One in three Americans dies in a nursing home (NH) or in a hospital, shortly following transfer from a long-term care facility. The proportion of deaths occurring in NHs is projected to increase to 40% by 2020. Excellence in palliative and end-of-life (EOL) care must become a priority for these long-term care institutions. However, findings from NHs point to high incidence of pain and poor management of other symptoms and excessive reliance on hospitalizations, indicating inadequate EOL care quality. Expert opinion and research have suggested that poor EOL quality in NHs may be due to lack of palliative care training among staff and absence of EOL care protocols or guidelines, but research demonstrating that attention to these factors improves outcomes is absent. While dedicated care teams have been shown to improve outcomes for NH residents in need of specialized care, the impact of palliative care teams in improving resident outcomes has remained largely unstudied and untested. This will be the first randomized controlled trial to evaluate the impact of palliative care teams (PCTeam) on resident and staff outcomes, and care processes, in NHs.

Our objective is to demonstrate, using a randomized controlled trial design and a difference in difference analytic approach, that nursing home-based palliative care practice guidelines implemented through PCTeams will improve quality of care processes and outcomes for residents at the end of life.

We will adapt existing palliative care guidelines for EOL care, endorsed by the National Quality Forum (NQF), to the NH environment, deploy the adapted practice guidelines through a PCTeam model, and evaluate the effectiveness of this intervention on resident EOL outcomes and staff care processes and outcomes. The specific aims (SA) will address the following questions:

SA 1: Is PCTeam intervention effective in improving NH residents' EOL outcomes?

SA 2: Is PCTeam intervention effective in improving NH staff EOL care processes and outcomes?

In the context of these specific aims we will test the following hypotheses:

H1: Residents in NHs in the intervention arm, compared to the control, will achieve better EOL risk-adjusted outcomes and care processes with regard to:

• Pain
• Dyspnea
• Depression
• In-hospital deaths
• Hospitalizations
• Advance directives

H2: Direct care staff in NHs in the intervention arm, compared to the control, will achieve better EOL processes and outcomes measured by:

• Assessment of EOL symptoms
• Delivery of EOL care
• Communication/coordination among providers
• Communication with residents/families
• Teamwork effectiveness
• Staff satisfaction

H3: Family caregivers of decedent residents in the intervention NHs, compared to the control, will report receiving more patient and family centered care as measured by higher levels of satisfaction with:

• Shared decision making between providers, the patient and the family

• Care that is respectful of the patient wishes and dignity

• Attention to the emotional and spiritual needs of the family.

  31 NHs in upstate New York have been recruited for the study (letters of support). Stakeholders include residents, family members, staff, policy makers, and others. The intervention will deploy theTeamSTEPPS, a team development model created by the Department of Defense and the Agency for Healthcare Research and Quality.

Inclusion Criteria:

• long-stay nursing home residents (>90 days)

Exclusion Criteria:

• rehabilitative, post-acute residents (<90 days)

All decedents with data who resided in assigned facilities.

1 year

We only collected and used data on decedents in our enrolled nursing homes, therefore ALL individual level outcomes measures are based on residents who died during the study period. However, the intervention was not at the individual level, but rather a facility-level. The study was not directly associated with, or a cause of, any of these deaths. Rather, we aimed to teach nursing home staff in the intervention facilities how to improve the quality of care for residents who were dying.

1. We were unable to distinguish the relative importance of the individual intervention components.
2. Our findings were confined to regional facilities, and may not be generalized elsewhere.