|Home | About | Journals | Submit | Contact Us | Français|
Pneumonia hospitalization rates are frequently reported as a measure of pneumonia disease burden in the United States (US). However, a detailed understanding of pneumonia burden in all healthcare settings, including the emergency department (ED), is essential for measuring the full effect of this disease on the population and planning and evaluating interventions to reduce pneumonia-related morbidity. The aim of this study was to quantify pneumonia-attributable ED visits in the US among children and adults during the three year period, July 2006 through June 2009.
Rates of pneumonia ED visits were calculated using the Nationwide Emergency Department Sample (NEDS), the largest source of US ED data. Pneumonia ED visits were identified using International Classification of Diseases codes within NEDS. A pneumonia ED visit was defined by a primary (first-listed) pneumonia discharge diagnosis or a secondary pneumonia diagnosis with an accompanying primary diagnosis of respiratory failure, shock, septicemia, a sign or symptom consistent with pneumonia, another acute respiratory infection, or an acute exacerbation of a chronic pulmonary disease. Population-based annual rates of pneumonia ED visits stratified by age group and geographic region from July 2006 through June 2009 were calculated. The percentages of pneumonia ED visits resulting in treat-and-release outpatient ED visits were also calculated within each age stratum.
During the study period, 6,917,025 ED visits for pneumonia were identified, representing 2.2% of all US ED visits. During the three study years, defined as July through June of 2006–2007, 2007–2008, and 2008–2009, pneumonia ED visit rates per 1,000 person-years were 7.4 (95% confidence interval [CI]: 7.0–7.8), 7.8 (95% CI: 7.3–8.2), and 7.6 (95% CI: 7.1–8.0), respectively. Annual rates were stable over the three years within each age group and geographic region. Overall, 39.3% of pneumonia ED visits, including 74.5% of pediatric and 28.1% of adult visits, were managed as treat-and-release outpatient visits.
Pneumonia accounts for 2.2% of ED visits in the US and results in approximately 7 to 8 ED visits per 1000 persons per year. A substantial proportion of pneumonia cases diagnosed in the ED are managed in treat-and-release ED outpatient visits, highlighting that enumeration of ED visit rates provides important complementary information to hospitalization rates for the assessment of pneumonia burden.
Pneumonia is one of the most common reasons for hospitalization and death in the United States (US), accounting for approximately 1.2 million hospitalizations and 56,000 deaths annually.1,2 Several national public health interventions have been introduced aimed at reducing pneumonia-related morbidity and mortality, including seasonal influenza vaccines and the new 13-valent pneumococcal conjugate vaccine (PCV-13) program.3–4 Robust measurements of pneumonia burden are essential for understanding the extent of pneumonia-related disease, evaluating the impact of these interventions, and planning for future public health programs. Historically, pneumonia hospitalization rates have been used to measure pneumonia burden.5 However, a majority of pneumonia cases are managed in out-patient settings;6 therefore, quantifying only hospitalization rates underestimates the true burden of pneumonia on the US population. With annual visits to US emergency departments (EDs) growing to over 136 million in 2009,7 the ED is an increasingly important venue for healthcare utilization both as an entry point for hospitalizations and for outpatient visits. Therefore, pneumonia-attributable ED visits are an important consideration when estimating overall pneumonia burden. There are limited published data on national, population-based rates of pneumonia-attributable ED visits. Therefore, we used the Nationwide Emergency Department Sample (NEDS),8 the largest source of ED data in the US, to estimate age-specific pneumonia ED visit rates during three consecutive years, July 2006 through June 2009.
We utilized the NEDS administrative database for calculating population-based rates of pneumonia-attributable ED visits. The study was approved by the local institutional review board as non-human research.
NEDS is an US administrative database maintained by the Agency for Healthcare Research and Quality (AHRQ) as a component of the Healthcare Cost and Utilization Project.8 NEDS is the largest source of US ED data, and contains information from 25–30 million ED visits to non-federal hospitals annually starting in 2006. NEDS includes data from 29 states and approximately 20% of US ED visits annually. NEDS has a complex sampling design that allows calculation of national estimates.8 NEDS includes a sampling weight for each observation and is stratified by geographic region, trauma center designation, urban-rural status, teaching hospital status, and hospital ownership. EDs represent the primary sampling units and all visits from sampled EDs are included.8 For each ED visit, NEDS contains up to 15 diagnoses coded according to the International Classification of Diseases, Ninth Revision Clinical Modification (ICD-9 CM). Additionally, NEDS includes data for more than 50 variables for each ED visit, including: month and year of visit, patient disposition from the ED, patient demographics, and hospital characteristics, including hospital location according to geographic region of the country (Northeast, Midwest, South, West).
For this study, a pneumonia ED visit was defined as an ED visit with a primary (first-listed) pneumonia diagnosis (ICD-9-CM codes: 480.xx – 486.xx, 487.0), or with a secondary pneumonia diagnosis (listed in diagnosis fields 2–15) with an accompanying primary diagnosis of respiratory failure, shock, septicemia, a sign or symptom consistent with pneumonia, another acute respiratory infection, or an acute exacerbation of a chronic pulmonary disease (Supplemental Table 1). The case definition of a pneumonia ED visit was limited to visits with a primary diagnosis of pneumonia or a secondary diagnosis of pneumonia with a closely related infectious or pulmonary primary diagnosis in an effort to select cases in which pneumonia was the driving factor for ED evaluation.
NEDS data were used to estimate the number of pneumonia ED visits nationwide for three years, July 2006 through June 2009. This time frame was divided into three 12–month periods (July 2006–June 2007; July 2007–June 2008; and July 2008–June 2009) to capture respiratory infection seasons, typically November through April, which span calendar years. Weighted pneumonia ED visit counts were calculated using the NEDS sampling design for eight age groups: <2 years; 2–4 years; 5–17 years; 18–39 years; 40–64 years; 65–74 years; 75–84 years; and ≥ 85 years. Patients <18 years old were considered children, while those ≥18 years old were classified as adults. Visits resulting in ED discharge without hospitalization (“treat-and-release”) were considered outpatient ED visits. Age group-specific annual (July–June) rates of pneumonia ED visits were calculated by dividing the weighted annual count of pneumonia ED visits by July population estimates from the US Census Bureau9 and expressed as ED visits per 1,000 person-years. Monthly rates were calculated by dividing monthly weighted counts of pneumonia ED visits by July population estimates, dividing by 1,000, and multiplying by the proportion of days in a year within each month to express them as annualized rates per 1,000 person-years. Rates were also separately calculated for US geographic regions (Northeast, Midwest, South, West), as classified by the US Census Bureau. Statistical analyses accounted for the NEDS sampling design, including the weighting factor, the stratification variable, and the EDs as the primary sampling units. Analyses were conducted per AHRQ recommendations8 and using SAS 9.2 (Cary, NC) and Stata 12 IC (College Station, TX) survey packages.
From July 2006 to June 2009, there were an estimated 6,917,025 ED visits in the US with a primary ICD-9 diagnosis of pneumonia or a secondary diagnosis of pneumonia with an accompanying primary diagnosis of sepsis, respiratory failure or another acute respiratory condition. These visits, defined as pneumonia-attributable ED visits in this study, accounted for 2.2% of all ED visits in the three year study period. Pneumonia was listed as the first ICD-9 diagnosis in 5,221,115 (75.5%) of these visits. The percentage of overall ED visits attributed to pneumonia ranged from 0.7% in the 18–39 year old age group to 6.6% in those ≥ 85 years old (Supplemental Table 2). Overall, 2,719,545 (39.3%) pneumonia ED visits resulted in ED discharge without hospital admission (treat-and-release outpatient ED visits), including 74.5% of pediatric and 28.1% of adult visits. Among adults, increasing age was associated with higher proportions of ED visits leading to hospitalization (Supplemental Table 2). Medicare or Medicaid was the primary payer for 4,457,627 (64.6%) pneumonia ED visits. Patients residing in zip codes with a median household income in the lowest national quartile accounted for 31.0% of pneumonia ED visits, compared to 17.6% of visits among those in the highest income quartile.
During 2006–2007, 2007–2008, and 2008–2009, overall pneumonia ED visit rates per 1,000 person-years were 7.4 (95% confidence interval [CI]: 7.0 to 7.8), 7.8 (95% CI: 7.3 to 8.2), and 7.6 (95% CI: 7.1 to 8.0), respectively (Table). Rates varied by geographic region, with the lowest rates in the Western US and the highest in the Midwest. Rates were highest at the extremes of age and in the winter months, with consistent peaks during December through February each year (Supplemental Figure).
Pneumonia represents an enormous public health problem in the US.1–2 Although many reports have focused on pneumonia hospitalizations,5 a comprehensive quantification of pneumonia burden in all healthcare settings is necessary to appreciate the full impact of this disease. Due to high and increasing ED utilization in the US,7 consideration of ED visits is essential when estimating pneumonia disease burden in the US. Since 2006, AHRQ has compiled detailed information about US ED utilization in the NEDS database.8 We used this new data source to estimate rates of pneumonia ED visits from July 2006 through June 2009. Our results demonstrate consistent year-to-year rates of pneumonia ED visits within all age groups and geographic regions. Furthermore, our study revealed that 39.3% of all pneumonia ED visits resulted in ED discharge. These outpatient pneumonia ED visits, which are not captured in evaluations of pneumonia burden based on hospitalization rates, are important to consider when evaluating the impact of public health interventions aimed at reducing pneumonia burden, such as the recently-implemented PCV-13 immunization program.3
Using data from the National Hospital Ambulatory Medical Care Survey (NHAMCS) from 1993 through 2008, Neuman et al10 reported an overall rate of 5.2 (95% CI: 4.8 to 5.6) pneumonia ED visits per 1000 person-years for adults and found increasing rates with increasing patient age. Our study used a newly-available and substantially larger data source. For 2009, NEDS included data from 964 hospitals and 28 million ED visits, whereas NHAMCS included data from 356 hospitals and approximately 35,000 ED visits.7,8 Our study adds to this literature with updated and precise estimates of pneumonia ED visit rates for well-defined age groups including both children and adults and demonstrate the substantial pneumonia burden that is managed via treat-and-release outpatient ED visits.
Identification of pneumonia ED visits was dependent on ICD-9 coding. The optimal case definition using ICD-9 codes to retrospectively identify pneumonia-attributable ED visits is unknown. In addition to ED visits with a primary diagnosis of pneumonia, we included visits with a secondary diagnosis of pneumonia with selected primary diagnoses to capture cases with a primary diagnosis reflecting a consequence of pneumonia, such as sepsis or shortness of breath; these visits accounted for 24.5% of pneumonia ED visits included in this study. All ED visits with pneumonia listed in any ICD-9 diagnosis position regardless of the primary diagnosis were not included in an attempt to avoid visits in which a non-pneumonia illness was the major reason for ED presentation and pneumonia was listed as a secondary diagnosis not directly related to the ED evaluation. Accuracy of the case definition used in this study for identifying visits with pneumonia as the primary reason for ED presentation is unknown. If visits primarily for pneumonia were coded with pneumonia ICD-9 codes outside the first-listed diagnosis field in NEDS, rates reported in this study underestimate the true ED burden of pneumonia. NEDS does not contain individual identifiers, and a patient may have contributed more than one pneumonia ED visit in our calculations. Finally, NEDS does not include ED visits from federal or military hospitals, and thus, our calculated rates may underestimate the true pneumonia burden in the US population.
The rate of pneumonia ED visits in the US population during the three consecutive years including the 2006–2008 respiratory infection seasons was 7 to 8 visits per 1,000 person-years. Approximately 39% of pneumonia cases diagnosed in the ED were managed on a treat-and-release outpatient basis; therefore, evaluating the rate of pneumonia ED visits provides important complementary information to pneumonia hospitalization rates for the study of pneumonia burden in the US.
Funding/Support: This study was supported in part by Intergovernmental Personnel Agreements from CDC (11-IPA1110211 and 12-IPA1210402). Dr. Self received salary support from the grant KL2TR000446 from the National Center for Advancing Translational Sciences.
Role of the Sponsor: CDC participated in review and approval of the manuscript; CDC had no role in the design or conduct of the study, or the collection, management or analysis of study data. The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the CDC.
We thank Matthew R. Moore, MD, MPH, and Cynthia G. Whitney, MD, MPH, in the Respiratory Diseases Branch at CDC for critical review of the manuscript.