Archana Bhaskaran, MD

With dramatically increasing costs of healthcare, it has become increasingly necessary for healthcare providers to demonstrate value in the delivery of care. Porter and Teisberg have strongly advocated that healthcare reform efforts should focus on improving value rather than limiting cost, with value being defined as quality per unit cost.1 However, it has been pointed out that value means different things to different people.2 The biggest challenge in defining value stems mainly from the difficulty in defining quality, because it, too, means vastly different things to different people. Modern medicine is increasingly characterized by multidisciplinary care. With limited or shrinking resources, it will become necessary for individual specialists to describe and articulate, in quantitative terms, their specific contributions to the overall outcome of individual patients.

Previous publications have provided broad descriptions of the value provided by infectious disease (ID) specialists in the domains of sepsis, infection control, outpatient antibiotic therapy, antimicrobial stewardship, and directive care and teaching.3, 4 Studies have also shown the value of ID physicians in specific disease conditions. ID consultation is associated with lower mortality5, 6 and lower relapse rates7 in hospitalized patients with Staphylococcus aureus bacteremia. In another study evaluating the impact of ID consultants, patients seen by ID consultants had longer lengths of hospital stay, longer intensive care unit lengths of stay, and higher antibiotic costs than matched controls not seen by ID consultants.8 It can be argued that a major limitation of the study was that controls were not matched for the ID diagnosis, nor for the causative microorganisms, but it is clear that ID physicians are challenged to demonstrate their contribution to the care of patients.

A unique activity of ID physicians is the management of community‐based parenteral anti‐infective therapy (CoPAT). At Baystate Medical Center, a policy of mandatory ID consultation was instituted for patients leaving hospital on parenteral antibiotics. A study was conducted on the impact of predischarge ID consultation for 44 patients who were not already being followed by the ID service. The study documented change from intravenous (IV) to oral formulation, change of antibiotic choice, and change of dose/duration of treatment in a substantial proportion of patients.9 These are significant changes, but ID consultation contributes more than the themes explored in the study.

The purpose of this study was to evaluate the contribution of ID consultation when consulted for CoPAT, an activity specific to ID practice, in a different institution, and using an expanded definition of medical contribution.

METHODS

The Cleveland Clinic's Department of Infectious Disease has 24 staff physicians and 11 inpatient ID consultative services. These include: 2 solid organ transplant services; a bone marrow transplant and oncologic service; 2 infective endocarditis/cardiac device infection services; an intensive care unit (ICU) service; a bone and joint infection service; a neuroinfection service; and 3 general ID consult services. Consultative services are provided 7 days a week. At the Cleveland Clinic, ID consultation is required prior to discharge on parenteral antibiotic therapy.10, 11 ID consultation for CoPAT usually occurs when the primary service deems the patient is close to being discharged from hospital. This circumstance allows for assessing the specific contribution of ID physicians beyond that of the primary service and other consulting services.

RESULTS

In the time period studied, there were 1326 CPOE requests for ID consultation. The response to the question, CoPAT consult? was Yes for 304, No for 507, and Not sure for 515 requests. Of the 304 consultation requests marked Yes, 41 were excluded. Reasons for exclusion were: no ID consultation note (21), wrong service consulted (8), consultation request placed while the ID service was already following the patient (7), and duplicate consultation request (5). The remaining 263 consultation requests corresponded to 1 or more CoPAT consultation requests for 249 patients (across different hospitalizations). Of the 263 consultation requests, 172 were initial consultations, while the remaining 91 were reconsultations (patients not actively being followed by the ID service, but previously seen during the same hospitalization).

Consultation characteristics are outlined in Table 2. The most common group of infections for which CoPAT was sought was bone and joint infections, accounting for over 20% of the consultation requests. CoPAT consultations were requested a median of 4 days after hospitalization. Patients were discharged from hospital a median of 3 days after they were seen by the ID service. ID consultation did not delay discharge. The ID service usually saw the patient the same day, and followed the patient in hospital for a median of 1 day. There was no difference in hospital days after consult for patients who did not need antibiotics versus those who did.

ID consultation provided value in at least 1 domain in 260 of the 263 consultations. This included optimization of antimicrobial treatment in 84%, significant alteration of patient assessment in 52%, and additional medical care contribution in 71% of consultations. Substantial contributions were made in all domains in both initial consultations and in reconsultations. Specific ID contributions within each of the domains are shown in Figure 1. There was wide overlap of contributions across the 3 domains for individual consultations (Figure 2), with contributions in all domains occurring in 34% of consultations. CoPAT was deemed not to be necessary in 27% of consultations. Among patients who did not require CoPAT, 60% received oral antibiotics and 40% were deemed not to need any antibiotics at hospital discharge. Among the patients discharged on CoPAT, a follow‐up appointment with a Cleveland Clinic ID physician familiar with the patient was set up 86% of the time; the rest either followed up with another physician or it was deemed that a scheduled follow‐up ID visit was not necessary.0

Figure 1

Figure 2

DISCUSSION

Physicians practicing in the specialty of infectious diseases face challenges and opportunities, as they adapt to changing demands within hospital practice in regard to reimbursement in an Accountable Care environment. Other challenges include emerging infections, antimicrobial resistance, need for antimicrobial stewardship, and increasing numbers of immunocompromised patients.12 From a health systems perspective, the overall value of care provided by the entire organization, and overall outcomes, are ultimately what matter. However, healthcare administrators need an appreciation of contributions of individual providers and specialties to fairly allocate resources and compensation for care provided. Articulating unique contributions is particularly challenging for individuals or services that provide purely cognitive input. Shrinking healthcare resources makes it critically important for cognitive specialists to be able to define their unique role in the care of patients with complex problems.

Our study found that a major contribution of ID consultation for CoPAT is that the process identifies a large number of patients who do not need CoPAT, thus effecting a powerful antimicrobial stewardship function. In our study, CoPAT was deemed unnecessary 27% of the time. The Infectious Diseases Society of America practice guidelines on outpatient parenteral antimicrobial therapy emphasize the importance of careful evaluation of patients considered for parenteral antibiotics outside the hospital setting.13 The focus on careful selection of appropriate patients for CoPAT has been a cornerstone of the Cleveland Clinic model of care. Nearly 30 years ago, we found that outpatient parenteral antibiotic therapy was unnecessary or not feasible in 40% of the patients referred for evaluation.10 If we adjust the numbers with the assumption that reimbursement issues present at that time are now less of an issue, the proportion of patients who were referred for CoPAT but not discharged on it was 29%, a figure remarkably similar to that found in the current study.

Another major contribution of ID consultation is the provision of effective transition of care from the inpatient to the outpatient setting. Frequent occurrence of postdischarge adverse events has been recognized as a problem in clinical practice.14 Primary care physicians are rarely involved in discussions about hospital discharge.15 A consensus conference including the American College of Physicians, Society of Hospital Medicine, and Society of General Internal Medicine, convened in July 2007 to address quality gaps in transitions of care between inpatient and outpatient settings. It identified 5 principles for effective care transitions: accountability, communication, timeliness, patient and family involvement, and respect for the hub of coordination of care.16 Recognizing gaps in care transition, hospitalists in a hospital‐based infusion program developed a model of care that successfully bridged the hospital‐to‐home care transition for patients who could return to hospital for daily antimicrobial infusions.17 In our system, ID physicians take ownership for directing parenteral antibiotic therapy for the episode of illness, specifying the physician, date, and time of follow‐up before the patient is discharged from hospital, thereby essentially satisfying the principles of effective care transitions identified. The purpose of the ID follow‐up is not to replace other follow‐up care for patients but to ensure safe transition of care while treating an episode of infection.

Attribution of identified contributions to the ID consultation could be done because our study was limited to CoPAT consultations. Such consultations typically occur when patients are deemed close to hospital discharge by the primary service. There should be little controversy about attribution of cognitive input in such consultations, because from the primary service's perspective, the patient is ready or almost ready to be discharged from hospital. It would be fair to state that most of the identified contributions in the study would not have occurred had it not been for the ID consultation.

We acknowledge that the study suffers from many limitations. The biggest limitation is that the contribution elements are defined by ID physicians and sought in the medical record by physicians from the same specialty. This arrangement certainly has potential for significant bias. To limit this bias, data collection was performed by physicians who had not participated in the care of the patient. In addition, we only could assess what was documented in the electronic health record. Our study found that alteration of antibiotic therapy was a substantial contribution, however, documentation of recommendation to change antibiotics in the medical record rarely specified exactly why the change was recommended. Reasons for antibiotic change recommendations included bug‐drug mismatch, minimum inhibitory concentration (MIC) considerations, pharmacokinetic considerations, adverse effects, convenience of dosing, drug interactions, and insurance coverage. However, it is not possible to quantify the specific contribution of each of these reasons, in a retrospective study, without making assumptions about why specific ID physicians made specific antibiotic change recommendations. There may have been more contributions that might not have been apparent on a retrospective chart review. The lack of a control group also lessens the impact of our findings. We could not have a control group, because no patient is discharged from the Cleveland Clinic on CoPAT without having been seen by an ID physician. Mandatory ID consultation for CoPAT has previously been shown to reduce costs,9 however, our study was not designed to evaluate cost.

The perceived value of ID consultation in our institution can be appreciated when one considers the longstanding institutional policy of requiring ID consultation for CoPAT.10, 11 The perpetuation of this tradition in the hospital is testament to the presumption that mandatory ID consultation is seen to be of value by the institution.

In summary, ID consultation in our institution contributes to the care of inpatients being considered for CoPAT by substantially reducing unnecessary parenteral antibiotic use, optimizing antibiotic therapy, recognizing need for additional testing before discharge from hospital, and by providing effective transition of care from the inpatient to the outpatient setting.

With dramatically increasing costs of healthcare, it has become increasingly necessary for healthcare providers to demonstrate value in the delivery of care. Porter and Teisberg have strongly advocated that healthcare reform efforts should focus on improving value rather than limiting cost, with value being defined as quality per unit cost.1 However, it has been pointed out that value means different things to different people.2 The biggest challenge in defining value stems mainly from the difficulty in defining quality, because it, too, means vastly different things to different people. Modern medicine is increasingly characterized by multidisciplinary care. With limited or shrinking resources, it will become necessary for individual specialists to describe and articulate, in quantitative terms, their specific contributions to the overall outcome of individual patients.

Previous publications have provided broad descriptions of the value provided by infectious disease (ID) specialists in the domains of sepsis, infection control, outpatient antibiotic therapy, antimicrobial stewardship, and directive care and teaching.3, 4 Studies have also shown the value of ID physicians in specific disease conditions. ID consultation is associated with lower mortality5, 6 and lower relapse rates7 in hospitalized patients with Staphylococcus aureus bacteremia. In another study evaluating the impact of ID consultants, patients seen by ID consultants had longer lengths of hospital stay, longer intensive care unit lengths of stay, and higher antibiotic costs than matched controls not seen by ID consultants.8 It can be argued that a major limitation of the study was that controls were not matched for the ID diagnosis, nor for the causative microorganisms, but it is clear that ID physicians are challenged to demonstrate their contribution to the care of patients.

A unique activity of ID physicians is the management of community‐based parenteral anti‐infective therapy (CoPAT). At Baystate Medical Center, a policy of mandatory ID consultation was instituted for patients leaving hospital on parenteral antibiotics. A study was conducted on the impact of predischarge ID consultation for 44 patients who were not already being followed by the ID service. The study documented change from intravenous (IV) to oral formulation, change of antibiotic choice, and change of dose/duration of treatment in a substantial proportion of patients.9 These are significant changes, but ID consultation contributes more than the themes explored in the study.

The purpose of this study was to evaluate the contribution of ID consultation when consulted for CoPAT, an activity specific to ID practice, in a different institution, and using an expanded definition of medical contribution.

METHODS

The Cleveland Clinic's Department of Infectious Disease has 24 staff physicians and 11 inpatient ID consultative services. These include: 2 solid organ transplant services; a bone marrow transplant and oncologic service; 2 infective endocarditis/cardiac device infection services; an intensive care unit (ICU) service; a bone and joint infection service; a neuroinfection service; and 3 general ID consult services. Consultative services are provided 7 days a week. At the Cleveland Clinic, ID consultation is required prior to discharge on parenteral antibiotic therapy.10, 11 ID consultation for CoPAT usually occurs when the primary service deems the patient is close to being discharged from hospital. This circumstance allows for assessing the specific contribution of ID physicians beyond that of the primary service and other consulting services.

RESULTS

In the time period studied, there were 1326 CPOE requests for ID consultation. The response to the question, CoPAT consult? was Yes for 304, No for 507, and Not sure for 515 requests. Of the 304 consultation requests marked Yes, 41 were excluded. Reasons for exclusion were: no ID consultation note (21), wrong service consulted (8), consultation request placed while the ID service was already following the patient (7), and duplicate consultation request (5). The remaining 263 consultation requests corresponded to 1 or more CoPAT consultation requests for 249 patients (across different hospitalizations). Of the 263 consultation requests, 172 were initial consultations, while the remaining 91 were reconsultations (patients not actively being followed by the ID service, but previously seen during the same hospitalization).

Consultation characteristics are outlined in Table 2. The most common group of infections for which CoPAT was sought was bone and joint infections, accounting for over 20% of the consultation requests. CoPAT consultations were requested a median of 4 days after hospitalization. Patients were discharged from hospital a median of 3 days after they were seen by the ID service. ID consultation did not delay discharge. The ID service usually saw the patient the same day, and followed the patient in hospital for a median of 1 day. There was no difference in hospital days after consult for patients who did not need antibiotics versus those who did.

ID consultation provided value in at least 1 domain in 260 of the 263 consultations. This included optimization of antimicrobial treatment in 84%, significant alteration of patient assessment in 52%, and additional medical care contribution in 71% of consultations. Substantial contributions were made in all domains in both initial consultations and in reconsultations. Specific ID contributions within each of the domains are shown in Figure 1. There was wide overlap of contributions across the 3 domains for individual consultations (Figure 2), with contributions in all domains occurring in 34% of consultations. CoPAT was deemed not to be necessary in 27% of consultations. Among patients who did not require CoPAT, 60% received oral antibiotics and 40% were deemed not to need any antibiotics at hospital discharge. Among the patients discharged on CoPAT, a follow‐up appointment with a Cleveland Clinic ID physician familiar with the patient was set up 86% of the time; the rest either followed up with another physician or it was deemed that a scheduled follow‐up ID visit was not necessary.0

Figure 1

Figure 2

DISCUSSION

Physicians practicing in the specialty of infectious diseases face challenges and opportunities, as they adapt to changing demands within hospital practice in regard to reimbursement in an Accountable Care environment. Other challenges include emerging infections, antimicrobial resistance, need for antimicrobial stewardship, and increasing numbers of immunocompromised patients.12 From a health systems perspective, the overall value of care provided by the entire organization, and overall outcomes, are ultimately what matter. However, healthcare administrators need an appreciation of contributions of individual providers and specialties to fairly allocate resources and compensation for care provided. Articulating unique contributions is particularly challenging for individuals or services that provide purely cognitive input. Shrinking healthcare resources makes it critically important for cognitive specialists to be able to define their unique role in the care of patients with complex problems.

Our study found that a major contribution of ID consultation for CoPAT is that the process identifies a large number of patients who do not need CoPAT, thus effecting a powerful antimicrobial stewardship function. In our study, CoPAT was deemed unnecessary 27% of the time. The Infectious Diseases Society of America practice guidelines on outpatient parenteral antimicrobial therapy emphasize the importance of careful evaluation of patients considered for parenteral antibiotics outside the hospital setting.13 The focus on careful selection of appropriate patients for CoPAT has been a cornerstone of the Cleveland Clinic model of care. Nearly 30 years ago, we found that outpatient parenteral antibiotic therapy was unnecessary or not feasible in 40% of the patients referred for evaluation.10 If we adjust the numbers with the assumption that reimbursement issues present at that time are now less of an issue, the proportion of patients who were referred for CoPAT but not discharged on it was 29%, a figure remarkably similar to that found in the current study.

Another major contribution of ID consultation is the provision of effective transition of care from the inpatient to the outpatient setting. Frequent occurrence of postdischarge adverse events has been recognized as a problem in clinical practice.14 Primary care physicians are rarely involved in discussions about hospital discharge.15 A consensus conference including the American College of Physicians, Society of Hospital Medicine, and Society of General Internal Medicine, convened in July 2007 to address quality gaps in transitions of care between inpatient and outpatient settings. It identified 5 principles for effective care transitions: accountability, communication, timeliness, patient and family involvement, and respect for the hub of coordination of care.16 Recognizing gaps in care transition, hospitalists in a hospital‐based infusion program developed a model of care that successfully bridged the hospital‐to‐home care transition for patients who could return to hospital for daily antimicrobial infusions.17 In our system, ID physicians take ownership for directing parenteral antibiotic therapy for the episode of illness, specifying the physician, date, and time of follow‐up before the patient is discharged from hospital, thereby essentially satisfying the principles of effective care transitions identified. The purpose of the ID follow‐up is not to replace other follow‐up care for patients but to ensure safe transition of care while treating an episode of infection.

Attribution of identified contributions to the ID consultation could be done because our study was limited to CoPAT consultations. Such consultations typically occur when patients are deemed close to hospital discharge by the primary service. There should be little controversy about attribution of cognitive input in such consultations, because from the primary service's perspective, the patient is ready or almost ready to be discharged from hospital. It would be fair to state that most of the identified contributions in the study would not have occurred had it not been for the ID consultation.

We acknowledge that the study suffers from many limitations. The biggest limitation is that the contribution elements are defined by ID physicians and sought in the medical record by physicians from the same specialty. This arrangement certainly has potential for significant bias. To limit this bias, data collection was performed by physicians who had not participated in the care of the patient. In addition, we only could assess what was documented in the electronic health record. Our study found that alteration of antibiotic therapy was a substantial contribution, however, documentation of recommendation to change antibiotics in the medical record rarely specified exactly why the change was recommended. Reasons for antibiotic change recommendations included bug‐drug mismatch, minimum inhibitory concentration (MIC) considerations, pharmacokinetic considerations, adverse effects, convenience of dosing, drug interactions, and insurance coverage. However, it is not possible to quantify the specific contribution of each of these reasons, in a retrospective study, without making assumptions about why specific ID physicians made specific antibiotic change recommendations. There may have been more contributions that might not have been apparent on a retrospective chart review. The lack of a control group also lessens the impact of our findings. We could not have a control group, because no patient is discharged from the Cleveland Clinic on CoPAT without having been seen by an ID physician. Mandatory ID consultation for CoPAT has previously been shown to reduce costs,9 however, our study was not designed to evaluate cost.

The perceived value of ID consultation in our institution can be appreciated when one considers the longstanding institutional policy of requiring ID consultation for CoPAT.10, 11 The perpetuation of this tradition in the hospital is testament to the presumption that mandatory ID consultation is seen to be of value by the institution.

In summary, ID consultation in our institution contributes to the care of inpatients being considered for CoPAT by substantially reducing unnecessary parenteral antibiotic use, optimizing antibiotic therapy, recognizing need for additional testing before discharge from hospital, and by providing effective transition of care from the inpatient to the outpatient setting.

With dramatically increasing costs of healthcare, it has become increasingly necessary for healthcare providers to demonstrate value in the delivery of care. Porter and Teisberg have strongly advocated that healthcare reform efforts should focus on improving value rather than limiting cost, with value being defined as quality per unit cost.1 However, it has been pointed out that value means different things to different people.2 The biggest challenge in defining value stems mainly from the difficulty in defining quality, because it, too, means vastly different things to different people. Modern medicine is increasingly characterized by multidisciplinary care. With limited or shrinking resources, it will become necessary for individual specialists to describe and articulate, in quantitative terms, their specific contributions to the overall outcome of individual patients.

Previous publications have provided broad descriptions of the value provided by infectious disease (ID) specialists in the domains of sepsis, infection control, outpatient antibiotic therapy, antimicrobial stewardship, and directive care and teaching.3, 4 Studies have also shown the value of ID physicians in specific disease conditions. ID consultation is associated with lower mortality5, 6 and lower relapse rates7 in hospitalized patients with Staphylococcus aureus bacteremia. In another study evaluating the impact of ID consultants, patients seen by ID consultants had longer lengths of hospital stay, longer intensive care unit lengths of stay, and higher antibiotic costs than matched controls not seen by ID consultants.8 It can be argued that a major limitation of the study was that controls were not matched for the ID diagnosis, nor for the causative microorganisms, but it is clear that ID physicians are challenged to demonstrate their contribution to the care of patients.

A unique activity of ID physicians is the management of community‐based parenteral anti‐infective therapy (CoPAT). At Baystate Medical Center, a policy of mandatory ID consultation was instituted for patients leaving hospital on parenteral antibiotics. A study was conducted on the impact of predischarge ID consultation for 44 patients who were not already being followed by the ID service. The study documented change from intravenous (IV) to oral formulation, change of antibiotic choice, and change of dose/duration of treatment in a substantial proportion of patients.9 These are significant changes, but ID consultation contributes more than the themes explored in the study.

The purpose of this study was to evaluate the contribution of ID consultation when consulted for CoPAT, an activity specific to ID practice, in a different institution, and using an expanded definition of medical contribution.

METHODS

The Cleveland Clinic's Department of Infectious Disease has 24 staff physicians and 11 inpatient ID consultative services. These include: 2 solid organ transplant services; a bone marrow transplant and oncologic service; 2 infective endocarditis/cardiac device infection services; an intensive care unit (ICU) service; a bone and joint infection service; a neuroinfection service; and 3 general ID consult services. Consultative services are provided 7 days a week. At the Cleveland Clinic, ID consultation is required prior to discharge on parenteral antibiotic therapy.10, 11 ID consultation for CoPAT usually occurs when the primary service deems the patient is close to being discharged from hospital. This circumstance allows for assessing the specific contribution of ID physicians beyond that of the primary service and other consulting services.

RESULTS

In the time period studied, there were 1326 CPOE requests for ID consultation. The response to the question, CoPAT consult? was Yes for 304, No for 507, and Not sure for 515 requests. Of the 304 consultation requests marked Yes, 41 were excluded. Reasons for exclusion were: no ID consultation note (21), wrong service consulted (8), consultation request placed while the ID service was already following the patient (7), and duplicate consultation request (5). The remaining 263 consultation requests corresponded to 1 or more CoPAT consultation requests for 249 patients (across different hospitalizations). Of the 263 consultation requests, 172 were initial consultations, while the remaining 91 were reconsultations (patients not actively being followed by the ID service, but previously seen during the same hospitalization).

Consultation characteristics are outlined in Table 2. The most common group of infections for which CoPAT was sought was bone and joint infections, accounting for over 20% of the consultation requests. CoPAT consultations were requested a median of 4 days after hospitalization. Patients were discharged from hospital a median of 3 days after they were seen by the ID service. ID consultation did not delay discharge. The ID service usually saw the patient the same day, and followed the patient in hospital for a median of 1 day. There was no difference in hospital days after consult for patients who did not need antibiotics versus those who did.

ID consultation provided value in at least 1 domain in 260 of the 263 consultations. This included optimization of antimicrobial treatment in 84%, significant alteration of patient assessment in 52%, and additional medical care contribution in 71% of consultations. Substantial contributions were made in all domains in both initial consultations and in reconsultations. Specific ID contributions within each of the domains are shown in Figure 1. There was wide overlap of contributions across the 3 domains for individual consultations (Figure 2), with contributions in all domains occurring in 34% of consultations. CoPAT was deemed not to be necessary in 27% of consultations. Among patients who did not require CoPAT, 60% received oral antibiotics and 40% were deemed not to need any antibiotics at hospital discharge. Among the patients discharged on CoPAT, a follow‐up appointment with a Cleveland Clinic ID physician familiar with the patient was set up 86% of the time; the rest either followed up with another physician or it was deemed that a scheduled follow‐up ID visit was not necessary.0

Figure 1

Figure 2

DISCUSSION

Physicians practicing in the specialty of infectious diseases face challenges and opportunities, as they adapt to changing demands within hospital practice in regard to reimbursement in an Accountable Care environment. Other challenges include emerging infections, antimicrobial resistance, need for antimicrobial stewardship, and increasing numbers of immunocompromised patients.12 From a health systems perspective, the overall value of care provided by the entire organization, and overall outcomes, are ultimately what matter. However, healthcare administrators need an appreciation of contributions of individual providers and specialties to fairly allocate resources and compensation for care provided. Articulating unique contributions is particularly challenging for individuals or services that provide purely cognitive input. Shrinking healthcare resources makes it critically important for cognitive specialists to be able to define their unique role in the care of patients with complex problems.

Our study found that a major contribution of ID consultation for CoPAT is that the process identifies a large number of patients who do not need CoPAT, thus effecting a powerful antimicrobial stewardship function. In our study, CoPAT was deemed unnecessary 27% of the time. The Infectious Diseases Society of America practice guidelines on outpatient parenteral antimicrobial therapy emphasize the importance of careful evaluation of patients considered for parenteral antibiotics outside the hospital setting.13 The focus on careful selection of appropriate patients for CoPAT has been a cornerstone of the Cleveland Clinic model of care. Nearly 30 years ago, we found that outpatient parenteral antibiotic therapy was unnecessary or not feasible in 40% of the patients referred for evaluation.10 If we adjust the numbers with the assumption that reimbursement issues present at that time are now less of an issue, the proportion of patients who were referred for CoPAT but not discharged on it was 29%, a figure remarkably similar to that found in the current study.

Another major contribution of ID consultation is the provision of effective transition of care from the inpatient to the outpatient setting. Frequent occurrence of postdischarge adverse events has been recognized as a problem in clinical practice.14 Primary care physicians are rarely involved in discussions about hospital discharge.15 A consensus conference including the American College of Physicians, Society of Hospital Medicine, and Society of General Internal Medicine, convened in July 2007 to address quality gaps in transitions of care between inpatient and outpatient settings. It identified 5 principles for effective care transitions: accountability, communication, timeliness, patient and family involvement, and respect for the hub of coordination of care.16 Recognizing gaps in care transition, hospitalists in a hospital‐based infusion program developed a model of care that successfully bridged the hospital‐to‐home care transition for patients who could return to hospital for daily antimicrobial infusions.17 In our system, ID physicians take ownership for directing parenteral antibiotic therapy for the episode of illness, specifying the physician, date, and time of follow‐up before the patient is discharged from hospital, thereby essentially satisfying the principles of effective care transitions identified. The purpose of the ID follow‐up is not to replace other follow‐up care for patients but to ensure safe transition of care while treating an episode of infection.

Attribution of identified contributions to the ID consultation could be done because our study was limited to CoPAT consultations. Such consultations typically occur when patients are deemed close to hospital discharge by the primary service. There should be little controversy about attribution of cognitive input in such consultations, because from the primary service's perspective, the patient is ready or almost ready to be discharged from hospital. It would be fair to state that most of the identified contributions in the study would not have occurred had it not been for the ID consultation.

We acknowledge that the study suffers from many limitations. The biggest limitation is that the contribution elements are defined by ID physicians and sought in the medical record by physicians from the same specialty. This arrangement certainly has potential for significant bias. To limit this bias, data collection was performed by physicians who had not participated in the care of the patient. In addition, we only could assess what was documented in the electronic health record. Our study found that alteration of antibiotic therapy was a substantial contribution, however, documentation of recommendation to change antibiotics in the medical record rarely specified exactly why the change was recommended. Reasons for antibiotic change recommendations included bug‐drug mismatch, minimum inhibitory concentration (MIC) considerations, pharmacokinetic considerations, adverse effects, convenience of dosing, drug interactions, and insurance coverage. However, it is not possible to quantify the specific contribution of each of these reasons, in a retrospective study, without making assumptions about why specific ID physicians made specific antibiotic change recommendations. There may have been more contributions that might not have been apparent on a retrospective chart review. The lack of a control group also lessens the impact of our findings. We could not have a control group, because no patient is discharged from the Cleveland Clinic on CoPAT without having been seen by an ID physician. Mandatory ID consultation for CoPAT has previously been shown to reduce costs,9 however, our study was not designed to evaluate cost.

The perceived value of ID consultation in our institution can be appreciated when one considers the longstanding institutional policy of requiring ID consultation for CoPAT.10, 11 The perpetuation of this tradition in the hospital is testament to the presumption that mandatory ID consultation is seen to be of value by the institution.

In summary, ID consultation in our institution contributes to the care of inpatients being considered for CoPAT by substantially reducing unnecessary parenteral antibiotic use, optimizing antibiotic therapy, recognizing need for additional testing before discharge from hospital, and by providing effective transition of care from the inpatient to the outpatient setting.