Discharge Destination As a Marker of Mobility Impairment in Survivors of Acute Respiratory Distress Syndrome

Objectives: Discharge destination is a commonly used surrogate for long-term recovery in rehabilitation studies. We determined the accuracy of discharge destination as a surrogate marker for 6-month mobility impairment in acute respiratory distress syndrome survivors. Design/Setting: Secondary analysis of the Economic Analysis of Pulmonary Artery Catheters study, a long-term observational sub-study of the National Institutes of Health/National Heart, Lung, and Blood Institute Acute Respiratory Distress Syndrome Network Fluid and Catheter Treatment Trial. Patients: Patients underwent functional assessment using the Health Utilities Index-2 mobility domains at 6 months. A score greater than or equal to 3 (i.e., need for assistive device) defined mobility impairment. Discharge to any institutional care constituted a care facility discharge. We used logistic regression to explore the association between discharge destination and mobility impairment. We generated test characteristics and receiver operating characteristics to assess the accuracy of discharge destination as a surrogate for mobility impairment. Interventions: None. Measurements and Main Results: A total of 591 patients were enrolled in Economic Analysis of Pulmonary Artery Catheters in whom 328 had functional measurements at 6 months with 116 (35%) of these patients discharged to a care facility. Care facility patients were older (mean age 53 vs 44 yr; p < 0.001) and had longer hospital durations (length of stay 29 vs 17 d; p < 0.001). Care facility discharge was associated with greater 6-month mobility impairment. Sensitivity and specificity of discharge to a care facility for mobility impairment were 40.5% (95% CI, 32.0–49.6%) and 79.3% (95% CI, 73.3–84.2%) at 6 months. Discharge destination alone was a poor discriminator of long-term mobility impairment (receiver operating characteristic area under the curve: 0.61 at 6 mo). Conclusions: Discharge to a care facility was strongly associated with mobility impairment 6 months after acute respiratory distress syndrome but discharge destination alone performed poorly as a surrogate for mobility impairment.