Objective To identify existing prognostic delirium prediction models and evaluate their validity and statistical methodology in the older adult (≥60 years) acute hospital population. Design Systematic review. Data sources and methods PubMed, CINAHL, PsychINFO, SocINFO, Cochrane, Web of Science and Embase were searched from 1 January 1990 to 31 December 2016. The Preferred Reporting Items for Systematic Reviews and Meta-Analyses and CHARMS Statement guided protocol development. Inclusion criteria: age >60 years, inpatient, developed/validated a prognostic delirium prediction model. Exclusion criteria: alcohol-related delirium, sample size ≤50. The primary performance measures were calibration and discrimination statistics. Two authors independently conducted search and extracted data. The synthesis of data was done by the first author. Disagreement was resolved by the mentoring author. Results The initial search resulted in 7,502 studies. Following full-text review of 192 studies, 33 were excluded based on age criteria (<60 years) and 27 met the defined criteria. Twenty-three delirium prediction models were identified, 14 were externally validated and 3 were internally validated. The following populations were represented: 11 medical, 3 medical/surgical and 13 surgical. The assessment of delirium was often non-systematic, resulting in varied incidence. Fourteen models were externally validated with an area under the receiver operating curve range from 0.52 to 0.94. Limitations in design, data collection methods and model metric reporting statistics were identified. Conclusions Delirium prediction models for older adults show variable and typically inadequate predictive capabilities. Our review highlights the need for development of robust models to predict delirium in older inpatients. We provide recommendations for the development of such models.

While delirium is associated with cognitive decline and dementia, there is limited evidence to support causality for this relationship. Clarification of how delirium may cause cognitive decline, perhaps through evidence of contemporaneous neuronal injury, would enhance plausibility for a causal relationship. Dose-dependence of neuronal injury with delirium severity would further enhance the biological plausibility for this relationship. We tested whether delirium is associated with neuronal injury in 114 surgical patients recruited to a prospective biomarker cohort study. Patients underwent perioperative testing for changes in neurofilament light, a neuronal injury biomarker, as well as a panel of 10 cytokines, with contemporaneous assessment of delirium severity and incidence. A subset of patients underwent preoperative MRI. Initially we confirmed prior reports that neurofilament light levels correlated with markers of neurodegeneration [hippocampal volume (ΔR² = 0.129, P = 0.015)] and white matter changes including fractional anisotropy of white matter (ΔR² = 0.417, P < 0.001) with similar effects on mean, axial and radial diffusivity) in our cohort and that surgery was associated with increasing neurofilament light from preoperative levels [mean difference (95% confidence interval, CI) = 0.240 (0.178, 0.301) log₁₀ (pg/ml), P < 0.001], suggesting putative neuronal injury. Next, we tested the relationship with delirium. Neurofilament light rose more sharply in participants with delirium compared to non-sufferers [mean difference (95% CI) = 0.251 (0.136, 0.367) log₁₀ (pg/ml), P < 0.001]. This relationship showed dose-dependence, such that neurofilament light rose proportionately to delirium severity (ΔR² = 0.199, P < 0.001). Given that inflammation is considered an important driver of postoperative delirium, next we tested whether neurofilament light, as a potential marker of neurotoxicity, may contribute to the pathogenesis of delirium independent of inflammation. From a panel of 10 cytokines, the pro-inflammatory cytokine IL-8 exhibited a strong correlation with delirium severity (ΔR² = 0.208, P < 0.001). Therefore, we tested whether the change in neurofilament light contributed to delirium severity independent of IL-8. Neurofilament light was independently associated with delirium severity after adjusting for the change in inflammation (ΔR² = 0.040, P = 0.038). These data suggest delirium is associated with exaggerated increases in neurofilament light and that this putative neurotoxicity may contribute to the pathogenesis of delirium itself, independent of changes in inflammation.

Management of delirium in intensive care units is challenging because effective therapies are lacking. Music is a promising nonpharmacological intervention. To determine the feasibility and acceptability of personalized music (PM), slow-tempo music (STM), and attention control (AC) in patients receiving mechanical ventilation in an intensive care unit, and to estimate the effect of music on delirium. A randomized controlled trial was performed in an academic medical-surgical intensive care unit. After particular inclusion and exclusion criteria were applied, patients were randomized to groups listening to PM, relaxing STM, or an audiobook (AC group). Sessions lasted 1 hour and were given twice daily for up to 7 days. Patients wore noise-canceling headphones and used mp3 players to listen to their music/audiobook. Delirium and delirium severity were assessed twice daily by using the Confusion Assessment Method for the Intensive Care Unit (CAM-ICU) and the CAM-ICU-7, respectively. Of the 1589 patients screened, 117 (7.4%) were eligible. Of those, 52 (44.4%) were randomized, with a recruitment rate of 5 patients per month. Adherence was higher in the groups listening to music (80% in the PM and STM groups vs 30% in the AC group; P = .01), and 80% of patients surveyed rated the music as enjoyable. The median number (interquartile range) of delirium/coma-free days by day 7 was 2 (1-6) for PM, 3 (1-6) for STM, and 2 (0-3) for AC (P = .32). Median delirium severity was 5.5 (1-7) for PM, 3.5 (0-7) for STM, and 4 (1-6.5) for AC (P = .78). Music delivery is acceptable to patients and is feasible in intensive care units. Further research testing use of this promising intervention to reduce delirium is warranted.

Delirium is a sudden disturbance in attention and orientation to the environment that develops over a short period of time and tends to fluctuate in severity during the course of the day.¹ The acute confusional state of delirium occurs in 50–80% of critically ill patients and postoperatively (from the day after surgery onwards) in up to 54% of elective major non-cardiac surgical patients.¹ It incurs a huge societal burden, because of, in part, a result of its association with increased morbidity and mortality; each additional day of delirium has been independently associated with a 10% increased risk of death.² Increased morbidity contributes to prolonged hospital length of stay and significant financial implications: delirium is estimated to total $4–16 billion annually.³ Its...