In the News

Readmissions data, use of ICUs for heart failure, and more.


Readmission causes, timing and reductions analyzed

Readmissions of recently hospitalized patients were the focus of several studies published in the Jan. 23/30 Journal of the American Medical Association.

The first study assessed Medicare patients who were hospitalized between 2007 and 2009 for heart failure (HF), acute myocardial infarction (MI) or pneumonia. Within 30 days of discharge, 24.8% of HF patients had been readmitted, as had 19.9% of MI patients and 18.3% of pneumonia patients. However, most of these patients were readmitted for a reason other than the primary diagnosis of their initial hospitalizations. Researchers also looked at the timing of readmissions, finding that the median time to readmission was 10 days for MI and 12 for HF and pneumonia, but that readmissions were frequent throughout the month. Based on these findings, the authors called for broader approaches to preventing readmissions, rather than efforts focused on specific diseases or limited time periods after discharge.

The second study looked at how often recently hospitalized patients had treat-and-release visits to emergency departments (EDs), in addition to being readmitted. The prospective study included more than four million patients age 18 and older hospitalized in California, Florida and Nebraska in 2008 to 2009. Researchers found that 17.9% of hospitalizations resulted in another acute care encounter within 30 days. Most of these encounters were hospital readmissions, but 39.8% were treat-and-release ED visits. The rate of ED visits varied by condition, with digestive disorders and psychosis being the highest-volume causes. The most common reasons for returning to the ED were related to the diagnosis of the initial hospitalization. Study authors concluded that the many studies and interventions focused solely on 30-day readmissions miss these common ED encounters.

Finally, the third study reported on a successful effort to reduce readmissions, conducted by the Centers for Medicare and Medicaid Services. A multicomponent care transitions intervention was applied in 14 communities in 2009 to 2010, and 30-day readmission rates were compared before and after in the intervention communities and 50 control communities. The mean rate of 30-day readmissions dropped in both groups, but more significantly in the intervention communities; intervention communities dropped from 15.21 readmissions per 1,000 beneficiaries per quarter in 2006 to 2008 to 14.34 in 2009 to 2010 versus from 15.03 per 1,000 to 14.72 in control communities over the same time periods (P=0.03). The overall mean rate of hospitalizations followed a similar trend, with the result that readmissions didn't decrease as a percentage of all hospitalizations. The authors noted that the project's interventions (such as better elder care in the community, palliative care counseling, disease management, and care plans) were intended to improve the quality of care transitions.

Combined, the findings of these studies support a patient-centered approach to improving care and reducing readmissions, concluded an accompanying editorial. The studies highlight the fragmentation of care under the current system. Attempts to correct this and reduce readmissions should be broadly focused, using multiple solutions and engaging the community, the editorialist wrote.

Wide variation in use of ICUs for heart failure admissions

Hospitals vary widely in their use of the intensive care unit (ICU) for heart failure patients, a study found.

Researchers analyzed more than 160,000 heart failure (HF) discharges from 341 hospitals in 2009 and 2010. They assessed what percentage of the patients discharged from each hospital had been admitted to the ICU and compared use of various interventions and mortality among the hospitals. Results were published in the Feb. 26 Circulation.

The percentage of heart failure patients admitted to the ICU varied from 0% to 88% with a median of 10% (interquartile range, 6% to 16%). At the hospitals in the top quartile of ICU use, a lower percentage of ICU patients received certain interventions, including mechanical ventilation (6% of ICU patients in top-quartile hospitals vs. 15% in others), noninvasive positive-pressure ventilation (8% vs. 19%), vasopressors and/or inotropes (9% vs. 16%), vasodilators (6% vs. 12%), or any of the above (26% vs. 51%).

The difference in the use of these interventions suggests that the top-quartile hospitals may be admitting relatively healthier patients to their ICUs, the researchers concluded. They noted that those ICUs also had lower mortality among their heart failure patients than the lower-quartile hospitals' ICUs, although there wasn't a difference in overall inpatient mortality. In other words, “hospitals that most frequently triage patients with HF to the ICU may be engaging in a high-cost behavior that does not improve patient outcomes,” the authors said.

They noted that there are no clear standards for ICU admission, and that improved heart failure triage guidelines are needed. However, there are validated risk-assessment models for ICU triage that have not been widely adopted, the authors said. Some individual hospitals have successfully improved their ICU triage systems, they reported. One limitation of the study was that the participating hospitals were voluntary participants in a quality improvement data collection project, so the results might actually underestimate the extent of the problem.

The February issue of ACP Hospitalist featured an article on the challenges of ICU triage decisions. Read it online.

Geriatric conditions may increase risk for heart failure hospitalization

Certain geriatric conditions in older persons with heart failure may be modifiable risk factors for hospital admission, a study found.

Researchers used data from the population-based Cardiovascular Heart Study to determine whether and how geriatric conditions are related to long-term risk for all-cause hospitalization in patients with heart failure. Community-dwelling older patients who had a new diagnosis of heart failure were included, and data from annual exams and medical records were examined. The authors defined geriatric conditions as those that occur in older patients, typically have multiple causes, and are not necessarily related to a specific disease. They looked at slow gait, muscle weakness (i.e., weak grip), cognitive impairment and depressive syndromes and used Anderson-Gill regression modeling to examine whether any of these were related to hospital admission after a heart failure diagnosis. The study results were published in the Feb. 12 Journal of the American College of Cardiology.

A total of 758 patients (mean age at diagnosis, 79.7 years) had a new diagnosis of heart failure, with a mean hospital admission rate of 7.9 per 10 person-years (95% CI, 7.4 to 8.4 per 10 person-years). Diabetes mellitus, New York Heart Association functional class III or IV, chronic kidney disease, slow gait, depressed ejection fraction, depression and muscle weakness were independently associated with hospitalization (hazard ratios, 1.36, 1.32, 1.32, 1.28, 1.25, 1.23 and 1.19, respectively).

The authors noted that data for the study were first collected in 1989 and that heart failure management and possibly risk factors for hospitalization have changed substantially since then. They also acknowledged that patients were censored at the time of death, that the full effect of comorbid conditions was not known, and that the mean patient age was higher than that in most heart failure registries. However, they concluded that three potentially modifiable geriatric conditions—slow gait, depression and muscle weakness—are associated with risk for hospitalization in patients with heart failure and that such risk factors should be assessed routinely at diagnosis.

The author of an accompanying editorial agreed that it is important to examine hospitalization risk factors in older patients but pointed out that the evidence to support risk factor modification is currently weak. Exercise training has only been studied in younger patients, resistance training to improve weak grip has been evaluated only in small cohort studies of exercise endurance, and no consensus has been reached on the effect of treatment for depression, he noted.

The editorialist suggested that based on the current results, it “seems reasonable” for physicians to evaluate muscle strength, gait speed and psychological status in clinically stable patients with newly diagnosed heart failure. “At this time, however, response to slow gait or weak grip is limited to risk stratification, with possible increased follow-up intensity,” he wrote.

Vaccine recommendations updated for pneumococcus, Tdap and flu

The Centers for Disease Control and Prevention's Advisory Committee on Immunization Practices (ACIP) recently issued a new adult immunization schedule that adds the pneumococcal conjugate vaccine, updates Tdap regimens for the elderly and pregnant women, and eliminates the egg-allergy exception for influenza vaccines.

Because current vaccination rates are low, ACIP also urges clinicians to regularly assess patient vaccination histories and implement intervention strategies to increase adherence, the recommendations state. They were published in the Feb. 5 Annals of Internal Medicine.

For the first time, the 13-valent pneumococcal conjugate vaccine (PCV13) was added to the adult schedule. PCV13 should be used with the 23-valent pneumococcal polysaccharide vaccine (PPSV23) for immunocompromised adults, or those with diseases such as HIV, cancer, advanced kidney disease, functional or anatomic asplenia, cerebrospinal fluid leaks or cochlear implants. The schedule also clarifies which adults would need one or two doses of PPSV23 before the age of 65.

Recommendations for the Tdap vaccine have expanded to include routine vaccination of adults age 65 or older and vaccination of pregnant women with each pregnancy. The ideal timing of Tdap vaccination during pregnancy is in the third trimester, between 27 and 36 weeks' gestation. This recommendation was made to safeguard the pregnant woman and her baby, as protective maternal antibodies will pass to the fetus. Infants are too young for the vaccination but are at the highest risk for severe illness or death from pertussis, the recommendations noted. All patients age six months and over should be vaccinated against influenza. Mild egg allergy (hives) is no longer a contraindication, but patients with an egg allergy should get the inactivated flu shot since that is what has been studied.

The FDA has approved a quadrivalent influenza vaccine that contains influenza A (H3N2), influenza A (H1N1), and two influenza B vaccine virus strains, one from each lineage of circulating influenza B viruses. This is meant to increase the likelihood that the vaccine provides cross-reactive antibody against a higher proportion of circulating influenza B viruses.

Beginning with the 2013-2014 season, it is expected that only the quadrivalent formulation will be available. Because a mix of quadrivalent and trivalent influenza vaccines may be available in 2013-2014, the abbreviation for inactivated influenza vaccine has been changed from TIV (trivalent inactivated influenza vaccine) to IIV (inactivated influenza vaccine). The abbreviation for live-attenuated influenza vaccine (LAIV) remains unchanged.

ACP and 16 other medical societies comprise ACIP, which annually reviews the vaccination schedule to ensure it reflects current clinical recommendations for licensed vaccines.