in the past: bundling services into fixed payments for a prospectively defined episode of care, asking patients to pay more of the costs of care, and simply not paying for, or paying less for, any given type of care. Hospitalists already have dealt with many of these approaches in one form or another. Medicare's prospective payment system and the payment of fixed annual fees for the care of patients in health maintenance organizations have given all hospitalists some exposure to the pressure for lower hospital resources use under prospective payment systems. Proposals for demonstration projects within healthcare reform to study the effects of bundling inpatient and outpatient care or even hospital and professional fees suggest that hospitalists may need to be open to new incentive structures in the years to come. For example, reduced incentives for rapid discharge if costs pushed into the outpatient setting are borne by the hospital, there may be co‐management models if professional and hospital fees are bundled. Increases in patient copayments may also play some role in healthcare reform, and the paper by Ross et al.6 should be a reminder to hospitalists that we may do our patients a great disservice if we fail to recognize the effects of our decisions on their out‐of‐pocket costs. Indeed, while doctors and patient both recognize the importance of discussing out‐of‐pocket costs, they both agree that these discussions rarely occur.7 That such discussions are not reimbursed explicitly suggests one of the many challenges of controlling healthcare costs; if physician payments are decreased to control costs and physicians respond by attempting to see even more patients in any given time period, discussions of important but less urgent issues such as out‐of‐pocket costs seem likely to be reduced. Such dilemmas arise frequently as the healthcare system devises increasingly complex approaches to the control of costs and suggest to many that fundamental reform of the payment and delivery system with greater reliance on integrated health systems paid through full capitation will eventually need to become the nation's approach to healthcare cost containment.8

The United States spends more on healthcare than any country in the world, and it is widely believed that the Nation could spend less while achieving comparable or better outcomes. The recent debate over healthcare reform in the United States, the large Federal budget deficit in the context of the current economic recession, and the prospect of widening gaps in Medicare funding with the increasing entry of baby boomers into old age suggest that the issue of healthcare cost will remain intense for many years to come. What roles hospitalists will play in the nation's struggle to control health care costs remain to be seen. Six papers in this issue of the Journal of Hospital Medicine discuss issues related to costs, and reflect several of the ways in which hospital medicine can contribute to understanding, and ultimately, controlling healthcare costs.

Two papers, one by Whelan et al.1 examining the costs associated with upper vs. lower GI bleeding and one by Lorch et al.2 examining the costs associated with herpex simplex virus (HSV) infections among neonates with and without congenital abnormalities, focus on epidemiologic determinants of healthcare costs. Such studies can identify subgroups of patients with high costs who may be logical targets for efforts to control costs. One tension in the use of such analyses to control cost is that total cost for any patient group is the product of both the cost per patient and the number of patients falling into each group. In the case of gastrointestinal (GI) bleeding, the surprise compared to past reports is that lower GI bleeding is about as common among hospitalized patients as upper GI bleeding. This may be because pharmacotherapy for conditions that cause upper GI bleeding is reducing the rate at which disease progresses to the point where hospitalization is required. The importance of prevalence is reinforced even in the findings about HSV infection, where despite 2‐ to 3‐fold higher average costs among babies with HSV who have congenital abnormalities, the fact that 90% of babies hospitalized with HSV lack congenital abnormalities implies that the clear majority of costs are due to babies without congenital abnormalities. In seeking strategies to control costs, it is important to pay attention to both the prevalence and cost per case of specific conditions. Because hospitalists are generalist physicians who typically care for few patients with any given diagnosis, the importance of prevalence implies that disease‐specific efforts to control costs may produce smaller total gains than those that cross diseases, such as efforts to improve communication between inpatient and outpatient physicians.

Moreover, the presence of high costs for some condition does not, of course, imply that effective interventions exist to reduce those costs. Two other papers, one by Mudge et al.3 examining a disease management program for heart failure, and one by Go et al.4 examining the effects of hospitalists on the costs of hospitalization for GI bleeding, reinforce the idea that interventions to reduce hospital costs are not always as effective as hoped. Even worse, efforts to control costs can have unintended effects, such as the delays in antimicrobial administration with antimicrobial approval policies that are reported by Winters et al.5 These studies also illustrate that analyses of the effectiveness of interventions can be performed using a variety of experimental designs (eg, the before/after comparison used by Mudge et al,3 and the natural experiments based on assignment of patients to physicians based on day of admission used by Go et al.4 or based on time of day used by Winters et al.)5 The role of hospitalists as clinical leaders in hospitals often places them in positions to design and execute experiments, but the role of hospitalists as astute clinicians who can recognize the presence of natural experiments in their clinical environment can be every bit as powerful in producing valid research designs.

As society seeks strategies to control healthcare costs in the years ahead, it will almost certainly turn to the same general strategies that have been used in the past: bundling services into fixed payments for a prospectively defined episode of care, asking patients to pay more of the costs of care, and simply not paying for, or paying less for, any given type of care. Hospitalists already have dealt with many of these approaches in one form or another. Medicare's prospective payment system and the payment of fixed annual fees for the care of patients in health maintenance organizations have given all hospitalists some exposure to the pressure for lower hospital resources use under prospective payment systems. Proposals for demonstration projects within healthcare reform to study the effects of bundling inpatient and outpatient care or even hospital and professional fees suggest that hospitalists may need to be open to new incentive structures in the years to come. For example, reduced incentives for rapid discharge if costs pushed into the outpatient setting are borne by the hospital, there may be co‐management models if professional and hospital fees are bundled. Increases in patient copayments may also play some role in healthcare reform, and the paper by Ross et al.6 should be a reminder to hospitalists that we may do our patients a great disservice if we fail to recognize the effects of our decisions on their out‐of‐pocket costs. Indeed, while doctors and patient both recognize the importance of discussing out‐of‐pocket costs, they both agree that these discussions rarely occur.7 That such discussions are not reimbursed explicitly suggests one of the many challenges of controlling healthcare costs; if physician payments are decreased to control costs and physicians respond by attempting to see even more patients in any given time period, discussions of important but less urgent issues such as out‐of‐pocket costs seem likely to be reduced. Such dilemmas arise frequently as the healthcare system devises increasingly complex approaches to the control of costs and suggest to many that fundamental reform of the payment and delivery system with greater reliance on integrated health systems paid through full capitation will eventually need to become the nation's approach to healthcare cost containment.8