Hospital discharge is an important transition point for patients who have been in ICUs. The physical, functional, and psychosocial changes associated with recovery from critical illness present challenges for patients, families, and healthcare providers involved in the patients' care. While the complex discharging needs of ICU survivors have been increasingly recognized1-3, little rigorous research or evaluation has been conducted in this area.
The objectives of this study are to 1) assess the feasibility of constructing outcome measures of care transitions for patients discharged from intensive/critical units (ICUs) using existing VA national databases; and 2) examine the associations of patient and organizational factors with post-ICU discharge outcomes, including mortality, post discharge clinic visit, hospital readmission, and medication prescribing patterns.
For Objective 1, we searched PubMed for literature on outcome measures that have been used to evaluate care transitions for ICU survivors. Two investigators independently reviewed identified literature. Available data elements in VA data systems were determined from definitions provided in technical and user guides of VA data sources. Feasibility will be evaluated by comparing operational definitions of outcome measures from literature with descriptions of data elements from VA data sources. For Objective 2, we used an observational cross-sectional design to assess contributing factors of post-ICU health outcomes. We included the first admission of patients who had an ICU stay and were discharged between 4/06 and 3/07 from 20 randomly selected VAMCs. Outcome and independent variables were drawn from VA Patient Treatment File and DSS pharmacy national data extracts. Patient outcome analyses were conducted at the patient level. Data were analyzed using descriptive statistics, multilevel mixed effect models for continuous outcomes (e.g., number of post discharge clinic visit and hospital readmissions) and generalized linear latent and mixed models for binary outcomes (e.g., mortality after discharge).
The study included 11303 first admissions of patients during the study period. The majority of patients were white race (67%), male (97%), and not married (54%), with an average age of 66 years (SD=12). The average ICU and hospital length of stay were 4 days (SD=5) and 10 days (SD=24), respectively. The top 5 discharge diagnosis groups were coronary atherosclerosis, acute myocardial infarction, cardiac dysrhythmias, congestive heart failure, and occlusion/stenosis of precerebral arteries, accounted for 27% of all discharges. 95% of patients had at least one outpatient visit within one year following the index discharge, with 75% of the first outpatient visit occurred within 14 days of the index discharge. Among patients discharged with the top 10 diagnosis categories, 17% of the first outpatient visits were for the purposes of administrative/social admission, aftercare, and routine general medical examine. 43% of patients were readmitted to an acute care hospital within one year from the index discharge. The 14- and 30-day all cause readmission rates to an acute care hospital were 11% and 17%, respectively. Approximated 17% of the first readmission was for the same diagnostic category as the index discharge.
This analysis will assist in determining information resources that are available or needed to provide better care to patients admitted with critical illness. Findings from this pilot study will provide initial evidence to standardize tools for use in ICUs, and for measuring recovery after an ICU stay.
- Li YF, Helfrich CD. A Multigroup Invariance Analysis of the Organizational Culture Scale. Poster session presented at: VA HSR&D National Meeting; 2009 Feb 13; Washington, DC.