To our knowledge, this is the first study examining the association of delirium and short-term outcomes of older adults admitted to SICUs. We found that nearly one third of older adults admitted to a SICU had a complication during their ICU stay. There was a strong association between SICU delirium and complication occurrence. Unexpectedly, most subjects in this sample had a complication before delirium was diagnosed. This suggests that ICU complications may serve as a risk factor for delirium. Further study of this area is warranted. This is particularly important as the Centers for Medicare and Medicaid Services consider expanding the list of hospital-acquired conditions for nonpayment. The proposed list for 2009 includes delirium and conditions (ie
, complications) such as extreme blood sugar derangement, iatrogenic pneumothorax, ventilator-associated pneumonia, and deep vein thrombosis/pulmonary embolism.24
We also found that SICU delirium was associated with a decrease in functional ability at hospital discharge and greater likelihood of being discharged to a place other than home. Even after controlling for a number of important variables including age, severity of illness, and mechanical ventilation use, an older adult with SICU delirium was seven times more likely to be discharged to a place other than home. While not statistically significant, delirium increased the odds of functional decline over threefold. It is likely, that if the sample had been larger, this finding would have reached significance. These results are consistent with prior studies of the effect of delirium on hospitalized (non-ICU) older adults' functional ability and discharge placement.25–28
Our findings have significant clinical and research implications. First, they support the mounting evidence that delirium is an important, independent prognosticator of poor outcomes in the critically ill older adult. We suggest both routine delirium assessment and the judicious use of sedatives and analgesics (as offered in the SCCM guidelines)29
be incorporated into the plan of care of all critically ill older adults. We also suggest that strategies intended to decrease the frequency and severity of delirium in hospitalized older adults (ie
, restraint reduction, early device removal, frequent mobilization, hearing and visual aids, and efforts to improve patient communication through assistive strategies) be implemented in the critical care setting. For example, a study30
of older patients with femoral neck fracture demonstrated that the application of comprehensive geriatric assessment, management, and rehabilitation decreased the number of days of postoperative delirium, decreased complications, and shortened length of hospitalization. Similar success has been demonstrated in the Hospitalized Elder Life Program.31
Despite the lack of randomized control trials in the ICU setting, there is clearly a need for evidence-based practices to be rigorously employed during the perioperative period. Practical interventions focused on optimal fluid and electrolyte management, nutrition, sleep, and mobility may help reduce delirium and its impact in the ICU.
Our investigation has several limitations. The inclusion of only older, English-speaking surgical patients recruited from a single institution limits generalizability of study findings. We are unable to establish a true “cause/effect” relationship between delirium and the selected outcomes; however, we do provide evidence of the negative effect of SICU delirium on important outcomes. The Katz ADL instrument was selected because of its wide use in the geriatric research literature; however, this instrument is limited by its relatively coarse measure of function. Using surrogates to obtain information regarding the subjects' preadmission health status, functional and cognitive ability, and living arrangements creates the possibility that the surrogates either overestimated or underestimated the older adults' actual status. As with many clinical research studies, the number of covariates measured and examined was limited. It is possible that other factors may impact older adults' short-term outcomes. For example, delirium is treated in diverse ways (ie, restraint/sedation/analgesic use), and these different methods of managing delirium have the potential to influence the outcome variables that were assessed. Finally, in our multivariable models SICU delirium was handled as a yes/no event. It is possible our results may have differed if delirium severity was measured or the time-varying aspect of this disorder was considered.
The strengths of this study include its prospective design, the frequency of delirium assessment, and the focus on older SICU patients, a population that had not been studied previously. The study is also strengthened by the use of a wide variety of valid and reliable tools to measure delirium, dementia, agitation and sedation, severity of illness, functional ability, and comorbidities.
In this study of older adults admitted to SICUs, complications were associated with increased incidence of delirium. Delirium significantly increased the likelihood that older adults would be discharged to a place providing higher levels of dependent care. The consequences of SICU delirium are costly from a resource, economic, and social perspective. Clinicians should routinely assess for delirium in the ICU and aggressively manage it. Further research is needed on how to better prepare elders for surgery, elucidate risk factors for delirium in the SICU, and prevent/treat delirium in the acute care setting.