This is a 52-month study (8 months preparation; 36 months to conduct the trial; 8 months data analyses and manuscript preparation) of a cluster randomized controlled trial (RCT) of an intervention to improve infection management for suspected UTIs and LRIs among residents with advanced dementia (N=480; N=240/arm) living in NHs (N=24; N=12/arm). The NH is the unit of randomization as the intervention must be delivered at the facility level to avoid contamination and because this is how it would be employed in the real-world. Analyses will be at the patient level.

The final stage of dementia is characterized by recurrent suspected infections. Research has shown these episodes are widely mismanaged, leading to adverse patient and public health outcomes. Antimicrobials are extensively prescribed in advanced dementia, most often in the absence of clinical evidence to support a bacterial infection. Antimicrobial exposure is the main risk factor for multidrug-resistant organisms (MDROs). Nursing home (NH) residents with advanced dementia are three times more likely to be colonized with MDROs compared to other residents. Moreover, as these patients are in the terminal phase of dementia, evidence suggests they may not clinically benefit from antimicrobials. Comfort is the stated goal of care for 90% of advanced dementia patients, and the risks and burdens associated with work-up and treatment of suspected infections generally do not promote that goal, particularly when hospitalization is involved. Taken together, there is a clear need to improve infection management in advanced dementia both to provide better end-of-life care to these patients and reduce the public health threat of MDROs.

This is a 52-month study (8 months preparation; 36 months to conduct the trial; 8 months data analyses and manuscript preparation) of a cluster randomized controlled trial (RCT) of an intervention to improve infection management for suspected UTIs and LRIs among residents with advanced dementia (N=410; N=205/arm) living in NHs (N=28; N=14/arm). The NH is the unit of randomization as the intervention must be delivered at the facility level to avoid contamination and because this is how it would be employed in the real-world. Analyses will be at the patient level.

Inclusion Criteria:

Facility inclusion criteria

1. More than 60 beds
2. Within 60 miles of Boston

Resident inclusion eligibility criteria

• Age > or = to 60 years
• A diagnosis of dementia (any type)
• Global Deterioration Scale (GDS) score of 7
• NH length of stay >90 days
• An individual who can communicate in English has been formally or informally designated as a health care proxy
• Not comatose GDS stage 7 features include: profound memory deficits (cannot recognize family), total functional dependence, speech <= 5 words, incontinence, and non-ambulatory. GDS 7 was chosen to define advanced dementia as it has been successfully operationalized and validated in or prior studies, and experts agreed to use this definition in research studies. A 90 day minimum length of stay was chosen to exclude short-stay patients.

Provider inclusion criteria

• Direct care provider of advanced dementia residents (a nurse, nurse practitioner, physician or physician assistant identified by a senior administrator as an individual who cares for residents with advanced dementia)
• Can communicate in English (because intervention materials are all in English),
• Over 21 years of age.

Exclusion Criteria:

Residents with cognitive impairment due to causes other than dementia (e.g., head trauma and in short-term, sub-acute SNFs) will be excluded.

Residents' whose proxies cannot communicate in English will be excluded from the study, because the information directed to proxies are only in English.

Providers that do not provide direct care to residents with advanced dementia or who do not speak English.

-

Each residents was followed for up to one year

There were no pre-specified adverse events in this minimal risk study

The research assistants collecting outcome data through chart reviews and follow-up proxy interviews will be blinded to nursing home randomization. The Principal Investigator, data programmers, and statistician will be blinded to treatment assignment