LTAC use, med adherence after MI, and more

Summaries from ACP Hospitalist Weekly.


LTAC use varies more by region and hospital than by patient-level factors

Some hospitals discharge significantly more of their patients to long-term acute care hospitals (LTACs) than others, according to a study.

The retrospective observational cohort study included 65,525 hospitalized patients ages 65 years or older who were transferred from an acute care hospital to an LTAC or a skilled nursing facility (SNF) during fiscal year 2012; 64.8% were women, and 60.9% were 85 years or older. Results were published by JAMA Internal Medicine on Feb. 5 and appeared in the March issue.

Overall, 4.7% of the patients were transferred to an LTAC. The study found 29 patient-level, three hospital-level, and five region-level independent predictors of LTAC transfer; the strongest predictors were having a tracheostomy (adjusted odds ratio [aOR], 23.8; 95% CI, 15.8 to 35.9) and being hospitalized in close proximity to an LTAC (aOR for 0 to 2 miles vs. >42 miles, 8.4; 95% CI, 6.1 to 11.5). After adjustment for case mix, differences between patients explained only 52.1% of the variation in LTAC use. Regional differences explained 32.9%, with use of LTACs being much higher in the South (17% to 37%) than the Pacific Northwest, North, and Northeast (<2.2%).The remainder of the variation (15.0%) was attributable to differences between hospitals. Even within a region, adjusted hospital LTAC transfer rates varied substantially.

The results support the findings of previous similar studies and “highlight opportunities to optimize the use of LTACs for patients who could not be effectively treated in less intensive settings, such as SNFs,” according to the authors. From a clinical perspective, LTACs do offer benefits to patients such as daily physician care and better nurse-to-patient ratios, but they also carry risks such as higher hospital-acquired infection rates. “The demand for LTACs is in part because of a perceived lack of adequately staffed, high-quality SNFs capable of providing complex nursing and medical therapy,” the authors said.

They noted that as of October 2018 LTACs will receive lower, site-neutral payments for patients who haven't had an ICU stay of at least three days or prolonged mechanical ventilation during hospitalization. However, it isn't clear whether this policy will lead to clinically appropriate reductions in LTAC use, according to the study authors, who called on payers and accountable care organizations to partner with high-quality SNFs and hospitals with low LTAC use to develop more efficient networks.

A separate study, published by Annals of Internal Medicine online on Feb. 13 and in the May 15 issue, looked at the use of postacute care (PAC) facilities by Medicare beneficiaries at the end of life. Among more than 8 million patients who died between 2006 and 2011, 23.3% received care in a PAC facility in the last 90 days of life. Of those patients, 60.4% returned to an acute care hospital before death, most directly from the PAC facility. Among all studied patients, 39.0% used hospice services in the last year of life, and those patients had lower use of acute care and PAC facilities in the days and weeks before death.

The results show high rates of hospital readmission from PAC facilities at the end of life, which may lead to lower-quality end-of-life care for these patients and higher costs for the health care system, according to the study authors. The findings “could inform conversations with patients near the end of life about their setting and goals of care,” they said. “Policies that require documentation of advanced care planning before transfer to a PAC facility, including discussion of do-not-hospitalize orders, could help facilitate these conversations.”

Med adherence, outcomes after MI vary by hospital

By 90 days postdischarge after acute myocardial infarction (MI), patient outcomes and adherence to secondary prevention medications appear to vary by hospital, a study found.

Researchers used data from nearly 20,000 Medicare patients discharged after MI from 347 U.S. hospitals participating in the Acute Coronary Treatment and Intervention Outcomes Network Registry–Get With The Guidelines from Jan. 2, 2007, to Oct. 1, 2010. The study used Medicare Part D prescription filling data to determine patients' medication adherence, defined as the proportion of days covered more than 80% within 90 days after discharge, which was compared with two-year major adverse cardiovascular events. Results were published online on Jan. 31 by Circulation and appeared in the May 15 issue.

Overall rates of adherence were 68% for beta-blockers, 63% for statins, 64% for angiotensin-converting enzyme (ACE) inhibitors and angiotensin receptor blockers (ARBs), and 72% for thienopyridines. Adherence rates varied significantly among hospitals: 59% to 75% for beta-blockers, 55% to 69% for statins, 57% to 69% for ACE inhibitors and ARBs, and 64% to 77% for thienopyridines.

Compared to hospitals in the lowest quartile of 90-day composite medication adherence, patients at hospitals with the highest adherence had lower two-year risk of major adverse cardiovascular events (27.5% vs. 35.3%; adjusted hazard ratio, 0.88; 95% CI, 0.80 to 0.96). Hospitals with the highest adherence also had lower adjusted rates of patients' dying or being readmitted compared to those with the lowest adherence (hazard ratio, 0.90; 95% CI, 0.85 to 0.96).

The study authors noted limitations of the analysis, including that they could not determine whether patients actually took the dispensed medications and a lack of data on individual hospital practices. They added that they did not capture adherence data for patients not enrolled in Medicare Part D or on alternate prescription coverage.

“Patient adherence is a critical component of an evolving healthcare system that is now prioritizing quality of care over traditional fee-for-service. We believe an understanding of factors associated with adherence on a hospital-level, in addition to those on a patient-level, provide a novel framework on which to further engage with the issues of poor medication adherence in healthcare,” the authors concluded.

No change in inpatient mortality with increased emergency calls

Increases in medical emergency (or rapid-response) calls were not associated with improvements in hospital mortality, an Australian study found.

Researchers conducted two retrospective analyses. The first included all 441,029 patients discharged from one hospital in Melbourne from January 2008 to June 2016. The second included all 3.3 million patients discharged from 15 hospitals in the state of Victoria from July 2010 to June 2015. Results were published by Critical Care Medicine on March 29 and appeared in the July issue.

In the first study, patients' median age was 61 years, 57.2% were men, and 0.70% died. Monthly emergency call rates varied between 9.21 and 30.69 per 1,000 discharges, with a median of 18.4 per 1,000 discharges. Being treated at the hospital during a month with a higher calling rate was not associated with any improvement in mortality, after adjustment for age, gender, emergency status, same-day admission, year of discharge, and comorbidities.

The second study confirmed the findings of the first. Its population also had a median age of 61 years; 51.4% were men, and the hospital mortality rate was 0.83%. Emergency call rates varied from 18.46 to 33.40 per 1,000 discharges, with a median of 25.75 per 1,000 discharges, and their frequency was not associated with mortality rates. The study authors noted that previous research has shown improvements in mortality with higher calling rates but that these studies failed to adjust for overall improvements in hospital mortality over time, patient factors, and case mix.

“The success of medical emergency or rapid response teams has led to an increasing number of calls in many hospitals. This increase has often occurred naturally but sometimes encouraged in the belief that increased calling rates lead to reduced mortality,” the authors wrote. The worldwide generalizability of their study is limited and it may need to be confirmed with local data, but the current results suggest that efforts to increase rapid-response call rates do not seem warranted, the authors said.

Haloperidol increases in-hospital mortality after acute MI compared to atypical antipsychotics

In patients admitted with acute myocardial infarction (MI) who receive antipsychotic drugs, haloperidol may increase the risk of in-hospital death compared to an atypical antipsychotic, a study found.

Researchers used a health care database to assess 6,578 adult patients (mean age, 75.2 years) who, between 2003 and 2014, received antipsychotic drugs during an admission for a primary diagnosis of acute MI at more than 700 U.S. hospitals. The primary outcome was in-hospital mortality within seven days of initiating oral haloperidol or oral atypical antipsychotics (olanzapine, quetiapine, or risperidone). Results were published March 28 by The BMJ.

Overall, 1,668 (25.4%) patients received haloperidol, and 4,910 (74.6%) received atypical antipsychotics. Although the groups had similar mean times from admission to treatment initiation (5.3 d vs. 5.6 d, respectively) and length of stay (12.5 d vs. 13.6 d, respectively), the mean duration of treatment was shorter in the haloperidol group (2.4 d vs. 3.9 d). The researchers used 1:1 propensity score matching to adjust for confounding factors.

In intention-to-treat analyses, the absolute rate of death per 100 person-days was 1.7 in the haloperidol group (129 deaths) and 1.1 in the atypical antipsychotic group (92 deaths) during seven days of follow-up. The unadjusted and adjusted hazard ratios of death for patients on haloperidol were 1.51 (95% CI, 1.22 to 1.85) and 1.50 (95% CI, 1.14 to 1.96), respectively. The adjusted hazard ratio peaked on the day of antipsychotic initiation (3.28; 95% CI, 1.75 to 6.23) and decreased by 19% each day. By day 5, the increased risk disappeared (1.12, 95% CI, 0.79 to 1.59).

In an as-treated analysis, the unadjusted hazard ratio for death within seven days was 1.90 (95% CI, 1.43 to 2.53), and the adjusted hazard ratio was 1.93 (95% CI, 1.34 to 2.76). The larger hazard ratios in this analysis suggested that haloperidol's potential adverse effects may be more pronounced while patients are taking the drug, the study authors noted.

They noted limitations of the study, such as a lack of information about patients prior to admission and about the true indication for antipsychotic use. “[A]lthough haloperidol has long been used to manage agitation or related symptoms for patients admitted to hospital, our findings suggest that atypical antipsychotics may be a less harmful option in older populations with acute myocardial infarction who require an off-label antipsychotic for severe agitation,” the authors concluded.

Frailty associated with failure to rescue after surgery, study finds

Frail patients who undergo surgery appear to be at higher risk for failure to rescue, a study found.

Photo by Thinkstock
Photo by Thinkstock

Researchers used data from the National Surgical Quality Improvement Program to assess patients who had inpatient general, vascular, thoracic, cardiac, and orthopedic surgery between Jan. 1, 2005, and Dec. 31, 2012. Patients were stratified into five frailty groups according to Risk Analysis Index (RAI) score (≤10, 11 to 20, 21 to 30, 31 to 40, and >40), and procedures were categorized as having a high or low mortality risk (>1% or ≤1%, respectively). The number of postoperative complications and inpatient failure to rescue were the primary outcome measures. Study results were published March 21 by JAMA Surgery and appeared in the May 16 issue.

Overall, 984,550 patients were included in the study. Mean age was 58.2 years, and 55.8% were women. Most of the patients (72.7%) had general surgery; 16.2% had vascular surgery, 1.2% had thoracic surgery, 1.3% had cardiac surgery, and 8.7% had orthopedic surgery. Major complications after low-risk or high-risk surgery became more likely as RAI score increased, with rates of 3.2% and 13.5% in the lowest RAI group and 36.4% and 54.4% in the highest group, respectively. When the researchers stratified by number of complications, failure to rescue rates increased significantly across RAI groups after low- and high-risk procedures.

Compared with patients who had an RAI score of 10 or less, odds ratios for failure to rescue after a low-risk procedure and one major complication were 5.3 for patients with RAI scores of 11 to 20, 8.1 for those with RAI scores of 21 to 30, 22.3 for those with RAI scores of 31 to 40, and 43.9 for those with RAI scores above 40. The same increased risk was seen for patients who had a high-risk procedure and one major complication (odds ratios, 2.5, 5.1, 8.9, and 18.4, respectively).

The authors noted that the National Surgical Quality Improvement Program was not designed to assess frailty and that the data file does not provide hospital identifiers or information on clinician or hospital factors that could affect failure to rescue, among other limitations. However, they concluded that patient frailty before surgery appears to be a significant risk factor for postoperative complications and failure to rescue, even when surgical procedures are considered to be low risk. They recommended that assessment of frailty should be built into quality improvement efforts and preoperative conversations about informed consent and that future research should aim to establish clear definitions of frailty to help clinicians better identify at-risk patients.

“In the meantime, our data suggest the risk-benefit ratio of performing even minor elective procedures needs to be carefully considered in individuals who are considered frail,” the authors wrote.

The authors of an accompanying editorial praised the study's strengths, including its inclusion of low-risk procedures, but noted that it does not identify a frailty cutoff at which surgical risk becomes prohibitive and that the RAI has not been directly compared with established clinical markers of frailty. They also noted that the fact that frail patients are at higher risk is not surprising and that the more interesting question is whether that risk can be modified. “The challenge now is how to increase access to resource-intensive multi-disciplinary prehabilitation programs for more patients to benefit from effective surgical therapies,” the editorialists wrote.

Endovascular treatment as effective in routine clinical practice as in trials

Endovascular treatment for patients with acute ischemic stroke due to proximal intracranial vessel occlusion in the anterior circulation is at least as effective and safe in routine clinical practice as in the setting of a randomized controlled trial.

Researchers conducted an ongoing, prospective, observational cohort study among 16 centers that perform endovascular treatment in The Netherlands between March 2014 and June 2016. The Multicentre Randomised Controlled Trial of Endovascular Treatment for Acute Ischaemic Stroke in the Netherlands (MR CLEAN) Registry tracked 1,488 patients who had received endovascular treatment, including stent retriever thrombectomy, aspiration, and all alternative methods for acute ischemic stroke, within 6.5 hours from onset of symptoms.

The primary outcome was the modified Rankin Scale (mRS) score, ranging from 0 (no symptoms) to 6 (death) at 90 days after the onset of symptoms. Outcomes and safety variables in the MR CLEAN Registry were compared with the MR CLEAN trial intervention and control arms. Results were published by The BMJ on March 9.

Better functional outcomes were found in patients in the MR CLEAN Registry compared with the MR CLEAN trial intervention arm (adjusted common odds ratio [OR], 1.30; 95% CI, 1.02 to 1.67; P=0.03) and the MR CLEAN trial control arm (adjusted common OR, 1.85; 1.46 to 2.34; P<0.01). The reperfusion rate, with successful reperfusion defined as a score of 2B-3 on the extended thrombolysis in cerebral infarction score, was the same for patients in both groups (58.7%).

Other outcomes included duration from onset of stroke to start of endovascular treatment and from onset of stroke to successful reperfusion or last contrast bolus, and these statistics were one hour shorter for patients in the MR CLEAN Registry. Symptomatic intracranial hemorrhage occurred in 5.8% of patients in the MR CLEAN Registry compared with 7.7% in the MR CLEAN trial intervention arm and 6.4% in the MR CLEAN trial control arm.

The findings confirm that endovascular treatment should be standard of care for patients with acute ischemic stroke due to proximal intracranial vessel occlusion in the anterior circulation, the authors wrote. They noted that the registry included a broad group of patients without imaging selection, while clinical trials select patients before treatment with a favorable prognosis based on imaging characteristics.

The authors called for additional research to further improve the implementation of endovascular treatment in routine practice. “Research is needed to improve existing patient selection paradigms and tools, particularly those combining clinical and imaging variables, and to provide consensus on how patient selection for endovascular treatment can be reliably and consistently implemented,” they wrote.

Hospitalizations rose for acute heart failure in past decade

Hospitalizations for acute decompensated heart failure (HF) increased between 2005 and 2014, but survival rates improved for black patients.

The Atherosclerosis Risk in Communities (ARIC) Study sampled HF-related hospitalizations among patients age 55 years and older in four U.S. communities using ICD-9-CM codes. Hospitalizations were validated by standardized physician review and/or computer algorithm, yielding 40,173 events after accounting for sampling design. Results were published by Circulation on March 8 and appeared in the July 3 issue.

Of the acute decompensated HF hospitalizations, 50% of patients had reduced ejection fraction and 39% had preserved ejection fraction. Reduced ejection fraction was more common in black men and white men, whereas preserved ejection fraction was most common in white women. Average age-adjusted rates of acute decompensated HF were highest in black patients (38.1 per 1,000 black men, 30.5 per 1,000 black women), with rates differing by HF type and sex.

Acute decompensated HF rates increased over the 10 years, with average annual percent changes (AAPC) of 4.3% in black women, 3.7% in black men, 1.9% in white women, and 2.6% in white men. These figures mostly reflect more acute preserved ejection fraction HF, the authors wrote. Age-adjusted 28-day and one-year case-fatality proportions were approximately 10% and 30%, respectively, and they were similar across race-sex groups and HF types.

Only black patients showed decreased one-year mortality over time (AAPC, −5.4% for black women and −4.6% for black men), with rates differing by heart failure type (AAPC, −7.1% for black women with preserved ejection fraction and −4.7% for black men with reduced ejection fraction). Survival at one year was poor regardless of ejection fraction but improved over time for black patients, the authors noted.

“Given the focus of providers, payors, and health care systems to decrease readmission rates for ADHF [acute decompensated heart failure], seeing these survival trends in both first and recurrent hospitalized ADHF may translate to more attention on the medical management of hospitalized patients with HF,” the authors wrote.