Population aging challenges health care systems, particularly emergency departments (EDs) and acute care hospitals, where older adults frequently present with high vulnerability and less favorable outcomes.1,2 Discussing and documenting goals of care (GoC) is essential for aligning treatment with patient preferences.3,4 GoC discussions initiated in EDs and during hospitalization are poorly documented in countries where advance care planning (ACP) is common, and might be particularly scarce in less developed areas.1,5–8 To address these gaps, we investigated its prevalence in a tertiary hospital in Brazil.
Methods
This study, drawing on secondary data from a prospective cohort, included patients aged ≥65 years, consecutively admitted through the ED, between November 2021 and April 2022. At admission, sociodemographic and clinical variables were collected, including clinical severity according to the National Early Warning Score 2 (NEWS 2)9 and geriatric vulnerability according to the PROAGE score.10 Trained medical investigators retrieved information on GoC discussions and other relevant data through electronic health records review. Telephone interviews were conducted at 30, 90, and 180 days to assess deterioration in quality of life and mortality. The study was approved by the local research ethics committee, and written informed consent was provided. Analyses were performed using SPSS22.0 (IBM). We compared baseline and outcome variables based on documentation of GoC.
Results
A total of 830 patients were included (Supplementary Figure 1), with a mean age of 79.7 years; 47% were women, and 92% White. The prevalence of GoC was 8% (n = 67). Patients with documented GoC were older (88 vs 80 years) and more often female (63% vs 45%). They had lower educational level (11 vs 15 years) and were less likely to be married/in a stable union (31% vs 59%). They had higher unemployment rates (90% vs 63%) and were less likely to live with their families (64% vs 76%).
Specialty care differed between groups: cardiologists were the primary physicians for 36% of patients without GoC (vs 22%), whereas geriatricians predominated in the GoC group (43% vs 14%). Among 67 documented GoC discussions, only 7 (10%) involved patient participation, whereas 55 (82%) involved family members (Supplementary Figure 2). Most discussions occurred during hospitalization (55%), some occurred before admission (39%), and a minority occurred in the ED (6%) (Supplementary Figure 2).
Regarding health care utilization, patients with documented GoC visited the ED (57% vs 39%) and were hospitalized (55% vs 40%) more often in the previous 6 months. They exhibited greater functional dependence, requiring more assistance for self-care (91% vs 28%) and hired caregivers (85% vs 24%). They also demonstrated worse clinical severity (National Early Warning Score 2 4 vs 1), geriatric vulnerability (PRO-AGE 6 vs 2), and quality of life scores at admission (EuroQol-5 Dimensions, EQ-5D, 17 vs 9). Rates of prolonged hospitalization (70% vs 30%) and in-hospital adverse events (19% vs 5%) were also higher among patients with documented GoC. They also had higher mortality during hospitalization (45% vs 1%) and at 30 (8% vs <1%), 90 (11% vs 3%), and 180 (19% vs 13%) days. Readmission rates were similar, except at 90 days, when those with documented GoC had higher readmissions (54% vs 27%). Table 1 summarizes these results.
Table 1.
Characteristics and Outcomes of the Study Population
| Characteristic | Total Sample | GoC Documentation | P Value | |
|---|---|---|---|---|
| Yes | No | |||
| N = 830 | (n = 67; 8.1%) | (n = 763; 91.9%) | ||
| Sociodemographic characteristics | ||||
| Age, y | 79.7 ± 8.7 | 87.7 ± 9.3 | 79 ± 8.2 | <.001 |
| Sex | .007 | |||
| Female | 387 (46.6) | 42 (62.7) | 345 (45.2) | |
| Male | 443 (53.4) | 25 (37.3) | 418 (54.8) | |
| Ethnicity | .452 | |||
| White | 762 (91.8) | 61 (91) | 701 (92.1) | |
| Black | 11 (1.3) | 2 (3) | 9 (1.2) | |
| Multiracial | 30 (3.6) | 3 (4.5) | 27 (3.5) | |
| Asian | 25 (3) | 1 (1.5) | 24 (3.2) | |
| Education, y, median (IQR) | 15 (6) | 11 (8) | 15 (6) | .021 |
| Marital status | <.001 | |||
| Married/common law partnership | 470 (56.7) | 21 (31.3) | 449 (58.9) | |
| Other | 360 (43.4) | 46 (68.7) | 314 (41.2) | |
| Religion | .865 | |||
| Catholic and Christian | 656 (79) | 55 (82.1) | 601 (79.7) | |
| Other religions | 78 (9.4) | 7 (10.4) | 71 (9.4) | |
| No religion | 87 (10.5) | 5 (7.5) | 82 (10.9) | |
| Inactive work status | 538 (64.9) | 60 (89.6) | 478 (62.7) | <.001 |
| Relative’s financial support | 51 (6.1) | 6 (9) | 45 (5.9) | .291 |
| Cohabitation | <.001 | |||
| Family | 621 (74.8) | 43 (64.2) | 578 (75.8) | |
| Long-term care facility | 10 (1.2) | 5 (7.5) | 5 (0.7) | |
| Alone | 199 (24) | 19 (28.4) | 180 (23.6) | |
| Clinical features | ||||
| Regular medical follow-up | 781 (91.9) | 63 (94.0) | 718 (94.1) | >.999 |
| Cardiologist | 270 (32.5) | 14 (22.2) | 256 (35.9) | <.001 |
| Geriatrician | 129 (15.5) | 27 (42.9) | 102 (14.3) | |
| General practitioner | 97 (11.7) | 4 (6.3) | 93 (13) | |
| Neurologist | 58 (7) | 4 (6.3) | 54 (7.6) | |
| Oncologist | 33 (4) | 4 (6.3) | 29 (4.1) | |
| Other | 197 (25.3) | 10 (15.9) | 180 (25.2) | |
| Last routine visit to a physician | .504 | |||
| < 1 mo | 527 (63.5) | 48 (76.2) | 479 (66.8) | |
| Between 1 and 3 mo | 116 (14) | 9 (14.3) | 107 (14.9) | |
| ED consultation in the last month | 178 (21.4) | 20 (52.6) | 158 (52.7) | 1.000 |
| ED consultation in the last 6 mo | 339 (40.8) | 38 (56.7) | 301 (39.4) | .007 |
| Hospitalization in the last 6 mo | 340 (41) | 37 (55.2) | 303 (39.7) | .014 |
| Symptoms of anxiety or depression | 490 (59) | 43 (64.2) | 447 (58.6) | .619 |
| Fall in the last year | 250 (30.1) | 23 (35.4) | 227 (32.2) | .678 |
| Needs help with self-care | 277 (33.4) | 61 (91) | 216 (28.3) | <.001 |
| Full-time help | 227 (27.3) | 59 (96.7) | 168 (77.8) | <.001 |
| Hired caregiver | 239 (28.8) | 57 (85.1) | 182 (23.9) | <.001 |
| Katz Index, median (IQR) | 0 (3) | 5 (1) | 0 (1) | <.001 |
| Lawton and Brody, median (IQR) | 27 (13) | 10 (3) | 27 (9) | <.001 |
| NEWS 2, median (IQR) | 1 (3) | 4 (3) | 1 (3) | <.001 |
| PRO-AGE, median (IQR) | 2 (3) | 6 (3) | 2 (3) | <.001 |
| Quality of life (EQ-5D), median (IQR) | 7 (10) | 11 (13) | 7 (9) | <.001 |
| Outcomes | ||||
| Hospital length of stay, d, median (IQR) | 5.2 (6.4) | 12.9 (15.3) | 5 (6.3) | <.001 |
| Prolonged hospitalization (top tertile) | 277 (33.4) | 47 (70.1) | 230 (30.1) | <.001 |
| In-hospital adverse events | 52 (6.3) | 13 (19.4) | 39 (5.1) | <.001 |
| Worsening quality of life | 566 (68.2) | 24 (35.8) | 542 (71.0) | <.001 |
| In-hospital mortality | 36 (4.3) | 30 (44.8) | 6 (0.8) | <.001 |
| 30-d mortality | 4 (0.6) | 2 (7.7) | 2 (0.3) | .009 |
| 90-d mortality | 20 (2.9) | 3 (10.7) | 17 (2.6) | .044 |
| 180-d mortality | 33 (5.1) | 4 (16) | 29 (4.7) | .034 |
EQ-5D, Euro Quality of Life Instrument—5D; IQR, interquartile range; NEWS 2, National Early Warning Score 2. Values are mean ± SD, n (%), or as otherwise indicated. Boldface is used for significant results P < .05.
Discussion
This is the first Latin American study analyzing GoC documentation prevalence in older patients admitted from the ED. The low prevalence echoes previous studies6,11,12; however, lower rates might be expected given regional socioeconomic disparities,7 and highlight a gap in addressing care preferences during acute encounters.
Individuals without stable relationships were more likely to have documented preferences. This finding is supported by a previous report,11 but not by systematic reviews.7,12 Documentation was also more frequent among those with lower education levels, reflecting regional specificities, contrary to those observed in other countries.7,8
Patients with documented GoC were older, had higher health care utilization, had more functional dependence, had worse clinical severity and vulnerability scores, had longer hospital stays, had more in-hospital adverse events, and had increased mortality. These characteristics suggest that discussions occur mainly when vulnerability is easily recognizable, often too late to fully benefit from ACP.8
Only 10% of patients participated in their GoC discussions, likely because of cognitive impairments, acute ED stress, or cultural factors7,8 favoring collective family decision-making. Such dynamics raise important concerns about patient autonomy and warrant further studies.
Geriatrician care predominated in the group with documented discussions, indicating its holistic approach might encompass quality of life and care preferences. The usual disease-specific focus of other specialties might be amenable to ACP training.7,8 Only 6% of GoC discussion and documentation occurred in the ED, suggesting missed opportunities for palliative approach in this setting.2
Study limitations included its single-center and observational design. It is also noteworthy that the scenario described was followed by system-level improvements in our hospital with mandatory documentation of patient’s GoC within all electronic health records clinical notes. This structural change aimed to promote earlier discussions, increase documentation rates, and ultimately enhance patient-centered care. In conclusion, GoC documentation for older adults admitted through the ED in this Latin American setting was critically low, often reactive, and frequently excluded patient participation. Addressing this significant gap through proactive communication strategies and systemic changes is vital for ensuring truly patient-centered care.
Supplementary Material
Supplementary data related to this article can be found online at https://doi.org/10.1016/j.jamda.2025.105795.
Acknowledgments
The authors thank the ProAGE team for patient care and Geriatric ED implementation, and the Geriatric Center for Advanced Medicine (Hospital Sirio-Libanes, Sao Paulo) for sponsorship.
This study was supported by the Geriatric Center for Advanced Medicine, Hospital Sírio-Libanês, São Paulo. Dr Aliberti is supported by the Alzheimer’s Association (23AARFD-1028868) and a postgraduate scholarship from the Maria Emília Foundation. Dr Avelino-Silva is supported by the National Institute on Aging (T32AG000212). The sponsor had no role in study design, data collection/analysis, or manuscript preparation.
Footnotes
Disclosure
The authors declare no conflicts of interest.
Contributor Information
Luciana Pricoli-Vilela, Geriatric Emergency Department Research Group (ProAGE), Hospital Sírio-Libanês, São Paulo, Brazil; Geriatric Center for Advanced Medicine, Hospital Sírio-Libanês, São Paulo, Brazil; Teaching and Research Institute Post-Graduate Program, Hospital Sírio-Libanês, São Paulo, Brazil.
Luiz F.G.L. De Carvalho, Geriatric Emergency Department Research Group (ProAGE), Hospital Sírio-Libanês, São Paulo, Brazil; Geriatric Center for Advanced Medicine, Hospital Sírio-Libanês, São Paulo, Brazil.
Marlon J.R. Aliberti, Geriatric Emergency Department Research Group (ProAGE), Hospital Sírio-Libanês, São Paulo, Brazil; Geriatric Center for Advanced Medicine, Hospital Sírio-Libanês, São Paulo, Brazil; Laboratório de Investigação Médica em Envelhecimento (LIM-66), Serviço de Geriatria, Hospital das Clínicas HCFMUSP, Faculdade de Medicina, Universidade de São Paulo, São Paulo, Brazil; Teaching and Research Institute Post-Graduate Program, Hospital Sírio-Libanês, São Paulo, Brazil.
Márcia M.P. Serra, Teaching and Research Institute Post-Graduate Program, Hospital Sírio-Libanês, São Paulo, Brazil.
Thiago J. Avelino-Silva, Geriatric Emergency Department Research Group (ProAGE), Hospital Sírio-Libanês, São Paulo, Brazil; Geriatric Center for Advanced Medicine, Hospital Sírio-Libanês, São Paulo, Brazil; Laboratório de Investigação Médica em Envelhecimento (LIM-66), Serviço de Geriatria, Hospital das Clínicas HCFMUSP, Faculdade de Medicina, Universidade de São Paulo, São Paulo, Brazil; Division of Geriatrics, University of California, San Francisco, CA, USA.
Christian V. Morinaga, Geriatric Emergency Department Research Group (ProAGE), Hospital Sírio-Libanês, São Paulo, Brazil; Teaching and Research Institute Post-Graduate Program, Hospital Sírio-Libanês, São Paulo, Brazil.
Pedro K. Curiati, Geriatric Emergency Department Research Group (ProAGE), Hospital Sírio-Libanês, São Paulo, Brazil; Geriatric Center for Advanced Medicine, Hospital Sírio-Libanês, São Paulo, Brazil; Teaching and Research Institute Post-Graduate Program, Hospital Sírio-Libanês, São Paulo, Brazil.
References
- 1.Ouchi K, George N, Schuur JD, et al. Goals-of-care conversations for older adults with serious illness in the emergency department: challenges and opportunities. Ann Emerg Med 2019;74:276–284. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.American College of Emergency Physicians. American Geriatrics Society. Emergency Nurses Association. Society for Academic Emergency Medicine. Geriatric Emergency Department Guidelines Task Force. Geriatric emergency department guidelines. Ann Emerg Med 2014;63:e7–e25. [DOI] [PubMed] [Google Scholar]
- 3.Knight T, Malyon A, Fritz Z, et al. Advance care planning in patients referred to hospital for acute medical care: results of a national day of care survey. EClinicalMedicine 2020;19:100235. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Walker LE, Liwonjo A, Goyal DG. Interventions to increase utilisation of advanced care planning documentation for hospitalised older adults. BMJ Open Qual 2025;14:e002703. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Jimenez G, Tan WS, Virk AK, Low CK, Car J, Ho AHY. Overview of systematic reviews of advance care planning: summary of evidence and global lessons. J Pain Symptom Manage 2018;56:436–459.e25. [DOI] [PubMed] [Google Scholar]
- 6.Brinkman-Stoppelenburg A, Rietjens JAC, van der Heide A. The effects of advance care planning on end-of-life care: a systematic review. Palliat Med 2014;28:1000–1025. [DOI] [PubMed] [Google Scholar]
- 7.Rezaei MA, Zahiri A, Kianian T, et al. Factors related to patients’ readiness for advance care planning: a systematic review. BMC Public Health 2025;25:78. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.McMahan RD, Tellez I, Sudore RL. Deconstructing the complexities of advance care planning outcomes: what do we know and where do we go? A scoping review. J Am Geriatr Soc 2021;69:234–244. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Royal College of Physicians. National Early Warning Score (NEWS) 2: stand-ardising the assessment of acute-illness severity in the NHS. Accessed April 30, 2025. https://www.rcp.ac.uk/media/a4ibkkbf/news2-final-report_0_0.pdf
- 10.Curiati PK, Gil-Junior LA, Morinaga CV, Ganem F, Curiati JAE, Avelino-Silva TJ. Predicting hospital admission and prolonged length of stay in older adults in the emergency department: the PRO-AGE scoring system. Ann Emerg Med 2020;76:255–265. [DOI] [PubMed] [Google Scholar]
- 11.Panozzo L, Doherty Z, Leach M, Ward B. Prevalence and correlates of advance care planning among Australian hospital inpatients. Intern Med J 2023;53:610–614. [DOI] [PubMed] [Google Scholar]
- 12.Oulton J, Rhodes SM, Howe C, Fain MJ, Mohler MJ. Advance directives for older adults in the emergency department: a systematic review. J Palliat Med 2015; 18:500–505. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.