Delirium is one of the most commonly occurring complications in hospitalized elderly patients. At present, there is no evidence for a pharmacologic treatment being effective in prevention. Non‑pharmacologic multicomponent approaches targeting risk factors proved to be the most effective strategy for the prevention and the treatment of delirium.

The aim of this study was to evaluate the effects of non-pharmacological psychosocial interventions in geriatric in-patients on delirium development and new prescriptions of centrally active delirants (NPD).

This observational pre-post implementation study was approved by the local ethics committee and took place at three geriatric wards of the Johanniter Hospital in Bonn. The study consisted of two phases. During the data collection in phase 1 (2016) patients received standard geriatric care. In phase 2 (2018), three new non-pharmacological psychosocial interventions were implemented comprising cognitive stimulation, reminiscence and music therapy. After the second data collection period, an overall evaluation of the study was performed by comparing patients with an intervention to those without an intervention from phase 2 and to the study population from phase 1.

In phase 2, 139 patients were assessed for eligibility and 106 patients were included into the study population, with 10 patients having delirium and 26 patients receiving at least one intervention based on clinical indication (cognitive impairment, delirium). No differences regarding the demographic data were observed comparing the different groups. Patients with and without an intervention were compared and the following characteristics were observed in the intervention group: a higher prevalence of delirium (26.9% vs. 3.9%; p < 0.01), dementia (38.5% vs. 7.5%; p < 0.01) and depression (53.8% vs. 28.8%; p = 0.02); at admission a lower median Mini-Mental State Examination (MMSE) score (22 vs. 27; p < 0.01) and a higher median Clock-Drawing Test (CDT) score (4 vs. 3; p = 0.02), at discharge a lower median Barthel Index (BI) score (62.5 vs. 70; p = 0.02) as well as a significantly higher anticholinergic burden until the second week of hospital stay. Importantly, no significant differences were observed regarding the NPD.

In comparison to the study population from phase 1, patients with an intervention from phase 2, did not differ regarding the MMSE, CDT, BI, potentially inappropriate medications or the anticholinergic burden, but they received more newly prescribed benzodiazepines. This observation is a general underlying pattern, also documented in comparing the study populations of the two phases, and corresponds to a high heterogeneity in geriatric patients.

This observational study indicates that patients with cognitive impairment and delirium on a geriatric ward can be subjected to a non-pharmacological psychosocial intervention. Although functional impairment at discharge in these patients is greater than in controls the amount of NPD is similar. These results suggest that the interventions may have beneficial effects potentially leading to sparing of NPD. Future studies are needed to verify these findings.