The state and county estimates displayed in this web site were derived from the U.S. costs of asthma in the current literature.1, 2 Cost estimates are based on estimates of prevalence. Yet, data on asthma prevalence does not exist at the state and county level. For purposes of this web site, national rates of asthma prevalence were applied to state/county census data‑‑taking into account key demographic characteristics specifically, age, gender, and race—to come up with local prevalence estimates. National estimates of the cost of asthma per adult/child were then applied to the local area prevalence rates to obtain state/county level cost estimates. For all calculations, asthma was defined as code 493 of the International Classification of Disease, ninth revision (ICD‑9). Unless specified otherwise, the calculations were based on 1994 data.
The following is a more detailed discussion of the methodology behind the calculations.
Population and Prevalence
National estimates of asthma prevalence were calculated from data sets of the National Health Interview Survey (NHIS). The NHIS is an annual, self‑reported, home‑interview of a sample of the U.S. population (proxies are used for persons under 19 years of age). 3 Persons with asthma were defined as those respondents who reported having asthma in the past 12 months. Since the NHIS surveys for asthma prevalence in one sixth of their sample per year, several years of data (1993‑1995) were averaged in order to obtain stable figures. In this manner, asthma prevalence estimates were calculated for 42 different combinations of age, gender and race (e.g. white males age 18‑24).
County estimates of the 1994 population by gender, race, and age group were obtained from the U.S. Census Bureau and aggregated into the same 42 demographic groups.4 Local prevalence rates were then calculated as a simple ratio of prevalence and population.
Local Area Estimates of Costs
With the prevalence data in hand, asthma‑related costs of illness were then derived for both direct and indirect medical expenditures.5 Direct medical expenditures are composed of charges for inpatient and outpatient hospital services, emergency room services, physician services (both inpatient care and office visits), and medications. Indirect costs include the value of time lost from school and work as a result of asthma morbidity and mortality.
The local cost estimates (direct and indirect) are a product of the local prevalence rates (described in the section above) multiplied by the national estimates of the cost of asthma for either adults or children (under 18 years).
Demographic and Cost Estimation for Places
In order to estimate the costs of asthma for cities with populations of 100,000 or more it was necessary to first obtain estimates of the 1994 populations of these cities stratified into the 42 demographic groups used in the main analysis. As these were not available from third party sources, estimates were developed from available sources by interpolating between 1990 US Census data and commercially developed 1999 estimates . Estimates were developed for the 200 incorporated cities and census defined places (CDPs) with 1990 populations of 100,000 or more. For 1990, census block group level population counts were aggregated to the city or CDP level for the 42 demographic groups. The 1999 population estimates were already aggregated to the city or CDP level, however, it was necessary to disaggregate the hispanic population into its white, black, and other race components. This was done on the basis of the proportions of these racial groups in the hispanic population in each individual city or CDP, in 1990. The 1999 population estimates where then aggregated into the 42 demographic groups. For each group, the 1994 population was estimated as the 1990 population plus 4/9ths of the difference between 1990 and 1999 population values.
The estimated costs of asthma for each place was calculated as the sum of the national per capita costs for the 42 demographic groups weighted by the estimated populations for each group.
(1. Both set of data were obtained from Applied Geographic Solutions.)
Due to insufficient data at the local level, three key assumptions guided the calculating of the state/county estimates.
(1) The prevalence of asthma within the 42 demographic groups (defined by age, gender, and race) is constant across the country.
(2) Mean resource utilization per person with asthma is constant across the country (adults and children considered separately). Resource utilization includes obvious components, such as medications and hospital facilities, but also includes less obvious components, such as mortality.
(3) Wage and cost rates are constant across the country. This includes wage rates of health care providers as well as the general population.
There are a number of factors that limit the usefulness of the data presented in this web site. These include the timeliness of the data, the exclusion of many sources of local variation, and restriction to point estimates without estimates of error. Therefore, these estimates should only be used for broadly comparative purposes‑‑to access the relative magnitude of asthma prevalence and cost.
Many of the estimates required multiple years of data to provide a stable figure. At the initiation of this project, 1994 was the most recent year for which data on the various components were available.
The data presented in these web site tables adapts national estimates to local conditions based explicitly on local demographic variation. It is recognized that local conditions may vary across other dimensions. The cost and wage estimates do not reflect state, regional, or urban/rural differences which may or may not be significant. The prevalence rates were also calculated on the basis of demographic groups defined at the national level; local variations were not controlled for in this process. For example, several published reports suggest that certain very low‑income urban populations have rates of asthma prevalence that exceed the rates used in this analysis. It would be reasonable to conclude that the asthma prevalence may be underestimated for counties with large urban populations of low income.
Accuracy and reliability of the data
The data for both costs and prices were obtained from national sources thought to be most reliable and commonly used for this type of economic analysis. However, most of the published data provide only point estimates and include no method for determining confidence intervals. Without measures of standard error for the cost estimates, it is not possible to assess error quantitatively.
Notes on the National Estimates of Costs
Whereas the local cost estimates were only calculated at the level of direct and indirect costs, national estimates were calculated for the major components that comprise these major categories. What follows is a brief description of the methodology behind each component of the national cost estimates.
The data on health care utilization, morbidity and mortality came from surveys conducted by the National Center for Health Statistics (NCHS), of the Centers for Disease Control and Prevention.
Direct costs are comprised of hospital care, both inpatient and outpatient, emergency care, physician care, and medications.
Asthma hospitalizations were identified according to the first‑listed discharge diagnosis provided in the National Hospital Discharge Survey (NHDS). 6 Cost estimates of hospital inpatient care were calculated by multiplying the number of asthma‑related hospitalization days reported in the NHDS by the figure for adjusted expenses per inpatient day as reported by the American Hospital Association. 7 ,8
Emergency room and hospital outpatient utilization data were obtained from the NHIS survey. The emergency room and hospital outpatient cost information came from the 1987 National Medical Expenditure Survey‑2 (NMES‑2), projected to 1994 dollars (note: this survey estimates charges/expenditures rather than actual costs of care). 9 For emergency room estimates, the number of emergency room visits reported in the NHIS was multiplied by the average charge per visit from the NMES data (adjusted to 1994 dollars). Similarly, for hospital outpatient estimates, the number of visits as reported in the NHIS was multiplied by the average charge per visit per the NMES data (adjusted to 1994 dollars).
Utilization data on ambulatory care in the non‑hospital outpatient setting came from the National Ambulatory Medical Care Survey (NAMCS).10, 11 This survey studies a sample of non‑federal, office‑based physicians. As with the hospitalization data, asthma visits were identified using the first‑listed diagnosis. The cost estimates for physicians' office visits were calculated by multiplying the number of asthma visits (according to physician specialty) reported in the NAMCS, by the charge per new or established patient visit based on data from the American Medical Association's national survey of physicians. 12 ,13
For costs associated with physician inpatient services, it was assumed that an inpatient visit occurred with each hospital day. Therefore, estimates of physicians' inpatient services were calculated by multiplying the number of visit days by the charge per initial or follow‑up hospital visit based on data from the American Medical Association's national survey of physicians.
Medication costs were calculated by multiplying the average wholesale prescription price for each class of asthma drugs, by an average annual dose based on prescription mentions in the NAMCS. The wholesale prescription prices were obtained from the 1995 Drug Topics Red Book. 14 , 15
The NHIS also provides information on the burden of illness related to time lost from either school or work (defined to include both outside work and housekeeping).
In calculating costs related to loss of school days, the child's mother was assumed to be the caretaker. NHIS data on the number of school days lost were equated with the number of days lost from work for the primary caretaker. Costs were derived as the value of time lost from either outside employment (based on average annual earnings) or housekeeping (based on the imputed value of housework).16 ,17
Loss of work calculations were based on NHIS information about the number of workdays lost. Costs were derived as the value of time lost from either outside employment (based on average annual earnings) or housekeeping (based on the imputed value of housework).
Asthma mortality was defined based on underlying cause of death as listed in the U.S. vital records. Unlike the NCHS surveys that are based on samples, the vital records contain information on the entire U.S. population. 18, 19 In calculating the cost estimates, the number of asthma deaths was multiplied by the age‑ and sex‑specific current values of estimated lifetime earnings discounted to current value (3 percent for the 1994 estimates).20
Note: The assumptions associated with calculating these national estimates appear in the published literature. 1 , 2
Updating 1994 Estimates to 1998
The 1998 update projects the 1994 findings forward on the basis of some new 1998 data. Specifically, 1998 to 1994 ratios of cost components were applied to 1994 estimates of the total cost components for adults and children. 2, 21 The inflated total costs were divided by the estimated total number of children and adults with asthma to yield an updated estimate of cost per person with asthma. US Census estimates for the population in 1998 were used to update the estimates for the total US population, and for the county and place level estimates of the 42 demographic groups (defined by age, gender, and race). The 1994 estimates of incidence rates for the 42 demographic groups were not updated.
1. Weiss KB, Gergen PJ, Hodgson TA. An economic evaluation of asthma in the United States. New Engl J Med 1992, 326: 862‑866.
2. Weiss KB, Sullivan SD, Lyttle CS. Trends in the costs of asthma in the United States, 1985‑1994. J Allergy Clin Immunol 2000;106:493-9.
3. National Center for Health Statistics. Current estimates from the National Health Interview Survey, United States 1983, 1987, 1994. Vital and health statistics. Series 10. Nos. 154, 166, 193. Washington, D.C.: Government Printing Office, 1983, 1987, 1996.
4. U.S. Census Bureau. Estimates of the Population of Counties by Age, Sex, Race and Hispanic Origin: 1990 to 1998. Internet Release date: September 15, 1999.
5. Rice DP, Hodgson TA, Kopstein AN. The economic costs of illness: a replication and update. Health Care Financ Rev 1985;7(1):61‑80.
6. National Center for Health Statistics. Utilization of short‑stay hospitals, United States, 1983,1984,1985, 1987, 1994, annual summary. Vital and health statistics. Series 13. Nos. 83,84,91,96,99, 128. Washington, D.C.: Government Printing Office 1983‑1987, and 1997.
7. American Hospital Association. Hospital statistics, 1986 ed. Chicago: American Hospital Association, 1986.
8. American Hospital Association. Hospital statistics: emerging trends in hospitals, 1995‑96 ed. Chicago: American Hospital Association, 1995.
9. Smith DH, Malone DC, Lawson KA, Okamoto LJ, Battista C, Saunders WB. A national estimate of the economic costs of asthma. Am J Respir Crit Care Med 1997;156(3 Pt 1):787‑93.
10. National Center for Health Statistics, Nelson C, McLemore T. The National Ambulatory Medical Care Survey: United States, 1975‑81 and 1985 trends. Vital and health statistics. Series 13. No. 93 Washington, D.C.: Government Printing Office, 1988. (DHHS publication no. (PHS)88‑1754.)
11. National Center for Health Statistics. Ambulatory care visits for asthma: United States, 1993‑94. Advance data from vital and health statistics report no. 277. Washington DC: Government Printing Office, 1996. (DHHS publication no. (PHS) 96‑1250).
12. Gonzalez, ML, Emmons DW, Slora EJ. Socioeconomic characteristics of medical practice 1988. Chicago: American Medical Association, 1988.
13. Gonzalez, ML Socioeconomic characteristics of medical practice 1995. Chicago: American Medical Association, 1995.
14. Drug topics red book: annual pharmacists' reference. Oradell, N. J.: Medical Economics 1989.
15. Drug topics red book: annual pharmacists' reference. Oradell, N. J.: Medical Economics 1995.
16. Max W, Rice DP, MacKenzie EJ. The lifetime cost of injury. Inquiry 1990;27:332‑43.
17. Personal Communications, D. Rice. 1994 estimates of earnings, imputed value of housework, and current value of estimated future lifetime earnings.
18. National Center for Health Statistics. Vital statistics of the United States, 1985. Vol. 2. Mortality. Parts A and B. Washington, D.C.: Government Printing Office, 1988, (DHHS publication nos. (PHS)88‑1101(2)).
19. National Center for Health Statistics. Advance report of final mortality statistics, 1994. Vol. 45, No. 3 supplement. Washington DC: Government Printing Office, 1996. (DHHS publication no. (PHS) 96‑1120).
20. Lipscomb J, Weinstein MC, Torrance GW. Time preference. In: Gold ME, Siegel JE, Russell LB, Weinstein MC, eds. Cost‑effectiveness in health and medicine. New York: Oxford University Press, 1996:214‑246.
21. Weiss KB, Sullivan SD. The health economics of asthma and rhinitis. I. Assessing the economic impact. J Allergy Clin Immunol 2001;107:3-8.