Lack of Useful Predictors of Dignity-Related Distress Among the Critically Ill as Assessed With the Patient Dignity Inventory

Abstract

BACKGROUND: Many intensive care unit patients are awake (ie, alert and engaging in conversation), actively experiencing many facets of their critical care. The Patient Dignity Inventory can be used to elicit sources of distress in these patients. We examined the administrative question as to which awake intensive care unit patients should be evaluated and potentially treated (eg, through palliative care consultation) for distress. Should the decision to screen for distress be based on patient demographics or treatment conditions? METHODS: This was a retrospective cohort study of 155 adult patients from 5 intensive care units of one hospital from 2019 to 2020. Each patient had ≥48 hours without delirium, dementia, or sedation. The Patient Dignity Inventory has 25 items to which patients responded on a 1 (not a problem) to 5 (an overwhelming problem) scale. Multiple complete, stepwise forward, and stepwise backward logistic regression models were created among patient and treatment variables for predicting thresholds of the mean among the 25 items. RESULTS: There were 50% (78/155; 95% confidence interval [CI], 42–58) of patients with significant dignity-related distress (mean score ≥1.60). There were 34% (52/155; CI, 26–42) of patients with severe dignity-related distress (mean score ≥1.92; previously associated with often feeling like wanting to die). Models including combinations of vasopressor medication (protective of distress), tracheostomy (greater risk of distress), and female gender (greater risk of distress) had some predictive value. However, all combinations of potential predictors had misclassification rates significantly >20%. CONCLUSIONS: Identification of subsets of patients with little potential benefit to screening for dignity-related distress would have a reduced workload of palliative care team members (eg, nurses or social workers). Our results show that this is impractical. Given that approximately one-third of critical care patients who are alert and without delirium demonstrate severe dignity-related distress, all such patients with prolonged intensive care unit length of stay should probably be evaluated for distress.