Effect of Cognitively Stimulating Activities on Symptom Management of Delirium Superimposed on Dementia

A Randomized Controlled Trial

Ann Kolanowski, PhD; Donna Fick, PhD; Mark Litaker, PhD; Paula Mulhall, RN; Linda Clare, PhD; Nikki Hill, PhD; Jacqueline Mogle, PhD; Malaz Boustani, MD; David Gill, MD; Andrea Yevchak-Sillner, PhD


J Am Geriatr Soc. 2016;64(12):2424-2432. 

In This Article


This is the first clinical trial to test a nonpharmacological intervention recommended in current guidelines for management of delirium.[11] A daily 30-minute session of individualized cognitively stimulating activities, delivered with high fidelity, did not decrease delirium duration or severity. There was a small effect on executive function favoring intervention, and intervention participants spent fewer days in PAC than control participants.

Most participants experienced subsyndromal delirium during their PAC stay, but most days were delirium free, although delirium features persisted through discharge for the majority of participants. A recent prospective cohort study found that DSD is more protracted than previously thought.[13] The majority of participants in that study had no or partial delirium recovery 3 months after enrollment despite significant improvements in global cognition and function. These findings and the current results underscore the need for longer follow-up to capture possible delayed effects of intervention on resolution of DSD.

Few studies have investigated whether specific cognitive domains improve more than others when interventions for delirium are instituted. The current found a small effect on executive function that favored intervention. This was above gains that may be attributed to rehabilitation therapies, which can also improve this outcome and which the majority of both groups received.

Even the small improvement observed in executive function may be notable given the underlying cognitive impairments of the participants and literature that suggests an association between this higher-order function and delirium pathophysiology.[36,37] In clinical studies, frontoexecutive dysfunction independently predicted postoperative delirium even in the absence of cognitive impairment.[38] Individuals who recover from delirium have short-[5] and long-term[4] impairments in executive function. A previous study[18] reported greater benefits in executive function with cognition-focused and physical therapy than with physical therapy alone in individuals who survived the intensive care unit. Unlike the current study study, the combined intervention in that study continued for 12 weeks, the participants were younger, and none had dementia. Executive function may be an important cognitive domain that reflects central nervous system integrity after delirium. As such, it may be a sensitive indicator of the effectiveness of delirium interventions.

In unadjusted analyses, the intervention was found to have a small to moderate effect on constructional praxis. These visual-spatial skills are important for hazard perception, and deficits are found in individuals with frontal dysfunction,[39] but after adjusting for baseline differences, the group comparison was no longer statistically significant.

Although the intervention did not reduce delirium, it may have a role in future delirium prevention trials by strengthening important cognitive domains such as executive function and providing cognitive reserve as a protection against delirium risk factors. Affecting executive function also has implications for practice because these higher-order functions are important for independence.[40] In the current study, groups were distributed equally across sites, and length-of-stay analyses controlled for the effect of facility on outcomes. Length of stay was significantly shorter in intervention than control participants. Although there was no difference in discharge disposition, the ability to transition more quickly to a lower level of care can result in cost savings that may result from even small effects on cognitive function. Additionally, as discussed below, the outcome measures used may not have been sensitive enough to demonstrate important differences between the groups attributed to the intervention. Instrument selection is an important area for future research.

Attention, memory, and orientation may have been less amenable to intervention than executive function, but may also recover earlier, making improvement difficult to detect in PAC. A previous study[5] found that, 1 week after recovery from delirium, the performance of hospitalized older adults on attention, memory, and orientation tasks was similar to that of those who did not experience delirium, although executive function, praxis, and language remained significantly worse in the group with delirium than in the group without, and in unadjusted analyses in the current study, the intervention had a small effect on two of these domains.

Both groups received physical and occupational therapy, which may account for the lack of difference in physical function. Although the aggressive rehabilitation available in PAC facilitates recovery, the large percentage of participants who had delirium features on discharge and were admitted to long-term care (57.3% of those alive at 3 months) raised the question of whether the PAC environment influences the persistence of delirium and higher-order cognitive symptoms.[6]

There are several limitations of this study. It is likely that cognition in PAC is linked to preillness cognition. Because participants were enrolled at admission to PAC, it was not possible to test function before the illness that precipitated hospitalization. This limitation was addressed by using the Modified Blessed Dementia Rating Scale and CDR at baseline to estimate preexisting cognitive function. By design, only individuals with DSD were eligible for the study, so it is not known whether the intervention would be more effective in those without dementia. For individuals with dementia, the duration of the intervention and length of follow-up may need to be extended beyond a relatively short PAC stay. The instruments used to measure the outcomes were selected to reduce participant burden; a larger effect might be found using a more-in-depth neuropsychological test battery.

There are also several strengths of this study. A theory-based intervention was tested in a population that is not routinely included in studies that test interventions that optimize rehabilitation. Specific cognitive domains were also examined rather than one global outcome, adding to the literature on delirium resolution.

All interventions require staff time; using only components with known benefits will improve the quality and cost-effectiveness of care. Cognitive stimulation improved cognition and decreased length of stay, which has implications for quality of life and cost of care. Resolution of delirium may require more-intense nonpharmacological management when an individual has dementia.