Risk Factors for Emergency Room Visits Among Patients With Head and Neck Cancer: A Longitudinal Cohort Study Within the Korean Healthcare System
Article information
Abstract
Objectives.
A substantial proportion of patients with head and neck cancer (HNC) require emergency room (ER) visits or unplanned hospitalizations during or after treatment with various modalities. We investigated HNC cases that necessitated ER visitation after cancer treatment, aiming to identify potential risk factors in the context of the Korean healthcare system.
Methods.
This single-center cohort study examined patients with HNC who received cancer treatments at Samsung Medical Center in 2019 (n=566). Treatment modalities included surgery alone (n=184), surgery and adjuvant therapy (n=138), curative non-surgical treatment such as radiation or chemoradiation (n=209), and palliative treatments (n=35). We followed these cases for up to 3 years, focusing on those who visited the ER during or after cancer treatment, and analyzed the primary reasons and risk factors associated with these visits.
Results.
The ER visitation rate was 8.0% (n=45) among patients with HNC, with a total of 70 ER visits (12.4%; mean, 1.56; range, 1–4). The rate of treatment-related ER visitation was 4.6%. Common reasons for ER visits included surgical site or wound complications (31.1% of patients visiting the ER, 22.9% of ER visits) and issues with oral intake or feeding (22.2% of patients, 31.4% of visits). Significant risk factors for ER visits included tumor subsite (with hypopharyngeal cancer associated with a 17.9% rate of treatment-related ER visits), tumor stage (T2–4, 8.6%–12.2%; N+ status, 6.7%), and treatment modality (surgery with adjuvant chemoradiation, 19.4%). Patient age and comorbidities did not represent significant factors.
Conclusion.
The most frequent reasons for ER visits among patients with HNC included complications with wounds and feeding. Additionally, tumor characteristics and treatment modality were independent risk factors for ER visits. Adequate planning and management to address these issues could potentially decrease the number of ER visits, lower costs, and improve patient care.
INTRODUCTION
Patients with head and neck cancer (HNC) are often elderly, have multiple chronic conditions, and experience high rates of re-hospitalization and emergency room (ER) visits [1-3]. Approximately 10%–15% of patients with HNC present to the ER with an advanced tumor stage, which is associated with comparatively poor survival outcomes [4]. ER visits by individuals with HNC who have undergone surgery contribute to increased healthcare costs and inferior patient outcomes [5]. Moreover, the morbidity and mortality rates for patients with HNC in the ER are notably high [6]. Specifically, in the postoperative period, these patients exhibit significant rates of readmission, unexpected hospitalizations, and ER visits, ranging from 0.50 to 0.86 per 100 patient days [2]. Beyond ER visits, unplanned readmission rates within 30 days following HNC surgery range from 5.1% to 26.5% [3], displaying associations with old age [7-9] and the presence of comorbidities [10]. The readmission rate for patients with HNC exceeds that of patients with all cancer types [11].
The most common reasons for ER visits among individuals with HNC are symptom management (particularly pain), respiratory distress, and gastrointestinal issues related to feeding [12,13]. Disease burden and treatment modality (surgery versus chemoradiation) represent risk factors for ER visits due to respiratory complications in patients with HNC [14]. Additionally, low socioeconomic status has been identified as a risk factor for ER visits or readmissions among patients with cancer [7,8,15]. Other medical conditions, such as weight loss, hypertension, and depression, are also associated with a higher likelihood of ER visits in the HNC population [14]. Visits to the ER during or after treatment for HNC can disrupt ongoing therapies, acutely diminish quality of life, and lead to higher medical costs. However, most such visits are potentially avoidable [16].
Notably, Korea’s national health insurance system provides coverage for nearly all residents. This sets it apart from other medical systems, such as those in the United States or Europe, for which data on ER visit incidence are available. Consequently, socioeconomic status may not pose a barrier to ER attendance for patients with HNC in Korea. Given this context, it is necessary to evaluate the incidence and risk factors of ER visits among these patients to better assess the medical and social burdens associated with ER visits by individuals with HNC under the Korean national health insurance system.
An extensive literature search yielded only three articles that explored ER visits among patients with cancer in Korea. One study indicated that 6.8% of these visits were made by patients with true emergencies, while urgent cases accounted for 30.5%. Notably, 62.7% of the ER visits among 2,699 patients with cancer at Korean ERs were by individuals not facing emergencies (National Cancer Center, Korea) [17]. Another study, utilizing the National Emergency Department Information System database from 2017 to 2019, determined that visits by patients with cancer constituted only 5.5% of all ER visits in Korea [18]. The authors noted an interesting trend: longer ER stays were associated with a higher proportion of hospital admissions for cancer patients, implying that these patients often visit the ER for admission purposes in addition to seeking acute care. This trend may reflect the distinctive aspects of Korea’s medical system [18]. In one study of a Korean hospital, the most common chief concern among ER attendees with cancer was pain, followed by gastrointestinal symptoms, respiratory symptoms, high fever, and weakness [19]. However, no prior studies have specifically examined ER visits by Korean patients with HNC.
Therefore, we conducted a cohort study to investigate the incidence of and risk factors for ER visits among Korean patients with HNC. Our findings may help elucidate the demand for ER services, inform resource distribution, and identify necessary measures to improve HNC care in Korea.
MATERIALS AND METHODS
The study protocol was approved by the Institutional Review Board of Samsung Medical Center before data collection (No. 2024-07-135) and was conducted under the guidelines established by the Declaration of Helsinki. Because there was only minimal risk to subjects in this study, the requirement for written informed consent was waived, and all patient data were deidentified for the analyses.
Study subjects
This was a single-center cohort study of HNC patients treated at Samsung Medical Center during 2019. All patients had been newly diagnosed with pathologically confirmed HNC (squamous cell carcinomas and others) and had undergone cancer treatment for their diseases at our center. Patients who had previous HNC and underwent treatments before 2019 were excluded from the study. The total number of HNC patients was 616; however, 50 patients were excluded due to loss to follow-up, incomplete medical information, and/or additional treatments and followup at other hospitals, leaving 566 HNC patients in our analysis (Fig. 1).
Study diagram. SMC, Samsung Medical Center; ER, emergency room.
Standard of care
Alongside the initial diagnostic workups, optimal treatments were determined by multi-disciplinary discussions, based on the current guideline (National Comprehensive Cancer Network, 2024, v4). The TNM status of each tumor was classified according to the 8th Edition of the American Joint Committee on Cancer (AJCC) Staging Manual. Treatments with curative intent were applied for 531 patients: surgery alone (n=184), surgery plus adjuvant therapy (n=138), and curative non-surgical treatment (radiation or chemo-radiation) (n=209). Initial palliative treatments were given to 35 patients with synchronous systemic metastasis. Upfront surgery and radiation were applied to primary sites and neck nodes depending on the risk of regional metastasis. Definitive concurrent chemo-radiation (three cycles of cisplatin-based regimen) was applied to cases with advanced T or N status when surgery might cause significant surgical morbidity and functional loss.
Based on surgical pathology information after surgery, patients had received adjuvant radiation. The adjuvant radiation was delivered by three-dimensional conformal radiation therapy, or intensity-modulated radiation therapy with 54 to 66 Gy. In cases with high risk-adverse features (presence of cancer cells at resection margin and extra-nodal extension of lymph node metastasis) after surgery, adjuvant concurrent chemo-radiation was the main adjuvant treatment modality. Palliative treatments consisted of various chemotherapeutic regimens, targeted agents and immune checkpoint inhibitors, or combined.
Outcome measurement and analysis
We followed up all cases after (during) cancer treatments for up to 3 years to identify ER visit rate and reviewed the main reasons for ER visits. The main causes of ER visits were classified into two categories: treatment-related and -unrelated (or indirectly related). Clear discrimination of these two reasons was somewhat difficult in some cases. We meticulously reviewed the ER medical records, the chief complaints for ER visit, and the area affected by symptoms. Then, we determined the main causes of ER visits as either treatment-related or -unrelated.
Statistical analysis
The events of interest were treatment-related ER visits and total ER visits. Clinical and pathological characteristics were compared between treatment-related ER visit versus others and total ER visit versus no ER visit, using chi-square test or Fisher’s exact test, two-sample t-test, or Wilcoxon rank sum test, as appropriate. The variables were patient variables (age, sex, smoking and alcohol status, and Eastern Cooperative Oncology Group Performance Status [ECOG PS] for non-surgical patients and American Society of Anesthesiologists Physical Status [ASA PS] Classification for surgical patients), tumor factors (tumor subsites, T, N, M status, and tumor pathology), and treatment modalities. Univariable and multivariable logistic regression analyses were conducted for treatment-related ER visits and total ER visits. The multivariable models were constructed with variables that were significant in the univariable analyses (P<0.05) or backward stepwise selection. All statistical analyses were performed using R software version 4.3. (R Foundation for Statistical Computing). A two-tailed P-value <0.05 was considered statistically significant.
RESULTS
Baseline characteristics of the HNC cohort
Regarding the baseline characteristics of patients with HNC, the most common subsites were the oral cavity (23.5%) and the nasopharynx (16.3%). These were followed by the oropharynx (14.1%) and the paranasal sinus and nasal cavity (11.3%) (Table 1). The T, N, and M statuses at diagnosis are also detailed in Table 1. Disease categorized as cT1–2 and cN0 (TNM stage I–II) was present in 48.3% of the patients. As anticipated, squamous cell carcinoma comprised 70% of the tumor pathology. Regarding treatment modality, 32.5% of patients underwent surgery alone with curative intent. Another 24.4% received surgery and adjuvant therapy, which included radiation or concurrent chemoradiation. Meanwhile, 36.9% of patients were treated with nonsurgical curative methods, specifically definitive radiation or concurrent chemoradiation.
Baseline characteristics of the 1-year (2019) cohort of patients with HNC in this study (n=566)
Frequency and causes of ER visits
Among 566 patients with HNC in 2019, 8.0% (n=45) visited the ER in general, while 4.6% (n=26) experienced treatment-related ER visits (Table 2). The 45 patients had a total of 70 ER visits, with a mean of 1.56 visits per patient (range, 1–4). Ten patients had multiple treatment-related ER visits, while four had multiple treatment-unrelated ER visits. In most cases, the reasons for multiple ER visits were consistent, except for two patients with treatment-related visits (Supplementary Fig. 1). Each of these two patients made four ER visits: one for surgical site bleeding (twice) and jejunostomy tube issues (twice) and the other for complications related to concurrent chemoradiation (twice) and pleural effusion (twice). All ER visits occurred during the post-treatment period following procedures with curative intent, which included surgery, radiation, or concurrent chemoradiation. No ER visits occurred during treatment, apart from palliative chemotherapy.
Causes of ER visits (n=45)
Surgical site or wound issues were the most common reasons for ER visits, accounting for 31.1% of patients (n=14; 22.9% of ER visits), followed by problems with oral intake or feeding (22.2% of patients [n=10]; 31.4% of ER visits). Notably, only one patient visited the ER with airway problems or dyspnea as the main cause. Treatment-unrelated issues involving body areas other than the head, neck, or donor site were varied and presented in 19 patients (3.3% of the 566 patients with HNC). Most ER visits occurred within the first 3 months post-treatment, after which the frequency of visits plateaued (Fig. 2).
Timing of emergency room (ER) visits in patients with head and neck cancer.
Risk factor analysis for treatment-related ER visits in patients with HNC
To identify potential risk factors for treatment-related ER visits in patients with HNC, we compared medical information between patients who experienced such visits and those who did not (Table 3). No significant differences were present between the two groups regarding patient variables, including age, sex distribution, and performance status. However, patients with primary cancer originating in the nasopharynx and oropharynx experienced fewer ER visits (P<0.001 and P=0.014, respectively), while more frequent ER visits were observed in patients with hypopharyngeal cancer (P=0.027). Regarding treatment method, surgery with adjuvant therapies was associated with a higher frequency of ER visits (P=0.014), whereas non-surgical treatment was associated with fewer ER visits (P<0.001).
Comparisons of clinical variables for treatment-related ER visits
Subsequent multivariable analysis confirmed that a primary tumor originating from the hypopharynx or larynx, T2–4 and N+ status, and surgery-based treatment were independent risk factors for treatment-related ER visits among patients with HNC in Korea (Table 4). In summary, significant risk factors included tumor subsite (with hypopharyngeal cancer associated with a 17.9% rate of treatment-related ER visits), tumor stage (T2–4, 8.6%–12.2%; N+ status, 6.7%), and treatment modality (surgery with adjuvant chemoradiation, 19.4%). Notably, patient age and comorbidities were not significant risk factors.
Multivariable analysis for treatment-related ER visits
Risk factors for overall ER visits among patients with HNC
Even ER visits unrelated to treatment may be indirectly associated with HNC or related interventions. Consequently, we conducted a risk factor analysis for all ER visits, encompassing both treatment-related and unrelated visits. Similar to the treatment-related results, our findings indicated that patients with nasopharyngeal and oropharyngeal cancers experienced fewer total ER visits (P<0.001 and P=0.003, respectively), while those with hypopharyngeal cancer had more frequent visits (P=0.049) (Supplementary Table 1). Additionally, patients presenting with advanced T status were more likely to visit the ER (P=0.038). Consistent with previous findings, surgery with adjuvant treatments was linked to a higher frequency of total ER visits (P=0.015), whereas non-surgical treatments were associated with fewer visits (P<0.001). Multivariable analysis revealed that hypopharyngeal or laryngeal cancer, T2–4 status, N+ status, and surgery-based treatment were independent risk factors for total ER visits (Supplementary Table 2).
DISCUSSION
This longitudinal cohort study included all patients diagnosed with HNC and treated at our hospital in 2019. We monitored these patients to understand the frequency and reasons for their ER visits over up to 3 years. Our findings indicate that after treatment, 4.6% of patients with HNC visited the ER due to treatment-related complications. Unplanned ER visits can severely undermine quality of life for patients and lead to unnecessary healthcare expenditures [20]. Previous research has documented rates of 0.50 to 0.86 ER visits or unplanned hospital admissions per 100 patient-days within 30 days of undergoing curative treatment for HNC [2], and approximately 10% of patients who undergo HNC and reconstructive surgery are readmitted within 30 days [3]. Rates of ER visits or admissions have been found to encompass one-third of patients with HNC receiving radiation therapy [21] or outpatient chemotherapy [22]. In the present study, we included all patients newly diagnosed with HNC in 2019 who received any form of treatment, including surgery, radiation, or chemotherapy. Building on previous research, our overview of ER visits among patients with HNC may contribute to the development of effective preventive measures aimed at reducing ER visits, serving as an indicator of the quality and cost-effectiveness of care [23].
The features of the health insurance system and the economic burden on patients with cancer substantially influence ER visits and treatment outcomes [15,24-26]. Korea’s national health insurance system is unique, covering all citizens, and patients with HNC receiving medical care face relatively low economic burdens. For context, ER visits in the United States cost on average between 1,200 to 1,300 U.S. dollars, whereas in Korea, patients with national health insurance paid approximately 50 to 100 U.S. dollars for ER visits in 2022 (www.afcurgentcare.com, www.shhosp.co.kr). Additionally, patients with HNC undergoing treatment or post-treatment care at the same hospital have unrestricted ER access in terms of distance and policy. When comparing the rates of ER visits between patients covered by national medical insurance (n=550, 97.2%) and those eligible for Medical Aid (n=16, 2.8%), no significant difference in ER visit rates was observed (8.6% vs. 6.3%, respectively). Consequently, our analysis focused on patient, tumor, and treatment factors, rather than socioeconomic status. Our main findings indicate that certain tumor subsites, advanced T and N status (≥T2 or N1), and multimodal treatment that includes surgery were risk factors for ER visits among Korean patients undergoing treatment for HNC. Most ER visits occurred within 3 months after treatment completion, suggesting that treatment-related issues are the primary reasons for ER visits in this patient population.
Tumor burden, as indicated by tumor stage, is a well-established risk factor for ER visitation [1,2,27]. Treatment volume (including the extent of surgery or radiation, as well as the treatment modality applied) tends to increase proportionally to the tumor stage. Consequently, the likelihood of treatment-related ER attendance depends on the therapeutic intensity. Among treatment modalities, surgical intervention was identified as a significant risk factor for ER visits in patients with HNC. Contrary to our findings, earlier studies have posited that curative concurrent chemoradiation was the primary driver of ER visits in patients with HNC [2,26]. However, surgical site infections and wound problems have been frequently cited as main reasons for ER visits in other research [3,27-30]. Our post-hoc analysis revealed that patients with HNC who underwent both surgery and postoperative concurrent chemoradiation experienced the highest frequency of ER visits, aligning with a prior study [2]. Therefore, clinicians should carefully monitor post-treatment complications in high-risk HNC cases involving advanced tumors and multiple forms of treatment, with the aim of minimizing unplanned ER visits.
HNC can occur in various anatomical locations. Among these, the hypopharynx and larynx have been reported as high-risk areas for ER visits following treatment [2,31], a finding supported by our results. Compared to other subsites of the head and neck, the hypopharynx and larynx are critical for respiration and swallowing. Surgical treatments for HNCs in these areas are typically complex, often necessitating reconstructive surgery, and are associated with higher rates of postoperative complications [3,32]. The second most common reason for treatment-related ER attendance was issues with oral intake or tube feeding. Gastrointestinal problems are among the most frequent reasons for ER visits in patients who have undergone general surgery [30] and those with cancer [12], and medications for gastrointestinal symptoms are the most commonly prescribed category of drug in the ER [13]. Nutrition support is crucial during and after HNC interventions, and alternative feeding methods are often used for patients undergoing treatment for HNC. However, patients may struggle to adapt to these new feeding methods after being discharged from the hospital, leading to feeding tube malfunctions or complications related to alternative feeding. To mitigate this issue, strengthening patient education regarding oral intake and feeding during HNC treatment or before hospital discharge could reduce the number of unplanned ER visits among patients with HNC.
One specific finding from our study was that patient comorbidities did not represent significant risk factors for ER visits in patients with HNC. This contrasts with many previous studies that have identified comorbidity as a key factor contributing to ER visits in cancer cases [1,2,8,27,33]. This discrepancy may be attributable to the distinctive health insurance system in Korea. Korean patients with HNC can be well-managed and monitored within the healthcare system. They may benefit from close follow-up care, which allows for regular medical check-ups when symptoms change, potentially reducing the need for ER visits. Another factor is the possibility of incomplete data on comorbidities. We were only able to obtain information on ECOG PS for 56.6% of non-surgical patients and ASA PS for 86.6% of surgical patients. Consequently, the impact of comorbidities may be underrepresented in our analysis, highlighting the need for a future validation study. Similarly, age did not emerge as a significant risk factor for ER visits among patients with HNC in our study. In Korea, being 65 years old or older is a criterion for receiving long-term healthcare, particularly for those with chronic illnesses. Patient support provided by care workers is partially reimbursed by the Korean system, especially for aged patients with HNC and chronic conditions. This may mitigate the influence of age and comorbidity on the frequency of ER visits within this population. However, the real effect sizes of these variables remain unclear in our research.
This study represents the first 1-year (2019) longitudinal cohort analysis of patients with HNC who visited the ER in Korea. However, the study has several limitations. The data were derived from a single hospital in Korea over a single year, necessitating caution when attempting to generalize the findings. Admission and discharge criteria vary across institutions and could influence the frequency of ER visits following hospital discharge or during HNC treatment. Nevertheless, trend plots from 2008 to 2019 indicate that ER visit rates for oral/pharyngeal cancer have remained relatively stable (24.2%±3.0%, without much fluctuation) each year, except for 2020, which was impacted by the outbreak of coronavirus disease 2019 (Supplementary Fig. 2). Consequently, our 1-year data may still reasonably reflect overall ER visit patterns among patients with HNC, regardless of the study year. A second limitation is the relatively small number of patients enrolled in the study, which precluded a separate analysis of each cause of ER visitation. The sample size was insufficient for conducting multivariable analysis to investigate potential risk factors for each cause while adjusting for confounders. Our findings would benefit from validation in a larger cohort, potentially utilizing nationwide data. One of the most critical weaknesses of the study was the issue of missing data, as mentioned previously. The retrospective nature of the analysis meant that the raw data were incomplete, lacking detailed information on various comorbidities and medical conditions. Additionally, potential confounding factors that went unexamined in this study, such as family or social support for patient care, patient mental status, and patient occupation or religion, could have influenced the results. These limitations should be addressed in future research using a large prospective cohort.
Nevertheless, our study reflects real-world practice regarding HNC treatments in Korea and identified a specific frequency (8.0%) of ER visits among patients undergoing treatment for HNC. This information can inform the allocation of medical resources. Our findings suggest practical measures to minimize unplanned and costly ER visits for these patients [34]. Establishing a direct line of communication for patients after hospital discharge and arranging short-term follow-up for those who have undergone HNC surgery could help prevent unnecessary ER visits for minor issues. Furthermore, wound care can be improved, along with patient education on self-management of wounds. We observed that ER visits by patients with HNC were most frequent immediately following hospital discharge, as well as after the first outpatient clinic appointment for treatment-related ER visits (Supplementary Fig. 3). Consequently, diligent monitoring of the treatment site during outpatient visits is crucial until the wound has stabilized or matured, particularly in patients at high risk. As many ER visits are precipitated by complications with oral intake or tube feeding, thorough training and regular check-ups to address feeding problems should help reduce the number of unplanned ER visits in patients with HNC.
In conclusion, our study revealed that wound issues and feeding problems were the most common reasons for ER visits among patients with HNC. Tumor characteristics, including subsite and stage, and treatment modality were identified as risk factors for ER visits in the Korean HNC cohort. Proper planning and management of these issues could reduce the frequency of ER visits, lower healthcare costs, and improve patient care.
HIGHLIGHTS
▪ The rate of emergency room (ER) visits and associated risk factors were assessed in a longitudinal cohort of patients with head and neck cancer (HNC).
▪ Among patients with HNC, 8.0% experienced ER visits, with 4.6% doing so for treatment-related issues.
▪ The most frequent reasons for ER visits in this population were issues related to wounds and feeding.
▪ Tumor factors and treatment modalities were identified as independent risk factors for ER visits.
▪ Adequate planning and management of these problems could help reduce ER visits and associated costs.
Notes
Han-Sin Jeong is an editorial board member of the journal but was not involved in the peer reviewer selection, evaluation, or decision process of this article. No other potential conflicts of interest relevant to this article were reported.
ACKNOWLEDGMENTS
This work was supported by a grant from Samsung Medical Center (No. OTX0001031).
AUTHOR CONTRIBUTIONS
Conceptualization: HSJ. Methodology: HSJ. Formal analysis: HY, NC. Investigation: HK, YK, YJS, JHP. Resources: HK, YK, YJS, JHP. Data curation: HY, HK, YK, YJS, JHP, NC. Visualization: NC. Supervision: HSJ. Funding acquisition: HSJ. Writing–original draft: HY, HK, HSJ. Writing–review & editing: YK, NC, HSJ. All authors read and agreed to the published version of the manuscript.
SUPPLEMENTARY MATERIALS
Supplementary materials can be found online at https://doi.org/10.21053/ceo.2024.00257.
Comparison of clinical variables for total ER visits (treatment-related or unrelated)
Multivariable analysis for total ER visits (treatment-related or unrelated)
Causes and timing of multiple emergency room (ER) visits. In our series, instances of multiple ER visits occurred in 10 patients with treatment-related ER visits and four patients with treatment-unrelated visits. Most of these multiple ER visits were due to the same causes at each visit, apart from two patients in the treatment-related category. These two patients, identified as no. 8 and no. 9, each visited the ER four times. Patient no. 8 presented with surgical site bleeding on two occasions and experienced issues with a jejunostomy tube for tube feeding on two other occasions. Patient no. 9 visited the ER twice due to complications related to concurrent chemoradiation and twice due to pleural effusion.
Trend plot of emergency room (ER) visit rate in patients with oral and pharyngeal cancer from 2008 to 2020 [1]. In this plot, the annual rate of ER visits for patients with oral/pharyngeal cancer following a diagnosis of head and neck cancer (HNC) ranged from 19.5% to 29.5%. This rate reflects the number of ER visits rather than the rate of individual patients. Consequently, the rate of ER visits was higher than that observed in our research, which was 7.6% for oral/pharyngeal cancer. In our study, the rate represented the number of patients who visited the ER during or after HNC treatment.
Number of outpatient clinic visits prior to emergency room (ER) visit. The number 0 indicates that patients visited the ER immediately following hospital discharge and prior to their first outpatient clinic appointment. Overall, patients attended a mean of 2.60 outpatient clinic visits before presenting to the ER (standard error [SE], 2.65 visits). For treatment-related ER visits, the mean number of outpatient clinic visits preceding the ER visit was 2.65 (SE, 2.67). For ER visits unrelated to treatment, the average was 2.52 (SE, 2.70). This difference was not statistically significant (P=0.921). OPD, outpatient department.