To evaluate availability, effectiveness, and implementation of interventions for integrating palliative care into ambulatory care for U.S.-based adults with serious life-threatening chronic illness or conditions other than cancer and their caregivers We evaluated interventions addressing identification of patients, patient and caregiver education, shared decision-making tools, clinician education, and models of care.
We searched key U.S. national websites (March 2020) and PubMed, CINAHL, and the Cochrane Central Register of Controlled Trials (through May 2020). We also engaged Key Informants.
We completed a mixed-methods review; we sought, synthesized, and integrated Web resources; quantitative, qualitative and mixed-methods studies; and input from patient/caregiver and clinician/stakeholder Key Informants. Two reviewers screened websites and search results, abstracted data, assessed risk of bias or study quality, and graded strength of evidence (SOE) for key outcomes: health-related quality of life, patient overall symptom burden, patient depressive symptom scores, patient and caregiver satisfaction, and advance directive documentation. We performed meta-analyses when appropriate.
We included 46 Web resources, 20 quantitative effectiveness studies, and 16 qualitative implementation studies across primary care and specialty populations. Various prediction models, tools, and triggers to identify patients are available, but none were evaluated for effectiveness or implementation. Numerous patient and caregiver education tools are available, but none were evaluated for effectiveness or implementation. All of the shared decision-making tools addressed advance care planning; these tools may increase patient satisfaction and advance directive documentation compared with usual care (SOE: low). Patients and caregivers prefer advance care planning discussions grounded in patient and caregiver experiences with individualized timing. Although numerous education and training resources for nonpalliative care clinicians are available, we were unable to draw conclusions about implementation, and none have been evaluated for effectiveness. The models evaluated for integrating palliative care were not more effective than usual care for improving health-related quality of life or patient depressive symptom scores (SOE: moderate) and may have little to no effect on increasing patient satisfaction or decreasing overall symptom burden (SOE: low), but models for integrating palliative care were effective for increasing advance directive documentation (SOE: moderate). Multimodal interventions may have little to no effect on increasing advance directive documentation (SOE: low) and other graded outcomes were not assessed. For utilization, models for integrating palliative care were not found to be more effective than usual care for decreasing hospitalizations; we were unable to draw conclusions about most other aspects of utilization or cost and resource use. We were unable to draw conclusions about caregiver satisfaction or specific characteristics of models for integrating palliative care. Patient preferences for appropriate timing of palliative care varied; costs, additional visits, and travel were seen as barriers to implementation.
For integrating palliative care into ambulatory care for serious illness and conditions other than cancer, advance care planning shared decision-making tools and palliative care models were the most widely evaluated interventions and may be effective for improving only a few outcomes. More research is needed, particularly on identification of patients for these interventions; education for patients, caregivers, and clinicians; shared decision-making tools beyond advance care planning and advance directive completion; and specific components, characteristics, and implementation factors in models for integrating palliative care into ambulatory care.
