Do community specialist palliative care services that provide home nursing increase rates of home death for people with life-limiting illnesses? A systematic review and meta-analysis of comparative studies

Tim Luckett
Patricia M. Davidson
Lawrence Lam, University of Notre Dame Australia
Jane L. Phillips, University of Notre Dame Australia
David C. Currow
Meera Agar


Context: Systematic reviews and meta-analyses suggest that community specialist palliative care services (SPCSs) can avoid hospitalizations and enable home deaths. But more information is needed regarding the relative efficacies of different models. Family caregivers highlight home nursing as the most important service, but it is also likely the most costly.

Objectives: To establish whether community SPCSs offering home nursing increase rates of home death compared with other models.

Methods: We searched MEDLINE, AMED, EMBASE, CINAHL, the Cochrane Database of Systematic Reviews, and CENTRAL on March 2 and 3, 2011. To be eligible, articles had to be published in English-language peer-reviewed journals and report original research comparing the effect on home deaths of SPCSs providing home nursing vs. any alternative. Study quality was independently rated using Cochrane grades. Maximum likelihood estimation of heterogeneity was used to establish the method for meta-analysis (fixed or random effects). Potential biases were assessed.

Results: Of 1492 articles screened, 10 articles were found eligible, reporting nine studies that yielded data for 10 comparisons. Study quality was high in two cases, moderate in three and low in four. Meta-analysis indicated a significant effect for SPCSs with home nursing (odds ratio 4.45, 95% CI 3.24–6.11; P < 0.001). However, the high-quality studies found no effect (odds ratio 1.40, 95% CI 0.97–2.02; P = 0.071). Bias was minimal.

Conclusion: A meta-analysis found evidence to be inconclusive that community SPCSs that offer home nursing increase home deaths without compromising symptoms or increasing costs. But a compelling trend warrants further confirmatory studies. Future trials should compare the relative efficacy of different models and intensities of SPCSs.


Link to Publisher Version (DOI)