Closing the quality gap: revisiting the state of the science (vol. 8: improving health care and palliative care for advanced and serious illness).

Objective: To systematically review the evidence on the effectiveness of health care and palliative care interventions to improve outcomes for patients with advanced and serious illness.

Data sources: We searched MEDLINE®, CINAHL, PsycINFO, Cochrane, and DARE from 2000 through 2011. We identified additional studies from reference lists of eligible articles and relevant reviews, as well as from technical experts.

Review methods: We developed questions in collaboration with technical experts. We excluded retrospective and uncontrolled studies. Two investigators independently screened search results and abstracted data from eligible studies. We adapted previous frameworks to categorize included studies (e.g., by improvement target, setting). Because many studies did not report effect sizes and almost all studies were small (lt 200 studies), in order to be able to quantitatively describe the literature, we calculated the percentage of studies with a significant improvement in outcomes with the intervention compared to control group for each category. We also checked that all other studies did not report significant results in the opposite direction and checked that there were not differences between larger and smaller studies.

Results: We included 90 studies described in 96 articles. Of the 23 studies targeting continuity, coordination, and transitions, 33 percent of studies that evaluated quality of life as an outcome, 67 percent that evaluated patient satisfaction, and 31 percent that evaluated health care utilization (admissions and length of stay) found a statistically significant improvement with the intervention. Of the 21 studies targeting pain, almost all focused on patient education and self-management; 48 percent of them found a statistically significant improvement with the intervention. Findings for larger (>100) and smaller (le 100) studies were similar. For distress, only 29 percent of the seven included studies found a statistically significant impact. Of the 20 studies in communication and decisionmaking, only 22 percent of studies addressing patient or family satisfaction found a statistically significant improvement for this outcome, compared to 73 percent for the outcome of health care utilization. We found only two studies within hospice programs, both of which found a statistically significant improvement in at least one outcome; nine studies were in nursing homes, 78 percent of which demonstrated a significant improvement with the intervention. In terms of types of quality improvement, for the target of continuity, studies including patient-centered quality improvement types, such as education and self-management, had the strongest evidence of effectiveness on patient- and family-centered domains such as satisfaction and quality of life. Studies of provider-focused interventions (e.g., education, reminders) were more likely to have an impact on health care utilization. Only one of five studies addressing multiple targets and focusing on facilitated relay of clinical data to providers demonstrated a statistically significant improvement in either quality of life or satisfaction. In terms of consultative and integrative interventions, for the target of communication and decisionmaking, three-quarters of consultative interventions showed a statistically significant improvement with the intervention, compared to half of integrative interventions. The literature was too heterogeneous and effect sizes were too infrequently reported for quantitative synthesis. There was moderate strength of evidence for the target of continuity, coordination, and transitions and the outcome of patient and caregiver satisfaction but low strength of evidence for other outcomes. For the target of pain, there was moderate strength of evidence for pain as an outcome. For the target of communication and decisionmaking, there was moderate strength of evidence for the outcome of health care utilization but low strength of evidence for other outcomes.

Conclusions: We found that evidence was strongest (moderate strength of evidence) for interventions for pain, and for the targets of communication and decisionmaking and continuity for selected outcomes. While a few high- and medium-quality, well-designed health care and palliative care interventions have been conducted to improve outcomes for patients with advanced and serious illness, this report highlights the continued presence of variable findings, quality deficiencies, vaguely defined interventions, and variable outcome measurement tools and reporting in much of this intervention literature. The evidence has a number of gaps, including few studies in the hospice setting or pediatrics. Future research needs include techniques for improving recruitment and retention to assure adequate sample size, better development and description of interventions, and further development and standardization of outcome measures and tools.