Mixed Results With Plan to Cut Antimicrobial Use in Advanced Dementia

A multi-pronged antimicrobial stewardship intervention failed to show statistically significant benefits in nursing home residents with advanced dementia, although it did offer some clinical wins, researchers reported.

In the cluster randomized TRAIN-AD trial of 28 Boston-area nursing homes, the intervention to improve the management of suspected urinary tract infections (UTIs) and lower respiratory infections (LRIs) in this patient population led to a nonsignificant 33% reduction in antimicrobial treatment courses for suspected UTIs or LRIs per person-year in the intervention versus the control arm, for an adjusted marginal rate difference of −0.27 95% CI, −0.71 to 0.17), according to Susan L. Mitchell, MD, MPH, of Hebrew SeniorLife at the Hinda and Arthur Marcus Institute for Aging Research, in Boston, and co-authors.

But that 33% could be seen as “a clinically significant… absolute reduction in antimicrobial courses for suspected LRIs and UTIs among facilities assigned to the intervention,” noted Preeti N. Malani, MD, and Shiwei Zhou, MD, both of the University of Michigan in Ann Arbor, in an invited commentary accompanying the study.

There were some other boons as well, the authors reported, such as the reduction “primarily attributable to reduced antimicrobial use for LRIs,” and a significantly lower use of chest x-ray in the intervention arm (adjusted marginal rate difference −0.56, 95% CI −1.10 to −0.03), they wrote in JAMA Internal Medicine.

While Malani and Zhou called the study “unique,” they acknowledged that it was “underpowered to show a statistically significant reduction” in terms of the main outcomes, which included “burdensome procedures,” such as bladder catheterization. The authors noted that data collection ended early in March 2020 because of pandemic-related nursing home lockdown.

Mitchell and co-authors explained that the “intervention content integrated best practices from infectious diseases and palliative care for management of suspected UTIs and LRIs in residents with advanced dementia.”

These components were geared to nursing home practitioners (physicians, physician assistants, nurse practitioners, and nurses, who made up the majority at about 78%), and included an in-person seminar, an online course, management algorithms (posters), communication tips (pocket cards), and feedback reports on antimicrobial prescribing.

In terms of the algorithms, they “operationalized 2 main considerations for antimicrobial use: (1) alignment with residents’ preferences, and (2) presence of consensus-based minimal criteria for treatment initiation,” the authors explained.

There was a 3-month start-up period to implement the intervention that included a 1-hour in-person training seminar, or a 10-minute one-on-one orientation with project directors, for the “targeted practitioners.” Seminars were repeated every 6 months. Participants also took a 45-minute online course on infection management in advanced dementia, with pre- and post-test knowledge assessment.

The intervention did offer “rewards,” for completion such as a $50 gift card, CME credits, and the opportunity to enter a raffle for a new laptop.

The “residents’ health care proxies received a booklet about infections in advanced dementia. Nursing homes in the control arm continued routine care,” they wrote. The study was done from August 2017 to April 2020, with outcomes measured for as many as 12 months, they added.

The intervention arm had 199 residents with a mean age of 87.7, 81.9% of whom were white women. The control arm had 227 residents (mean age 85.3; 83.7% women, 88.1% white). Also, dementia related to Alzheimer’s disease was present in 44.7% of the patients in the intervention arm and 57.3% in the control arm. “Directives to withhold antimicrobial use were rare,” the authors noted.

During the 12-month study period, 40.7% and 49.3% of residents in the intervention and control arms, respectively, had at least one burdensome procedure for a suspected UTI or LRI, but secondary outcomes did not differ significantly between arms:

• Bladder catheterizations: 24.6% in the intervention arm versus 19.8% in the control arm, adjusted marginal rate of procedures per person-year 0.47 (95% 0.21 to 0.73) versus 0.37 (95% 0.25 to 0.49), adjusted marginal rate difference 0.11 (95% −0.18 to 0.39).
• Venous blood sampling: 30.2% versus 32.6%, 0.64 (95% 0.28 to 1.00) versus 0.76 (95% CI 0.41 to 1.11), −0.12 (95% CI -0.62 to 0.38.
• Hospital transfers: 11.6% versus 8.8%, 0.18 (95% CI 0.04 to 0.31) versus 0.14 (95% CI 0.07 to 0.20), 0.04 (95% −0.11 to 0.19).
• Antimicrobials initiated when minimal criteria were absent: 20.6% versus 22.0%, 0.37 (95% 0.15 to 0.60) versus 0.43 (95% 0.23 to 0.62), −0.05 (95% CI −0.35 to 0.24).

In terms of healthcare professional participation, 88.4% of practitioners completed either the online course or the training seminar, while 63.8% completing both activities. Among nurses, 95% completed one or the other, while 65% did both, leading Mitchell’s group to conclude that there “was high practitioner adherence to the training components of the TRAIN-AD intervention.”

As for the reduction in suspected LRIs. these are common events, and more “aggressive treatment of LRIs has been associated with greater discomfort and little clinical benefit in this population…a program integrating infectious disease and palliative principles that promotes less aggressive care for suspected LRIs has potential clinical relevance,” they wrote.

Malani and Zhou pointed out that the intervention used in TRAIN-AD was “resource and time-intensive… it may be months to years before there is adequate bandwidth to shift focus to antimicrobial stewardship.”

However, they stressed that the “response from nursing homes during the COVID-19 pandemic has showcased their ability to be nimble and to adapt” so Mitchell’s group has provided “a roadmap to help curb inappropriate antibiotic use. The intervention appeared to be particularly effective in decreasing inappropriate antibiotic prescribing for suspected LRIs, which has increased during the pandemic.”

1. An antimicrobial stewardship program led to a nonsignificant reduction in antimicrobial use in a pragmatic cluster randomized clinical trial among nursing home residents with advanced dementia.

2. The use of chest x-ray use was significantly lower in the intervention arm, but other burdensome procedures, such as bladder catheterization and hospital transfers, were unchanged.

Shalmali Pal, Contributing Writer, BreakingMED™

Cat ID: 728

Topic ID: 498,728,282,404,494,728,791,730,190,520,33,192,255,195,241,925