Tobacco Documents Online

I I I I I r I I I I I I I I r I I The Carter Center of Emory University Health Policy Consultation Atlanta, November 26-28, 1984 CLOSING THE GAP: Cross-Sectional Analysis of Unnecessary Morbidity and Mortality in the United States* Robert W. Amler, M.D.,1 Craig C. White, M.D.,1 Michael K. Berry,l Donald L. Eddins,' Nancy N. Fajman, M.M.Sc.,2 Daniel McGee, Jr.1 1. Centers for Disease Control 2. F.mory University School of Medicine 9

Presented November 26, 1984 We wish to acknowledge the technical assistance of Nancy Pearce, Ronald W. Wilson (National Center for Health Statistics), Judith Biamey (Emory University), William H. Foege, and James S. Marks (Centers for Disease Control). _ The consultants' reports have covered most deaths, or mortality, and most significant illnesses, or morbidity, in the United States. Each year, these 13 health problems account for about about 70% of hospitalizations, 80% of all deaths, and 85% of direct personal health-care expenditures. For each health problem, we have been told what is currently known about the size of the problem, generic causes or reasons for the problem, and the potential prevention of death, suffering, and cost that is projected if those generic factors could be eliminated. Certain factors are especially important because they affect multiple health problems. Often the seriousness of such factors is not immediately obvious. A risk factor normally is viewed as affecting only one or two health problems at a time. But the true impact of a factor may not be recognized unless viewed with respect to multiple health problems in a cross-sectional analysis. For example, the health impact of alcohol cannot be fully measured by looking only at deaths from liver disease, or drunk driving, or the cost of detoxification units. For this reason, emphasis was placed on generic risk factors as they pertained to multiple health problems, to identify those factors most responsible for the gap. MAJOR HEALTH PROBLEMS Unintentional injuries. Dr. Smith and colleagues identified specific injury risks, as well as alcohol, tobacco, and socio-economic position. Diabetes mellitus. Serious complications -- blindness, amputation, kidney failure, and stroke -- could be-reduced substantially by specific preventive health care. The role of tobacco as a cause of vascular disease must also be considered, although difficult to quantify precisely. Digestive diseases. Generic factors include preventive health care (for cancer of the rectum and colon), tobacco (for ulcers) and alcohol (for liver diseases). Cancer. Many Americans fear cancer because it seems mysterious and uncontrollable, yet at least one-quarter of all cancer deaths are caused by ~ known external factors: tobacco, alcohol, improper nutrition, and Q occupational exposures. Other respected estimates suggest one-quarter of all N cancer deaths are due to tobacco-_alone. LrI ~ .~ CG N W - C2 -

I I I I I r I r I I r I I I I I 0 Violence. Many homicides and__suicid_e_s are_rel_ate_d to hand guns and use of alcohol. In addition, there is a complex interaction of substance abuse, cultu rl beliefs, and socioeconomic factors that may be partly amenable to specific prevention services and to improved community supports for youth and for early drug users. Infectious diseases. These deaths are largely preventable through services such as immunization, early diagnosis and treatment, and surveillance. Although tobacco and alcohol probably exacerbate many respiratory infections, the precise effect is difficult to quantify. Dental diseases. The most important dental_diseases -- tooth decay and periodontal disease -- are amenable to preventive health care in the form of fluoridation and specific programs of early diagnosis and treatment. Res irator diseases. Most chronic lung disease is caused by tobacco. Virtually all cases and deaths could be eliminated if tobacco use and certain occupational exposures were eliminated. In addition, specific prevention services can reduce the impact of asthma, tuberculosis, influenza, and pneumonia. Arthritis. The health impact of arthritis and other musculoskeletal diseases is substantial, yet reducible by specific prevention services such as calcium supplementation to prevent osteoporosis. Depression and alcoholism are complex problems that may partially be addressed by early detection and by specific prevention services. Infant mortality and morbidity. Several factors contribute to these problems, including unintended pregnancy, tobacco, alcohol, improper nutrition, preventive health care, and socioeconomic position. Cardiovascular disease. A_s many as two-thirds of these deaths are preventable. These are largely attributable to tobacco, high blood pressure, diet, and lack of exercise. GENERIC RISK FACTORS The goal of this project was to look at risk factors cross-sectionally over multiple health problems. Tobacco is a risk factor for morbidity and mortality resulting from cardiovascular disease and diabetes, cancer, respiratory diseases, digestive diseases, injuries and, perhaps, infections. Alcohol is responsible for__uninten_tional injuries, cancer, violence, infant morbidity and mortality, depression and alcoholism, and infections. -C3-

Injury risks. There_a_re generic risks associated with both unintentional and intentional injuries. Unintended pregnancy is associated with infant morbidity and mortality, and, though difficult to quantify, must impact on maternal health, violence, and depression. Prevention services, though diverse in type and application, clearly are of major importance in reducing morbidity and mortality for almost all of the health problems discussed. Violence, depression, and substance abuse are in fact generic problems which impact broadly not only on the physical and mental health of the individual and the family, but on the health of society as well. ANALYSIS We began this study with a cross-sectional review of the 13 high-priority health problems as reported by the consultants. We standardized definitions of the criteria that were measured to enhance comparability of the data, and made adjustments to minimize duplication between different health problems. Where risk factors overlapped, a "cascade" priority system was used to assign attributable proportions of morbidity and mortality. This model was used with the recognition that it tends to underestimate the impact of factors that are lower in the cascade because it assigns a single underlying cause for each death or event. Although the analysis was driven primarily by mortality data, morbidity was felt to be roughly parallel in most instances. Exceptions to this were arthritis, dental disease, depression and violence. The relative importance of these conditions was substantial when morbidity measures were applied. FINDINGS In 1980, the base-year used in the analysis, the US resident population was just over 227 million. There were 3.6 million births and 2 million deaths, including nearly 47,000 deaths among infants under 1 year old, for an infant mortality rate of 12.6 per 1,000 live births. Premature deaths in 1980 accounted for an estimated 12 million potential years of life lost before age 65. This estimate -- years of_potential life lost before age 65 -- is commonly used to measure premature death, and is derived from the number of deaths that occured in each age group during the year. In 1980, patients received approximately 277 million days of in-patient hospitalization in non-Federal, non-psychiatric, short-stay hospitals. The total national health expenditure was $249 billion. Of this total, $173 billion was expended for direct personal health care: defined as short-term -C4-

I I I I I I I I I I I I I I I I I hospitalization ($92 billion), physician, ($63 bitlion),and drugs ($18 billion). dentist, and other professional care Fil;ure I represents all deaths due to the 13 priority health problems. Sixty-six percent of these deaths are potentially preventable. The risk factors associated with these deaths include: Tobacco, high blood pressure, nutrition, screening, alcohol, injury risks, and others (including prevention services, hand guns, unintended pregnancy, occupational risks). Looking at premature death, as measured by years of potential life lost before age 65, 64.7% of the total is potentially preventable (Figure 2). Either way, about two-thirds of mortality is unnecessary. The distribution of risk fnctors associated with these deaths is somewhat different: Tobacco, alcohol, injury risks, prevention services, screening, unintended pregnancy, high blood pressure, and others (including hand guns, nutrition, occupational risks). The totals are so large -- 1.2 million deaths and 8.4 million years of life -- that minor variations in disease occurrence or in the model used for analysis have little impact on these distributions. Tobacco, high blood pressure, improper nutrition, and screening are important factors where death numbers are concerned. Other factors, such as alcohol, injury risks, prevention services and unintended pregnancy, are equally important because they affect the survival of young people. Impact of tobacco. Tobacco -is the leading cause of death in the United States. Tobacco causes approximately 360,000 deaths each year according to the Surgeon General's reportz or nearly 1,000 unnecessary deaths every day. Most of these deaths occur as heart attacks, strokes, and diabetes, cancer, and chronic lung disease. Note that tobacco causes more deaths by cardiovascular diseases than by cancer. Still, the cancer problem is large. Tobacco leads all other substances as the greatest carcinogen known to man (and to woman) and is responsible for the fact that more women will now die of lung cancer than breast cancer. Furthermore, tobacco causes almost all chronic lung disease -- more than asbestos and coal dust, even among workers who are regularly exposed to those substances. Nearly one-third of Americans who die of heart disease and stroke are less than 65 years old, and one-quarter of these deaths are attributable to tobacco (Figure 3). At least two additional health problems are important causes of premature mortality: Infant_mortality (resulting from low birth weight), and fires/burns. In fact, cigarettes are the leading cause of deaths from house fires -- over 2,000 killed every year, mostly children. And tobacco has other serious effects, such as ulcers and vascular disease. Though seldom resulting in death, these conditions are disabling and often necessitate surgery or amputation of a limb. Impact of alcohol. Alcohol is_the second leading cause of premature death, about I arrd a half million years of potential life lost before age 65 (Figure 4). The most commonly associated conditions are injuries (mostly car -C5- i

 crashes), liver diseases (mostly cirrhosis), alcohol-related violence (homicide and suicide), and cancer (mostly mouth, larynx, and esophagus). All of these conditions -- including cirrhosis -- are particularly tragic because they primarily kill young people. When used excessively alcohol also has serious effects on the heart and nervous systems and on the fetus. Impact of prevention services._ Much unnecessary, premature death is preventable through appropriate prevention services. Specific prevention services include prenatal care and appropriate newborn care, to reduce infant mortality; immunizations, rapid diagnosis and treatment, and surveillance, to reduce infectious diseases; targeted interventions to prevent homicides and suicides; and specific programs to maintain cardiovascular fitness and prevent complications of diabetes. Other appropriate services may prevent or reduce arthritis and osteoporosis, dental diseases, depression, alcoholism, drug abuse, and respiratory diseases. In addition, screening is of considerable importance in preventing deaths from cancer of the breast, cervix, and colon. Beyond preventing unnecessary death and suffering, many of these services have proven cost-benefit, but are not universally available to all Americans, hence, we pay more. Impact of unintended pregnancy is serious, particularly for teenagers. Teenage mothers earn half the income of those who first give birth in their 20's, and teenage fathers are less likely to complete high school than other men. Families in which the mother gave birth as a teenager account for about half of the 9.4 billion dollars paid by AFDC, or Aid to Families with Dependent Children. ,Impact of violence, depression, and substance abuse is substantial, though difficult to gauge from mortality statistics alone. The predilection of these problems for the young and for minority groups has important social implications beyond what statistics can show. Impact of socioeconomic level_has been alluded to repeatedly. Dr. Kaplan has provided some examples of the cross-cutting effect of socioeconomic position on multiple health problems. Each working group may find it useful to consider the role of socioeconomic position when considering specific recommendations. S U24MARY We have defined the gap; it is large and largely closeable. The data presented by the consultants__ clearly indicate that we already know the factors responsible for more than half of all deaths and major illnesses in this country, and those factors are within the realm of existing technology. The challenge now is to use our diverse talents to close the gap. 0 N    N a N   r a U -C6-

a a a a a a K a a a a K a a a a Figure I D EAT H~ ! f',\! U.~. P 0 P U LAT 0 [",T, 1980 1 3 LEADI NG' CAUSES BL.rL.!ZhL.lJSsS.lZ

Figure 2 `1'E,A,R,S) Cj'E hi FE: L(-DST BEEOP E(C-)5 1 ;.~ LEADING CAUSES, U.S., 1980 6ZSZf,OSZQZ I

a a a a a a a a A m m m m m m m m Figure 3 SMO KING - LOSS OF LIFE YEARS LOST BEFORE AGE 65 CIRCUL: (4-1.8%) oE8zhoSzOz

Figure 4 ALCOHOL - LOSS OF LIFE YEARS LOST BEFORE AGE 65 CANCER: (1 .9%) ALCOHOLISM: (12.H%) CIRRHOSIS: (16.9%) [EazfiaszOz