Deciding to limit treatment (TLD) is a complicated process influenced by patients’ other medical conditions and cultural considerations. We hypothesized that delirium is associated with the issuance of TLD orders in hospitalized patients. To determine whether the proportion of patients with at least one TLD order differs between those who have delirium and those who do not, to identify which factors are linked to TLD orders, and to investigate how TLD and delirium relate to the 90-day chance of dying or being readmitted to hospital. Every patient admitted to the internal medicine ward of one hospital between November 2019 and January 2020 received a formal delirium assessment by a neuropsychologist within the first 48 hours of admission. Information on TLD orders, deaths within 90 days, and hospital readmissions was collected. The study included 217 patients. Of these, 119 (54.9%) had a “do not resuscitate” order, 107 (49.3%) had a “do not intubate” order, and 77 (35.5%) had a “do not admit to intensive care unit” order. Older age, higher Charlson comorbidity index, cognitive impairment, kidney failure, use of antidepressants or neuroleptics, and malnutrition were all linked to the presence of TLD orders. Patients with delirium were more likely to have at least one TLD order (24 out of 32 patients, 77.4%) than patients without delirium (95 out of 185 patients, 51.1%; OR = 3.3, 95%CI: 1.3-8.0; p < 0.01). After adjustment, this association was no longer statistically significant (aOR = 2.0; 95%CI: 0.7–5.6; p = 0.20). Having a TLD order (aHR = 1.8; 95%CI: 1.1–3.0) and having delirium (aHR = 1.8; 95%CI: 1.1–3.1) were both independently associated with a higher risk of death or hospital readmission within 90 days. Patients found to have delirium in the first 48 hours after hospital admission often receive orders limiting treatment. However, this relationship disappears once other medical conditions are taken into consideration.
