In the current study, more than two-thirds (72.2%) of the community-dwelling older adults with dementia were exposed to at least one sedative or anticholinergic. A significant majority of these (38.1%) had high exposure to these medications. In addition, the prevalence of delirium (up to 33.2%) showed a significant upward trend with increased exposure. Moreover, the most common delirium subtype was hyperactive delirium in both the low- and high-exposure groups. One of the most striking findings of the current analysis was the significant association of high DBI levels with DSD, even after adjusting for potential covariates. To our knowledge, this is the first study to show that high exposure to these drugs increases the risk of delirium four-fold in this population.
In addition, a previous study showed that people with dementia have a higher anticholinergic burden than those without dementia [35]. Moreover, high anticholinergic burden is an independent risk factor for delirium and is associated with worsening cognitive and physical function in patients with DSD [36, 37]. However, studies investigating the association between delirium and anticholinergic burden have neglected sedative load as a potential mediator. Recently, it was shown that increasing sedative load with minimal anticholinergic burden at baseline was associated with a higher incidence of delirium in community-dwelling older adults with dementia [11]. Consistent with these findings, our results showed that the total anticholinergic and sedative components of DBI increased the overall risk of delirium in patients with dementia. A previous population-based cross-sectional study of 2087 participants showed that the use of these drugs increases the risk of frailty [38]. However, this study did not evaluate delirium as an outcome. Our findings on the association between delirium and exposure to anticholinergics and sedatives are unique.
One possible explanation for these findings, as a neurotransmitter hypothesis, is the deterioration of cholinergic neurotransmission and expansion of neuroinflammation due to activation of microglial cells in patients with dementia [39, 40]. Because of these pathological changes, patients may be more susceptible to the possible adverse effects of drugs. In addition, the cholinergic system plays an important role in inflammation and oxidative stress [41]. Accordingly, high exposure to anticholinergics and sedatives may exacerbate systemic inflammatory status in this vulnerable population. Furthermore, high exposure results in significantly more drug-induced deterioration than low exposure when considered together with variables contributing to a complex clinical condition [42]. We speculated that cumulative high exposure to anticholinergics and sedatives may have a more detrimental effect via these possible changes. Therefore, these underlying inflammatory mechanisms may shed light on the possible relationship between drug exposure and DSD pathogenesis [41].
Although older adults with dementia are more susceptible to adverse effects of anticholinergics and sedatives, these medications are commonly prescribed in clinical practice. This is consistent with recent reports that the prevalence of taking at least one anticholinergic or sedative drug was 10–60% of patients with dementia in a different clinical setting [35, 43–46, 11]. In our study, the higher percentage (72.2%) of patients who were exposed to these medications reflected the greater ability of the DBI to detect anticholinergic and/or sedative medications.
Despite increasing attempts to identify modifiable risk factors, age, gender, educational level, comorbidity, and severity of illness have been significantly and independently associated with delirium in community-dwelling older adults with dementia [4, 47, 48]. For example, a recent study showed that the severity of dementia according to the CDR score increased the risk of delirium after adjusting for age, gender, and education level [49]. Accordingly, these results are consistent with the findings of the present study that the CDR score is one of the most important risk factors for delirium after adjusting for relevant confounders.
This study has certain limitations. Although the baseline risk of high drug exposure was established on admission, this study did not examine the association between changes in the DBI and the risk of delirium over time. However, recent evidence from animal studies suggests that ingestion of drugs with anticholinergic or sedative properties may alter the brain structure [50]. Therefore, the cumulative effects of these drugs may persist over a long period, even when these drugs are discontinued or switched. Although we carefully accounted for several covariates, such as age, gender, CCI score, Lawton-Brody IADL score, CDR score, and polypharmacy; residual confounding remains possible.
The present study has several strengths. Subjects underwent a comprehensive geriatric assessment and were concurrently assessed with the DBI at enrollment, eliminating potential selection bias. In addition, we were able to account for the duration (at least one month) and the dosage of anticholinergic and sedative medications. This study had a large sample size, and a validated scale was used to objectively calculate drug exposure. In addition, the subtypes of dementia and delirium, and the possible etiology of delirium according to the drug exposure level are described in detail. Our study extends the findings of previous studies that reported an association with anticholinergic or sedative drugs and shows that total cumulative dose-related medication exposure should be considered in patients with dementia.