Effect size estimates of risk factors for post-intensive care syndrome: A systematic review and meta-analysis.

Abstract

Objective: To provide updated evidence on the risk factors and accurately quantify the effect size of the risk factors associated with post-intensive care syndrome (PICS), encompassing cognitive, mental, physical and socio-economic domains.

Research methodology: We conducted a systematic review of literature from January 2010 to October 2023. The meta-analysis was conducted to calculate an effect size for every risk factor, and odds ratio and 95% confidence intervals were used as summary statistics.

Results: Of 67,468 retrieved studies, 160 were included in qualitative synthesis, 102 were included in quantitative synthesis. A total of 60 factors were identified, categorized into 17 person-related, 23 disease-related and 20 ICU-related categories. The strongest correlations with cognitive health were observed for previous cognitive problems, sedatives and delirium. Factors most strongly correlated with mental health included previous mental problems, delirium, lack of social support, illicit drug and bad experience in ICU. The strongest correlations with physical health involved previous mental problem, delirium, organ dysfunction of neurologic and respiratory support. For socio-economic domains, older age and female were identified as significant risk factors.

Conclusions: This systematic review and meta-analysis identified and quantified the person, disease, and ICU-related risk factors associated with PICS. These findings may enable clinicians to better recognize the patient at high risk for PICS at an early stage during their stay in ICU.

Implications for clinical practice: A thorough investigation of risk factors across the four domains of PICS is necessary to gain a holistic understanding. The identification and integration of risk factors associated with PICS empower critical care multidisciplinary teams to optimize management strategies, thereby assisting ICU survivors a better recovery. Since multiple risk factors may be simultaneously associated with the four domains of post-intensive care syndrome, it is imperative to develop a comprehensive prediction algorithm.