- Email: [email protected]
Epistaxis health disparities in the United States pediatric population
International Journal of Pediatric Otorhinolaryngology 114 (2018) 20–25
Contents lists available at ScienceDirect
International Journal of Pediatric Otorhinolaryngology journal homepage: www.elsevier.com/locate/ijporl
Epistaxis health disparities in the United States pediatric population a
b
b,c
Lindsay Yang , Kevin Hur , Jeffrey Koempel , Elisabeth H. Ference a b c
T
b,c,∗
Keck School of Medicine of the University of Southern California, Los Angeles, United States Rick and Tina Caruso Department of Otolaryngology-Head and Neck Surgery, Keck School of Medicine of the University of Southern California, Los Angeles, United States Division of Otolaryngology-Head and Neck Surgery, Department of Surgery, Children's Hospital Los Angeles, Los Angeles, United States
A R T I C LE I N FO
A B S T R A C T
Meeting: Poster Presentation at the American Rhinologic Society, Combined Otolaryngology Spring Meeting, on April 19–20, 2018 in National Harbor, Maryland
Objective: Despite epistaxis occurring in up to 60% of the population, few studies have investigated health status disparities in the pediatric epistaxis population. The aim of this study was to evaluate sociodemographic risk factors associated with epistaxis visits for pediatric patients. Methods: Data were extracted from the National Ambulatory Medical Care Survey and National Hospital Ambulatory Medical Care Survey Outpatient Department from 2001 to 2010. Outpatient visits of children less than 18 years who received a primary, secondary, or tertiary diagnosis of epistaxis (ICD-9CM code 784.7X) were included. Bivariate and stepwise multivariate regressions were conducted to develop a final model for epistaxis visits described by sociodemographics. Results: Epistaxis visits accounted for 5 ± 0.6 million visits in children less than 18 years. 51% and 33% of children presenting with epistaxis had private insurance and Medicaid, respectively (p = 0.001). 69% of epistaxis visits were evaluated at a pediatric clinic, 18% at an ENT/surgery clinic, and 13% at a general/family medicine clinic (p < 0.0001). After multivariate adjustment, epistaxis visits were associated with older age (p = 0.006). Black children were more likely to present with epistaxis (95% CI 1.3–4.1, p = 0.005) compared to white children. Allergic rhinitis, present in 11% of epistaxis visits, was a significant comorbidity associated with visits (95%CI 1.3–4.6, p = 0.008). Patients were also more likely to present to an ENT/surgery clinic (95% CI 4.5–16.5, p < 0.0001) compared to a general/family medicine clinic. Conclusions: Epistaxis visits by children are associated with age, race, and specialty. Targeted interventions to help reduce this common presentation should be developed.
Keywords: Epistaxis Pediatric Otolaryngology Public health Rhinology Manuscript
1. Introduction Epistaxis is a very common condition among children. Though it is rare in children younger than two years old [1], at least one episode of epistaxis has occurred in 30% of children by the age of five and in over 50% of children greater than five years old [2]. Most cases in children and young adults are self-limited and due to dry nasal mucosa; some are anterior, originating from Little's area in the anterior portion of the nasal septum where Kiesselbach's plexus forms. However, if more severe, patients may present to an emergency care provider, primary care provider, or otolaryngologist. Children reported high stress scores for bleeds longer than the average time of 5–10 min, and caregivers' fear of excessive blood loss was a significant factor causing high distress [3]. Despite the frequency of epistaxis, few studies have investigated health status disparities in the pediatric population. While healthcare
disparities focus on the differences in access to or availability of medical facilities and services, health status disparities refer to the variation in rates of disease occurrence and disabilities between socioeconomic and/or geographically defined population groups [4,5]. Some studies have examined health status disparities in presentation to the emergency department (ED) of certain states [3,6–9], but none have captured other sources of patient care or provided a national pediatric outlook. The aim of this study was to evaluate sociodemographic risk factors associated with epistaxis visits for pediatric patients in a nationally representative sample in the outpatient setting. 2. Methods 2.1. Data sources Data were extracted from the National Ambulatory Medical Care
∗ Corresponding author. Rick and Tina Caruso Department of Otolaryngology-Head and Neck Surgery, Keck School of Medicine of the University of Southern California, 1540 Alcazar Street, Suite 204M, Los Angeles, CA, 90033 United States. E-mail addresses: [email protected] (L. Yang), [email protected] (K. Hur), [email protected] (J. Koempel), [email protected] (E.H. Ference).
https://doi.org/10.1016/j.ijporl.2018.08.025 Received 2 July 2018; Received in revised form 20 August 2018; Accepted 20 August 2018 Available online 23 August 2018 0165-5876/ © 2018 Elsevier B.V. All rights reserved.
International Journal of Pediatric Otorhinolaryngology 114 (2018) 20–25
L. Yang et al.
identified on the survey; unknown category includes those surveys listed as unknown or uncompleted. Due to missing data for median household income, payment source was used as an alternative. Provider demographics included practice region (Northeast, Midwest, South, West); practice location (metropolitan statistical area, non-metropolitan statistical area); practice type (general/family medicine, pediatrics, ENT/surgery); new patient to practice; and number of visits in the last year (no visits, 1–5 visits, ≥6 visits). A stepwise multivariate logistic regression was conducted to assess the percentage of epistaxis visits with a refined list of covariates. Variables statistically significant at the 5% level, that modified the association between provider and outcome by more than 10% (when included in the model), or were clinically meaningful were retained in the final model. Epistaxis visits were also examined for their association with various comorbid conditions, including history of chronic disease, history of asthma, diagnosis of allergic rhinitis (ICD-9CM code 477.9X), diagnosis of anemia, and diagnosis of anemic etiologies. Anemic etiologies included iron deficiency anemias (ICD-9CM codes 280.8X, 280.9X), megaloblastic or nutritional deficiency anemias (ICD-9CM codes 281. XX), hereditary anemias (ICD-9CM code 282. XX), sickle cell anemias (ICD-9CM code 283. XX), aplastic anemias (ICD-9CM code 284. XX) and other chronic disease anemias (285. XX). Lastly, procedures performed at epistaxis visits were analyzed to evaluate their relationship to race (ICD-9CM codes 210. XX for epistaxis control, 212. XX for diagnostic procedure on nose, 22X.XX for operations on nasal sinuses. Statistical analyses were performed using the SURVEYFREQ, SURVEYREG, SURVEYLOGISTIC, and GENMOD procedures in SAS statistical software (version 9.4, SAS Institute Inc, Cary, North Carolina, USA).
Survey (NAMCS) and National Hospital Ambulatory Medical Care Survey (NHAMCS) Outpatient Department (OPD) from 2001 to 2010. The Centers for Disease Control and Prevention (CDC) and National Center for Health Sciences collect the NAMCS and NHAMCS annually to measure the utilization and provision of ambulatory care services, which is the predominant method of providing health care in the United States [10,11]. The NAMCS uses a 2-stage probability design that samples physicians who represent geographical regions and patient visits within these practices. Data were collected over a 1-week period randomized to the physician. The NHAMCS is intended to offer more complete ambulatory data than previously covered by the NAMCS. The NHAMCS uses a 4-stage probability design that samples approximately 500 nationally representative primary sampling units (PSUs) of the 50 states and the District of Columbia, hospitals within the PSUs, clinics within the hospitals, and patient visits within the clinics. Data was collected over a 4-week period randomized by facility and included patient, visit, and provider characteristics. Both the NAMCS and NHAMCS are weighted to represent national usage, with visits adjusted for the provider specialty, visits to the provider, nonresponse, and smoothing. Both surveys have undergone changes since their inception. Within the 2001–2010 window, phrasing of questions and the availability of variables have changed. History of chronic disease and urban-rural classification of a patient's zip code were introduced in 2005 [10]. Practice specialty type is missing in the 2005 data tables. The survey uses the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9CM) codes for recording diagnoses. Providers surveyed can complete up to 3 diagnoses. 2.2. Inclusion criteria Outpatient visits of children less than 18 years who received a primary, secondary, or tertiary diagnosis of epistaxis (ICD-9CM code 784.7X) were included. Patients whose diagnosis was missing in all 3 instances were excluded. The primary outcome measure was the weighted proportion of epistaxis visits. Sociodemographic variables were collected in addition to the type of clinic visited and type of provider seen. Clinic types were categorized as general/family medicine, pediatrics clinic, or ENT/surgery clinic. Provider types included a physician, nurse practitioner, or physician assistant. Results indicate variables that were not available during all years and the first available year.
3. Results Epistaxis visits accounted for 5.0 ± 0.6 million visits by children less than 18 years from 2001 to 2010. The ages of patients with a diagnosis of epistaxis were similar across the 3 divisions 0–5, 6–11, and 12–18 years old (36%, 31%, and 33%, respectively), though this was significantly different from the age distribution across non-epistaxis visits, which had a greater predominance toward 0–5 year olds (49%) (p = 0.022) (Table 1). Males (61%) accounted for more epistaxis visits than females, but a similar pattern was true across visits overall (Table 1). The race of patients by frequency of epistaxis visits were Whites (37%), Blacks (17%), Other/unknown (24%), and Hispanics (22%) (Table 1). Blacks accounted for a significantly greater share of epistaxis visits than they do for non-epistaxis visits (11%) (p = 0.005) (Table 1). Payment source distribution was also significantly different between epistaxis and non-epistaxis visits, with a decreased share of private insurance visits (51% and 60%, respectively) (p = 0.001) (Table 1). A diagnosis of allergic rhinitis (11%) was also significantly associated with epistaxis visits (p < 0.001) (Table 1). There were no significant differences in other patient demographics, including history of chronic disease, history of asthma, diagnosis of anemia, or diagnosis of anemic etiologies between children with epistaxis versus those who did not have epistaxis. Practice location is predominantly Metropolitan across all visits (88%), but even more so for epistaxis visits (95%) (p = 0.005). There was a greater percentage of epistaxis visits seen at an ENT/surgery clinic (18%) rather than general/family medicine clinic (13%), compared to non-epistaxis visits, but the majority of visits were still seen with pediatrics providers (69%) (p < 0.001) (Table 1). Practice region, first visit, and visit frequency were not significantly different based on bivariate comparison. The final multivariate model included covariates age, race, payment source, history of chronic disease, diagnosis of allergic rhinitis, region, practice location, and practice type. After multivariate adjustment,
2.3. Statistical analysis Bivariate and stepwise multivariate regressions were conducted to develop a final model for epistaxis visits described by sociodemographics. The number of epistaxis visits were first weighted based on the surveys' PSU, strata, and visit weights. The weighted proportion of epistaxis visits were compared to patient and provider characteristics using Chi-square tests. Patient demographic included age (0–5, 6–11, 12–18 years old). These categories were selected to best capture an even distribution of frequency (1,795,057 weighted visits, 1,545,384 weighted visits, 1,656,646 weighted visits, respectively); moreover, these age ranges carry clinical significance in early childhood, middle childhood, and adolescence. Other demographic information included sex (female, male); race (White, Hispanic, Black, and other/unknown); payment source (private insurance, Medicare/Medicaid, other, unknown); history of chronic disease; history of asthma; diagnosis of anemia; or diagnosis of various anemic etiologies. The category of other/unknown race includes all patients who do not fall under White non-Hispanic, Black non-Hispanic, and Hispanic categories; patients may have identified as Asian, Native Hawaiian/Other Pacific Islander, American, or Indian/Alaska Native, or may have reported more than one race. Under payment source, other includes worker's compensation, self-pay, no charge/charity, or other 21
International Journal of Pediatric Otorhinolaryngology 114 (2018) 20–25
L. Yang et al.
4. Discussion
Table 1 Characteristics of visits by pediatric patients with epistaxis, based on weighted visits.
Unweighted visits Estimated weighted visits Patient demographics, % Age 0–5 years 6–11 years 12–18 years Gender, male Race White Hispanic Black Other/unknown Payment Source Private Insurance Medicare/Medicaid Other Unknown History of chronic disease History of asthma Diagnosis of allergic rhinitis Diagnosis of anemia Diagnosis of anemia causes Provider demographics, % Region Northeast Midwest South West Practice location Non-MSA Metropolitan Practice type General/family medicine Pediatrics ENT/surgery New patient to practice Past visits to practice in 12 mos No visits 1–5 visits ≥6 visits
Epistaxis (SE)
Non-Epistaxis (SE)
274 4,997,087 (100)
107,983 1,812,827,680 (100)
35.9 30.9 33.2 60.9
(5.2) (5.0) (5.4) (5.2)
49.2 20.4 30.4 51.9
(0.5) (0.3) (0.4) (0.3)
37.2 22.1 16.8 24.0
(5.4) (5.6) (3.7) (4.6)
56.5 15.9 10.6 17.0
(1.3) (1.3) (0.7) (1.0)
This is the first national study to report that pediatric epistaxis visits are associated with age, race, practice location, and practitioner type. First, our study finds that epistaxis visits to providers are more likely to occur with older ages, 6–11 year olds in relation to 0–5 year olds and slightly more so with 12–18 year olds. The average age of children presenting with epistaxis in our study was 8.32 years, which is generally consistent with the existing literature. Shay et al., 2017, who studied pediatric visits to EDs for four states, reported a mean age of 7.54 [6]; Pallin et al., 2005, who reviewed NAMCS and NHAMCS from an earlier timeframe (1992–2001) across all populations, found a bimodal distribution peaking among those younger than 10 and then again at 70–79 years old [7]; and Patel et al., 2014 had an average age of 9.1 years with 175 patients [12]. Second, epistaxis visits are significantly more associated with black children than white children, with and without adjustments for other patient factors and provider demographics. Epistaxis visits are also more significantly associated with children of other/unknown ethnicity than white children, though the clinical significance is unclear given that this category was generated due to insufficient sample size when disaggregated. In general though, our findings differ from other studies of racial disparities in epistaxis. Hispanic patients represented the highest proportion of children with epistaxis in Shay et al., (2017), though the population of this study was representative of state ED databases for New York, Florida, Iowa, and California for 2010 and may not reflect the national trends for other care settings [6]. When racial disparities are expanded to the adult population, Pallin et al., 2005 noted no significant difference between ED utilization rates among black versus white epistaxis patients and Hispanic versus non-Hispanic epistaxis patients, with higher rates of ED use among blacks for all conditions [7]. Chaaban et al., 2016 did find that blacks were significantly more likely to present with epistaxis than non-Hispanic whites (1.23 fold), though the study covered only Medicare beneficiaries who represent a much older population [9]. Nevertheless, the concept that disparities exist in pediatric otolaryngology has been tested in other fields [13]. Black and Hispanic children have more revisits after a tonsillectomy with a greater likelihood of acute pain than do white children [14]. The low number of procedures performed at ambulatory visits did not allow us to assess if there was racial disparity as well in the treatment received, which Shay et al., 2017 noted in their ED study [14]. In future studies, racial disparities and the difference in findings between ambulatory and ED settings can be further explored through a longitudinal database including both ED and ambulatory visits; this could further elucidate referral patterns regarding minority populations. If there is a racial difference in referral patterns or ED visits, understanding current utilization would allow for better resource use and patient management by correcting this differential. Third, it has long since been realized that allergic rhinitis is a common cause of recurrent epistaxis [15], and this occurrence has been followed in larger studies as well [16]. This study's findings are consistent with prior literature, as a diagnosis of allergic rhinitis indicates a pediatric patient is 2.4 times more likely to present with epistaxis. Fourth, practice location epistaxis visits were more likely to present in the Northeast than the Midwest. This was consistent with literature that describes how epistaxis rates are higher in the northern United States than the south, perhaps reflecting temperature effects on the nasal mucosa, use of central heating, or upper respiratory infection frequency [9]. Patients in the Mountain and Alaska regions are 1.42 and 1.78 times more likely to present with epistaxis, respectively, whereas patients in the South Atlantic, East South Central, West South Central, and California regions are less likely to present with epistaxis, though these studies do not cover the pediatric population [9]. Fifth, ENT/surgery provided a significantly greater portion of care for epistaxis visits, compared to general/family medicine practitioners. The distribution of care providers for epistaxis visits differs significantly
P-Value
0.022
0.092 0.005
0.001 50.9 (6.0) 32.5 (5.6) 4.8 (2.0) 11.8 (4.5) 36.8 (5.8) 6.6 (4.0) 10.6 (2.9)
60.3 (1.1) 31.0 (1.1) 5.1 (0.3) 3.5 (0.3) 43.4 (1.1) 4.6 (0.2) 3.2 (0.2)
0.265 0.545 < 0.001
1.2 (1.1) 0.5 (0.4)
0.4 (0.0) 0.2 (0.0)
0.228 0.121
27.3 13.0 39.4 20.2
20.4 22.1 37.3 20.1
0.144
(5.5) (3.2) (5.9) (4.9)
(1.8) (2.4) (2.7) (2.1)
0.005 5.3 (2.0) 94.7 (2.0)
12.3 (2.6) 87.7 (2.6)
13.2 (3.5)
24.0 (1.2)
68.9 (4.5) 18.0 (2.9) 12.6 (3.0)
72.3 (1.2) 3.7 (0.2) 11.7 (0.3)
19.0 (3.7) 66.6 (4.5) 14.4 (3.7)
17.4 (0.4) 59.4 (0.5) 23.2 (0.6)
< 0.001
0.760 0.111
epistaxis visits were associated with older age, 6–11 year olds (95%CI 1.2–3.5, p = 0.009) and 12–18 year olds (95%CI 1.2–3.7, p = 0.006) (Table 2). Black children (95%CI 1.3–4.1, p = 0.005) and children of other/unknown ethnicity (95%CI 1.1–3.4, p = 0.029) were more likely to present with epistaxis compared to White children (Table 2). Hispanic children compared to white children presented with higher rates of epistaxis, however the observed difference was not significant after adjustment for cofounders. Additionally, a diagnosis of allergic rhinitis was significantly associated with epistaxis visits (95%CI 1.3, 4.6, p = 0.008) (Table 2). Epistaxis patients were less likely to be from the Midwest (95%CI 0.2–0.9, p = 0.014) compared to the Northeast (Table 2). Patients with epistaxis were also more likely to present to an ENT/surgery clinic (95% CI 4.5–16.5, p < 0.0001) compared to a general/family medicine clinic (Table 2). In relation to comorbid conditions, and after adjustment for patient characteristics (age, gender, and race), epistaxis visits were not significantly associated with history of chronic disease, history of asthma, diagnosis of anemia, or diagnosis of anemic etiologies (Table 3). Procedures related to epistaxis control performed at epistaxis visits were low overall (2.83% of epistaxis visits), so no further analysis was performed stratifying the relationship of procedures to race. 22
International Journal of Pediatric Otorhinolaryngology 114 (2018) 20–25
L. Yang et al.
Table 2 Weighted epistaxis visits by pediatric patients, with multivariate adjustment. Weighted Total Epistaxis Visits Patient demographics Age 0–5 years 6–11 years 12–18 years Race White Hispanic Black Other/unknown Payment Source Private Insurance Medicare/Medicaid Other Unknown History of chronic disease Diagnosis of allergic rhinitis Provider demographics Region Northeast Midwes South West Practice location Non-MSA Metropolitan Practice type General/family medicine Pediatrics ENT/surgery a
Unadjusted P-Value
Adjusted ORa (95% CI)
Adjusted P-Value
Reference category 2.0 (1.2, 3.5) 2.1 (1.2, 3.7)
0.009 0.006
Reference category 1.8 (0.9, 3.7) 2.3 (1.3, 4.1) 1.9 (1.1, 3.4)
0.094 0.005 0.029
Reference category 1.1 (0.7, 1.8) 1.3 (0.5, 2.9) 3.5 (1.5, 8.2) 0.8 (0.7, 2.0) 2.4 (1.3, 4.6)
0.663 0.603 0.005 0.462 0.008
Reference category 0.5 (0.2, 0.9) 0.8 (0.4, 1.5) 0.7 (0.3, 1.4)
0.014 0.435 0.306
Reference category 0.5 (0.2, 1.1)
0.072
Reference category 1.7 (0.9, 3.3) 8.6 (4.5, 16.5)
0.117 < 0.001
0.022 1,795,057 1,545,384 1,656,646 0.005 1,856,617 1,102,237 839,606 1,198,627 0.001 2,541,091 1,624,483 239,487 592,026 1,836,863 528,988
0.265 < 0.001 0.144
1,366,042 651,782 1,969,218 1,010,045 0.005 4,734,500 262,587 < 0.001 616,889 3,228,068 843,343
Final model included age, race, payment source, history of chronic disease, region, practice location, and practice type.
providers involved, the high rate of referral to ENT/surgery may indicate a need for a more specialized level of care, beyond the initial ambulatory visit for epistaxis. It is unclear why black children may be more likely to present with epistaxis compared to white children. Previous studies of the adult population have proposed this disparity may be due to co-morbidities, as African Americans in the United States have a higher prevalence of controlled and uncontrolled hypertension, diabetes mellitus, renal disease, heart failure, and coronary artery disease, though the role of hypertension in epistaxis remains unclear [9]. However, we would not expect chronic conditions to play as significant role at a young age and believe environmental factors to play a larger role at this stage. For instance, sociodemographic factors such as cigarette smoke and air pollution exposure contribute to racial and ethnic health disparities in
from the care pattern for non-epistaxis visits. A greater percentage of visits are to ENT/surgery practitioners for epistaxis (18.0%) compared to non-epistaxis (3.7%). This finding may reveal referral patterns that literature thus far has yet to characterize, as prior studies have not examined how pediatric visits for epistaxis distribute across specialties and have primarily focused on patient presentation to the ED. Furthermore, this finding in the context of low procedures performed at epistaxis visits (2.8%) suggests that children may be over-referred to ENT providers. With these findings, population interventions can be better targeted. Providers can better understand their patient demographics and focus treatments for epistaxis. For example, recommendations regarding nasal mucosa humidification may be more useful for parents with older children, especially those living in the Northeast. In terms of the Table 3 Weighted epistaxis visits in relation to comorbidities, with multivariate adjustment. Weighted Total Epistaxis Visits Patient Demographics Age 0–5 years 6–11 years 12–18 years Gender, male Race White Hispanic Black Other/unknown Comorbid Conditions History of chronic disease History of asthma Diagnosis of allergic rhinitis Diagnosis of anemia Diagnosis of anemia causes
Unadjusted P-Value
Adjusted OR (95% CI)
Adjusted P-Value
0.5 (0.4, 1.1) 0.6 (0.3, 0.9) 0.7 (0.4, 1.1)
Ref 0.002 0.039 0.094
0.5 (0.2, 0.8) 0.4 (0.2, 0.7) 0.5 (0.3, 0.9)
Ref 0.013 0.002 0.017
1.4 0.7 0.3 0.4 0.3
0.218 0.512 < 0.001 0.307 0.117
0.318 1,676,525 1,802,336 1,399,694 3,045,121
0.092 0.005
1,856,617 1,102,237 839,606 1,198,627 1,836,863 592,258 528,988 58,520 25,846
0.265 0.158 < 0.001 0.228 0.121
Bolded terms indicate P-Value < 0.05. 23
(0.8, (0.2, (0.2, (0.1, (0.1,
2.3) 2.3) 0.6) 2.4) 1.4)
International Journal of Pediatric Otorhinolaryngology 114 (2018) 20–25
L. Yang et al.
other hand, we may be overweighting particular patients, as the data is tracked on a visit-level entry rather than a patient-level basis; patient characteristics that predispose to a greater frequency of visits may represent a larger share of our analysis. Other patient demographics, provider demographics, and related comorbidities can be explored, but at the expense of increasing the probability of a chance or clinically less meaningful finding. Moreover, this study is able to connect possible associations with epistaxis visits, but further exploration for the causation of why these factors are associated is needed.
longitudinal asthma outcomes, though do not explain the disparity in its entirety [17–20]. Smoke exposure and air pollution may also lead to differences in the rate of chemical irritation of the nasal mucosa, predisposing some children to epistaxis [21,22]. Based on our comparison of epistaxis versus non-epistaxis visits, racial disparities in epistaxis are likely due to a difference in rates of epistaxis rather than in healthcare utilization. This conclusion further supported by the literature, where utilization rates in the ED for epistaxis were similar across different racial populations [7]. Despite decreased frequency of epistaxis in white children compared to black children, prior studies have found that white children are more likely to undergo an epistaxis control procedure than black children [6]. Villwock and Goyal 2014 found that in the general population black race was associated with early intervention with ligation, however children were a minority in the study population [23]. A difference in care based on race and ethnicity is not unique to epistaxis. This is a trend reflected across other pediatric conditions and particularly well established in asthma [20,24,25]. The study also demonstrates that other factors were nonsignificant contributors to epistaxis visits. First, Medicare/Medicaid, more so Medicaid in the context of pediatric patients, was not significantly associated with epistaxis visits. Payment source serves as a stand-in for household income, and suggests that perhaps socioeconomic factors for visit payment may not influence whether or not a patient presents for epistaxis. When focused only on ED visits, however, income status does play a significant role in epistaxis visits per literature. Pallin et al., 2004 in their study of epistaxis visits to the ED that public insurance rate was 44% in epistaxis visits compared to the 32% in nonepistaxis visits [7], and children from the lowest income quartile comprised a greater proportion of epistaxis presentations (39%) [6]. After presentation, children from the lowest income quartile or patients with public insurance were least likely to undergo an epistaxis control procedure in the ED (p < 0.001) [6]. Or, even if extrapolated to tonsillectomies, increasing household income, for instance, has been found to be associated with decreasing rates of complications [14]. In the larger context of all outpatient clinic visits, other factors, such as those mentioned above, may play a greater role in patient care for epistaxis. Second, the practice location for an urban versus rural center did not appear inhibitory nor beneficiary toward epistaxis presentation. Third, comorbidities aside from allergic rhinitis, such as history of chronic disease, asthma, or anemia-related causes, are not significantly associated with epistaxis visits. Although these conditions may seem clinically relevant to determining if someone may also have concomitant nasal inflammation or increased risk of bleeding, the lack of significance of these co-morbidities with and without adjustment for other demographics indicate that epistaxis rates may be more greatly influenced by other factors, such as age. This study has several limitations. Data collection and sample size are intended to reflect the accuracy of visits across the nation, though NAMCS and NHAMCS are collected via survey of practitioners without confirmation of accuracy of patient characteristics or diagnoses. Sources for voluntary information, such as household income, are frequently missing, and other representative measures, such as payment source, must be used in analyses as a stand-in feature. This study aims to examine epistaxis in the ambulatory setting; however, this may limit data to reviewing cases of a lower severity level, as opposed to those presenting in an ED. Community Health Clinic sampling was increased in 2006, so data from 2001 to 2005 may underestimate the population of patients who use these outpatient care centers. It is estimated that the NHAMCS sampling frame covers about 85% of hospitals in the country [7], though this is more than the 20% sample of U.S. hospitals covered by alternative sources, such as the Nationwide Inpatient Sample [23]. Moreover, the NHAMCS records only the first three diagnoses, so even though epistaxis may be concomitant comorbidity, it may not be listed. Therefore, we may be underestimating the frequency of epistaxis in patients for whom it is a chronic but not primary complaint. On the
5. Conclusion Epistaxis visits by children are associated with age, race, geographic region, and provider specialty but not with payment source and urban location. Examining populations presenting for epistaxis is important for developing targeted interventions to help reduce this common presentation. Further areas of research can explore the etiologies for populations at higher risk for epistaxis. Financial support This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Conflicts of interest None. References [1] P. Dubey, P. Bds, N. Kumar, G. Bhardwaj, N. Kumar, A rare but important entity: epistaxis in infants, J. Emerg. Med. 52 (2017) 231–234, https://doi.org/10.1016/j. jemermed.2016.10.017. [2] A. Qureishi, M.J. Burton, Interventions for recurrent idiopathic epistaxis (nosebleeds) in children, Cochrane Database Syst. Rev. (2012), https://doi.org/10.1002/ 14651858.CD004461.pub3. [3] K. Davies, K. Batra, R. Mehanna, I. Keogh, Pediatric epistaxis: epidemiology, management & impact on quality of life, Int. J. Pediatr. Otorhinolaryngol. 78 (2014) 1294–1297 http://ovidsp.ovid.com/athens/ovidweb.cgi?T=JS&CSC=Y& NEWS=N&PAGE=fulltext&D=medl&AN=24882453 http://tf5lu9ym5n.search. serialssolutions.com?sid=OVID:medline&id=pmid:24882453&id=doi:10.1016% 2Fj.ijporl.2014.05.013&issn=0165-5876&isbn=&volume=78&issue=8&spag. [4] Medical Subject Headings (MeSH), Healthcare disparities, natl. Cent. Biotechnol. Information, U.S. Natl. Libr. Med. (2008), https://www.ncbi.nlm.nih.gov/mesh/? term=health+care+disparities. [5] Medical Subject Headings (MeSH), Health status disparities, natl. Cent. Biotechnol. Information, U.S. Natl. Libr. Med. (2008), https://www.ncbi.nlm.nih.gov/mesh/? term=health+status+disparities. [6] S. Shay, N.L. Shapiro, N. Bhattacharyya, Epidemiological characteristics of pediatric epistaxis presenting to the emergency department, Int. J. Pediatr. Otorhinolaryngol. 103 (2017) 121–124, https://doi.org/10.1016/j.ijporl.2017.10.026. [7] D.J. Pallin, Y.M. Chng, M.P. McKay, J.A. Emond, A.J. Pelletier, C.A. Camargo, Epidemiology of epistaxis in US emergency departments, 1992 to 2001, Ann. Emerg. Med. 46 (2005) 77–81, https://doi.org/10.1016/j.annemergmed.2004.12. 014. [8] N.J. Brown, R.G. Berkowitz, Epistaxis in healthy children requiring hospital admission, Int. J. Pediatr. Otorhinolaryngol. 68 (2004) 1181–1184, https://doi.org/ 10.1016/j.ijporl.2004.04.015. [9] M.R. Chaaban, D. Zhang, V. Resto, J.S. Goodwin, Demographic, seasonal, and geographic differences in emergency department visits for epistaxis, Otolaryngol. Head Neck Surg. 156 (2017) 81–86, https://doi.org/10.1177/0194599816667295. [10] J. Tenney, P.H.K. White, J. Williamson, National ambulatory medical care survey: background and methodology, Natl. Cent. Heal. Stat. Vital Heal. Stat. 2 (1974). [11] M.T. L. McCaig, Plan and operation of the national hospital ambulatory medical care survey, Natl. Cent. Heal. Stat. Vital Heal. Stat. 1 (1994). [12] N. Patel, J. Maddalozzo, K.R. Billings, An update on management of pediatric epistaxis, Int. J. Pediatr. Otorhinolaryngol. 78 (2014) 1400–1404 http://ovidsp. ovid.com/athens/ovidweb.cgi?T=JS&CSC=Y&NEWS=N&PAGE=fulltext&D= medl&AN=24972938 http://tf5lu9ym5n.search.serialssolutions.com?sid= OVID:medline&id=pmid:24972938&id=doi:10.1016%2Fj.ijporl.2014.06.009& issn=0165-5876&isbn=&volume=78&issue=8&spag. [13] N. Bhattacharyya, N.L. Shapiro, Associations between socioeconomic status and race with complications after tonsillectomy in children, Otolaryngol. Head Neck Surg. 151 (2014) 1055–1060, https://doi.org/10.1177/0194599814552647. [14] S. Shay, N.L. Shapiro, N. Bhattacharyya, Pediatric otolaryngologic conditions: racial and socioeconomic disparities in the United States, Laryngoscope 127 (2017)
24
International Journal of Pediatric Otorhinolaryngology 114 (2018) 20–25
L. Yang et al.
[21] L. Calderón‐Garcidueñas, A. Rodriguez‐Alcaraz, R. García, L. Ramírez, G. Barragan, Nasal inflammatory responses in children exposed to a polluted urban atmosphere, J. Toxicol. Environ. Health 45 (1995) 427–437, https://doi.org/10.1080/ 15287399509532006. [22] D. Shusterman, Toxicology of nasal irritants, Curr. Allergy Asthma Rep. 3 (2003) 258–265, https://doi.org/10.1007/s11882-003-0048-z. [23] J.A. Villwock, P. Goyal, Early versus delayed treatment of primary epistaxis in the United States, Int. Forum Allergy Rhinol. 4 (2014) 69–75, https://doi.org/10.1002/ alr.21236. [24] M. Trivedi, V. Fung, E.O. Kharbanda, E.K. Larkin, M.G. Butler, K. Horan, T.A. Lieu, A.C. Wu, Racial disparities in family-provider interactions for pediatric asthma care, J. Asthma (2017), https://doi.org/10.1080/02770903.2017.1337790. [25] K. Parikh, J. Berry, M. Hall, G.M. Mussman, A. Montalbano, J. Thomson, R. Morse, K.M. Wilson, S.S. Shah, T.H.E.J. Ournal, O.F.P. Ediatrics, K. Parikh, J. Berry, M. Hall, G.M. Mussman, A. Montalbano, J. Thomson, R. Morse, K.M. Wilson, S.S. Shah, Racial and ethnic differences in pediatric readmissions for common chronic conditions, J. Pediatr. 186 (2017) 158–164.
746–752, https://doi.org/10.1002/lary.26240. [15] L.S. Girsh, Allergic rhinitis, a common cause of recurrent epistaxis in children, Arch. Pediatr. Adolesc. Med. 99 (1960) 819–821, https://doi.org/10.1001/archpedi. 1960.02070030821017. [16] M.R. Purkey, Z. Seeskin, R. Chandra, Seasonal variation and predictors of epistaxis, Laryngoscope 124 (2014) 2028–2033, https://doi.org/10.1002/lary.24679. [17] A. Aligne, P. Auinger, R. Byrd, M. Weitzman, Risk factors for pediatric asthma, Am. J. Respir. Crit. Care Med. 162 (2000) 873–877, https://doi.org/10.1164/ajrccm. 162.3.9908085. [18] J. Gern, The urban environment and childhood asthma study, J. Allergy Clin. Immunol. 125 (2010) 545–549, https://doi.org/10.1016/j.jaci.2010.01.037. [19] M. Guarnieri, J. Balmes, Outdoor air pollution and asthma, Lancet 383 (2014) 1581–1592, https://doi.org/10.1016/S0140-6736(14)60617-6. [20] D.M. Washington, L.M. Curtis, K. Waite, M.S. Wolf, M.K. Paasche-Orlow, Sociodemographic factors mediate race and ethnicity-associated childhood asthma health disparities: a longitudinal analysis, J. Racial Ethnic Health Dispar. (2017) 1–11 https://doi.org/10.1007/s40615-017-0441-2.
25
Contents lists available at ScienceDirect
International Journal of Pediatric Otorhinolaryngology journal homepage: www.elsevier.com/locate/ijporl
Epistaxis health disparities in the United States pediatric population a
b
b,c
Lindsay Yang , Kevin Hur , Jeffrey Koempel , Elisabeth H. Ference a b c
T
b,c,∗
Keck School of Medicine of the University of Southern California, Los Angeles, United States Rick and Tina Caruso Department of Otolaryngology-Head and Neck Surgery, Keck School of Medicine of the University of Southern California, Los Angeles, United States Division of Otolaryngology-Head and Neck Surgery, Department of Surgery, Children's Hospital Los Angeles, Los Angeles, United States
A R T I C LE I N FO
A B S T R A C T
Meeting: Poster Presentation at the American Rhinologic Society, Combined Otolaryngology Spring Meeting, on April 19–20, 2018 in National Harbor, Maryland
Objective: Despite epistaxis occurring in up to 60% of the population, few studies have investigated health status disparities in the pediatric epistaxis population. The aim of this study was to evaluate sociodemographic risk factors associated with epistaxis visits for pediatric patients. Methods: Data were extracted from the National Ambulatory Medical Care Survey and National Hospital Ambulatory Medical Care Survey Outpatient Department from 2001 to 2010. Outpatient visits of children less than 18 years who received a primary, secondary, or tertiary diagnosis of epistaxis (ICD-9CM code 784.7X) were included. Bivariate and stepwise multivariate regressions were conducted to develop a final model for epistaxis visits described by sociodemographics. Results: Epistaxis visits accounted for 5 ± 0.6 million visits in children less than 18 years. 51% and 33% of children presenting with epistaxis had private insurance and Medicaid, respectively (p = 0.001). 69% of epistaxis visits were evaluated at a pediatric clinic, 18% at an ENT/surgery clinic, and 13% at a general/family medicine clinic (p < 0.0001). After multivariate adjustment, epistaxis visits were associated with older age (p = 0.006). Black children were more likely to present with epistaxis (95% CI 1.3–4.1, p = 0.005) compared to white children. Allergic rhinitis, present in 11% of epistaxis visits, was a significant comorbidity associated with visits (95%CI 1.3–4.6, p = 0.008). Patients were also more likely to present to an ENT/surgery clinic (95% CI 4.5–16.5, p < 0.0001) compared to a general/family medicine clinic. Conclusions: Epistaxis visits by children are associated with age, race, and specialty. Targeted interventions to help reduce this common presentation should be developed.
Keywords: Epistaxis Pediatric Otolaryngology Public health Rhinology Manuscript
1. Introduction Epistaxis is a very common condition among children. Though it is rare in children younger than two years old [1], at least one episode of epistaxis has occurred in 30% of children by the age of five and in over 50% of children greater than five years old [2]. Most cases in children and young adults are self-limited and due to dry nasal mucosa; some are anterior, originating from Little's area in the anterior portion of the nasal septum where Kiesselbach's plexus forms. However, if more severe, patients may present to an emergency care provider, primary care provider, or otolaryngologist. Children reported high stress scores for bleeds longer than the average time of 5–10 min, and caregivers' fear of excessive blood loss was a significant factor causing high distress [3]. Despite the frequency of epistaxis, few studies have investigated health status disparities in the pediatric population. While healthcare
disparities focus on the differences in access to or availability of medical facilities and services, health status disparities refer to the variation in rates of disease occurrence and disabilities between socioeconomic and/or geographically defined population groups [4,5]. Some studies have examined health status disparities in presentation to the emergency department (ED) of certain states [3,6–9], but none have captured other sources of patient care or provided a national pediatric outlook. The aim of this study was to evaluate sociodemographic risk factors associated with epistaxis visits for pediatric patients in a nationally representative sample in the outpatient setting. 2. Methods 2.1. Data sources Data were extracted from the National Ambulatory Medical Care
∗ Corresponding author. Rick and Tina Caruso Department of Otolaryngology-Head and Neck Surgery, Keck School of Medicine of the University of Southern California, 1540 Alcazar Street, Suite 204M, Los Angeles, CA, 90033 United States. E-mail addresses: [email protected] (L. Yang), [email protected] (K. Hur), [email protected] (J. Koempel), [email protected] (E.H. Ference).
https://doi.org/10.1016/j.ijporl.2018.08.025 Received 2 July 2018; Received in revised form 20 August 2018; Accepted 20 August 2018 Available online 23 August 2018 0165-5876/ © 2018 Elsevier B.V. All rights reserved.
International Journal of Pediatric Otorhinolaryngology 114 (2018) 20–25
L. Yang et al.
identified on the survey; unknown category includes those surveys listed as unknown or uncompleted. Due to missing data for median household income, payment source was used as an alternative. Provider demographics included practice region (Northeast, Midwest, South, West); practice location (metropolitan statistical area, non-metropolitan statistical area); practice type (general/family medicine, pediatrics, ENT/surgery); new patient to practice; and number of visits in the last year (no visits, 1–5 visits, ≥6 visits). A stepwise multivariate logistic regression was conducted to assess the percentage of epistaxis visits with a refined list of covariates. Variables statistically significant at the 5% level, that modified the association between provider and outcome by more than 10% (when included in the model), or were clinically meaningful were retained in the final model. Epistaxis visits were also examined for their association with various comorbid conditions, including history of chronic disease, history of asthma, diagnosis of allergic rhinitis (ICD-9CM code 477.9X), diagnosis of anemia, and diagnosis of anemic etiologies. Anemic etiologies included iron deficiency anemias (ICD-9CM codes 280.8X, 280.9X), megaloblastic or nutritional deficiency anemias (ICD-9CM codes 281. XX), hereditary anemias (ICD-9CM code 282. XX), sickle cell anemias (ICD-9CM code 283. XX), aplastic anemias (ICD-9CM code 284. XX) and other chronic disease anemias (285. XX). Lastly, procedures performed at epistaxis visits were analyzed to evaluate their relationship to race (ICD-9CM codes 210. XX for epistaxis control, 212. XX for diagnostic procedure on nose, 22X.XX for operations on nasal sinuses. Statistical analyses were performed using the SURVEYFREQ, SURVEYREG, SURVEYLOGISTIC, and GENMOD procedures in SAS statistical software (version 9.4, SAS Institute Inc, Cary, North Carolina, USA).
Survey (NAMCS) and National Hospital Ambulatory Medical Care Survey (NHAMCS) Outpatient Department (OPD) from 2001 to 2010. The Centers for Disease Control and Prevention (CDC) and National Center for Health Sciences collect the NAMCS and NHAMCS annually to measure the utilization and provision of ambulatory care services, which is the predominant method of providing health care in the United States [10,11]. The NAMCS uses a 2-stage probability design that samples physicians who represent geographical regions and patient visits within these practices. Data were collected over a 1-week period randomized to the physician. The NHAMCS is intended to offer more complete ambulatory data than previously covered by the NAMCS. The NHAMCS uses a 4-stage probability design that samples approximately 500 nationally representative primary sampling units (PSUs) of the 50 states and the District of Columbia, hospitals within the PSUs, clinics within the hospitals, and patient visits within the clinics. Data was collected over a 4-week period randomized by facility and included patient, visit, and provider characteristics. Both the NAMCS and NHAMCS are weighted to represent national usage, with visits adjusted for the provider specialty, visits to the provider, nonresponse, and smoothing. Both surveys have undergone changes since their inception. Within the 2001–2010 window, phrasing of questions and the availability of variables have changed. History of chronic disease and urban-rural classification of a patient's zip code were introduced in 2005 [10]. Practice specialty type is missing in the 2005 data tables. The survey uses the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9CM) codes for recording diagnoses. Providers surveyed can complete up to 3 diagnoses. 2.2. Inclusion criteria Outpatient visits of children less than 18 years who received a primary, secondary, or tertiary diagnosis of epistaxis (ICD-9CM code 784.7X) were included. Patients whose diagnosis was missing in all 3 instances were excluded. The primary outcome measure was the weighted proportion of epistaxis visits. Sociodemographic variables were collected in addition to the type of clinic visited and type of provider seen. Clinic types were categorized as general/family medicine, pediatrics clinic, or ENT/surgery clinic. Provider types included a physician, nurse practitioner, or physician assistant. Results indicate variables that were not available during all years and the first available year.
3. Results Epistaxis visits accounted for 5.0 ± 0.6 million visits by children less than 18 years from 2001 to 2010. The ages of patients with a diagnosis of epistaxis were similar across the 3 divisions 0–5, 6–11, and 12–18 years old (36%, 31%, and 33%, respectively), though this was significantly different from the age distribution across non-epistaxis visits, which had a greater predominance toward 0–5 year olds (49%) (p = 0.022) (Table 1). Males (61%) accounted for more epistaxis visits than females, but a similar pattern was true across visits overall (Table 1). The race of patients by frequency of epistaxis visits were Whites (37%), Blacks (17%), Other/unknown (24%), and Hispanics (22%) (Table 1). Blacks accounted for a significantly greater share of epistaxis visits than they do for non-epistaxis visits (11%) (p = 0.005) (Table 1). Payment source distribution was also significantly different between epistaxis and non-epistaxis visits, with a decreased share of private insurance visits (51% and 60%, respectively) (p = 0.001) (Table 1). A diagnosis of allergic rhinitis (11%) was also significantly associated with epistaxis visits (p < 0.001) (Table 1). There were no significant differences in other patient demographics, including history of chronic disease, history of asthma, diagnosis of anemia, or diagnosis of anemic etiologies between children with epistaxis versus those who did not have epistaxis. Practice location is predominantly Metropolitan across all visits (88%), but even more so for epistaxis visits (95%) (p = 0.005). There was a greater percentage of epistaxis visits seen at an ENT/surgery clinic (18%) rather than general/family medicine clinic (13%), compared to non-epistaxis visits, but the majority of visits were still seen with pediatrics providers (69%) (p < 0.001) (Table 1). Practice region, first visit, and visit frequency were not significantly different based on bivariate comparison. The final multivariate model included covariates age, race, payment source, history of chronic disease, diagnosis of allergic rhinitis, region, practice location, and practice type. After multivariate adjustment,
2.3. Statistical analysis Bivariate and stepwise multivariate regressions were conducted to develop a final model for epistaxis visits described by sociodemographics. The number of epistaxis visits were first weighted based on the surveys' PSU, strata, and visit weights. The weighted proportion of epistaxis visits were compared to patient and provider characteristics using Chi-square tests. Patient demographic included age (0–5, 6–11, 12–18 years old). These categories were selected to best capture an even distribution of frequency (1,795,057 weighted visits, 1,545,384 weighted visits, 1,656,646 weighted visits, respectively); moreover, these age ranges carry clinical significance in early childhood, middle childhood, and adolescence. Other demographic information included sex (female, male); race (White, Hispanic, Black, and other/unknown); payment source (private insurance, Medicare/Medicaid, other, unknown); history of chronic disease; history of asthma; diagnosis of anemia; or diagnosis of various anemic etiologies. The category of other/unknown race includes all patients who do not fall under White non-Hispanic, Black non-Hispanic, and Hispanic categories; patients may have identified as Asian, Native Hawaiian/Other Pacific Islander, American, or Indian/Alaska Native, or may have reported more than one race. Under payment source, other includes worker's compensation, self-pay, no charge/charity, or other 21
International Journal of Pediatric Otorhinolaryngology 114 (2018) 20–25
L. Yang et al.
4. Discussion
Table 1 Characteristics of visits by pediatric patients with epistaxis, based on weighted visits.
Unweighted visits Estimated weighted visits Patient demographics, % Age 0–5 years 6–11 years 12–18 years Gender, male Race White Hispanic Black Other/unknown Payment Source Private Insurance Medicare/Medicaid Other Unknown History of chronic disease History of asthma Diagnosis of allergic rhinitis Diagnosis of anemia Diagnosis of anemia causes Provider demographics, % Region Northeast Midwest South West Practice location Non-MSA Metropolitan Practice type General/family medicine Pediatrics ENT/surgery New patient to practice Past visits to practice in 12 mos No visits 1–5 visits ≥6 visits
Epistaxis (SE)
Non-Epistaxis (SE)
274 4,997,087 (100)
107,983 1,812,827,680 (100)
35.9 30.9 33.2 60.9
(5.2) (5.0) (5.4) (5.2)
49.2 20.4 30.4 51.9
(0.5) (0.3) (0.4) (0.3)
37.2 22.1 16.8 24.0
(5.4) (5.6) (3.7) (4.6)
56.5 15.9 10.6 17.0
(1.3) (1.3) (0.7) (1.0)
This is the first national study to report that pediatric epistaxis visits are associated with age, race, practice location, and practitioner type. First, our study finds that epistaxis visits to providers are more likely to occur with older ages, 6–11 year olds in relation to 0–5 year olds and slightly more so with 12–18 year olds. The average age of children presenting with epistaxis in our study was 8.32 years, which is generally consistent with the existing literature. Shay et al., 2017, who studied pediatric visits to EDs for four states, reported a mean age of 7.54 [6]; Pallin et al., 2005, who reviewed NAMCS and NHAMCS from an earlier timeframe (1992–2001) across all populations, found a bimodal distribution peaking among those younger than 10 and then again at 70–79 years old [7]; and Patel et al., 2014 had an average age of 9.1 years with 175 patients [12]. Second, epistaxis visits are significantly more associated with black children than white children, with and without adjustments for other patient factors and provider demographics. Epistaxis visits are also more significantly associated with children of other/unknown ethnicity than white children, though the clinical significance is unclear given that this category was generated due to insufficient sample size when disaggregated. In general though, our findings differ from other studies of racial disparities in epistaxis. Hispanic patients represented the highest proportion of children with epistaxis in Shay et al., (2017), though the population of this study was representative of state ED databases for New York, Florida, Iowa, and California for 2010 and may not reflect the national trends for other care settings [6]. When racial disparities are expanded to the adult population, Pallin et al., 2005 noted no significant difference between ED utilization rates among black versus white epistaxis patients and Hispanic versus non-Hispanic epistaxis patients, with higher rates of ED use among blacks for all conditions [7]. Chaaban et al., 2016 did find that blacks were significantly more likely to present with epistaxis than non-Hispanic whites (1.23 fold), though the study covered only Medicare beneficiaries who represent a much older population [9]. Nevertheless, the concept that disparities exist in pediatric otolaryngology has been tested in other fields [13]. Black and Hispanic children have more revisits after a tonsillectomy with a greater likelihood of acute pain than do white children [14]. The low number of procedures performed at ambulatory visits did not allow us to assess if there was racial disparity as well in the treatment received, which Shay et al., 2017 noted in their ED study [14]. In future studies, racial disparities and the difference in findings between ambulatory and ED settings can be further explored through a longitudinal database including both ED and ambulatory visits; this could further elucidate referral patterns regarding minority populations. If there is a racial difference in referral patterns or ED visits, understanding current utilization would allow for better resource use and patient management by correcting this differential. Third, it has long since been realized that allergic rhinitis is a common cause of recurrent epistaxis [15], and this occurrence has been followed in larger studies as well [16]. This study's findings are consistent with prior literature, as a diagnosis of allergic rhinitis indicates a pediatric patient is 2.4 times more likely to present with epistaxis. Fourth, practice location epistaxis visits were more likely to present in the Northeast than the Midwest. This was consistent with literature that describes how epistaxis rates are higher in the northern United States than the south, perhaps reflecting temperature effects on the nasal mucosa, use of central heating, or upper respiratory infection frequency [9]. Patients in the Mountain and Alaska regions are 1.42 and 1.78 times more likely to present with epistaxis, respectively, whereas patients in the South Atlantic, East South Central, West South Central, and California regions are less likely to present with epistaxis, though these studies do not cover the pediatric population [9]. Fifth, ENT/surgery provided a significantly greater portion of care for epistaxis visits, compared to general/family medicine practitioners. The distribution of care providers for epistaxis visits differs significantly
P-Value
0.022
0.092 0.005
0.001 50.9 (6.0) 32.5 (5.6) 4.8 (2.0) 11.8 (4.5) 36.8 (5.8) 6.6 (4.0) 10.6 (2.9)
60.3 (1.1) 31.0 (1.1) 5.1 (0.3) 3.5 (0.3) 43.4 (1.1) 4.6 (0.2) 3.2 (0.2)
0.265 0.545 < 0.001
1.2 (1.1) 0.5 (0.4)
0.4 (0.0) 0.2 (0.0)
0.228 0.121
27.3 13.0 39.4 20.2
20.4 22.1 37.3 20.1
0.144
(5.5) (3.2) (5.9) (4.9)
(1.8) (2.4) (2.7) (2.1)
0.005 5.3 (2.0) 94.7 (2.0)
12.3 (2.6) 87.7 (2.6)
13.2 (3.5)
24.0 (1.2)
68.9 (4.5) 18.0 (2.9) 12.6 (3.0)
72.3 (1.2) 3.7 (0.2) 11.7 (0.3)
19.0 (3.7) 66.6 (4.5) 14.4 (3.7)
17.4 (0.4) 59.4 (0.5) 23.2 (0.6)
< 0.001
0.760 0.111
epistaxis visits were associated with older age, 6–11 year olds (95%CI 1.2–3.5, p = 0.009) and 12–18 year olds (95%CI 1.2–3.7, p = 0.006) (Table 2). Black children (95%CI 1.3–4.1, p = 0.005) and children of other/unknown ethnicity (95%CI 1.1–3.4, p = 0.029) were more likely to present with epistaxis compared to White children (Table 2). Hispanic children compared to white children presented with higher rates of epistaxis, however the observed difference was not significant after adjustment for cofounders. Additionally, a diagnosis of allergic rhinitis was significantly associated with epistaxis visits (95%CI 1.3, 4.6, p = 0.008) (Table 2). Epistaxis patients were less likely to be from the Midwest (95%CI 0.2–0.9, p = 0.014) compared to the Northeast (Table 2). Patients with epistaxis were also more likely to present to an ENT/surgery clinic (95% CI 4.5–16.5, p < 0.0001) compared to a general/family medicine clinic (Table 2). In relation to comorbid conditions, and after adjustment for patient characteristics (age, gender, and race), epistaxis visits were not significantly associated with history of chronic disease, history of asthma, diagnosis of anemia, or diagnosis of anemic etiologies (Table 3). Procedures related to epistaxis control performed at epistaxis visits were low overall (2.83% of epistaxis visits), so no further analysis was performed stratifying the relationship of procedures to race. 22
International Journal of Pediatric Otorhinolaryngology 114 (2018) 20–25
L. Yang et al.
Table 2 Weighted epistaxis visits by pediatric patients, with multivariate adjustment. Weighted Total Epistaxis Visits Patient demographics Age 0–5 years 6–11 years 12–18 years Race White Hispanic Black Other/unknown Payment Source Private Insurance Medicare/Medicaid Other Unknown History of chronic disease Diagnosis of allergic rhinitis Provider demographics Region Northeast Midwes South West Practice location Non-MSA Metropolitan Practice type General/family medicine Pediatrics ENT/surgery a
Unadjusted P-Value
Adjusted ORa (95% CI)
Adjusted P-Value
Reference category 2.0 (1.2, 3.5) 2.1 (1.2, 3.7)
0.009 0.006
Reference category 1.8 (0.9, 3.7) 2.3 (1.3, 4.1) 1.9 (1.1, 3.4)
0.094 0.005 0.029
Reference category 1.1 (0.7, 1.8) 1.3 (0.5, 2.9) 3.5 (1.5, 8.2) 0.8 (0.7, 2.0) 2.4 (1.3, 4.6)
0.663 0.603 0.005 0.462 0.008
Reference category 0.5 (0.2, 0.9) 0.8 (0.4, 1.5) 0.7 (0.3, 1.4)
0.014 0.435 0.306
Reference category 0.5 (0.2, 1.1)
0.072
Reference category 1.7 (0.9, 3.3) 8.6 (4.5, 16.5)
0.117 < 0.001
0.022 1,795,057 1,545,384 1,656,646 0.005 1,856,617 1,102,237 839,606 1,198,627 0.001 2,541,091 1,624,483 239,487 592,026 1,836,863 528,988
0.265 < 0.001 0.144
1,366,042 651,782 1,969,218 1,010,045 0.005 4,734,500 262,587 < 0.001 616,889 3,228,068 843,343
Final model included age, race, payment source, history of chronic disease, region, practice location, and practice type.
providers involved, the high rate of referral to ENT/surgery may indicate a need for a more specialized level of care, beyond the initial ambulatory visit for epistaxis. It is unclear why black children may be more likely to present with epistaxis compared to white children. Previous studies of the adult population have proposed this disparity may be due to co-morbidities, as African Americans in the United States have a higher prevalence of controlled and uncontrolled hypertension, diabetes mellitus, renal disease, heart failure, and coronary artery disease, though the role of hypertension in epistaxis remains unclear [9]. However, we would not expect chronic conditions to play as significant role at a young age and believe environmental factors to play a larger role at this stage. For instance, sociodemographic factors such as cigarette smoke and air pollution exposure contribute to racial and ethnic health disparities in
from the care pattern for non-epistaxis visits. A greater percentage of visits are to ENT/surgery practitioners for epistaxis (18.0%) compared to non-epistaxis (3.7%). This finding may reveal referral patterns that literature thus far has yet to characterize, as prior studies have not examined how pediatric visits for epistaxis distribute across specialties and have primarily focused on patient presentation to the ED. Furthermore, this finding in the context of low procedures performed at epistaxis visits (2.8%) suggests that children may be over-referred to ENT providers. With these findings, population interventions can be better targeted. Providers can better understand their patient demographics and focus treatments for epistaxis. For example, recommendations regarding nasal mucosa humidification may be more useful for parents with older children, especially those living in the Northeast. In terms of the Table 3 Weighted epistaxis visits in relation to comorbidities, with multivariate adjustment. Weighted Total Epistaxis Visits Patient Demographics Age 0–5 years 6–11 years 12–18 years Gender, male Race White Hispanic Black Other/unknown Comorbid Conditions History of chronic disease History of asthma Diagnosis of allergic rhinitis Diagnosis of anemia Diagnosis of anemia causes
Unadjusted P-Value
Adjusted OR (95% CI)
Adjusted P-Value
0.5 (0.4, 1.1) 0.6 (0.3, 0.9) 0.7 (0.4, 1.1)
Ref 0.002 0.039 0.094
0.5 (0.2, 0.8) 0.4 (0.2, 0.7) 0.5 (0.3, 0.9)
Ref 0.013 0.002 0.017
1.4 0.7 0.3 0.4 0.3
0.218 0.512 < 0.001 0.307 0.117
0.318 1,676,525 1,802,336 1,399,694 3,045,121
0.092 0.005
1,856,617 1,102,237 839,606 1,198,627 1,836,863 592,258 528,988 58,520 25,846
0.265 0.158 < 0.001 0.228 0.121
Bolded terms indicate P-Value < 0.05. 23
(0.8, (0.2, (0.2, (0.1, (0.1,
2.3) 2.3) 0.6) 2.4) 1.4)
International Journal of Pediatric Otorhinolaryngology 114 (2018) 20–25
L. Yang et al.
other hand, we may be overweighting particular patients, as the data is tracked on a visit-level entry rather than a patient-level basis; patient characteristics that predispose to a greater frequency of visits may represent a larger share of our analysis. Other patient demographics, provider demographics, and related comorbidities can be explored, but at the expense of increasing the probability of a chance or clinically less meaningful finding. Moreover, this study is able to connect possible associations with epistaxis visits, but further exploration for the causation of why these factors are associated is needed.
longitudinal asthma outcomes, though do not explain the disparity in its entirety [17–20]. Smoke exposure and air pollution may also lead to differences in the rate of chemical irritation of the nasal mucosa, predisposing some children to epistaxis [21,22]. Based on our comparison of epistaxis versus non-epistaxis visits, racial disparities in epistaxis are likely due to a difference in rates of epistaxis rather than in healthcare utilization. This conclusion further supported by the literature, where utilization rates in the ED for epistaxis were similar across different racial populations [7]. Despite decreased frequency of epistaxis in white children compared to black children, prior studies have found that white children are more likely to undergo an epistaxis control procedure than black children [6]. Villwock and Goyal 2014 found that in the general population black race was associated with early intervention with ligation, however children were a minority in the study population [23]. A difference in care based on race and ethnicity is not unique to epistaxis. This is a trend reflected across other pediatric conditions and particularly well established in asthma [20,24,25]. The study also demonstrates that other factors were nonsignificant contributors to epistaxis visits. First, Medicare/Medicaid, more so Medicaid in the context of pediatric patients, was not significantly associated with epistaxis visits. Payment source serves as a stand-in for household income, and suggests that perhaps socioeconomic factors for visit payment may not influence whether or not a patient presents for epistaxis. When focused only on ED visits, however, income status does play a significant role in epistaxis visits per literature. Pallin et al., 2004 in their study of epistaxis visits to the ED that public insurance rate was 44% in epistaxis visits compared to the 32% in nonepistaxis visits [7], and children from the lowest income quartile comprised a greater proportion of epistaxis presentations (39%) [6]. After presentation, children from the lowest income quartile or patients with public insurance were least likely to undergo an epistaxis control procedure in the ED (p < 0.001) [6]. Or, even if extrapolated to tonsillectomies, increasing household income, for instance, has been found to be associated with decreasing rates of complications [14]. In the larger context of all outpatient clinic visits, other factors, such as those mentioned above, may play a greater role in patient care for epistaxis. Second, the practice location for an urban versus rural center did not appear inhibitory nor beneficiary toward epistaxis presentation. Third, comorbidities aside from allergic rhinitis, such as history of chronic disease, asthma, or anemia-related causes, are not significantly associated with epistaxis visits. Although these conditions may seem clinically relevant to determining if someone may also have concomitant nasal inflammation or increased risk of bleeding, the lack of significance of these co-morbidities with and without adjustment for other demographics indicate that epistaxis rates may be more greatly influenced by other factors, such as age. This study has several limitations. Data collection and sample size are intended to reflect the accuracy of visits across the nation, though NAMCS and NHAMCS are collected via survey of practitioners without confirmation of accuracy of patient characteristics or diagnoses. Sources for voluntary information, such as household income, are frequently missing, and other representative measures, such as payment source, must be used in analyses as a stand-in feature. This study aims to examine epistaxis in the ambulatory setting; however, this may limit data to reviewing cases of a lower severity level, as opposed to those presenting in an ED. Community Health Clinic sampling was increased in 2006, so data from 2001 to 2005 may underestimate the population of patients who use these outpatient care centers. It is estimated that the NHAMCS sampling frame covers about 85% of hospitals in the country [7], though this is more than the 20% sample of U.S. hospitals covered by alternative sources, such as the Nationwide Inpatient Sample [23]. Moreover, the NHAMCS records only the first three diagnoses, so even though epistaxis may be concomitant comorbidity, it may not be listed. Therefore, we may be underestimating the frequency of epistaxis in patients for whom it is a chronic but not primary complaint. On the
5. Conclusion Epistaxis visits by children are associated with age, race, geographic region, and provider specialty but not with payment source and urban location. Examining populations presenting for epistaxis is important for developing targeted interventions to help reduce this common presentation. Further areas of research can explore the etiologies for populations at higher risk for epistaxis. Financial support This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Conflicts of interest None. References [1] P. Dubey, P. Bds, N. Kumar, G. Bhardwaj, N. Kumar, A rare but important entity: epistaxis in infants, J. Emerg. Med. 52 (2017) 231–234, https://doi.org/10.1016/j. jemermed.2016.10.017. [2] A. Qureishi, M.J. Burton, Interventions for recurrent idiopathic epistaxis (nosebleeds) in children, Cochrane Database Syst. Rev. (2012), https://doi.org/10.1002/ 14651858.CD004461.pub3. [3] K. Davies, K. Batra, R. Mehanna, I. Keogh, Pediatric epistaxis: epidemiology, management & impact on quality of life, Int. J. Pediatr. Otorhinolaryngol. 78 (2014) 1294–1297 http://ovidsp.ovid.com/athens/ovidweb.cgi?T=JS&CSC=Y& NEWS=N&PAGE=fulltext&D=medl&AN=24882453 http://tf5lu9ym5n.search. serialssolutions.com?sid=OVID:medline&id=pmid:24882453&id=doi:10.1016% 2Fj.ijporl.2014.05.013&issn=0165-5876&isbn=&volume=78&issue=8&spag. [4] Medical Subject Headings (MeSH), Healthcare disparities, natl. Cent. Biotechnol. Information, U.S. Natl. Libr. Med. (2008), https://www.ncbi.nlm.nih.gov/mesh/? term=health+care+disparities. [5] Medical Subject Headings (MeSH), Health status disparities, natl. Cent. Biotechnol. Information, U.S. Natl. Libr. Med. (2008), https://www.ncbi.nlm.nih.gov/mesh/? term=health+status+disparities. [6] S. Shay, N.L. Shapiro, N. Bhattacharyya, Epidemiological characteristics of pediatric epistaxis presenting to the emergency department, Int. J. Pediatr. Otorhinolaryngol. 103 (2017) 121–124, https://doi.org/10.1016/j.ijporl.2017.10.026. [7] D.J. Pallin, Y.M. Chng, M.P. McKay, J.A. Emond, A.J. Pelletier, C.A. Camargo, Epidemiology of epistaxis in US emergency departments, 1992 to 2001, Ann. Emerg. Med. 46 (2005) 77–81, https://doi.org/10.1016/j.annemergmed.2004.12. 014. [8] N.J. Brown, R.G. Berkowitz, Epistaxis in healthy children requiring hospital admission, Int. J. Pediatr. Otorhinolaryngol. 68 (2004) 1181–1184, https://doi.org/ 10.1016/j.ijporl.2004.04.015. [9] M.R. Chaaban, D. Zhang, V. Resto, J.S. Goodwin, Demographic, seasonal, and geographic differences in emergency department visits for epistaxis, Otolaryngol. Head Neck Surg. 156 (2017) 81–86, https://doi.org/10.1177/0194599816667295. [10] J. Tenney, P.H.K. White, J. Williamson, National ambulatory medical care survey: background and methodology, Natl. Cent. Heal. Stat. Vital Heal. Stat. 2 (1974). [11] M.T. L. McCaig, Plan and operation of the national hospital ambulatory medical care survey, Natl. Cent. Heal. Stat. Vital Heal. Stat. 1 (1994). [12] N. Patel, J. Maddalozzo, K.R. Billings, An update on management of pediatric epistaxis, Int. J. Pediatr. Otorhinolaryngol. 78 (2014) 1400–1404 http://ovidsp. ovid.com/athens/ovidweb.cgi?T=JS&CSC=Y&NEWS=N&PAGE=fulltext&D= medl&AN=24972938 http://tf5lu9ym5n.search.serialssolutions.com?sid= OVID:medline&id=pmid:24972938&id=doi:10.1016%2Fj.ijporl.2014.06.009& issn=0165-5876&isbn=&volume=78&issue=8&spag. [13] N. Bhattacharyya, N.L. Shapiro, Associations between socioeconomic status and race with complications after tonsillectomy in children, Otolaryngol. Head Neck Surg. 151 (2014) 1055–1060, https://doi.org/10.1177/0194599814552647. [14] S. Shay, N.L. Shapiro, N. Bhattacharyya, Pediatric otolaryngologic conditions: racial and socioeconomic disparities in the United States, Laryngoscope 127 (2017)
24
International Journal of Pediatric Otorhinolaryngology 114 (2018) 20–25
L. Yang et al.
[21] L. Calderón‐Garcidueñas, A. Rodriguez‐Alcaraz, R. García, L. Ramírez, G. Barragan, Nasal inflammatory responses in children exposed to a polluted urban atmosphere, J. Toxicol. Environ. Health 45 (1995) 427–437, https://doi.org/10.1080/ 15287399509532006. [22] D. Shusterman, Toxicology of nasal irritants, Curr. Allergy Asthma Rep. 3 (2003) 258–265, https://doi.org/10.1007/s11882-003-0048-z. [23] J.A. Villwock, P. Goyal, Early versus delayed treatment of primary epistaxis in the United States, Int. Forum Allergy Rhinol. 4 (2014) 69–75, https://doi.org/10.1002/ alr.21236. [24] M. Trivedi, V. Fung, E.O. Kharbanda, E.K. Larkin, M.G. Butler, K. Horan, T.A. Lieu, A.C. Wu, Racial disparities in family-provider interactions for pediatric asthma care, J. Asthma (2017), https://doi.org/10.1080/02770903.2017.1337790. [25] K. Parikh, J. Berry, M. Hall, G.M. Mussman, A. Montalbano, J. Thomson, R. Morse, K.M. Wilson, S.S. Shah, T.H.E.J. Ournal, O.F.P. Ediatrics, K. Parikh, J. Berry, M. Hall, G.M. Mussman, A. Montalbano, J. Thomson, R. Morse, K.M. Wilson, S.S. Shah, Racial and ethnic differences in pediatric readmissions for common chronic conditions, J. Pediatr. 186 (2017) 158–164.
746–752, https://doi.org/10.1002/lary.26240. [15] L.S. Girsh, Allergic rhinitis, a common cause of recurrent epistaxis in children, Arch. Pediatr. Adolesc. Med. 99 (1960) 819–821, https://doi.org/10.1001/archpedi. 1960.02070030821017. [16] M.R. Purkey, Z. Seeskin, R. Chandra, Seasonal variation and predictors of epistaxis, Laryngoscope 124 (2014) 2028–2033, https://doi.org/10.1002/lary.24679. [17] A. Aligne, P. Auinger, R. Byrd, M. Weitzman, Risk factors for pediatric asthma, Am. J. Respir. Crit. Care Med. 162 (2000) 873–877, https://doi.org/10.1164/ajrccm. 162.3.9908085. [18] J. Gern, The urban environment and childhood asthma study, J. Allergy Clin. Immunol. 125 (2010) 545–549, https://doi.org/10.1016/j.jaci.2010.01.037. [19] M. Guarnieri, J. Balmes, Outdoor air pollution and asthma, Lancet 383 (2014) 1581–1592, https://doi.org/10.1016/S0140-6736(14)60617-6. [20] D.M. Washington, L.M. Curtis, K. Waite, M.S. Wolf, M.K. Paasche-Orlow, Sociodemographic factors mediate race and ethnicity-associated childhood asthma health disparities: a longitudinal analysis, J. Racial Ethnic Health Dispar. (2017) 1–11 https://doi.org/10.1007/s40615-017-0441-2.
25