Testing for cellulitis, opioid use recommendations, and more

Summaries from ACP Hospitalist Weekly.


Unnecessary blood tests and imaging common in cellulitis hospitalizations, study finds

The majority of patients with uncomplicated cellulitis seen in an ED received more testing than recommended by guidelines, a recent single-center study found.

To evaluate the appropriateness of tests and imaging for cellulitis, the retrospective cohort study included 183 patients treated in the Massachusetts General Hospital ED between October 2014 and February 2017. All had a presumed diagnosis of uncomplicated cellulitis and were subsequently admitted to either an inpatient medicine or ED observation unit. Results were published as a research letter on April 2 by JAMA Internal Medicine.

Almost a third (32.8%) of the patients received blood cultures, but growth was detected in only one case. Guidelines from the Infectious Diseases Society of America (IDSA) would have called for cultures based on history and vital signs in only 18 patients (10%). Over two-thirds of the patients (67.8%) received imaging. None of those patients were febrile or neutropenic, so all of the imaging was contraindicated by IDSA guidelines. While most patients received no or one form of imaging, 30 (16.4%) underwent two forms, nine (4.9%) underwent three forms, and two (1.1%) underwent four imaging modalities. Ultrasonography was the most common form (46.0%), followed by radiography (29.0%), CT (16.0%), and MRI (6.0%).

The results of imaging changed diagnosis and management in only eight patients: one with hematoma, five with abscesses, and two with osteomyelitis. The researchers looked for patient factors that predicted the use of imaging and found that higher rates of lymphedema and higher serum glucose levels were the only factors that differentiated the patients who received imaging from those who didn't.

The authors gathered data on costs of blood tests and imaging, as well as national rates of hospitalization for cellulitis, to extrapolate the costs of this non-guideline-concordant care. Finding that “the cost of these largely clinically useless diagnostic studies is approximately $226.9 million dollars annually,” they recommended that “imaging and blood cultures should be pursued only in patients who are severely immunocompromised or experiencing systemic toxic effects.”

SHM releases recommendations on opioid use in hospitalized adults with acute, noncancer pain

The Society of Hospital Medicine (SHM) recently released a consensus statement that includes 16 recommendations on the use of opioids in hospitalized adults with acute, noncancer pain.

The consensus statement, which was based on a systematic review of relevant guidelines, applies to patients who are not in palliative, end-of-life, or intensive care settings. The full list of recommendations was published in the April 2018 Journal of Hospital Medicine.

Notable recommendations include the following:

  • Clinicians should limit the use of opioids to patients with either severe pain or moderate pain that has not responded to nonopioid therapy or where nonopioid treatment is contraindicated or anticipated to be ineffective.
  • When using opioids, use the lowest effective dose for the shortest possible duration. To treat acute pain, use immediate-release formulations and avoid long-acting or extended-release formulations, including transdermal fentanyl.
  • Use oral opioids whenever possible. IV opioids should be reserved for patients who cannot take food or medications by mouth, those who may have gastrointestinal malabsorption, and when immediate pain control and/or rapid dose titration is necessary.
  • When initiating opioid therapy, changing from one route of administration to another, or changing from one opioid to another, use an opioid equivalency table or calculator.
  • Clinicians should pair opioids with scheduled nonopioid analgesic medications, unless contraindicated, and always consider using nonpharmacologic pain management strategies.
  • To prevent opioid-induced constipation, clinicians should order a bowel regimen, such as stimulant laxatives, for all hospitalized patients receiving opioids (unless contraindicated). Stool softeners are not recommended.
  • Limit coadministration of opioids with other central nervous system depressant medications to the extent possible.
  • At the start of opioid therapy, clinicians should educate patients, families, and caregivers about the potential risks and side effects, as well as establish realistic goals and expectations for recovery.
  • If issuing an opioid prescription at discharge, clinicians should ask patients about any existing opioid supply at home and account for any such supply. They should prescribe the minimum quantity of opioids anticipated to be necessary.
  • At discharge, clinicians should also ensure that patients, families, and caregivers receive information on how to minimize the risks of opioid therapy (e.g., taking opioids correctly, taking the minimum quantity necessary, safeguarding their supply and disposing any unused supply, and avoiding agents that may potentiate sedative effects).

Medicare readmission penalties don't reflect hospitals' overall readmission rates

Hospitals' performance on Medicare readmission measures are not concordant with their readmission rates for non-Medicare patients or Medicare patients admitted for other conditions, a recent study found.

Researchers used 2013 and 2014 data from the Healthcare Cost and Utilization's all-payer Nationwide Readmission Database to compare hospital-level 30-day readmissions for the conditions that are publicly reported and penalized by Medicare (heart failure, acute myocardial infarction, and pneumonia) to readmission rates for non-Medicare patients with those conditions and Medicare patients with other conditions. Results were published March 27 by Annals of Internal Medicine and appeared in the May 1 issue.

The study found wide variation among the three groups within the same hospitals. The excess readmission ratios for the targeted conditions differed from those for Medicare patients with other conditions by more than 0.1 for 29% of hospitals. The difference between the ratios for Medicare patients with the reported conditions and those for non-Medicare patients with the same conditions was more than 0.1 in 46% of hospitals. The authors concluded that the readmission measures used by Medicare to determine financial penalties have poor agreement with readmission rates for other patients at the same hospitals.

They also noted that among the hospitals with higher readmission rates, the rate for the Medicare-targeted conditions would overestimate that for the non-Medicare patients but underestimate those for the other Medicare patients. Hospitals with larger gaps between their readmission rates for targeted conditions and for other Medicare patients were also found to be more likely to be privately owned, and hospitals with lower readmissions for the Medicare targeted group than either of the other groups were more likely to be metropolitan teaching hospitals, suggesting that the financial incentives may have led these types of hospitals to focus on interventions for the patients and conditions targeted by Medicare, the authors said.

They calculated that more than half of all hospitals would have a change in penalty status if their readmission rates were based on the statistics in the other patient populations. “These findings suggest that measures currently used to evaluate hospital performance on 30-day readmissions often are poor reflectors of hospital performance for other conditions or non-Medicare populations,” they wrote. “Further public reporting about other groups, perhaps through all-payer databases, would present a more accurate representation of hospital performance on readmission among diverse insurance groups.”

An accompanying editorial critiqued some of the study authors' conclusions, noting that the three conditions targeted for readmission penalties “were not intended to define the totality of a hospital's performance.” However, the editorialists agreed with the study authors that current policies could use improvement. “We already have a publicly reported measure that captures the overall hospital quality signal; building on this measure might be the best way to acknowledge the idea that financial incentives under a program to reduce readmissions should be based on more than a set of selected conditions,” the editorial said.