Most COI studies distinguish between direct costs and indirect costs. Direct costs measure the resources used to treat an illness while indirect costs measure lost productivity i.e., the effects of the illness on the ability of either patients or their caregivers to work (e.g., lost income) or engage in other activities (e.g., cleaning the house).
The earliest studies used the human capital method to measure these indirect losses on the country’s gross domestic product (GDP). That is, costs often were narrowly defined as they are in the national income and product accounts: thus, the impact of disease on wages and salaries would be counted as costs, but the effects of ill health on retirees or other nonworkers would be excluded since the value of their “household production” (e.g., housekeeping) is not included in GDP. The results of such analyses could be interpreted very roughly as how much greater GDP would be if an illness were eradicated (see discussion of prevalence-based costs below for a caveat).
More recent studies have measured the burden of illness using the willingness-to-pay method which takes into account intangible costs of illness such as pain and suffering, not just productivity losses.
A good primer on COI has recently been released by RTI full text pdf.
COI studies vary on two critical dimensions: how losses to health are valued and how costs are allocated across periods of time.
The two basic approaches to valuing health are the human capital approach and willingness-to-pay approach. These are described in detail below. In terms of time-frame, there are 3 alternative methods to estimating the cost of illness: prevalence-based method; b) incidence-based method; and c) present cost method.
Prevalence-based Method. This approach is still the most common, allocating direct costs and morbidity losses in the year in which they occur (the base year) and adding to this the net present value of lost earnings for all individuals who died in that same base year. Thus, the prevalence-based method does not account for people who died either in the past or the future, but the mortality losses largely represent lost earnings from future years rather than the actual earnings that were lost in the base year from people who died that year or in the past. Strictly speaking, therefore, such estimates should not be interpreted as the reduction in base year GDP that is associated with an illness or condition. Moreover, even if all instances of new distance could be instantly prevented, this would not reduce the COI for that disease to zero since people who already had the disease would continue to experience direct and indirect losses.
Incidence-based Method. This method is future-oriented. It focuses on those who are first diagnosed with a disease (e.g., cancer) or condition (e.g., brain injury) and then calculates the lifetime stream of direct and indirect costs for those individuals, discounting all such costs back to their present value in the base year. While more complicated to calculate, incidence-based estimates are most useful in thinking about the pay-off from prevention since if a new vaccine could completely eradicate a disease, for example, this would completely eliminate any incidence-based costs associated with it.
Present Cost Method. This is used only rarely, but it is the only COI measure that with reasonable accuracy measures the loss in base year GDP that can be attributed to a disease or condition. As with the prevalence-based method, it includes all direct and morbidity losses in the base year. But then it adds only the mortality losses that can be attributed to individuals who would otherwise have been alive that year. Thus, like the prevalence-based method, it accounts for lost earnings from decedents for the base year (but excludes any future earnings from this group), but unlike any other method, it looks backwards to consider the productivity losses in the base year of all the people who have died prematurely and who otherwise would have been alive in the base year but for that disease or condition. Of course, if health losses are measured in willingness-to-pay terms and/or a value attached to unpaid household labor (which is not counted in GDP), then even this measure should not properly be compared to GDP.
Direct costs are the costs associated with prevention or treatment of an illness. These costs generally include: a) formal (i.e., paid) care such as hospitalization, physician visits and medications; b) informal care costs such as the time losses of family or friends who care for a patient (which are considered to be direct costs even though these represent productivity losses); c) non-health costs incurred by the family such as for transportation or lodging related to assisting an ill family member (i.e., any costs that would not have been incurred but for the illness); and d) patient time costs related to treatment, including time spent traveling, waiting time and actual time spent receiving treatment.
Core direct costs are those connected with the use of medical care in the prevention, diagnosis, and treatment of disease and in the continuing care, rehabilitation, or terminal care of patients. Examples include expenditures for hospitalization, outpatient clinical care, nursing home care, and home health care; services of primary physicians and specialists, dentists, and other health practitioners; drugs and drug sundries; and rehabilitation counseling and other rehabilitation costs, such as for prostheses, appliances, eyeglasses, hearing aids, and other devices to overcome impairments resulting from illness or disease. Collectively, these expenditures represent the personal health care component of the United States National Health Accounts.
Other related direct costs are borne by patients or other payers but are not included in the National Health Expenditures Accounts. Examples of such costs are expenditures for transportation to hospitals, to physicians’ offices or to other health providers; certain household expenditures (e.g., help for cleaning, laundering, and cooking); special diets and clothing; relocation and moving expenses; and certain property losses, including the destruction of property resulting from alcohol abuse and criminal activity due to drug addiction.
Formal (paid) care costs may include the following. The following are abbreviated versions of the definitions of what is included in each cost “bucket” according to the Centers for Medicare and Medicaid Services **National Health Expenditure Accounts**:
Hospital care. Includes room and board, operating fees, costs for resident physician services, inpatient pharmacy costs, hospital-based nursing home costs and costs for any services bill by a hospital.
Physician services. Includes offices of MD’s, Doctors of Osteopathy and independent medical labs.
Dental services. Includes offices of doctors of dental surgery (DDS), or doctors of dental medicine (DM).
Other professional services. Includes licensed health practitioners other than physicians or dentists such as private-duty nurses, chiropractors, podiatrists, optometrists and expenditures for services in outpatient clinics (e.g., physical therapy).
Home health care. Includes services delivered in the home by non-facility-based home health agencies (Medicare approved). Prescription drugs.
Other medical nondurables. Includes nondurable products (e.g., bandages, heating pads, nonprescription drugs, analgesics, rubber medical sundries).
Vision products and other medical durables. Includes items such as eyeglasses, hearing aids, surgical appliances/supplies, wheelchairs and other equipment.
Nursing home care. Includes inpatient nursing home care provided through skilled nursing facilities (SNFs), intermediate care facilities (ICFs), and intermediate care facilities for the mentally retarded (ICF-MR), including VA-operated nursing facilities. This category does not include domiciliary care since it is not medical in nature.
Other personal health care. This includes industrial in-plant health facilities provided by employers and all other government expenditures for health care not broken down into categories.
Transportation. While CMS does not break this out as a separate category, note that this would include ambulance and helicopter services (if these were not otherwise accounted for under hospital services), but CMS does not track personal transportation expenses for medical purposes in the NHEA.
Informal Care. These include time losses of family or friends who care for a patient. These may include “medical” services (e.g., changing dressings), but also any sort of personal care (e.g., bathing) or even household services (e.g., cooking) necessitated by illness-related impairments in a patient’s ability to independently perform activities of daily living. The same methods for converting indirect time losses into dollar equivalents (see below) can be used to monetize informal care and patient time costs.
Orosz: OECD System of Health Accounts and the US National Health Account (PDF, 250KB)
DeParle: How Policy Makers Use the National Health Accounts (PDF, 32KB)
Display and Categorization of Source of Funds Estimates in the National Health Expenditure Accounts: Incorporating the MMA (PDF, 105KB)
Donahoe: Capital in the National Health Accounts (PDF, 415KB)
Historical Office of National Cost Estimates. “Revisions to the National Health Accounts and Methodology.” Health Care Financing Review 11, No. 4 (Summer 1990): 42-54.
NCHS, Appendix V. “Methodology Employed in Calculating Direct and Indirect Costs of Illness,” in Costs of Illness United States, 1980. Hyattsville, MD: National Center for Health Statistics, April 1986: 78-86. This contains a very detailed, step-by-step description of how direct costs are calculated.
How are indirect costs measured?
Indirect costs include productivity costs related to lost or impaired ability to work (or engage in leisure activities) due to morbidity and lost economic productivity related to early death. For cost-effectiveness analysis, it is recommended that these indirect losses be subtracted from all other health effects accounted for in the denominator, but for COI calculations, time losses can be monetized using market wages to represent the opportunity cost of lost time. For those who are not working or retired, the value of time can be monetized using the hourly wage of individuals with similar characteristics (this provides a lower bound) or by using a “replacement cost” approach which uses the market price of an equivalent service (e.g., maid), which provides an upper bound on the value of the time lost (see examples):
- Median Earnings of Full-time, Year-Round Workers by Selected Characteristics: 1998
- Willingness to Pay Estimates of Value of Human Life
- Present Value of Expected Lifetime Productivity by Age, Gender, and Discount Rate: 1992
- Max W, Rice DP, Sung H-Y, Michel M. Valuing Human Life: Estimating the Present Value of Lifetime Earnings, 2000. Available at http://repositories.cdlib.org/ctcre/esarm/PVLE2000/.
Source: Marthe R. Gold, Joanna E. Siegel, Louise B. Russell, Milton C. Weinstein. Cost-effectiveness in health and medicine. Oxford University Press, 1996.
There are 3 alternative methods used in estimating the burden of illness using WTP: a) labor market method (these are also known as wage-risk estimates); b) consumer market method (these are also known as price-risk estimates); and c) survey method. There are three different types of survey methods in use: a) contingent valuation; b) pairwise comparisons; and c) standard gamble. See Mean Annual Value of Housekeeping Services: 1985 .