Costs of Illness for Environmentally
Related Health Effects
in Older Americans
Final Report
Prepared for
Kathy Sykes
Senior Advisor, Aging Initiative
Office of Children's Health Protection
EPA (Mail Code 1107A)
1200 Pennsylvania Ave. NW
Room 2512N (Ariel Rios North)
Washington, DC 20460
(202) 564-3651
FAX (202) 564-2733
Prepared by
George Van Houtven
Amanda Honeycutt
Boyd Gilman
Nancy McCall
Wanda Throneburg
RTI International
Health, Social, and Economics Research
Research Triangle Park, NC 27709
RTI Project Number 08687.029.001
April 2005
HRTI
INTERNATIONAL
-------
RTI Project Number
08687.029.001
Costs of Illness for Environmentally
Related Health Effects
in Older Americans
Final Report
April 2005
Prepared for
Kathy Sykes
Senior Advisor, Aging Initiative
Office of Children's Health Protection
EPA (Mail Code 1107A)
1200 Pennsylvania Ave. NW
Room 2512N (Ariel Rios North)
Washington, DC 20460
(202) 564-3651
FAX (202) 564-2733
Prepared by
George Van Houtven
Amanda Honeycutt
Boyd Gilman
Nancy McCall
Wanda Throneburg
RTI International*
Health, Social, and Economics Research
Research Triangle Park, NC 27709
*RTI International is a trade name of Research Triangle Institute.
-------
Table of Contents
Section Page
1. Introduction 1-1
2. Background 2-1
2.1 Selection of Health Effects 2-1
2.2 Overview of the COI Approach 2-3
2.3 COI Literature Review 2-3
2.4 Discussion 2-4
3. Cost Analysis Data and Methods 3-1
3.1 Direct Costs 3-1
3.1.1 Medical Costs 3-1
3.1.2 Self-Administered Prescription Drug Costs 3-3
3.1.3 Nursing Home Costs 3-4
3.2 Indirect Costs 3-5
3.2.1 Increased Morbidity 3-5
3.2.2 Premature Mortality 3-6
4. Cost-of-Illness Estimates for Selected Conditions in Older Americans 4-1
4.1 Chronic Lung Disease 4-1
4.1.1 Direct Costs: Medical 4-1
4.1.2 Direct Costs: Self-Administered Prescription Drugs and Nursing Home
Care 4-1
4.1.3 Indirect Costs: Morbidity 4-3
4.1.4 Indirect Costs: Mortality 4-3
4.2 Ischemic Heart Disease 4-4
4.2.1 Direct Costs: Medical 4-4
4.2.2 Direct Costs: Self-Administered Prescription Drugs and Nursing Home
Care 4-5
4.2.3 Indirect Costs: Morbidity 4-6
4.2.4 Indirect Costs: Mortality 4-7
4.3 Stroke 4-7
4.3.1 Direct Costs: Medical 4-8
4.3.2 Direct Costs: Self-Administered Prescription Drugs and Nursing Home
Care 4-8
4.3.3 Indirect Costs: Morbidity 4-9
4.3.4 Indirect Costs: Mortality 4-10
4.4 Lung Cancer 4-10
4.4.1 Direct Costs: Medical 4-11
4.4.2 Direct Costs: Self-Administered Prescription Drugs and Nursing Home
Care 4-11
4.4.3 Indirect Costs: Morbidity 4-12
4.4.4 Indirect Costs: Mortality 4-13
4.5 Pneumonia 4-13
4.5.1 Direct Costs: Medical 4-14
iii
-------
4.5.2 Direct Costs: Self-Administered Prescription Drugs and Nursing Home
Care 4-14
4.5.3 Indirect Costs: Morbidity 4-16
4.5.4 Indirect Costs: Mortality 4-16
4.6 Gastrointestinal Illness 4-16
4.6.1 Direct Costs: Medical 4-17
4.6.2 Direct Costs: Self-Administered Prescription Drugs and Nursing Home
Care 4-17
4.6.3 Indirect Costs: Morbidity 4-18
4.6.4 Indirect Costs: Mortality 4-19
5. Discussion of Results 5-1
6. References 6-1
Appendixes
A: Summaries of Methods, Data Sources, and Findings by Disease A-l
B: Detailed Methods for Estimating Missed Work Days and Bed Days Attributable to
Stroke, Heart Disease, and Chronic Lung Disease B-l
C: Annual Medicare Payments in 2000 by Region and State C-l
iv
-------
List of Tables
Number Page
1-1 Estimated Aggregate Annual Costs of Illness Among Older Americans for Selected
Health Conditions in 2000 (in millions of 2000$) 1-2
2-1 Selected Health Conditions (by ICD-9 Code) 2-2
4-1 Costs of Illness Associated with Chronic Lung Disease—Estimated Aggregate Costs in
2000 for Individuals 65 and Older 4-1
4-2 Direct Medical Costs Associated with Chronic Lung Disease—Number of Beneficiaries
and Average Medical Payments by Type of Service and Age-Gender Category 4-2
4-3 Other Direct Costs Associated with Chronic Lung Disease—Incremental Cost Estimates
for Prescription Medications and Nursing Home Services by Age-Gender Category 4-2
4-4 Morbidity-Related Indirect Costs Associated with Chronic Lung Disease—Estimated
Annual Per-Person Productivity Losses by Age-Gender Category 4-3
4-5 Mortality-Related Indirect Costs Associated with Chronic Lung Disease—Estimated
Aggregate Productivity Losses by Age Category 4-4
4-6 Costs of Illness Associated with Ischemic Heart Disease—Estimated Aggregate Costs in
2000 for Individuals 65 and Older 4-4
4-7 Direct Medical Costs Associated with Ischemic Heart Disease—Number of Beneficiaries
and Average Medical Payments by Type of Service and Age-Gender Category 4-5
4-8. Other Direct Costs Associated with Ischemic Heart Disease—Incremental Cost Estimates
for Prescription Medications and Nursing Home Services by Age-Gender Category 4-6
4-9 Morbidity-Related Indirect Costs Associated with Ischemic Heart Disease—Estimated
Annual Per-Person Productivity Losses by Age-Gender Category 4-6
4-10 Mortality-Related Indirect Costs Associated with Ischemic Heart Disease—Estimated
Aggregate Productivity Losses by Age Category 4-7
4-11 Costs of Illness Associated with Stroke—Estimated Aggregate Costs in 2000 for
Individuals 65 and Older 4-7
4-12 Direct Medical Costs Associated with Stroke—Number of Beneficiaries and Average
Medical Payments by Type of Service and Age-Gender Category 4-8
4-13 Other Direct Costs Associated with Stroke—Incremental Cost Estimates for Prescription
Medications and Nursing Home Services by Age-Gender Category 4-9
4-14 Morbidity-Related Indirect Costs Associated with Stroke—Estimated Annual Per-Person
Productivity Losses by Age-Gender Category 4-9
4-15 Indirect Costs Resulting from Increased Mortality Associated with Stroke—Estimated
Aggregate Productivity Losses by Age Category 4-10
V
-------
4-16 Costs of Illness Associated with Lung Cancer—Estimated Aggregate Costs in 2000 for
Individuals 65 and Older 4-10
4-17 Direct Medical Costs Associated with Lung Cancer—Number of Beneficiaries and
Average Medical Payments by Type of Service and Age-Gender Category 4-11
4-18 Other Direct Costs Associated with Lung Cancer—Incremental Cost Estimates for
Prescription Medications and Nursing Home Services by Age-Gender Category 4-12
4-19 Morbidity-Related Indirect Costs Associated with Lung Cancer—Estimated Annual Per-
Person Productivity Losses by Age-Gender Category 4-13
4-20 Mortality-Related Indirect Costs Resulting Associated with Lung Cancer—Estimated
Aggregate Productivity Losses by Age Category 4-13
4-21 Costs of Illness Associated with Pneumonia—Estimated Aggregate Costs in 2000 for
Individuals 65 and Older 4-14
4-22 Direct Medical Costs Associated with Pneumonia—Number of Beneficiaries and
Average Medical Payments by Type of Service and Age-Gender Category 4-15
4-23 Other Direct Costs Associated with Pneumonia—Incremental Cost Estimates for
Prescription Medications and Nursing Home Services by Age-Gender Category 4-15
4-24 Mortality-Related Indirect Costs Associated with Pneumonia—Estimated Aggregate
Productivity Losses by Age Category 4-16
4-25 Costs of Illness Associated with Gastrointestinal Illness—Estimated Aggregate Costs in
2000 for Individuals 65 and Older 4-16
4-26 Direct Medical Costs Associated with Gastrointestinal Illness—Number of Beneficiaries
and Average Medical Payments by Type of Service and Age-Gender Category 4-17
4-27 Other Direct Costs Associated with Gastrointestinal Illness—Incremental Cost Estimates
for Prescription Medications and Nursing Home Services by Age-Gender Category 4-18
4-28 Mortality-Related Indirect Costs Associated with Gastrointestinal Illness—Estimated
Aggregate Productivity Losses by Age Category 4-19
vi
-------
1. Introduction
With the U.S. population expected to age rapidly over the next half century, environmental health
effects among older Americans are becoming an increasingly important public health concern. Forecasts
suggest that the elderly population is likely to grow to 70 million people by 2030, and the number of
individuals over 85 years is expected to increase to 19 million by 2050—an almost fivefold increase
between 2000 and 2050.
For several reasons, a rapidly growing elderly population presents unique challenges in the area
of environmental health. First, because of diminished immunity associated with aging, older adults are
often more susceptible than younger adults to environmental hazards. Second, the prevalence of diseases
that can further compromise immunity, such as Alzheimer's and diabetes, tend to be highest among the
elderly. Finally, older persons have accumulated a lifetime of exposures that persist in the body and may
manifest as health problems long after the time period of exposure.
The U.S. Environmental Protection Agency's (EPA's) Aging Initiative has been established
precisely to address these challenges. EPA launched the Aging Initiative in 2002 to "prioritize and study
environmental health hazards to older persons and examine the effect that a rapidly growing aging
population will have on our environment" (EPA, 2003). Through this initiative, the Agency is developing
a national agenda on the environment and aging, with efforts focused on research that could lead to
developing and implementing policies to better protect older Americans from the health effects of
environmental exposures. One such research effort is to develop a better understanding of the burden of
disease associated with environmental exposures among the elderly.
The purpose of this report is to contribute to this research effort by assessing the economic burden
of specific illnesses among the elderly. The analysis focuses on health conditions for which
environmental exposure are known or suspected to be an important contributing factor. In particular, we
analyzed costs for specific health conditions within the following six illness categories:
• chronic respiratory disease • lung cancer
• heart disease • pneumonia
• stroke • gastrointestinal illness
In Section 2, we provide a more detailed description of the specifically selected health conditions,
and we also discuss the main reasons for selecting these conditions. The discussion in Section 2 includes
a summary of the evidence linking these conditions with environmental exposures.
To assess the economic burden of the selected illnesses among the elderly, we applied a cost-of
illness (COI) approach. This approach is well established in the field of health valuation and has been
widely applied to assess losses for a variety of illnesses. In Section 2 we describe the conceptual
framework underlying COI and the primary issues associated with conducting this type of analysis. We
also summarize the evidence from previous COI research related to these conditions. Although we
identified over 20 articles published since 1990 that have estimated costs for these (or related) conditions,
the methods and results of these studies varied widely and few of them specifically focused on health
effects in older Americans. Our analysis was designed to address the limitations and gaps in this
literature.
By applying a consistent set of COI methods, data sources, and disease classifications and by
limiting our attention to older Americans, our analysis provides estimates of the burden of illness that are
directly comparable across conditions and are targeted to the population of interest for EPA's Aging
Initiative. Based on available data, we estimated the main health care costs (direct costs) and productivity
losses (indirect costs) associated with the prevalence of the six conditions among adults 65 years and
older in 2000. In Section 3, we describe the data and the methods we used to construct our cost estimates.
To estimate direct costs associated with each condition, we primarily relied on Medicare claims data from
a nationally representative sample of Medicare fee-for-service beneficiaries. We supplemented these
direct cost estimates using national survey data to estimate incremental prescription drug (self-
1-1
-------
administered) and nursing home costs. To estimate indirect costs, we combined national health and
earnings data to estimate both morbidity and mortality related productivity losses for the 65 and older
population. Section 3 also describes the main limitations and caveats associated with these data and
methods, which must be kept in mind when interpreting the results.
It is important to emphasize that the cost estimates developed with these approaches should not
be interpreted as those specifically attributable to environmental exposures. Unfortunately, the science
and empirical evidence regarding the epidemiological links between environmental exposures and these
health outcomes are not sufficiently advanced to reliably estimate this attributable fraction.
Consequently, the results are more appropriately interpreted as upper-bound estimates of environmentally
related costs of illness for these conditions.
Section 4 describes the results separately for each of the six health categories. We break down
medical costs by type of medical service (e.g., inpatient, physician visits, hospital outpatient) and by age
and gender category. Estimates of prescription drug and nursing home costs are separately reported by
age and gender category. The indirect costs are divided into losses associated with morbidity and those
due to mortality, and they are also subdivided according to age and gender category.
Table 1-1 summarizes
Table 1-1. Estimated Aggregate Annual Costs of Illness
Among Older Americans for Selected Health
Conditions in 2000 (in millions of 2000$)
our results by reporting
estimated aggregate direct and
indirect costs for each of the six
illness categories. Comparisons
of these aggregate cost
estimates across illness
categories must be made with
caution. Due to data
limitations, certain components
of direct and indirect costs
could not be reliably estimated
for different illnesses (see
footnotes to Table 1-1).
Nevertheless, the results clearly
suggest that chronic lung
disease and ischemic heart
disease are the two categories of
illness with the largest
aggregate costs in 2000. The
point estimate for chronic lung
Chronic Lung
Ischemic Heart
Disease
Disease
Stroke
Direct Costs
27,294
42,239
ll,840a
Indirect Costs
6,605
7,913
6,292
Total Costs
33,898
50,152
18,133
Gastrointestinal
Lung Cancer
Pneumonia
Illness
Direct Costs
4,277b
10,936
l,006a
Indirect Costs
173c
4.7d
0.039d
Total Costs
4,450
10,941
1,006
aDoes not include nursing home costs.
bDoes not include self administered prescription drug or nursing home costs.
cDoes not include lost household productivity for morbidity.
dDoes not include any morbidity related productivity losses.
disease exceeds $35 billion and, even without nursing home costs included, the point estimate for
ischemic heart disease is close to $52 billion. Gastrointestinal illness is estimated to impose the lowest
aggregate cost of the six conditions, but even without estimates of nursing home or morbidity related
productivity losses, its aggregate costs are estimated to exceed $1 billion. For each of the six conditions,
direct medical costs comprise the largest component of estimated aggregate costs.
The numbers reported in Table 1-1 represent our best point estimates of aggregate costs for the
selected illnesses; however, these estimates are best interpreted as midpoints within a range of
uncertainty. This uncertainty arises from several different sources and for several different reasons which
are discussed throughout the report and summarized in the concluding section (Section 6). To partially
quantify this uncertainty, the results described in Section 4 also include confidence intervals for several of
the key values that went into constructing the aggregate cost estimates.
1-2
-------
2. Background
Several of the most prevalent and costly adverse health conditions among older Americans—
including chronic respiratory disease, heart disease, cerebrovascular disease, cancer,
pneumonia/influenza, and gastrointestinal illness—are known to be associated, at least in part, with
environmental exposures. To improve understanding of the magnitude of the cost burden imposed by
these types of environmentally related health effects among older Americans, we selected specific health
conditions for economic analysis. In this section, we define the selected conditions and discuss why they
were chosen for this analysis. We also provide a general description of the COI method, and we review
the existing evidence from the COI literature regarding the direct and indirect costs associated with the
selected conditions. We then discuss the limitations of using results from the existing literature for
assessing the direct and indirect costs of conditions related to environmental exposures in older adults.
2.1 Selection of Health Effects
Some of the key considerations in selecting conditions for analysis were:
• prevalence among the older population,
• expected average costs associated with illness,
• evidence of linkages to environmental exposures, and
• data availability.
We targeted conditions that were expected to impose a relatively high burden (through high
prevalence and/or high cost) among the elderly and conditions for which environmental exposures are
expected to be important risk factors. Using the commonly applied International Classification of Disease
(ICD-9) system for coding illnesses, we also defined conditions in a way that could be easily identified
and matched from multiple data sources.
Table 2-1 lists the selected health conditions according to their ICD-9 codes and groups the
conditions into six general categories of illness. The first category—chronic respiratory illness—is
common among older Americans and strongly associated with indoor and outdoor air-quality conditions.
Based on data for the Centers for Disease Control and Prevention (Lucas et al., 2004), in 2001, 6.7
percent of the 65 and older population suffered from chronic bronchitis and 5.1 percent suffered from
emphysema, compared to 5.3 and 0.8 percent respectively for those under 65. Prevalence of asthma
among the elderly (8.7 percent) is lower than in children and young adults; however, the illness is
typically more severe in older age groups. Exposures to common air pollutants, such as particulate matter
(PM) and ozone, have been found in several studies to cause and/or exacerbate these conditions.1
Coronary heart disease, including heart attacks and angina pectoris, is among the most common
conditions among the 65 and older population in the U.S., affecting roughly 21 percent of this population
in 2001. Several studies have also shown that exposures to common pollutants such as PM can increase
risks associated with heart disease, and exposures to less common toxic pollutants such as lead have also
been found to increase these risks. Because of the chronic nature of both heart disease and chronic
respiratory disease, both diseases are expected to be associated with frequent and costly health-care
utilization.
The prevalence of stroke is also very high among seniors, affecting 9 percent of the 65 and older
population in 2001. In that same year, more than 404,000 Medicare fee-for-service beneficiaries were
admitted to the hospital with a primary diagnosis of stroke (Trisolini et al., 2002). The Medicare hospital
payments associated with those admissions were quite high—approximately $2 billion in 2001. Because
1 For a summary of the epidemiological evidence regarding links between common air pollutants and
respiratory and cardiovascular illness, see for example EPA (1999).
2-1
-------
Table 2-1. Selected Health Conditions (by ICD-9 Code)
ICD-9 Code
Health Condition
stroke victims often require a
great deal of therapy and care
beyond the initial hospital
admission, the full cost of stroke
among older Americans is likely
to be much higher than $2
billion. Regarding the
environmental etiology of stroke,
studies have shown that
exposures to pollutants such as
PM, ozone, and lead may
increase risks of stroke-related
deaths (ATSDR, 1999).
Lung cancers, which
according to CDC estimates
(Lucas et al., 2004) affect less
than one percent of the 65 and
older population, are less
prevalent in the elderly
population than other conditions
and certain other cancers.
Nevertheless, lung cancers have
the 2nd highest prevalence of
cancers among those 65 and
older, and it is known to impose
significant costs on a per case
basis. Furthermore, links
between environmental
exposures and lung cancers are
well established. Known or
suspected environmental causes
of lung cancer include, among
others, exposures to PM (Pope et
al., 2002), asbestos (ATSDR,
2001), and radon (ATSDR,
1990).
Pneumonia is a relatively
high-prevalence condition
among older individuals, but
more importantly, when older individuals are afflicted with this condition, they generally require more
costly medical care than younger individuals. Among Medicare fee-for-service beneficiaries, the rate of
hospital admission for pneumonia (and/or influenza) was 20 per 1,000 in 2001.2 The annual cost of
hospitalizations related to pneumonia is also typically very high, and pneumonia infections are also more
likely to result in death when elderly persons become infected. As with chronic lung disease, exposures
to common air pollutants, such as PM and ozone, have been found in several studies to cause and/or
exacerbate pneumonia.
Chronic Lung Disease
Chronic bronchitis
Emphysema
Asthma
Bronchiectasis
Chronic airway obstruction, not elsewhere classified
Ischemic Heart Disease
Acute myocardial infarction
Other acute and subacute forms of ischemic heart
disease
Old myocardial infarction
Angina pectoris
Other forms of chronic ischemic heart disease
Stroke
Subarachnoid hemorrhage
Intracerebral hemorrhage
Other and unspecified intracranial hemorrhage
Occlusion and stenosis of precerebral arteries
Occlusion of cerebral arteries
Acute, but ill-defined, cerebrovascular disease
Lung Cancer
Malignant neoplasm of bronchus and lung
Secondary malignant neoplasm of respiratory and
digestive systems
Carcinoma in situ of respiratory system
Pneumonia
Viral pneumonia
Pneumococcal pneumonia
Other bacterial pneumonia
Pneumonia due to other specified organism
Bronchopneumonia, organism unspecified
Pneumonia, organism unspecified
Influenza with pneumonia
Gastrointestinal Illness
Intestinal Infectious Diseases
Other and unspecified noninfectious gastroenteritis and
colitis
491
492
493
494
496
410
411
412
413
414
430
431
432
433
434
436
162.2-162.9
197
231
480
481
482
483
485
486
487
001-009*
558.9
*Excluding 008.45 (Clostridium difficile colitis)
2 Age- and gender-adjusted; based on RTI analysis of the Medicare Quality Monitoring System special
analytical files (Trisolini et al., 2002).
2-2
-------
Gastrointestinal illness (GI) is common in all age groups in the U.S.; however, the elderly face
particularly high risks of hospitalization and death due to GI. According to discharge statistics for 1979-
1995, hospitalization rates for GI were more than twice as high for those older than 65, and over 75
percent of hospital deaths from diarrheal disease were among this age group (Mounts et al., 1999).
Exposures to waterborne pathogens are suspected to be significant contributors to overall rates of GI
(Morris and Levin, 1995; Payment et al., 1991; and Bennett et al., 1987), although the attributable fraction
is highly uncertain.
2.2 Overview of the COI Approach
The COI approach may be used to assess both direct and indirect costs of illness. Direct costs
represent the dollar value of goods and services consumed as a result of illness and for which payment is
made. They include payments for treatment, diagnosis, continuing care, rehabilitation, and terminal care
and are typically measured as costs related to hospital stays, physician services, nursing homes,
prescription drugs, and in-home health care services. Indirect costs represent costs for which no payment
changes hands but for which an economic effect is still observed. They include primarily productivity
losses associated with illness and premature death, and are typically measured as the value of lost
productivity (labor and household) due to illness.
Most COI studies use a prevalence-based approach for estimating costs. Prevalence-based cost
estimates include all costs related to a condition for the prevalent population over a given time period,
usually a year. This approach includes costs for newly diagnosed cases as well as for those in the
advanced stage of disease. Prevalence-based COI estimates are useful primarily for quantifying and
highlighting the burden of a particular disease or condition. Incidence-based COI estimates represent the
lifetime cost of disease, from onset to death. Incidence-based estimates require a great deal of data on
disease progression in addition to costs at each stage of disease, and provide a useful measure of the cost
savings of preventing or delaying onset of a disease for use in economic evaluations of preventive
interventions. Because EPA is currently interested in understanding and quantifying the burden of
diseases associated with environmental exposures in older Americans, a prevalence-based approach is
most appropriate.
In this study, we used a prevalence-based approach to develop COI estimates that focus on the
direct costs and labor, as well as household productivity losses, associated with each of the six conditions
of interest. Our methods and data sources for the COI analysis are described in Section 3, and results are
shown and described in Section 4.
2.3 COI Literature Review
Prior to generating our own estimates of the costs associated with heart disease, stroke, chronic
lung disease, lung cancer, pneumonia, and gastrointestinal illness in individuals older than 65 years, we
reviewed the literature on the costs of these conditions. We initially searched MEDLINE and PubMED to
identify articles on the cost, cost-of-illness, expenditure, or economic impact or burden of each of the
health conditions in a U.S. population 65 years and older. We limited our review to articles that were
published in peer-reviewed journals from 1990 to the present. Following the decision to focus on the
specific conditions and associated ICD-9 codes listed in Table 2-1 for COI analysis, we refined our
literature search to include only those conditions. However, we also eliminated the requirement that
studies provide cost estimates specifically for the 65 and older population. Our final review of articles on
the six conditions consisted of 24 articles on the direct or indirect costs of chronic lung disease, 7 on heart
disease, 9 on stroke, 7 on lung cancer, 9 on pneumonia, and 6 on GI illness. Brief summaries of methods,
data sources, and findings from each study reviewed are provided in Tables A-l to A-6 in Appendix A.
The methods, data sources, disease classifications, age groups considered, and specific cost
categories estimated in each study varied widely. For chronic lung disease, specific studies focused on a
2-3
-------
range of different cost outcomes, including hospitalization costs only, expenditures for all health care
services, work-loss costs only, employer costs for absenteeism and employees' health care services, and
informal caregiving costs. Data sources for estimating chronic lung disease costs also varied, and
included survey data sources, such as the National Hospital Discharge Survey and Medical Expenditure
Panel Survey, Medicare or private insurance claims data, and hospital-specific data on costs.
Consequently, cost estimates also varied widely. In 2000 dollars, estimated annual costs for all health
care services associated with chronic lung disease were approximately $1,000 to $7,800 per person (See
Table A-l).
Estimated costs for heart disease also varied widely. Estimated direct medical costs associated
with heart disease varied from about $5,900 to $7,400 per person. Total estimated medical costs
associated with heart disease ranged from $67 billion for females over 45 years of age to $90.3 billion for
all circulatory diseases in those 65 years and older. Large differences in estimated costs arise primarily
because of differences in the cost categories analyzed or in the age groups considered.
For stroke, many of the studies reviewed focused on hospital charges, but one of these estimated
costs only for those with diagnoses of both stroke and myocardial infarction. One study estimated
hospital charges of approximately $13,000 per person for stroke alone. The estimated total direct and
indirect cost of stroke was approximately $46 billion per year.
The estimated costs of lung cancer in the literature range from $5.6 billion (females over 45 years
only) to $8.2 billion per year. One study estimated average Medicare payments per year for several
cancers, including lung cancer, of about $28,000 per affected beneficiary (Riley et al., 1995).
Pneumonia cost studies primarily focused on hospitalization costs. Mean or median costs
associated with pneumonia ranged from about $5,900 to $10,000 per person. Variations in these
estimates reflect differences in the definitions used for pneumonia and the methods used (i.e., primary
diagnosis costs versus attributable costs). Total direct medical cost estimates associated with pneumonia
ranged from $5.7 to $19.4 billion per year.
Only one of the GI studies used the same disease classifications used in our COI analysis (Sandler
et al., 2002). We selected the condition codes shown in Table 2-1 because they reflect the GI most
closely associated with environmental exposures. Sandler et al. (2002) estimated direct medical costs
associated with GI of $1.7 million per year for all age groups, and additional indirect costs of $540
million per year. Just over half of those costs were estimated to result from foodborne illnesses.
2.4 Discussion
Our review of the COI literature on the six conditions revealed large differences across studies in
the methods used, the specific condition codes considered, the data sources analyzed, the age groups
included, and the categories for which costs were estimated. Such differences make it difficult to
compare cost estimates from different studies without first attempting to adjust them. We therefore
recommended original COI analyses for the six conditions, using a consistent set of methods and data
sources and focusing on the same cost categories across conditions for those 65 years and older.
In this subsection, we discuss features of COI studies in the literature that make it difficult to
compare cost estimates across conditions or across studies on the same condition. We then describe how
our COI approach produces comparable estimates across the six conditions.
First, the methods used to estimate disease costs vary across studies in the literature. Some
analyses estimate costs for a primary diagnosis of a condition, while others provide cost estimates for any
diagnosis of a condition. Still other studies generate cost estimates at the person level, as opposed to the
condition level, by calculating the costs attributable to a condition (i.e., those costs that could presumably
be saved if the condition were prevented) or by calculating incremental costs for people with the
condition as compared to costs for people without the condition. No one approach is better than the
others. In most cases, the available data dictate which methods may be used to assess costs. For
2-4
-------
example, if condition codes are assigned to individuals with the condition as opposed to being assigned to
specific health care services, it is generally preferred to estimate the incremental, or excess, cost of care
for people with the condition.
COI studies in the literature rarely defined the conditions using the same set of ICD-9 codes as
we selected for focus, making it difficult to compare cost estimates across studies and to our COI
estimates. Our selection of ICD-9 codes for each condition was in part based on identifying health effects
that are more likely to be associated with environmental exposures in older adults.
The data available for COI analysis ranged from private health insurance or Medicare claims data
to national household-based survey data. Data from insurance claims are likely to provide the best source
of information for identifying costs for each of the ICD-9 codes of interest. In contrast, survey data
generally contain ICD-9 codes that are assigned to individuals based on self-reports of a health condition.
However, because claims data provide cost and utilization information only for direct medical costs and
exclude nonreimbursable medical care costs and nonmedical and indirect costs, survey data are often
needed to estimate costs in these categories.
Finally, in reviewing COI analyses in the literature, we discovered that many studies focus on
narrow categories of costs, such as hospitalization or employer-based costs, which makes it difficult to
compare estimates across studies without accounting for the differences in the cost categories included.
In our COI analyses, we attempted to estimate costs for the same set of cost categories for all six of the
conditions. For some categories of costs, however, there were not enough observations with a specific
condition to be able to generate reliable condition-specific cost estimates.
Differences in the information available from each data source used meant that data-specific
methods were applied to estimate each category of costs. These methods were consistently applied across
the six conditions. For example, because claims data contain information about the costs to diagnose or
treat a specific condition, we estimated inpatient, physician, outpatient, home health, and durable medical
equipment costs from Medicare claims by estimating the costs of a primary or any diagnosis of the health
condition of interest (see Section 3 for details). In contrast, survey data generally contain costs for
individuals but not for specific diagnoses. We therefore used survey data to estimate the incremental
prescription drug and nursing home costs for people with each condition and to estimate the work loss
and bed days attributable to each condition.
Because consistent COI methods, data sources, and disease classifications were used in our COI
analyses for the six conditions, EPA can use the resulting COI estimates to make comparisons across the
conditions. Such comparisons should help in assessing the relative burden of each condition among older
adults and in making decisions about where to direct resources for the prevention or treatment of disease
associated with environmental health exposures in the elderly.
2-5
-------
3. Cost Analysis Data and Methods
To estimate the annual cost burden associated with each of the six selected conditions in 2000
among individuals 65 and older, we separately estimated direct and indirect costs for each illness. In this
section, we describe the data and methods used to construct these cost estimates.
3.1 Direct Costs
To make best use of the available national data, we separated the estimation of direct costs into
three mutually exclusive components:
• medical costs,
• self-administered prescription drug costs, and
• nursing home costs.
Each of these components is discussed below.
3.1.1 Medical Costs
For individuals 65 and older, Medicare claims provide the most comprehensive source of data on
costs for medical services. National samples of claims data are accessible in different ways. After
reviewing the alternatives, we selected the Consumer Assessment of Health Plans Surveys (CAHPS) for
Medicare Fee-for-Service (MFFS) beneficiaries as the primary data source for our analysis.
3.1.1.1 Data Description
CAPHS-MFFS was selected for this part of the analysis because it was relatively easy to access
and is well suited for this analysis. These data include detailed Medicare claims data and other potentially
relevant demographic and health-related information for a large and nationally representative sample of
older Americans.
The sampling frame for CAHPS-MFFS was drawn from CMS's Enrollment Data Base (EDB),
which comprises approximately 30 million persons enrolled in fee-for-service Medicare and residing in
the United States or Puerto Rico. Beneficiaries with less than 6 months' continuous enrollment in MFFS
and those with a representative payee or living in an institution were excluded. To select the CAHPS-
MFFS sample, a total of 276 geographic primary sampling areas were constructed, each consisting of one
or more counties, and the sample was allocated to these areas to achieve a minimum of 300 respondents
in each area.
The CAHPS-MFFS sample for this study included 145,875 noninstitutionalized Medicare
beneficiaries, age 65 years or older, who were enrolled in MFFS and eligible for Part A and Part B
benefits for at least part of 2000. Beneficiaries with proxy respondents were excluded from the analysis.
It should be noted that, because the study was based on a retrospective analysis of self-reported survey
data, the study sample has a disproportionate share of nondecedents. The disproportionate share of
nondecedents in the sample may slightly bias the cost estimates from our analysis. However, because
health care costs tend to escalate during the last few months of a person's life and drop to zero thereafter,
the overall direction of this bias is not known
3.1.1.2 Analysis Methods
To estimate medical costs associated with the selected illness categories, we began by extracting
Medicare claims information for the year 2000 for all Medicare-covered medical services and for all
survey respondents and nonrespondents. Medicare-covered services include
3-1
-------
• acute and nonacute inpatient services,
• hospital outpatient and other ambulatory services,
• professional fees,
• home health services, and
• durable medical equipment.
Medicare does not cover self-injected medications administered in a home or outpatient setting
and nursing home or other long-term care services; therefore, costs for these services are not included in
this claims-based analysis.
Costs reported in the Medicare claims data include
• payments made by Medicare,
• beneficiary payments in the form of deductibles and copayments, and
• any third-party payments.
Total medical payments are calculated as the sum of the three payments types.
We then identified medical costs associated with each of the illness categories by matching the
ICD-9 codes recorded for each claim with the condition-specific ICD-9 codes listed in Table 2-1. For
inpatient services, each claim consolidates payments at an admissions level. Consequently, each of these
claims may be associated with multiple ICD-9 codes—one for the primary diagnosis and potentially
several more codes for related secondary diagnoses. Similarly, claims for outpatient and home health
services report payments that may be associated with multiple ICD-9 codes. In contrast, claims for
professional fees and durable medical equipment are all condition specific and are therefore associated
with a single ICD-9 code.
For this analysis, we selected all claims that were recorded with at least one of the ICD-9 codes
listed in Table 2-1, including both primary and secondary diagnostic codes for inpatient, outpatient, and
home health services. Because payments for these types of services are consolidated, they may include
costs that are attributable to conditions other than the selected ICD-9 codes. As a result, for inpatient,
outpatient, and home health services, the cost estimates developed in this analysis are likely to overstate
the costs that are specifically attributable to the six conditions of interest. The extent of this
overestimation is not known. For professional fees and durable medical equipment, no similar
overestimation of costs is expected because they are all based on condition-specific claims.
Using the CAHPS-MFFS (with appropriate geographic weights for the stratified sample), we
were therefore able to obtain unbiased nationally representative estimates for the roughly 26 million
nonistitutionalized Medicare beneficiaries over the age of 64 enrolled in fee-for-service in 2000. Based
on the sample of 145,875 respondents, we are able to estimate the number and percentage (prevalence
rate) of beneficiaries, 65 and older, who were diagnosed with conditions in each of the six illness
categories in 2000. It should be noted these prevalence estimates are based on claims for services in
2000; therefore, they do not include individuals who had the underlying condition in 2000 but did not for
whatever reason seek care for the condition during that period. Consequently, these claims based
prevalence estimates are likely to underestimate by a small amount the overall prevalence of each
condition in the noninstitutionalized fee-for-service elderly population.
For these beneficiaries, we are also able to estimate the average (per capita) and aggregate
medical costs associated with these illnesses in 2000. With the information in CAHPS-MFFS, it is also
possible to disaggregate these estimates by
3-2
-------
• type of medical service,
• age/gender categories, and
• Census region and state.
The results for each illness category are summarized in Section 4, with additional detail provided in
Appendix C.
To extrapolate these direct medical cost estimates to the entire non-institutionalized population of
individuals 65 and older in 2000, we assumed that the prevalence and average costs for each condition
based on the CAHPS-FFS data were also applicable to those not participating in fee for service Medicare
and not in nursing homes. According to U.S. Census data, the total 65 and older population in 2000 was
35 million, and according to the National Center of Health Statistics (NCHS, 2003) the number of nursing
home residents was 1.5 million. Therefore, we applied the prevalence and average cost estimates for
each condition to an assumed total population of 33.5 million. This extrapolation is like to slightly
overestimate prevalence and costs for those not in fee-for service Medicare (and not institutionalized)
because they are on average somewhat younger and in somewhat better health than those in Medicare fee
for service (Brown et al., 1993 ).
3.1.2 Self-Administered Prescription Drug Costs
One of the inherent limitations in using Medicare claims data (including CAHPS-MFFS as
discussed in Section 3.1.1) to assess direct costs of illness is that expenditures for self-administered
prescription drugs are not reimbursable by Medicare and are therefore excluded from the claims-based
cost estimates. To estimate these prescription drug costs for each of the six conditions, we used data from
the 2000 Medical Expenditures Panel Survey (MEPS).
3.1.2.1 Data Description
The MEPS is a nationally representative subsample of the National Health Interview Survey
(NHIS, described in some detail below in Section 3.2). One important feature of the MEPS is that it
allows for the linkage of demographic data on individuals to their health care utilization, spending, and
sources of payment.
The 2000 survey requested information about the number of prescription medication purchases
(including refills) and total expenditures for prescription drugs by sources of payment. The sources of
payment for which information was collected include self or family, Medicare, Medicaid, private
insurance, TRICARE, the Veterans' Administration, workers' compensation, and other sources. The
survey also contains a conditions file that includes information about all conditions reported for
household residents.
Health conditions included in MEPS are those reported by a household respondent and those
listed as the reason for a medical provider visit or prescribed medication. Medical conditions were
recorded by the interviewer as verbatim text; professional coders later assigned three-digit ICD-9 codes.
3.1.2.2 Analysis Methods
To estimate the prescription drug costs associated with each of the six conditions, we first created
a sample of individuals 65 years and older from the 2000 MEPS and then calculated mean prescription
drug expenditures for each condition by age group (65 to 74, 75 to 84, and 85 years and older) and by sex.
Mean expenditures and standard errors were estimated for total prescription drug spending, for Medicare
spending, for total spending except Medicare, and for total spending except Medicare and self- or family-
paid costs. To avoid double-counting, our reported estimates focus on the cost estimates that exclude
prescription drugs where Medicare was reported as the source of payment. However, cost estimates from
3-3
-------
our analysis of Medicare claims may also include co-pays or deductibles that represent out-of-pocket
expenses. Consequently, our prescription drug cost estimates may include some out-of-pocket expenses
that are also included in our cost estimates from Medicare claims analysis.
Because estimates of mean total prescription drug costs cannot be taken as attributable cost
estimates, we also estimated the incremental, or additional, costs of prescription drugs for individuals
with each condition as compared to persons without the condition, by age group and by sex. Incremental
cost estimates represent the additional costs for those with the condition as compared to those without the
condition. We chose to consider incremental costs because it allows us to interpret our estimates as
representing excess costs for people with the conditions of interest. However, because it is not
uncommon for individuals in the 65 and older age group to have high medical care expenses, incremental
cost estimates in an older population are often close to zero or even negative. In other words, individuals
in this age group who do not have the condition of interest have a fairly high likelihood of having another
costly condition or disability. In comparison, total cost estimates are limited in that they represent total
costs for individuals with a condition and do not account for the high degree of comordity in the over-65
population.
3.1.3 Nursing Home Costs
Another gap in our cost estimates from Medicare claims data (using CAHPS-MFFS) is the cost
associated with care in a nursing home. We address this gap by analyzing nursing home charges using
the 1999 National Nursing Home Survey (NNHS).
3.1.3.1 Data Description
The 1999 NNHS sample consisted of 1,496 of the approximately 18,400 nursing homes in the
United States that had at least three beds and were either certified by Medicare or Medicaid or had a state
license to operate as a nursing home. A sample of up to six current residents was selected from each
participating nursing home for inclusion in the current resident file. Interviews were then conducted with
nursing staff familiar with residents' medical records to collect demographic information, diagnoses
(ICD-9 codes assigned at admission and based on current records), total charges for a specified period,
and indicators for a range of different sources of payment (e.g., self pay, Medicare, Medicaid). The
responding nurse referred to the resident's medical record when answering the interview questions.
3.1.3.2 Analysis Methods
To develop an estimate of the nursing home costs associated with each of the six conditions, we
used the NNHS current resident file and limited our analysis to those people 65 years and older. We then
calculated mean total charges by sex, age group (65 to 74, 75 to 84, and 85 years and older), and
condition. These estimated charges were converted to costs by using the Medicare cost-to-charge ratio
for skilled nursing facilities, which was 0.78777. Estimated costs were updated to 2000 dollars using the
consumer price index for all urban consumers.
Incremental cost estimates were then generated for each condition by taking the difference of
mean total costs by age group and sex for those with and without the condition of interest. We also
examined frequencies of Medicare as a source of payment and the inclusion of drugs and medical supplies
in the total charges reported.
Applying estimates of mean incremental costs for each health condition to calculate aggregate
nursing home costs also requires estimates of the number of elderly in nursing homes with each condition.
Data on nursing home trends (NCHS, 2003) indicates that there were approximately 1.5 million nursing
home residents in 2000. To estimate the portion of this population that experienced each condition, we
applied the prevalence rates derived from the CAPHS-MFFS data. These claims based prevalence
estimates are likely to somewhat underestimate prevalence among nursing home residents because rates
3-4
-------
of illness are typically higher in the institutionalized population and because the claims based estimates
do not include those who did not seek care for their condition in 2000.
3.2 Indirect Costs
Indirect costs are those for which no payment is made, but for which an economic effect is still
observed. In our analysis and in most COI studies, a human capital approach is used to value the indirect
cost of time lost from productive activities due to excess morbidity or premature mortality. The human
capital approach values productivity losses based on market earnings and an imputed value for household
production (Drummond et al., 1997). Advantages of the human capital approach are that it assumes a
societal perspective and relies on data that are readily available. A disadvantage is that estimates of the
value of life and health among the retired elderly will be lower than comparable estimates for working-
age adults because the estimates are based largely on measures of market productivity. The human
capital approach also ignores the costs of pain and suffering associated with an illness and excludes the
value of time lost from consumption activities (Tolley, Kenkel, and Fabian, 1994).
The concept of willingness to pay (WTP) is useful for addressing the limitations of the human
capital approach. However, the implementation of stated or revealed preference approaches to estimate
the value of health effects, or WTP, requires far more time and resources than does a human capital
approach using existing data. For this reason, we used the human capital approach to value indirect costs
associated with the six conditions. These estimates may be thought of as lower-bound estimates for the
full indirect costs of each condition.
3.2.1 Increased Morbidity
We used data from the 2001 National Health Interview Survey (NHIS) to estimate the impact of
each condition on labor force participation rates, number of work loss days for those who were working
during the survey year, and number of bed days. These estimates were combined with age- and sex-
specific earnings and household productivity values for 2000 to generate estimates of the increased
morbidity costs associated with each condition.
3.2.1.1 Data Description
The NHIS is an annual household interview survey designed primarily to collect data about the
health status, health conditions, and health care utilization of household members. The survey also
requests information about basic demographics and about days lost from work, bed days, and functional
limitations. Data are collected annually from about 43,000 households and about 106,000 persons within
the households. Information about diseases and other health conditions among people was also collected.
However, the NHIS did not contain information about pneumonia or gastrointestinal illness; as a result,
these conditions are excluded from our morbidity cost estimates.
Data from the NHIS were combined with productivity estimates from Grosse (2003). Grosse
provides estimates of annual earnings for those in the workforce and household productivity for all adults
by age group (65 to 74 and 75 and older) and by sex.
3.2.1.2 Analysis Methods
We used the 2001 NHIS to estimate two components of labor productivity losses: the losses
associated with being completely unable to work (out of the labor force) and the losses associated with
missing work days. For the first component, we used estimates of the percentage of those who report that
they are not currently working but worked previously as a proxy for the probability of being unable to
work due to poor health. We estimated the probability of being completely unable to work because of
one of the six health conditions as the excess percentage of those with the condition as compared to those
3-5
-------
without the condition who are not currently working. Separate estimates of the probability of being
unable to work were developed for all those 65 years and older for each condition.
We then estimated the per-person morbidity cost associated with being completely unable to
work as the estimated probability of being unable to work because of the condition multiplied by annual
earnings. Annual labor market earnings estimates were by sex for the 65 to 74 and 75 years and older age
groups.
For the second component of labor productivity losses, we limited our analysis to those over 65
years of age who reported that they currently work. We then used a regression approach to estimate the
impact of each condition on missed work days, controlling for a number of variables, including education,
poverty status, self-reported health status, occupation, and smoking status.
Results from our regression analysis are shown in Appendix B. Using the regression results, we
generated predictions of the number of missed work days attributable to each of the six conditions. These
estimates were generated by condition and by sex for all age groups combined. We then estimated labor
productivity losses by multiplying predicted work-loss days by sex-specific average daily earnings for
those 65 to 74 years and 75 years and older in the workforce (Grosse, 2003). Per-person estimates for all
those with the condition were generated by multiplying estimated productivity losses due to missed work
days by the percentage with each condition that currently works. Our methods for estimating the
household productivity losses associated with being sick in bed were similar.
Our bed days analysis included all those 65 years and older. We used a regression approach to
estimate the impact of each condition on number of bed days, controlling for education, poverty status,
self-reported health status, labor force participation status, and smoking status. Based on these results, we
predicted the number of bed days attributable to each condition by sex. Household productivity losses
were estimated by multiplying predicted bed days for each condition by sex-specific average household
productivity for those 65 to 74 years and 75 years and older.
The estimates of labor productivity losses shown in Section 4 represent the sum of the expected
costs due to being unable to work and the expected costs due to missing work for people with each
condition. Household productivity losses represent the value of lost household work attributable to each
condition.
Our estimates for lung cancer are limited by the small number of observations in the 65 years and
older age group in the 2001 NHIS. Only 26 males and 15 females were identified with lung cancer. The
number with lung cancer who reported being in the labor force was even smaller and forced us to exclude
lung cancer from our work days-loss analysis. We also excluded lung cancer from our analysis of bed
days attributable to lung cancer. The only morbidity costs for lung cancer included in our analysis are the
costs associated with being completely unable to work.
We do not provide measures of dispersion for our indirect cost estimates, in large part because
our source for the earnings and household productivity estimates did not provide standard errors or other
measures of dispersion. Because the variation in earnings among older individuals who work is much
greater than the variation in the percentage of older Americans who work, we chose not to estimate
standard errors that would necessarily treat earnings as having no variation.
3.2.2 Premature Mortality
We used data from the 1998 National Vital Statistics Report (NVSR) to determine the number of
annual deaths with one of the six conditions listed as the cause of death. These data were combined with
estimates of the present value of earnings and household productivity from Grosse (2003) to estimate the
cost of mortality due to each condition.
3-6
-------
3.2.2.2 Data Description
The 1998 NVSR provides calculations from all death certificates filed in the 50 states and the
District of Columbia in 1996 (Peters, Kockanek, and Murphy, 1998). We relied on the 1998 report
primarily because it uses ICD-9 codes to categorize causes of death. Later reports began using the revised
ICD-10 classification scheme. The NVSR provides statistics on deaths and death rates by age group (65
to 74, 75 to 84, and 85 years and older) for groupings of ICD-9 codes that closely match the disease
groupings we selected for COI analysis. It also provides death rates for the overall population by age
group and sex.
Grosse (2003) provides estimates of the present value of earnings and household productivity by
age for several different discount rates. In our analysis, we use the Grosse (2003) present value estimates
for earnings only and for earnings and household production combined. For both measures, we assume
that the value of future production is discounted at an annual rate of 5 percent.
3.2.2.3 Analysis Methods
To estimate the indirect costs resulting from premature mortality among those 65 years and older,
we multiplied the cause- and age group-specific number of deaths for each condition by the age group-
specific present value of earnings (earnings and household production). Because Grosse (2003) does not
provide an estimate for expected productivity among those 85 and older, we were not able to estimate
losses (labor or household) for deaths in the 85 and older age group. The present value of productivity
losses in this age group is expected to be lower than for the younger age groups.
For ischemic heart disease and pneumonia, the health condition categories provided in the NVSR
exactly matched our characterization of the condition using ICD-9 codes. However, for chronic lung
disease (CLD), stroke, and lung cancer, the NVSR condition categories included some ICD-9 codes that
were not used in our assessment of direct costs. For gastrointestinal illness, the NVSR condition
categories excluded some of the ICD-9 codes used in our direct cost analyses. Because of these
discrepancies, our estimates of mortality costs for CLD, stroke, and lung cancer are slight overestimates
of mortality costs, while our estimate for gastrointestinal illness is an underestimate of actual mortality
costs for the conditions of interest.
3-7
-------
4. Cost-of-lllness Estimates for Selected Conditions in Older
Americans
Using the data and methods described in Section 3, we estimated the direct and indirect costs of
illness associated with each of the six selected conditions in 2000 among individuals 65 and older. In this
section, we separately summarize the results for CLD, ischemic heart diseases (IHD), stroke, lung cancer,
pneumonia, and gastrointestinal illness. All costs reported in the results tables are expressed in 2000
dollars.
4.1 Chronic Lung Disease
As summarized in
Table 4-1, we estimated that
approximately 4.4 million older
Americans suffered from CLD in
2000, and it was the primary
cause associated with almost
100,000 deaths in individuals 65
and older. Based on these
estimates, we also calculated
aggregate direct and indirect
costs from CLD in this
population to be approximately
$33.9 billion. These results are
further described and
disaggregated in the sections
below.
Table 4-1. Costs of Illness Associated with Chronic
Lung Disease—Estimated Aggregate Costs in
2000 for Individuals 65 and Older
Cost Category
Number of
Affected
Individuals
Mean Cost
($/person)
Aggregate Costs
Direct Costs
Medical
4,190,512
$5,786
$24,246,304,937
Prescription Drug
4,190,512
$574
$2,405,354,137
Nursing Home
185,000
$3,471
$642,135,000
Indirect Costs
Morbidity
4,190,512
$1,574
$6,594,623,730
Mortality
98,326
—
$9,895,808
Total Costs of Illness
—
—
$33,898,313,612
4.1.1 Direct Costs: Medical
Based on CAHPS-MFFS data, roughly 3.3 million (12.5 percent of) noninstitutionalized
Medicare fee-for-service elderly beneficiaries suffered from CLD in 2000, and the annual medical costs
for these beneficiaries averaged about $5,800 per year. As shown in Table 4-2, the largest component of
these costs was associated with inpatient services, which averaged over $13,000 for nearly 1.2 million
beneficiaries. The aggregate medical cost estimates shown in Table 4-1, were extrapolated from these
estimates, assuming the same prevalence and average costs for CLD among the noninstitutionalized
elderly who do not participate in Medicare fee-for-service. As described in Section 3, this extrapolation is
likely to somewhat overestimate aggregate medical costs in the noninstitutionalized elderly population.
Rates of CLD among the male elderly population were found to be significantly higher than
among the female elderly population in each of the three age categories listed in Table 4-2. Furthermore,
whereas rates of CLD among the female population were between 10 and 12 percent for all three age
categories, the rate for males climbed from 13 percent for those between 65 and 74 years to 19 percent for
those 85 and older. Compared to this variation in prevalence rates, there was considerably less variation
in average medical costs for CLD across age and gender categories, ranging between $5,000 and $6,500.
4.1.2 Direct Costs: Self-Administered Prescription Drugs and Nursing Home Care
Having CLD was associated with significantly higher prescription drug costs for males and
females in the 65 to 74 years age group. These estimates are described in Table 4-3. Estimated costs for
this age group were approximately $600 to $1,200. Estimated incremental prescription drug costs were
$574 for the total population over 65 years with CLD. Excess costs for those in the two oldest age groups
4-1
-------
Table 4-2. Direct Medical Costs Associated with Chronic Lung Disease—Number of
Beneficiaries and Average Medical Payments by Type of Service and Age-
Gender Category
Beneficiaries with a Diagnosis
Direct Medical Costs
($/Person/Year)
Number
Percent
95% CI (±)
Mean
95% CI (±)
Total
3,264,847
12.5
(12.3, 12.7)
5,786
(5,568, 6,005)
Type of Medical Service
Inpatient
1,196,221
4.6
(4.5, 4.7)
13,019
(12,535, 13,504)
Physician
1,611,419
8.8
(8.6, 8.9)
461
(439, 483)
Hospital outpatient
1,069,976
4.1
(3.9, 4.2)
570
(535,610)
Home health
201,731
7.7
(0.7,0.8)
2,643
(2,406, 2,880)
Durable medical equipment
720,079
2.7
(2.7, 2.8)
2,220
(2,142, 2,298)
Age-Gender Category
Age 65-74 Years
Male
754,596
12.9
(12.5,0.13.3)
5,596
(5,146, 6,046)
Female
706,640
10.1
(9.8, 10.5)
5,122
(4,655, 5,588)
Age 75-84 Years
Male
649,188
17.1
(16.6, 17.7)
6,030
(5,539, 6,520)
Female
687,217
11.6
(11.2, 11.9)
6,461
(5,901,7,021)
Age 85+ Years
Male
192,794
19.0
(17.8,20.1)
5,131
(4,610, 5,652)
Female
273,812
10.5
(10.0, 11.1)
6,219
(5,625, 6,814)
Note: Geographic weights used to obtain unbiased nationally representatives estimates for Medicare elderly
population.
Table 4-3. Other Direct Costs Associated with Chronic Lung Disease—Incremental Cost
Estimates for Prescription Medications and Nursing Home Services by Age-
Gender Category
Prescription Drug Incremental Costs
Nursing Home Incremental Costs
Mean
Mean
Sample
($/Person/
Sample
($/Person/
Age
Size
Year)
95% CI (±)
Size
Year)
95% CI (±)
Age 65+ years
174
$574
(264, 884)
764
$3,471
(-3,138, 10,081)
Age 65-74 years
Male
37
$563
(124, 1,003)
71
$28,200
M3,331, 99,730)
Female
59
$1,215
(588, 1,843)
64
$9,574
(222, 18,927)
Age 75-84 years
Male
24
$128
(-251, 506)
102
($651)
(-6,507, 5,206)
Female
40
$141
(-325, 607)
175
$505
(-3,286, 4,296)
Age 85+ years
Male
4
($429)
(-701,-157)
104
$1,102
(-2,846, 5,049)
Female
10
$42
(-609, 694)
248
($634)
(-2,925, 1,657)
Notes: Prescription drug estimates exclude those with source of payment listed as Medicare. () denotes a negative
value.
4-2
-------
were not statistically significantly different from zero. The lower estimated excess drug costs for the two
oldest age groups probably reflects the higher rate of comorbidity and prescription drug use among all
individuals 75 years and older, which tends to make the prescription drug spending of those with CLD
look similar to those without.
Table 4-3 also shows estimates of nursing home costs associated with CLD. Although the
standard errors on the estimated incremental nursing home costs were large, incremental costs were fairly
large for males and females in the 65 to 74 years age group, especially for males. Estimated excess
nursing home costs for the other two age groups were much smaller and were not statistically different
from zero. Our findings of higher excess costs for the youngest age group with CLD is consistent with
the hypothesis of a large degree of costly comorbidities within the 75 years and older population.
4.1.3 Indirect Costs: Morbidity
Table 4-4.
Morbidity-Related Indirect Costs Associated
with Chronic Lung Disease—Estimated
Annual Per-Person Productivity Losses by
Age-Gender Category
As shown in Table 4-4,
we estimate that, for those older
Americans suffering from CLD,
the annual per person labor
productivity losses resulting the
illness vary across age and sex
subgroups from about $900 to
$2,000. These estimates include
the probability of being out of
the labor force, which is about 4
percentage points higher in
people with CLD than for the
average older American, and
work-loss days among the 10
percent of older people with
CLD who continue to work. Our
estimates of work-loss days
attributable to CLD are 4.6 for
females and 3.4 for males.
Estimates of annual household productivity losses resulting from increased morbidity in those
with CLD are fairly low and average $154 per person for all those over 65 years of age. Estimates range
from $95 per year for males older than 75 years up to $211 per year for females in the 65 to 74 years age
group. Our estimates indicate that the number of bed days per year attributable to CLD is about 5 for
females and 3 for males.
Total productivity losses are calculated as the sum of estimated labor and household productivity
losses. They are estimated to average about $1600 for those 65 and older with CLD.
Labor
Household
Total
Productivity
Productivity
Productivity
Age
Losses
Losses
Losses
Age 65+ years
$1,420
$154
$1,574
Age 65-74 years
Male
$1,867
$103
$1,971
Female
$836
$211
$1,047
Age 75+ years
Male
$1,566
$95
$1,660
Female
$752
$181
$933
4.1.4 Indirect Costs: Mortality
Using data from NSVR, Table 4-5 shows that the annual number of deaths with CLD as the cause
was estimated to be about 98,000 in 2000, with 76 percent occurring in people between 65 and 84 years
of age. Because many of the expected deaths resulting from CLD are in the 65 to 74 and 75 to 84 years
age groups, the present value of labor productivity losses associated with CLD is high—over $2 billion
per year. The combined value of labor and productivity losses is also high—about $9.9 billion per year.
As discussed in Section 3, the value of productivity losses for those 85 and older could not be estimated
due to lack of data on expected productivity for this age group.
4-3
-------
Table 4-5. Mortality-Related Indirect Costs Associated with Chronic Lung Disease—
Estimated Aggregate Productivity Losses by Age Category
Age
Number of
Deaths
Aggregate Labor
Productivity Losses
(in 1,000s)
Aggregate Labor and Household
Productivity Losses
(in 1,000s)
Age 65+ years
98,326
$2,120,309
$9,895,808
Age 65-74 years
29,869
$1,523,485
$5,685,484
Age 75-84 years
44,651
$596,824
$4,210,324
Age 85+ years
23,806
—
—
Notes: Chronic lung disease diagnosis in mortality data is based on ICD-9 codes of 490-496. Mortality related
indirect costs for 85+ not estimated due to lack of data on expected productivity for this age group.
Cost Category
4.2 Ischemic Heart Disease
Of the six conditions Table 4-6.
analyzed in this report, IHD is
the most prevalent condition
among older Americans. We
estimate that approximately 6.9
million older Americans
experienced IHD in 2000, and
other data indicate that almost
450,000 deaths were primarily
attributable to the disease. As
reported in Table 4-6, we
estimate the aggregate direct
(except nursing home) and
indirect costs from CLD in this
population to be approximately
$50.1 billion in 2000. These
results are further described and
disaggregated in the sections below
Costs of Illness Associated with Ischemic
Heart Disease—Estimated Aggregate Costs in
2000 for Individuals 65 and Older
Number of
Affected
Individuals
Mean Cost
($/person)
Aggregate Costs
Direct Costs
Medical
Prescription drug
Nursing home
Indirect Costs
Morbidity
Mortality
Total Costs of Illness
6,604,030
6,604,030
290,080
6,604,030
443,494
$5,716
$680
$1,193
$37,748,632,961
$4,490,740,100
$7,878,384,878
$34,535,761
— $50,152,293,701
4.2.1 Direct Costs: Medical
Our analysis of direct medical costs, summarized in Table 4-7, indicates that over 5.1 million (20
percent of) Medicare fee-for-service elderly beneficiaries experienced IHD in 2000. The prevalence of
IHD is therefore almost 60 percent greater than for CLD in this population; however, the annual medical
costs per beneficiary are very similar, averaging about $5,700 per year. As with all of the conditions
analyzed in this report, the largest component of estimated medical costs was associated with inpatient
services, in this case averaging about $13,000 for over 1.8 million beneficiaries. When the estimated
medical costs of IHD for fee-for-service beneficiaries ($5716) are extrapolated to the entire
noninstitutionalized population of adults over 65 in 2000 (33.5 million), they sum to almost $38 billion,
as shown in Table 4-5.
As is the case with CLD, the prevalence of IHD was found to be significantly higher among the
male elderly population than the female elderly population in each of the three age categories listed in
Table 4-7. However, rates of IHD for both males and females were significantly higher in the older age
categories. The rate for males was 22 percent for those between 65 and 74 years and 32 percent for those
4-4
-------
Table 4-7. Direct Medical Costs Associated with Ischemic Heart Disease—Number of
Beneficiaries and Average Medical Payments by Type of Service and Age-
Gender Category
Beneficiaries with a Diagnosis
Direct Medical Costs
($/Person/Year)
Number
Percent
95% CI (±)
Mean
95% CI (±)
Total
5,145,229
19.6
(19.4,19.9)
5,716
(5,557, 5,875)
Type of Medical Service
Inpatient
1,878,134
7.2
(7.0, 7.3)
13,181
(12,852, 13,511)
Physician
4,084,632
15.6
(0.1,0.2)
641
(621, 661)
Hospital outpatient
1,859,793
7.1
(7.0, 7.2)
407
(389,425)
Home health
265,519
1.0
(1.0,1.1)
2,063
(1,913,2,213)
Durable medical equipment
40,886
0.2
(0.1,0.2)
831
(584, 1,079)
Age-Gender Category
Age 65-74 Years
Male
1,280,121
21.9
(21.4, 22.3)
5,707
(5,359, 6,056)
Female
832,829
11.9
(11.6, 12.3)
5,803
(5,385, 6,220)
Age 75-84 Years
Male
1,169,587
30.8
(30.1, 31.5)
5,770
(5,429, 6,110)
Female
1,028,219
17.3
(16.8, 17.8)
5,792
(5,439, 6,146)
Age 85+ Years
Male
320,908
31.5
(30.2, 32.9)
4,957
(4,530, 5,385)
Female
513,565
19.8
(19.0, 20.5)
5,793
(5,400, 6,186)
Note: Geographic weights used to obtain unbiased nationally representatives estimates for Medicare elderly
population.
85 and older. For women, the rates of IHD were 12 and 20 percent, respectively. The average medical
costs for IHD across age and gender categories varied by less than $1,000 (i.e., between $4,900 and
$5,800).
4.2.2 Direct Costs: Self-Administered Prescription Drugs and Nursing Home Care
Our estimates of the excess prescription drug costs associated with heart disease were positive for
all age and sex subgroups. As reported in Table 4-8, on average, prescription drug costs for those over 65
years with IHD were $680 higher than for those without. These findings suggest that heart disease drug
costs exceed the cost of drugs for many other diseases that afflict the elderly. Estimated incremental
prescription costs for the 65 to 74 years age group were $700 to $1,100 and statistically different from
zero. For the two older age groups, estimated excess drug costs were generally positive, but much lower
and not significantly different from zero.
Estimates of excess nursing home costs for individuals with heart disease are not significantly
different from zero for any of the age and sex subgroups. These findings suggest that among nursing
home residents older than 64 years, those with heart disease have nursing home costs that are no higher
than those without the condition. The estimated negative mean value for incremental nursing home costs
reported in Table 4-8 was assumed to be implausible and therefore not included in the aggregate estimates
for IHD reported in Table 4-6.
4-5
-------
Table 4-8. Other Direct Costs Associated with Ischemic Heart Disease—Incremental Cost
Estimates for Prescription Medications and Nursing Home Services by Age-
Gender Category
Prescription Drug Incremental Costs
Nursing Home Incremental Costs
Mean
Mean
Sample
($/Person/
Sample
($/Person/
Age
Size
Year)
95% CI (±)
Size
Year)
95% CI (±)
Age 65+ years
156
$680
(400, 961)
1,042
($1,357)
(-3,187, 474)
Age 65-74 years
Male
45
$722
(203, 1,240)
55
($9,572)
(-27,005, 7,860)
Female
37
$1,120
(467, 1,773)
55
$3,240
(-3,486, 9,967)
Age 75-84 years
Male
22
$18
(-233, 269)
116
($1,286)
(-6,962,4,389)
Female
35
$675
(260, 1,091)
231
($1,969)
(-5,371, 1,433)
Age 85+ years
Male
8
$311
M60, 1,082)
127
($905)
M,847, 3,038)
Female
9
$237
(-167, 642)
458
($645)
(-3,011, 1,722)
Notes: Heart disease diagnosis based on ICD-9 code of 410-414. Sample weights used and complex survey design
taken into account in calculating standard errors. Prescription drug estimates exclude those with a source of
payment listed as Medicare. () denotes a negative value.
4.2.3 Indirect Costs: Morbidity
The estimated labor productivity losses resulting from increased morbidity in people with IHD
are a bit lower than estimated losses for those with CLD. Table 4-9 reports estimated labor productivity
losses per person with IHD of about $470 to $1,200, with a portion of those losses arising because people
with heart disease have a 3 percentage point higher probability of being unable to work than the
analogous population without heart disease. Among those with heart disease who do work, we estimate
work-loss days attributable to heart disease of 1.7 for females and 1 for males. The average labor
productivity loss across age and sex subgroups is $880 per year.
Table 4-9. Morbidity-Related Indirect Costs Associated with Ischemic Heart Disease—
Estimated Annual Per-Person Productivity Losses by Age-Gender Category
Age
Labor Productivity
Losses
Household Productivity
Losses
Total Productivity Losses
Age 65+ years
$878
$315
$1,193
Age 65-74 years
Male
$1,156
$210
$1,366
Female
$517
$449
$966
Age 75+ years
Male
$969
$192
$1,162
Female
$465
$385
$850
4-6
-------
Estimated household productivity losses in older people with heart disease are $315 and range
from $192 to $449 per year within specific age and sex subgroups. These estimates are based on
estimated bed days attributable to heart disease of 10 for females and 6 for males.
4.2.4 Indirect Costs: Mortality
The estimated annual
number of deaths in 2000
with IHD as the cause is over
443,000—over four times the
number of deaths resulting
from CLD in this age group
and almost three times the
number of stroke deaths. The
large number of deaths caused
by heart disease is associated
with high labor and household
productivity losses. As
shown in Table 4-10, the
estimated annual labor
productivity losses resulting
from heart disease mortality
are $7 billion, while the
estimated annual labor and
household productivity losses are almost $35 billion.
Table 4-10. Mortality-Related Indirect Costs Associated with
Ischemic Heart Disease—Estimated Aggregate
Productivity Losses by Age Category
Age
Number
of Deaths
Aggregate Labor
Productivity
Losses
(in 1,000s)
Aggregate Labor
and Household
Productivity Losses
(in 1,000s)
Age 65+ years
443,494
$7,231,200
$34,535,761
Age 65-74 years
97,226
$4,959,013
$18,506,512
Age 75-84 years
169,993
$2,272,187
$16,029,249
Age 85+ years
176,275
—
—
Notes: Ischemic heart disease diagnosis in mortality data is based on ICD-9
codes of 410-414. Mortality related indirect costs for 85+ not estimated due
to lack of data on expected productivity for this age group.
4.3 Stroke
As summarized in
Table 4-11, we estimate that
approximately 2.6 million
older Americans experienced
strokes in 2000. We also find
that over 150,000 deaths in
individuals 65 and older were
primarily attributable to
stroke. Based on these
estimates, we calculated
aggregate direct (except for
nursing home) and indirect
costs from stroke in this
population to be
approximately $18.1 billion.
These results are further
described and disaggregated
in the sections below.
Table 4-11. Costs of Illness Associated with Stroke—
Estimated Aggregate Costs in 2000 for
Individuals 65 and Older
Cost Category
Direct Costs
Medical
Prescription drug
Nursing home
Indirect Costs
Morbidity
Mortality
Total Costs of Illness
Number of
Affected
Individuals
Mean Cost
($/person)
Aggregate Costs
2,535,378
2,535,378
111,000
2,535,378
156,000
$3,725
$945
$2,478
$9,444,281,366
$2,395,931,783
$6,281,881,402
$10,423,691
— $18,132,518,242
4-7
-------
4.3.1 Direct Costs: Medical
Table 4-12 summarizes our estimates of stroke-related medical costs among Medicare fee-for-
service elderly beneficiaries. We estimate that roughly 2 million (7.5 percent of) beneficiaries
experienced stroke in 2000. The annual medical costs per beneficiary averaged $3,700 per year, which is
significantly lower than for CLD and IHD. Extrapolating these prevalence and average cost estimates to
the entire population of noninstitutionalized adults over 65 in 2000 results in an aggregate estimate of
stroke-related medical costs of $9.4 billion (shown in Table 4-11).
Table 4-12. Direct Medical Costs Associated with Stroke—Number of Beneficiaries and
Average Medical Payments by Type of Service and Age-Gender Category
Beneficiaries with a Diagnosis
Direct Medical Costs
($/Person/Year)
Number
Percent
95% CI (±)
Mean
95% CI (±)
Total
1,975,324
7.5
(7.4,7.7)
3,725
(3,528, 3,923)
Type of Medical Service
Inpatient
479,335
1.8
(1.8, 1.9)
11,555
(10,959, 12,151)
Physician
1,161,419
6.1
(6.0, 6.3)
461
(439, 483)
Hospital outpatient
576,421
2.2
(2.1,2.3)
615
(568, 662)
Home health
143,443
0.5
(0.5,0.6)
3,282
(2,792, 3,592)
Durable medical equipment
202,001
0.8
(0.7, 0.8)
1,243
(1,091, 1,395)
Age-Gender Category
Age 65-74 Years
Male
376,615
6.4
(6.1,6.7)
3,658
(3,102, 4,214)
Female
334,302
4.8
(4.6, 5.0)
3,663
(3,227, 4,100)
Age 75-84 Years
Male
433,262
11.4
(10.9, 11.9)
3,682
(3,209, 4,155)
Female
460,068
7.7
(7.4,8.1)
3,785
(3,397, 4,172)
Age 85+ Years
Male
136,725
13.4
(12.5, 14.4)
3,248
(2,765, 3,732)
Female
234,353
9.0
(8.5,9.5)
4,165
(3,748, 4,581)
Note: Geographic weights used to obtain unbiased nationally representatives estimates for Medicare elderly
population.
As is the case with CLD and IHD, the prevalence of stroke was found to be significantly higher
for males than for females in each of the three age categories. Similar to IHD, the prevalence of stroke
for both males and females was significantly higher in the older age categories, going from 6 percent of
men and 5 percent of women between 65 and 74 years to 13 percent and 9 percent, respectively, for those
85 and older. The average medical costs again show relatively little variation across age and gender
categories, varying between $3,600 and $4,200 for stroke.
4.3.2 Direct Costs: Self-Administered Prescription Drugs and Nursing Home Care
Our assessment of the prescription drug and nursing home costs associated with stroke,
summarized in Table 4-13, resulted in generally positive incremental cost estimates for prescription
drugs. For the total population 65 years and older, a stroke diagnosis in the past year was associated with
significantly higher prescription drug costs of about $945 per person per year. For males and females in
the 65 to 74 years age group, costs were even higher—$1,200 to $1,700 per year. Although a stroke
diagnosis was also associated with positive incremental prescription drug costs for the 75 to 84 years age
group, the magnitude was much smaller—approximately $500 to $1,000. For the oldest age group, stroke
was not associated with significantly higher prescription drug costs.
4-8
-------
Table 4-13. Other Direct Costs Associated with Stroke—Incremental Cost Estimates for
Prescription Medications and Nursing Home Services by Age-Gender
Category
Prescription Drug Incremental Costs
Nursing Home Incremental Costs
Mean
Mean
Sample
($/Person/
Sample
($/Person/
Age
Size
Year)
95% CI (±)
Size
Year)
95% CI (±)
Age 65+ years
114
$945
(596, 1,294)
73
($403)
(-5,089, 4,283)
Age 65-74 years
Male
30
$1,708
(687, 2,728)
7
($2,103)
(-30,046, 25,840)
Female
23
$1,215
(353, 1,622)
8
$3,632
(-14,938, 22,203)
Age 75-84 years
Male
16
$507
(-133, 1,147)
10
($9,402)
(-14,138, 4,665)
Female
23
$988
(353, 1,622)
19
($2,991)
(-9,744, 3,762)
Age 85+ years
Male
11
($177)
(-612, 257)
4
$21,157
(-14,511, 56,826)
Female
11
$34
(-386, 455)
25
($670)
(-7,049, 5,709)
Notes:. Stroke diagnosis based on ICD-9 code of 430-434 and 436. Sample weights used and complex survey
design taken into account in calculating standard errors. Prescription drug estimates exclude those with a source
of payment listed as Medicare. () denotes a negative value.
Somewhat surprisingly, nursing home costs for those with a reported stroke were no higher than
costs for those without stroke as a diagnosis. Although the estimated excess cost for males in the 85 years
and older age group was $21,000, this estimate is not significantly different from zero. Moreover,
aggregate nursing home costs for individuals in the NNHS with a stroke diagnosis were no higher than
nursing home costs for those without a stroke diagnosis. Due to the small sample size and high standard
errors, these estimates of stroke related nursing home costs are not included in our aggregate cost
estimates.
4.3.3 Indirect Costs: Morbidity
According to the results
shown in Table 4-14, the annual
labor productivity losses
resulting from increased
morbidity in older people who
experience a stroke are
relatively high, ranging from
about $1,000 to $2,400 per
person. These losses result in
large part from the estimated
increase in the probability of
being unable to work among
people with a reported stroke—
6 percentage points higher than
average for the over-65
population. For those with a
stroke diagnosis who did work (4.8
could be attributed to stroke, while
Table 4-14. Morbidity-Related Indirect Costs Associated
with Stroke—Estimated Annual Per-Person
Productivity Losses by Age-Gender Category
Labor
Household
Total
Productivity
Productivity
Productivity
Age
Losses
Losses
Losses
Age 65+ years
$1,820
$658
$2,478
Age 65-74 years
Male
$2,403
$455
$2,858
Female
$1,069
$885
$1,954
Age 75+
Male
$2,015
$417
$2,432
Female
$961
$759
$1,720
percent), we estimated that females had about 3.2 work-loss days that
males had 1.7.
4-9
-------
Our estimates of annual household productivity losses among those with a stroke diagnosis range
from approximately $420 to $890. Although these estimates appear to be fairly low, because of the
relatively low value of household productivity in older adults ($11,000 to $16,000 per year), we estimated
that the annual number of bed-loss days attributable to stroke was 20.4 for females and 13.5 for males.
These estimates suggest that stroke has a large impact on the ability of older people to perform simple
household duties, including cooking, cleaning, or even providing care for grandchildren. Total labor and
household productivity losses associated with stroke for the 65 years and older population are almost
$2,500 per person per year.
4.3.4 Indirect Costs: Mortality
As reported in
Table 4-15, the productivity
losses associated with stroke
as the cause of death are very
high. We estimate that
approximately 156,000
Americans over 65 years
would have likely died from
stroke in 2000. Over 40
percent of those estimated
deaths were among people
over 85 years of age. The
present value of labor
productivity losses resulting
from stroke deaths exceeds $2
billion. However, because a
Table 4-15. Indirect Costs Resulting from Increased
Mortality Associated with Stroke—Estimated
Aggregate Productivity Losses by Age Category
Aggregate Labor
Aggregate Labor and
Number
Productivity
Household
of
Losses
Productivity Losses
Age
Deaths
(in 1,000s)
(in 1,000s)
Age 65+ years
156,000
$2,079,434
$10,423,691
Age 65-74 years
25,053
$1,277,826
$4,768,713
Age 75-84 years
59,972
$801,608
$5,654,978
Age 85+ years
70,975
—
—
Notes: Stroke diagnosis in mortality data is based on ICD-9 codes 430-438.
Mortality related indirect costs for 85+ not estimated due to lack of data on
expected productivity for this age group.
high percentage of the 65 years and older population is retired (about 73 percent based on our analysis of
NHIS), the present value of labor and household productivity losses is far higher and is estimated to
exceed $10 billion annually.
4.4 Lung Cancer
Compared to the other health
conditions examined in this report,
the prevalence of lung cancer among
the elderly is relatively low;
however, the mortality rate and
average costs of illness are
comparatively high. As shown in
Table 4-16, roughly 116,000 deaths
in the 65 and older population were
primarily attributable to lung cancer
in 2000, and the aggregate costs of
illness were estimated to be $4.4
billion, even excluding prescription
drug costs, nursing home costs, and
most morbidity-related productivity
losses (due to data limitations).
Further details regarding the
estimated direct and indirect costs
Table 4-16. Costs of Illness Associated with Lung
Cancer—Estimated Aggregate Costs in
2000 for Individuals 65 and Older
Cost Category
Number of
Affected
Individuals
Mean Cost
($/person)
Aggregate
Costs
Direct Costs
Medical 393,853
Prescription drug 393,853
Nursing home 17,760
Indirect Costs
Morbidity 393,853
Mortality 116,360
Total Costs of Illness —
$10,859 $4,276,854,803
$40 la
$157,935,240
$15,182,419
$4,449,972,462
aDoes not include household productivity losses. Prescription drug and
nursing home costs not included due lack of sufficient data.
associated with lung cancer are described below.
4-10
-------
4.4.1 Direct Costs: Medical
Estimates of medical costs associated with lung cancer among Medicare fee-for-service elderly
beneficiaries are summarized in Table 4-17. Using CAHPS-MFFS data, we estimate that about 300,000
(1.2 percent of) beneficiaries suffered from lung cancer in 2000. This rate is lower than any of the other
conditions examined, but the average medical costs per affected beneficiary, estimated to be nearly
$11,000, are almost twice as high as even CLD and IHD. As shown in Table 4-16, extrapolating these
prevalence and average cost estimates to the entire noninstitutionalized population of adults over 65 in
2000 results in an aggregate estimate of medical costs from lung cancer of $4.3 billion..
Table 4-17. Direct Medical Costs Associated with Lung Cancer—Number of Beneficiaries
and Average Medical Payments by Type of Service and Age-Gender Category
Beneficiaries with a Diagnosis
Direct Medical Costs
($/Person/Year)
Number
Percent
95% CI (±)
Mean
95% CI (±)
Total
306,853
1.2
(1.1,1.2)
10,859
(9,963, 11,756)
Type of Medical Service
Inpatient
139,141
0.5
(0.5, 0.6)
15,308
(14,030, 16,587)
Physician
234,149
0.9
(0.8,0.9)
3,070
(2,502, 3,638)
Hospital outpatient
148,165
0.6
(0.5, 0.6)
2,725
(2,334, 3,116)
Home health
25,928
0.1
(0.1,0.1)
2,093
(1,570, 2,616)
Durable medical equipment
21,118
0.1
(0.1,0.1)
1,195
(885, 1,505)
Age-Gender Category
Age 65-74 Years
Male
75,239
1.3
(1.2, 1.4)
11,949
(10,356, 13,542)
Female
71,305
1.0
(0.9, 1.1)
10,920
(9,119, 12,721)
Age 75-84 Years
Male
66,395
1.8
(1.6, 1.9)
10,575
(8,363, 12,788)
Female
61,153
1.0
(0.9, 1.2)
11,341
(9,083, 13,599)
Age 85+ Years
Male
14,594
1.4
(1.1, 1.8)
7,223
(4,851, 9,595)
Female
18,167
0.7
(0.6,0.9)
8,443
(5,235, 11,652)
Note: Geographic weights used to obtain unbiased nationally representatives estimates for Medicare elderly
population.
As with all of the other conditions examined in this report, the prevalence of lung cancer among
older Americans is significantly higher for males than for females; however, the rates of lung cancer are
relatively stable across the three age categories for females they vary between 0.7 and 1 percent and
for males they vary between 1.3 and 1.8 percent. In contrast, the estimated average medical costs are not
significantly different between men and women. Estimated average costs are somewhat lower for men
and women in the highest age group (85 and older), going from about $11,000 to $8,000, but this
difference is only statistically significant for men.
4.4.2 Direct Costs: Self-Administered Prescription Drugs and Nursing Home Care
The 2000 MEPS contained information about only a handful of older people with lung cancer.
Because of the small number of observations in MEPS on people with lung cancer, we were unable to
generate reliable estimates of the excess prescription drug costs for age and sex subgroups for people with
lung cancer. However, to be consistent with our analysis for the other five conditions, we generated
4-11
-------
incremental cost estimates for age and sex subgroups with at least three observations, and these estimates
are reported in Table 4-18. The estimated incremental cost of prescription drugs for the 65 years and
older population with lung cancer is about $1,400 per year. This estimate is not quite statistically
significant at the 95 percent confidence level, but the large mean suggests that the prescription drug costs
associated with lung cancer among the elderly are substantially larger than prescription drug costs in
those without lung cancer. Due to the very small sample sizes associated with these estimates, they are
not included in our aggregate cost of illness estimates.
Table 4-18. Other Direct Costs Associated with Lung Cancer—Incremental Cost
Estimates for Prescription Medications and Nursing Home Services by Age-
Gender Category
Prescription Drug Incremental Costs
Nursing Home Incremental Costs
Mean
Mean
Sample
($/Person/
Sample
($/Person/
Age
Size
Year)
95% CI (±)
Size
Year)
95% CI (±)
Age 65+ years
14
$1,351
(-117, 2,819)
29
$5,367
M,085, 14,820)
Age 65-74 years
Male
4
$2,279
(-1,993, 6,550)
5
($14,492)
(-36,609, 7,625)
Female
2
—
—
4
$13,049
(-13,390, 39,489)
Age 75-84 years
Male
2
—
—
3
($8,267)
(-16,502, -33)
Female
3
$2,414
(1,470, 3,358)
6
$16,867
(-18,764, 52,499)
Age 85+ years
Male
1
—
—
5
$11,099
(-614, 22,812)
Female
2
—
—
6
($717)
(-13,883, 12,448)
Notes: Lung cancer diagnosis based on ICD-9 codes of 162, 197, and 231. Sample weights used and complex
survey design taken into account in calculating standard errors. () denotes a negative value.
The number of individuals in the NNHS with lung cancer listed as a diagnosis was also relatively
small (three to six observations per age and sex subgroup). Our estimates of incremental nursing home
costs for people with lung cancer vary widely, and none is significantly different from zero. Because of
the small samples and resulting large standard errors, it is difficult to draw any conclusions about the
likely impact of lung cancer on nursing home costs from these findings.
4.4.3 Indirect Costs: Morbidity
The indirect costs resulting from increased morbidity in people with a lung cancer diagnosis, as
summarized in Table 4-19, are based solely on the estimated increased probability of being unable to
work among people with lung cancer. Among the 41 respondents in NHIS with a lung cancer diagnosis,
the probability of being unable to work was about 1.3 percentage points higher than for analogously
defined individuals without lung cancer. Resulting cost estimates are $211 to $530 per year, and average
$400 per year across all age and sex subgroups. The work loss days attributable to lung cancer could not
be estimated precisely because of the small percentage of those over 65 years with lung cancer who
reported working (about 7 percent). Further, lung cancer was not considered in our analysis of the
number of bed days attributable to a health condition.
4-12
-------
Table 4-19. Morbidity-Related Indirect Costs Associated with Lung Cancer—Estimated
Annual Per-Person Productivity Losses by Age-Gender Category
Age
Labor Productivity Household Productivity
Total Productivity Losses
Age 65+ years
$401 —
—
Age 65-74 years
Male
$530 —
—
Female
$234 —
—
Age 75+ years
Male
$444 —
—
Female
$211 —
—
4.4.4 Indirect Costs: Mortality
Mortality estimates for lung cancer are presented in Table 4-20. The estimated number of deaths
among those 65 years and older with lung cancer as the cause is about 116,000 in 2000—somewhat
higher than the estimated number of deaths from CLD yet lower than the estimated number of stroke
deaths. Almost 50 percent of these deaths were in the 65 to 74 years age group and resulted in high
estimated productivity losses. Estimated labor productivity losses resulting from lung cancer mortality
have a present value of $3.5 billion; estimated labor and household productivity losses have a present
value of $15 billion.
Table 4-20. Mortality-Related Indirect Costs Resulting Associated with Lung Cancer—
Estimated Aggregate Productivity Losses by Age Category
Age
Number of
Deaths
Aggregate Labor
Productivity Losses
(in 1,000s)
Aggregate Labor and
Household Productivity Losses
(in 1,000s)
Age 65+ years
116,360
$3,507,104
$15,182,419
Age 65-74 years
56,402
$2,876,801
$10,735,918
Age 75-84 years
47,156
$630,303
$4,446,501
Age 85+ years
12,802
—
—
Notes: Lung cancer diagnosis in mortality data is based on ICD-9 codes of 160-161 and 163-165 and excludes
197 and 231
4.5 Pneumonia
As summarized in Table 4-21, we estimate that approximately 1.9 million older Americans
experienced pneumonia in 2000, and pneumonia was the primary cause of about 83,000 deaths in
individuals 65 and older. Based on these estimates, we calculated aggregate direct and indirect (except
for morbidity) costs from pneumonia in this population to be approximately $11 billion. These results are
further described and disaggregated in the sections below.
4-13
-------
Table 4-21. Costs of Illness Associated with Pneumonia—Estimated Aggregate Costs
in 2000 for Individuals 65 and Older
Number of Affected Mean Cost
Cost Category Individuals ($/person) Aggregate Costs
Direct Costs
Medical 1,779,693 $5,623 $10,007,215,290
Prescription drug 1,779,693 $302 $537,467,369
Nursing home 78,440 $4,987 $391,180,280
Indirect Costs
Morbidity 1,779,693 — —
Mortality 83,264 — $4,695,729
Total Costs of Illness — — $10,940,558,668
Notes: Morbidity related indirect costs not estimated due to lack of data.
4.5.1 Direct Costs: Medical
Table 4-22 summarizes our estimates of medical costs associated with pneumonia. Among
Medicare fee-for-service beneficiaries in 2000, we estimate that almost 1.4 million (5.3 percent of)
beneficiaries aged 65 and older suffered from pneumonia. The annual medical costs per beneficiary
averaged $5,600 per year, which is very similar to the average estimates for CLD and IHD. Extrapolating
these prevalence and average cost estimates to the entire noninstitutionalized population of adults over 65
in 2000, we estimate aggregate medical costs associated with pneumonia to be $10 billion (shown in
Table 4-21).
The prevalence of pneumonia was found to be significantly higher among males and among those
in the higher age categories. For women 65 to 74, rates of pneumonia were about 3.3 percent compared
to 4.1 percent for males of the same ages. For those 85 years and older, rates of pneumonia were 8.2 and
12.3 percent, respectively, for females and males. The average medical costs for pneumonia showed
relatively little variation across age and gender categories, varying between $5,100 and $6,000.
4.5.2 Direct Costs: Self-Administered Prescription Drugs and Nursing Home Care
We used the 2000 MEPS to assess the incremental prescription drug costs associated with
pneumonia and the 1999 NNHS to estimate incremental nursing home costs associated with pneumonia.
As described in Table 4-23, our findings suggest that excess prescription drug costs for people who
reported having pneumonia during the study year were fairly low and were only significantly different
from zero for females in the 75 to 84 years age group. However, estimated incremental costs for the full
over-65 population with a pneumonia diagnosis are significantly different from zero—about $300 per
person per year.
Incremental nursing home costs for people who had pneumonia were generally positive and quite
large, ranging from about $3,500 for the oldest females to $15,000 for 75 to 84 year-old males. Although
none of the estimates is significantly different from zero, the large standard errors on the mean estimate
are due in large part to the wide variation in nursing home costs and the relatively small number of
individuals who had pneumonia as a diagnosis. For the total population over 65 years, the incremental
cost estimates suggest that a pneumonia diagnosis is associated with significantly higher nursing home
costs, averaging about $5,000 per person per year.
4-14
-------
Table 4-22. Direct Medical Costs Associated with Pneumonia—Number of Beneficiaries
and Average Medical Payments by Type of Service and Age-Gender Category
Beneficiaries with a Diagnosis
Direct Medical Costs
($/Person/Year)
Number
Percent
95% CI (±)
Mean
95% CI (±)
Total
1,386,567
5.3
(5.2,5.4)
5,623
(5,305, 5,940)
Type of Medical Service
Inpatient
646,530
2.5
(2.4, 2.6)
11,028
(10,436, 11,619)
Physician
1,078,170
4.1
(4.0, 4.2)
310
(295, 324)
Hospital outpatient
321,993
1.2
(1.2, 1.3)
298
(252, 344)
Home health
111,684
0.4
(0.4, 0.5)
1,837
(1,662, 2,012)
Durable medical equipment
28,733
0.1
(0.1,0.1)
1,089
(834, 1,345)
Age-Gender Category
Age 65-74 Years
Male
241,755
4.1
(3.9, 4.4)
5,626
(4,903, 3,650)
Female
229,631
3.3
(3.1,3.5)
5,103
(4,341, 5,866)
Age 75-84 Years
Male
270,826
7.1
(6.8, 7.5)
5,915
(4,935, 6,894)
Female
306,271
5.2
(4.9, 5.4)
5,638
(5,018, 6,2596)
Age 85+ Years
Male
125,536
12.3
(11.4, 13.3)
5,394
(4,707, 6,082)
Female
212,548
8.2
(8.0, 9.0)
5,950
(5,275, 6,564)
Note: Geographic weights used to obtain unbiased nationally representatives estimates for Medicare elderly
population.
Table 4-23. Other Direct Costs Associated with Pneumonia—Incremental Cost Estimates
for Prescription Medications and Nursing Home Services by Age-Gender
Category
Prescription Drug Incremental Costs
Nursing Home Incremental Costs
Mean
Mean
Sample
($/Person/
Sample
($/Person/
Age
Size
Year)
95% CI (±)
Size
Year)
95% CI (±)
Age 65+ years
145
$302
(64, 540)
175
$4,987
(824, 9,151)
Age 65-74 years
Male
26
($199)
M90, 91)
11
($5,160)
(-23,978, 13,658)
Female
48
$348
(-193, 889)
8
$3,995
M,739, 12,730)
Age 75-84 years
Male
18
$214
(-166, 595)
24
$14,890
(-868, 30,648)
Female
35
$882
(489, 1276)
30
$4,182
(-7,699, 16,064)
Age 85+ years
Male
6
($293)
(-665, 79)
34
$4,968
(-2,685, 12,621)
Female
12
$42
(-530, 616)
68
$3,486
(-2,057, 9,028)
Notes:. Pneumonia diagnosis based on ICD-9 code of 480-486 and 487. Sample weights used and complex survey
design taken into account in calculating standard errors. () denotes a negative value.
4-15
-------
4.5.3 Indirect Costs: Morbidity
Because pneumonia was not reported in the NHIS, we were unable to estimate the costs of
increased morbidity attributable to pneumonia.
4.5.4 Indirect Costs: Mortality
Mortality costs represent the present value of all future earnings (earnings and household
productivity) losses resulting from pneumonia, discounted at 5 percent. These costs are reported in
Table 4-24 and are estimated to exceed $916 million in labor market earnings losses alone. The present
value of earnings and household productivity losses resulting from pneumonia deaths is almost $4.7
billion. The estimated number of deaths with pneumonia indicated as the cause of death is approximately
83,000, with over half of deaths occurring among those 85 years and older.
Table 4-24. Mortality-Related Indirect Costs Associated with Pneumonia—Estimated
Aggregate Productivity Losses by Age Category
Age
Number of
Deaths
Aggregate Labor Productivity
Losses
(in 1,000s)
Aggregate Labor and
Household Productivity Losses
(in 1,000s)
Age 65+ years
83,264
$915,661
$4,695,729
Age 65-74 years
10,408
$530,854
$1,981,091
Age 75-84 years
28,789
$384,807
$2,714,638
Age 85+ years
44,067
$0
$0
Notes: Pneumonia diagnosis in mortality data is based on ICD-9 code of 480-486
4.6 Gastrointestinal Illness
Table 4-25.
Costs of Illness Associated with Gastrointestinal
Illness—Estimated Aggregate Costs in 2000 for
Individuals 65 and Older
Cost Category
Number of
Affected
Individuals
Mean Cost
($/person)
Aggregate
Costs
Of the six health
conditions examined in this
report for older Americans,
gastrointestinal illnesses are
estimated to impose the
lowest aggregate cost
burden; however, the costs
are still substantial. As
summarized in Table 4-25,
we estimated that in 2000
approximately 0.8 million
older Americans
experienced gastrointestinal
illnesses that were severe
enough to seek medical
attention, and we calculated
aggregate direct and
indirect costs from
gastrointestinal illness in
this population to be approximately $1 billion (even excluding nursing home and morbidity related
indirect costs)..
Direct Costs
Medical
Prescription drug
Nursing home
Indirect Costs
Morbidity
Mortality
813,519
813,519
35,520
813,519
573
$1,160
$76
$943,681,600
$61,827,415
$38,675
Total Costs of Illness
$1,005,547,690
Notes: Nursing home costs and morbidity related indirect costs not estimated
due to lack of sufficient data.
4-16
-------
4.6.1 Direct Costs: Medical
Table 4-26 summarizes our estimates of medical costs associated with gastrointestinal illness for
Medicare fee-for-service beneficiaries in 2000. We estimate that over 600,000 (2.4 percent of)
beneficiaries aged 65 and older experienced gastrointestinal illness that was severe enough to seek
medical attention. Of the six conditions analyzed in this report, gastrointestinal illness imposes the lowest
annual medical costs per affected beneficiary, averaging less than $1,200 per year. When these
prevalence and average cost estimates were extrapolated to the entire noninstitutionalized population of
adults over 65 in 2000, we estimate aggregate medical costs associated with gastrointestinal illness to be
$0.9 billion (shown in Table 4-25).
Table 4-26. Direct Medical Costs Associated with Gastrointestinal Illness—Number of
Beneficiaries and Average Medical Payments by Type of Service and Age-
Gender Category
Beneficiaries with a Diagnosis
Direct Medical Costs
($/Person/Year)
Number
Percent
95% CI (±)
Mean
95% CI (±)
Total
633,816
2.4
(2.3,2.5)
1,160
(1,057, 1,264)
Type of Medical Service
Inpatient
153,738
0.4
(0.4, 0.5)
5,217
(4,860, 5,574)
Physician
474,861
1.8
(1.7, 1.9)
132
(125, 139)
Hospital outpatient
184,115
0.7
(0.7, 0.7)
316
(280, 353)
Home health
11,066
0.0
(0.0,0.1)
2,505
(1,546, 3,464)
Durable medical equipment
188
0.0
(0.0, 0.0)
793
(-1,012, 2,597)
Age-Gender Category
Age 65-74 Years
Male
10,530
1.8
(1.6, 1.9)
642
(386, 900)
Female
175,163
2.5
(2.3, 2.7)
1,133
(908, 1,358)
Age 75-84 Years
Male
100,098
2.6
(2.4, 2.9)
1,031
(824, 1,238)
Female
156,930
2.6
(2.5,2.8)
1,375
(1,175, 1,574)
Age 85+ Years
Male
29,546
3.0
(2.4, 3.4)
1,272
(1828, 1,717)
Female
67,049
2.3
(2.3, 2.9)
1,685
(1,372, 1,999)
Note: Geographic weights used to obtain unbiased nationally representatives estimates for Medicare elderly
population.
Prevalence rates for gastrointestinal illness were found to be very similar across age and gender
categories. Only in the 65 to 74 age category were prevalence rates significantly different between men
and women, with rates for females at 2.5 percent compared to 1.8 percent for males. As with most other
conditions, the average medical costs for gastrointestinal illness showed relatively little variation across
age and gender categories, varying between $600 and $1,700 per beneficiary in 2000.
4.6.2 Direct Costs: Self-Administered Prescription Drugs and Nursing Home Care
Our analysis of prescription drug costs from the 2000 MEPS, summarized in Table 4-27, suggests
that the excess costs associated with gastrointestinal illness among those 65 to 74 years of age is $200 to
$750 per person for year. For the other age groups, our incremental cost estimates and the associated
confidence intervals indicate that prescription drug costs for those with a gastrointestinal illness diagnosis
are not significantly different from the prescription costs of those without a gastrointestinal diagnosis.
4-17
-------
Table 4-27. Other Direct Costs Associated with Gastrointestinal Illness—Incremental
Cost Estimates for Prescription Medications and Nursing Home Services by
Age-Gender Category
Prescription Drug Incremental Costs
Nursing Home Incremental Costs
Mean
Mean
Sample
($/Person/
Sample
($/Person/
Age
Size
Year)
95% CI (±)
Size
Year)
95% CI (±)
Age 65+ years
165
$76
(-101,253)
32
$22,495
(-593, 45,582)
Age 65-74 years
Male
34
$749
(438, 1,059)
2
—
—
Female
56
$215
(-167, 596)
1
—
—
Age 75-84 years
Male
22
$181
(-327, 690)
4
$45,552
(-57,356, 148,461)
Female
44
($198)
M89, 93)
11
$11,281
(-6,671,29,233)
Age 85+ years
Male
3
($685)
(_944, -425)
1
—
—
Female
6
$266
(-605, 1,137)
13
$35,123
(-13,532, 83,779)
Notes: Gastrointestinal illness diagnosis based on ICD-9 codes of 001-009 and 558. Sample weights used and
complex survey design taken into account in calculating standard errors. () denotes a negative value.
Our analysis of the excess nursing home costs associated with gastrointestinal illness also
produced small age and sex subgroups and large resulting standard errors. For the 65 to 74 years age
group, only three individuals had a diagnosis of gastrointestinal illness; these observations were excluded
from our analysis. For the over-65 nursing home population with a diagnosis of gastrointestinal illness,
estimated incremental costs exceeded $22,000. This estimate was not quite statistically significant and
may be driven in large part by the presence of other comorbidities that make nursing home residents more
susceptible to gastrointestinal illnesses. Given the small number of observations and large standard
errors, it is difficult to draw any conclusions about the likely impact of gastrointestinal illness on nursing
home costs; therefore, these estimates are not included in our aggregate cost estimates.
4.6.3 Indirect Costs: Morbidity
Gastrointestinal illness was not reported in the NHIS. Consequently, we were unable to estimate
the costs of increased morbidity attributable to gastrointestinal illness.
4-18
-------
4.6.4 Indirect Costs: Mortality
As shown in
Table 4-28, the estimated
number of deaths in 2000 with
gastrointestinal illness as the
cause was much lower than
estimated deaths for the other
five conditions. We estimated
that about 570 deaths among
those 65 years and older were
due to gastrointestinal illness.
Because over half of those
deaths occurred in the 85 years
and older age group, the
estimated indirect costs
resulting from gastrointestinal
illness mortality are much lower
than for the other conditions—a
present value of $7.8 million in
labor productivity losses and
$38.7 million in labor and
household productivity losses combined.
Table 4-28. Mortality-Related Indirect Costs Associated
with Gastrointestinal Illness—Estimated
Aggregate Productivity Losses by Age
Category
Aggregate
Aggregate Labor
Labor
and Household
Productivity
Productivity
Number of
Losses
Losses
Age
Deaths
(in 1,000s)
(in 1,000s)
Age 65+ years
573
$7,837
$38,675
Age 65-74 years
98
$4,999
$18,656
Age 75-84 years
212
$2,838
$20,019
Age 85+ years
263
—
—
Notes: Gastrointestinal illness diagnosis in mortality data is based on ICD-9
codes of 004 and 006-009 and excludes 001-003, 005, and 558.9.
Mortality related indirect costs for 85+ not estimated due to lack of data
on expected productivity for this age group.
4-19
-------
5. Discussion of Results
The population of older Americans is expected to increase rapidly over the next 50 years, and this
trend presents important challenges in the area of environmental health. Several of the most prevalent
and costly adverse health conditions among older Americans are known to be associated, at least in part,
with environmental exposures, and older adults are often more vulnerable to environmental hazards.
The purpose of this report is to contribute to a better understanding of the cost burden imposed by
environmentally related health effects among older Americans. In collaboration with EPA, we selected
six general health conditions for which environmental exposures are known or suspected to be important
contributing factors. Using available national data, we estimated direct and indirect costs for each of
these conditions among individuals 65 and older in 2000.
Of the six conditions analyzed, chronic lung disease and ischemic heart disease are found to have
the highest prevalence among the older population and to impose the largest aggregate cost. Both
conditions were found to affect between 10 percent and 20 percent of the elderly population and to
impose total costs of illness of over $35 billion in 2000. The aggregate medical costs for CLD and IHD
represent roughly 15 percent and 22 percent respectively of estimated total medical costs among
individuals 65 and older in the U.S.
The estimated prevalence and aggregate costs of stroke and pneumonia were both somewhat
lower, in each case affecting between 5 and 10 percent of older Americans and costing between $10
billion and $20 billion on aggregate in 2000. As expected lung cancer was found to have the lowest
prevalence among the elderly population, affecting roughly one percent, but the aggregate costs estimates
are nonetheless substantial. Even without estimates of self-administered prescription drug costs, nursing
home costs, or lost household productivity due to morbidity (again due to data limitations), aggregate
costs were estimated to be $4.5 billion.
Gastrointestinal illnesses were found to impose the lowest cost burden of the six conditions.
Affecting approximately two to three percent of the elderly population in 2000, we estimated aggregate
costs of illness for GIs (without nursing home or indirect morbidity costs) of $1 billion.
Although the estimates summarized above and described in detail in previous sections provide
useful and important insights into the costs of illness for these six conditions, several limitation and
uncertainties associated with these estimates must be recognized. Each of these issues has been discussed
in previous sections; however, we reiterate them here to ensure that they are properly interpreted.
First, the cost estimates developed in this report should not be interpreted as those specifically
attributable to environmental exposures. Unfortunately, the science and empirical evidence regarding the
epidemiological links between environmental exposures and these health outcomes are not sufficiently
advanced to estimate reliably this attributable fraction. Consequently, our estimates are more
appropriately interpreted as upper-bound estimates of environmentally related costs of illness for these
conditions.
Second, our estimates of the prevalence of each condition are based on a nationally representative
sample of Medicare fee-for-service beneficiaries, using ICD-9 codes (as listed in Table 2-1) to identify
the presence of each condition in 2000. These prevalence estimates are based on claims for services in
2000. They do not include individuals who had the underlying condition in 2000 but did not seek care;
therefore, they are likely to underestimate overall prevalence. Moreover, since Medicare FFS covered
roughly 75 percent of the 65 and older population in 2000, these estimates were extrapolated to the
remaining 25 percent of the population. This extrapolation is likely to overestimate prevalence in the
noninstitutionalized non-FFS population and underestimate prevalence in the nursing home population.
Third, portions of the direct medical cost estimates based on the CAPHS-MFFS data cannot be
interpreted as being exclusively attributable to the selected conditions. Medicare claims data for
inpatient, outpatient, and home care costs, include potentially more than one diagnosis code per condition;
therefore, portions of the reported costs may be attributable to other co-morbid conditions. For this
5-1
-------
reason, the direct medical cost estimates are best interpreted as providing upper-bound estimates of these
costs for each condition.
Fourth, the estimates of prescription drug and nursing home costs are based on considerably
smaller samples than were used for estimating medical costs; therefore, the confidence intervals for these
estimates are considerably larger. Estimates based on samples smaller than 100 are reported in Section 4,
but they were not included in our aggregate estimates (i.e., they were set equal to zero) due to their lack of
precision. Excluding these values implies that the aggregate direct-costs estimates for stroke, lung cancer,
and gastrointestinal illness are underestimated; however, the extent of underestimation is not known.
Fifth, to approximate the portion of prescription drug and nursing home costs that are specifically
attributable to the selected conditions, we calculated incremental costs (i.e., the difference between
average costs for those with and without the condition). This approach provides a rather crude
approximation of attributable costs, and, in some cases, the estimated incremental costs are negative. The
negative estimates for nursing home costs for IHD were assumed to be implausible and replaced with a
zero value in our aggregate direct cost estimates.
Sixth, our estimates of morbidity-related productivity losses are based in part on an analysis of
additional work loss and bed days using NHIS data. Due to data limitations associated with NHIS, we
were not able to estimate losses for pneumonia or gastrointestinal illness, and we were able to only
partially estimate losses for lung cancer. For this reason, the indirect costs for these conditions are
underestimates.
Seventh, to estimate the indirect costs resulting from premature mortality, we used NSVR data
that defined health conditions using slightly different ICD-9 codes. Because of these discrepancies, our
estimates of mortality costs for CLD, stroke, and lung cancer are slight overestimates of mortality costs,
while our estimate for gastrointestinal illness is a slight underestimate of actual mortality costs for the
conditions of interest.
Finally, to estimate mortality-related costs, we also applied age-specific productivity estimates
from Grosse (2003). Unfortunately, this source does not provide an estimate for expected productivity
among those 85 and older. By not including productivity losses (labor or household) for deaths in the 85
and older age group, we underestimated indirect costs of illness. However, these excluded losses are
expected to be small relative to the productivity losses in the younger age groups.
5-2
-------
6. References
Agency for Toxic Substance Disease Registry (ATSDR). 1990. "Toxicological Profile for Radon."
PB/91/180422/AS.
Agency for Toxic Substance Disease Registry (ATSDR). 1999. "Toxicological Profile for Lead."
PB/99/166704.
Agency for Toxic Substance Disease Registry (ATSDR). 2001. "Toxicological Profile for Asbestos."
PB/2001/109101.
Bennett, John V., Scott D. Holmberg, Martha F. Rogers, and Steven L. Solomon. 1987. "Infections and
Parasitic Diseases." American Journal of Preventive Medicine 3:102-114.
Brown, R.S., Clement, DG, Hill, JW and Retchin, SM: Do Health Maintenance Organizations
Work for Medicare? Health Care Financing Review, 15(l):7-23. 1993.
Drummond, M.F., B. O'Brien, G.L. Stoddart, and G.W. Torrance. 1997. Methods for the Economic
Evaluation of Health Care Programmes. 2nd Edition. Oxford: Oxford University Press.
Grosse, S.D. 2003. "Productivity Loss Tables (Appendix)." In Prevention Effectiveness: A Guide to
Decision Analysis and Economic Evaluation, 2nd Edition, A.C. Haddix, S.M. Teutsch, P.S.
Corso, eds. London: Oxford University Press.
Lucas, J.W., Schiller, J.S., and V. Benson. 2004. Summary Health Statistics for U.S. Adults: National
Health Interview Survey, 2001. National Center for Health Statistics, Vital Health Stat 10(218).
Morris, R.D., and R. Levin. 1995. "Estimating the Incidence of Waterborne Infectious Disease Related
to Drinking Water in the United States." In Assessing and Managing Health Risks from Drinking
Water Contamination: Approaches and Applications. Proceedings of the Rome Symposium,
September 1994.
Mounts, A.W., R.C. Holman, M.J. Clarke, J.S. Bresee, and R.I. Glass. 1999. "Trends in Hospitalizations
Associated with gastroenteritis among Adults in the United States, 1979-1995." Epidemiology
and Infection 123-8.
National Center for Health Statistics. Health, United States, 2003 With Chartbook on Trends in the Health
of Americans. Hyattsville, Maryland: 2003.
Payment, P., L. Richardson, J. Siemiatycki, R. Dewar, M. Edwardes, and E. Franco. 1991. "A
Randomized Trial to Evaluate the Risk of Gastrointestinal Disease due to Consumption of
Drinking Water Meeting Current Microbiological Standards." American Journal of Public
Health 81(6):703-708.
Peters, K.D., K.D. Kochanek, and S.L. Murphy. 1998. "Deaths: Final Data for 1996." National Vital
Statistics Reports 47(9). Hyattsville, MD: National Center for Health Statistics.
Pope, C.A., R.T. Burnett, M.J. Thun, E.E. Calle, D. Krewski, K. Ito, and G.D. Thurston. 2002. "Lung
Cancer, Cardiopulmonary Mortality, and Long-Term Exposure to Fine Particulate Air Pollution."
Journal of the American Medical Association 287(9): 1132-1141.
6-1
-------
Riley, G.F., A.L. Potosky, J.D. Lubitz, and L.G. Kessler. 1995. "Medicare Payments from Diagnosis to
Death for Elderly Cancer Patients by Stage at Diagnosis." Medical Care 33(8):828-841.
Sandler, R.S., J.E. Everhart, M. Donowitz, E. Adams, K. Cronin, C. Goodman, E. Gemmen, S. Shah, A.
Avdic, and R. Rubin. 2002. "The Burden of Selected Digestive Diseases in the United States."
Gastroenterology 122(5): 1500-1511.
Tolley, G., D. Kenkel, and R. Fabian. 1994. Valuing Health for Policy: An Economic Approach.
Chicago, IL: University of Chicago.
Trisolini M, M Klosterman, NT McCall, et al. 2002. Medicare Quality Monitoring System Version 4
Specifications - M4-Frequency, Rate, Length of Stay, and Costs to Hospital Discharges for
HCQIP Clinical Priority Areas of Heart Failure, Stroke, Pneumonia, and AMI by Age Group,
Sex, Race, Medicare Eligibility Status, Medicaid Enrollment Status, and Geographic Area,
Calendar Years 1992-2001. CMS Contract No. 500-95-0058, T.O. #16, February 2002.
U.S. Environmental Protection Agency (EPA). 1999. The Benefits and Costs of the Clean Air Act, 1990
to 2010. EPA410-R-99. Washington, DC: Office of Air and Radiation.
U.S. Environmental Protection Agency. What is the EPA Aging Initiative?
. Accessed on December 21, 2003.
6-2
-------
Appendix A
Review of Cost of Illness Studies for Selected Health Conditions:
Summaries of Methods, Data Sources, and Findings by Disease
-------
Table A-1. Chronic Lung Disease
Article Title
(Authors)
Relevant Medical
Condition(s)
Study Population
Description
Cost
Categories
Included
Annual Cost Base Cost Estimate(s)
Data Source(s) Estimate(s) Year 2000 Dollarsb
A comparison of
three approaches for
attributing
hospitalizations to
specific diseases in
cost analyses
(Ward, M. M„ H. S.
Javitz, W. M. Smith,
and A. Bakst)
COPD:
-COPD and
bronchitis (ICD-9
491)
-Emphysema (ICD-
9 492)
-Unspecified
obstructive
pulmonary disease
(ICD-9 496)
U.S. population (data
source population), age
45 or older
Prevalence-based
Hospitalization
National Hospital
Discharge Survey
1990
Attribute all to the
disease if primary
diagnosis:
-$3,449 per
hospitalization
-$711 million annual
total
Attribute portion to
disease based on
position in diagnoses
and comorbidities:
-$3,205 per
hospitalization
-$2.2 billion annual
total
Incremental analysis by
age and gender cohort:
-$2,361 per
hospitalization
-$1.6 billion annual
total
2000a Attribute all to the
disease if primary
diagnosis:
-$3,449 per
hospitalization
-$711 million
annual total
Attribute portion to
disease based on
position in
diagnoses and
comorbidities:
-$3,205 per
hospitalization
-$2.2 billion annual
total
Incremental analysis
by age and gender
cohort:
-$2,361 per
hospitalization
-$1.6 billion annual
total
A national study of
medical care
expenditures for
respiratory conditions
(Yelin, E., et. al.)
Respiratory
conditions
-All respiratory
conditions (ICD-9
491-494, 496, 500,
501)
-Asthma (ICD-9
493)
Data source population
Prevalence-based
Hospital stays
Physician visits
Nonphysician
visits
ER visits
Prescriptions
filled
Home health
care services
Other
Medical
Expenditure Panel
Survey (1996,
household data)
-All respiratory mean
for all services $3,753
-Asthma mean for all
services $2,973
1996 -All respiratory
mean for all
services $4,289
-Asthma mean for
all services $3,398
(continued)
-------
Table A-1. Chronic Lung Disease (continued)
Article Title
(Authors)
Relevant Medical
Condition(s)
Study Population
Description
Cost
Categories
Included
Annual Cost Base Cost Estimate(s)
Data Source(s) Estimate(s) Year 2000 Dollarsb
Acute exacerbation
of chronic bronchitis:
disease-specific
issues that influence
the cost-effectiveness
of antimicrobial
therapy
(Saint, S., et. al.)d
-Acute
exacerbations of
chronic bronchitis
(AECB)
-All acute and
chronic bronchitis
12,379 with AECB and
13,904 with all
bronchitis
Prevalence-based (per
hospitalization)
Inpatient -University
hospitalization HealthSystem
Consortium
Clinical Database
-University of
Michigan Health
System
AECB per
hospitalization:
-$6,285 from UMI data
-$6,625 from UHSC
data
All acute and chronic
bronchitis per
hospitalization:
-$6,287 from UMI data
-$6,524 from UHSC
data
1999 AECB per
hospitalization:
-$6,541 from UMI
data
-$6,895 from
UHSC data
All acute and
chronic bronchitis
per hospitalization:
-$6,543 from UMI
data
-$6,790 from
UHSC data
American Lung
Association fact
sheet: chronic
obstructive
pulmonary disease
(COPD), October
2003
(American Lung
Association)
COPD:
-Chronic bronchitis
-Emphysema
U.S.
Prevalence-based
Healthcare
expenditures
Indirect costs
General info from
ALA
-Annual cost to nation
$32.1 billion
-Healthcare
expenditures $18.0
billion
-Indirect cost of $14.1
billion
2003a -Annual cost to
nation $28.2 billion
-Healthcare
expenditures $15.8
billion
-Indirect cost of
$12.4 billion
(continued)
-------
Table A-1. Chronic Lung Disease (continued)
Cost
Article Title
Relevant Medical
Study Population
Categories
Annual Cost
Base
Cost Estimate(s)
(Authors)
Condition(s)
Description
Included
Data Source(s)
Estimate(s)
Year
2000 Dollarsb
An economic
Asthma (ICD-9
Adults (18 or over)
Hospital
-National Center
-Direct $1,197.8
1985
-Direct $2,752.3
evaluation of asthma
493)
Prevalence-based
inpatient care
for Health
million
million
in the United States
Emergency
Statistics
-Indirect $1,268.0
-Indirect $2,913.6
(Wiess, K. B., P. J.
room
-National
million
million
Gergen, and T. A.
Hospital
Hospital
-All $2,465.8 million
-All $5,665.9
Hodgson)
outpatient
Inpatient
physician
services
Visits to
physician
office
Medications
Loss of school
days
Loss of work
Mortality loss
Discharge Survey
-National
Ambulatory
Medical Care
Survey
-National Health
Interview Survey
-National
Medical Care
Utilization and
Expenditure
Survey
million
Capitation, managed
COPD:
5% sample of Medicare
Hospital
Medicare data
Expenditures per capita
1992
Expenditures per
care, and chronic
-Chronic
beneficiaries, age 65
inpatient
1992
for all aged Medicare
capita for all aged
obstructive
obstructive
and over
Hospital
beneficiaries with
Medicare
pulmonary disease
bronchitis with or
Prevalence-based (total
outpatient
COPD $8,482
beneficiaries with
(Grasso, M. E., et.
without mention of
expenditures for one
Physician
(compared to $3,511
COPD $11,637
al.)
acute exacerbation
(ICD-9 491.2)
-Emphysema (ICD-
9 492.x)
year)
Cost of
comorbid
condition
for all Medicare
beneficiaries)
(compared to
$4,817 for all
Medicare
beneficiaries)
(continued)
-------
Table A-1. Chronic Lung Disease (continued)
Cost
Article Title
Relevant Medical
Study Population
Categories
Annual Cost
Base
Cost Estimate(s)
(Authors)
Condition(s)
Description
Included
Data Source(s)
Estimate(s)
Year
2000 Dollarsb
Comparing the
-Mood disorders
U.S. nationally
Direct (health
1996 Medical
Direct per capita health
1996
Direct per capita
national economic
-Diabetes
representative sample
costs)
Expenditure Panel
costs for treatment of
health costs for
burden of five
-Heart disease
(noninstitutionalized)
Work loss
Survey
condition:
treatment of
chronic conditions
-Asthma
Prevalence-based
Total societal
-Heart disease $6,463
condition:
(Druss, B. G., et. al)
-Hypertension
costs
-Hypertension $569
-Asthma $663
Estimated work-loss
costs for persons with
condition:
-Heart disease $3.8
billion
-Hypertension $11.5
billion
-Asthma $3.4 billion
-Heart disease
$7,386
-Hypertension $650
-Asthma $758
Estimated work-loss
costs for persons
with condition:
-Heart disease $4.3
billion
-Hypertension
$13.1 billion
-Asthma $3.9
billion
Cost estimates for
Asthma (ICD-9
Total U.S.
Directlndirect
1987 National
-Total economic cost
1996
-Total economic
environmentally
493)
populationPrevalence-
include: -
Medical
$14 billion -Direct cost
cost $16 billion -
related
based
Mortality
Expenditure
$12 billion-Indirect
Direct cost $14
asthma(Farquhar, I.,
costs-
Survey
cost $2 billion
billion-Indirect cost
et. al.)c
Morbidity
$2 billion
costs (lost
workdays of
patient and
reduced
productivity of
patient)
(continued)
-------
Table A-1. Chronic Lung Disease (continued)
Cost
Article Title
Relevant Medical Study Population
Categories
Annual Cost
Base
Cost Estimate(s)
(Authors)
Condition(s) Description
Included
Data Source(s)
Estimate(s)
Year
2000 Dollarsb
Costs of
-COPD Representative sample
Hospital
National health
COPD All ages (> 34
1996
COPD all ages (>
Occupational COPD
-Asthma of U.S. population, 65+
inpatient
expenditures for
years)
34 years)
and Asthma
(ICD-9, 490-496) subgroup not included
Outpatient
1996
-Total = $18.6 trillion
-Total = $21.3
(Leigh, J., P.
Prevalence-based
(including ER)
Expenditures
-Per capita = $ 1,165
trillion
Romano, M.
Home health
allocated to
(total cost/estimated
-Per capita = $1,331
Schenker, and K.
care
COPD and
1996 COPD
(total cost/
Kreiss)
Physician
asthma using data
prevalence)
estimated 1996
office visits
from:
Asthma all ages (>19
COPD prevalence)
Nursing home
-National
years)
Asthma all ages (>
care
Hospital
-Total = $6.5 trillion
19 years)
Prescription
Discharge Survey
-Per capita = $445
-Total = $7.4
drugs
-Healthcare Cost
(total cost/ estimated
trillion
Medical
and Utilization
1996 asthma
-Per capita = $509
supplies (e.g.,
Project
prevalence)
(total cost/
oxygen)
-National
estimated 1996
Other medical
Ambulatory Med
asthma prevalence)
services
Care Survey
-National
Hospital
Ambulatory Med
Care Survey
(continued)
-------
Table A-1. Chronic Lung Disease (continued)
Cost
Article Title
Relevant Medical
Study Population
Categories
Annual Cost Base
Cost Estimate(s)
(Authors)
Condition(s)
Description
Included
Data Source(s)
Estimate(s) Year
2000 Dollarsb
Direct medical cost
COPD:
Data source population
Hospitalization
-1990 National
Total annual U.S. 1994
Total annual U.S.
of chronic obstructive
-Chronic bronchitis
Prevalence-based
Inpatient
Hospital
payment for care $6.6
payment for care
pulmonary disease in
(ICD-9 491)
physician
Discharge Survey
billion
$8.2 billion
the U.S.A.
-Emphysema (ICD-
services
-1993 Medicare
(Ward, M. M„ et. al.)
9 492)
Emergency
Public Use Files
-Unspecified
department
-1993 National
obstructive
visits
Hospital
pulmonary disease
Outpatient
Ambulatory
(ICD-9 496)
diagnostic/scre
Medical Care
ening
Survey
procedures
-1990 National
Nursing home
Ambulatory
stays
Medical Care
Hospice care
Survey
Home
-National
healthcare
Hospital
Prescription
Ambulatory
medications
Medical Care
Long-term
Survey
oxygen
(outpatient visits)
therapy.
-1992 National
Home and
Hospice Care
Survey (home
healthcare
component)
-1985 National
Nursing Home
Survey
-1987 National
Medical
Expenditure
Survey
(continued)
-------
Table A-1. Chronic Lung Disease (continued)
Cost
Article Title Relevant Medical Study Population Categories Annual Cost Base Cost Estimate(s)
(Authors) Condition(s)
Description
Included
Data Source(s)
Estimate(s)
Year
2000 Dollarsb
Direct medical costs COPD:
U.S. (population of data
Inpatient care
-1993 National
-COPD cost per
1996
-COPD cost per
of chronic obstructive -Chronic bronchitis
sources)
(hospital stays
Hospital
prevalent case: total
prevalent case: total
pulmonary disease: (ICD-9 490, 491)
Prevalence-based
and hospital
Discharge Survey
direct costs $896
direct costs $1,024
chronic bronchitis -Emphysema (ICD-
inpatient
-Healthcare Cost
-Chronic bronchitis
-Chronic bronchitis
and emphysema 9 492)
doctor visits)
and Utilization
cost per prevalent case:
cost per prevalent
(Wilson, L„ E. B.
Outpatient care
Project
total direct costs $816
case: total direct
Devine, and K. So)
(doctor visits,
-1993 National
-Emphysema cost per
costs $933
laboratory tests
Ambulatory
prevalent case: total
-Emphysema cost
and
Medical Care
direct costs $1341
per prevalent case:
medications)
Survey
total direct costs
Emergency
-National
$1,533
care and home
Ambulatory
and long-term
Medical Care
care (home
Survey:
healthcare,
Outpatient
including
Department
equipment and
-National
supplies; and
Ambulatory
nursing home
Medical Care
visits)
Survey:
Emergency
Department 1993
-National Home
and Hospice Care
Survey
-Other sources of
cost estimates
(such as
clinicians)
(continued)
-------
Table A-1. Chronic Lung Disease (continued)
Cost
Article Title
Relevant Medical
Study Population
Categories
Annual Cost
Base
Cost Estimate(s)
(Authors)
Condition(s)
Description
Included
Data Source(s)
Estimate(s)
Year
2000 Dollarsb
Economic analysis of
-COPD
U.S. population, mean
Inpatient
Confronting
-Direct cost $4,119
2002
-Direct cost $3,761
the Confronting
-Chronic bronchitis
age 63
hospitalization
COPD survey
-Indirect cost $1,527
-Indirect cost
COPD survey: an
-Emphysema
Prevalence-based
Oxygen
(telephone
-Societal cost $5,646
$1,394
overview of results
-Had symptoms
therapy
interviews with
-Societal cost
(Wouters, E. F.)
consistent with
Prescribed
patients and
$5,156
chronic bronchitis
medication
physicians)
Influenza
vaccination
Laboratory
tests
Scheduled
visits to PCP or
specialist
Unscheduled
visits to PCP or
specialist
ER visits
Lost
productivity
(work loss
days)
Economic burden of
Respiratory
Beneficiaries with at
Actual cash
1997 claims data
Health-care and work-
1997
Health-care and
respiratory infections
infections:
least one of the
payments by
loss costs $4,397 per
work-loss costs
in an employed
-Pneumonia
conditions of a national
the employer
beneficiary (and $6,838
$4,888 per
population
-Chronic bronchitis
Fortune 100
for:
per employee)
beneficiary (and
(Birnbaum, H. G., M.
-Other specific
manufacturer (excludes
-Inpatient
$7,602 per
Morley, P. E.
respiratory
people over 65)
-Outpatient
employee)
Greenberg, and G. L.
infections
Prevalence-based
-Prescription
Colice)
(ICD-9 786, 465,
drug
462-463,466,461,
-Office
473, 490-491, 034,
-Other
472, 474, 480-486,
Disability costs
460, 464, 487)
Absenteeism
(continued)
-------
Table A-1. Chronic Lung Disease (continued)
Cost
Article Title
Relevant Medical
Study Population
Categories
Annual Cost
Base
Cost Estimate(s)
(Authors)
Condition(s)
Description
Included
Data Source(s)
Estimate(s)
Year
2000 Dollarsb
Informal caregiving
Chronic lung
Age 70 years or older, 3
Informal
1993 Asset and
-Additional yearly cost
2000
-Additional yearly
for chronic lung
disease
cohorts: no lung
caregiving
Health Dynamics
of informal care per
cost of informal
disease among older
disease, lung disease
Study (AHEAD)
case (lung disease and
care per case (lung
Americans
without associated
activity limitation
disease and activity
(Langa, K. M., et. al.)
activity limitations, and
lung disease with
limited associated
activity limitations
Prevalence-based
compared to informal
care for individuals
with no lung disease)
$2,200
-National annual cost
of >$2 billion for
informal caregiving for
chronic lung disease
limitation compared
to informal care for
individuals with no
lung disease) $2,200
-National annual
cost of >$2 billion
for informal
caregiving for
chronic lung disease
Lost income and
Chronic respiratory
Civilian,
Lost earnings
-Survey of
Average annual
1991
Average annual
work limitations in
diseases:-Lung or
noninstitutionalized
(SlPP)Work
Income and
earnings loss for adults
earnings loss for
persons with chronic
respiratory problem
U.S.
disability
Program
with chronic respiratory
adults with chronic
respiratory disorders
(general)-Asthma -
populationPrevalence-
(NHIS,
Participation
conditions $3,143 -
respiratory
(Ward, M. M„ H. S.
Bronchitis-
based
percentage
(October 91-
$5,272, 25-64—$1,267,
conditions $4,631 -
Javitz, W. M. Smith,
Emphysema-
only - no dollar
January 92)-
65+
$7,768, 25-64-
and M. A. Whan)
Allergic rhinitis-
Tuberculosis-Other
lung disease
amounts)
National Health
Interview Survey
(1993-1994)
$1,867, 65+
(continued)
-------
Table A-1. Chronic Lung Disease (continued)
Article Title
(Authors)
Cost
Relevant Medical
Study Population
Categories
Condition(s)
Description
Included
Data Source(s)
Broad range of
U.S. Population and
Hospital
HCFA Personal
diseases (ICD-9),
age- specific subgroups
inpatient
Health Care
including:
Prevalence-based
Outpatient
Expenditures for
-Neoplasms (140-
(including ER)
1995
239)
Home health
Expenditures for
-Nervous system
care
each health
and sense organs
Physician
service allocated
(320-389)
office visits
across age-, sex-,
-Circulatory system
Nursing home
and diagnosis-
(390-459)
care
specific groups
-Respiratory system
Lab and x-ray
using Medicare
(460-519)
Dental
and VA claims
-Skin and
Ambulance
and survey data,
subcutaneous tissue
services
such as:
(680-709)
Other
-National
professional
Hospital
services (e.g.,
Discharge Survey
podiatrists,
-National
optometrists)
Medical
Expenditure
Survey
-National Health
Interview Survey
-National
Ambulatory Med
Care Survey
-National
Nursing Home
Survey
Annual Cost Base Cost Estimate(s)
Estimate(s) Year 2000 Dollarsb
Medical
Expenditures for
Major Diseases, 1995
(Hodgson, T., and A.
Cohen)
>
i
O
65+:
Total = $300.2 billion
Per capita =
-$6,194, 65-74 yrs
-$10,365, 75-84 yrs
-$18,877, 85+
1995 65+:
Total = $355.1
billion
Per capita =
-$7,326, 65-74 yrs
-$12,259, 75-84 yrs
-$22,327, 85+
(continued)
-------
Table A-1. Chronic Lung Disease (continued)
Cost
Article Title
Relevant Medical
Study Population
Categories
Data
Annual Cost
Base
Cost Estimate(s)
(Authors)
Condition(s)
Description
Included
Source(s)
Estimate(s)
Year
2000 Dollarsb
Pharmacoeconomic
-COPD
Patients at Creighton
Drugs
Original
Median treatment costs
2000a
Median treatment
evaluation of COPD
-Emphysema
University Medical
Oxygen therapy
research
incurred over the entire
costs incurred over
(Hilleman, D. E„ N.
-Chronic bronchitis
Center Hospital and/or
Laboratory tests
duration of follow-up
the entire duration
Dewan, M. Malesker,
outpatient clinics age
Diagnostic tests
(per patient per year):
of follow-up (per
and M. Friedman)
35-80, with COPD (any
diagnosis), smokers (at
least 20 pack years),
and filled at least 70%
of their prescriptions in
the past year for
pulmonary medication
Prevalence-based
Procedures
Clinic visits
Emergency
department
visits
Hospitalizations
-Stage I $1,681
-Stage II $5,037
-Stage III $10,812
patient per year):
-Stage I $1,681
-Stage II $5,037
-Stage III $10,812
The burden of COPD
COPD:
Age 42-89 (mean age of
PCP visits
Confronting
Estimated mean annual
2002
Estimated mean
in the U.S.A.: results
-Chronic bronchitis
64.08), all patients
(scheduled and
COPD survey
cost per patient with
annual cost per
from the Confronting
-Emphysema
current or former
unscheduled)
(telephone
COPD:
patient with COPD:
COPD survey
smokers
Specialists visits
interviews with
-Total direct cost
-Total direct cost
(Halpern, M. T., R.
Prevalence-based
(scheduled and
patients and
$4,120
$3,762
H. Stanford, and R.
unscheduled),
physicians)
-Indirect $1,527
-Indirect $1,394
Borker)
Inpatient
hospitalization
ER visits
Prescription
medication
(specific Rxs)
Laboratory tests
General
categories of
Indirect (work
loss) and
Societal
(direct+indirect)
costs
-Societal $5,646
-Societal $5,156
(continued)
-------
Table A-1. Chronic Lung Disease (continued)
Relevant Cost
Article Title Medical Study Population Categories Annual Cost Base Cost Estimate(s)
(Authors) Condition(s) Description Included Data Source(s) Estimate(s) Year 2000 Dollarsb
The cost of asthma in
the emergency
department and
hospital
(Stanford, R„ T.
McLaughlin, and L.
J. Okamoto)
Asthma (ICD-9
493)
Patients from 27
hospitals across the U.S.
with asthma, Oct 1996-
Sept 1997, age 18 years
or older
Prevalence-based (per
visit)
Emergency
department
Hospitalization
Premier's
Perspective
Comparative
Database (PCD)
-Emergency
department only total
$234.48
-Emergency
department and
Hospitalization total
$3,102.53
1997 -Emergency
department only
total $260.67
-Emergency
department and
Hospitalization total
$3,449.02
The cost of health
-Asthma
Adult members of an
Hospital
Original
Annual costs
1996
Annual costs
conditions in a health
-Cerebrovascular
HMO (Kaiser
Laboratory
research
attributable to condition
attributable to
maintenance
disease
Permanente, Northern
Radiology
(after adjustment for
condition (after
organization
-COPD
California) between July
Outpatient visit
age, gender, and
adjustment for age.
(Ray, G. T„ et. al.)
-Congestive heart
1995 and June 1996
Home health
comorbidities):
gender, and
failure
Prevalence-based
Pharmacy
-Asthma $1,009
comorbidities):
-Ischemic heart
Cost of care
-Cerebrovascular
-Asthma $1,153
disease
from non-Kaiser
disease $7,114
-Cerebrovascular
-Lung cancer
Permanente
-COPD $6,859
disease $8,130
-Pneumonia
vendors
-Congestive heart
failure $7,176
-Ischemic heart disease
$5,169
-Lung cancer $8,612
-Pneumonia $9,499
-COPD $7,839
-Congestive heart
failure $8,201
-Ischemic heart
disease $5,907
-Lung cancer
$9,842
-Pneumonia
$10,856
(continued)
-------
Table A-1. Chronic Lung Disease (continued)
Cost
Article Title
Relevant Medical
Study Population
Categories
Annual Cost
Base
Cost Estimate(s)
(Authors)
Condition(s)
Description
Included
Data Source(s)
Estimate(s)
Year
2000 Dollarsb
The costs of treating
COPD:-Chronic
Noninstitutionalized
Inpatient
1987 National
Mean per person direct
1987
Mean per person
COPD in the United
bronchitis (ICD-9
population with COPD,
admissionsOutp
Medical
medical expenditure for
direct medical
States (Strassels, S.
491 )-Emphysema
many former or current
atient clinic
Expenditure
people with COPD
expenditure for
A., D. H. Smith, S.
(ICD-9 492)-
smokers, over 40 years
visitsOffice
Survey
$6,469 (approximately
people with COPD
D. Sullivan, P. S.
Chronic airway
old (n=228)Prevalence-
visits
(households)
25% COPD related)
$12,968
Mahajan)
obstruction, not
based (expenditures for
(specialists,
(approximately 25%
elsewhere
one year)
generalists, and
COPD related)
classified (ICD-9
other)Prescribed
496)
medicationsEme
rgency
department
visitsBed
daysRestricted
activity
daysLost
workdays
The economic burden
-COPD
U.S.
Medical
Division of
-COPD average cost
1993
-COPD average
of COPD
-Asthma
Prevalence-based
management
Epidemiology
per person per year
cost per person per
(Sullivan, S. D„ S. D.
-Influenza
Medical care
National Heart,
$1,522
year $1,971
Ramsey, and T. A.
-Pneumonia
services
Lung, and Blood
Total annual cost
Total annual cost
Lee)
-Tuberculosis
Hospitalization
Institute 1996
(direct and indirect):
(direct and indirect):
-Respiratory
Indirect:
report
-COPD $23.9 billion
-COPD $30.9
cancer
Mortality and
-Asthma $12.6 billion,
billion
morbidity (loss
-Pneumonia $7.8
-Asthma $16.3
of work time
billion
billion,
and
-Respiratory cancer
-Pneumonia $10.1
productivity)
$25.1 billion
billion
-Respiratory cancer
$32.5 billion
(continued)
-------
Table A-1. Chronic Lung Disease (continued)
Article Title
(Authors)
Relevant Medical
Condition(s)
Study Population
Description
Cost
Categories
Included
Annual Cost Base Cost Estimate(s)
Data Source(s) Estimate(s) Year 2000 Dollarsb
The economic burden
of non-influenza-
related viral
respiratory tract
infection in the
United States
(Fendrick, A. M., A.
S. Monto, B.
Nightengale, and M.
Sarnes)
Non-influenza
related viral
respiratory
infection (VRTI)
(cold with
complications
including acute
sinusitis, otitis
media, and lower
respiratory
infections)
Nationwide telephone
survey of U.S.
households
Prevalence-based
Outpatient
physician visits
Physician
encounters in an
emergency
department
Treatment
OTC medication
Prescription
medication
Missed
caregiver
workdays
Missed
workdays/Absen
teeism
-Medical
Expenditure
Panel Survey
(1997)
-National Health
Interview
Survey
-Bureau of
Labor Statistics
-Epidemiolog-
ical survey
(utilization only)
Extrapolated to U.S.
population:
-Total $40 billion
-Direct $17 billion
-Indirect $22.5 billion
2001 Extrapolated to U.S.
population:
-Total $38 billion
-Direct $16 billion
-Indirect $21.5
billion
The Morbidity and
Mortality Chartbook
on Cardiovascular,
Lung, and Blood
Diseases, 1996,
National Heart, Lung,
and Blood Institute,
May 1996c
COPD and allied
conditions (ICD-9
490-496)
Total U.S. population
Prevalence-based
Direct
Indirect:
-Mortality
-Morbidity (loss
of workdays)
Survey data
from:
-National Center
for Health
Statistics
-Health Care
Financing
Administration
-Other
-Total economic cost
$37.3 billion
-Direct cost $21.6
billion
-Indirect cost $16.2
billion
1998 -Total economic
cost $40.2 billion
-Direct cost $23.3
billion
-Indirect cost $17.5
billion
(continued)
-------
Table A-1. Chronic Lung Disease (continued)
Cost
Article Title
Relevant Medical
Study Population
Categories
Annual Cost
Base
Cost Estimate(s)
(Authors)
Condition(s)
Description
Included
Data Source(s)
Estimate(s)
Year
2000 Dollarsb
Treatment cost of
Acute
280,830 COPD patients
Hospital
-Medicare
Mean hospital cost:
1995
Mean hospital cost:
acute exacerbations
exacerbations of
(Medicare and non-
Physician
claims data
-$5,497 over 65
-$6,502 over 65
of chronic bronchitis
chronic bronchitis
Medicare)
Drug
-National
-$5,561 under 65
-$6,577 under 65
(Niederman, M. S.,
(AECB)
Prevalence-based
Healthcare and
-$5,516 all ages
-$6,524 all ages
et. al.)d
Cost Utilization
Project
-National
Ambulatory
Med Care
Survey
-National
Hospital
Ambulatory
Medical Care
Survey
Publication year used as base year
bAnnual Medical Care Consumer Price Index used for conversion. Average of January and February Medical Care CPI used for 2004.
cAs reviewed in Kirschstein, R. 2000. "Disease-specific estimates of direct and indirect costs of illness and NIH support." Office of the Director,. National Institutes of Health,
Dept of Health and Human Services.
dAs reviewed in Halpern MT, Higashi MK, Bakst AW, Schmier JK. 2003. "The economic impact of acute exacerbations of chronic bronchitis in the United States and Canada: a
literature review." J Manag Care Pharm Jul-Aug; 9(4): 353-9. Review.
-------
Table A-2. Ischemic Heart Disease
Study
Cost
Article Title
Relevant Medical
Population
Categories
Annual Cost
Base
Cost Estimate(s)
(Authors)
Condition(s)
Description
Included
Data Source(s)
Estimate(s)
Year
2000 Dollarsb
American Heart
Cardiovascular
U.S.
Hospital
-Hodgson, T. A. and A. J.
-Heart disease: direct -
2004
-Heart disease: direct -
Association.
disease (ICD-9 390-
population
Nursing home
Cohen. 1999. "Medical
$130.6 billion, indirect -
$88.6 billion, indirect -
Heart Disease and
459, 745-747):
(depends on
Physicians/other
care expenditures for
$108.0 billion, total -
$92.4 billion, total -
Stroke Statistics -
-Heart disease
data source)
professionals
selected circulatory
$238.6 billion
$204.2 billion
2004 Update.
-Coronary heart
Prevalence-
Drugs/other
diseases: opportunities for
-Coronary heart
-Coronary heart disease:
Economic cost of
disease
based
medical durables
reducing national health
disease: direct - $66.3
direct - $56.7 billion,
cardiovascular
-Stroke
Home health care
expenditures." Medical
billion, indirect - $66.9
indirect - $57.2 billion,
diseases.
-Hypertensive
Lost
Care 37:994-1012.
billion, total - $133.2
total - $114.0 billion
disease
productivity/morb
-National Health
billion
-Stroke: direct - $28.2
-Congestive heart
idity
Expenditures Amounts,
-Stroke: direct - $33.0
billion, indirect - $17.6
failure
Lost
and Average Annual
billion, indirect - $20.6
billion, total - $45.9
-Total cardiovascular
productivity/mort
Percent Change, by Type
billion, total - $53.6
billion
disease
ality
of Expenditures: Selected
billion
-Hypertensive disease:
Calendar Years 1980-2012
-Hypertensive disease:
direct - $35.5 billion,
(cms.hhs.gov).
direct - $41.5 billion,
indirect - $12.0 billion,
-Rice, D. P., T. A.
indirect - $14.0 billion,
total - $47.5 billion
Hodgson, and A. N.
total - $55.5 billion
-Congestive heart
Kopstein. 1985. "The
-Congestive heart
failure: direct - $22.8
economic cost of illness: a
failure: direct - $26.7
billion, indirect - $1.8
replication and update.
billion, indirect - $2.1
billion, total - $24.6
Health Care Finane Rev
billion, total - $28.8
billion
7:61-80.
billion
-Total cardiovascular
-Historic Income tables -
-Total cardiovascular
disease: direct - $194.0
People (census.gov)
disease: direct - $226.7
billion, indirect - $121.2
-Deaths for 358 Selected
billion, indirect - $141.7
billion, total - $315.2
Causes by 5-Year Age
billion, total - $368.4
billion
Groups, Race, and Sex,
billion
United States, 2000
(cdc.nchs/default/htm).
-Rice, Max, Michel, and
Sung. 2003. "Present
Value of Lifetime
Earnings, U.S. 2000."
Unpublished tables,
Institute for Health and
Aging, University of
California, San Francisco.
(continued)
-------
Table A-2. Ischemic Heart Disease (continued)
Study
Cost
Article Title
Relevant Medical
Population
Categories
Annual Cost
Base
Cost Estimate(s)
(Authors)
Condition(s)
Description
Included
Data Source(s)
Estimate(s)
Year
2000 Dollarsb
Comparing the
-Mood disorders
U.S.
Direct (health
1996 Medical Expenditure
Direct per capita health
1996
Direct per capita health
national economic
-Diabetes
nationally
costs)
Panel Survey
costs for treatment of
costs for treatment of
burden of five
-Heart disease
representative
Work loss
condition:
condition:
chronic
-Asthma
sample
Total societal
-Heart disease $6,463
-Heart disease $7,386
conditions
-Hypertension
(noninstitutio
costs
-Hypertension $569
-Hypertension $650
(Druss, B. G., et.
nalized)
-Asthma $663
-Asthma $758
al)
Prevalence-
based
Estimated work-loss
costs for persons with
condition:
-Heart disease $3.8
billion
-Hypertension $11.5
billion
-Asthma $3.4 billion
Estimated work-loss
costs for persons with
condition:
-Heart disease $4.3
billion
-Hypertension $13.1
billion
-Asthma $3.9 billion
Cost of
Ischemic heart
Patients at
Average monthly
Original research
Total cost $220.31 per
1999a
Total cost $229.28 per
medications for
disease (chronic)
Camden-On-
medication
month for medication
month for medication
patients with
(ICD 410-414)
Gauley
(cardiac and
($104.77 of that for
($109.03 of that for
ischemic heart
Medical
noncardiac, out-
cardiac medication)
cardiac medication)
disease in a rural
Center in
of-pocket and
family practice
West Virginia
reimbursed)
center
with chronic
(Patricoski, C.T.,
ischemic
and G. Steiner)
heart disease,
age 30-97
(104 patients)
Prevalence-
based
(continued)
-------
Table A-2. Ischemic Heart Disease (continued)
Article Title
(Authors)
Relevant Medical
Condition(s)
Study
Population
Description
Cost
Categories
Included
Data Source(s)
Annual Cost Base Cost Estimate(s)
Estimate(s) Year 2000 Dollarsb
Healthcare Use
among U.S.
Women Aged 45
and Older
(Hoerger, T., et.
al.)
-Cardiovascular
disease
-Breast neoplasms
-Gynecological
neoplasms
-Osteoporosis
Representativ
e sample of
U.S. women
45 and older
Prevalence-
based
Hospital inpatient
Physician office
Hospital
outpatient
Emergency
department
Nursing home
Home and
hospice care
For utilization:
-Healthcare Cost and
Utilization Project
-National Ambulatory Medical
Care Survey
-National Hospital Ambulatory
Medical Care Survey
-National Nursing Home
Survey
-National Home and Hospice
Survey
For cost:
-Converted charges to costs
using cost-to-charge ratio
-Medicare fee calculated using
Medicare's resource-based
relative value scale (RBRVS)
-Or used information from
literature
All ages (45+),
Total:
-Cardiovascular
disease = $60.4
billion
-Breast and
gynecological
neoplasms = $5.0
billion
1997 All ages (45+), total:
-Cardiovascular disease
= $67.1 billion
-Breast and
gynecological neoplasms
= $5.6 billion
Medical
Broad range of
U.S.
Hospital inpatient
HCFA Personal Health Care
65+:
1995 65+:
Expenditures for
diseases (ICD-9),
Population
Outpatient
Expenditures for 1995
Total = $300.2
Total = $355.1 billion
Major Diseases,
including:
and age-
(including ER)
Expenditures for each health
billion
Per capita =
1995
-Neoplasms (140-
specific
Home health care
service allocated across age-,
Per capita =
-$7,326, 65-74 yrs
(Hodgson, T., and
239)
subgroups
Physician office
sex-, and diagnosis-specific
-$6,194, 65-74 yrs
-$12,259, 75-84 yrs
A. Cohen)
-Nervous system and
Prevalence-
visits
groups using Medicare and VA
-$10,365, 75-84
-$22,327, 85+
sense organs (320-
based
Nursing home
claims and survey data, such
yrs
389)
care
as:
-$18,877, 85+
-Circulatory system
Lab and x-ray
-National Hospital Discharge
(390-459)
Dental
Survey
-Respiratory system
Ambulance
-National Medical Expenditure
(460-519)
services
Survey
-Skin and
Other
-National Health Interview
subcutaneous tissue
professional
Survey
(680-709)
services (e.g.,
podiatrists,
optometrists)
-National Ambulatory Med
Care Survey
-National Nursing Home
Survey
(continued)
-------
Table A-2. Ischemic Heart Disease (continued)
Study
Cost
Article Title
Relevant Medical
Population
Categories
Annual Cost
Base
Cost Estimate(s)
(Authors)
Condition(s)
Description
Included
Data Source(s)
Estimate(s)
Year
2000 Dollarsb
The cost of health
-Asthma-
Adult
Ho spitalLaborato
Original research
Annual costs
1996
Annual costs attributable
conditions in a
Cerebrovascular
members of
ryRadiologyOutp
attributable to condition
to condition (after
health
disease-COPD-
an HMO
atient visitHome
(after adjustment for
adjustment for age,
maintenance
Congestive heart
(Kaiser
healthPharmacyC
age, gender, and
gender, and
organization
failure-Ischemic
Permanente,
ost of care from
comorbidities): -
comorbidities): -Asthma
(Ray, G. T„ et.
heart disease-Lung
Northern
non-Kaiser
Asthma $1,009-
$1,153-Cerebro vascular
al.)
cancer-Pneumonia
California)
Permanente
Cerebrovascular disease
disease $8,130-COPD
between July
vendors
$7,114-COPD $6,859-
$7,839-Congestive heart
1995 and June
Congestive heart failure
failure $8,201-Ischemic
1996Prevalen
$7,176-Ischemic heart
heart disease $5,907-
ce-based
disease $5,169-Lung
Lung cancer $9,842-
cancer $8,612-
Pneumonia $10,856
Pneumonia $9,499
The Morbidity
-Heart diseases
U.S.
Direct
Survey data from:
Total economic cost:
1999
Total economic cost:
and Mortality
(ICD-9 390-398, 402,
population
Indirect:
-National Center for
-Heart disease $183.1
-Heart disease $190.6
Chartbook on
404-429)
Prevalence-
mortality and
Health Statistics
billion
billion
Cardiovascular,
-Coronary heart
based
morbidity (loss of
-Health Care Financing
-Coronary heart disease
-Coronary heart disease
Lung, and Blood
diseases (ICD-9 410-
workdays)
Administration
$99.8 billion
$103.9 billion
Diseases, 1998,
414)
-Other
Direct:
Direct:
National Heart,
Indirect costs from Dr.
-Heart disease $101.8
-Heart disease $105.9
Lung, and Blood
Dorothy Rice.
billion
billion
Institute, October
-Coronary heart disease
-Coronary heart disease
1998.c
$53.1 billion
$55.3 billion
Indirect:
Indirect:
-Heart disease $81.3
-Heart disease $84.6
billion
billion
-Coronary heart disease
-Coronary heart disease
$46.7 billion
$48.6 billion
Publication year used as base year
bAnnual Medical Care Consumer Price Index used for conversion. Average of January and February Medical Care CPI used for 2004.
cAs reviewed in Kirschstein, R. 2000. "Disease-specific estimates of direct and indirect costs of illness and NIH support." Office of the Director,. National Institutes of Health,
Dept of Health and Human Services.
-------
Table A-3. Stroke (Hemorrhagic or Ischemic)
Study
Cost
Article Title
Relevant Medical
Population
Categories
Annual Cost
Base
Cost Estimate(s)
(Authors)
Condition(s)
Description
Included
Data Source(s)
Estimate(s)
Year
2000 Dollarsb
American Heart
Cardiovascular
U.S.
Hospital
-Hodgson, T. A. and A. J.
-Heart disease: direct -
2004
-Heart disease: direct -
Association.
disease (ICD-9 390-
population
Nursing home
Cohen. 1999. "Medical
$130.6 billion, indirect -
$88.6 billion, indirect -
Heart Disease and
459, 745-747):
(depends on
Physicians/other
care expenditures for
$108.0 billion, total -
$92.4 billion, total -
Stroke Statistics -
-Heart disease
data source)
professionals
selected circulatory
$238.6 billion
$204.2 billion
2004 Update.
-Coronary heart
Prevalence-
Drugs/other
diseases: opportunities for
-Coronary heart
-Coronary heart disease:
Economic cost of
disease
based
medical durables
reducing national health
disease: direct - $66.3
direct - $56.7 billion,
cardiovascular
-Stroke
Home health care
expenditures." Medical
billion, indirect - $66.9
indirect - $57.2 billion,
diseases.
-Hypertensive
Lost
Care 37:994-1012.
billion, total - $133.2
total - $114.0 billion
disease
productivity/morb
-National Health
billion
-Stroke: direct - $28.2
-Congestive heart
idity
Expenditures Amounts,
-Stroke: direct - $33.0
billion, indirect - $17.6
failure
Lost
and Average Annual
billion, indirect - $53.6
billion, total - $45.9
-Total cardiovascular
productivity/mort
Percent Change, by Type
billion, total - $53.6
billion
disease
ality
of Expenditures: Selected
billion
-Hypertensive disease:
Calendar Years 1980-2012
-Hypertensive disease:
direct - $35.5 billion,
(cms.hhs.gov).
direct - $41.5 billion,
indirect - $12.0 billion,
-Rice, D. P., T. A.
indirect - $14.0 billion,
total - $47.5 billion
Hodgson, and A. N.
total - $55.5 billion
-Congestive heart
Kopstein. 1985. "The
-Congestive heart
failure: direct - $22.8
economic cost of illness: a
failure: direct - $26.7
billion, indirect - $1.8
replication and update.
billion, indirect - $2.1
billion, total - $24.6
Health Care Finane Rev
billion, total - $28.8
billion
7:61-80.
billion
-Total cardiovascular
-Historic Income tables -
-Total cardiovascular
disease: direct - $194.0
People (census.gov)
disease: direct - $226.7
billion, indirect - $121.2
-Deaths for 358 Selected
billion, indirect - $141.7
billion, total - $315.2
Causes by 5-Year Age
billion, total - $368.4
billion
Groups, Race, and Sex,
billion
United States, 2000
(cdc.nchs/default/htm).
-Rice, Max, Michel, and
Sung. 2003. "Present
Value of Lifetime
Earnings, U.S. 2000."
Unpublished tables,
Institute for Health and
Aging, University of
California, San Francisco.
(continued)
-------
Table A-3. Stroke (Hemorrhagic or Ischemic) (continued)
Study
Cost
Article Title
Relevant Medical
Population
Categories
Annual Cost
Base
Cost Estimate(s)
(Authors)
Condition(s)
Description
Included
Data Source(s)
Estimate(s)
Year
2000 Dollarsb
Effects of stroke
Stroke associated
Sample of US
Hospital
Global Utilization of
Average cost in
1993
Average cost in baseline
on medical
with acute
GUSTO I
Physician
Streptokinase and Tissue
baseline hospitalization:
hospitalization:
resource use and
myocardial infarction
patients,
Stroke procedures
Plasminogen Activator for
-Stroke $29,242
-Stroke $37,867
costs in acute
-Hemorrhagic
average age
Outpatient visits
Occluded Coronary
-No stroke $20,301
-No stroke $26,288
myocardial
-Non-hemorrhagic
of stroke
Institutional care
Arteries Study (GUSTO)
Baseline medical costs:
Baseline medical costs:
infarction.
cohort 69 and
-Hemorrhagic $26,619
-Hemorrhagic $34,470
GUSTO I
no stroke
-Nonhemorrhagic
-Nonhemorrhagic
Investigators.
cohort 60
$33,799
$43,768
Global Utilization
Prevalence-
-No stroke $20,301
-No stroke $26,288
of Streptokinase
based
and Tissue
Plasminogen
Activator for
Occluded
Coronary Arteries
Study
(Tung, C. Y„ et.
al.)
Inpatient costs of
Cerebrovascular
Administrativ
All hospital
-AMCC 192 Uniform
Mean cost per
1992
Mean cost per discharge:
specific
events (stroke):
e data set of
charges incurred
Hospital Discharge Data
discharge:
-SAH $54,868
cerebrovascular
-Subarachnoid
all hospital
during
Set
-SAH $39,994
-ICH $9,544
events at five
hemorrhage (SAH)
discharges
hospitalization
-Medicare cost to charge
-ICH $21,535
-ICI $13,557
academic medical
(ICD-9 430)
from 5
(changed to costs
ratios
-ICI $9,882
-TIA $6,383
centers.
-Intracerebral
academic
using cost to
-TIA $4,653
Mean cost per admission
(Holloway, R. G.,
hemorrhage (ICH)
medical
charge ratio)
Mean cost per
by age:
et. al.)
(ICD-9 431)
centers in
(omitted
admission by age:
-SAH $52,705 under 65
-Ischemic cerebral
1992
ambulance cost)
-SAH $38,417 under 65
-ICH $33,099 under 65
infarction (ICI) (ICD-
Prevalence-
-ICH $24,126 under 65
-ICI $13,811 under 65
9 434 or 436)
based
-ICI $10,067 under 65
-TIA $6,327 under 65
-Transient ischemic
-TIA $4,612 under 65
-SAH $62,958 over 65
attack (TIA) (ICD-9
-SAH $45,891 over 65
-ICH $26,594 over 65
435)
-ICH $19,385 over 65
-ICI $9,777 over 65
-ICI $13,413 over 65
-TIA $6,418 over 65
-TIA $4,678 over 65
(continued)
-------
Table A-3. Stroke (Hemorrhagic or Ischemic) (continued)
Study
Cost
Article Title
Relevant Medical
Population
Categories
Annual Cost
Base
Cost Estimate(s)
(Authors)
Condition(s)
Description
Included
Data Source(s)
Estimate(s)
Year
2000 Dollarsb
Lifetime cost of
Stroke:
U.S. (separate
Acute-care costs
Medicare claims data
Lifetime costs per
1990
Lifetime costs per person
stroke in the
-Subarachnoid
sources for
Long-term
1987 National Medical
person for first stroke
for first stroke occurring
United States
hemorrhage (SAH)
under and
ambulatory care
Expenditures Survey
occurring in 1990:
in 1990:
(Taylor, T. N., et.
-Intracerebral
over 65)
costs
Insurance claims data
-SAH $228,030
-SAH $365,296
al.)
hemorrhage (ICH)
Incidence-
Nursing home
U.S. Bureau of Economic
-ICH $123,565
-ICH $197,947
-Ischemic stroke
based
costs
Analysis
-ISC $90,981
-ISC $145,748
(ISC)
(lifetime cost)
Costs attributable
to stroke
recurrence
Prescription drug
costs
Indirect costs
(market and
nonmarket)
Medical
Broad range of
U.S.
Hospital
HCFA Personal Health
65+: Total = $300.2
1995
65+: Total = $355.1
Expenditures for
diseases (ICD-9),
Population
inpatientOutpatie
Care Expenditures for
billionPer capita = -
billionPer capita = -
Major Diseases,
including :-
and age-
nt (including
1995Expenditures for each
$6,194, 65-74 yrs-
$7,326, 65-74 yrs-
1995(Hodgson,
Neoplasms (140-
specific
ER)Home health
health service allocated
$10,365, 75-84 yrs-
$12,259, 75-84 yrs-
T., and A. Cohen)
239)-Nervous system
subgroupsPre
carePhysician
across age-, sex-, and
$18,877, 85+
$22,327, 85+
and sense organs
valence-based
office
diagnosis-specific groups
(320-389)-
visitsNursing
using Medicare and VA
Circulatory system
home careLab
claims and survey data,
(390-459)-
and x-rayDental
such as:-National Hospital
Respiratory system
Ambulance
Discharge Survey-
(460-519)-Skin and
servicesOther
National Medical
subcutaneous tissue
professional
Expenditure Survey-
(680-709)
services (e.g.,
National Health Interview
podiatrists,
Survey-National
optometrists)
Ambulatory Med Care
Survey-National Nursing
Home Survey
(continued)
-------
Table A-3. Stroke (Hemorrhagic or Ischemic) (continued)
Study
Cost
Article Title
Relevant Medical
Population
Categories
Annual Cost
Base
Cost Estimate(s)
(Authors)
Condition(s)
Description
Included
Data Source(s)
Estimate(s)
Year
2000 Dollarsb
National Center
Cerebrovascular
Total U.S.
Direct
Framingham Heart Study
-Total economic cost
1998
-Total economic cost
for Health
disease (stroke)
population
Indirect:
base data
$43.3 billion
$46.6 billion
Statistics, Centers
(ICD-9 430-438)
Prevalence-
-Mortality
-Direct cost $28.3
-Direct cost $30.5 billion
for Disease
based
-Morbidity (loss
billion
-Indirect cost $16.2
Control and
of workdays)
-Indirect cost $15
billion
Prevention
billion
(Thomas A.
Hodgson)0
National Heart,
Lung, and Blood
Institute, NIH
(Thomas J.
Thom)c
Outcomes and
Stroke (ICD-9 342.9,
Patients
Rehabilitation
Original research
Average Medicare
199T
Average Medicare
costs after hip
430,431,432.0,
diagnosed
facility
(Medicare reimbursements
Reimbursement during
Reimbursement during
fracture and
432.1,432.9, 434.0,
with
Skilled nursing
to sample)
the 6 months after
the 6 months after
stroke: A
434.1,434.9, 436,
condition,
facility
admission:
admission:
comparison of
437.3,438)
Medicare
Acute hospital
-RF total $23,133
-RF total $25,716
rehabilitation
coverage, 65
Outpatient care
-Subacute SNF total
-Subacute SNF total
settings
or older, acute
Physician
$15,522
$17,255
(Kramer, A. M.,
hospital stay
Durable medical
-Traditional SNF total
-Traditional SNF total
et. al.)
in the last 30
equipment
$11,299
$12,561
days, and no Home health or
previous SNF hospice care
or rehab Other
admission for
this event
Prevalence-
based
(continued)
-------
Table A-3. Stroke (Hemorrhagic or Ischemic) (continued)
Study
Cost
Article Title
Relevant Medical
Population
Categories
Annual Cost
Base
Cost Estimate(s)
(Authors)
Condition(s)
Description
Included
Data Source(s)
Estimate(s)
Year
2000 Dollarsb
The cost of health
-Asthma
Adult members
Hospital
Original research
Annual costs
1996
Annual costs attributable
conditions in a
-Cerebrovascular
of an HMO
Laboratory
attributable to condition
to condition (after
health
disease
(Kaiser
Radiology
(after adjustment for
adjustment for age,
maintenance
-COPD
Permanente,
Outpatient visit
age, gender, and
gender, and
organization
-Congestive heart
Northern
Home health
comorbidities):
comorbidities):
(Ray, G. T„ et.
failure
California)
Pharmacy
-Asthma $1,009
-Asthma $1,153
al.)
-Ischemic heart
between luly
Cost of care
-Cerebrovascular
-Cerebrovascular disease
disease
1995 and lune
from non-Kaiser
disease $7,114
$8,130
-Lung cancer
1996
Permanente
-COPD $6,859
-COPD $7,839
-Pneumonia
Prevalence-
based
vendors
-Congestive heart
failure $7,176
-Ischemic heart disease
$5,169
-Lung cancer $8,612
-Pneumonia $9,499
-Congestive heart failure
$8,201
-Ischemic heart disease
$5,907
-Lung cancer $9,842
-Pneumonia $10,856
Variations in
-Stroke (ICD-9 430-
Stroke or TIA
Inhospital
MetLife Study
Average inhospital
1995
Average inhospital
average charges
434.9)
as primary
charge (hospital
charge:
charge:
for strokes and
-Transient ischemic
diagnosis, age
stay and
-Stroke $11,010
-Stroke $13,022
TIAs: United
attack (TIA) (ICD-9
40 or over
physician
-TIA $4,940
-TIA $5,843
States, 1995
435-435.9)
(approximate
charge)
(Mushinski, M.)
median of 70)
Prevalence-
based (per
hospitalization)
"Publication year used as base year
bAnnual Medical Care Consumer Price Index used for conversion. Average of January and February Medical Care CPI used for 2004.
cAs reviewed in Kirschstein, R. 2000. "Disease-specific estimates of direct and indirect costs of illness and NIH support." Office of the Director,. National
Institutes of Health, Dept of Health and Human Services.
-------
Table A-4. Lung Cancer
Study
Cost
Article Title
Relevant Medical
Population
Categories
Annual Cost
Base
Cost Estimate(s)
(Authors)
Condition(s)
Description
Included
Data Source(s)
Estimate(s)
Year
2000 Dollarsb
Healthcare Use
-Cardiovascular
Representative
Hospital inpatient
For utilization:
All ages (45+), Total:
1997
All ages (45+), total:
among U.S.
disease
sample of U.S.
Physician office
-Healthcare Cost and
-Cardiovascular
-Cardiovascular disease
Women Aged
-Breast neoplasms
women 45 and
Hospital
Utilization Project
disease = $60.4 billion
= $67.1 billion
45 and Older
-Gynecological
older
outpatient
-National Ambulatory
-Breast and
-Breast and
(Hoerger, T.,
neoplasms
Prevalence-
Emergency
Medical Care Survey
gynecological
gynecological neoplasms
et. al.)
-Osteoporosis
based
department
-National Hospital
neoplasms = $5.0
= $5.6 billion
Nursing home
Ambulatory Medical Care
billion
Home and
Survey
hospice care
-National Nursing Home
Survey
-National Home and Hospice
Survey
For cost:
-Converted charges to costs
using cost-to-charge ratio
-Medicare fee calculated
using Medicare's resource-
based relative value scale
(RBRVS)
-Or used information from
literature
Medical
Broad range of
U.S.
Hospital inpatient
HCFA Personal Health Care
65+:
1995
65+:
Expenditures
diseases (ICD-9),
Population and
Outpatient
Expenditures for 1995
Total = $300.2 billion
Total = $355.1 billion
for Major
including:
age- specific
(including ER)
Expenditures for each health
Per capita =
Per capita =
Diseases, 1995
-Neoplasms (140-
subgroups
Home health care
service allocated across age-,
-$6,194, 65-74 yrs
-$7,326, 65-74 yrs
(Hodgson, T.,
239)
Prevalence-
Physician office
sex-, and diagnosis-specific
-$10,365, 75-84 yrs
-$12,259, 75-84 yrs
and A. Cohen)
-Nervous system
based
visits
groups using Medicare and
-$18,877, 85+
-$22,327, 85+
and sense organs
Nursing home
VA claims and survey data,
(320-389)
care
such as:
-Circulatory system
Lab and x-ray
-National Hospital Discharge
(390-459)
Dental
Survey
-Respiratory system
Ambulance
-National Medical
(460-519)
services
Expenditure Survey
-Skin and
Other
-National Health Interview
subcutaneous tissue
professional
Survey
(680-709)
services (e.g.,
-National Ambulatory Med
podiatrists,
Care Survey
optometrists)
-National Nursing Home
Survey
(continued)
-------
Table A-4. Lung Cancer (continued)
Study
Cost
Article Title
Relevant Medical
Population
Categories
Annual Cost
Base
Cost Estimate(s)
(Authors)
Condition(s)
Description
Included
Data Source(s)
Estimate(s)
Year
2000 Dollarsb
Medicare
Lung cancer (also
Age 65 and
Inpatient hospital
-Medicare payments
-Average payments per
1990
-Average payments per
payments from
includes breast,
older
Physician/supplie
-National Cancer
year $17,371
year $27,828
diagnosis to death
prostate,
Incidence and
r
Institute's Surveillance,
-Diagnosis to death
-Diagnosis to death
for elderly cancer
colon/rectum, and
Prevalence
Outpatient
Epidemiology, and End
$29,184
$46,752
patients by stage
bladder cancer)
(includes one
Other
Results (SEER) Program
Average Medicare
Average Medicare
at diagnosis
estimate of
Costs include
payments total (also
payments total (also lists
(Riley, G. F„ A.
annual costs,
Medicare
lists by age and gender):
by age and gender):
L. Potosky, J. D.
but all others
payments for all
-Less than 1 year
-Less than 1 year
Lubitz, and L. G.
are by phase
covered services,
survival $19,199
survival $30,756
Kessler)
or from
including care for
-Initial $17,518
-Initial $28,063
diagnosis to
conditions
-Continuing care
-Continuing care $6,896
death)
unrelated to
$4,305
-Pre-final $15,996
cancer.
-Pre-final $9,985
-Final $21,173
-Final $13,217
The costs of
Lung cancer
Data source
Direct
1990 National Center for
Estimated annual costs
1990
Estimated annual costs of
cancer care in the
population
Overall
Health Statistic study
of cancer care (U.S.
cancer care (U.S. 1990
United States:
Prevalence-
-Morbidity
1990 Malignant
Malignant neoplasm
implications for
based
-Mortality
neoplasm only):
only):
action
-Direct by value
-Mortality $58,773
-Mortality $94,152
(Schuette, H. L.,
and percentage
million (61.1%)
million (61.1%)
T. C. Tucker, M.
-Morbidity $8,895
-Morbidity $14,249
L. Brown, and A.
million (10.3%)
million (10.3%)
L. Potosky)
-Direct costs $27,458
-Direct costs $43,987
(28.6%)
(28.6%)
(continued)
-------
Table A-4. Lung Cancer (continued)
Article Title
(Authors)
Relevant Medical
Condition(s)
Study
Population
Description
Cost
Categories
Included
Data Source(s)
Annual Cost
Estimate(s)
Base
Year
Cost Estimate(s)
2000 Dollarsb
The economic
burden of cancer
(Brown, M. L.,
and L. Fintor)a
Lung cancer (ICD-9
162.2-162.9)
Total U.S.
population
Prevalence-
based
Direct
Indirect
-National Cancer
Institute's Surveillance
Epidemiology and End
Results (SEER) program
-Medicare claims records
(diagnosed between 1974
and 1981)
Direct $5.1 billion
1990
Direct $8.2 billion
The economic
burden of COPD
(Sullivan, S. D.,
S. D. Ramsey,
and T. A. Lee)
-COPD
-Asthma
-Influenza
-Pneumonia
-Tuberculosis
-Respiratory cancer
U.S.
Prevalence-
based
Medical
management
Medical care
services
Hospitalization
Indirect:
Mortality and
morbidity (loss of
work time and
productivity)
Division of Epidemiology
National Heart, Lung, and
Blood Institute 1996
report
-COPD average cost
per person per year
$1,522
Total annual cost (direct
and indirect):
-COPD $23.9 billion
-Asthma $12.6 billion,
-Pneumonia $7.8
billion
-Respiratory cancer
$25.1 billion
1993
-COPD average cost per
person per year $1,971
Total annual cost (direct
and indirect):
-COPD $30.9 billion
-Asthma $16.3 billion,
-Pneumonia $10.1
billion
-Respiratory cancer
$32.5 billion
The Period
Prevalence and
Costs of Treating
Nonmelanoma
Skin Cancers in
Patients over 65
Nonmelanoma skin
cancers
(ICD-9 173)
U.S.
Medicare
population
age 65 years
and older
Prevalence-
Physician office
treatment charges
(v. costs)
5% Sample of Medicare
claims for 1995
Total = $285 million
Per capita = $329
1995
Total = $337 million Per
capita = $389
Years of Age based
Covered by
Medicare
(Joseph A., T.
Mark, and C.
Mueller)
aAs reviewed in Kirschstein, R. 2000. "Disease-specific estimates of direct and indirect costs of illness and NIH support." Office of the Director. National
Institutes of Health, Dept of Health and Human Services.
bAnnual Medical Care Consumer Price Index used for conversion. Average of January and February Medical Care CPI used for 2004.
-------
Table A-5. Pneumonia
Article Title
(Authors)
Relevant Medical
Condition(s)
Study
Population
Description
Cost
Categories
Included
Data Source(s)
Annual Cost Base Cost Estimate(s)
Estimate(s) Year 2000 Dollarsb
Efficacy and
Cost-
Effectiveness
of V accination
against
Influenza
among Elderly
Persons Living
in the
Community
(Nichol, K, K.
Margolis, J.
Wuorenma, and
T. Von
Sternberg)
-Influenza
-Pneumonia
(ICD-9 480-487)
Sample of
enrollees of HMO
in Minneapolis; 65
years or older
(approx. 25,000 in
each claims year)
Prevalence-based
Hospital
inpatient
charges (v.
costs)
Minnesota HMO
administrative data
Cost per enrollee
(adjusted for health
status):
-With influenza
immunization = $46
-Without influenza
immunization = $67
1993 Cost per enrollee:
-With influenza
immunization = $60
-Without influenza
immunization = $87
Hospitalization
Pneumonia and
7,527 community
Hospitalization
Longitudinal study of aging
Pneumonia primary
1984 Pneumonia primary
for pneumonia
influenza (ICD-9
dwelling adults
(LSOA) (some link to
discharge diagnosis:
discharge diagnosis:
among older
480-487.8, 507-
aged 70 and older
NHIS) connected to
-Median cost $5,100
-Median cost $12,454
adults
507.1, 507.8, 997.3)
in 1984 (mean age
Medicare Automated Data
per hospitalization
per hospitalization
(Callahan, C.
of 76.5 in 1984)
Retrieval System (Medicare
Pneumonia secondary
Pneumonia secondary
M., and F. D.
Prevalence-based
Claims data)
discharge diagnosis:
discharge diagnosis:
Wolinsky)
(per
hospitalization)
-Median cost $10,100
per hospitalization
-Median cost $24,664
per hospitalization
Hospitalized
Community-
Age 65+ with CAP
Hospital cost
-1997 Medicare Provider
Mean hospital cost
1997 Mean hospital cost
community-
acquired Pneumonia
as admission and
Analysis and Review
$6,949
$7,725
acquired
(ICD-9 481, 482,
diagnosis
hospital discharge database
pneumonia in
485, or 486) or
discharge or
(MedPAR)
the elderly:
bacterial pneumonia
bacterial
-CB97-64, Population
age- and sex-
with pulmonary
pneumonia listed
Estimates Program, U.S.
related patterns
complaint at
as discharge
Bureau of the Census
of care and
admission
diagnosis coupled
-1997 Provider Specific
outcome in the
with pulmonary
File
United States
complaint at
-Hospital Cost Report
(Kaplan, V., et.
admission
Minimum Dataset
al.)
Prevalence-based
(cost per hospital
admission)
IV3
00
(continued)
-------
Table A-5. Pneumonia (continued)
Relevant
Study
Cost
Article Title
Medical
Population
Categories
Annual Cost
Base
Cost Estimate(s)
(Authors)
Condition(s)
Description
Included
Data Source(s)
Estimate(s)
Year
2000 Dollarsb
Outcome and
Ventilator
ICU patients at
Hospitalization
Original research
Estimated attributable
2003a
Estimated attributable
attributable cost
associated
Missouri Baptist
cost of ventilator
cost of ventilator
of ventilator-
pneumonia
Medical Center in
associated pneumonia
associated pneumonia
associated
St. Louis (10 bed
(controlling for other
(controlling for other
pneumonia
surgical ICU and
factors) $11,897
factors) $10,443
among intensive
10 bed medical
Hospital cost of
Hospital cost of patients,
care unit patients
ICU) from Jan. 19,
patients, mean:
mean:
in a suburban
1998, to Dec. 31,
-With VAP $70,568
-With VAP $61,946
medical center
1999, requiring
-Without $21,620
-Without $18,978
(Warren, D. K.,
mechanical
et. al.)
ventilation, mean
age 69
Prevalence-based
(per
hospitalization)
Relation between
Community-
Patients at 3
Hospital
Hospital bills (obtained for
Median cost of
2000a
Median cost of
length of hospital
acquired
hospital sites (two
roomLaboratory
each participant)
hospitalization -Total
hospitalization -Total
stay and costs of
pneumonia
in Pittsburgh, PA
tests and
$5,942 -Hospital room
$5,942 -Hospital room
care for patients
and one in Boston,
proceduresEmer
$3,465-Non-room
$3,465-Non-room
with community-
MA) in the
gency
$2,422
$2,422
acquired
Pneumonia Patient
departmentPhar
pneumonia(Fine,
Outcomes
macy and
M. J., et. al.)
Research Team
(PORT) cohort
study, age 18 years
or
intravenous
solutionsRadiolo
gy tests and
proceduresOther
olderPrevalence-
based (cost per
hospital stay)
(continued)
-------
Table A-5. Pneumonia (continued)
Relevant
Study
Cost
Article Title
Medical
Population
Categories
Annual Cost
Base
Cost Estimate(s)
(Authors)
Condition(s)
Description
Included
Data Source(s)
Estimate(s)
Year
2000 Dollarsb
The cost of health
-Asthma
Adult members of
Hospital
Original research
Annual costs
1996
Annual costs attributable
conditions in a
-Cerebrovascular
an HMO (Kaiser
Laboratory
attributable to
to condition (after
health
disease
Permanente,
Radiology
condition (after
adjustment for age,
maintenance
-COPD
Northern
Outpatient visit
adjustment for age,
gender, and
organization
-Congestive heart
California)
Home health
gender, and
comorbidities):
(Ray, G. T„ et.
failure
between July 1995
Pharmacy
comorbidities):
-Asthma $1,153
al.)
-Ischemic heart
and June 1996
Cost of care
-Asthma $1,009
-Cerebrovascular disease
disease
Prevalence-based
from non-Kaiser
-Cerebrovascular
$8,130
-Lung cancer
Permanente
disease $7,114
-COPD $7,839
-Pneumonia
vendors
-COPD $6,859
-Congestive heart
failure $7,176
-Ischemic heart
disease $5,169
-Lung cancer $8,612
-Pneumonia $9,499
-Congestive heart failure
$8,201
-Ischemic heart disease
$5,907
-Lung cancer $9,842
-Pneumonia $10,856
The Cost of
Community
Representative
Hospital
For 65+:
65+:
1995
65+:
Treating
acquired
sample of U.S.
inpatient
-National Health and
-Total = $4.8 billion
-Total = $5.7 billion
Community-
pneumonia
population and 65+
Inpatient
Nutrition Examination
-Per capita =
-Per capita =
Acquired
(ICD-9 480-487)
subgroup
physician
Survey III (to estimate
approximately $8000
approximately $9,500
Pneumonia
Prevalence-based
services
incidence)
(total cost/0.6 million
(total cost/ 0.6 million
(Niederman, M.,
Physician office
-Medicare Standard
people with inpatient
people with inpatient
J. McCombs, A.
visits
Analytical Files (Parts A
stay)
stay)
Unger, A. Kumar,
Emergency
and B; to estimate
and R. Popovian)
department
Outpatient clinic
Other facilities
(e.g., skilled
utilization/costs)
nursing, nursing
home)
(continued)
-------
Table A-5. Pneumonia (continued)
Relevant
Study
Cost
Article Title
Medical
Population
Categories
Annual Cost
Base
Cost Estimate(s)
(Authors)
Condition(s)
Description
Included
Data Source(s)
Estimate(s)
Year
2000 Dollarsb
The economic
-COPD
U.S.
Medical
Division of Epidemiology
-COPD average cost
1993
-COPD average cost per
burden of COPD
-Asthma
Prevalence-based
management
National Heart, Lung, and
per person per year
person per year $1,971
(Sullivan, S. D.,
-Influenza
Medical care
Blood Institute 1996 report
$1,522
Total annual cost (direct
S. D. Ramsey,
-Pneumonia
services
Total annual cost
and indirect):
and T. A. Lee)
-Tuberculosis
-Respiratory
cancer
Hospitalization
Indirect:
Mortality and
morbidity (loss
of work time
and
productivity)
(direct and indirect):
-COPD $23.9 billion
-Asthma $12.6 billion,
-Pneumonia $7.8
billion
-Respiratory cancer
$25.1 billion
-COPD $30.9 billion
-Asthma $16.3 billion,
-Pneumonia $10.1
billion
-Respiratory cancer
$32.5 billion
Unpublished
Pneumonia and
Total U.S.
Direct
Survey data from:
-Total economic cost
1999
-Total economic cost
(Thomas Thorn,
influenza (ICD-9
population
Indirect:
-National Center for Health
$25.6 billion
$26.6 billion
NHLBI)C
480-487)
Prevalence-based
-Mortality
Statistics
-Health Care Financing
Administration
-Other
Indirect costs from Dr.
Dorothy Rice.
-Direct cost $18.6
billion
-Indirect $7.0 billion
-Direct cost $19.4 billion
-Indirect $7.3 billion
"Publication year used as base year
bAnnual Medical Care Consumer Price Index used for conversion. Average of January and February Medical Care CPI used for 2004.
cAs reviewed in Kirschstein, R. 2000. "Disease-specific estimates of direct and indirect costs of illness and NIH support." Office of the Director, National
Institutes of Health, Dept of Health and Human Services.
-------
Table A-6. Gastrointestinal Illness
Article Title
(Authors)
Relevant Medical
Condition(s)
Study
Population
Description
Cost
Categories
Included
Data Source(s)
Annual Cost Base Cost Estimate(s)
Estimate(s) Year 2000 Dollarsb
Cost of Digestive -Digestive diseases Total U.S.
Diseases in the
U.S.
(Broun, D. M, and
J. E. Everhart)*
(ICD-9 01-09, 070, population
150, 151, 153-154,
155, 157, 152, 156,
158, 159, 211,455,
456.0-456.2, 530-
579, 787, 789)
-Gallbladder (ICD-9
574-575)
Prevalence-
based
Direct
Indirect:
-Mortality
-Morbidity (loss
of workdays)
Hospital statistics
Digestive disease:
-Total economic cost
$56.2 billion
-Direct cost $41.5 billion
-Indirect cost $14.7
billion
Gall bladder
-Total economic cost $4.7
billion
-Direct cost $4.4 billion
-Indirect cost $0.4 billion
1985 Digestive disease:
-Total economic cost
$129.1 billion
-Direct cost $95.4 billion
-Indirect cost $33.8
billion
Gall bladder
-Total economic cost
$10.8 billion
-Direct cost $10.1 billion
-Indirect cost $0.9
billion
Excess costs from
-Gastrointestinal
Tennessee
Hospital stay
Medicare/Medicaid Adjusted mean annual
1996a Adjusted mean annual
gastrointestinal
disease
Medicaid
Outpatient visits
payments payment for all types of
payment for all types of
disease associated
-Gastroduodenal
enrollees age
Prescription
medical care for study
medical care for study
with nonsteroidal
ulcer disease (ICD-9
65 or older (in
medication
gastrointestinal disorders:
gastrointestinal
anti-inflammatory
531-534)
1989)
-Nonuser $134
disorders:
drugs
-Gastritis/duodenitis
Prevalence-
-Occasional user $180
-Nonuser $153
(Smalley, W. E„
(ICD-9 535)
based
-Regular user $244
-Occasional user $206
M. R. Griffin, R.
-Gastrointestinal
(payments per
-Regular user $279
L. Fought, and W.
hemorrhage (ICD-9
person year)
A. Ray)
578)
-Nonneoplastic
gastrointestinal
disease (ICD-9 146-
171, 174-190) (by
procedure for
outpatient visits and
by type of medication
for prescriptions)
(continued)
-------
Table A-6. Gastrointestinal Illness (continued)
Study
Cost
Article Title
Relevant Medical
Population
Categories
Annual Cost
Base
Cost Estimate(s)
(Authors)
Condition(s)
Description
Included
Data Source(s)
Estimate(s)
Year
2000 Dollarsb
Gastrointestinal
Gastrointestinal
50,000 patients
Professional
Claims data in a
Total claims submitted
199T
Total claims submitted
illness in
illness (ICD-9 used
in database
charges
proprietary health
$152 million
$169 million
managed care:
to separate into
who met
Nonadmission
services research
healthcare
subcategories of GI):
eligibility
charges
database
utilization and
-Peptic ulcer disease
criteria
Admission
costs
-Nonulcer peptic
Prevalence-
charges
(Lim, D., et. al.)
disease
-Lower GI tract
disease
-Abdominal pain
-Hepatitis
-Liver disease
-Gallbladder/biliary
tract
-GI tract hemorrhage
-Other GI tract
problems
based (per
episode)
Prescriptions
Health Impact of
Digestive diseases
Total U.S.
Direct
Hospital statistics
-Total economic cost
1989
-Total economic cost
Peptic Ulcer in
-Peptic Ulcer (ICD-9
population
(including other
$4.92 billion
$8.59 billion
the United States
531-534)
Prevalence-
related
-Direct cost $3.55 billion
-Direct cost $6.20 billion
(Sonnenberg, A.,
based
nonhealth costs)
-Indirect cost $1.37
-Indirect cost $2.39
et. al.)*
Indirect:
billion
billion
Cost of Digestive
-Mortality
Diseases in the
-Morbidity (loss
U.S.
of workdays)
(Broun, D. M, et.
al.)*
(continued)
-------
Table A-6. Gastrointestinal Illness (continued)
Study
Cost
Article Title
Relevant Medical
Population
Categories
Annual Cost
Base
Cost Estimate(s)
(Authors)
Condition(s)
Description
Included
Data Source(s)
Estimate(s)
Year
2000 Dollarsb
Prevalence and
Upper
Patients with
Hospitalization
-Healthcare Cost &
Average charge per
1991,
Average charge per
cost of
gastrointestinal
upper
Utilization Project
admission all diagnoses:
1992
admission all diagnoses:
hospitalization for
complications (peptic
gastrointestinal
(HCUP-3)
-$8,661 in 1991
-$12,762 in 1991 (2000
gastrointestinal
and gastroduodenal
complications
-MarketScan
-$9,742 in 1992
dollars)
complications
ulcers with bleeding
as diagnosis,
Average charge per
-$13,365 in 1992 (2000
related to peptic
or perforation) (ICD-
average age 66
admission with UGIC as
dollars)
ulcers with
9 531.0-6, 532.0-6,
(HCUP-3) and
primary diagnosis:
Average charge per
bleeding or
533.0-6, 534.0-6)
52
-$12,970 in 1991
admission with UGIC as
perforation:
(MarketScan)
-$14,294 in 1992
primary diagnosis:
comparison of
Prevalence-
-$19,111 in 1991 (2000
two national
based (charge
dollars)
databases
per admission)
-$19,610 in 1992 (2000
(Kong, S. X., et.
al.)
dollars)
The burden of
-Nonfoodborne
Data source
Direct/IndirectH
-National Hospital
Nonfoodborne: -Total
1998
Nonfoodborne: -Total
selected digestive
gastroenteritis and
populationPrev
ospital
Discharge Survey-
$2,107 million-Direct
$2,270 million-Direct
diseases in the
other intestinal
alence-based
facilitylnpatient
National Ambulatory
$1,602 million-Indirect
$1,726 million-Indirect
United
infections (ICD-9
physician
Medical Care Survey-
$505 millionFoodborne: -
$544 millionFoodborne:
States(Sandler, R.
001-009 (subset),
servicesHospital
National Hospital
Total $1,119 million-
-Total $1,205 million-
S„ et. al.)
558.9)-Foodborne
OPDHospital
Ambulatory Medical
Direct $886 million-
Direct $954 million-
illness (ICD-9 001-
emergency
Care Survey
Indirect $233 million
Indirect $251 million
009 (subset), 558.9,
roomOffice
070.0, 070.1)-
visits DrugsLost
Additional digestive
work days
diseases
"Publication year used as base year
bAnnual Medical Care Consumer Price Index used for conversion. Average of January and February Medical Care CPI used for 2004.
cAs reviewed in Kirschstein, R. 2000. "Disease-specific estimates of direct and indirect costs of illness and NIH support." Office of the Director, National
Institutes of Health, Dept of Health and Human Services.
-------
Appendix B
Detailed Methods for Estimating Missed Work Days and Bed Days
Attributable to Stroke, Heart Disease, and Chronic Lung Disease
In Section 3, we provided a general description of our methods for estimating the indirect costs
resulting from increased morbidity. This appendix provides additional details on our approach for
estimating the missed work days attributable to stroke, heart disease, and chronic lung disease.
We used the 2001 NHIS to predict days of work missed and bed days attributable to each
condition. For the work-loss days analysis, we restricted our sample to those 65 years and older who
were employed in the labor market. We also excluded those with missing data. Our final analysis sample
included only 524 adults. For our bed-days analysis, we restricted the sample to all those 65 years and
older, regardless of employment status. Our final sample for this analysis included 3,930 older adults.
Because of the discrete nature of the variables of interest for our analysis (i.e., counts of days of
work lost and days spent in bed), we used a negative binomial regression with a long link to estimate
missed work days and bed days for people 65 years and older. By including indicator variables for stroke,
heart disease, and chronic lung disease, we were able to estimate the impact of each condition on work
loss and bed days. The regressions also controlled for other factors expected to influence the number of
missed work days, including sex, race, education, income, smoking status, and a variable indicating the
presence of a functional limitation (e.g., difficulty walking or standing). The work-loss days regression
also included variables to control for hourly versus salaried workers and specific categories of
occupations. The bed-days regression included variables to control for labor-force participation status
(i.e., working, retired, never worked, etc.).
Results from these regressions are shown in Tables B-l and B-2. In the work-loss days analysis,
our estimates are positive, suggesting that people with a reported stroke, chronic lung disease, and/or
heart disease have more work-loss days than those with none of these conditions. However, these
estimates are not statistically significant for any of the conditions. Results from the bed-days regression
are positive and highly statistically significant for stroke and heart disease, suggesting that those with
these conditions experience more bed days than those without, even controlling for a number of other
factors likely to affect serious illness.
To estimate the dollar value of work-loss and bed days attributable to stroke, heart disease, and
chronic lung disease, we used the following procedure:
1. Predicted annual work-loss (bed) days for each person in the sample.
2. For those with reported stroke, heart disease, and/or chronic lung disease, repeat Step 1
setting the indicator variable for the condition to zero. The resulting predictions represent
estimated work-loss (bed) days for those without the condition of interest who have similar
characteristics (i.e., sex, race, employment status, etc.).
3. Calculate the difference between predicted work-loss (bed) days for those with and without
the condition of interest (i.e., subtract value calculated in Step 2 from Step 1 value). This
calculation produces an estimate of the work-loss (bed) days attributable to each condition.
4. Multiply estimated sex-specific labor earnings (household productivity values) (Grosse,
2003) by estimates of work-loss days attributable to each condition from Step 3.
B-1
-------
Table B-1. Missed Work Days Analysis—Negative Binomial Regression (NBR) Results
Variable Descriptions. All except constant term
are indicator variables that = 1 if:
NBR
Coefficients
Robust
Standard
Errors
P-value
High school grad
-1.04
0.40
0.01
Had some college
-0.77
0.42
0.07
College grad
-0.72
0.59
0.22
Graduate degree
0.08
0.81
0.92
Female
0.20
0.43
0.64
Low income (<200% of poverty line)
-1.74
0.90
0.06
Middle income (<500% of poverty line)
-1.72
0.84
0.04
High income (>500% of poverty line)
-1.47
0.87
0.09
Former smoker
0.31
0.52
0.55
Never smoked
0.33
0.48
0.50
Non-Hispanic white
-0.88
0.53
0.10
Non-Hispanic black
0.28
0.68
0.68
Non-Hispanic other
-2.64
1.04
0.01
Salaried employee
0.31
0.37
0.40
No functional limitation reported
-1.05
0.31
0.00
Professional specialty occupation
1.47
0.64
0.02
Technician or related support occupation
1.33
0.95
0.16
Sales occupation
1.28
0.53
0.02
Administrative support occupation
0.91
0.45
0.04
Private household occupation
-3.26
0.89
0.00
Protective services occupation
-1.76
0.85
0.04
Household
0.27
0.53
0.61
Farming, forestry, and fishing occupation
0.27
0.95
0.77
Precision production, craft, and repair occupation
0.56
0.60
0.35
Operator, fabricator, or laborer
1.37
0.59
0.02
Transportation or material-moving occupation
0.33
0.82
0.68
Handler, equipment cleaner, helper, or laborer
-0.30
0.89
0.74
Self-report of STROKE
0.54
0.92
0.56
Self-report of CHRONIC LUNG DISEASE
0.98
0.69
0.15
Self-report of HEART DISEASE
0.30
0.39
0.44
Constant term
2.53
1.14
0.03
Notes: Estimates only for those 65 years and older in the 2001 NHIS who were currently working. Unweighted
sample size is 524.
B-2
-------
Table B-2. Bed-Days Analysis—Negative Binomial Regression (NBR) Results
Variable Descriptions. All except constant term
are indicator variables that = 1 if:
NBR
Coefficients
Robust
Standard
Errors
P-value
High school grad
-0.19
0.23
0.40
Had some college
-0.13
0.25
0.60
College grad
-0.47
0.26
0.07
Graduate degree
-0.11
0.31
0.72
Female
0.27
0.16
0.10
Low income (<200% of poverty line)
0.04
0.24
0.86
Middle income (<500% of poverty line)
0.11
0.26
0.68
High income (>=500% of poverty line)
-0.49
0.31
0.12
Former smoker
0.32
0.21
0.13
Never smoked
0.04
0.21
0.85
Smoking status unknown
-16.14
0.83
0.00
Non-Hispanic white
-0.26
0.23
0.26
Non-Hispanic black
0.19
0.29
0.50
Non-Hispanic other
0.36
0.44
0.42
Retired
0.81
0.22
0.00
Not working but previously worked
2.10
0.30
0.00
Never worked
1.38
0.38
0.00
No functional limitation reported
-1.34
0.18
0.00
Self-report of STROKE
0.96
0.26
0.00
Self-report of CHRONIC LUNG DISEASE
0.32
0.21
0.12
Self-report of HEART DISEASE
0.78
0.18
0.00
Constant term
0.57
0.38
0.14
Notes: Estimates only for those 65 years and older in the 2001 NHIS. Unweighted sample size is 3930.
Our estimates of work loss and bed days attributable to each condition are shown by sex in
Tables B-3 and B-4, respectively. Among older adults who worked, chronic lung disease was responsible
for the largest impact on work loss, or about 3 to 4.5 days of missed work per year. Considering bed
days, stroke was responsible for the largest number of bed days among the three conditions—13 annual
bed days for males and 20 annual bed days for females.
Note that, although our estimates for attributable work-loss days are not statistically different
from zero, we used them to estimate the cost associated with work loss in the 65 and older population that
works. We chose to include these work-related cost estimates because it is likely that our lack of
statistically significant results is due to the small overall sample size and the small number with each
condition of interest. In addition, because only about 5 to 11 percent of those with any of the conditions
worked, the estimated cost of losing 1 to 4.6 work days per year was low.
B-3
-------
Table B-3. Predicted Missed-Work Days for Workers 65 Years and Older by Condition
and Sex
Number of Predicted Missed-Work Days
Condition and Sex Respondents Attributable to Condition
Stroke
Male 9 1.72
Female 4 3.19
Chronic Lung Disease
Male 30 3.42
Female 26 4.59
Ischemic Heart Disease
Male 72 0.93
Female 46 1.74
Note: Predictions of work days missed were calculated using NBR results in Table B-l.
Table B-4. Predicted Bed Days for Respondents 65 Years and Older by Condition and
Sex
Number of Predicted Missed Work Days
Condition and Sex Respondents Attributable to Condition
Stroke
Male 135 13.45445
Female 209 20.43143
Chronic Lung Disease
Male 209 3.036422
Female 316 4.85797
Ischemic Heart Disease
Male 561 6.198227
Female 669 10.35484
Note: Predictions of bed days were calculated using NBR results in Table B-2.
B-4
-------
Appendix C
Direct Medical Costs Associated with Selected Health Conditions:
Annual Medicare Payments in 2000 by Region and State
-------
Table C-1. Direct Medical Costs Associated with Pneumonia—Annual Medicare
Payments in 2000 by Region and State
Direct Medical Costs
Beneficiaries with a Diagnosis
($/Person/Y ear)
Region (State)
Number
Percent
CI (±)
Mean
ci(±)
New England
Total
75,925
5.5
(5.0, 6.0)
6,719
(5,050, 7,989)
Connecticut
20,277
6.0
(5.0, 7.0)
6,163
(4,122, 8,204)
Maine
7,102
4.1
(2.9, 5.3)
5,887
(2,921, 8,843)
Massachusetts
35,954
6.3
(5.3,7.3)
8,079
(5,802, 10,355)
New Hampshire
5,850
4.3
(3.0, 5.5)
2,978
(1,102, 4,854)
Rhode Island
4,234
4.7
(2.8, 6.5)
6,494
(2,907, 10,081)
Vermont
2,508
3.5
(1.9, 5.1)
3,192
(245, 6,139)
Middle Atlantic
Total
229,597
5.9
(6.0, 6.3)
6,359
(5,218,7,501)
New Jersey
49,307
5.8
(5.1, 6.5)
8,740
(4,677, 12,803)
New York
115,794
6.5
(5.9,7.1)
6,065
(4,693, 7,436)
Pennsylvania
64,496
5.2
(4.7, 5.7)
5,068
(4,143, 5,993)
East North Central
Total
252,991
5.3
(5.0, 5.7)
4,899
(4,349, 5,449)
Indiana
33,873
5.0
(4.3, 5.7)
4,963
(3,959, 5,967)
Illinois
58,323
4.8
(4.0, 5.6)
4,877
(3,590, 6,163)
Michigan
70,181
6.5
(5.8,7.3)
4,865
(3,635, 6,094)
Ohio
65,599
5.7
(5.1, 6.3)
5,133
(4,079, 6,187)
Wisconsin
25,014
4.0
(3.4, 4.6)
4,347
(3,424, 5,270)
West North Central
Total
116,909
5.4
(5.0, 5.8)
5,892
(4,800, 6,984)
Iowa
21,523
5.4
(4.4, 6.3)
5,015
(3,741, 6,288)
Kansas
19,598
6.4
(5.2, 7.6)
4,905
(3,551, 6,260)
Minnesota
19,272
4.1
(3.3,4.8)
8,023
(2,882, 13,163)
Missouri
34,387
5.8
(5.0, 6.6)
6,582
(4,641, 8,523)
Nebraska
10,011
4.8
(3.6,6.1)
4,683
(2,360, 7,006)
North Dakota
5,162
5.8
(3.8,7.8)
2,801
(957, 4,645)
South Dakota
6,958
6.8
(4.7, 8.9)
6,104
(3,902, 8,306)
South Atlantic
Total
249,410
4.8
(4.5, 5.0)
5,076
(4,526, 5,627)
Delaware
5,529
6.3
4.7, 7.9)
6,843
(4,033, 9,653)
District of Columbia
2,620
4.8
(2.9, 6.6)
15,097
(2,331, 27,863)
Florida
72,173
4.3
(3.8,4.7)
4,568
(3,766, 5,389)
Georgia
35,091
5.2
(4.5, 5.9)
5,309
(3,808, 6,810)
Maryland
24,246
5.0
(4.2, 5.9)
6,423
(4,551, 8,295)
North Carolina
45,245
5.3
(4.7, 5.9)
4,187
(3,414, 4,960)
South Carolina
20,371
4.6
(3.7, 5.4)
5,261
(3,078, 7,443)
Virginia
29,932
4.3
(3.7, 5.0)
5,221
(2,623,7,818)
West Virginia
14,204
6.0
(4.8, 7.2)
4,515
(3,257, 5,773)
(continued)
C-1
-------
Table C-1. Direct Medical Costs Associated with Pneumonia—Annual Medicare
Payments in 2000 by Region and State (continued)
Direct Medical Costs
Beneficiaries with a Diagnosis
($/Person/Y ear)
Region (State)
Number
Percent
CI (±)
Mean
ci(±)
East South Central
Total
101,233
5.5
(5.0, 6.0)
5,363
(4,1538, 6,187)
Alabama
23,355
4.8
(4.1, 5.6)
5,773
(4,359, 7,187)
Kentucky
28,643
6.5
(5.5, 7.6)
5,062
(3,537, 6,586)
Mississippi
16,875
5.4
(4.4, 6.4)
7,021
(4,004, 10,038)
Tennessee
32,360
5.3
(4.5, 6.0)
4,469
(3,284, 5,653)
West South Central
Total
152,602
5.8
(5.4, 6.2)
5,985
(4,950, 7,020)
Arkansas
18,737
5.7
(4.7, 6.7)
4,665
(2,731, 6,599)
Louisiana
21,588
5.6
(4.7, 6.5)
6,409
(4,684, 8,133)
Oklahoma
22,113
5.9
(4.9, 6.9)
5,346
(3,776, 6,917)
Texas
90,165
5.9
(5.3, 6.4)
6,314
(4,702, 7,926)
Mountain
Total
59,498
4.5
(4.0, 5.0)
5,048
(4,170, 5,925)
Arizona
15,785
4.5
(3.4, 5.6)
4,370
(2,835, 5,904)
Colorado
11,948
5.1
(4.0, 6.2)
6,821
(4,921, 9,352)
Idaho
5,589
4.6
(2.8, 6.4)
5,109
(2,304, 7,914)
New Mexico
6,777
4.5
(3.3,5.8)
4,730
(2,649, 6,811)
Montana
3,873
3.4
(1.9, 5.0)
4,683
(1,533, 7,834)
Utah
8,171
4.8
(3.5, 6.1)
3,296
(1,841,4,750)
Nevada
5,530
4.3
(3.0, 5.5)
5,374
(2,173, 8,576)
Wyoming
1,815
3.4
(1.9,5.0)
7,946
(-2,312, 18,205)
Pacific
Total
131,330
5.0
(4.5, 5.4)
6,304
(5,073, 7,534)
Alaska
2,080
6.1
(4.1, 8.2)
5,984
(3,352, 8,416)
California
91,048
5.0
(4.4, 5.6)
7,496
(5,766, 9,225)
Hawaii
4,606
5.0
(3.5, 6.5)
4,389
(1,710, 7,069)
Oregon
11,671
4.8
(3.7, 5.8)
3,828
(2,434, 5,222)
Washington
21,925
5.0
(4.1, 5.8)
3,104
(2,064, 4,143)
C-2
-------
Table C-2. Direct Medical Costs Associated with Stroke —Annual Medicare Payments
in 2000 by Region and State
Direct Medical Costs
Beneficiaries with a Diagnosis
($/Person/Y ear)
Region (State)
Number
Percent
CI (±)
Mean
ci(±)
New England
Total
94,443
5.9
(5.4, 6.5)
4,242
(3,288, 5,197)
Connecticut
25,113
7.5
(6.3,8.6)
3,045
(1,833,4,256)
Maine
10,735
6.2
(4.7, 7.7)
2,770
(1,646, 3,894)
Massachusetts
40,135
7.1
(6.0,8.1)
5,334
(3,400, 7,268)
New Hampshire
9,934
7.2
(5.6, 8.9)
3,111
(1,486, 4,736)
Rhode Island
4,763
5.3
(3.3,7.2)
4,731
(1,743,7,719)
Vermont
3,762
5.3
(3.3, 7.2)
7,162
(1,017, 13,307)
Middle Atlantic
Total
307,495
6.4
(6.0, 6.7)
3,686
(3,237, 4,135)
New Jersey
71,158
8.4
(7.5, 9.2)
4,478
(3,532, 5,424)
New York
138,898
7.8
(7.1, 8.4)
3,374
(2,617, 4,131)
Pennsylvania
97,439
7.8
(7.2, 8.5)
3,552
(2,945, 4,159)
East North Central
Total
367,642
7.8
(7.4, 8.2)
3,823
(3,222, 4,424)
Indiana
52,894
7.8
(7.0, 8.6)
4,014
(3,081,4,948)
Illinois
86,491
7.2
(6.2, 8.2)
4,512
(2,593, 6,431)
Michigan
102,437
9.5
(8.7, 10.4)
2,613
(2,043, 3,184)
Ohio
91,660
8.0
(7.3, 8.7)
4,380
(3,081, 5,579)
Wisconsin
34,160
5.4
(4.7, 6.2)
3,911
(2,948, 4,873)
West North Central
Total
133,861
6.2
(5.7, 6.6)
3,653
(2,992, 4,314)
Iowa
25,497
6.4
(5.3,7.4)
2,585
(1,702, 3,467)
Kansas
21,262
6.9
(5.7, 8.2)
3,907
(2,397, 5,417)
Minnesota
17,996
3.8
(3.1,4.5)
3,461
(1,988,4,934)
Missouri
44,835
7.6
(6.7, 8.5)
3,821
(2,749, 4,893)
Nebraska
13,316
6.4
5.0, 7.9)
5,854
(1,571, 10,137)
North Dakota
5,323
6.0
(4.0, 8.0)
2,053
(659, 3,446)
South Dakota
5,453
5.3
(3.4, 7.2)
3,105
(1,218,4,992)
South Atlantic
Total
449,303
8.6
(8.3, 9.0)
3,433
(3,084, 3,783)
Delaware
8,462
9.6
(7.8, 11.5)
3,862
(2253, 5,471)
District of Columbia
2,620
4.8
(2.9, 6.6)
4,587
(1,224, 7,949)
Florida
178,328
10.5
(9.9, 11.2)
3,028
(2,617, 3,440)
Georgia
54,206
8.1
(7.2, 8.9)
3,322
(2,585, 4,060)
Maryland
91,854
19.1
(17.6, 20.6)
6,882
(5,687, 8,077)
North Carolina
61,526
7.2
(6.5, 7.9)
3,076
(2,360, 3,972)
South Carolina
33,093
7.4
(6.4, 8.4)
4,681
(2,604, 6,758)
Virginia
55,077
7.9
(7.1, 8.8)
3,391
(1,898,4,885)
West Virginia
19,069
8.0
(6.6, 9.4)
3,689
(2,316, 5,062)
(continued)
C-3
-------
Table C-2. Direct Medical Costs Associated with Stroke (Ischemic or Hemorrhagic)—
Annual Medicare Payments in 2000 by Region and State (continued)
Direct Medical Costs
Beneficiaries with a Diagnosis
($/Person/Y ear)
Region (State)
Number
Percent
CI (±)
Mean
ci(±)
East South Central
Total
149,178
8.0
(7.6, 8.6)
4,271
(3,699, 4,843)
Alabama
42,206
8.8
(7.7, 9.8)
4,096
(3,126, 5,067)
Kentucky
35,404
8.1
(7.0 , 9.2)
4,077
(2,850, 5,303)
Mississippi
23,622
7.5
(6.4, 8.7)
3,844
(2,795, 4,892)
Tennessee
47,946
7.8
(6.9, 8.7)
4,788
(3,604, 5,952)
West South Central
Total
212,888
6.4
(7.6, 8.6)
4,017
(3,441,4,592)
Arkansas
27,484
8.4
(7.1,9.6)
3,821
(2,748, 4,893)
Louisiana
32,981
8.6
(7.5, 9.7)
3,271
(2,393,4,149)
Oklahoma
27,133
7.3
(6.2, 8.4)
3,456
(1,855, 5,058)
Texas
125,290
8.2
(7.5, 8.8)
4,377
(3,521, 5,233)
Mountain
Total
65,373
5.0
(4.4, 5.5)
3,430
(2,685, 4,174)
Arizona
19,835
56.7
(4.5, 6.9)
2,962
(1,546, 4,378)
Colorado
10,657
45.7
(3.5,5.6)
3,319
(1,694, 4,943)
Idaho
5,356
4.4
(2.6 . 6.2)
6,798
(2,061, 11,535)
New Mexico
6,128
4.1
(2.9, 5.3)
3,027
(1,437, 4,617)
Montana
6,024
5.3
(3.4, 7.3)
4,295
(1,536, 7,055)
Utah
8,184
4.8
(3.5, 6.2)
2,652
(813,4,491)
Nevada
7,174
5.5
(4.1,7.0)
2,995
(1,426, 4,563)
Wyoming
2,017
3.8
(2.2, 5.4)
3,011
(314, 5,708)
Pacific
Total
173,089
6.6
(6.0,7.1)
3,557
(2,764, 4,350)
Alaska
1,950
5.7
(3.7, 7.7)
5,253
(2,260, 8,426)
California
127,335
7.0
(6.2, 7.7)
3,600
(2,565, 4,635)
Hawaii
7,004
7.6
(5.8, 9.4)
5,164
(2,196, 8,133)
Oregon
10,118
4.1
(3.1,5.1)
3,569
(1,418, 5,720)
Washington
26,681
6.0
(5.1,7.0)
2,798
(1,878, 3,718)
C-4
-------
Table C-3. Direct Medical Costs Associated with Chronic Lung Disease—Annual
Medicare Payments in 2000 by Region and State
Direct Medical Costs
Beneficiaries with a Diagnosis
($/Person/Y ear)
Region (State)
Number
Percent
CI (±)
Mean
ci(±)
New England
Total
173,635
12.6
(11.9, 13.4)
6,648
(5,703, 7,593)
Connecticut
40,571
12.1
(10.7, 13.4)
6,682
(5,200, 8,164)
Maine
20,917
12.1
(10.0, 14.1)
6,367
(4,357, 8,378)
Massachusetts
75,950
13.4
(11.9, 14.8)
7,380
(5,521, 9,238)
New Hampshire
16,796
12.3
(10.2, 14.3)
6,004
(4,317, 7,692)
Rhode Island
10,761
11.9
(9.1, 14.7)
5,851
(3,363, 8,340)
Vermont
8,640
12.1
(9.3, 14.9)
2,981
(1,545, 4,418)
Middle Atlantic
Total
467,332
12.0
(11.6, 12.5)
6,228
(5,588, 6,867)
New Jersey
102,374
12.0
(11.0, 13.0)
7,300
(5,343, 9,258)
New York
210,086
11.8
(11.0, 12.5)
5,447
(4,581, 6,314)
Pennsylvania
154,872
12.5
(11.7, 13.3)
6,577
(5,755, 7,400)
East North Central
Total
581,380
12.3
(11.8, 12.8)
5,761
(5,252, 6,270)
Indiana
86,337
12.7
(11.7, 13.7)
5,375
(4,147, 6,604)
Illinois
133,561
11.1
(1.0, 12.2)
6,474
(4,922, 8,026)
Michigan
152,208
14.2
(13.1, 15.2)
5,877
(5,000, 6,754)
Ohio
147,656
12.9
(12.0, 13.7)
5,583
(4,823, 6,342)
Wisconsin
61,618
9.8
(8.9, 10.8)
4,897
(4,034, 5,561)
West North Central
Total
240,049
11.0
(10.5, 11.6)
6,485
(5,286, 7,684)
Iowa
47,337
11.8
(10.4, 13.2)
4,784
(3,670, 5,899)
Kansas
36,878
12.0
(10.4, 13.6)
5,971
(4,244, 7,697)
Minnesota
42,074
8.9
(7.8, 1.0)
5,779
(3,725, 7,834)
Missouri
74,433
12.6
(11.5, 13.7)
8,086
(4,717, 11,455)
Nebraska
18,689
9.0
(7.3, 11.0)
8,485
(4,672, 12,298)
North Dakota
10,485
11.8
(9.1, 14.5)
5,355
(3,207, 7,503)
South Dakota
10,154
9.9
(7.4, 12.4)
4,960
(3,018, 6,903)
South Atlantic
Total
680,253
13.0
(12.6, 13.4)
5,579
(5,205, 5,593)
Delaware
9,172
10.4
(8.5, 12.4)
9,498
(5,608, 13,387)
District of Columbia
4,191
7.6
(5.3, 9.9)
9,841
(3,268, 16,414)
Florida
250,009
14.7
(14.0, 15.5)
5,382
(4,731, 6,033)
Georgia
84,409
12.6
(11.5, 13.6)
5,356
(4,514, 6,198)
Maryland
58,363
12.1
(10.9, 13.4)
7,561
(5,878, 9,244)
North Carolina
106,788
12.5
(11.5, 13.5)
5,452
(4,666, 6,238)
South Carolina
51,419
11.5
(10.3, 12.8)
4,649
(3,662, 5,637)
Virginia
82,241
11.8
(10.8, 12.9)
5,151
(3,926, 6,376)
West Virginia
33,660
14.1
(12.4, 15.9)
5,431
(4,140, 6,721)
(continued)
C-5
-------
Table C-3. Direct Medical Costs Associated with Chronic Lung Disease—Annual
Medicare Payments in 2000 by Region and State (continued)
Direct Medical Costs
Beneficiaries with a Diagnosis
($/Person/Y ear)
Region (State)
Number
Percent
CI (±)
Mean
ci(±)
East South Central
Total
252,568
13.7
(13.0, 14.3)
4,974
(4,534, 5,415)
Alabama
68,419
1.4
(13.0, 15.5)
5,244
(4,370, 6118)
Kentucky
65,299
14.9
(13.4, 16.4)
4,863
(4,042, 5,685)
Mississippi
38,944
12.4
(11.0, 13.9)
4,891
(3,699, 6,083)
Tennessee
79,906
13.0
(11.8, 14.1)
4,874
(4,098, 5,651)
West South Central
Total
336,899
12.8
(12.3, 13.4)
6,317
(5,675, 6,959)
Arkansas
42,442
12.9
(11.4, 14.4)
5,712
(4,427, 7,176)
Louisiana
47,496
12.4
(11.1, 13.7)
6,710
(5,574, 7,845)
Oklahoma
50,308
13.5
(12.0, 15.0)
5,841
(4,345, 7,337)
Texas
196,653
12.8
(12.0, 13.6)
6,475
(5,531,7,419)
Mountain
Total
167,846
12.7
(11.9, 13.5)
4,817
(4,238, 5,396)
Arizona
45,171
12.9
(11.2, 14.6)
4,016
(2,839, 5,192)
Colorado
36,154
15.5
(13.7, 17.3)
6,174
(4,850, 7,497)
Idaho
14,205
11.7
(8.9, 14.4)
4,780
(2,566, 6,993)
New Mexico
18,911
12.7
(10.6, 14.7)
3,878
(2,713, 5,043)
Montana
13,984
12.4
(9.6, 15.2)
4,662
(2,913, 6,410)
Utah
13,909
8.2
(6.5, 9.9)
3,423
(2,145, 4,701)
Nevada
18,857
15.0
(12.3, 16.7)
5,802
(3,578, 8,027)
Wyoming
6,655
12.5
(9.7, 15.3)
6,094
(3,473, 8,714)
Pacific
Total
308,164
11.7
(11.0, 12.3)
5,911
(4,942, 6,881)
Alaska
4,550
13.4
(10.5, 16.3)
6,569
(3,716, 9,421)
California
213,616
11.7
(10.8, 12.6)
6,447
(5,093, 7,802)
Hawaii
8,847
9.6
(7.5, 11.6)
6,550
(3,141, 9,960)
Oregon
26,258
10.7
(9.2, 12.3)
4,660
(3,497, 5,823)
Washington
54,893
12.4
(11.1, 13.8)
4,266
(3,247, 5,285)
C-6
-------
Table C-4. Direct Medical Costs Associated with Ischemic Heart Disease—Annual
Medicare Payments in 2000 by Region and State
Direct Medical Costs
Beneficiaries with a Diagnosis
($/Person/Y ear)
Region (State)
Number
Percent
CI (±)
Mean
ci(±)
New England
Total
282,552
20.5
(19.6,21.5)
6,303
(5,632, 6,973)
Connecticut
73,319
21.8
(20.0, 23.6)
5,973
(4,954, 6,992)
Maine
36,360
21.0
(18.5, 23.6
5,116
(3,880, 6,353)
Massachusetts
119,789
21.1
(19.4, 22.7)
7,413
(6,100, 8,726)
New Hampshire
24,898
18.2
(15.7, 20.6)
4,628
(3,305, 5,591)
Rhode Island
16,759
18.5
(15.1,21.9)
5,247
(2,989, 7,505)
Vermont
11,427
16.0
(12.8, 19.2)
5,750
(3,077, 8,422)
Middle Atlantic
Total
885,001
22.8
(22.2, 23.4)
5,886
(5,400, 6,372)
New Jersey
208,261
24.5
(23.1,25.8)
6,426
(5,182, 7,670)
New York
404,025
22.6
(21.6, 23.6)
5,691
(4,935, 6,446)
Pennsylvania
272,715
22.0
(21.0, 23.0)
5,764
(5,184, 6,343)
East North Central
Total
977,693
20.6
(20.1,21.2)
5,892
(5,508, 6,275)
Indiana
133,417
2.0
(18.5, 20.9)
5,559
(4,904, 6,213)
Illinois
224,414
18.6
(17.1, 20.1)
6,544
(5,285, 7,802)
Michigan
285,749
26.6
(25.3, 27.9)
5,364
(4,766, 5,962)
Ohio
233,021
20.3
(19.3,21.3)
6,075
(5,511, 6,639)
Wisconsin
101,092
16.1
(14.9, 17.3)
5,953
(5,169, 6,737)
West North Central
Total
381,285
17.5
(16.9, 18.2)
5,760
(5,304, 6,217)
Iowa
65,222
16.3
(14.7, 17.8)
4,874
(3,996, 5,753)
Kansas
61,420
20.0
(18.1,22.0)
4,746
(3,871, 5,622)
Minnesota
64,542
13.6
(12.3, 14.9)
6,070
(5,096, 7,043)
Missouri
122,870
20.8
(19.4, 22.1)
6,283
(5,447, 7,119)
Nebraska
33,317
16.1
(1.4,1.8)
6,274
(4,573, 7,974)
North Dakota
16,614
18.7
(15.4, 22.0)
6,651
(4,212, 9,091)
South Dakota
17,300
16.8
(13.7, 20.0)
5,994
(1,906, 10,082)
South Atlantic
Total
1,066,047
20.4
(19.9, 20.9)
5,370
(5,093, 6,648)
Delaware
21,009
23.9
(21.1,26.7)
5,469
(4,090, 6,848)
District of Columbia
6,811
12.4
(9.6, 15.2)
10,614
(3,861, 17,367)
Florida
432,873
25.5
(24.6, 26.4)
4,750
(4,370, 5,131)
Georgia
115,271
17.2
(16.0, 18.4)
5,960
(5,065, 6,854)
Maryland
91,854
19.1
(17.6, 20.6)
6,882
(5,687, 8,077)
North Carolina
146,238
17.1
(16.0, 18.2)
5,140
(4,500, 5,780)
South Carolina
81,512
18.3
(16.8, 19.8)
5,936
(4,717, 7,154)
Virginia
116,923
16.8
(15.6, 18.1)
5,257
(4,286, 6,227)
West Virginia
53,556
22.5
(20.4, 24.6)
5,835
(4,805, 6,864)
(continued)
C-7
-------
Table C-4. Direct Medical Costs Associated with Ischemic Heart Disease—Annual
Medicare Payments in 2000 by Region and State (continued)
Direct Medical Costs
Beneficiaries with a Diagnosis
($/Person/Y ear)
Region (State)
Number
Percent
CI (±)
Mean
ci(±)
East South Central
Total
346,972
18.8
(18.2, 19.5)
5,930
(5,342, 6,518)
Alabama
87,164
18.1
(16.7, 19.5)
5,657
(4,853, 6,462)
Kentucky
88,595
20.2
(18.6,21.9)
5,993
(5,051, 6,935)
Mississippi
55,910
17.9
(16.2, 19.6)
6,000
(4,914, 7,087)
Tennessee
115,302
18.7
(17.4, 20.0)
6,054
(4,648, 7,460)
West South Central
Total
500,632
19.1
(18.4, 19.8)
5,879
(5,444,6,314)
Arkansas
65,705
20.0
(18.2,21.8)
5,182
(4,283, 6,081)
Louisiana
75,143
19.6
(18.1,21.2)
5,951
(5,069, 6,833)
Oklahoma
73,744
19.7
(18.1,21.4)
6,757
5,364, 8,151)
Texas
286,040
18.6
(17.7, 19.5)
5,794
(5,198, 6,391)
Mountain
Total
210,157
15.9
(15.1, 16.8)
5,014
(4,459, 5,586)
Arizona
66,016
18.9
(16.8, 20.9)
4,074
(3,118, 5,030)
Colorado
34,250
14.7
(12.9, 16.5)
5,929
(4,566, 7,292)
Idaho
19,561
16.1
(12.9, 19.2)
6,791
(4,049, 9,533)
New Mexico
21,949
14.7
(12.5, 16.9)
5,062
(3,458, 6,666)
Montana
16,351
14.5
(11.5, 17.5)
4,916
(3,122, 6,710)
Utah
23,429
13.8
(11.7, 15.9)
4,119
(2,856, 5,383)
Nevada
22,349
17.2
(14.8, 19.6)
5,408
(3,737, 7,079)
Wyoming
6,252
11.7
(9.7, 14.5)
6,392
(3,588, 9,196)
Pacific
Total
429,332
16.3
(15.5, 17.0)
5,955
(5,263, 6,640)
Alaska
4,355
12.8
(9.9, 15.7)
10,150
6,256, 14,043)
California
317,040
17.4
(16.3, 18.4)
6,284
(5,384, 7,185)
Hawaii
14,599
15.8
(13.3, 18.3)
5,069
(3,139, 6,999)
Oregon
30,314
12.4
(10.8,14.0)
4,460
(3,262, 5,659)
Washington
63,023
14.3
(12.8, 15.7)
4,911
(3,992, 5,829)
C-8
-------
Table C-5. Direct Medical Costs Associated with Lung Cancer—Annual Medicare
Payments in 2000 by Region and State
Direct Medical Costs
Beneficiaries with a Diagnosis
($/Person/Y ear)
Region (State)
Number
Percent
CI (±)
Mean
ci(±)
New England
Total
16,862
1.2
(1.0, 1.4)
11,928
(6,172, 17,685)
Connecticut
3,116
0.9
(0.5, 1.3)
14,866
(2,288, 27,445)
Maine
2,424
1.4
(0.7,2.1)
7,629
(2,366, 12,892)
Massachusetts
8,212
1.4
(1.0, 1.9)
14,367
(3,563,25,171)
New Hampshire
1,142
0.8
(0.3, 1.4)
7,589
(-822, 16,001)
Rhode Island
1,411
1.6
(0.5, 2.6)
6,836
(-2,159, 15,830)
Vermont
557
0.8
(0.0, 1.5)
51
(-13, 115)
Middle Atlantic
Total
53,771
1.4
(1.2, 1.6)
11,950
(9,833, 14,067)
New Jersey
10,976
1.3
(0.9, 1.6)
12,854
(8,754, 16,954)
New York
28,147
1.6
(1.3, 1.9)
11,920
(8,860, 15,281)
Pennsylvania
14,648
1.2
(0.9, 1.4)
11,330
(8,101, 14,560)
East North Central
Total
55,466
1.2
(1.0, 1.3)
8,431
(6,968, 9,893)
Indiana
7,460
1.1
(0.8, 1.4)
8,552
(5,639, 11,466)
Illinois
10,714
0.9
(0.6, 1.2)
7,188
(3,882, 10,494)
Michigan
18,217
1.7
(1.3,2.1)
9,181
(5,949, 12,413)
Ohio
13,489
1.2
(0.9, 1.4)
7,852
(5,295, 10,410)
Wisconsin
5,586
0.9
(0.6, 1.2)
9,603
(6,000, 13,205)
West North Central
Total
22,121
1.0
(0.8, 1.2)
9,076
(6,975, 11,177)
Iowa
3,390
0.8
(0.5, 1.2)
9,481
(5,420, 13,543)
Kansas
4,372
1.4
(0.8, 2.0)
10,908
(4,969, 16,847)]
Minnesota
5,483
11.6
(0.8, 1.6)
8,587
(5,146, 12,028)
Missouri
5,575
0.9
(0.6, 1.3)
8,561
(3,763, 13,359)
Nebraska
1,742
0.8
(0.3, 1.4)
9,284
M12, 18,980)
North Dakota
807
0.9
(0.1, 1.7)
10,094
(-15,752, 35,939)
South Dakota
752
0.7
(0.0, 1.4)
2,402
M,962, 9,765)
South Atlantic
Total
61,018
1.2
(1.0, 1.3)
11,552
(9,432, 13,671)
Delaware
1,319
1.5
(0.7,2.3)
9,030
(3,166, 14,894)
District of Columbia
524
1.0
(0.1, 1.8)
8,287
(-5,511,22,085)
Florida
20,028
1.2
(1.0, 1.4)
10,145
(7,637, 12,653)
Georgia
5,321
0.8
(0.5, 1.1)
7,350
(4,223, 10,477)
Maryland
6,006
1.2
(0.8, 1.6)
14,596
(8,176,21,016)
North Carolina
11,084
1.3
(1.0, 1.6)
13,871
(9,656, 18,086)
South Carolina
6,199
13.9
(0.9, 1.8)
11,061
(6,458, 15,664)
Virginia
7,695
1.1
(0.8, 1.4)
10,012
(201, 19,823)
West Virginia
2,841
1.2
(0.6, 1.7)
20,863
(-3,824, 45,550)
(continued)
C-9
-------
Table C-5. Direct Medical Costs Associated with Lung Cancer—Annual Medicare
Payments in 2000 by Region and State (continued)
Direct Medical Costs
Beneficiaries with a Diagnosis
($/Person/Y ear)
Region (State)
Number
Percent
CI (±)
Mean
ci(±)
East South Central
Total
23,270
1.3
(1.0, 1.5)
9,662
(7,446, 11,879)
Alabama
6,982
1.4
(1.0, 1.9)
9,649
(5,304, 13,993)
Kentucky
4,260
1.0
(0.6, 1.4)
8,928
(3,573, 14,284)
Mississippi
5,013
1.6
(1.0,2.2)
11,549
(5,677, 17,421)
Tennessee
7,015
1.1
(0.8, 1.5)
8,773
(5,425, 12,121)
West South Central
Total
29,891
1.1
(1.0, 1.3)
14,719
(11,150, 18,289)
Arkansas
2,838
0.9
(0.5, 1.3)
13,599
(6,611,20,587)
Louisiana
4,089
1.1
(0.7, 1.5)
13,295
(7,863, 18,727)
Oklahoma
4,395
1.2
(0.7, 1.6)
14,258
(7,372,21,145)
Texas
18,569
1.2
(0.9, 1.5)
15,313
(9,938, 20,688)
Mountain
Total
11,858
0.9
(0.7, 1.1)
8,655
(5,685, 11,625)
Arizona
3,550
1.0
(0.5, 1.6)
3,860
(777, 6,942)
Colorado
1,142
0.5
(0.1,0.9)
6,321
(-3,060, 15,701)
Idaho
1,863
1.5
(0.5, 2.6)
14,291
(5,763,22,818)
New Mexico
992
0.7
(0.2, 1.2)
13,289
(-54, 26,633)
Montana
645
0.6
(0.0, 1.2)
14,372
(-17,078, 45,821)
Utah
951
0.6
(0.1, 1.0)
3,225
(-163, 6,613)
Nevada
1,808
1.4
(0.6, 2.2)
11,900
(-2,033, 25,833)
Wyoming
908
1.7
(0.6, 2.8)
8,873
(2,650, 15,097)
Pacific
Total
31,000
1.2
(0.9, 1.4)
10,985
(7,363, 14,607)
Alaska
390
1.1
(0.2, 0.2)
16,169
(-494, 32,831)
California
21,098
11.5
(0.9, 1.5)
11,712
6,504, 16,920)
Hawaii
1,871
2.0
(1.0, 3.1)
11,613
(5,164, 18,062)
Oregon
2,964
12.1
(0.7, 1.8)
10,368
(5,778, 14,957)
Washington
4,677
1.1
(0.6, 1.5)
7,415
(3,610, 11,221)
C-10
-------
Table C-6. Direct Medical Costs Associated with Gastrointestinal Illness—Annual
Medicare Payments in 2000 by Region and State
Direct Medical Costs
Beneficiaries with a Diagnosis
($/Person/Y ear)
Region (State)
Number
Percent
CI (±)
Mean
ci(±)
New England
Total
37,400
2.7
(2.3
3.1)
1,341
(735, 1,947)
Connecticut
7,866
2.3
(1.7
3.0)
872
(385, 1,359)
Maine
4,563
2.6
(1.6
3.6)
1,300
(373, 2,227)
Massachusetts
17,945
3.2
(2.4
3.9)
1,776
(565,2,987)
New Hampshire
3,125
2.3
(1.3
3.3)
586
(12, 1,259)
Rhode Island
2,646
2.9
(1.5
4.4)
845
(-11, 1,700)
Vermont
1,254
1.8
(0.6
2.9)
1,134
(1,132, 3,401)
Middle Atlantic
Total
99,429
2.6
(2.5
2.8)
1,144
(866, 1,422)
New Jersey
22,230
2.6
(2.1
3.1)
904
(538, 1,270)
New York
50,104
2.8
(2.4
3.2)
1,085
(681, 1,489)
Pennsylvania
27,095
2.2
(1.8
2.5)
1,449
(821, 2,077)
East North Central
Total
104,447
2.2
(2.0
2.4)
1,193
(912, 1,474)
Indiana
11,516
1.7
(1.3
2.1)
1,376
(715, 2,038)
Illinois
34,024
2.8
(2.2
3.5)
1,396
(713, 2,079)
Michigan
25,309
2.4
(1.9
2.8)
663
(368, 957)
Ohio
22,868
2.0
(1.6
2.3)
1,511
(893,2,129)
Wisconsin
10,730
1.7
(1.3
2.1)
924
(588, 1,260)
West North Central
Total
47,352
2.2
(2.0
2.4)
1,042
(781, 1,303)
Iowa
8,672
2.2
(1.5
2.8)
834
(428, 1,238)
Kansas
9,397
3.1
(2.2
3.9)
862
(394, 1,330)
Minnesota
6,038
1.3
(0.8
1.7)
1,617
(525, 2,710)
Missouri
14,474
2.4
(1.9
3.0)
1,279
(681, 1,877)
Nebraska
5,332
2.6
(1.6
3.5)
5,332
(216, 1,025)
North Dakota
1,936
2.2
(1.0
3.4)
1,082
(277, 1,888)
South Dakota
1,504
1.5
(0.5
2.5)
210
(23, 396)
South Atlantic
Total
125,383
2.4
(2.2
2.6)
1,102
(925, 1,278)
Delaware
2,529
2.9
(1.8
3.9)
776
(305, 1,248)
District of Columbia
524
1.0
(0.1
1.8)
91
M, 181)
Florida
40,424
2.4
(2.0
2.7)
896
(604, 1,187)
Georgia
18,115
2.7
(2.2
3.2)
1,502
(1,034, 1,971)
Maryland
12,907
2.7
(2.1
3.3)
1,092
(595, 1,590)
North Carolina
19,257
2.3
(1.8
2.7)
1,187
(754, 1,620)
South Carolina
9,420
2.1
(1.5
2.7)
1,106
(304, 1,908)
Virginia
14,948
2.2
(1.7
2.6)
927
(407, 1,447)
West Virginia
7,261
3.0
(2.2
3.9)
1,582
(517, 2,646)
(continued)
C-11
-------
Table C-6. Direct Medical Costs Associated with Gastrointestinal Illness—Annual
Medicare Payments in 2000 by Region and State (continued)
Direct Medical Costs
Beneficiaries with a Diagnosis
($/Person/Y ear)
Region (State)
Number
Percent
CI (±)
Mean
ci(±)
East South Central
Total
50,148
2.7
(2.4, 3.0)
1,267
(756, 1,778)
Alabama
10,006
2.1
(1.6,2.6)
538
(263,813)
Kentucky
12,340
2.8
(2.1,3.5)
2,126
(175, 4,078)
Mississippi
8,936
2.9
(2.1, 3.6)
1,081
(655, 1,507)
Tennessee
18,867
3.1
(2.5, 3.6)
1,180
(758, 1,603)
West South Central
Total
73,659
2.8
(2.5, 3.1)
1,384
(1,099, 1,669)
Arkansas
12,024
3.7
(2.8,4.5)
963
(456, 1,470)
Louisiana
7,207
1.9
(1.4,2.4)
1,531
(583, 2,480)
Oklahoma
10,738
2.9
(2.2, 3.6)
1,464
(536, 2,392)
Texas
43,689
2.8
(2.5, 3.2)
1,456
(1,082, 1,829)
Mountain
Total
29,821
2.3
(1.9,2.6)
841
(536, 1,146)
Arizona
8,874
2.5
(1.7,3.3)
914
(249, 1,580)
Colorado
5,029
2.2
(1.4,2.9)
1,316
(369, 2,263)
Idaho
2,096
1.7
(0.6,2.8)
657
(-253, 1,566)
New Mexico
2,594
1.7
(0.9,2.5)
1,966
(118, 3815)
Montana
3,657
3.2
(1.7,4.8)
544
(86, 1,002)
Utah
2,440
1.4
(0.7, 2.2)
210
(43, 378)
Nevada
3,586
2.8
(1.7,3.8)
177
(118,236)
Wyoming
1,613
3.0
(1.6,4.5)
487
(-128, 1,102)
Pacific
Total
54,108
20.5
(1.8,2.3)
1,154
(793, 1,515)
Alaska
520
1.5
(0.5, 2.6)
1,336
(-1,132, 3,803)
California
40,392
2.2
(1.8,2.6)
1,265
(801, 1,728)
Hawaii
2,257
2.4
(1.4, 3.5)
722
(31, 1,413)
Oregon
3,595
1.5
(0.8,2.1)
626
(46, 1,205)
Washington
7,343
1.7
(1.2,2.2)
924
(226, 1,622)
C-12
------- |