Tobacco Documents Online

( PRELL~INARY DRAFT SMOKING-REIATED DEATHS AND FINANCIAL COSTS Karl Kronebusch Office of Technology Assessment U.S. Congress May 10, 1985 PEELIMINAEY DRAFT "~A~AA'~ ........... 4" " '~A'~'A~A ....... ~A~'A ............. * NOTE: This is a PRELIMINARY DRAFT. * * lU has not Been approved for release By OTA. * * IU is Being circulated for review purposes * * only, and should nou be quoted, disuribuued, * * or reproduced. The mauerial is Being re- * * viewed and should noU Be considered final. * ¢D

#'| DRAFT. (519185) DO NOT QUOTE, CITE, OK EEPRODUCE Iq~FRODUCTI ON Smoking is ~he larges~ single preventable cause of death in ~he United States. The use of cigarettes, cigars, and pipes has been associated with a diverse group of diseases, including cancers of the lung, lip and mouth, esophagus, pancreas and bladder; heart disease; and chronic lung disease. The Subcommittee on Health of ~he House Ways and Means Committee has asked ~hat OTA prepare an analysis of ~he financial costs of smoking uo societT, and ~he costs borne by the Medicare and Medicaid programs. This analysis includes a brief review of previous estimates of the costs of smoking-related disease. Pas~ cost estimates have generally used ~he fraction of mortality related to smoking (attributable risk) and ~hen multiplied ~hau fraction times ~he costs of a particular illness category. The estimates in ~his paper will follow ~he same general approach. The first step is to estimate ~he number of deaths related to smoking in each of the three major disease categories that has been associated wi~h smoking--cancers, heart disease, and chronic respiratory disease. Most previous estimates have not calculated ~he number of deaths by age group. The estimates in ~his paper will do so, and will ~hus be able to present the age distribution of smoking-related deaths as well as to calculate ~he number of life-years lost due to smoking-rela~ed disease. Life-years saved is a measure that is increasingly being used for evaluating ~he effects of health interventions. After all, death is inevitable for each of us, but l

q DR&.~rT (5/9/85) ( DO NOT QUOTE, CITE, OR REPEODUCE the reduction of ~he incidence of premauure death is an achievable goal. Life-years lost is one measure of =he extent of premature mortallry. The second step is to apply =he estimates of the fraction of mortality for each major disease category to =he heal=h care costs and lost produculvicy costs for each category. Thus, in the first step an estimate will be prepared of the fraction of cancers, cardiovascular disease, and respiratory system disease =hat are related to smoking. In =he second step, those fractions are applied to =he total heal=h care costs and los~ productivity costs for each class of disease. The third step of =his analysis will be to apply information on the share of heal=h care costs for the elderly paid for by government programs to =he esulmated costs of smoking-related disease among the elderly. Finally, there will be a discussion of how health care costs and the costs of other social programs might change as a result of reduced smoking. PReViOUS ESTIMATES OF THE EXTENT OF SMOKING-RELATED DISEASE Table I presents the range of a~ributablerisk estimates for smoking- related disease. "Attributable risk" is a concept from epidemiology. Put most simply, attributable risk is =he fraction or percen=age of disease that is associated with a specified risk factor. In this case, the risk factor is smoking of tobacco products, most generally cigarettes. O tO

e~ (519185) DO NOT QUOTE, CITE, 0RREPEODUCE The fraction of major disease categories a~ributed =o smoking ranges from 20 ¢o A0 percenu of cancers, ii =o 30 percenu of cardiovascular disease, and 20 ~o A0 percent of all respiratory system disease (including 80-90 percen~ of chronic obsuruc~ive lung disease)I. AU firs= blush, ~hese ranges appear to be fairly narrow--a factor of 2 or 3 separates ~he lower end of each range from the upper. Bun because the number of =oral deaths in each of ~hese categories is qu/=e large, the difference between an estimate based on the lower bound and one based on an upper can amounu =o =ens of thousands of deaths. Some analysus have also also au~ribuued some dear/%s from digesuive system disease (e.g. s~omach and duodenal ulcers), accidental injuries (e.g. fires), and perinaual morrmliry (mothers who smoke during pregnancy tend ~o have higher ra~es ofmlscarriages and lower bir~hweight babies). Table I also presents several of the esuimates of the number of deauhs from smoking-realted disease. These range from around 300,000 deaths each year ¢o 485,000. The Surgeon General's reports on smoking have, in the last few years, presenned estimates of 129,000 deaths from cancer, 170,000 deaths from hear~ disease, and 50,000 deaths from chronic obstructive lung disease. Together, these ~oual to abouu 350,000 deaths annually. PREVTOUS ESTIMATES OF THE COSTS OF SMOKING-RELATED DTSEASE Generally, two broad cauegories of costs have been considered in IChronic obsuructive lung disease is a general category of disease than includes chronic bronchi=is, emphysema, and asthma. Physicians compleuing death ceruifica=es now more commonly use this general category, rather ~han refering specifically =o bronchi=is or emphysema. &q O C¢ &1

? DRm'Z (519185) DO NOT QUOTE, CITE, OR REPRODUCE previosus esuimaces of smoking-related disease costs. These are usually called ~he direct and indlrecu cosus of illness. The direct cosus are goods and services ~hat are used in caring for and treating ~hose with disease. Generally, ~he only direct costs r_hac are estimated are ~hose directly associated wir/Imedical care. Indirect costs are generally represented by the lose producuivi~y of ~hose who suffer from disease. Lost productivity is generally measured using lost wages, often with some adjustment for ~he household services provided by housewives. The literat-ure on r~he cost of illness has used ~o differen~ approaches to estimate direcu and indirecU costs. In ~he prevalence approach, all medical cosus are attributed co ~he year in which ~he money is act~ally spenu. . In ~he incidence approach, present and future medical costs2 are aucributed co r~he year in which r~he disease flrsu becomes manifesU. Thus, if r_he course of a disease involves medical crea~menu over r_hree differenu years before the pauienc dies, ~he prevalence approach would assign ~he costs separauely co each year. The incidence approach, on ~he ocher hand, would calculate a presenn value for ~he scream of costs over nhe ~hree-year period, and assign r~hau single sum to ~he firs~ year ~he disease was treated. For diseases chac involve less than one year of Crea~menu, ~he ~wo approaches are essenuially the same. 2These are expressed as present values, i.e. ~hey have been discounted to cake account of ~he time value of money. This is the notion ~hat, even in an economy witch stable prices, a dollar today is worth more than a dollar tomorrow. This is so because wi~h a dollar ~oday, an investment can be made CO earn a return. The appropriate race of in~erest to use for discountin~ future effects, however, has been a matzer of some dispute.

• ° ,L " i. * ,r DRAFT (5/9/S5) L~ DO NOT qUOTE, CITE, OE R22KODUCE For uhe indirect costs, ~he prevalence approach assigns ~he costs of los~ productlvi~y due Uo morbidity co ~he year in which ~he producclvicy is lose. In ~he prevalence approach, ~he future earnlnEs lost due co mortallry are discounted Co uhe presen~ and assigned uo ~he year of death. In ~he incidence approach, all indirect costs due Co morbldicy and ~he lose fut,~re earnings due Co mortality are discounued Co ~he present and assigned Co ~he year ~he disease is first manifesu. The landmark work of rice and her colleagues (Rice, 1967; Cooper & Rice, 1976), which discusses ~he costs of all diseases, used ~he prevalence approach, as does ~he mosu recenu update of ~his method by Hodgson and Kopscein (Hodgson & Kopscein, 198A). Only recently have researchers ~-ried co implement uhe incidence approach. For example, Hartnmian and colleagues, have measured and compared ~he costs of coronary hearu disease, stroke, cancer, and motor vehicle injuries (Haruunian, et el. ). Table 2 presents previous estimates of ~he costs of smoking. One of ~he earliest estimate of ~he costs of smoking was ~hac of Hedrick. He used ~he resul~s of a Canadian s~udy of ~he cosrm of smoking in Canada for lung cancer, coronary heart disease, chronic bronchiuis and emphysema. The resulting costs of $278 million was increased hy 50 percenu co cake account of diseases not included. AfTer adding in estimates of -~he costs of morbidicy and of fires (from =he Canadian study), the total amounted to $526.5 million. An esclmate of U.S. costs was made by multiplying this figure for Canadian costs by !0--=he ratio of =he U.S. Gross National Produce to ~ha~ of Canada (Hedrick, 1971).

DRAYT, (5/9185) DO NOT QUOTE, CZTE, OR REPRODUCE The most commonly cited estimate of the costs of smoking is one prepared by Lute and Schweiczer (1978). Followlng =he prevalence approach, they used the Cooper and Rice (1976) cost of illness statistics for the year 1972, inflated them Co 1975 dollars,3 and then attributed 20 percent of =he costs of neoplasms, 25 percent of circulatory system costs, A0 percent of respiratory system disease costs, and 1.i percent of the costs of fires. The total estimated heal=h care costs were about $8.2 billion, which, according to =heir calculations represented 7.8 percent of =he Coral health care expenditures in the U.S. The indlrecC costs of lost earnings amounted Co $6.2 billion for morbidity, and 12.9 billlon for mortality. Their coral was $27.5 billion.A Then HHS Secretary Joseph Califano, in the forward co the 1979 Surgeon General's report on smoking, estimated thaC smoking resulted in $5-8 billion in health care expenses--2.5 Co A percent of the Nation's health care costs of $205 billion. Lost produccivlt-y, wages, and absenteeism due co smoking related illness were estimated Uo amount Co $12-18 billion.5 Sidney Wolfe estimated =he morbidiuy, mortality, and direct healr_h care costs of smoking to be $18.9A billion in 1976. The morbidity costs were calculated using =he results of a Depar=menc of Heal=h, Education, and Welfare study thaC found =hac 19 percent of days lost from work were related to smoking. This resulted in a figure of $2.96 billion. To this he added $I.18 3Using the medical care component of the Consumer Price Index (~or medical costs) and the Gross National Product Implicit Price Deflator (for lost earnings. Aincluding $176 million for property costs due to fires. 51 have no= ye= determined how these es=imaces were generated, it seems curious that they are lower than the estimates Lute and Schweiuzer had published the previous year! 1'¢ 0 CO t'O GO

. DRAFT (519/S5) DO NOT QUOTE, CITE, 0R REPRODUCE billion =o account for morbidity among ~hose unable ~o work, for a to~al morbidity cost of $4.14 billion. Mortality costs and direct medical care costs were estimated using a==ibuuable risks derived from a Nauional Science Foundation report. Mortalit-y costs amounted =o $7.7 billion. Direcu heal=h care costs were esuimated to be $.93 billion for cancers, $1.99 billion for cardiovascular disease, and $1.67 billion for respiratory diesease in 1972. An adjustment for the increase in medical costs between 1972 and 1976 resulted in an es~ima=e of $7.1 billion for 1976 (Wolfe, 1977). Marvin Kristein derived his estimate of heal~h care costs from informa=ion ~hat implied ~hat smokers of one or more packs per day had a 50 percen~ ~rea~er hospi~alizauion raue r~han nonsmokers. Using data on ~he number of such smokers in 1975, and ~he total nauional spending on health care, he calculated r.hat smokers used $5.2 billion in heal~h care services in 1975. He used a National Center for Heal~h Statistics esuima=e ~ha= 77 million days were lost from work in 1965 due to cigaret=e smoking. These he valued at $40 per day ~o generate an estimate of $3.1 billion in losu productlviUy due to morbidity. Using an es~ima=e ~hau 300,000 deaths in 1975 were associated witch smoking, he calculated that =he los= earnings amoun=ed to $12 billion (Kristein, 1977). In a second article, Kristein esuima~ed ~he costs of borne by business firms for r~he "average" smoker. He included estimates of r.he cosus of heal~h insurance, fire losses, workers' compensation, absenueeism, productivity losses, and involuntary exposure ~o tobacco smoke. Added ~ogether, ~hese amounted ~o be~aeen $336 and $601 (1980 dollars) per smoker (Kris~ein, 1983). Co

DRAF (5/9/S5) DO NOT QUOTE, CITE, OR REPRODUCE Osier and colleagues estimated r~e costs of smoking-related cases of lung cancer, coronary heart disease, and chronic obstr~u:tive pulmonary disease using the incidence approach (Osier, ec al, 1984). They, however, only estimated r.he "expected value" of costs for age and sex-specific groups of smokers. For example, men aged ~0-4~ who smoke more than two packs per day incur, on average, a discounted ~otal of $56,670 in direct medical care costs and indirect costs due Uo lost productiviuy compared ~o non-smokers of the same age and sex.6 They did not attempt ~o agEregate these group specific costs for all of society. Rice and Hodgson (1983) have also developed estimates for r.he costs of smoking, using ~he prevalence approach. They developed more detailed estimates of the fractions of mornality associated wir.h smoking than ~he ones used by Lute and Schweitzer. The result was that they attributed about 22 percent of cancer dear~s, 16 percent of circulatory system dearths, 20 percent of respiratory system deanhs, 16 percent of digestive system deaths, and 3.5 percent of infectious and parasitic disease deaths (specifically, ~uberculosis) to smoking. These fractions were ~hen multiplied by ~he costs of medical care and lost productivity for each of these categories. In 1980 dollars, ~he direct medical costs amounted to $16.1 billion, r~he indirect costs of morbidity were about 6.9 .billion, and the indirec~ costs of mortality, about $19.2 billion. The total costs were thus about $A2.2 billion (Rice and Hodgson, 1983). 6Costs are in 1980 dollars and r_he discounu rate used was 5 percent. For women smokers of the same age Croup, the costs are lower, $19,000, largely because the indirec= costs are based on the average eaz-nings of women, which are substantially less than those of men. ba O O

• JC • i (- ( DO NOT QUOTE, CITE, OK KEPEODUCE The Minnesota Department of Heal~h has developed estimates of the costs of smoking-related disease for r, he state of Hinnesota. Their approach is generally r/~e same as ~hat used by Rice and Hodgson, wi~h ~he addition of specific information on the prevalence of smoking in M/nnesota. They estimate ~hat r.he direct medical care cost of smoking-related disease in Minnesota in. 1983 was $37&.6 million and ~he indirect cost of lost income due to premature deanh was $303.3 million. These total to $677.9 million or about $807 for each smoker in t/~e s~aue and $I.A8 for each pack of cigarettes sold in Minnesota (Minn. Dept. of Heal~h, 198&). Table 2 also presents ~hese estimated costs with adjustments for ~he effects of inflation since~hese estimates were made. After this adjustment, most of ~hese estimates appear to fall within a limited range: $15 to $30 billion in direct health care costs, and $25 to SA0 billion fn indirect productivity losses (in 1985 dollars). In addition, researchers have used information from surveys and the medical care data collection system of a Health Maintenance Organization, to compare smokers and nonsmokers in their use of medical services and the frequency of lost worktime and disability. Based on data from the 196A-5 and 197& National Health Interview Surveys, r/le Surgeon General's 1979 report on Smoking and Heal~h found ~hat current cigarette smokers tend to report more chronic conditions, such as emphysema and arteriosclerotic heart disease, than persons who never smoked. The incidence of acute conditions, such as influenza, was higher among smokers ~han among nonsmokers. Current smokers also reported an excess of work loss days and bed disability days than did the nonsmokers. Finally, current smokers and former smokers reported more