Out-of-hours discharge from intensive care, in-hospital mortality and intensive care readmission rates: a systematic review protocol

Most patients are discharged from an intensive care unit with an expectation that they will survive their hospital stay, yet these patients have high subsequent in-hospital mortality. Patients are frequently discharged from an intensive care unit to a lower level of hospital care in the evenings and at night (out-of-hours). By affecting the care that patients receive, out-of-hours discharge may alter post-intensive care in-hospital mortality rates.

Two searches will be conducted—the first a general search for all factors associated with post-intensive care in-hospital mortality and a second focused specifically on out-of-hours discharges. Searches will be performed in multiple databases, including Medline, Embase, Web of Knowledge, Cumulative Index of Nursing and Allied Health Literature (CINAHL) and the Cochrane Library. OpenGrey will also be searched, to ensure any unpublished ‘grey’ data are accessed. Language and date restrictions will not be applied. Assessment for inclusion and data extraction will be undertaken by two independent reviewers. Methodological quality will be assessed using the ACROBAT-NRSI tool. The primary outcome measure will be post-intensive care in-hospital mortality. To provide a clearer picture of this problem, studies reporting readmission to the intensive care unit (ICU) will also be included, even in the absence of report of in-hospital mortality.

The primary outcome data will be synthesised and summarised using a random-effects meta-analysis. Where possible, subgroup meta-analyses will assess associated factors such as discharge destination, palliative care discharges and severity of illness scores.

To the best of our knowledge, a systematic review of the association of out-of-hours discharge with in-hospital mortality has never been undertaken. Synthesis of the available information is important because out-of-hours discharge remains common and, if associated with post-intensive care unit mortality, is highly amenable to system change.

PROSPERO CRD42014010321