Recent Research

Severity of illness and patient satisfaction scores, obstructive sleep apnea, CA-UTI prevention, and more.


Sickest patients score hospital clinicians lower on responsiveness, communication

Patients who were judged to be at high risk of dying scored their hospitals lowered on multiple domains in postdischarge satisfaction surveys, a recent study found.

Researchers used data from the Hospital Consumer Assessment of Healthcare Providers and Systems (HCAHPS) survey for 5 community hospitals in Michigan. The study included 17,509 HCAHPS medical and surgical respondents, 14.4% of whom were judged to be at high risk of dying, based on age, gender, prior hospitalizations, clinical laboratory values, and diagnoses present on admission. Results were published in the September Journal of Hospital Medicine.

Overall, the high-risk patients were less likely to choose the highest (top box) score for all HCAHPS domains. After the ratings were adjusted for factors including gender, standard HCAHPS risk-adjustment variables, and individual hospital differences, significant differences remained between low- and high-risk patients' scores on 3 domains. High-risk patients were less likely to give a top box score for doctor communication (odds ratio [OR], 0.85; 95% CI, 0.77 to 0.94) or responsiveness of hospital staff (OR, 0.77; 95% CI, 0.69 to 0.85). However, they were more likely to report having received adequate discharge information (OR, 1.30; 95% CI, 1.14 to 1.48).

The study was limited by the small number of hospitals included and the limited availability of HCAHPS scores from high-risk patients. Those who were admitted to hospice or a skilled nursing facility after hospital discharge were excluded from surveying, and only 7% of high-risk patients surveyed responded, compared to 13% of low-risk patients. Still, the results highlight how high-risk patients may have “additional needs for communication, coordination of specialty care, and timely response to the call button,” said the authors, who suggested that this and future studies be used to design a care model to improve both outcomes and patient experience.

An accompanying editorial noted that “it can be challenging to know what to make of HCAHPS scores” in general. But this study serves as a reminder of the importance of being responsive to the needs of seriously ill patients and communicating well with them about their illness, prognosis, and care. “It is a matter of respect,” the editorialist wrote.

Treating obstructive sleep apnea may reduce rapid response events

Treating obstructive sleep apnea in high-risk patients may help reduce rapid response events, according to a recent study.

Over 15 months, researchers at a tertiary care medical center screened 2,590 patients with a BMI of 30 kg/m2 or greater for obstructive sleep apnea at hospital admission. Patients who were admitted to the hospital on weekends or holidays and those who were not obese were not screened, and those who were already being treated for obstructive sleep apnea were excluded. Based on the results of the STOP (Snoring, Tiredness, Observed apneas and Blood Pressure) questionnaire, patients were characterized as at high or low risk for obstructive sleep apnea. For those at high risk, the admitting team could obtain a pulmonary/sleep medicine consultation, consisting of a comprehensive sleep evaluation and inpatient management with positive airway pressure (PAP) therapy if needed. Patients who received PAP therapy were considered adherent if a respiratory therapist noted that they had used PAP for at least 4 hours while they slept. The researchers compared rates of rapid response system use in high-risk and low-risk patients, as well as the effect of PAP therapy on rapid response events. The study results were published online May 11 by PLOS One.

Among the patients screened, 1,973 (76%) were identified as at high risk for obstructive sleep apnea. Rapid response system rates per 1,000 admissions were 43.60 in the high-risk group and 25.91 in the low-risk group. Sixty-nine percent of rapid response system events in the high-risk group had cardiopulmonary causes (e.g., arrhythmia, hypoxia, hypotension, syncope, and respiratory distress) compared with 50% in the low-risk group. High-risk patients who adhered to PAP therapy had significantly lower rapid response event rates than high-risk patients who did not adhere to PAP therapy or did not receive it (16.99, 53.40, and 56.21, respectively; P<0.01).

The authors noted that their study was observational and that their results were not adjusted for baseline characteristics, severity of illness, or reason for hospital admission, among other limitations. However, they concluded that rapid response rates are increased in hospitalized patients at high risk for obstructive sleep apnea and that PAP therapy can help reduce these rates. Because the STOP questionnaire is simple and inexpensive, the authors hypothesized that implementing hospital-wide screening for obstructive sleep apnea would be cost-effective. They called for additional prospective, multicenter studies to confirm their results.

CA-UTI prevention project worked outside ICUs, but not inside

A national effort to reduce catheter-associated urinary tract infections (CA-UTIs) was effective in non-ICU units, but not in ICUs, according to a recent study.

The Comprehensive Unit-based Safety Program (CUSP) was funded by the Agency for Healthcare Research and Quality in 926 inpatient units (59.7% non-ICU, 40.3% ICU) in 603 hospitals around the U.S. The study compared rates of catheter use and CA-UTIs during a 3-month baseline phase, a 2-month CUSP implementation phase, and a year-long sustainability phase. Results were published in the June 2 New England Journal of Medicine.

Overall, implementation of CUSP was associated with decreases in CA-UTI, in both unadjusted analyses (a decrease from 2.82 to 2.19 per 1,000 catheter-days) and adjusted analyses (from 2.40 to 2.05 per 1,000 catheter-days). In non-ICUs, catheter use decreased from 20.1% to 18.8% (incidence rate ratio, 0.93; 95% CI, 0.90 to 0.96; P<0.001) and CA-UTIs dropped from 2.28 to 1.54 per 1,000 catheter-days (incidence rate ratio, 0.68; 95% CI, 0.56 to 0.82; P<0.001). However, in ICUs, both catheter use and CA-UTI rates were largely unchanged after the intervention.

The study's results “suggest that non-ICUs benefited from participating in the program, whereas ICUs did not,” the authors said, adding that the reasons for this difference are “unclear.” Possible explanations include that ICU patients are sick enough to require an indwelling catheter for close monitoring of urine output and that many of them have fever, which leads to culturing of urine and detection of CA-UTI. The study was limited by its nonrandomized design, which could lead to confounding by secular trends in CA-UTI. However, recent CDC data showed a national trend of increasing CA-UTI rates over the studied time period, the authors noted.

A change in the definition of CA-UTI in 2012 could also have affected the study's findings but would likely lead to an underestimate of the program's influence, said an accompanying editorial. The editorialist speculated that since the protocols to prevent CA-UTI have long been well-known, the CUSP program's “overt instruction to build a multidisciplinary team, focus on camaraderie and communication, and attend to failure points may be requisites to success.”

Public reporting did not affect mortality rates for Medicare patients

CMS’ Hospital Compare program has not improved hospital mortality for older patients hospitalized with acute myocardial infarction (MI), congestive heart failure (CHF), or pneumonia, a recent analysis found.

Researchers compared Medicare patients with these conditions to Medicare patients with 12 of the other most common nonsurgical discharge diagnoses. Results were published in the Aug. 2 Annals of Internal Medicine.

The study sample included all hospitalizations for Medicare fee-for-service enrollees (n=20,707,266; mean age, 79 to 80 years; 41% male) with these 15 conditions between 2005 and 2012 at 3,970 hospitals (roughly 85% of acute care hospitals in the U.S.). The Hospital Compare program began reporting mortality for MI, CHF, and pneumonia in 2008, so researchers compared the pre- and post-reporting periods to examine trends in 30-day mortality rates.

None of the targeted 3 conditions showed greater improvement in mortality rates after public reporting, but researchers noted a statistically significant difference between the reported and nonreported conditions (difference in change in trend, 0.07% per quarter; 95% CI, 0.068% to 0.072%). Between the 2 time periods, change in mortality for the 3 publicly reported conditions slowed from −0.23% per quarter to −0.09% per quarter (change, 0.13% per quarter; 95% CI, 0.12% to 0.14%).

Patients in the nonreported group had stroke, sepsis, gastroenteritis and esophagitis, gastrointestinal bleeding, urinary tract infection, metabolic disorder, arrhythmia, and renal failure (those with chronic obstructive pulmonary disease, respiratory infection, and chest pain were excluded due to clinical overlap with the reported conditions, and those with lower-extremity fracture were also excluded). For these conditions, change in mortality slowed slightly from −0.17% per quarter to −0.11% per quarter (change, 0.06% per quarter, 95% CI, 0.05% to 0.07%).

The study authors noted limitations to their analysis, such as how the administrative data used may be limited in ability to account for differences in illness severity between hospitals and across time. They added that they did not use an identical control group, sociodemographic data, or data from hospitals that have opened since 2009.

“We are unsure why public reporting of mortality rates has not accelerated overall improvements in this outcome for reported conditions in U.S. hospitals,” they wrote, noting the manner in which CMS calculates and displays mortality results (with most hospitals being labeled as average or above average) as a possible explanation. “Our findings suggest that expectations for performance improvement from reporting alone should remain limited,” they concluded.

Standard or high-flow nasal oxygen better than non-invasive ventilation for immunocompromised, study finds

For immunocompromised patients with non-hypercapnic acute respiratory failure, non-invasive ventilation was associated with worse outcomes than standard oxygen or high-flow nasal cannula, an analysis found.

Researchers conducted a post-hoc subgroup analysis of a previously published randomized controlled trial comparing standard oxygen, high-flow nasal cannula alone, or non-invasive ventilation interspersed with high-flow nasal cannula. The trial included ICU patients with acute respiratory failure, and this analysis focused on 82 immunocompromised patients. Results were published in the August The Lancet Respiratory Medicine.

Of the 82 patients, 30 received standard oxygen, 26 received high-flow nasal cannula alone, and 26 received non-invasive ventilation (interspersed with high-flow). In the 28 days after randomization, intubation was required for 43% of the standard oxygen patients, 31% of the high-flow patients, and 65% of the non-invasive ventilation patients (P=0.04). The odds ratio for intubation was not significantly different between the standard and high-flow groups, but for the non-invasive ventilation patients compared to the high-flow patients, the odds ratio for intubation was 4.25 (95% CI, 1.33 to 13.56). Researchers found 2 factors independently associated with intubation and mortality: age and non-invasive ventilation as first-line therapy.

The authors concluded that non-invasive ventilation “might be associated with an increased risk of intubation and mortality” in immunocompromised patients with acute hypoxic respiratory failure and “should be used cautiously.” The negative effects might be due to ventilator-induced lung injury, the authors speculated.

The analysis's results contrast with 2 previous studies, they noted, citing recent improvements in ICU care and differences in studied patients as possible explanations. This analysis is limited by its post-hoc nature and the small numbers of patients, so the findings should be confirmed with a randomized controlled trial, the authors said. An accompanying editorial noted that high-flow nasal cannulas may also offer benefits in comfort, convenience, and compliance.

Study examines time to death, predictive factors after palliative extubation

Low systolic blood pressure and high comorbidities may predict death after palliative extubation, a recent study found.

Researchers used data from 148 patients (mean age 78 years; 60% female) who, between September 2010 to August 2013, had undergone palliative extubation in any clinical unit of a 500-bed ethnically diverse community hospital in Queens, N.Y. Results were published in the June issue of Critical Care Medicine.

One-quarter of patients had tracheostomy on admission, 16% were ventilator-dependent, and the top diagnostic categories were sepsis (47%) and primary respiratory failure (22%). The main outcome measures were in-hospital mortality rate, survival rate to discharge, time from palliative extubation to death, and place of discharge for surviving patients.

Of the 148 patients, 114 (77%) died in the hospital, and 34 (23%) were discharged alive, primarily to hospice or nursing home settings. Median survival was 8.9 hours (range, 4 minutes to 7 days) among those who died in the hospital, and for those surviving to discharge, median time to discharge was about 4 days. After palliative extubation, mortality rates were 43% at 12 hours, 56% at 24 hours, 68% at 72 hours, and 79% at 96 hours. Age (<65 or ≥65 years) was not significantly associated with death (P=0.18).

Charlson Comorbidity Index (CCI) scores showed a U-shaped link with death. Patients with high (CCI=6 to 13) comorbidities or no (CCI=0) comorbidities but with a single devastating event had an 83% higher risk of death than those with moderate (CCI=1 to 5) comorbidities (P=0.01). Median survival was longer in patients with normal blood pressure than in patients with hypotension (19 vs. 9 hours), and patients with systolic blood pressure below 90 mm Hg had a 2.5-times higher risk of death (P=0.01).

The study authors noted limitations to the study, such as its retrospective single-center design, the high percentage of chronic ventilator patients in the cohort, and the potential difficulty of using the CCI and systolic blood pressure to prognosticate mortality risk at the bedside. “In a society where the population is aging, we note that age itself was not a significant factor or predictor in mortality versus survival after palliative extubation,” they wrote. “More important is the concept of ‘physiologic age,’ as evidenced by number of comorbidities and hypotension.”

Ceftazidime-avibactam and doripenem equally effective for complicated inpatient UTI

Ceftazidime-avibactam cured complicated urinary tract infections in hospitalized patients as well as doripenem, according to an industry-funded, non-inferiority study.

The trial included 1,033 hospitalized adults with suspected or confirmed complicated urinary tract infections, including acute pyelonephritis. They were randomized to combined ceftazidime (2,000 mg) and avibactam (500 mg) or doripenem (500 mg or adjusted for renal function), both dosed every 8 hours for 10 to 14 days, with the possibility of switching to oral antibiotics at 5 days. Results were published in the Sept. 15 Clinical Infectious Diseases.

The primary efficacy analyses included 393 ceftazidime-avibactam patients and 417 doripenem patients and showed similar rates of patients reporting symptomatic resolution at day 5 (70.2% vs. 66.2%) and having both symptomatic resolution and microbiological eradication at the test of cure, about 3 weeks after initial treatment (71.2% vs. 64.5%). The ceftazidime-avibactam group had a higher rate of eradication at the test of cure (77.4% vs. 71.0%; difference, 6.4% [95% CI, 0.33% to 12.36%]). The ceftazidime-avibactam combination showed a similar safety profile to ceftazidime alone.

The study authors concluded that the combined drug was highly effective for the empiric treatment of complicated urinary tract infections and could be an alternative to carbapenems, reducing use of that drug class. The findings are consistent with previous preclinical and clinical studies, the authors said. They noted that 19.6% of the study's patients had pathogens that weren't susceptible to ceftazidime, and although overall eradication rates were lower in those patients, the compared drugs showed similar effectiveness. The study was funded by a manufacturer of ceftazidime-avibactam.

Score predicts mortality from acute respiratory distress syndrome

Researchers created and tested a new score to predict mortality in patients with acute respiratory distress syndrome (ARDS) in a recent study.

The score was developed and validated in 600 ICU patients who met Berlin criteria for moderate to severe ARDS and were receiving mechanical ventilation. Half of the patients were in the derivation cohort, and half were in the validation cohort. The score was calculated 24 hours after ARDS diagnosis. The primary outcome of the study was inpatient mortality. Results were published in the July Critical Care Medicine.

The score the researchers chose had a minimum of 3 points and a maximum of 9, based on 3 variables: age in years (<47, 47 to 66, >66), Pao2/Fio2 in mm Hg (>158, 105 to 158, <105), and plateau pressure in cm H2O (<27, 27 to 30, >30). Seven other variables were evaluated for potential inclusion in the score but were not selected. Patients with a score greater than 7 were found to have an inpatient mortality rate of 83.3%, compared to 14.5% in patients with a score below 5, the study found.

The score is simple and could potentially be used in real time to calculate the prognosis of patients with ARDS, the study authors said, noting that the results would apply only to those with moderate to severe cases who are on mechanical ventilation. Use of the score could facilitate the rapid implementation of clinical decisions that could alter ARDS management, identify patients who are most likely to benefit from treatment, and help select patients for research trials, the authors said.

“Our findings illustrate that ARDS cannot be viewed as a homogenous disorder. On the contrary, when scoring patients after 24 hours of usual care, patients were grouped more uniformly in three categories of increasing mortality that were also associated with increasing lung dysfunction,” they wrote.

Genetic variants may predict effectiveness of clopidogrel in some populations, study finds

Patients with certain genetic variants were more likely to have a stroke despite clopidogrel treatment after a minor stroke or transient ischemic attack, a recent study from China found.

Researchers conducted a substudy in 2,933 Chinese patients who had an acute minor ischemic stroke or transient ischemic attack. The patients were randomized within 24 hours of symptom onset to either aspirin alone (loading dose of 75 to 300 mg followed by 75 mg daily for 3 months) or the same dose of aspirin plus clopidogrel (loading dose of 300 mg followed by 75 mg daily for 3 months). They were followed for 90 days. Results were published in the July 5 JAMA.

More than half of the studied patients (58.8%) carried variants on the CYP2C19 gene known as loss-of-function alleles. In the 90-day follow-up period, receiving clopidogrel was associated with reduced risk of stroke only among patients without the genetic variant (nonvariant patients: 6.7% in the clopidogrel-aspirin group vs. 12.4% in the aspirin group; variant patients: 9.4% vs. 10.8%; P=0.02 for interaction). The results for the study's secondary outcome—a composite of ischemic stroke, hemorrhagic stroke, myocardial infarction, or vascular death—were similar. The genetic variant did not appear to have any association with bleeding, which was higher in the clopidogrel-aspirin group than in the aspirin group.

Previous small studies have looked at the association between CYP2C19 variants and clopidogrel efficacy in preventing stroke and found differing results in different ethnic groups, the study authors noted. A strength of this study was its inclusion of an aspirin-only control group, but a limitation was its inclusion of only Chinese patients. This and other studies indicate that the relevant genetic variations are more common in East Asian populations than other populations, with previous research showing variant prevalence rates of 18% in Mexican patients and 33% in African-American patients.

“This study provided evidence supporting genetic testing that may allow clinicians to personalize antiplatelet therapy, especially in East Asian patient populations for whom the prevalence of CYP2C19 loss-of-function allele is high,” the authors wrote. They noted that a similar trial based primarily in North America and Europe is currently enrolling patients and may provide the information needed to extrapolate these results to non-Chinese patients.

Acute kidney injury rates similar whether stroke patients received CTA or noncontrast CT

Patients with acute stroke who underwent computed tomographic angiography (CTA) had similar rates of acute kidney injury and creatinine changes as those who had noncontrast head CTs, a new retrospective study found.

The study included 289 ED patients who were treated by a single center's acute stroke intervention team between December 2013 and September 2014. The study was devised primarily to evaluate the safety of empirically using CT contrast due to concerns that time spent measuring patients' serum creatinine levels could delay stroke treatment. The team was encouraged to use CTA when there were no clear contraindications. In total, 157 patients received CTA and 132 had noncontrast head CTs instead. Results were published in the August Stroke.

Although there was a difference between mean creatinine values on presentation, 1.06 mg/dL for the CTA group versus 1.39 mg/dL for the noncontrast head CT group (P=0.004), the groups had similar mean serum creatinine levels between 24 and 48 hours after treatment: 1.06 mg/dL in the CTA group versus 1.40 mg/dL in the noncontrast group (P=0.059). The number of patients with acute kidney injury was also similar between groups (5 vs. 7; P=0.422), as were the changes in creatinine level (−0.07 mg/dL vs. −0.11 mg/dL; P=0.489). The CTA group's door-to-needle time was shorter, but not significantly so, at 68.11 minutes compared to 81.36 minutes (P=0.577). Vascular anomalies were identified in 10.2% of the CTA patients, and high-grade stenosis or occlusion was found in 35%.

American Heart Association guidelines now include a recommendation for noninvasive vascular imaging as part of acute stroke evaluation, and CTA is the most easily and rapidly obtained test for many hospitals, the authors said. In this study, CTA appeared safe with regards to renal function, although the authors noted that patients with diabetes and/or chronic kidney disease are at highest risk for contrast-induced nephropathy. In addition, CTA offered clinical value in rapid identification of vascular abnormalities, according to the authors. They also noted that some study patients may have been selected for noncontrast CT because of a history of renal dysfunction identified through the medical record or patient report; 39.4% of the noncontrast group had a history of chronic kidney disease compared with 15.3% of those undergoing CTA (P<0.001).

Other potential benefits of CTA include that, at least in this study, it was associated with a slightly faster delivery of intravenous tissue plasminogen activator and that it eliminated the need for additional vascular imaging, specifically magnetic resonance angiography, later in the care process, the authors said. They noted that the study was limited by its retrospective and single-center design and that contrast-induced nephropathy may develop more than 48 hours after contrast exposure.