Assessment of Intensive Care Unit Outreach Transition Programs effect on readmissions and mortality among ICU survivors discharged to ward in Calgary – a Time Series Study.

Introduction: Potentially preventable errors and adverse events can occur as a consequence of patient transfer of care between health care providers. Intensive Care Unit transition programs (ICUTP) are a type of Rapid Response System Model, which presents a very simple idea: if a patient shows signs of imminent clinical deterioration, a team of providers is called up to the bedside to immediately assess and treat the patient in order to prevent intensive care unit transfer, cardiac arrest, or death. ICUTP incorporate ICU transition assignments, having a role in facilitating discharge and providing a smooth transition for complex convalescent patients to a general hospital ward. However, when assessing the ICU transition assignments for these teams, there is a substantial lack of available data, essential for assessing their effectiveness. This led us to inquire into the real role of a program in which significant investments have been placed. Objectives: Primary: To assess if ICU transition programs decrease the risk of ICU readmission when compared to standard care among patients who survive their initial admission to an adult ICU. Secondary: To assess if ICU transition programs decrease the risk of in-hospital mortality when compared to standard care among patients who survive their initial admission to an adult ICU. Methods: We performed an interrupted time series (ITS) study, a variation of time-series studies classified as quasi-experiments, involving all adults older than 18 years old, who survived their first ICU admission and were discharged to ward between 2002 and 2010 in Calgary, Alberta. The outcomes (ICU readmission and hospital mortality) were measured at every 3 months, before and after the implementation of our ICUTP - the ICU outreach team (ICUOT). Multivariable segmented logistic regression was used to adjust the estimates of the odds ratio (OR) of each outcome measure before and after the intervention. Data were reported as odds ratios (OR) and proportions with 95% confidence intervals (CI) and were evaluated for multicollinearity and autocorrelation. Results: At the start of the study 6.0% (95% confidence interval [CI] 4.9% to 7.0%) of the patients of our study population were readmitted to ICU. During the pre-intervention period, we could see a non-significant decrease of 0.02%(-0.02%, 95% CI -0.10% to +0.07%) in the proportion of patients readmitted to ICU per quarter of study. After implementation of the ICUOT, there was a 2.0% significant increase in the proportion of patients readmitted to ICU (+2.0%, 95% CI +0.5% to +3.2%). Subsequently, we saw a non-significant decrease in that proportion (-0.04% per quarter; 95% CI, -0.2% to +0.1%). At the end of the study, the proportion of patients readmitted in the ICU was 6.0% (95% CI, 4.8% to 7.0%). Regarding hospital mortality, 7.0% (95% confidence interval [CI] 6.0% to 9.0%) of the patients of our study population died in the hospital at the start of the study period. In the pre-intervention period, there was a non-significant decrease of 0.01% (-0.01% 95% CI, -0.09% to +0.07%) in the proportion of patients who died in the hospital per quarter. After implementation of the ICUOT, there was an immediate non-significant increase of 1.0% in the proportion of patients dying in the hospital, (+1.0%, 95% CI -0.3% to +2.4%). Subsequently, there was a significant small decrease in that proportion (-0.2% per quarter; 95% CI, -0.3% to -0.05%). At the end of the study, the proportion of in-hospital deaths was 4.0% (95% CI, 3.0% to 5.0%). Conclusion: This work, based on a robust methodology that uses an interrupted time series with segmented logistic regression, showed that ICU readmission rates remain the same based on the estimated changes in intercept and slope when comparing pre and post-intervention periods. There is insufficient evidence of a statistically significant effect of the ICUOT on ICU readmissions. However, it is possible that closer monitoring and faster actions on those ICU survivors who required or did not required ICU readmission have been lead to a small but significant improvement in mortality.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.