HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:
Guest Access | Sign In via User Name/Password

This Article Abstract Full Text (PDF) Free Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Article Archive Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (13) Citing Articles via Google Scholar Google Scholar Articles by Birnbaum, H. G. Articles by Colice, G. L. Search for Related Content PubMed PubMed Citation Articles by Birnbaum, H. G. Articles by Colice, G. L.

Economic Burden of Respiratory Infections in an Employed Population^*

^* From Analysis Group/Economics (Dr. Birnbaum and Mr. Greenberg), Boston, MA; Brandeis University (Ms. Morley), Waltham, MA; and Washington Hospital Center (Dr. Colice), Washington, DC.

Correspondence to: Howard Birnbaum, PhD, Analysis Group/Economics, 111 Huntington Ave, Tenth Floor, Boston, MA 02199; e-mail: hbirnbaum{at}analysisgroup.com

	Abstract

TOP Abstract Introduction Materials and Methods Results Discussion References

 
Context: While respiratory infections are a leading cause of morbidity, there is little information on the costs of medically treating these conditions, or on their workplace impact.

Objective: The purpose of this study was to estimate the economic burden of respiratory infections from the perspective of an employer.

Design, setting, and participants: A total of 63,890 patients with at least one diagnosis for a respiratory infection in 1997 were identified in a claims database of a national Fortune 100 company. Outcome measures were compared to those of a 10% random sample of beneficiaries in the overall beneficiary population.

Main outcome measures: The annual per capita costs for each category of respiratory infections were determined for beneficiaries of this major employer by analyzing all medical, prescription drug, and disability claims in 1997.

Results: In 1997, the total cost to the employer per patient, as well as medical-service utilization, were higher among patients with respiratory infections than among beneficiaries in the overall beneficiary population. Significant variations exist in costs across the 11 selected respiratory infections. For example, annual per capita employer expenditures for patients with respiratory infections totaled $4,397, while expenditures for patients with pneumonia and patients with acute tonsillitis/pharyngitis were $11,544 and $2,180, respectively, as compared with costs for the average beneficiary, which was $2,368.

Conclusions: Patients with respiratory infections present an important financial burden to employers. We estimate that the cost to employers of patients with respiratory infections in the United States in 1997 was $112 billion, including costs of medical treatment and time lost from work.

Key Words: direct costs • economic burden • employer costs • indirect costs • respiratory infections • work loss

	Introduction

TOP Abstract Introduction Materials and Methods Results Discussion References

 
In 1996, 208 million episodes of respiratory disorders (including infectious and noninfectious respiratory conditions, such as asthma) were reported in the United States.¹ According to the Centers for Disease Control and Prevention, in 1996, the incidence per 100 persons for respiratory infections such as pneumonia, common cold, and influenza were 1.8, 23.6, and 36.0, respectively.¹ The incidence of respiratory infections is higher among the very young and the very old.² Given the high rate of respiratory conditions, understanding the medical-resource allocation and the economic burden incurred for these patients is important for payers demanding cost-effective health care.

The focus of the research reported here is on the subset of respiratory infections, which are the leading cause of morbidity in developed countries.³ Research on the epidemiology and treatment of various respiratory infections has provided important insights into prognostic factors for these conditions.^{4 5 6} However, there is little information on the economic costs of treating respiratory infections. Furthermore, very little research exists on the indirect costs of respiratory infections. While it was reported that morbidity due to respiratory infections led to 152 days lost from school per 100 youths, and 99 days lost from work per 100 employed persons in 1996,¹ no cost estimates were provided to quantify these economic losses.

Most studies on the economic cost of treating respiratory infections have focused on the direct health-care costs. Moreover, these studies rely on data from epidemiologic surveys, including disease-specific questionnaires, and national health statistics, in addition to imputed cost estimates, to determine cost of illness, rather than using actual cost data. Rice⁷ developed this widely used methodology to estimate the national cost of illness.

One study⁸ of pneumonia found that the total direct cost, in 1995 dollars, of treating pneumonia patients < 65 years old in the United States was $3.6 billion per year. A similar study⁹ was performed for acute exacerbations of chronic bronchitis, estimating the cost of both inpatient and outpatient services. Both studies relied on incidence data for patients with community-acquired respiratory infections from a variety of national survey databases, while costs were computed using the average Medicare cost per hospital day or physician visit and applying these costs to the number of hospital and outpatient visits for patients.

Other similar studies have been conducted for respiratory conditions. For example, a study¹⁰ of chronic bronchitis and emphysema found that the total direct health-care costs for these illnesses were $11.7 billion, in 1996 dollars. This study, too, relied primarily on survey data from multiple national, state, and local sources, and costs for hospital and physician visits were derived from the Healthcare Cost and Utilization Project and national Medicare physician fee estimates in 1996. Another study¹¹ of this sort estimated the 1996 health-care costs for sinusitis to be $5.8 billion, of which $1.8 billion was for children 12 years old.

Cost-of-illness studies such as those reported above are significantly limited by their reliance on self-reported data and assigned costs per day and per visit, as opposed to actual cost data. These studies are also confined to analysis of direct health-care costs. Our research provides a broader, more complete view of the economic burden associated with the treatment of respiratory infections, from the perspective of an employer. Using claims data from a national Fortune 100 company, we are able to compile the actual employer payments for the treatment of respiratory infections. Based on these data, we report a more accurate estimate of total costs by including not only direct costs (medical and prescription drugs), but also indirect costs (disability and sporadic absenteeism), which have not been previously quantified. In addition, findings are reported on the demographic characteristics of patients with respiratory infections and their treatment patterns.

	Materials and Methods

TOP Abstract Introduction Materials and Methods Results Discussion References

 
The data come from a rich source of medical, prescription drug, and disability claims, which can be merged at the patient level, from distinct parts of one corporate benefits system. The data include all 1997 claims related to at least one of the 11 respiratory infection diagnoses (Table 1 ) for all beneficiaries (employees, spouses, dependents, and retirees) of a national Fortune 100 manufacturer. The company offers comprehensive health insurance and has a predominantly unionized workforce, 90% of whom are eligible for disability benefits. More than 100,000 beneficiaries, distributed across the country, were enrolled in one of the managed indemnity insurance plans of this company in 1997. Complete data are available for these individuals. Data for those enrolled in health maintenance organizations (who accounted for approximately 20% of enrollees) are not available, and these individuals are excluded from the sample. To ensure completeness of records, we have also excluded subjects > 65 years old in 1997, due to their enrollment in Medicare.

Direct and indirect costs are the actual cash payments by the employer, reported on an average, per-person basis annually. Our methods are similar to the approach taken by Burton and Conti¹³ and Burton et al,¹⁴ who used a "data warehouse" to analyze a range of other illnesses and document their effects on employee productivity. While data available here on the workforce burden of illness do not include measures of on-the-job productivity, they do include periods of disability and daily payments received by the employee. Data on shorter, medically related illness absences are not available, but are imputed in part, based on days when medical care was provided. If an employee was not receiving disability when medical care occurred during work days, these days are counted as sickness work-loss days in the case of hospital care, or as a half day in the case of office visits. Because the disability claims cover missed work days due to illness for periods of 6 consecutive days for eligible workers, patients with disability claims also are assigned 5 sickness work-loss days. Work-loss costs refer here to employer payments for the sum of disability plus imputed medically related absence time. Barnett et al¹⁵ provide a more complete description of these data.

Analyses were performed on the full respiratory infections sample, as well as separately on patients in each of the 11 respiratory infection diagnoses.¹⁶ Claims data for patients with respiratory infections were contrasted with the overall beneficiary population, reflected in a 10% random sample of the employer’s overall beneficiary population (ie, employees, spouses, and dependents), which included those with respiratory infections. Data are also provided on the subset of just employees with respiratory infections (n = 19,156) and a 10% sample of employees.

	Results

TOP Abstract Introduction Materials and Methods Results Discussion References

 
There were 63,890 beneficiaries with at least one claim for the treatment of at least 1 of the 11 defined respiratory infections in 1997 (Table 1) . There was considerable overlap among patients in each of the respiratory infection categories. For example, 51% of patients with a diagnosis of pneumonia also had a claim for symptoms of the respiratory system. The incidence for respiratory infections in the beneficiary population was 36%. This rate is consistent with the rate of acute respiratory conditions described for the overall US population.¹ In comparing the demographics of beneficiaries with respiratory infections to those of the overall beneficiary population, we find the gender distribution to be similar. However, consistent with the fact that the incidence of respiratory infections is higher among the very young and very old, the respiratory infections beneficiary sample contains a higher percentage of subjects < 6 years old than the overall beneficiary population (p < 0.0001; Table 2 ). Recall that subjects > 65 years old are not considered due to their Medicare enrollment. All results are reported on an annual basis for 1997.

View larger version (25K): [in this window] [in a new window] [Download PPT slide]   Figure 1.. Medical claims per beneficiary, by type of service in 1997 (respiratory infections sample, n = 63,890). p < 0.0001 for differences between employer overall population and respiratory infections sample, with the exception of "other" (right, C) [p = 0.1715]. All p values are derived from t tests and 2 for proportions.

View larger version (23K): [in this window] [in a new window] [Download PPT slide]   Figure 2.. Average number of medical claims per beneficiary, by respiratory infection. Rank ordering of selected respiratory infections reflects the number of patients with a diagnosis for each infection. Disease categories are not mutually exclusive. Patients can be in more than one category. p < 0.0001 for differences between employer overall population and each of the respiratory infections samples.

View larger version (23K): [in this window] [in a new window] [Download PPT slide]   Figure 3.. Treated respiratory infections beneficiaries and overall beneficiaries population, by cost component, for 1997 employer payments. p < 0.0001 for differences between employer overall population and respiratory infections samples (both for patients and for subgroups). All p values are derived from t tests. *Health care includes both medical-care costs and prescription drug costs.

Analyzing the distribution of medical costs across the pneumonia patient population illustrates that expenses are concentrated among a relatively small group of patients who were disproportionately hospitalized. The most costly 10% of pneumonia patients (99% were hospitalized) accounted for 59% of the total cost of the entire pneumonia population. Meanwhile, the average hospitalization rate among patients with respiratory infections was significantly lower (15%). These findings are illustrated in Figure 4 , which shows that while the employer paid > $10,000 for approximately 25% of patients with pneumonia in 1997, similar expenditures were incurred for only 11% of patients with respiratory illness overall and an even smaller percentage (5%) of the overall beneficiary population.

View larger version (13K): [in this window] [in a new window] [Download PPT slide]   Figure 4.. Distribution of 1997 employer payments: beneficiaries with pneumonia and respiratory infections vs overall beneficiary population. *Indicates the median payment. Horizontal lines indicate the 25th and 75th percentiles of payments. The lower tail of each box indicates the fifth to 24th percentile of payments. The upper tail on each box indicates the 76th to 95th percentiles of payments. The highest and lowest 5% of payments are not presented here.

	Discussion

TOP Abstract Introduction Materials and Methods Results Discussion References

Employer payments for employees with respiratory infections are almost twice those for beneficiaries in the overall beneficiary population, on average. Underlying these costs are the substantial utilization of services by patients with respiratory infections. Besides respiratory-specific claims, patients with respiratory infections submitted more claims for other types of medical services than those in the overall beneficiary population. Although medical-service utilization varied across the 11 categories of respiratory infections, in all cases respiratory infection-specific claims (ie, claims associated with a diagnosis for respiratory infections) accounted for a small percentage of total medical-service utilization.

While this data set is for a single employer, it is possible to estimate the general magnitude of the costs of patients with respiratory infections in the United States, from the perspective of an employer. To the extent that the employed population in the nation is generally similar to the study population in the incidence of the respiratory infection, as well as health-care utilization and costs, the results of this study can be used to approximate the cost of patients with respiratory infections to the entire US employed population and their dependents. Our estimate is calculated by multiplying three numbers: (1) an estimate of the employed population and their dependents in the United States, (2) the incidence of respiratory infections, and (3) the incremental cost of respiratory infections (ie, the difference between the average cost of the respiratory-infections sample and that of the overall beneficiary population). In 1997, there were approximately 154 million employees and their dependents. (This estimate is derived by multiplying the number of families with at least one employee and no retirement income [48.3 million]¹² by the average family size [3.2 members].¹⁷ ) Multiplying this population estimate by the incidence rate for the 11 respiratory conditions considered here (36%) and the incremental cost of respiratory infections ($2,029; Fig 3 ), we estimate that the cost of patients with respiratory infections to employers in the United States in 1997 was $112 billion. This cost includes all of the health-care costs of treating patients with respiratory infections irrespective of the actual condition for which the patients were treated, as well as the work-loss costs to the employer. Hence, the costs include those of treating all of the comorbidities of respiratory infection claimants, as well as the costs of treatment explicitly for respiratory infections. However, dissecting the role of comorbidities in measuring the costs for treating respiratory infections (ie, whether respiratory infections exacerbate comorbid conditions, or comorbidities increase the susceptibility to or severity of respiratory infections) is a difficult issues and a topic for future research.

While substantial, the estimated costs of respiratory infections presented here may underestimate the true burden of respiratory infections for a variety of reasons. For one, we excluded individuals > 65 years old in 1997, thereby removing from our patient sample a population with a high incidence rate of respiratory infection. Also, claims for otitis media were not included, as this is a disease primarily in the nonemployed (ie, < 6 years old) population. Additionally, sick time at home was not fully measured; we captured only that part of work loss due to illness that was associated with disability or medical treatment. Respiratory infections of dependents can also lead to workplace absences for caregivers who miss work to care for ill children and spouses. In addition, the actual payments for disability that are reported here reflect only a fraction of the total opportunity cost of the employer for workforce disruptions due to disability. Other likely workplace costs include reduced on-the-job productivity, administrative and training expenses for replacement workers, as well as days missed for sick time. Finally, there are also contagion effects from ill employees who continue to work and spread the respiratory infections among their coworkers, leading to additional work absence. Using an industry standard rule of thumb (eg, for every dollar of disability claims paid out, an employer incurs an additional $1.50 in workforce disruption) would imply that the disability costs reported here represent only 40% of the actual disability costs to the employer. Indeed, data in Watson Wyatt¹⁸ suggest that a 1.5 multiple is a conservative estimate of the likely multiplier for employer disruption costs due to disability.

Alternatively, our analyses may have overestimated total costs of respiratory infections for two reasons. First, we included claims for "symptoms of the respiratory system" (ICD9 code 786.x). This is a condition (code) that captures a variety of several respiratory symptoms and signs, such as dyspnea, cough, abnormal sputum, chest pain, and abnormal chest sounds. Including these codes is reasonable because these symptoms and signs often represent acute respiratory infections and the coding reflects a nonspecific diagnosis. However, these symptoms and signs might also have represented noninfectious chest disorders. Thus, we may have included claims for comorbid conditions that do not directly relate to respiratory infections. As noted above, unraveling the issue of how comorbidities contribute to overall costs is complex.

This analysis also highlights the variation in employer payments among types of respiratory infections. For example, although pneumonia is not common, it is an extremely expensive condition, and per-patient costs for patients with this condition are almost 50% greater than those with the next most costly condition considered here, symptoms of the respiratory system. Similarly, for certain conditions, especially pneumonia, the major cost item is inpatient services, yet a small proportion of patients drive most of the costs (Table 3 , Fig 4 ). For other conditions (eg, acute upper respiratory tract infections and tonsillitis/pharyngitis), inpatient costs are a much smaller proportion of total costs. Although these are more commonly reported conditions (Table 1) , they are still not as big an overall burden as pneumonia on the aggregate employer costs (Table 3 , overall costs). Such information may provide a basis for cost savings strategies (eg, pneumonia vaccines as a mandatory treatment).

These findings substantially increase our understanding of the costs and characteristics of patients with respiratory infections. Nevertheless, these findings are limited by the available data. First, because this study relies on insurance claims data, the findings are subject to the usual limitations of administrative data sets. Such limitations include possibly inaccurate diagnoses and incomplete assembly of claims (eg, missing bills, multiple plan coverage).¹² Second, many patients in the respiratory infections sample had a claim for more than one type of respiratory infection. Of the respiratory infections included in the analysis, patients with pneumonia were the most likely to have another respiratory infection, and they also had the highest employer payments. Seventy-seven percent of patients with pneumonia had at least one claim for another respiratory infection in 1997. In addition to comorbidity with other respiratory infections, it is also necessary to consider comorbidity with other nonrespiratory conditions. Many claims from patients with respiratory infections were not specific for the respiratory infections. For example, the risks from pneumonia and influenza may be greatest for patients with underlying comorbid conditions, such as heart disease, diabetes, and HIV infection. Unraveling the unique costs of respiratory infections relative to other comorbid conditions and the benefits of various prescription medications are topics for future research.

	Footnotes

 
Abbreviation: ICD-9 = International Classification of Diseases, ninth revision

Analysis Group/Economics (which employed Dr. Birnbaum, Dr. Greenberg, and Ms. Morley at the time of this research) received an unconditional research grant from Aventis Pharmaceuticals to support this research.

Dr. Colice has been a consultant for Aventis Pharmaceuticals but received no compensation for this research.

	References

TOP Abstract Introduction Materials and Methods Results Discussion References

 

1. Vital and Health Statistics. Current estimates from the national health interview survey, 1996. Series 10, No. 200. Atlanta, GA: Centers for Disease Control and Prevention, National Center for Health Statistics, October 1999
2. MacFarlane, JT, Colville, A, Guion, A, et al Prospective study of aetiology and outcome of adult lower-respiratory-tract infections in the community. Lancet 1993;341,511-514[CrossRef][ISI][Medline]
3. International Conference on Acute Respiratory Infections. Available at http://nceph.anu.edu.au. Accessed July 17, 2002
4. Georges, H, Leroy, O, Vandenbussche, C, et al Epidemiological features and prognosis of severe community-acquired pneumococcal respiratory infection. Intensive Care Med 1999;25,198-206[CrossRef][ISI][Medline]
5. Sethi, S Infectious etiology of acute exacerbations of chronic bronchitis. Chest 2000;117,380S-385S[Abstract/Free Full Text]
6. White, CB, Foshee, WS Upper respiratory tract infections in adolescents. Adolesc Med 2000;11,225-249[Medline]
7. Rice DP. Estimating the cost of illness. Washington, DC: US Public Health Service, May 1996; publication No. 947–6
8. Neiderman, MS, McCombs, JS, Unger, AN, et al The cost of treating community-acquired respiratory infection. Clin Ther 1998;20,820-837[CrossRef][ISI][Medline]
9. Neiderman, MS, McCombs, JS, Unger, AN, et al Treatment cost of acute exacerbations of chronic bronchitis. Clin Ther 1999;21,576-591[CrossRef][ISI][Medline]
10. Wilson, L, Devine, EB, So, K Direct medical costs of chronic obstructive pulmonary disease: chronic bronchitis and emphysema. Respir Med 2000;94,204-213[CrossRef][ISI][Medline]
11. Ray, NF, Baraniuk, JN, Thamer, JN, et al Healthcare expenditures for sinusitis in 1996: contributions of asthma, rhinitis, and other airway disorders. J Allergy Clin Immunol 1999;57,92-109
12. Birnbaum, HG, Cremieux, PY, Greenberg, PE, et al Using healthcare claims data for outcomes research and pharmacoeconomic analyses. Pharmacoeconomics 1999;16,1-8[ISI][Medline]
13. Burton, WN, Conti, DJ Use of an integrated health data warehouse to measure the employer costs of five chronic disease states. Dis Manage Health 1998;2,17-26
14. Burton, WN, Conti, DJ, Chen, CY, et al The role of health risk factors and disease on worker productivity. J Occup Environ Med 1999;41,863-867[CrossRef][Medline]
15. Barnett, A, Birnbaum, H, Cremieux, PY, et al The costs of cancer to a major employer in the United States: a case-control analysis. Am J Manage Care 2000;11,1243-1251
16. Birnbaum, H, Morley, M, Greenberg, P, et al Economic burden of pneumonia in an employed population. Arch Intern Med 2001;161,2725-2731[Abstract/Free Full Text]
17. U.S. Department of Labor Statistics. Employment characteristics of families in 1997. Available at: http://stats.bls.gov/newrels.htm. Accessed July 16, 2001
18. Staying @ work: focusing on what works; Watson Wyatt 4th Annual Employer Survey. Catalog # W-289. Bos-ton MA: Watson Wyatt. Presented at the 13th Annual National Disability Management Conference, Washington Business Group on Health, October 27–29, 1999, Washington, DC

This article has been cited by other articles:

				G. L. Colice, M. A. Morley, C. Asche, and H. G. Birnbaum Treatment Costs of Community-Acquired Pneumonia in an Employed Population Chest, June 1, 2004; 125(6): 2140 - 2145. [Abstract] [Full Text] [PDF]

				R. Gonzales A 65-Year-Old Woman With Acute Cough Illness and an Important Engagement JAMA, May 28, 2003; 289(20): 2701 - 2708. [Full Text] [PDF]

This Article Abstract Full Text (PDF) Free Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Article Archive Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (13) Citing Articles via Google Scholar Google Scholar Articles by Birnbaum, H. G. Articles by Colice, G. L. Search for Related Content PubMed PubMed Citation Articles by Birnbaum, H. G. Articles by Colice, G. L.