Prediction of 90-day mortality risk after unplanned emergency department visits of advanced stage cancer patients.

Introduction

Despite a slight decline in incidence rates, cancer is the leading cause of death in high-income countries [ 1 – 3 ]. Cancer patients undergoing antineoplastic therapy are at increased risk of adverse events leading to unplanned emergency department visits [ 4 ]. According to observational studies, 50% of cancer patients receiving antineoplastic therapy will be hospitalized at some point [ 5 ].

Breast, lung, gastrointestinal (GI) tract and prostate cancer represent the most common cancer diagnoses at the emergency department (ED) [ 4 , 6 – 8 ]. The most frequent reasons for ED visits for cancer patients are respiratory distress, pain, GI symptoms, and fever [ 6 – 8 ]. In the last 10 years, the rate of emergency department visits due to complications of radiation therapy or systemic treatment has increased 5.5-fold [ 9 ]. Interestingly, hospitalization rates of cancer patients at the ED are significantly higher for “real world patients” compared to patients in randomized clinical trials (51% vs. 16%) [ 5 ]. Studies estimate that around 50% of unplanned ED visits of cancer patients can be preventable and are merely an expression of insufficient supportive care, especially in patients with advanced cancer [ 10 – 12 ].

Over 90% of treatment-related emergency visits by cancer patients ultimately lead to hospital admission and around 5% of these patients die during hospitalization [ 9 ].

Besides the high admission- and in-patient death rate, ED visits by oncologic patients represent a significant economic load for health care systems worldwide. In 2007, it was calculated that the cost for a chemotherapy-related ED visit was approximately $ 800. However, an inpatient treatment was estimated to cost roughly $ 22.000 [ 13 ]. Besides the financial burden, adverse events leading to unplanned ED visits and inpatient treatments may cause delays in chemotherapy administration, which can have detrimental effects on the survival of cancer patients [ 14 ].

Unplanned ED visits by cancer patients either due to therapy or cancer-related complications will not be fully preventable. Given the high frequency of ED visits and their burden on health care systems, it would be beneficial to guide the ED and inpatient facility resources to cancer patients at the highest risk for a detrimental outcome.

It is of paramount importance to identify cancer patients at the ED with a higher risk for an adverse outcome, thus enabling health care providers to adequately allocate resources. Both, low albumin and low BMI are associated with patient’s nutritional status, increased adverse events of systemic cancer therapy, and decreased survival [ 15 – 18 ]. Albumin and BMI are parameters that are easy to assess and obtain in an emergency medicine setting.

The goal of this retrospective pilot study was to identify easily accessible patient-derived risk factors for 90-day mortality after an unplanned ED visit. We aimed to investigate the influence of body mass index (BMI) and serum albumin at ED presentation on the mortality risk of advanced cancer patients at the ED.

Methods

We performed a retrospective chart review of all patients with a documented oncological diagnosis according to ICD-10 coding that was presented at the Department of Emergency Medicine of the General Hospital Vienna (AKH Wien) between August 2016 and July 2018. Patients who did not receive treatment at the AKH Wien or were treated at a Department other than the Department for Medicine I, Division of Oncology, were excluded. We also excluded patients in a curative setting and such patients who were regularly seen by physicians of our division but did not receive antitumor treatment (i.e., best supportive care, watch and wait). Active antitumor treatment was defined as regular administration of antitumoral substances through the outpatient infusion clinic or the inpatient ward or, in the case of oral antitumor agents, regular checkups at our outpatient department within the last 90 days before presenting at the emergency department. In patients who presented more than once at the emergency department, we focused on the first visit.

Information on tumor entity, type of treatment, as well as reasons for presentation were collected. Presenting symptoms were grouped into several subcategories as was the final diagnosis at the time of discharge. We also assessed whether the presentation was related to the tumor, the antitumor treatment, or for unrelated reasons. In case the date of death was not documented in our hospital documentation system, the insurance data check-up was performed (cutoff date 2022/06/20) to calculate 90-day mortality. Laboratory values, emergency severity index (ESI), and BMI were collected at the time of presentation.

Statistical methods

Qualitative variables are reported by absolute and relative frequencies, and quantitative variables by median and range. Correlations between 90-day mortality with single covariates were investigated with Chi-squared tests, and a logistic regression model for 90-day mortality with covariates albumin, BMI, sex, and oncological diagnosis was analyzed. Combined effects of low BMI and low albumin were analyzed by first introducing a new variable counting how many of the two values were lowered (0, 1, or 2), and then comparing Kaplan–Meier curves for each pair of values by log-rank tests with Bonferroni-Holm adjustment. A two-sided p -value of less than 0.05 was considered significant.

Results

A total of 2870 patients with cancer were identified. After the application of exclusion criteria, 448 patients were included in the final analysis. Median age was 64.5 (range 18–92) years, and 203 (45.3%) of patients were female. The most common oncological diagnoses were lung cancer (19.2%), pancreatobiliary carcinoma (15.6%), and colorectal cancer (11.2%). Almost half of the patients were treated in the setting of the outpatient infusion clinic (68.1%), either with chemotherapy alone (47.3%) or with the combination of chemotherapy and monoclonal antibodies (13.8%). In total, 42.4% of emergency department visits were due to problems related to the tumor itself, 22.1% to the antitumor therapy, and 35.5% to problems considered unrelated to the underlying neoplasia. The most common presenting complaints were pain (20.5%), fever (17.4%), or respiratory problems (14.7%). A summary can be found in Table 1 . Sixty-nine percent of patients had to be admitted, while 31.0% could be discharged to home care. The majority of patients were discharged with a diagnosis of infection (23.4%) or tumor progression (17.6%). Problems related to pain or gastrointestinal symptoms (i.e., diarrhea, nausea/vomiting, and constipation) could be most frequently handled in an outpatient setting (51.1% and 38.8%, respectively). Table 1 Baseline characteristics and distribution of all variables by patient according to their management. BMI body mass index, ESI emergency severity index Discharged from ED Required admission Total (% of total) 139 (31.0%) 309 (69.0%) 448 (100%) Sex Male 72 (51.8%) 173 (56.0%) 245 (54.7%) Female 67 (48.2%) 136 (44.0%) 203 (45.3%) Age (median, range) 62 (19–86) 66 (20–92) 65 (19–92) Oncological diagnosis Lung 21 (24.4%) 65 (75.6%) 86 (19.2%) Pancreatobiliary 20 (28.6%) 50 (71.4%) 70 (15.6%) Colorectal 20 (40.0%) 30 (60.0%) 50 (11.2%) Breast 22 (48.9%) 23 (51.1%) 45 (10.0%) Head/neck 6 (17.1%) 29 (82.9%) 35 (7.8%) Sarcoma 10 (33.3%) 20 (66.7%) 30 (6.7%) Upper GI 2 (8.7%) 21 (91.3%) 23 (5.1%) Lymphoma/multiple myeloma 9 (40.9%) 13 (59.1%) 22 (4.9%) Prim. CNS 13 (59.1%) 9 (40.9%) 22 (4.9%) Renal cell carcinoma 6 (33.3%) 12 (66.7%) 18 (4.0%) Neuroendocrine 4 (36.4%) 7 (63.6%) 11 (2.5%) Other 6 (16.7%) 30 (83.3%) 36 (8.0%) Presenting symptom Pain 47 (51.1%) 45 (48.9%) 92 (20.5%) Fever 14 (17.9%) 64 (82.1%) 78 (17.4%) Respiratory 16 (24.2%) 50 (75.8%) 66 (14.7%) Deterioration of general condition 6 (11.5%) 46 (88.5%) 52 (11.6%) Diarrhea/nausea/vomiting/constipation 19 (38.8%) 30 (61.2%) 49 (10.9%) Neurological symptoms 15 (32.6%) 31 (67.4%) 46 (10.3%) Bleeding 7 (30.4%) 16 (69.6%) 23 (5.1%) Cardiological symptom 5 (45.5%) 6 (54.5%) 11 (2.5%) Edema/effusion 3 (37.5%) 5 (62.5%) 8 (1.8%) Urological symptoms 2 (33.3%) 4 (66.7%) 6 (1.3%) Local problem at tumor site 1 (20.0%) 4 (80.0%) 5 (1.1%) Medical device 2 (66.7%) 1 (33.3%) 3 (0.7%) Other 2 (22.2%) 7 (77.8%) 9 (2.0%) Discharge diagnosis Infection 20 (19.0%) 85 (81.0%) 105 (23.4%) Tumor progression 21 (26.6%) 58 (73.4%) 79 (17.6%) Pain 25 (67.6%) 12 (32.4%) 37 (8.3%) Diarrhea/nausea/vomiting/constipation 15 (44.1%) 19 (55.9%) 34 (7.6%) Edema/effusion 7 (31.8%) 15 (68.2%) 22 (4.9%) Cardiological symptom 7 (38.9%) 11 (61.1%) 18 (4.0%) Bleeding 4 (23.5%) 13 (76.5%) 17 (3.8%) Blood clot/thrombosis 5 (29.4%) 12 (70.6%) 17 (3.8%) Surgical 1 (6.3%) 15 (93.8%) 16 (3.6%) Neurological 8 (57.1%) 6 (42.9%) 14 (3.1%) Hematological toxicity 2 (15.4%) 11 (84.6%) 13 (2.9%) Exsiccosis 4 (30.8%) 9 (69.2%) 13 (2.9%) Respiratory symptom 5 (50.0%) 5 (50.0%) 10 (2.2%) Acute kidney injury 1 (11.1%) 8 (88.9%) 9 (2.0%) Medical device problem 2 (22.2%) 7 (77.8%) 9 (2.0%) Drug related problem 2 (40.0%) 3 (60.0%) 5 (1.1%) Local problem at tumor site 1 (25.0%) 3 (75.0%) 4 (0.9%) Urological symptom 1 (25.0%) 3 (75.0%) 4 (0.9%) Deterioration of general condition 1 (33.3%) 2 (66.7%) 3 (0.7%) Other 7 (36.8%) 12 (63.2%) 19 (4.2%) Association Tumor 54 (28.4%) 136 (71.6%) 190 (42.4%) Therapy 29 (29.3%) 70 (70.7%) 99 (22.1%) Other 56 (35.2%) 103 (64.8%) 159 (35.5%) BMI (median, range) 24.7 (13.3–42.9) 23.0 (14.5–39.1) 23.5 (13.3–42.9) Albumin g/L (median, range) 39.3 (16.8–47.0) 34.45 (16.8–46.2) 35.7 (16.8–47.0) ESI 1 0 (0.0%) 2 (100.0%) 2 (0.4%) 2 29 (21.0%) 109 (79.0%) 138 (30.8%) 3 91 (37.9%) 149 (62.1%) 240 (53.6%) 4 8 (47.1%) 9 (52.9%) 17 (3.8%) Missing 11 (21.6%) 40 (78.4%) 51 (11.4%) 90-day mortality 31 (22.3%) 154 (49.8%) 185 (41.3%)

Median time between the last contact and the time of presentation at the emergency department was 7, 10, and 8 days for patients treated at the outpatient infusion clinic, the inpatient ward, and the outpatient department, respectively. In patients who had to be admitted, the median duration of admission was 8 days (0–89).

Of patients who had to be admitted, 54 (17.5%) patients died while being an inpatient. This proportion was highest in patients admitted for a tumor-related problem (48 of 136 (35.3%)) While the 90-day mortality rate in inpatients was 49.8%, it was only 22.3% in outpatients. Tumor-related cases had the highest 90-day mortality rate among both groups (29.5% in inpatients, 38.8% for outpatients).

ESI and BMI were available in 390 and 398 patients, respectively. Amongst them, the most common ESI grading was 3 (61.5%).

Impact of albumin and BMI on 90-day mortality

The median BMI was 23.5 (13.3–42.9). Albumin levels at the time of presentation were available in 355 (81.7%) patients; median albumin was at 35.7 g/L (16.8–47.0).

Patients with low (< 35 g/L) albumin levels showed higher 90-day mortality than those with high levels (> 35 g/L; 60.4% vs. 31.4%, p < 0.0001). No difference in 90-day mortality between weight classes was identified by the chi-squared test (low BMI 55.9%, normal BMI 42.9%, high BMI 39.5%, p = 0.22). However, Kaplan–Meier curves show separation between the three curves, and lower hazards were observed in patients with normal (HR 0.67, 95% CI 0.46–0.96) and high (HR 0.55, 95% CI 0.38–0.81) BMI than patients with low BMI. Log-rank tests with Bonferroni-Holm adjustment only identify a difference between high and low BMI ( p = 0.0065). Unlike BMI and albumin levels, age below or above 65 years was not a predictor of 90-day mortality (43.7% vs. 38.4%, p = 0.2972).

We then classified patients according to whether both albumin and BMI were under the critical values, whether one of both values was below the cutoff, or none of them. Pairwise differences identified significant differences in 90-day mortality between patients with both values high and patients with one value low (32.7% vs. 56.6%, p < 0.01) and between patients with both values high and patients with both values low (32.7% vs. 78.6%, p < 0.01), but not between patients with one or two values low ( p = 0.189). Regarding overall survival, hazards are higher for one or both levels low than for both levels high (one low HR 1.61, 95% CI 1.27–2.04; both low HR 4.01, 95% CI 2.30–7.02). Log-rank tests with Bonferroni-Holm adjustment identified differences between all three pairs of groups (Fig. 1 ). Fig. 1 Kaplan Meier curves for patients with neither albumin nor BMI lowered (“both high”), one or both values decreased

No difference between sexes in BMI ( p = 0.473) or albumin levels ( p = 0.072) could be identified by t -tests. BMI and albumin are weakly correlated ( r = 0.16, p = 0.004). We performed a logistic regression model for 90-day mortality with free variables sex, BMI (classified into low, normal, and high), albumin (classified as low or high), and tumor diagnosis with lung as the reference group due to the highest complete case number ( n = 86; estimated 90-day mortality, 0.45; 95% CI 0.35–0.56). High albumin levels and female sex conferred a better 90-day mortality. No significant difference was identified for any tumor diagnosis, with the exception of lymphoma/multiple myeloma, which had lower 90-day mortality (est, 0.18; 95% CI [0.07, 0.39], Table 2 ). Table 2 Logistic regression of 90-day mortality on albumin, BMI, and sex OR 95% CI p value (Intercept) 2.874 [0.985, 8.756] 0.057 Albumin > 35 g/L 0.312 [0.186, 0.515] 0 BMI normal 0.795 [0.303, 2.041] 0.634 BMI high 0.716 [0.263, 1.913] 0.505 Female sex 0.531 [0.313, 0.892] 0.018 Breast cancer 1.63 [0.607, 4.368] 0.329 Colorectal cancer 0.803 [0.312, 2.023] 0.643 Upper GI cancer 0.959 [0.321, 2.909] 0.94 Head and neck cancer 1.563 [0.6, 4.205] 0.365 Lymphoma/multiple myeloma 0.137 [0.02, 0.569] 0.015 Neuroendocrine carcinomas 0.541 [0.069, 2.955] 0.502 Pancreatobiliary cancer 0.697 [0.327, 1.468] 0.344 Primary CNS tumors 0.195 [0.01, 1.141] 0.133 Renal cell carcinoma 0.552 [0.098, 2.695] 0.469 Sarcoma 2.047 [0.722, 6.103] 0.184 Other 1.089 [0.417, 2.878] 0.862

Earlier analyses have shown that a significant proportion of unplanned ED visits could be preventable [ 19 , 20 ]. While individual approaches of the studies varied, presenting complaints such as pain and gastrointestinal symptoms like nausea, vomiting, diarrhea, or constipation were generally considered preventable, since their occurrence could be avoided through patient counseling and intensified supportive care. Applying this to our cohort, 141 (31.5%) of ED visits may have been avoidable. When applying our model to only these patients (both parameters were known in 79 patients with preventable complaints), more patients with one or two low parameters had to be admitted to inpatient (Table 3 ). This difference reached significance between patients with both values high when compared with patients with one value below the cutoff ( p = 0.034). Table 3 Patients with preventable diagnosis and their ED handling according to levels of albumin and BMI (both parameters were known in 79 patients) Discharge from ED Required admission Both high 18 (78.3%) 26 (46.4%) One low 5 (21.7%) 27 (48.2%) Both low 0 (0.0%) 3 (5.4%)

Discussion

Unplanned ED visits are a frequent occurrence in cancer patients. This puts a significant strain on ED services and medical services. It also reflects the poor experience many cancer patients have with adverse events during their antitumor treatment.

This analysis from a cohort of cancer patients undergoing palliative antineoplastic treatment at a tertiary care center highlights several points. Firstly, our data corroborate earlier results (4) suggesting that patients in a palliative setting have high admission rates (69%) upon presentation at the ED. Unsurprisingly in this patient cohort, most patients presented due to tumor-related problems, while roughly a fifth of patients presented with complications of antitumor-therapy. Regarding the 35.5% of patients whose presentation was regarded as unrelated (e.g., cardiologic symptoms), a degree of uncertainty remains, whether these problems could still be triggered by the underlying disease. The median time between the last contact and the ED presentation was between 7 and 10 days depending on the setting in which the therapy was admitted. This confirms the general clinical impression that the cancer patients’ vulnerability is highest after the first week after therapy.

Hospital admission correlated with a modest mortality rate of 17.5%; however, 90-day mortality for inpatients was twice as high than in patients discharged to home care (49.8% vs. 22.3%). The highest 90-day mortality was seen in tumor-related cases (most commonly tumor progression).

Given the marked inpatient mortality rate and the high 90-day mortality rate, especially in patients admitted to inpatient care, there is a need for simple yet effective tools to predict outcome in cancer patients requiring acute medical services.

Efforts have been made to apply clinical scores predicting the risk for patients undergoing antineoplastic therapy for unplanned ED visits [ 11 , 21 , 22 ]. As an example, the REDUCE Score incorporates comorbidities (COPD, congestive heart failure, renal failure) and blood parameters (hemoglobin, albumin) and takes prior ED visits (in the past 90 days) into account to highlight cancer patients at increased risk for side effects and ED visits [ 11 ]. Other scores that aim to predict the outcome of cancer patients like the (modified) Glasgow prognostic score (m)GPS, that is based on simple, readily available blood parameters CRP and albumin, are only validated to assess survival probability at baseline [ 23 – 25 ].

Further, CRP levels can be volatile and their elevation is multifactorial in the course of the disease, in particular as infection associated with rises in CRP levels is one of the most prevalent reasons for unplanned ED visits of cancer patients [ 4 ]. Therefore, we wanted to avoid incorporating CRP levels in our risk model.

Albumin and to a lesser degree low BMI have been recognized as markers of tumor-associated cachexia. In our model, subnormal albumin levels and low BMI were predictive for inferior survival, when compared to high levels and normal or high BMI or patients in which just one of these parameters were subnormal. Whether low albumin per se is a negative prognostic factor as an expression of high tumor burden or liver dysfunction due to metastatic disease, or as a surrogate parameter for cachexia is beyond the scope of this study.

One limitation of this study is its retrospective design, which did not allow us to evaluate our score for the prediction of ED visits by cancer patients.

Conclusion

In our cohort of 448 cancer patients, low albumin and low BMI were associated with higher 90-day mortality. Taken together, our data identify a cohort of patients who could benefit from more intensive supportive care in order to prevent unplanned ED visits. Further, they may help in selecting patients who would benefit from inpatient admission. In addition, these data may provide initial guidance on whether to discuss best supportive or hospice care in this patient population. However, this goes beyond the role of ED consultations and should be reserved for the treating physicians primarily involved in the oncological therapy.

More studies are urgently needed to predict ED visits and unfavorable outcomes of cancer patients in palliative care.

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Author contribution

GJ, CK, and CM designed the study. Material preparation, and data collection were performed by GJ, SP, and CM. Data analysis was carried out by GJ, CK, and CM. The first draft of the manuscript was written by GJ, CK, and CM. All authors read and critically reviewed the manuscript. All authors read and approved the final manuscript.

Funding

Open access funding provided by Medical University of Vienna.

Data availability

Data of this study is available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

The retrospective database on which the study was based was approved by the ethics committee of the Medical University of Vienna (1860/20171860/2017). Due to the retrospective nature of the study, no informed consent of the individual patients is necessary according to local regulations.

Competing interests

MP has received honoraria for lectures, consultation or advisory board participation from the following for-profit companies: Bayer, Bristol-Myers Squibb, Novartis, Gerson Lehrman Group (GLG), CMC Contrast, GlaxoSmithKline, Mundipharma, Roche, BMJ Journals, MedMedia, Astra Zeneca, AbbVie, Lilly, Medahead, Daiichi Sankyo, Sanofi, Merck Sharp & Dome, Tocagen, Adastra, Gan & Lee Pharmaceuticals, Janssen, Servier, Miltenyi, Böhringer-Ingelheim, Telix, Medscape. The other authors have no relevant financial or non-financial interests to disclose.

The other authors have no relevant financial or non-financial interests to disclose.

References

1. Malvezzi M Santucci C Boffetta P Collatuzzo G Levi F La Vecchia C Negri E European cancer mortality predictions for the year 2023 with focus on lung cancer Ann Oncol 2023 34 410 419 10.1016/j.annonc.2023.01.010 36882139 Malvezzi M, Santucci C, Boffetta P, Collatuzzo G, Levi F, La Vecchia C, Negri E (2023) European cancer mortality predictions for the year 2023 with focus on lung cancer. Ann Oncol 34:410–419 36882139 10.1016/j.annonc.2023.01.010

2. Siegel RL Miller KD Fuchs HE Jemal A (2022) Cancer statistics CA Cancer J Clin 2022 72 7 33 10.3322/caac.21708 35020204 Siegel RL, Miller KD, Fuchs HE (2022) Jemal A (2022) Cancer statistics. CA Cancer J Clin 72:7–33 35020204 10.3322/caac.21708

3. Mahase E Cancer overtakes CVD to become leading cause of death in high income countries BMJ 2019 366 l5368 10.1136/bmj.l5368 31481521 Mahase E (2019) Cancer overtakes CVD to become leading cause of death in high income countries. BMJ 366:l5368 31481521 10.1136/bmj.l5368

4. Caterino JM Adler D Durham DD Yeung SJ Hudson MF Bastani A Bernstein SL Baugh CW Coyne CJ Grudzen CR Henning DJ Klotz A Madsen TE Pallin DJ Reyes-Gibby CC Rico JF Ryan RJ Shapiro NI Swor R Venkat A Wilson J Thomas CR Jr Bischof JJ Lyman GH Analysis of diagnoses, symptoms, medications, and admissions among patients with cancer presenting to emergency departments JAMA Netw Open 2019 2 e190979 10.1001/jamanetworkopen.2019.0979 30901049 PMC6583275 Caterino JM, Adler D, Durham DD, Yeung SJ, Hudson MF, Bastani A, Bernstein SL, Baugh CW, Coyne CJ, Grudzen CR, Henning DJ, Klotz A, Madsen TE, Pallin DJ, Reyes-Gibby CC, Rico JF, Ryan RJ, Shapiro NI, Swor R, Venkat A, Wilson J, Thomas CR Jr, Bischof JJ, Lyman GH (2019) Analysis of diagnoses, symptoms, medications, and admissions among patients with cancer presenting to emergency departments. JAMA Netw Open 2:e190979 30901049 10.1001/jamanetworkopen.2019.0979 PMC6583275

5. Prince RM Atenafu EG Krzyzanowska MK Hospitalizations during systemic therapy for metastatic lung cancer: a systematic review of real world vs clinical trial outcomes JAMA Oncol 2015 1 1333 1339 10.1001/jamaoncol.2015.3440 26378774 Prince RM, Atenafu EG, Krzyzanowska MK (2015) Hospitalizations during systemic therapy for metastatic lung cancer: a systematic review of real world vs clinical trial outcomes. JAMA Oncol 1:1333–1339 26378774 10.1001/jamaoncol.2015.3440

6. Rivera DR Gallicchio L Brown J Liu B Kyriacou DN Shelburne N Trends in adult cancer-related emergency department utilization: an analysis of data from the nationwide emergency department sample JAMA Oncol 2017 3 e172450 10.1001/jamaoncol.2017.2450 28859189 PMC5824316 Rivera DR, Gallicchio L, Brown J, Liu B, Kyriacou DN, Shelburne N (2017) Trends in adult cancer-related emergency department utilization: an analysis of data from the nationwide emergency department sample. JAMA Oncol 3:e172450 28859189 10.1001/jamaoncol.2017.2450 PMC5824316

7. Whitney RL Bell JF Tancredi DJ Romano PS Bold RJ Wun T Joseph JG Unplanned hospitalization among individuals with cancer in the year after diagnosis J Oncol Pract 2019 15 e20 e29 10.1200/JOP.18.00254 30523749 PMC7010432 Whitney RL, Bell JF, Tancredi DJ, Romano PS, Bold RJ, Wun T, Joseph JG (2019) Unplanned hospitalization among individuals with cancer in the year after diagnosis. J Oncol Pract 15:e20–e29 30523749 10.1200/JOP.18.00254 PMC7010432

8. Mayer DK Travers D Wyss A Leak A Waller A Why do patients with cancer visit emergency departments? Results of a 2008 population study in North Carolina J Clin Oncol 2011 29 2683 2688 10.1200/JCO.2010.34.2816 21606431 PMC3139372 Mayer DK, Travers D, Wyss A, Leak A, Waller A (2011) Why do patients with cancer visit emergency departments? Results of a 2008 population study in North Carolina. J Clin Oncol 29:2683–2688 21606431 10.1200/JCO.2010.34.2816 PMC3139372

9. Jairam V Lee V Park HS Thomas CR Jr Melnick ER Gross CP Presley CJ Adelson KB Yu JB Treatment-related complications of systemic therapy and radiotherapy JAMA Oncol 2019 5 1028 1035 10.1001/jamaoncol.2019.0086 30946433 PMC6583836 Jairam V, Lee V, Park HS, Thomas CR Jr, Melnick ER, Gross CP, Presley CJ, Adelson KB, Yu JB (2019) Treatment-related complications of systemic therapy and radiotherapy. JAMA Oncol 5:1028–1035 30946433 10.1001/jamaoncol.2019.0086 PMC6583836

10. Brooks GA Abrams TA Meyerhardt JA Enzinger PC Sommer K Dalby CK Uno H Jacobson JO Fuchs CS Schrag D Identification of potentially avoidable hospitalizations in patients with GI cancer J Clin Oncol 2014 32 496 503 10.1200/JCO.2013.52.4330 24419123 PMC3918534 Brooks GA, Abrams TA, Meyerhardt JA, Enzinger PC, Sommer K, Dalby CK, Uno H, Jacobson JO, Fuchs CS, Schrag D (2014) Identification of potentially avoidable hospitalizations in patients with GI cancer. J Clin Oncol 32:496–503 24419123 10.1200/JCO.2013.52.4330 PMC3918534

11. Csik VP Li M Binder AF Handley NR Development of an oncology acute care risk prediction model JCO Clin Cancer Inform 2021 5 266 271 10.1200/CCI.20.00146 33720762 Csik VP, Li M, Binder AF, Handley NR (2021) Development of an oncology acute care risk prediction model. JCO Clin Cancer Inform 5:266–271 33720762 10.1200/CCI.20.00146

12. Delgado-Guay MO Kim YJ Shin SH Chisholm G Williams J Allo J Bruera E Avoidable and unavoidable visits to the emergency department among patients with advanced cancer receiving outpatient palliative care J Pain Symptom Manage 2015 49 497 504 10.1016/j.jpainsymman.2014.07.007 25131891 Delgado-Guay MO, Kim YJ, Shin SH, Chisholm G, Williams J, Allo J, Bruera E (2015) Avoidable and unavoidable visits to the emergency department among patients with advanced cancer receiving outpatient palliative care. J Pain Symptom Manage 49:497–504 25131891 10.1016/j.jpainsymman.2014.07.007

13. Pyenson BS, Fitch K (2010) Cancer patients receiving chemotherapy: opportunities for better management. Milliman Inc. https://www.milliman.com/en/insight/cancer-patients-receiving-chemotherapy-opportunities-for-better-management . Accessed 8 Oct 2024

14. Adelson K Dest V Velji S Lisitano R Lilenbaum R Emergency department (ED) utilization and hospital admission rates among oncology patients at a large academic center and the need for improved urgent care access J Clin Oncol 2014 32 19 19 10.1200/jco.2014.32.30_suppl.19 Adelson K, Dest V, Velji S, Lisitano R, Lilenbaum R (2014) Emergency department (ED) utilization and hospital admission rates among oncology patients at a large academic center and the need for improved urgent care access. J Clin Oncol 32:19–1924276775

15. Kichenadasse G Miners JO Mangoni AA Rowland A Hopkins AM Sorich MJ Association between body mass index and overall survival with immune checkpoint inhibitor therapy for advanced non-small cell lung cancer JAMA Oncol 2020 6 512 518 10.1001/jamaoncol.2019.5241 31876896 PMC6990855 Kichenadasse G, Miners JO, Mangoni AA, Rowland A, Hopkins AM, Sorich MJ (2020) Association between body mass index and overall survival with immune checkpoint inhibitor therapy for advanced non-small cell lung cancer. JAMA Oncol 6:512–518 31876896 10.1001/jamaoncol.2019.5241 PMC6990855

16. Liu X Shi H One-year mortality in patients with cancer cachexia: association of albumin and total protein Clin Nutr ESPEN 2021 46 S716 S717 10.2147/CMAR.S318728 PMC8412822 34512017 Liu X, Shi H (2021) One-year mortality in patients with cancer cachexia: association of albumin and total protein. Clin Nutr ESPEN 46:S716–S717 10.2147/CMAR.S318728 PMC8412822 34512017

17. Tu H McQuade JL Davies MA Huang M Xie K Ye Y Chow WH Rodriguez A Wu X Body mass index and survival after cancer diagnosis: a pan-cancer cohort study of 114 430 patients with cancer Innovation (Camb) 2022 3 100344 36353671 10.1016/j.xinn.2022.100344 PMC9638833 Tu H, McQuade JL, Davies MA, Huang M, Xie K, Ye Y, Chow WH, Rodriguez A, Wu X (2022) Body mass index and survival after cancer diagnosis: a pan-cancer cohort study of 114 430 patients with cancer. Innovation (Camb) 3:100344 36353671 10.1016/j.xinn.2022.100344 PMC9638833

18. Gatta A Verardo A Bolognesi M Hypoalbuminemia Intern Emerg Med 2012 7 Suppl 3 S193 199 10.1007/s11739-012-0802-0 23073857 Gatta A, Verardo A, Bolognesi M (2012) Hypoalbuminemia. Intern Emerg Med 7(Suppl 3):S193-199 23073857 10.1007/s11739-012-0802-0

19. Alishahi Tabriz A Turner K Hong YR Gheytasvand S Powers BD Elston Lafata J Trends and characteristics of potentially preventable emergency department visits among patients with cancer in the US JAMA Netw Open 2023 6 e2250423 10.1001/jamanetworkopen.2022.50423 36656584 PMC9857289 Alishahi Tabriz A, Turner K, Hong YR, Gheytasvand S, Powers BD, Elston Lafata J (2023) Trends and characteristics of potentially preventable emergency department visits among patients with cancer in the US. JAMA Netw Open 6:e2250423 36656584 10.1001/jamanetworkopen.2022.50423 PMC9857289

20. Panattoni L Fedorenko C Greenwood-Hickman MA Kreizenbeck K Walker JR Martins R Eaton KD Rieke JW Conklin T Smith B Lyman G Ramsey SD Characterizing potentially preventable cancer- and chronic disease-related emergency department use in the year after treatment initiation: a regional study J Oncol Pract 2018 14 e176 e185 10.1200/JOP.2017.028191 29452549 Panattoni L, Fedorenko C, Greenwood-Hickman MA, Kreizenbeck K, Walker JR, Martins R, Eaton KD, Rieke JW, Conklin T, Smith B, Lyman G, Ramsey SD (2018) Characterizing potentially preventable cancer- and chronic disease-related emergency department use in the year after treatment initiation: a regional study. J Oncol Pract 14:e176–e185 29452549 10.1200/JOP.2017.028191

21. Brooks GA Kansagra AJ Rao SR Weitzman JI Linden EA Jacobson JO A clinical prediction model to assess risk for chemotherapy-related hospitalization in patients initiating palliative chemotherapy JAMA Oncol 2015 1 441 447 10.1001/jamaoncol.2015.0828 26181251 PMC4576342 Brooks GA, Kansagra AJ, Rao SR, Weitzman JI, Linden EA, Jacobson JO (2015) A clinical prediction model to assess risk for chemotherapy-related hospitalization in patients initiating palliative chemotherapy. JAMA Oncol 1:441–447 26181251 10.1001/jamaoncol.2015.0828 PMC4576342

22. Hong JC Niedzwiecki D Palta M Tenenbaum JD Predicting emergency visits and hospital admissions during radiation and chemoradiation: an internally validated pretreatment machine learning algorithm JCO Clin Cancer Inform 2018 2 1 11 10.1200/CCI.18.00037 30652595 Hong JC, Niedzwiecki D, Palta M, Tenenbaum JD (2018) Predicting emergency visits and hospital admissions during radiation and chemoradiation: an internally validated pretreatment machine learning algorithm. JCO Clin Cancer Inform 2:1–11 30652595 10.1200/CCI.18.00037

23. Forrest LM McMillan DC McArdle CS Angerson WJ Dunlop DJ Evaluation of cumulative prognostic scores based on the systemic inflammatory response in patients with inoperable non-small-cell lung cancer Br J Cancer 2003 89 1028 1030 10.1038/sj.bjc.6601242 12966420 PMC2376960 Forrest LM, McMillan DC, McArdle CS, Angerson WJ, Dunlop DJ (2003) Evaluation of cumulative prognostic scores based on the systemic inflammatory response in patients with inoperable non-small-cell lung cancer. Br J Cancer 89:1028–1030 12966420 10.1038/sj.bjc.6601242 PMC2376960

24. Pantano Nde P Paiva BS Hui D Paiva CE Validation of the modified glasgow prognostic score in advanced cancer patients receiving palliative care J Pain Symptom Manage 2016 51 270 277 10.1016/j.jpainsymman.2015.09.010 26598040 Pantano Nde P, Paiva BS, Hui D, Paiva CE (2016) Validation of the modified glasgow prognostic score in advanced cancer patients receiving palliative care. J Pain Symptom Manage 51:270–277 26598040 10.1016/j.jpainsymman.2015.09.010

25. Proctor MJ Morrison DS Talwar D Balmer SM O'Reilly DS Foulis AK Horgan PG McMillan DC An inflammation-based prognostic score (mGPS) predicts cancer survival independent of tumour site: a Glasgow Inflammation Outcome Study Br J Cancer 2011 104 726 734 10.1038/sj.bjc.6606087 21266974 PMC3049591 Proctor MJ, Morrison DS, Talwar D, Balmer SM, O’Reilly DS, Foulis AK, Horgan PG, McMillan DC (2011) An inflammation-based prognostic score (mGPS) predicts cancer survival independent of tumour site: a Glasgow Inflammation Outcome Study. Br J Cancer 104:726–734 21266974 10.1038/sj.bjc.6606087 PMC3049591