Hospital Medicine 13

NATIONAL HARBOR, MD. – Hospital and physician efforts to improve patient satisfaction are just scratching the surface, according to Dr. Shaun Frost, the outgoing president of the Society of Hospital Medicine.

The missing ingredient, Dr. Frost said, is understanding the patient’s personal expectations about the hospital stay.

For instance, a patient might have a preference for tolerating pain in order to stay alert versus accepting the sedating effects of treatment with narcotics. Another patient might prefer to be discharged from the emergency department and have follow-up testing performed in an outpatient setting rather than stay overnight under observation status.

"Where we sometimes come up short in health care is failing to recognize that patient expectations transcend interests such as having their call lights answered in a timely manner or receiving warm meals on a scheduled basis," Dr. Frost said at the annual meeting of the Society of Hospital Medicine. "Although these are obviously important expectations that need to be addressed, they are relatively easy to identify and act on."

The challenge in uncovering these personally held patient expectations is that physicians have to ask about them, he said.

"Providers must empower patients to express their expectations by taking time to inquire about the social, economic, cultural, and environmental issues that dictate personally held patient values, goals, needs, interests, and preferences for their care," said Dr. Frost, who is the associate medical director for care delivery systems at HealthPartners in Bloomington, Minn.

But it’s not enough to know what the patient wants. Physicians also need to bring patients into the decision-making process by objectively explaining treatment options and avoiding the temptation to act paternalistically, Dr. Frost said.

Based on an experience from early in his career, Dr. Frost offered an example of what not to do: As a third-year medical student, Dr. Frost observed his supervising resident advising a patient with heart failure about establishing an advanced directive. The resident skillfully outlined the technical details about cardiopulmonary resuscitation, Dr. Frost said, but the explanation seemed aimed at influencing the patient to decide against resuscitation. When the resident asked the patient what he would like the team to do, he surprised everyone by emphatically stating that he wanted to be resuscitated.

The conversation ended with the resident accepting the patient’s decision, but also advising him to think about it more because they would talk about it again the next day. The following day the patient had not changed his mind, but he seemed notably disengaged, Dr. Frost said. From that point on, the patient appeared "guarded and withdrawn" and reluctant to participate in his treatments, Dr. Frost said.

"The fundamental problem here was that we spoke to this patient about our experience in running cardiac arrest code, but we didn’t speak with this patient about what it’s like for him and his family to live with severe congestive heart failure and furthermore why it was he’d been admitted to the hospital three times in the past 3 months," Dr. Frost said. "Instead of engaging this patient by inviting him into the conversation, we left him with the impression that he didn’t have a choice."

As the evidence grows that patient engagement has a positive influence on quality, safety, and affordability, Dr. Frost said it’s time for the specialty of hospital medicine to take the lead by setting standards for physician-patient collaboration.

"Enhanced collaboration with patients, their families, and their loved ones, which focuses attention on their personal unique needs and preferences, is the next great opportunity in health care team work," he said.

[email protected]

NATIONAL HARBOR, MD. – Hospital and physician efforts to improve patient satisfaction are just scratching the surface, according to Dr. Shaun Frost, the outgoing president of the Society of Hospital Medicine.

The missing ingredient, Dr. Frost said, is understanding the patient’s personal expectations about the hospital stay.

For instance, a patient might have a preference for tolerating pain in order to stay alert versus accepting the sedating effects of treatment with narcotics. Another patient might prefer to be discharged from the emergency department and have follow-up testing performed in an outpatient setting rather than stay overnight under observation status.

"Where we sometimes come up short in health care is failing to recognize that patient expectations transcend interests such as having their call lights answered in a timely manner or receiving warm meals on a scheduled basis," Dr. Frost said at the annual meeting of the Society of Hospital Medicine. "Although these are obviously important expectations that need to be addressed, they are relatively easy to identify and act on."

The challenge in uncovering these personally held patient expectations is that physicians have to ask about them, he said.

"Providers must empower patients to express their expectations by taking time to inquire about the social, economic, cultural, and environmental issues that dictate personally held patient values, goals, needs, interests, and preferences for their care," said Dr. Frost, who is the associate medical director for care delivery systems at HealthPartners in Bloomington, Minn.

But it’s not enough to know what the patient wants. Physicians also need to bring patients into the decision-making process by objectively explaining treatment options and avoiding the temptation to act paternalistically, Dr. Frost said.

Based on an experience from early in his career, Dr. Frost offered an example of what not to do: As a third-year medical student, Dr. Frost observed his supervising resident advising a patient with heart failure about establishing an advanced directive. The resident skillfully outlined the technical details about cardiopulmonary resuscitation, Dr. Frost said, but the explanation seemed aimed at influencing the patient to decide against resuscitation. When the resident asked the patient what he would like the team to do, he surprised everyone by emphatically stating that he wanted to be resuscitated.

The conversation ended with the resident accepting the patient’s decision, but also advising him to think about it more because they would talk about it again the next day. The following day the patient had not changed his mind, but he seemed notably disengaged, Dr. Frost said. From that point on, the patient appeared "guarded and withdrawn" and reluctant to participate in his treatments, Dr. Frost said.

"The fundamental problem here was that we spoke to this patient about our experience in running cardiac arrest code, but we didn’t speak with this patient about what it’s like for him and his family to live with severe congestive heart failure and furthermore why it was he’d been admitted to the hospital three times in the past 3 months," Dr. Frost said. "Instead of engaging this patient by inviting him into the conversation, we left him with the impression that he didn’t have a choice."

As the evidence grows that patient engagement has a positive influence on quality, safety, and affordability, Dr. Frost said it’s time for the specialty of hospital medicine to take the lead by setting standards for physician-patient collaboration.

"Enhanced collaboration with patients, their families, and their loved ones, which focuses attention on their personal unique needs and preferences, is the next great opportunity in health care team work," he said.

[email protected]

NATIONAL HARBOR, MD. – Hospital and physician efforts to improve patient satisfaction are just scratching the surface, according to Dr. Shaun Frost, the outgoing president of the Society of Hospital Medicine.

The missing ingredient, Dr. Frost said, is understanding the patient’s personal expectations about the hospital stay.

For instance, a patient might have a preference for tolerating pain in order to stay alert versus accepting the sedating effects of treatment with narcotics. Another patient might prefer to be discharged from the emergency department and have follow-up testing performed in an outpatient setting rather than stay overnight under observation status.

"Where we sometimes come up short in health care is failing to recognize that patient expectations transcend interests such as having their call lights answered in a timely manner or receiving warm meals on a scheduled basis," Dr. Frost said at the annual meeting of the Society of Hospital Medicine. "Although these are obviously important expectations that need to be addressed, they are relatively easy to identify and act on."

The challenge in uncovering these personally held patient expectations is that physicians have to ask about them, he said.

"Providers must empower patients to express their expectations by taking time to inquire about the social, economic, cultural, and environmental issues that dictate personally held patient values, goals, needs, interests, and preferences for their care," said Dr. Frost, who is the associate medical director for care delivery systems at HealthPartners in Bloomington, Minn.

But it’s not enough to know what the patient wants. Physicians also need to bring patients into the decision-making process by objectively explaining treatment options and avoiding the temptation to act paternalistically, Dr. Frost said.

Based on an experience from early in his career, Dr. Frost offered an example of what not to do: As a third-year medical student, Dr. Frost observed his supervising resident advising a patient with heart failure about establishing an advanced directive. The resident skillfully outlined the technical details about cardiopulmonary resuscitation, Dr. Frost said, but the explanation seemed aimed at influencing the patient to decide against resuscitation. When the resident asked the patient what he would like the team to do, he surprised everyone by emphatically stating that he wanted to be resuscitated.

The conversation ended with the resident accepting the patient’s decision, but also advising him to think about it more because they would talk about it again the next day. The following day the patient had not changed his mind, but he seemed notably disengaged, Dr. Frost said. From that point on, the patient appeared "guarded and withdrawn" and reluctant to participate in his treatments, Dr. Frost said.

"The fundamental problem here was that we spoke to this patient about our experience in running cardiac arrest code, but we didn’t speak with this patient about what it’s like for him and his family to live with severe congestive heart failure and furthermore why it was he’d been admitted to the hospital three times in the past 3 months," Dr. Frost said. "Instead of engaging this patient by inviting him into the conversation, we left him with the impression that he didn’t have a choice."

As the evidence grows that patient engagement has a positive influence on quality, safety, and affordability, Dr. Frost said it’s time for the specialty of hospital medicine to take the lead by setting standards for physician-patient collaboration.

"Enhanced collaboration with patients, their families, and their loved ones, which focuses attention on their personal unique needs and preferences, is the next great opportunity in health care team work," he said.

[email protected]

Half of patients admitted to U.S. hospitals for nonsurgical reasons were given opioids during their stay, according to a large national database review.

The medications were also given in relatively high doses – a mean of 70 mg oral morphine equivalents daily overall, but 100 mg or higher in about a quarter of patients, Dr. Shoshana Herzig said at the annual meeting of the Society of Hospital Medicine.

"Given the prevalence of high-dose exposure, the known risks of these drugs, and the older age of many recipients, opportunities exist to make opioid use safer in patients," said Dr. Herzig of Beth Israel Deaconess Medical Center, Boston.

Courtesy BIDMC

Dr. Herzig used data extracted from a large, national hospital database. The database contains administrative information from 600 U.S. hospitals, and reflects about 25% of the nation’s annual hospital discharges.

For this analysis, she selected 1.14 million nonsurgical admissions that occurred among 286 hospitals from July 2009 to June 2010. Opioid exposure was defined as at least one charge for an opioid medication during the admission or on the day of discharge. The database does not provide information about discharge prescriptions, nor does it describe in detail the pain complaint associated with the charge.

The patients’ median age was 64 years; 46% were male. Morphine was the most commonly prescribed (20%), followed by hydrocodone (14%), and hydromorphone (13%). Patients received the drugs parenterally (66%), orally (65%), or both.

Overall, opioids were used in 51% of admissions, including 45% of patients aged 65 years and older. Many patients (43%) received two or more different opioids. More than half of those who took the drugs during their stay (52%) also had them on the day of discharge.

The observed rate of opioid use varied considerably among the hospitals, ranging from 5% in the lowest-prescribing hospital to 72% in the highest-prescribing hospital, with a mean of 51%.

Dr. Herzig found regional differences as well, with the highest use in Western states (55% of all admissions), followed by Southern states (53% of admissions), the Midwest (50%), and the Northeast (39%).

A multivariate analysis that controlled for patient demographics, comorbidities, and hospital characteristics found a number of significant associations with opioid use.

Compared with whites and Hispanics, black patients were less likely to receive the drugs (odds ratio, 0.93). Patients with musculoskeletal injuries were most likely to receive opioids (OR, 2.0), as were patients with cancer (OR, 1.2). Psychiatric and alcohol admissions were significantly less likely to get opioids (OR, 0.37 and 0.46, respectively), but there was no significant relationship with an admission for substance abuse.

Some typically painful procedures significantly predicted opioid use, including cardiovascular procedures (OR, 1.8), gastrointestinal procedures (OR, 1.7), and mechanical ventilation (OR, 1.4).

The analysis didn’t allow any conclusions to be drawn about causality, Dr. Herzig said, but the findings call for more research. "Increased attention should be paid to the role that inpatient opioid prescribing plays in the increased rates of chronic opioid use and overdose-related deaths in the United States."

Dr. Herzig had no financial declarations.

[email protected]

Half of patients admitted to U.S. hospitals for nonsurgical reasons were given opioids during their stay, according to a large national database review.

The medications were also given in relatively high doses – a mean of 70 mg oral morphine equivalents daily overall, but 100 mg or higher in about a quarter of patients, Dr. Shoshana Herzig said at the annual meeting of the Society of Hospital Medicine.

"Given the prevalence of high-dose exposure, the known risks of these drugs, and the older age of many recipients, opportunities exist to make opioid use safer in patients," said Dr. Herzig of Beth Israel Deaconess Medical Center, Boston.

Courtesy BIDMC

Dr. Herzig used data extracted from a large, national hospital database. The database contains administrative information from 600 U.S. hospitals, and reflects about 25% of the nation’s annual hospital discharges.

For this analysis, she selected 1.14 million nonsurgical admissions that occurred among 286 hospitals from July 2009 to June 2010. Opioid exposure was defined as at least one charge for an opioid medication during the admission or on the day of discharge. The database does not provide information about discharge prescriptions, nor does it describe in detail the pain complaint associated with the charge.

The patients’ median age was 64 years; 46% were male. Morphine was the most commonly prescribed (20%), followed by hydrocodone (14%), and hydromorphone (13%). Patients received the drugs parenterally (66%), orally (65%), or both.

Overall, opioids were used in 51% of admissions, including 45% of patients aged 65 years and older. Many patients (43%) received two or more different opioids. More than half of those who took the drugs during their stay (52%) also had them on the day of discharge.

The observed rate of opioid use varied considerably among the hospitals, ranging from 5% in the lowest-prescribing hospital to 72% in the highest-prescribing hospital, with a mean of 51%.

Dr. Herzig found regional differences as well, with the highest use in Western states (55% of all admissions), followed by Southern states (53% of admissions), the Midwest (50%), and the Northeast (39%).

A multivariate analysis that controlled for patient demographics, comorbidities, and hospital characteristics found a number of significant associations with opioid use.

Compared with whites and Hispanics, black patients were less likely to receive the drugs (odds ratio, 0.93). Patients with musculoskeletal injuries were most likely to receive opioids (OR, 2.0), as were patients with cancer (OR, 1.2). Psychiatric and alcohol admissions were significantly less likely to get opioids (OR, 0.37 and 0.46, respectively), but there was no significant relationship with an admission for substance abuse.

Some typically painful procedures significantly predicted opioid use, including cardiovascular procedures (OR, 1.8), gastrointestinal procedures (OR, 1.7), and mechanical ventilation (OR, 1.4).

The analysis didn’t allow any conclusions to be drawn about causality, Dr. Herzig said, but the findings call for more research. "Increased attention should be paid to the role that inpatient opioid prescribing plays in the increased rates of chronic opioid use and overdose-related deaths in the United States."

Dr. Herzig had no financial declarations.

[email protected]

Half of patients admitted to U.S. hospitals for nonsurgical reasons were given opioids during their stay, according to a large national database review.

The medications were also given in relatively high doses – a mean of 70 mg oral morphine equivalents daily overall, but 100 mg or higher in about a quarter of patients, Dr. Shoshana Herzig said at the annual meeting of the Society of Hospital Medicine.

"Given the prevalence of high-dose exposure, the known risks of these drugs, and the older age of many recipients, opportunities exist to make opioid use safer in patients," said Dr. Herzig of Beth Israel Deaconess Medical Center, Boston.

Courtesy BIDMC

Dr. Herzig used data extracted from a large, national hospital database. The database contains administrative information from 600 U.S. hospitals, and reflects about 25% of the nation’s annual hospital discharges.

For this analysis, she selected 1.14 million nonsurgical admissions that occurred among 286 hospitals from July 2009 to June 2010. Opioid exposure was defined as at least one charge for an opioid medication during the admission or on the day of discharge. The database does not provide information about discharge prescriptions, nor does it describe in detail the pain complaint associated with the charge.

The patients’ median age was 64 years; 46% were male. Morphine was the most commonly prescribed (20%), followed by hydrocodone (14%), and hydromorphone (13%). Patients received the drugs parenterally (66%), orally (65%), or both.

Overall, opioids were used in 51% of admissions, including 45% of patients aged 65 years and older. Many patients (43%) received two or more different opioids. More than half of those who took the drugs during their stay (52%) also had them on the day of discharge.

The observed rate of opioid use varied considerably among the hospitals, ranging from 5% in the lowest-prescribing hospital to 72% in the highest-prescribing hospital, with a mean of 51%.

Dr. Herzig found regional differences as well, with the highest use in Western states (55% of all admissions), followed by Southern states (53% of admissions), the Midwest (50%), and the Northeast (39%).

A multivariate analysis that controlled for patient demographics, comorbidities, and hospital characteristics found a number of significant associations with opioid use.

Compared with whites and Hispanics, black patients were less likely to receive the drugs (odds ratio, 0.93). Patients with musculoskeletal injuries were most likely to receive opioids (OR, 2.0), as were patients with cancer (OR, 1.2). Psychiatric and alcohol admissions were significantly less likely to get opioids (OR, 0.37 and 0.46, respectively), but there was no significant relationship with an admission for substance abuse.

Some typically painful procedures significantly predicted opioid use, including cardiovascular procedures (OR, 1.8), gastrointestinal procedures (OR, 1.7), and mechanical ventilation (OR, 1.4).

The analysis didn’t allow any conclusions to be drawn about causality, Dr. Herzig said, but the findings call for more research. "Increased attention should be paid to the role that inpatient opioid prescribing plays in the increased rates of chronic opioid use and overdose-related deaths in the United States."

Dr. Herzig had no financial declarations.

[email protected]

NATIONAL HARBOR, MD. – A project aimed at improving the transition of care from hospitals to outpatient facilities significantly reduced 30-day readmission rates in a group of Illinois hospitals.

Hospitals that adopted the Project BOOST protocol also improved several quality measures of their discharge process, Dr. Jing Li said at the annual meeting of the Society of Hospital Medicine.

Project BOOST (Better Outcomes by Optimizing Safe Transitions) is a patient-centered, multicomponent intervention designed to enhance the hospital discharge transition. Integration of BOOST tools into the discharge process at a hospital is facilitated by physician-mentored implementation that provides expertise in care transitions, quality improvement, and outside support for internal change.

In 2011, a statewide survey determined that Illinois hospitals fell into the bottom quartile for 30-day readmission rates of Medicare beneficiaries when compared with other U.S. hospitals. In response to that report, BlueCross BlueShield of Illinois, the Illinois Hospital Association, and Northwestern University, Chicago, entered into the Project BOOST program to enhance care for patients at hospitals all across the state, said Dr. Li, director of Project BOOST at the university.

The pilot cohort in Illinois consisted of 14 hospitals, all of which received mentored implementation with coaching, monthly teleconferences, and regular data collection. Dr. Li presented results for six of these, because the other eight lacked adequate means of data collection during that first wave.

Of the six hospitals, five were community, nonteaching facilities; the other was also a community hospital, but it had a teaching component. Four of the facilities were urban, one suburban, and one rural. The mean bed size was 343, but that ranged from 255 to 501.

The midyear implementation survey showed 79% of BOOST teams implemented comprehensive patient readmission risk assessment; 57% implemented the discharge checklist; 79% started to use teach-back for patient and family education; 79% established mechanisms to ensure information was available to subacute providers at the time of discharge; and all hospitals conducted follow-up phone calls to more than 60% of their discharged patients.

The six hospitals for which BOOST and comparison unit data were available experienced a 25% relative reduction in 30-day readmission rates (from 16% to 12%), with a significant downward trend over time. Over the same period, the 30-day readmission rate in comparison units remained flat.

The program is now being implemented in 22 more facilities across the state, Dr. Li said.

The project is being funded by BlueCross BlueShield of Illinois. Dr. Li had no financial disclosures.

[email protected]

NATIONAL HARBOR, MD. – A project aimed at improving the transition of care from hospitals to outpatient facilities significantly reduced 30-day readmission rates in a group of Illinois hospitals.

Hospitals that adopted the Project BOOST protocol also improved several quality measures of their discharge process, Dr. Jing Li said at the annual meeting of the Society of Hospital Medicine.

Project BOOST (Better Outcomes by Optimizing Safe Transitions) is a patient-centered, multicomponent intervention designed to enhance the hospital discharge transition. Integration of BOOST tools into the discharge process at a hospital is facilitated by physician-mentored implementation that provides expertise in care transitions, quality improvement, and outside support for internal change.

In 2011, a statewide survey determined that Illinois hospitals fell into the bottom quartile for 30-day readmission rates of Medicare beneficiaries when compared with other U.S. hospitals. In response to that report, BlueCross BlueShield of Illinois, the Illinois Hospital Association, and Northwestern University, Chicago, entered into the Project BOOST program to enhance care for patients at hospitals all across the state, said Dr. Li, director of Project BOOST at the university.

The pilot cohort in Illinois consisted of 14 hospitals, all of which received mentored implementation with coaching, monthly teleconferences, and regular data collection. Dr. Li presented results for six of these, because the other eight lacked adequate means of data collection during that first wave.

Of the six hospitals, five were community, nonteaching facilities; the other was also a community hospital, but it had a teaching component. Four of the facilities were urban, one suburban, and one rural. The mean bed size was 343, but that ranged from 255 to 501.

The midyear implementation survey showed 79% of BOOST teams implemented comprehensive patient readmission risk assessment; 57% implemented the discharge checklist; 79% started to use teach-back for patient and family education; 79% established mechanisms to ensure information was available to subacute providers at the time of discharge; and all hospitals conducted follow-up phone calls to more than 60% of their discharged patients.

The six hospitals for which BOOST and comparison unit data were available experienced a 25% relative reduction in 30-day readmission rates (from 16% to 12%), with a significant downward trend over time. Over the same period, the 30-day readmission rate in comparison units remained flat.

The program is now being implemented in 22 more facilities across the state, Dr. Li said.

The project is being funded by BlueCross BlueShield of Illinois. Dr. Li had no financial disclosures.

[email protected]

NATIONAL HARBOR, MD. – A project aimed at improving the transition of care from hospitals to outpatient facilities significantly reduced 30-day readmission rates in a group of Illinois hospitals.

Hospitals that adopted the Project BOOST protocol also improved several quality measures of their discharge process, Dr. Jing Li said at the annual meeting of the Society of Hospital Medicine.

Project BOOST (Better Outcomes by Optimizing Safe Transitions) is a patient-centered, multicomponent intervention designed to enhance the hospital discharge transition. Integration of BOOST tools into the discharge process at a hospital is facilitated by physician-mentored implementation that provides expertise in care transitions, quality improvement, and outside support for internal change.

In 2011, a statewide survey determined that Illinois hospitals fell into the bottom quartile for 30-day readmission rates of Medicare beneficiaries when compared with other U.S. hospitals. In response to that report, BlueCross BlueShield of Illinois, the Illinois Hospital Association, and Northwestern University, Chicago, entered into the Project BOOST program to enhance care for patients at hospitals all across the state, said Dr. Li, director of Project BOOST at the university.

The pilot cohort in Illinois consisted of 14 hospitals, all of which received mentored implementation with coaching, monthly teleconferences, and regular data collection. Dr. Li presented results for six of these, because the other eight lacked adequate means of data collection during that first wave.

Of the six hospitals, five were community, nonteaching facilities; the other was also a community hospital, but it had a teaching component. Four of the facilities were urban, one suburban, and one rural. The mean bed size was 343, but that ranged from 255 to 501.

The midyear implementation survey showed 79% of BOOST teams implemented comprehensive patient readmission risk assessment; 57% implemented the discharge checklist; 79% started to use teach-back for patient and family education; 79% established mechanisms to ensure information was available to subacute providers at the time of discharge; and all hospitals conducted follow-up phone calls to more than 60% of their discharged patients.

The six hospitals for which BOOST and comparison unit data were available experienced a 25% relative reduction in 30-day readmission rates (from 16% to 12%), with a significant downward trend over time. Over the same period, the 30-day readmission rate in comparison units remained flat.

The program is now being implemented in 22 more facilities across the state, Dr. Li said.

The project is being funded by BlueCross BlueShield of Illinois. Dr. Li had no financial disclosures.

[email protected]

NATIONAL HARBOR, MD. – A hospital-wide rapid response protocol decreased the time to antibiotic administration and improved sepsis bundle compliance by creating a team approach to sepsis management.

Creating "Code Sepsis" involved every member of the clinical care team, from nursing assistants to attending physicians, Dr. Catherine Jones said at the annual meeting of the Society of Hospital Medicine. The pilot project has been so successful that it’s being implemented in other units of Wake Forest Baptist Medical Center, Winston-Salem, N.C.

"We’ve started a modified version of it in our surgical intensive care units and, in April, our emergency department rolled out this exact same process. Our goal is to roll it out throughout the organization, including our comprehensive cancer center, because it’s not unusual for patients to show up in clinic with severe sepsis," said Dr. Jones, a patient safety and hospital quality expert at the group.

The incidence of sepsis has been on an upward trajectory since 2000, she said, because of a variety of factors. An aging population and the more frequent use of invasive procedures, chemotherapy, and immunosuppressive drugs probably all contribute to the problem. Patients who develop sepsis are eight times more likely to die during a hospital admission.

The good news is that the international "Surviving Sepsis" campaign, which advocated a resuscitation bundle approach designed for early intervention, does seem to be making a difference in outcomes since it was inaugurated in 2002.

The key element seems to be the earlier timing of antibiotic delivery, with an estimate of one patient saved for every seven bundles implemented, Dr. Jones said.

"The problem is that, on the floor, early recognition of sepsis is difficult. It’s a spectrum of illness from infection to sepsis, severe sepsis, and septic shock, and none of these stages have simple, definable signs, symptoms, lab values, or imaging findings. Many other conditions mimic sepsis and present with the same criteria and organ failure as severe sepsis."

Several years ago, the facility began looking at its overall mortality figures, with a goal of reducing mortality by 15%. Since sepsis is an important driver of mortality, it became a target for intervention. A working group composed of interns, nurses, nursing assistants, pharmacists, and respiratory therapists met for 2 days to create an action plan – and Code Sepsis was born. The name was chosen for two reasons, Dr. Jones said: It connotes both a sense of urgency, and the understanding that there is a set protocol to follow.

The rapid response team is the protocol’s backbone. Consisting of two critical care nurses and a first-call provider, the team goes into action any time a patient meets the criteria for possible sepsis. The nurses activate Code Sepsis by calling emergency communications; they quickly the get blood work done and sent to the lab, and administer any initial treatments. The first call provider evaluates the patient and gives appropriate orders. Respiratory therapists obtain blood gases and manage any airway issues, and a pharmacist expedites the preparation and delivery of antibiotics. An ICU triage nurse locates a bed if the patient needs a transfer to a higher care unit.

The goal is to get broad-spectrum antibiotics flowing within an hour of the code being called. Fluid resuscitation is administered if there’s an elevated lactate level or a mean arterial pressure of less than 65 mmHg.

To make things even smoother, every nursing unit has prepackaged Code Sepsis equipment bundles complete with intravenous supplies and everything necessary to draw blood and send it off for rapid analysis.

The program, launched in March 2012, is now associated with a number of improved treatment parameters, Dr. Jones said. By September 2012, rapid response for potentially unstable patients rose from 30% to 70%. By last March, there were also significant increases in 100% bundle compliance. Antibiotics are now almost always administered in 1 hour or less of a Code Sepsis initiation. The risk-adjusted mortality index has fallen from a high of about 1.8 before the protocol to less than 1.25 – also a significant improvement.

The road to these improvements wasn’t always smooth, Dr. Jones said. The Code Sepsis model required some fine-tuning along the way. And the very valid concern of antibiotic overuse was not an easy obstacle to surmount.

"That was a big barrier. There is a lot of concern in our organization about the early use of antibiotics. To appropriately deliver those within 1 hour requires a leap of faith and the understanding that you may end up giving antibiotics to some who don’t need them, in order to make sure you give them within 1 hour to every patient who does."

Dr. Jones reported having no financial disclosures.

NATIONAL HARBOR, MD. – A hospital-wide rapid response protocol decreased the time to antibiotic administration and improved sepsis bundle compliance by creating a team approach to sepsis management.

Creating "Code Sepsis" involved every member of the clinical care team, from nursing assistants to attending physicians, Dr. Catherine Jones said at the annual meeting of the Society of Hospital Medicine. The pilot project has been so successful that it’s being implemented in other units of Wake Forest Baptist Medical Center, Winston-Salem, N.C.

"We’ve started a modified version of it in our surgical intensive care units and, in April, our emergency department rolled out this exact same process. Our goal is to roll it out throughout the organization, including our comprehensive cancer center, because it’s not unusual for patients to show up in clinic with severe sepsis," said Dr. Jones, a patient safety and hospital quality expert at the group.

The incidence of sepsis has been on an upward trajectory since 2000, she said, because of a variety of factors. An aging population and the more frequent use of invasive procedures, chemotherapy, and immunosuppressive drugs probably all contribute to the problem. Patients who develop sepsis are eight times more likely to die during a hospital admission.

The good news is that the international "Surviving Sepsis" campaign, which advocated a resuscitation bundle approach designed for early intervention, does seem to be making a difference in outcomes since it was inaugurated in 2002.

The key element seems to be the earlier timing of antibiotic delivery, with an estimate of one patient saved for every seven bundles implemented, Dr. Jones said.

"The problem is that, on the floor, early recognition of sepsis is difficult. It’s a spectrum of illness from infection to sepsis, severe sepsis, and septic shock, and none of these stages have simple, definable signs, symptoms, lab values, or imaging findings. Many other conditions mimic sepsis and present with the same criteria and organ failure as severe sepsis."

Several years ago, the facility began looking at its overall mortality figures, with a goal of reducing mortality by 15%. Since sepsis is an important driver of mortality, it became a target for intervention. A working group composed of interns, nurses, nursing assistants, pharmacists, and respiratory therapists met for 2 days to create an action plan – and Code Sepsis was born. The name was chosen for two reasons, Dr. Jones said: It connotes both a sense of urgency, and the understanding that there is a set protocol to follow.

The rapid response team is the protocol’s backbone. Consisting of two critical care nurses and a first-call provider, the team goes into action any time a patient meets the criteria for possible sepsis. The nurses activate Code Sepsis by calling emergency communications; they quickly the get blood work done and sent to the lab, and administer any initial treatments. The first call provider evaluates the patient and gives appropriate orders. Respiratory therapists obtain blood gases and manage any airway issues, and a pharmacist expedites the preparation and delivery of antibiotics. An ICU triage nurse locates a bed if the patient needs a transfer to a higher care unit.

The goal is to get broad-spectrum antibiotics flowing within an hour of the code being called. Fluid resuscitation is administered if there’s an elevated lactate level or a mean arterial pressure of less than 65 mmHg.

To make things even smoother, every nursing unit has prepackaged Code Sepsis equipment bundles complete with intravenous supplies and everything necessary to draw blood and send it off for rapid analysis.

The program, launched in March 2012, is now associated with a number of improved treatment parameters, Dr. Jones said. By September 2012, rapid response for potentially unstable patients rose from 30% to 70%. By last March, there were also significant increases in 100% bundle compliance. Antibiotics are now almost always administered in 1 hour or less of a Code Sepsis initiation. The risk-adjusted mortality index has fallen from a high of about 1.8 before the protocol to less than 1.25 – also a significant improvement.

The road to these improvements wasn’t always smooth, Dr. Jones said. The Code Sepsis model required some fine-tuning along the way. And the very valid concern of antibiotic overuse was not an easy obstacle to surmount.

"That was a big barrier. There is a lot of concern in our organization about the early use of antibiotics. To appropriately deliver those within 1 hour requires a leap of faith and the understanding that you may end up giving antibiotics to some who don’t need them, in order to make sure you give them within 1 hour to every patient who does."

Dr. Jones reported having no financial disclosures.

NATIONAL HARBOR, MD. – A hospital-wide rapid response protocol decreased the time to antibiotic administration and improved sepsis bundle compliance by creating a team approach to sepsis management.

Creating "Code Sepsis" involved every member of the clinical care team, from nursing assistants to attending physicians, Dr. Catherine Jones said at the annual meeting of the Society of Hospital Medicine. The pilot project has been so successful that it’s being implemented in other units of Wake Forest Baptist Medical Center, Winston-Salem, N.C.

"We’ve started a modified version of it in our surgical intensive care units and, in April, our emergency department rolled out this exact same process. Our goal is to roll it out throughout the organization, including our comprehensive cancer center, because it’s not unusual for patients to show up in clinic with severe sepsis," said Dr. Jones, a patient safety and hospital quality expert at the group.

The incidence of sepsis has been on an upward trajectory since 2000, she said, because of a variety of factors. An aging population and the more frequent use of invasive procedures, chemotherapy, and immunosuppressive drugs probably all contribute to the problem. Patients who develop sepsis are eight times more likely to die during a hospital admission.

The good news is that the international "Surviving Sepsis" campaign, which advocated a resuscitation bundle approach designed for early intervention, does seem to be making a difference in outcomes since it was inaugurated in 2002.

The key element seems to be the earlier timing of antibiotic delivery, with an estimate of one patient saved for every seven bundles implemented, Dr. Jones said.

"The problem is that, on the floor, early recognition of sepsis is difficult. It’s a spectrum of illness from infection to sepsis, severe sepsis, and septic shock, and none of these stages have simple, definable signs, symptoms, lab values, or imaging findings. Many other conditions mimic sepsis and present with the same criteria and organ failure as severe sepsis."

Several years ago, the facility began looking at its overall mortality figures, with a goal of reducing mortality by 15%. Since sepsis is an important driver of mortality, it became a target for intervention. A working group composed of interns, nurses, nursing assistants, pharmacists, and respiratory therapists met for 2 days to create an action plan – and Code Sepsis was born. The name was chosen for two reasons, Dr. Jones said: It connotes both a sense of urgency, and the understanding that there is a set protocol to follow.

The rapid response team is the protocol’s backbone. Consisting of two critical care nurses and a first-call provider, the team goes into action any time a patient meets the criteria for possible sepsis. The nurses activate Code Sepsis by calling emergency communications; they quickly the get blood work done and sent to the lab, and administer any initial treatments. The first call provider evaluates the patient and gives appropriate orders. Respiratory therapists obtain blood gases and manage any airway issues, and a pharmacist expedites the preparation and delivery of antibiotics. An ICU triage nurse locates a bed if the patient needs a transfer to a higher care unit.

The goal is to get broad-spectrum antibiotics flowing within an hour of the code being called. Fluid resuscitation is administered if there’s an elevated lactate level or a mean arterial pressure of less than 65 mmHg.

To make things even smoother, every nursing unit has prepackaged Code Sepsis equipment bundles complete with intravenous supplies and everything necessary to draw blood and send it off for rapid analysis.

The program, launched in March 2012, is now associated with a number of improved treatment parameters, Dr. Jones said. By September 2012, rapid response for potentially unstable patients rose from 30% to 70%. By last March, there were also significant increases in 100% bundle compliance. Antibiotics are now almost always administered in 1 hour or less of a Code Sepsis initiation. The risk-adjusted mortality index has fallen from a high of about 1.8 before the protocol to less than 1.25 – also a significant improvement.

The road to these improvements wasn’t always smooth, Dr. Jones said. The Code Sepsis model required some fine-tuning along the way. And the very valid concern of antibiotic overuse was not an easy obstacle to surmount.

"That was a big barrier. There is a lot of concern in our organization about the early use of antibiotics. To appropriately deliver those within 1 hour requires a leap of faith and the understanding that you may end up giving antibiotics to some who don’t need them, in order to make sure you give them within 1 hour to every patient who does."

Dr. Jones reported having no financial disclosures.

Dr. Patrick Conway, the chief medical officer at the Centers for Medicare and Medicaid Services, recently congratulated hospitalists for their role in bringing down readmission rates.

For decades, hospital readmissions were "rock solid" at between 19% and 19.5%, according to Dr. Conway. But starting in 2012 and continuing into 2013, readmission rates have fallen to about 17.7% nationally.

"That is hundreds of thousands of Medicare beneficiaries that are not readmitted every year," he said.

Speaking at the annual meeting of the Society of Hospital Medicine, Dr. Conway, who still works on a volunteer basis as a pediatric hospitalist on weekends, said the drop in readmissions is a credit to the work of hospitalists and other frontline providers.

The change coincides with the implementation in October 2012 of Medicare’s Readmissions Reduction Program, which reduces payments to hospitals with excess readmissions in the areas of pneumonia, acute myocardial infarction, and heart failure. But Dr. Conway disagrees with those who say that the penalty is solely responsible for the change.

"I think it’s a combination of interventions," Dr. Conway said. "We’re doing QI interventions in the field; there’s payment; there’s support structures; there’s investments in communities. But overall, the major point is we’re moving the national needle."

So what’s next? The CMS is putting readmission measures in place in every setting, including nursing homes, home health settings, and for physicians. The idea is to drive "shared accountability and focus," he said. Also coming down the pike are plans from the CMS to tie a small amount of hospital payments to lowering admission rates. Dr. Conway said that the Medicare officials don’t just want hospitals to be focused on readmissions, but to begin focusing on how to prevent admissions as well.

[email protected]

Dr. Patrick Conway, the chief medical officer at the Centers for Medicare and Medicaid Services, recently congratulated hospitalists for their role in bringing down readmission rates.

For decades, hospital readmissions were "rock solid" at between 19% and 19.5%, according to Dr. Conway. But starting in 2012 and continuing into 2013, readmission rates have fallen to about 17.7% nationally.

"That is hundreds of thousands of Medicare beneficiaries that are not readmitted every year," he said.

Speaking at the annual meeting of the Society of Hospital Medicine, Dr. Conway, who still works on a volunteer basis as a pediatric hospitalist on weekends, said the drop in readmissions is a credit to the work of hospitalists and other frontline providers.

The change coincides with the implementation in October 2012 of Medicare’s Readmissions Reduction Program, which reduces payments to hospitals with excess readmissions in the areas of pneumonia, acute myocardial infarction, and heart failure. But Dr. Conway disagrees with those who say that the penalty is solely responsible for the change.

"I think it’s a combination of interventions," Dr. Conway said. "We’re doing QI interventions in the field; there’s payment; there’s support structures; there’s investments in communities. But overall, the major point is we’re moving the national needle."

So what’s next? The CMS is putting readmission measures in place in every setting, including nursing homes, home health settings, and for physicians. The idea is to drive "shared accountability and focus," he said. Also coming down the pike are plans from the CMS to tie a small amount of hospital payments to lowering admission rates. Dr. Conway said that the Medicare officials don’t just want hospitals to be focused on readmissions, but to begin focusing on how to prevent admissions as well.

[email protected]

Dr. Patrick Conway, the chief medical officer at the Centers for Medicare and Medicaid Services, recently congratulated hospitalists for their role in bringing down readmission rates.

For decades, hospital readmissions were "rock solid" at between 19% and 19.5%, according to Dr. Conway. But starting in 2012 and continuing into 2013, readmission rates have fallen to about 17.7% nationally.

"That is hundreds of thousands of Medicare beneficiaries that are not readmitted every year," he said.

Speaking at the annual meeting of the Society of Hospital Medicine, Dr. Conway, who still works on a volunteer basis as a pediatric hospitalist on weekends, said the drop in readmissions is a credit to the work of hospitalists and other frontline providers.

The change coincides with the implementation in October 2012 of Medicare’s Readmissions Reduction Program, which reduces payments to hospitals with excess readmissions in the areas of pneumonia, acute myocardial infarction, and heart failure. But Dr. Conway disagrees with those who say that the penalty is solely responsible for the change.

"I think it’s a combination of interventions," Dr. Conway said. "We’re doing QI interventions in the field; there’s payment; there’s support structures; there’s investments in communities. But overall, the major point is we’re moving the national needle."

So what’s next? The CMS is putting readmission measures in place in every setting, including nursing homes, home health settings, and for physicians. The idea is to drive "shared accountability and focus," he said. Also coming down the pike are plans from the CMS to tie a small amount of hospital payments to lowering admission rates. Dr. Conway said that the Medicare officials don’t just want hospitals to be focused on readmissions, but to begin focusing on how to prevent admissions as well.

[email protected]

NATIONAL HARBOR, MD. – Angiotensin receptor blockade on the morning of a major orthopedic surgery was associated with a tripling in the risk of acute kidney injury – a link probably driven by the drugs’ association with perioperative hypotension, a retrospective analysis has found.

"Our study does support the idea of withholding beta-blockers on the morning of these major procedures," Dr. Eileen Hennrikus reported at the annual meeting of the Society of Hospital Medicine.

Extant literature has confirmed a link between beta-blockers and hypotension in connection with cardiovascular surgery, but not orthopedic surgery. Dr. Hennrikus, of the Milton S. Hershey Medical Center, Hershey, Penn., examined the link in patients undergoing elective spine fusions, total hip or total knee replacement – procedures typically associated with greater blood loss than cardiovascular surgeries.

She used retrospective data from 922 patients who had undergone any of the three surgeries during 2010. Induction hypotension was defined as a systolic blood pressure of 80 mm Hg for at least 10 minutes within a half hour of anesthesia induction. Intraoperative hypotension was defined as a systolic blood pressure of 80 mm Hg for at least 10 minutes during maintenance anesthesia. The measure of acute kidney injury was an increase in serum creative of at least 0.3 mg/dL, or a 50% increase over preoperative levels.

Of the entire cohort, 37% (343) patients received their prescribed angiotensin receptor blocker or angiotensin converting enzyme inhibitor on the morning of their surgery. The incidence of induction hypotension was significantly greater in those patients in patients who took the medication than in those who did not (12% vs. 6.7%; OR 1.93). Acute kidney injury was also significantly more common among those who had the medications, however (8% vs. 2%; OR 5.4). Intraoperative hypotension was not significantly different between the groups.

Dr. Hennrikus conducted a multivariate analysis to further tease out the relationship between the medications and acute kidney injury. The regression controlled for age, medical comorbidities (diabetes, coronary artery disease, hypertension, and congestive heart failure), body mass index, medications (diuretics, beta-blockers, calcium channel blockers, general anesthesia-induction agents, and vasopressors), and blood loss.

After adjustment for all of those factors, angiotensin blockade conferred a threefold increase in the risk of acute kidney injury (OR 2.97). Although in the primary analysis, intraoperative hypotension had not been significantly associated with the drugs, it more than doubled the risk of kidney injury the risk (OR 2.6) in the multivariate analysis.

The model also revealed something of a surprise, Dr. Hennrikus said: Body mass index was an independent risk factor for acute kidney injury. For every 5 kg/m² increase in BMI, the risk increased by 29% (OR 1.29).

Acute kidney injury exerted its own influence over hospital length of stay and mortality. The length of stay was significantly longer in patients who had kidney injury (6. days vs. 3 days). Two-year mortality was also significantly higher (6% vs. 2%).

Neither Dr. Hennrikus nor her coinvestigators reported having any financial conflicts.

[email protected]

NATIONAL HARBOR, MD. – Angiotensin receptor blockade on the morning of a major orthopedic surgery was associated with a tripling in the risk of acute kidney injury – a link probably driven by the drugs’ association with perioperative hypotension, a retrospective analysis has found.

"Our study does support the idea of withholding beta-blockers on the morning of these major procedures," Dr. Eileen Hennrikus reported at the annual meeting of the Society of Hospital Medicine.

Extant literature has confirmed a link between beta-blockers and hypotension in connection with cardiovascular surgery, but not orthopedic surgery. Dr. Hennrikus, of the Milton S. Hershey Medical Center, Hershey, Penn., examined the link in patients undergoing elective spine fusions, total hip or total knee replacement – procedures typically associated with greater blood loss than cardiovascular surgeries.

She used retrospective data from 922 patients who had undergone any of the three surgeries during 2010. Induction hypotension was defined as a systolic blood pressure of 80 mm Hg for at least 10 minutes within a half hour of anesthesia induction. Intraoperative hypotension was defined as a systolic blood pressure of 80 mm Hg for at least 10 minutes during maintenance anesthesia. The measure of acute kidney injury was an increase in serum creative of at least 0.3 mg/dL, or a 50% increase over preoperative levels.

Of the entire cohort, 37% (343) patients received their prescribed angiotensin receptor blocker or angiotensin converting enzyme inhibitor on the morning of their surgery. The incidence of induction hypotension was significantly greater in those patients in patients who took the medication than in those who did not (12% vs. 6.7%; OR 1.93). Acute kidney injury was also significantly more common among those who had the medications, however (8% vs. 2%; OR 5.4). Intraoperative hypotension was not significantly different between the groups.

Dr. Hennrikus conducted a multivariate analysis to further tease out the relationship between the medications and acute kidney injury. The regression controlled for age, medical comorbidities (diabetes, coronary artery disease, hypertension, and congestive heart failure), body mass index, medications (diuretics, beta-blockers, calcium channel blockers, general anesthesia-induction agents, and vasopressors), and blood loss.

After adjustment for all of those factors, angiotensin blockade conferred a threefold increase in the risk of acute kidney injury (OR 2.97). Although in the primary analysis, intraoperative hypotension had not been significantly associated with the drugs, it more than doubled the risk of kidney injury the risk (OR 2.6) in the multivariate analysis.

The model also revealed something of a surprise, Dr. Hennrikus said: Body mass index was an independent risk factor for acute kidney injury. For every 5 kg/m² increase in BMI, the risk increased by 29% (OR 1.29).

Acute kidney injury exerted its own influence over hospital length of stay and mortality. The length of stay was significantly longer in patients who had kidney injury (6. days vs. 3 days). Two-year mortality was also significantly higher (6% vs. 2%).

Neither Dr. Hennrikus nor her coinvestigators reported having any financial conflicts.

[email protected]

NATIONAL HARBOR, MD. – Angiotensin receptor blockade on the morning of a major orthopedic surgery was associated with a tripling in the risk of acute kidney injury – a link probably driven by the drugs’ association with perioperative hypotension, a retrospective analysis has found.

"Our study does support the idea of withholding beta-blockers on the morning of these major procedures," Dr. Eileen Hennrikus reported at the annual meeting of the Society of Hospital Medicine.

Extant literature has confirmed a link between beta-blockers and hypotension in connection with cardiovascular surgery, but not orthopedic surgery. Dr. Hennrikus, of the Milton S. Hershey Medical Center, Hershey, Penn., examined the link in patients undergoing elective spine fusions, total hip or total knee replacement – procedures typically associated with greater blood loss than cardiovascular surgeries.

She used retrospective data from 922 patients who had undergone any of the three surgeries during 2010. Induction hypotension was defined as a systolic blood pressure of 80 mm Hg for at least 10 minutes within a half hour of anesthesia induction. Intraoperative hypotension was defined as a systolic blood pressure of 80 mm Hg for at least 10 minutes during maintenance anesthesia. The measure of acute kidney injury was an increase in serum creative of at least 0.3 mg/dL, or a 50% increase over preoperative levels.

Of the entire cohort, 37% (343) patients received their prescribed angiotensin receptor blocker or angiotensin converting enzyme inhibitor on the morning of their surgery. The incidence of induction hypotension was significantly greater in those patients in patients who took the medication than in those who did not (12% vs. 6.7%; OR 1.93). Acute kidney injury was also significantly more common among those who had the medications, however (8% vs. 2%; OR 5.4). Intraoperative hypotension was not significantly different between the groups.

Dr. Hennrikus conducted a multivariate analysis to further tease out the relationship between the medications and acute kidney injury. The regression controlled for age, medical comorbidities (diabetes, coronary artery disease, hypertension, and congestive heart failure), body mass index, medications (diuretics, beta-blockers, calcium channel blockers, general anesthesia-induction agents, and vasopressors), and blood loss.

After adjustment for all of those factors, angiotensin blockade conferred a threefold increase in the risk of acute kidney injury (OR 2.97). Although in the primary analysis, intraoperative hypotension had not been significantly associated with the drugs, it more than doubled the risk of kidney injury the risk (OR 2.6) in the multivariate analysis.

The model also revealed something of a surprise, Dr. Hennrikus said: Body mass index was an independent risk factor for acute kidney injury. For every 5 kg/m² increase in BMI, the risk increased by 29% (OR 1.29).

Acute kidney injury exerted its own influence over hospital length of stay and mortality. The length of stay was significantly longer in patients who had kidney injury (6. days vs. 3 days). Two-year mortality was also significantly higher (6% vs. 2%).

Neither Dr. Hennrikus nor her coinvestigators reported having any financial conflicts.

[email protected]

NATIONAL HARBOR, MD. – How confident are you that the medication list your patient is discharged with is completely accurate?

Dr. Jason Stein, a hospitalist at Emory University in Atlanta, said that most physicians don’t know if they’re dealing with "gold" or "garbage."

Dr. Stein, who is a mentor in the Multi-Center Medication Reconciliation Quality Improvement Study (MARQUIS), said that hospitalists are often faced with either taking a "leap of faith" that the medication list is complete and accurate or throwing out the list at discharge and starting over.

MARQUIS, which was launched in 2010 by the Society of Hospital Medicine, is funded by a $1.5-million grant from the Agency for Healthcare Research and Quality. The study seeks to identify and disseminate evidence-based techniques for getting the best possible medication history from hospitalized patients with the ultimate goal of preventing medication errors. Data collection is scheduled to end this year.

The first part of the study was a literature review to cull the best available evidence on medication reconciliation practices in the hospital. That information was then used to develop a toolkit synthesizing those best practices for clinicians.

The toolkit, available online from the Society of Hospital Medicine, includes a manual on best practices and how to adapt them to individual sites, a video on how to take a good medication history, another video comparing the usual versus the optimal approach to discharge counseling, and pocket cards with questions to ask when taking a medication history.

Listen, ‘don’t lead’

"One thing that we’ve found really valuable is to not lead the witness," Dr. Jeffrey L. Schnipper, the MARQUIS principal investigator, said at the annual meeting of the Society of Hospital Medicine.

Don’t just sit there and read the medication list because the patient will likely just say yes to everything, said Dr. Schnipper, who is a hospitalist at Brigham and Women’s Hospital and assistant professor of medicine at Harvard Medical School in Boston. Instead, ask patients to explain what they are taking. The toolkit also provides some standard prompts to use to get patients to remember certain medications, such as ones they take only once a week or medications that are over the counter.

"You will learn a lot more," Dr. Schnipper said. "You will learn about the discrepancies that are really there and you will also be able to assess the patient’s knowledge."

The mentor method

Another big piece of MARQUIS is a mentored implementation project across six hospitals. The mentored implementation was kicked off about a year and a half ago, and the sites are now actively experimenting with interventions aimed at decreasing the number of unintentional medication discrepancies among their noncritical medical and surgical patients.

Each hospital identified a local champion, typically a hospitalist, who formed an interdisciplinary quality improvement team. That team then worked with a MARQUIS mentor, a hospitalist outside the institution with experience in both quality improvement and medication safety. The mentors had monthly phone calls with each of the sites to talk about their processes, successes, and challenges. There are also two site visits built into the study.

During the study, teams at each participating site chose from a menu of interventions recommended by the MARQUIS researchers to try to improve their medication reconciliation process. The menu includes:

• Defining medication reconciliation at their individual site.

• Assigning roles and responsibilities related to medication reconciliation.

• Improving access to preadmission medication information sources.

• Encouraging patients to keep their own up-to-date medication lists.

• Educating providers about how to take the best possible medication history.

• Implementing discharge counseling that includes patient education tools and teach back.

• Stratifying patients as low, intermediate, or high-risk based on disease state and number of medications.

• Improving information technology capability.

The sites are making progress, said Dr. Stein, who serves as a mentor to three of the six hospitals.

Presbyterian Medical Center in Charlotte already had a fairly robust program aimed at improving the medication reconciliation process. Pharmacy technicians, called medication reconciliation assistants (MRAs), are stationed in the emergency department to take a best possible medication history. The MRAs have a thorough process that includes interviewing the patient and then verifying the information with the pharmacy, the primary care physician, or the skilled nursing facility.

But they found that they were still missing patients who came in through the intensive care unit or were directly admitted. Now the center has trained a nurse to take the medication history for any patients who haven’t already seen the MRA and they’ve developed automated systems to identify those patients quickly. The hospital also created an automated system to identify high-risk patients who need a high-intensity discharge counseling session with a clinical pharmacist.

The hospital also has faced hurdles such as the fact that most of the clinicians involved in this process have never been trained on taking a thorough medication history. And even more have never received feedback on the quality of their preadmission medication histories. "This is a core competency that none of us have really ever been trained around," Dr. Stein said.

Another barrier is being clear about who is responsible for making sure each patient has an accurate medication list. "Is it the admitting provider? Unless you’ve actually explicitly outlined whose role that is, it’s nobody’s," Dr. Stein said.

The MARQUIS researchers are collecting data on the number of potentially harmful unintentional medication discrepancies per patient. The researchers collect the raw data from each of the sites and then send it on to physicians who consider the context for the discrepancies and the potential for harm.

Another goal of the project is to try to figure out why certain interventions work in certain places. The researchers are conducting surveys, interviews, direct observation, and focus groups to get these answers. "That will help us understand what components work in what settings," said MARQUIS investigator Amanda H. Salanitro, a hospitalist and health services researcher at Vanderbilt University Medical Center and the VA Tennessee Valley Healthcare System.

[email protected]

On Twitter @MaryEllenNY

NATIONAL HARBOR, MD. – How confident are you that the medication list your patient is discharged with is completely accurate?

Dr. Jason Stein, a hospitalist at Emory University in Atlanta, said that most physicians don’t know if they’re dealing with "gold" or "garbage."

Dr. Stein, who is a mentor in the Multi-Center Medication Reconciliation Quality Improvement Study (MARQUIS), said that hospitalists are often faced with either taking a "leap of faith" that the medication list is complete and accurate or throwing out the list at discharge and starting over.

MARQUIS, which was launched in 2010 by the Society of Hospital Medicine, is funded by a $1.5-million grant from the Agency for Healthcare Research and Quality. The study seeks to identify and disseminate evidence-based techniques for getting the best possible medication history from hospitalized patients with the ultimate goal of preventing medication errors. Data collection is scheduled to end this year.

The first part of the study was a literature review to cull the best available evidence on medication reconciliation practices in the hospital. That information was then used to develop a toolkit synthesizing those best practices for clinicians.

The toolkit, available online from the Society of Hospital Medicine, includes a manual on best practices and how to adapt them to individual sites, a video on how to take a good medication history, another video comparing the usual versus the optimal approach to discharge counseling, and pocket cards with questions to ask when taking a medication history.

Listen, ‘don’t lead’

"One thing that we’ve found really valuable is to not lead the witness," Dr. Jeffrey L. Schnipper, the MARQUIS principal investigator, said at the annual meeting of the Society of Hospital Medicine.

Don’t just sit there and read the medication list because the patient will likely just say yes to everything, said Dr. Schnipper, who is a hospitalist at Brigham and Women’s Hospital and assistant professor of medicine at Harvard Medical School in Boston. Instead, ask patients to explain what they are taking. The toolkit also provides some standard prompts to use to get patients to remember certain medications, such as ones they take only once a week or medications that are over the counter.

"You will learn a lot more," Dr. Schnipper said. "You will learn about the discrepancies that are really there and you will also be able to assess the patient’s knowledge."

The mentor method

Another big piece of MARQUIS is a mentored implementation project across six hospitals. The mentored implementation was kicked off about a year and a half ago, and the sites are now actively experimenting with interventions aimed at decreasing the number of unintentional medication discrepancies among their noncritical medical and surgical patients.

Each hospital identified a local champion, typically a hospitalist, who formed an interdisciplinary quality improvement team. That team then worked with a MARQUIS mentor, a hospitalist outside the institution with experience in both quality improvement and medication safety. The mentors had monthly phone calls with each of the sites to talk about their processes, successes, and challenges. There are also two site visits built into the study.

During the study, teams at each participating site chose from a menu of interventions recommended by the MARQUIS researchers to try to improve their medication reconciliation process. The menu includes:

• Defining medication reconciliation at their individual site.

• Assigning roles and responsibilities related to medication reconciliation.

• Improving access to preadmission medication information sources.

• Encouraging patients to keep their own up-to-date medication lists.

• Educating providers about how to take the best possible medication history.

• Implementing discharge counseling that includes patient education tools and teach back.

• Stratifying patients as low, intermediate, or high-risk based on disease state and number of medications.

• Improving information technology capability.

The sites are making progress, said Dr. Stein, who serves as a mentor to three of the six hospitals.

Presbyterian Medical Center in Charlotte already had a fairly robust program aimed at improving the medication reconciliation process. Pharmacy technicians, called medication reconciliation assistants (MRAs), are stationed in the emergency department to take a best possible medication history. The MRAs have a thorough process that includes interviewing the patient and then verifying the information with the pharmacy, the primary care physician, or the skilled nursing facility.

But they found that they were still missing patients who came in through the intensive care unit or were directly admitted. Now the center has trained a nurse to take the medication history for any patients who haven’t already seen the MRA and they’ve developed automated systems to identify those patients quickly. The hospital also created an automated system to identify high-risk patients who need a high-intensity discharge counseling session with a clinical pharmacist.

The hospital also has faced hurdles such as the fact that most of the clinicians involved in this process have never been trained on taking a thorough medication history. And even more have never received feedback on the quality of their preadmission medication histories. "This is a core competency that none of us have really ever been trained around," Dr. Stein said.

Another barrier is being clear about who is responsible for making sure each patient has an accurate medication list. "Is it the admitting provider? Unless you’ve actually explicitly outlined whose role that is, it’s nobody’s," Dr. Stein said.

The MARQUIS researchers are collecting data on the number of potentially harmful unintentional medication discrepancies per patient. The researchers collect the raw data from each of the sites and then send it on to physicians who consider the context for the discrepancies and the potential for harm.

Another goal of the project is to try to figure out why certain interventions work in certain places. The researchers are conducting surveys, interviews, direct observation, and focus groups to get these answers. "That will help us understand what components work in what settings," said MARQUIS investigator Amanda H. Salanitro, a hospitalist and health services researcher at Vanderbilt University Medical Center and the VA Tennessee Valley Healthcare System.

[email protected]

On Twitter @MaryEllenNY

NATIONAL HARBOR, MD. – How confident are you that the medication list your patient is discharged with is completely accurate?

Dr. Jason Stein, a hospitalist at Emory University in Atlanta, said that most physicians don’t know if they’re dealing with "gold" or "garbage."

Dr. Stein, who is a mentor in the Multi-Center Medication Reconciliation Quality Improvement Study (MARQUIS), said that hospitalists are often faced with either taking a "leap of faith" that the medication list is complete and accurate or throwing out the list at discharge and starting over.

MARQUIS, which was launched in 2010 by the Society of Hospital Medicine, is funded by a $1.5-million grant from the Agency for Healthcare Research and Quality. The study seeks to identify and disseminate evidence-based techniques for getting the best possible medication history from hospitalized patients with the ultimate goal of preventing medication errors. Data collection is scheduled to end this year.

The first part of the study was a literature review to cull the best available evidence on medication reconciliation practices in the hospital. That information was then used to develop a toolkit synthesizing those best practices for clinicians.

The toolkit, available online from the Society of Hospital Medicine, includes a manual on best practices and how to adapt them to individual sites, a video on how to take a good medication history, another video comparing the usual versus the optimal approach to discharge counseling, and pocket cards with questions to ask when taking a medication history.

Listen, ‘don’t lead’

"One thing that we’ve found really valuable is to not lead the witness," Dr. Jeffrey L. Schnipper, the MARQUIS principal investigator, said at the annual meeting of the Society of Hospital Medicine.

Don’t just sit there and read the medication list because the patient will likely just say yes to everything, said Dr. Schnipper, who is a hospitalist at Brigham and Women’s Hospital and assistant professor of medicine at Harvard Medical School in Boston. Instead, ask patients to explain what they are taking. The toolkit also provides some standard prompts to use to get patients to remember certain medications, such as ones they take only once a week or medications that are over the counter.

"You will learn a lot more," Dr. Schnipper said. "You will learn about the discrepancies that are really there and you will also be able to assess the patient’s knowledge."

The mentor method

Another big piece of MARQUIS is a mentored implementation project across six hospitals. The mentored implementation was kicked off about a year and a half ago, and the sites are now actively experimenting with interventions aimed at decreasing the number of unintentional medication discrepancies among their noncritical medical and surgical patients.

Each hospital identified a local champion, typically a hospitalist, who formed an interdisciplinary quality improvement team. That team then worked with a MARQUIS mentor, a hospitalist outside the institution with experience in both quality improvement and medication safety. The mentors had monthly phone calls with each of the sites to talk about their processes, successes, and challenges. There are also two site visits built into the study.

During the study, teams at each participating site chose from a menu of interventions recommended by the MARQUIS researchers to try to improve their medication reconciliation process. The menu includes:

• Defining medication reconciliation at their individual site.

• Assigning roles and responsibilities related to medication reconciliation.

• Improving access to preadmission medication information sources.

• Encouraging patients to keep their own up-to-date medication lists.

• Educating providers about how to take the best possible medication history.

• Implementing discharge counseling that includes patient education tools and teach back.

• Stratifying patients as low, intermediate, or high-risk based on disease state and number of medications.

• Improving information technology capability.

The sites are making progress, said Dr. Stein, who serves as a mentor to three of the six hospitals.

Presbyterian Medical Center in Charlotte already had a fairly robust program aimed at improving the medication reconciliation process. Pharmacy technicians, called medication reconciliation assistants (MRAs), are stationed in the emergency department to take a best possible medication history. The MRAs have a thorough process that includes interviewing the patient and then verifying the information with the pharmacy, the primary care physician, or the skilled nursing facility.

But they found that they were still missing patients who came in through the intensive care unit or were directly admitted. Now the center has trained a nurse to take the medication history for any patients who haven’t already seen the MRA and they’ve developed automated systems to identify those patients quickly. The hospital also created an automated system to identify high-risk patients who need a high-intensity discharge counseling session with a clinical pharmacist.

The hospital also has faced hurdles such as the fact that most of the clinicians involved in this process have never been trained on taking a thorough medication history. And even more have never received feedback on the quality of their preadmission medication histories. "This is a core competency that none of us have really ever been trained around," Dr. Stein said.

Another barrier is being clear about who is responsible for making sure each patient has an accurate medication list. "Is it the admitting provider? Unless you’ve actually explicitly outlined whose role that is, it’s nobody’s," Dr. Stein said.

The MARQUIS researchers are collecting data on the number of potentially harmful unintentional medication discrepancies per patient. The researchers collect the raw data from each of the sites and then send it on to physicians who consider the context for the discrepancies and the potential for harm.

Another goal of the project is to try to figure out why certain interventions work in certain places. The researchers are conducting surveys, interviews, direct observation, and focus groups to get these answers. "That will help us understand what components work in what settings," said MARQUIS investigator Amanda H. Salanitro, a hospitalist and health services researcher at Vanderbilt University Medical Center and the VA Tennessee Valley Healthcare System.

[email protected]

On Twitter @MaryEllenNY

NATIONAL HARBOR, MD. – Accountable Care Organizations are designed around wellness with primary care physicians at the center, but hospitalists still have a major role to play by improving quality and decreasing costs, according to Dr. Emily Mallin.

A look at the 33 quality measures that Medicare requires ACOs to meet shows how integral hospitalists are to the success of these new care delivery models, said Dr. Mallin, a hospitalist at one of the Pioneer ACOs approved by Medicare in 2012. Hospitalists are involved in nearly half of those quality measures, from preventive health to care coordination to the treatment of at-risk populations.

Photo by Mary Ellen Schneider

"While ACOs seem to be population based and outpatient based, the truth of the matter is that [hospitalists] are involved in this," Dr. Mallin, medical director of the academic medical service at Banner Good Samaritan Medical Center in Phoenix, said at the annual meeting of the Society of Hospital Medicine.

Dr. Mallin is part of Banner Health’s Pioneer ACO, a Medicare program that allows mature ACOs, usually located at integrated health systems, a chance to share in the savings generated through better-coordinated care. But this isn't the only type of ACO model.

"ACOs come in all shapes and sizes," Dr. Mallin said.

Officials at the Centers for Medicare and Medicaid Services have also approved more than 220 organizations for the Medicare Shared Savings program, which offers organizations the chance to share in savings only or to take on financial risk but potentially reap higher financial rewards. A third Medicare ACO model, called the Advance Payment Model, allows some small physician-run ACOs to receive up-front, monthly payments to help invest in care coordination. There are also several private payers that have been forming ACOs around the country.

Regardless of the specific structure, all ACOs are designed around the concept of making health care providers accountable for the quality, cost, and overall care of patients.

While the idea of taking on financial risk for the total care of patients sounds similar to the HMO heyday of the 1990s, Dr. Mallin said there are key differences: For instance, the move to HMOs was driven by insurance companies whose sole goal was reducing costs. With today’s ACOs, there is a greater attempt to involve physicians and other health care providers. And while there is a definite focus on reducing costs, ACOs are paid for outcomes, quality, and value. Another key difference, she said, is that there are now significantly more data available at the point of care to help coordinate care.

In the developing ACO world, hospitalists are considered "specialists," Dr. Mallin said. As specialists, hospitalists can participate in more than one ACO at a time. They are also entitled to share in the savings generated by the ACO but exactly how that is done depends on the where they work and how their ACO is structured, she said.

At Banner Health, hospitalists are taking on some new roles with the goal of better care coordination in the ACO. For example, they have an inpatient transitionalist physician who sees patients in the hospital before discharge and then follows them for 30 days after they leave the hospital. He makes house calls and follow-up phone calls to make sure they are receiving adequate care in the community and prevent them from bouncing back to the hospital. "He is extremely patient centered," Dr. Mallin said.

Banner also has a "triage-ist" who is stationed in the emergency department of the hospital and helps in deciding if patients should be admitted, Dr. Mallin said.

The biggest thing to keep in mind about the hospitalist role in ACOs is that this is evolving, said Dr. Edward J. Merrens, a hospitalist and the chief medical officer at Dartmouth-Hitchcock Medical Center in Hanover, N.H., one of the Medicare Pioneer ACOs.

"We don’t have it all figured out yet," he told attendees at the SHM meeting. "This is a new endeavor."

But hospitalists should start to think differently about the care they provide, he said. For instance, ACOs are reversing some long-standing ideas about health care costs, turning hospitals from "profit centers" to "cost centers," he said.

Under the ACO model, ordering an MRI is now a cost to the ACO, not a way to generate revenue for the hospital. Medicare is applying that same principle to hospital readmissions, by penalizing hospitals for having excess return hospitalizations. Dr. Merrins said hospitalists need to start thinking about admissions that way too and trying to prevent them in those patients who return frequently to the hospital for multiple reasons.

"Think about an admission as a readmission," he said.

[email protected]

On Twitter @MaryEllenNY

NATIONAL HARBOR, MD. – Accountable Care Organizations are designed around wellness with primary care physicians at the center, but hospitalists still have a major role to play by improving quality and decreasing costs, according to Dr. Emily Mallin.

A look at the 33 quality measures that Medicare requires ACOs to meet shows how integral hospitalists are to the success of these new care delivery models, said Dr. Mallin, a hospitalist at one of the Pioneer ACOs approved by Medicare in 2012. Hospitalists are involved in nearly half of those quality measures, from preventive health to care coordination to the treatment of at-risk populations.

Photo by Mary Ellen Schneider

"While ACOs seem to be population based and outpatient based, the truth of the matter is that [hospitalists] are involved in this," Dr. Mallin, medical director of the academic medical service at Banner Good Samaritan Medical Center in Phoenix, said at the annual meeting of the Society of Hospital Medicine.

Dr. Mallin is part of Banner Health’s Pioneer ACO, a Medicare program that allows mature ACOs, usually located at integrated health systems, a chance to share in the savings generated through better-coordinated care. But this isn't the only type of ACO model.

"ACOs come in all shapes and sizes," Dr. Mallin said.

Officials at the Centers for Medicare and Medicaid Services have also approved more than 220 organizations for the Medicare Shared Savings program, which offers organizations the chance to share in savings only or to take on financial risk but potentially reap higher financial rewards. A third Medicare ACO model, called the Advance Payment Model, allows some small physician-run ACOs to receive up-front, monthly payments to help invest in care coordination. There are also several private payers that have been forming ACOs around the country.

Regardless of the specific structure, all ACOs are designed around the concept of making health care providers accountable for the quality, cost, and overall care of patients.

While the idea of taking on financial risk for the total care of patients sounds similar to the HMO heyday of the 1990s, Dr. Mallin said there are key differences: For instance, the move to HMOs was driven by insurance companies whose sole goal was reducing costs. With today’s ACOs, there is a greater attempt to involve physicians and other health care providers. And while there is a definite focus on reducing costs, ACOs are paid for outcomes, quality, and value. Another key difference, she said, is that there are now significantly more data available at the point of care to help coordinate care.

In the developing ACO world, hospitalists are considered "specialists," Dr. Mallin said. As specialists, hospitalists can participate in more than one ACO at a time. They are also entitled to share in the savings generated by the ACO but exactly how that is done depends on the where they work and how their ACO is structured, she said.

At Banner Health, hospitalists are taking on some new roles with the goal of better care coordination in the ACO. For example, they have an inpatient transitionalist physician who sees patients in the hospital before discharge and then follows them for 30 days after they leave the hospital. He makes house calls and follow-up phone calls to make sure they are receiving adequate care in the community and prevent them from bouncing back to the hospital. "He is extremely patient centered," Dr. Mallin said.

Banner also has a "triage-ist" who is stationed in the emergency department of the hospital and helps in deciding if patients should be admitted, Dr. Mallin said.

The biggest thing to keep in mind about the hospitalist role in ACOs is that this is evolving, said Dr. Edward J. Merrens, a hospitalist and the chief medical officer at Dartmouth-Hitchcock Medical Center in Hanover, N.H., one of the Medicare Pioneer ACOs.

"We don’t have it all figured out yet," he told attendees at the SHM meeting. "This is a new endeavor."

But hospitalists should start to think differently about the care they provide, he said. For instance, ACOs are reversing some long-standing ideas about health care costs, turning hospitals from "profit centers" to "cost centers," he said.

Under the ACO model, ordering an MRI is now a cost to the ACO, not a way to generate revenue for the hospital. Medicare is applying that same principle to hospital readmissions, by penalizing hospitals for having excess return hospitalizations. Dr. Merrins said hospitalists need to start thinking about admissions that way too and trying to prevent them in those patients who return frequently to the hospital for multiple reasons.

"Think about an admission as a readmission," he said.

[email protected]

On Twitter @MaryEllenNY

NATIONAL HARBOR, MD. – Accountable Care Organizations are designed around wellness with primary care physicians at the center, but hospitalists still have a major role to play by improving quality and decreasing costs, according to Dr. Emily Mallin.

A look at the 33 quality measures that Medicare requires ACOs to meet shows how integral hospitalists are to the success of these new care delivery models, said Dr. Mallin, a hospitalist at one of the Pioneer ACOs approved by Medicare in 2012. Hospitalists are involved in nearly half of those quality measures, from preventive health to care coordination to the treatment of at-risk populations.

Photo by Mary Ellen Schneider

"While ACOs seem to be population based and outpatient based, the truth of the matter is that [hospitalists] are involved in this," Dr. Mallin, medical director of the academic medical service at Banner Good Samaritan Medical Center in Phoenix, said at the annual meeting of the Society of Hospital Medicine.

Dr. Mallin is part of Banner Health’s Pioneer ACO, a Medicare program that allows mature ACOs, usually located at integrated health systems, a chance to share in the savings generated through better-coordinated care. But this isn't the only type of ACO model.

"ACOs come in all shapes and sizes," Dr. Mallin said.

Officials at the Centers for Medicare and Medicaid Services have also approved more than 220 organizations for the Medicare Shared Savings program, which offers organizations the chance to share in savings only or to take on financial risk but potentially reap higher financial rewards. A third Medicare ACO model, called the Advance Payment Model, allows some small physician-run ACOs to receive up-front, monthly payments to help invest in care coordination. There are also several private payers that have been forming ACOs around the country.

Regardless of the specific structure, all ACOs are designed around the concept of making health care providers accountable for the quality, cost, and overall care of patients.

While the idea of taking on financial risk for the total care of patients sounds similar to the HMO heyday of the 1990s, Dr. Mallin said there are key differences: For instance, the move to HMOs was driven by insurance companies whose sole goal was reducing costs. With today’s ACOs, there is a greater attempt to involve physicians and other health care providers. And while there is a definite focus on reducing costs, ACOs are paid for outcomes, quality, and value. Another key difference, she said, is that there are now significantly more data available at the point of care to help coordinate care.

In the developing ACO world, hospitalists are considered "specialists," Dr. Mallin said. As specialists, hospitalists can participate in more than one ACO at a time. They are also entitled to share in the savings generated by the ACO but exactly how that is done depends on the where they work and how their ACO is structured, she said.

At Banner Health, hospitalists are taking on some new roles with the goal of better care coordination in the ACO. For example, they have an inpatient transitionalist physician who sees patients in the hospital before discharge and then follows them for 30 days after they leave the hospital. He makes house calls and follow-up phone calls to make sure they are receiving adequate care in the community and prevent them from bouncing back to the hospital. "He is extremely patient centered," Dr. Mallin said.

Banner also has a "triage-ist" who is stationed in the emergency department of the hospital and helps in deciding if patients should be admitted, Dr. Mallin said.

The biggest thing to keep in mind about the hospitalist role in ACOs is that this is evolving, said Dr. Edward J. Merrens, a hospitalist and the chief medical officer at Dartmouth-Hitchcock Medical Center in Hanover, N.H., one of the Medicare Pioneer ACOs.

"We don’t have it all figured out yet," he told attendees at the SHM meeting. "This is a new endeavor."

But hospitalists should start to think differently about the care they provide, he said. For instance, ACOs are reversing some long-standing ideas about health care costs, turning hospitals from "profit centers" to "cost centers," he said.

Under the ACO model, ordering an MRI is now a cost to the ACO, not a way to generate revenue for the hospital. Medicare is applying that same principle to hospital readmissions, by penalizing hospitals for having excess return hospitalizations. Dr. Merrins said hospitalists need to start thinking about admissions that way too and trying to prevent them in those patients who return frequently to the hospital for multiple reasons.

"Think about an admission as a readmission," he said.

[email protected]

On Twitter @MaryEllenNY