Funder requested change in contract which stopped enrollment at N = 104

Twenty percent of the US population will be age 65 or older by 2050, a surge from 8 percent in 1950 and 12 percent in 2000. The proportion of low-income older adults is also growing. Approximately 70 percent of people 65 and older are expected to need some level of long-term care, which may burden the nation's health and caregiver systems. While there are many models to care for chronically ill older adults, there is less agreement on how to support healthier, low-income older adults to stay independent in their communities.

This study will compare the effectiveness of an in-home preventive healthcare program delivered by nurses to on-site health and wellness classes for older adults living in low-income independent housing. We also want to understand adherence and preferences of older adults for these two options.

The three-year study will take place in up to 18 low-income independent older adult apartment buildings in Los Angeles. Half of buildings will be randomized to offer the health and wellness classes, and the other half will offer the in-home preventive healthcare program. All study activities will be provided on-site at the building locations, and individuals living in the buildings will be invited to participate. A total of 480 participants will be recruited to participate, 240 in each group. Residents from participating buildings will meet with the research team to provide feedback throughout the study.

Other stakeholders, including doctors, housing services, social agencies, hospital leaders, professional societies, advocacy groups, and city policymakers will also meet with the team as an advisory group to share input and concerns. This project seeks to compare how each option maintains or improves health and functional independence in low-income older adults, with the goal of limiting dependency, moves to nursing homes, and the use of costly health services, while improving health behaviors and promoting the use of preventive health and appropriate community services.

EPIC is designed as a cluster randomized trial of low-income independent older adult housing locations in Los Angeles, CA. A total of 18 apartment buildings managed by five housing providers are currently available to be randomized. Buildings will be randomly selected two at a time, independent of housing provider. Prior to any research activity at each building, a social event will be scheduled to introduce the core research team to the building tenants and begin to further identify tenant opinion leaders. Subsequently, a mini health fair will be conducted in each building as part of research participant recruitment efforts. Baseline enrollment and assessment will be conducted prior to the building pair being randomly assigned, one to each study arm, to avoid bias in study enrollment. This design will allow us to study the effect of each strategy with an 18-month intervention, over a three-year study duration. The inclusion of apartment buildings - being the congregate focus of the interventions - are best suited to a cluster randomized controlled trial.

Core research team members, working closely with building staff and resident advisors, will tailor recruitment strategies - including pre-implementation social events, on-site receptions and health fairs, invitation letters and opportunities for residents to speak with the researchers. All tenants in each housing location will be invited to participate in our study of preventive health interventions and will agree at the minimum to be interviewed at baseline, the end of nine months, and the end of 18 months. The study will enroll as many as agree to participate (a minimum of 640 with up to 25% attrition to render a complete sample of 480 participants).

In keeping with the pragmatic nature of the community-based research and the circumstances of emerging from COVID pandemic, we will plan to complete a focused review and evaluation of the research implementation processes after the first two building pairs have been enrolled and two months of interventions have been provided to the four buildings.

Treatment Arms as Complex Interventions. In the past two decades, researchers addressing implementation of health promotion and prevention programs are acknowledging that these interventions are complex. Programs typically have multiple interacting components, address multiple risk factors, are implemented in multiple sites, and allow a certain amount of flexibility to adapt the intervention, tailoring to sites or groups. Variation in reported effectiveness is another feature of complex interventions, given that they reflect the different realities of varying health needs and responses within disparate populations and contexts. To maintain integrity complex interventions use guided processes, also called forms. Research studying complex interventions is meant to provide not only results about adherence and effectiveness, but commentary on what works best for whom, and how it works.

Given their interacting components and the multiple risk domains targeted by these interventions; both approaches being studied in this project, comprehensive health assessment and health and wellness classes, can be considered to be complex interventions. They are both premised upon the common function of disability prevention and address intra-individual factors as well as extra-individual factors that modify and slow or prevent the disablement process as described in the Verbrugge and Jette model. As complex interventions they are adapted to meet the individual needs of the participants and the context in which they are implemented.

Both interventions use forms (guided processes) that are standardized, but the ultimate intervention is a negotiation between participants and practitioners (nurses or educators). The guide for the APRN intervention is a standardized multi-dimensional health risk appraisal (based on CGA), risk identification, standardized recommendations negotiated with the participant and followed for reinforcement. The guide in the Classes intervention is a standardized curriculum with flexibility to modify according to participants' needs and preferences. The two intervention arms have different profiles of what risk factors for which they are stronger or weaker, and, critically, different patterns of adherence and attendance, i.e, the "self-administered" dose of any preventive actions being taken by the participants. The use of a Comparative Effective Research design enables a first-ever direct comparison of the outcomes associated with each complex intervention, and a wealth of data to explore how the interventions are received and evaluated by their users.

Inclusion Criteria:

• Age and income qualified tenants in 18 identified older adult low income independent living apartment buildings.

Exclusion Criteria:

• Non-tenants in 18 identified buildings
• Assessed to be cognitively impaired and unable to consent.