Home care vs. SNF, sepsis mortality

Summaries from ACP Hospitalist Weekly.


Readmissions higher with discharge to home care versus SNF, but overall costs lower

Patients who were discharged to home care had higher readmissions but lower costs and no difference in mortality or function compared to similar patients sent to skilled nursing facilities (SNFs), a recent study found.

The retrospective cohort study used Medicare data from January 1, 2010, to December 31, 2016, on fee-for-service and Medicare Advantage beneficiaries who were discharged from the hospital to home with home health care or to a SNF. The researchers took an instrumental variables approach using the differential distance between the beneficiary's home ZIP code and the closest home health agency and the closest SNF. Results were published by JAMA Internal Medicine on March 11.

Of the more than 17 million studied hospitalizations (62.2% women; mean age, 80.5 years), 38.8% were discharged with home health care and 61.2% were discharged to a SNF. Discharge to home was associated with a 5.6-percentage point higher rate of readmission at 30 days compared with discharge to a SNF (95% CI, 0.8 to 10.3 percentage points; P=0.02). There were no significant differences between groups in 30-day mortality rate or improvement in functional status. Total Medicare payments for postacute care were significantly lower in the home health patients than in the SNF patients (−$5,384; 95% CI, –$6,932 to –$3,837; P<0.001). Total Medicare payments within 60 days of admission were also lower (−$4,514; 95% CI, –$6,932 to −$3,837; P<0.001), despite the higher rate of readmission.

The instrumental variable analysis allowed the study to focus on patients whose discharge location was likely determined by geography, and therefore the results apply to marginal patients (those whose need for home health versus a SNF is borderline and so either setting would be reasonable), the study authors said. The findings are important because incentives to reduce readmissions might push hospitals to discharge patients to SNFs, while alternative payment models such as bundled payments might encourage discharges to home care. “As payment incentives are refined to optimize provider response, balancing incentives to reduce costs with incentives to improve patient outcomes will be important,” the authors said. They noted that the study was limited by the risk of unmeasured confounding and that the results do not apply to non-Medicare patients.

An accompanying editorial considered how discharge destination decisions are made. “Although hospital officials seem to give precedence to patients' ‘choice’ of which SNF or which home health agency they might select (to the point that officials offer no information about provider quality), we know little about whether and how patients are informed about the differences between home and institutional care,” the editorial said. The editorialist also wondered whether the difference in readmissions between groups could be eliminated. “This is a very important question since it may be that a small improvement in clinician quality, better training to improve family engagement, or better targeting of who can use home health care might eliminate excess rehospitalizations without reducing the cost savings,” he wrote.

Door-to-antibiotic time associated with one-year mortality among sepsis patients

Faster antibiotic initiation was associated with improved one-year survival among patients who came to an ED with sepsis, according to a recent study.

The retrospective cohort study included 10,811 adult ED patients admitted with clinical sepsis to any of four U.S. hospitals from 2013 to 2017. The patients' median door-to-antibiotic time was 166 minutes (interquartile range, 115 to 230 minutes), and one-year mortality was 19%. The study found that, after adjustment, each additional hour from ED arrival to antibiotic initiation was associated with a 10% (95% CI, 5% to 14%; P<0.001) increased odds of one-year mortality. Results were published by CHEST on Feb. 16 and appeared in the May issue.

Image by Getty Images
Image by Getty Images

The association between earlier antibiotics and better survival remained linear when each one-hour interval of door-to-antibiotic time was compared to an hour or less. The association was also found on the outcomes of mortality in the hospital and at 30 and 90 days. One-year mortality was higher among patients with a door-to-antibiotic time over three hours compared to three hours or less (adjusted odds ratio, 1.27; 95% CI, 1.13 to 1.43), but the study did not find a significant mortality difference between patients who received antibiotics in under an hour versus those who received them at an hour or more (adjusted odds ratio, 1.26; 95% CI, 0.98 to 1.62).

Based on the results, the study authors calculated that delays in antibiotic administration for sepsis are associated with a 1.1% per hour increase in risk-adjusted absolute mortality. If this is confirmed, “decreasing average door-to-antibiotic time to 1.5 hours could prevent one death per 61 ED sepsis patients, or over four deaths per month just in the EDs included in this study,” they wrote. The linearity of the results also suggests that the improvement in sepsis outcomes associated with early antibiotics may be continuous, rather than equivalent for treatment within any specified time window.

Potential explanations for the apparent long-term effects of later antibiotic therapy could include more severe or enduring sepsis-associated organ failure, increased persistent inflammation, recurrent infection, or worse deconditioning, the authors suggested. They cautioned that the study, which they believe to be the first to analyze the impact of antibiotic timing on longer-term mortality, should be considered hypothesis-generating and will require confirmation and mechanistic investigation. As a retrospective observational study, it is subject to the risk of residual confounding, despite the authors' efforts at adjustment. They called for trials to confirm the findings, which they suggested could randomize patients to prehospital antibiotics or randomize EDs to interventions designed to accelerate sepsis care.

Penicillin skin testing de-escalated antibiotic prescribing, saved patients money

Penicillin skin testing in patients with self-reported allergy was associated with immediate antimicrobial de-escalation in the majority of patients who tested negative, which led to a significant increase in non-carbapenem beta-lactam usage, a quasi-experimental study found.

Researchers evaluated antimicrobial prescribing with and without penicillin skin testing for inpatients at a community hospital. They compared an intervention group of 100 patients who completed penicillin skin testing for a self-reported penicillin allergy to a matched control group of 100 patients who had a listed penicillin allergy as well as an infectious diseases consultation. Results were published online on Feb. 27 by Open Forum Infectious Diseases.

Overall, 98 of 100 patients tested negative for penicillin allergy. Of these, 70 (71%) had immediate changes directly made to their antimicrobial regimens, primarily through intervention of the antimicrobial stewardship program pharmacist working directly with the care team. The most common change was from carbapenems to penicillins (n=34), followed by a switch from a higher-generation cephalosporin to a lower-generation cephalosporin (n=13). In the penicillin skin testing group, beta-lactam (i.e., penicillin or cephalosporin) days of therapy were 666 out of 1,094 (60.88%, with 34.82% being a penicillin specifically). In contrast, beta-lactam days of therapy were 386 out of 984 (39.64%, with 6.4% being a penicillin specifically) in the control group (P<0.00001 for comparison).

Overall, changes to the antimicrobial regimen after penicillin skin testing saved the average patient $353.03 (including the $140 drug supply cost of performing the test) compared to no changes in the pretesting regimen (P=0.045). This calculation of savings included every patient tested, including those who tested positive and those without changes to their antimicrobial regimens. For patients who had direct changes to their regimens after testing, the cost savings increased to an average of $556.91 per patient. “These are likely conservative cost savings estimates considering the true cost of using non beta-lactam agents has been well documented with regards to morbidity and mortality due to decreased effectiveness or adverse effects,” the study authors wrote.

They noted limitations of the study, such as its observational design and the fact that it was performed at a single medical center with an established penicillin skin testing program. In addition, cost-savings calculations were based on local contract prices and did not take into account personnel time and overhead, nor antimicrobial use during future admissions.

The authors emphasized that penicillin skin testing is part of a broader initiative that includes penicillin allergy assessment. “Upon allergy assessment, many patients will not require [penicillin skin testing] and can safely be administered a beta-lactam in cases where allergy is incorrect or given a graded challenge if [the] patient is overall low risk for reaction,” they wrote, adding that skin testing remains the best option for patients with confirmed IgE-mediated reaction to penicillin.

Low-value procedures associated with hospital-acquired complications, study finds

Seven procedures considered low-value care were associated with higher rates of hospital-acquired complications (HACs) in a recent study.

Researchers in Australia performed a cohort study using hospital admission data from 225 public hospitals from July 1, 2014, through June 30, 2017, on seven low-value procedures:

  • endoscopy for dyspepsia in patients younger than age 55 years (3,689 episodes);
  • knee arthroscopy for osteoarthritis or meniscal tears (3,963 episodes);
  • colonoscopy for constipation in patients younger than age 50 years (665 episodes);
  • endovascular repair of abdominal aortic aneurysm (AAA) in asymptomatic, high-risk patients (508 episodes);
  • carotid endarterectomy in asymptomatic, high-risk patients (273 episodes);
  • renal artery angioplasty (176 episodes); and
  • spinal fusion for uncomplicated low back pain (56 episodes).

The study included a low-value care episode only if it was listed as a patient's principal procedure, which usually means it is the reason for hospital admission, the authors said. Harm was determined by rates of 16 associated HACs determined by the Australian Commission on Safety and Quality in Health Care to be appropriate measures for monitoring quality and safety of care. The rate of HACs with each low-value procedure and the associated change in mean length of stay were the study's main outcome measures. Results were published Feb. 25 by JAMA Internal Medicine.

Of the seven low-value procedures studied, endoscopy, knee arthroscopy, and colonoscopy had relatively low HAC rates (0.1% [95% CI, 0.02% to 0.2%], 0.5% [95% CI, 0.2% to 0.7%], and 0.3% [95% CI, 0.0% to 0.9%], respectively). Rates were higher for spinal fusion (7.1%; 95% CI, 2.2% to 11.5%), endovascular AAA repair (15.0%; 95% CI, 11.1% to 19.7%), carotid endarterectomy (7.7%; 95% CI, 5.2% to 10.1%), and renal artery angioplasty (8.5%; 95% CI, 5.8% to 11.5%). Among all of the HACs observed, 83 (26.3%; 95% CI, 21.8% to 31.5%) were health care-associated infections, which was the most common HAC for most of the seven procedures. Median length of stay was two or more times greater in patients with an HAC versus without for all seven of the low-value procedures.

The study looked only at a certain set of HACs during episodes of low-value care and did not consider other related hospital admissions or follow-up management, the researchers noted. They concluded that low-value care is associated with poor outcomes and said that future research should consider the consequences for the health care system as well as the potential financial, psychological, and psychosocial harm to patients. “The full burden of low-value care for patients and the health care system is yet to be quantified,” they wrote.