A retrospective cohort study of hospital discharge instructions following delirium episodes

Abstract

Delirium impacts a quarter of hospitalized adults and is associated with increased mortality and cognitive decline.¹ Symptoms may persist for months or recur, and timely recognition is critical.² Caregiver education may improve delirium recognition and ongoing management.^{2, 3} Prior work has demonstrated sub-optimal delirium documentation in discharge summaries, but the extent to which patients and/or caregivers receive written discharge instructions about delirium is unknown.^{4, 5}

Our primary aim was to characterize how clinicians document about delirium in patient/caregiver-directed written discharge instructions. We also assessed patient factors associated with receipt of delirium-related discharge instructions.

We selected charts using a random number generator from a cohort of 1851 older adults (ages ≥65) discharged from medical services at an academic hospital. Patients had ≥1 Confusion Assessment Method (CAM) screen, which are routinely performed each shift by nurses. We included the first 50 charts for which patients had “delirium” documented in their discharge summary.

Among our final cohort, we abstracted verbatim text from the discharge summary problem list and hospital course related to acute changes in cognition. We reviewed the patient/caregiver-directed written discharge instructions (an institutional requirement) and recorded any mention of “delirium” or potential synonyms (e.g., “confusion”). We also identified new prescriptions intended for delirium (e.g., anti-psychotics) and any new or expedited specialty outpatient referrals for delirium, as these are listed in written instructions and may reflect intentions for ongoing delirium care. Unclear cases were adjudicated between two physicians (BPG, EEV). Finally, we used Fisher's exact and Mann–Whitney testing to compare receipt of delirium-related written discharge instructions by patient factors. This study was approved by the UW-Madison Institutional Review Board.

In our sample of 50 adults (Table S1), 76% were female, 88% were White, and the median age was 82 (IQR: 74–87). Most (86%) were discharged from hospital medicine. Approximately 34% had dementia and 64% received inpatient geriatrics consultations (psychiatry and neurology are rarely consulted for delirium at our institution).

All patients had personalized patient/caregiver-directed written discharge instructions regarding a hospital diagnosis, but only 2 cases (4%) specifically mentioned “delirium” (Figure 1). Thirteen charts (26%) had synonyms for delirium (e.g., “confusion” (5), “altered mental status” (5)). One patient received a new anti-psychotic for agitation. Five charts (10%) listed new or expedited outpatient specialty referrals (e.g., geriatrics (3), neurology (2)). Considered together, 18 (36%) contained delirium-related discharge instructions or intended follow-up.

Table 1 contrasts documentation within discharge summary hospital courses and patient/caregiver-directed instructions for several representative patients. Some hospital courses contained detailed descriptions of delirium episodes, including suspected triggers, work-up and whether symptoms were persistent. However, most patient/caregiver-directed written instructions lacked any delirium-related information or post-discharge guidance.

We observed trends toward increased receipt of written instructions about delirium among patients receiving inpatient geriatrics consults (44% vs 11% without) or with >2 days of positive CAM scores (38% vs 21% with ≤2 days), but these associations were not statistically significant (Table S2).

Among a representative subcohort of patients with documented delirium in their discharge summary, approximately a third received any form of delirium-related information or follow-up within patient/caregiver-directed written discharge instructions. We observed discrepancies between what clinicians documented about delirium in clinician-facing hospital courses compared to patient/caregiver-directed discharge instructions. Our findings suggest that even when recognized and documented, the presence of delirium, its significance, and appropriate management instructions may not be regularly communicated to patients and caregivers, demonstrating an important communication gap. These results may reflect uncertainty regarding optimal follow-up care, inadequate evidence to support specific post-discharge interventions, or competing clinical priorities.

Prior work has focused on improving discharge summary quality and improving delirium discharge documentation.^4-6 Our study's focus on patient/caregiver-directed written discharge instructions among patients with recognized and documented delirium is novel. Additional strengths of this study include a representative sample drawn from a full cohort and a systematic chart review among a cohort of recognized delirium using CAM screening.

This is a small study at a single site where neurology and psychiatry consults do not typically manage delirium, which may influence practices. We did not evaluate whether in-person discharge counseling occurred, but such education should be an adjuvant for written instructions.⁷ Finally, our study was not powered to assess smaller, but potentially relevant, factors associated with receipt of discharge instructions.

Patient and caregiver education is an essential component of transitional care.⁸ There are major gaps between delirium communication to clinicians and patients/caregivers. Future work should examine how to provide meaningful delirium discharge instructions, including what information would be most valued and affect patient outcomes.

Study concept and design: Golden, Kaiksow, Inouye, and Vasilevskis. Data acquisition: Golden, Sonnentag, and Kaiksow. Analysis and interpretation of data: All authors. Manuscript preparation: All authors.

Dr. Inouye is the Editor in Chief of JAMA Internal Medicine, and holds the Milton and Shirley F. Levy Family Chair at Hebrew SeniorLife/Harvard Medical School.

The sponsor played no role in the design, methods, subject recruitment, data collections, analysis or preparation of this paper.

This work is supported by the National Institute on Aging of the National Institutes of Health under Award Number K23AG081458 (PI Golden). Dr. Inouye's role was supported by NIA Grant No. R33AG071744 (SKI). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.