Nutrition and Non-Nutrition-Related Challenges Predict Time to Death in Long-Term Care Residents: A Retrospective Chart Review
Bibliographic record
Abstract
Purpose: Signals of end-of-life decline observed in daily habits, such as mealtime participation, are important for moving towards comfort-focused goals of care in the final months of life of long-term care (LTC) residents. It is unclear how eating issues observed in real-time in LTC homes are used as indicators of suspected end of life. The study quantifies nutrition and key non-nutrition related signals (eg, general decline, unstable vitals) documented to describe end-of-life decline and the subsequent time to death. Patients and Methods: A retrospective chart review identified the first documented conversation where end-of-life decline was considered by members of the care team (eg, nurses, physicians, dietitian, family member) for 76 randomly selected decedents from 9 LTC homes in southwestern Ontario, Canada. Time (days) to death was calculated. A directed content analysis of the free-text description of the suspected end-of-life decline was used to categorize signals. Cox proportional hazards regression analysis tested the risk of mortality associated with each categorized signal. Results: Time to death of residents (mean age = 88 ± 7 years; 60% female) from the first documentation of potential end-of-life decline ranged from 0 days to over 2 years prior to death (median = 27.5 days). Seven nutrition-related and 18 non-nutrition related signals were identified. Swallowing difficulty (HR = 2.99; 95% CI = 1.41, 6.33), cognitive decline (HR = 0.40; 95% CI = 0.20, 0.77), delirium (HR = 13.23; 95% CI = 1.57, 111.69), and cancer (HR = 0.18; 95% CI = 0.07, 0.48) were associated with time to death. Conclusion: This study provides insight into the signals used by care providers in LTC to suspect that residents are declining towards the end of life and identifies four signals that were associated with time to death. When identified by care providers as indicators of end-of-life decline, swallowing difficulty and delirium predicted a shorter time to death, while cancer and cognitive decline predicted a longer time to death. Recognition of nutrition and non-nutrition related signals may be leveraged to systematically introduce timely comfort care conversations. Plain Language Summary: This study is part of a larger project exploring the experience of older adults living in residential long-term care (LTC) facilities around nutrition-related challenges (eg, refusing to eat, weight loss) in the final months of life. It is well known that residents experience more complex nutrition-related challenges with proximity to the end of life and that such challenges are associated with increased risk of death. Furthermore, LTC staff rely on observable signals, including nutrition-related challenges, to indicate that residents in their care are experiencing end-of-life decline and to reorient goals of care to emphasize comfort. As such, we carried out this study to identify how common nutrition-related challenges observed in residents may be leveraged to introduce comfort conversations between LTC staff, residents, and their family caregivers. To accomplish this, we performed an in-depth review of resident charts of 76 deceased LTC residents. In our review, we identified the point at which the LTC staff first noted a decline in the residents’ health that potentially indicated end-of-life decline. We took detailed notes of the circumstances surrounding the suspected decline and categorized the reasons given. Our study describes these documented reasons (ie, ‘signals’), which included seven nutrition-related challenges, only one of which (ie, swallowing difficulty) was associated with shorter time to death. From these findings, we can conclude that goals of care should be reassessed when LTC residents experience swallowing difficulty, while the other identified signals (eg, food refusal) may prompt LTC staff to initiate comfort-focused conversations despite uncertain imminence of death. Keywords: time to death, dysphagia, delirium, LTC
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".