Frequent Users of US Emergency Departments
Frequent Users of US Emergency Departments
Accurate characterisation of frequent ED users is an important step in shaping solutions to reduce cost while increasing access to, and quality of care for, this complex population. In this national study of ED utilisation, we found that frequent and super-frequent ED users in the US had lower socioeconomic status, higher rates of Medicaid coverage and presence of debilitating chronic diseases (ie, hypertension, diabetes, stroke, emphysema). This likely explains why these patients also use the primary care clinics frequently, with 85% having four or more outpatient visits during the previous 12 months. These findings are consistent with previously published studies suggesting frequent and super-frequent ED users in the US, Canada and Europe have complex medical, social and psychiatric needs, and rely heavily on other parts of the healthcare system. Here we expanded these findings to quantify the burden of chronic diseases for these patients and provided estimates that are representative of the entire US population. At this time we are unable to determine whether or not the reason for the ED visit is for an exacerbation of the patient's chronic disease, but it is possible that greater medication compliance and lifestyle modifications could improve these chronic conditions and reduce overall ED utilisation.
Challenging public perception that frequent ED users inappropriately consume ED resources, our results from the NHIS data suggest that frequent and super-frequent utilisers in the US have a perception of having poor general state of health. Unlike previous analyses, this study evaluated prevalence of specific chronic medical conditions among ED users. We were therefore not only able to demonstrate an association with self-reported poor health as previous studies have observed, but also document that frequent users do in fact have a higher rate of specific chronic medical conditions, such as emphysema, asthma, stroke and coronary artery disease. Additionally, frequent ED use was associated with outpatient mental health visits, indicating a higher prevalence of mental health conditions in this population.
Previous data from the USA suggests that adults with lower socioeconomic status, Medicaid insurance and comorbid illness had a higher rate of barriers to timely primary care access. Our results suggest that instead of abusing ED resources, many frequent and super-frequent users may visit the ED only when they are too sick to be seen in outpatient clinics, or when alternative sites of care cannot accommodate their needs within an adequate time frame. We were unable to assess for appropriateness of ED visits. However, given the aforementioned barriers to accessing primary care and other outpatient services, if these patients had the ability to access same day primary care appointments or improved access to other subspecialty services, the number of ED visits could decrease. Furthermore, programmes involving patient navigators have shown improvements in the management of chronic diseases, such as diabetes, hypertension and congestive heart failure. These types of programmes could help patients manage their chronic illnesses and potentially avoid ED visits and future hospitalisations.
Our study has several potential limitations. Although the NHIS sampling method was designed to include a nationally representative sample of US residents, it did not survey the homeless population, nursing homes, prisons or mental health facilities. We know from previously published data that these adults tend to have frequent ED utilisation. Thus our results may not be entirely generalisable to all ED users, and may underestimate the prevalence of mental illness, substance abuse and distribution of socioeconomic status among frequent users. For example, although previous studies reported a high prevalence of alcohol and drug abuse among frequent users, we found that frequent ED users in NHIS did not have a higher prevalence of alcohol or tobacco abuse. However, demographic characteristics from this sample were similar to those from US ED utilisation data.
The NHIS information is based on self-reported responses, so ED utilisation could not be confirmed and is subject to lack of response and recall bias. Additionally, similar to other survey datasets, our findings were limited by our inability to construct additional survey questions, such as incidence of recreational drug use and reasons for each visit to the ED and outpatient clinics. We were also unable to gather information on hospital admission or length of hospital stay, limiting our ability to draw any conclusions about the severity of illness. Reported associations may not be causally related and may be confounded by variables included in the analysis or unmeasured factors not included in the NHIS.
Discussion
Accurate characterisation of frequent ED users is an important step in shaping solutions to reduce cost while increasing access to, and quality of care for, this complex population. In this national study of ED utilisation, we found that frequent and super-frequent ED users in the US had lower socioeconomic status, higher rates of Medicaid coverage and presence of debilitating chronic diseases (ie, hypertension, diabetes, stroke, emphysema). This likely explains why these patients also use the primary care clinics frequently, with 85% having four or more outpatient visits during the previous 12 months. These findings are consistent with previously published studies suggesting frequent and super-frequent ED users in the US, Canada and Europe have complex medical, social and psychiatric needs, and rely heavily on other parts of the healthcare system. Here we expanded these findings to quantify the burden of chronic diseases for these patients and provided estimates that are representative of the entire US population. At this time we are unable to determine whether or not the reason for the ED visit is for an exacerbation of the patient's chronic disease, but it is possible that greater medication compliance and lifestyle modifications could improve these chronic conditions and reduce overall ED utilisation.
Challenging public perception that frequent ED users inappropriately consume ED resources, our results from the NHIS data suggest that frequent and super-frequent utilisers in the US have a perception of having poor general state of health. Unlike previous analyses, this study evaluated prevalence of specific chronic medical conditions among ED users. We were therefore not only able to demonstrate an association with self-reported poor health as previous studies have observed, but also document that frequent users do in fact have a higher rate of specific chronic medical conditions, such as emphysema, asthma, stroke and coronary artery disease. Additionally, frequent ED use was associated with outpatient mental health visits, indicating a higher prevalence of mental health conditions in this population.
Previous data from the USA suggests that adults with lower socioeconomic status, Medicaid insurance and comorbid illness had a higher rate of barriers to timely primary care access. Our results suggest that instead of abusing ED resources, many frequent and super-frequent users may visit the ED only when they are too sick to be seen in outpatient clinics, or when alternative sites of care cannot accommodate their needs within an adequate time frame. We were unable to assess for appropriateness of ED visits. However, given the aforementioned barriers to accessing primary care and other outpatient services, if these patients had the ability to access same day primary care appointments or improved access to other subspecialty services, the number of ED visits could decrease. Furthermore, programmes involving patient navigators have shown improvements in the management of chronic diseases, such as diabetes, hypertension and congestive heart failure. These types of programmes could help patients manage their chronic illnesses and potentially avoid ED visits and future hospitalisations.
Limitations
Our study has several potential limitations. Although the NHIS sampling method was designed to include a nationally representative sample of US residents, it did not survey the homeless population, nursing homes, prisons or mental health facilities. We know from previously published data that these adults tend to have frequent ED utilisation. Thus our results may not be entirely generalisable to all ED users, and may underestimate the prevalence of mental illness, substance abuse and distribution of socioeconomic status among frequent users. For example, although previous studies reported a high prevalence of alcohol and drug abuse among frequent users, we found that frequent ED users in NHIS did not have a higher prevalence of alcohol or tobacco abuse. However, demographic characteristics from this sample were similar to those from US ED utilisation data.
The NHIS information is based on self-reported responses, so ED utilisation could not be confirmed and is subject to lack of response and recall bias. Additionally, similar to other survey datasets, our findings were limited by our inability to construct additional survey questions, such as incidence of recreational drug use and reasons for each visit to the ED and outpatient clinics. We were also unable to gather information on hospital admission or length of hospital stay, limiting our ability to draw any conclusions about the severity of illness. Reported associations may not be causally related and may be confounded by variables included in the analysis or unmeasured factors not included in the NHIS.
