Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
s Characteristics of the Environment, Aging, and the Aged The toxic environments of the aging and the aged are diverse. A person's environment includes at least the following compo- nents: water, food, air, x rays, ultraviolet radiation, visible light, heat, life-style, pharmaceuticals, and health maintenance. Food and water can be considered separately, inasmuch as a federally established dichotomy assigns the analysis and regulation of water to the Environmental Protection Agency and the analysis and reg- ulation of food primarily to the Food and Drug Administration. Scientific considerations have followed this dichotomy. Life-style is included as a component of a person's environment because of the profound impact of such phenomena as smoking, recreational sun exposure, and recreational drug use on disease prevalence and longevity. Pharmaceuticals, even though they play an important role in health maintenance, are considered separately. Health main- tenance is a more generic and inclusive category, and changes in health care (with changes in diet) have been documented as the principal cause of extended life expectation in many countries (U.S. NTEHS, 1977, 1984~. Although a person might live in a general environment that seems constant, even over a long period, exposure to potentially toxic agents varies widely, that is, the toxic environment changes. 72
ENVIRONMENT, A GINO, AND THE A GED 73 Chemicals in the air, in food, and in water vary widely, even in isolated environments. Whether these changes are significant with respect to aging or the aged is one of the main questions in this study. In any event, it is both useful and correct to state that there is no average toxic environment. Therefore, no average toxic environment can be simulated ~ a laboratory, and there is no way to perform laboratory experiments whose results can be ex- trapolated in a general way to the composite human environment. However, the fact that human populations live in different environ- ments offers an opportunity for comparative studies in biochemical toxicology. All elements of the human environment contain toxic sum stances. The important question is whether exposures to these substances affect aging or produce toxic effects. This question can be framed in the context of each constituent of the environment. Water. The content of drinking water varies widely with locale. Generally, few substances are deliberately added to drink- ing water; those deliberately added include chlorine species and fluorides. Some chemicals found in trace concentrations in water supplies have been tested for some toxic effects and others have not. For instance, pesticides and other commercial chemicals have been detected ~ many water supplies, but at concentrations deemed by regulatory agencies to constitute only reasonable hazards to pub kc health. What is not clear is whether the toxicologic methods applied by regulatory agencies can evaluate age-associated toxic effects adequately. . Foodi. The term food, as opposed to diet, applies to the intake in foodstuffs of possibly toxic substances that can alter life span, as distinct from quantitative intake of putatively nontoxic normal foodstuffs. Food consists mainly of complex mixtures and is not well defined from a toxicologic perspective. The ingestion of food accounts for the bulk of the chemical substances that enter the body, of which most, but not all, are probably innocuous. Food acts both as an inducer of biologic processes relevant to aging and as a carrier of potentially deleterious substances. These substances can be divided grossly into four components: natural constituents, contaminants from the environment of its growth, contaminants from its processing, and contaminants from its preparation. · Air an] its contaminants. Inhalation has the third great- est responsibility (after ingestion of food and water) for bringing
74 AGING IN TODAY'S ENVIRONMENT chemical substances into contact with the body. Of the airborne agents that produce specific toxic effects, the major one is cigarette smoke. Inhalation can bring both gases and particles into contact with the lining of the airways. These substances differ with re- spect to absorption and clearance via the mucociliary escalator and possible toxic effects. Breathing volume, clearance, and am sorption vary with age. The makeup of air varies greatly among geographic locations and among indoor, outdoor, and occupational environments. The aged generally spend more time indoors and less time in occupational environments than do others. But the variation among indoor environments is so great that it precludes generalization about exposure. ~ Pharmaceutical exposure. Pharmaceutical agents, as op- posed to environmental agents, are designed to elicit biologic ef- fects. The toxicologic data on pharmaceuticals are extensive and usually include substantial human response data. Consideration of the pharmaceutical environment is ~rnportant to the problem being evaluated in this report for two main reasons. First, the aged are exposed to pharmaceuticals more than the young. Sec- ond, some pharmaceutical agents place biologic stress on exposed people. If an essential component of aging is a reduction in the capacity to respond to stress, then consideration of aged people's responses to pharmaceuticals could be important. There is no type ical or average pharmaceutical environment for the aged, but some groups with defined pharmaceutical exposure for extended periods might be important in studying the response of aged populations to toxic stress. . L,ife-style. Life-style including smoking, sun-tanning, and taking of recreational drugs is included as a separate category of environmental exposure for two reasons. First, life-style is consid- ered as creating optional environmental exposure. Second, expo- sures in this category are known or suspected to increase disease prevalence that might shorten life or mimic aging processes, or both. Thus, this category of environmental exposure must be considered avoidable toxic stress that is relevant for only some populations. . Bealth maintenance. This category is included only be- cause improvement in health care changes life expectancy dramat- ically in some human populations (Schneider and Reed, 1985~.
ENVIRONMENT, AGING, AND THE AGED 75 Improvement in health care is usually accompanied by improve- ment in nutrition; it has been difficult to separate these two factors as contributors to life-span extension. NUTRITION Until fairly recently, there has been a tendency to consider the issue of chemucal toxicity and the environment in terms of the contamination of the "natural" environment with man-made pollutants, such as DDT. Efforts like those of Higginson (1969) and Doll and Peto (1981) have led to a growing recognition that the environment includes such factors as diet and life style. Indeed, one of the most important factors in chronic human toxicity is natural components of the diet. Whereas man-made pollutants (so-called contaminants) are commonly measured in picograrns, dietary components that can be toxic are more often measured in milligrams or even grams. Common foods are replete with naturally occurring biologically active compounds, for example, vasoactive amines in bananas, cyanogens in corn, oxalates in spinach, goitrogens in cabbage, and anti-inflammatory compounds in licorice (Sapeika, 1978~. This is not surprising, in that the plants forrn~ng much of our diet have evolved mechanisms to prevent overpredation by insects (natural insecticides~and other animals(Ames,1983~. As a whole, it appears (Ames, 1983) that the most important chemical exposure of humans, both in quantity and in diversity, results from diet. Thus, diet is considered to be one of the most important aspects of the environment that should be taken into account in the studies of chemical toxicity and aging. Diet is a broad term that usually includes natural contaminants, such as mold-produced aflatoxin; the residues of food preparation prac- tices, such as mutagens produced by charring of foods (National Research Council, 1982~; and food additives. However, in focusing this discussion on nutrition, we will emphasize the portion of diet that maintains an organism's health or growth. An interesting topic that is excluded from the present discus- sion is the recently suggested importance of some environmental agents in the induction of parkinsonism, a disease considered char- acteristic of age. The suggestion was based on the development of a mode! for parkinsonism related to the production of a metabolite of an agent that causes parkinsonism in humans and animals, and
76 AGING IN TODAY'S ENVIRONMENT on the presence of antibodies to this metabolite in people who have parkinsonism but no history of exposure to the chern~cal (Lewin, 1986~. It is well known that the lack of a nutrient often leads to a deficiency syndrome and that an excess of the same component can sometimes lead to overt toxicity. The current discussion con- centrates on manipulations within what is generally considered to be a fairly normal range. Effects of Nutrition on CCbx~city Various nutritive components have major effects on the ex- pression of toxicity. For instance, by increasing the amount of casein given to rats for a month, one can increase the LD50 (the dose required to kill 50~o) of some herbicides by a factor of at least 6 (Shakman, 1974~. Protein quality and methionine content Knight also affect the detoxification of some pesticides (Boyd, 1972~. Compared with purified diets, cereal-based chow suppressed dipheny~hydantoin-induced cleft palate in mice (Mc- CIain and Rohrs, 1985~. Because chronic disease often has a long latency, it can be especially susceptible to the influence of nutrition. Nutrition is important in the pathogenesis of coronary heart disease, gallstones, appendicitis, varicose veins, obesity, hiatus hernia, and cancer (Burkitt, 1982~. Some of these relationships have become evident through the effect of a change to Western diet in groups whose traditional diets were different; for example, Eskimos who switched to Western diets suffered an increased rate of acute appendicitis (SincIair, 1982~. Other effects have been revealed by epidemiologic studies, such as the Framingham study, which related cholesterol to risk of coronary heart disease (Kanne! et al., 1971~. Experunental studies demonstrating the role of diet in chronic disease have been plentiful since the beginning of modern experi- mental nutrition; for example, some diets influence the progression of atherosclerotic lesions in rabbits (Anitschkow, 1913) and diet modifies induced tumorigenesis in mice (Watson and Melianby, 1930). In view of the long history of such studies, the lack of under- standing of the mechanisms by which nutrition influences chronic disease and toxicity is surprising. One reason is the complexity of the diet. As noted above, many substances are consumed in food.
ENVIRONMENT, A GINO, AND THE A GED 77 Changes in dietary composition can alter xenobiotic metabolism (Rogers and Newberne, 1971), change intestinal microflora quan- titatively and qualitatively (Stasse-Wolthuis, 1981) thereby mod- ifying transport and production of bacterial metabolites, affect bile acids (Ready et al., 1980), and change glutathione synthetase (Beutler, 1972~. With the number of substances in food, many of which are needed for maintenance of function, almost any step in the interaction of agent and organism is likely to be influenced by altering nutrition, which would thus lead to an effect on toxicity. The effects of individual macronutrients, such as dietary pro- tein and fat, on toxicity, longevity, and associated aging events have been explored by altering the dietary content of the macronu- trients being assessed. A major problem with this type of analysis is that such changes in dietary composition might influence the amount of food eaten. Unfortunately, the food intake is often not accurately determined or not taken into account In the interpreta- tion of findings. Because the amount and type of food eaten might profoundly influence both aging processes and toxicity of environ- mental agents, failure to address this issue reduces the value of a study. The micronutrients, that is, minerals and vitamins, are diffi- cult to study in the context of toxicity and aging partly because there are so many substances in each class of nutrients. In addi- tion, toxicity data are generally lacking on these compounds and elements. There is also little information on the role of these sum stances in aging, although considerable study has been aimed at ascertaining the relationship of specific components to age-related pathologic processes (e.g., calcium and vitamin D in relation to osteoporosis and calcium and sodium in relation to hypertension). Because a given nutrient is probably involved at several stages in the maintenance of body function, it might have different im- pacts on toxicity. This is especially true for the more complex toxic effects such as cancer. Although several factors in the diet might be involved in the toxicity, evidence that one agent can inhibit or promote the carcinogenic effects of another, depending on the sequence of exposures (e.g., Kitagawa et al., 1984), indicates that the effect of altering nutrient intake on complex toxic end points is difficult to determine.
78 AGING IN TODAY'S ENVIRONMENT Nutrition and Cancer Cancer is actually over 200 different diseases. It occurs in al- most all tissues that have the capacity to replicate. A recent review summarized much of the present consensus about the mechanisms of cancer (U.S. Interagency Staff Group on Carcinogens, 1986~. Simplistically, cancer can be considered as a multistage process that usually involves agent uptake; agent metabolism (for car- cinogen activation); cellular processes, such as DNA repair and proliferation; and organismic responses, such as attack by the immune system. Each of these major factors can be affected by changes in nutrition. A recent review (NRC, 1982) analyzed many of these issues. The following discussion gives some examples of the insights that have been gained on the relationship between nutrition and cancer. One of the most well-tested effects in modulation of carcino- genicity by nutrition is the inhibition of spontaneous and induced carcinogenesis by restriction of total caloric intake. In practice, this is usually accomplished by allowing ingestion of less food than is eaten by a control population. Many studies have demonstrated the effect of caloric restriction, including that of Tannenbaum (1945), who characterized the effect on benzota~pyrene-induced mammary tumors in mice, and Kritchevsky et al. (1984), who found a similar effect in 7,12-dimethy~benz~ajanthracene-induced mammary tumors in rats. Life-span studies have shown a decrease in the incidence of spontaneous tumors of many types. Human ev- idence is less clear because it is hard to separate the effects of changes in diet composition and total caloric content, socioeco- nomic status, and so on; but there is some evidence that the incidence of colorectal cancer is affected by caloric intake (e.g., Hill et al., 1979~. Lower dietary protein content is generally associated with lower incidences of tumors (NRC, 1982), such as 3-methy~cholan- threne-induced mammary adenocarcinoma in rats fed high-protein diets (Stray et al., 1964) and aflatoxin-induced liver tumors in rats (Newberne and Rogers, in press). In contrast, Clinton et al. (1979) found that a decrease in dietary protein increased the incidence of mammary tumors induced by 7,12-dimethy~benz~aJanthracene and that the protein effect depended on the sequence of nutritional changes and agent administration. Increased dietary fat has been implicated in an increasing
ENVIRONMENT, AGING, AND THE AGED 79 incidence of cancer at several sites (NRC, 1982~. For example, in- creasing dietary fat from To to 20~o usually increases the incidence of mammary tumors both spontaneous tumors and, depending on the carcinogen used, induced tumors. Interpretation of the ef- fect has been complicated by the observations that restricting total calories eliminates much of the high-fat effect (Kritchevsky et al., 1984), that energy intake, and body size play a role (Boissonneault et al., 1986), and that essential fatty acids are also important. Dietary fiber has been implicated in modulating toxicity. How- ever, there are many different dietary fibers, and they might act by various mechanisms, such as by altering the metabolism of di- etary agents (DeBethizy et al., 1983) or altering the turnover of gastrointestinal epithelium (Cassidy et al., 1981~. Most of the recent work on cancer and vitamins has focused on ~ritarnins A, C, and E. As a rule, vitamin A deficiency increases susceptibility to tumors, and an increase in its intake is protective. However, vitamin A also induces tumors in hamsters (Smith et al., 1985), and retiny] acetate, a related compound, increases hormone- induced mammary carcinogenesis in mice (Welsch et al., 1981~. Vitamin C has little effect on carcinogenesis except through its inhibition of the intestinal reaction of nitrites with Gaines to form nitrosamines (Mirvish, 1981~. There are a number of mechanisms by which vitamin E could inhibit cancer formation, particularly by its antioxidizing activity. Most animal studies have tended to confirm this effect, but others have found no such effect, and a few have found that vitamin E enhances carcinogenesis under some conditions (Newberne and Suphakarn, 1983; Toth and Patil, 1983; Wattenberg, 1972~. There has been little systematic evaluation of the effect of minerals on carcinogenesis. Selenium can be carcinogenic (USNCI, 1980~; but it also appears to protect against the formation of some induced tumors (Medina et al., 1983; Newberne et al., 1986~. Thus, the role of nutrition in carcinogenesis is far from fully understood. However, two observations are worth noting in re- gard to the effects of nutrition on aging. One is that the effects of changes in nutrition or carcinogenesis are not simple or even monotonic (i.e., an increasing effect with increasing nutrient con- tent). That is not surprising, inasmuch as the diet is a complex mixture and carcinogenesis is complex. Defining the components of and biologic response to a standard semisynthetic diet, such as
80 AGING IN TODAY'S ENVIRONMENT NIH-31, might eliminate some of the problems encountered with the cornrnonly used commercial diets. The second observation (discussed more extensively in the fol- lowing sections) is that, of the various nutritional modifications . . . . . . ~ , caloric restriction is likely, at least initially, to be the most gen- erally important for chronic end points. Although caloric restric- tion is a broad brush that involves many integrative, physiologic, metabolic, and cellular processes, it appears to have the most reproducible effects on chronic end points. Therefore, it is reason- able to assume that it is highly likely to provide a route to the understanding of mechanisms of aging. Nutrition and Agmg Although the discussion of environmental influences on toxic- ity and aging has not traditionally considered nutrition explicitly, eating and drinking are the greatest sources of chemical exposure of humans. Diet has an effect on toxic response, both acute and chronic. Nutrition is also known to be a factor in the develop- ment of chronic disease in humans and animals. However, the complexity of diet involves many steps in the interaction of agent and organism. The more complicated the interaction, the more complicated the influence of diet. Indications that this is true are the multiphasic responses of agents that interact with carcinogens and that can function as promoters or inhibitors, depending on the timing of exposure. Such complex behavior is manifest when cancer is used to mode! the interaction of toxicity and nutrition in regard to aging processes. . .. ~ .. . . . Background and Criteria for Evaluating Aging The data from the cancer model on the effect of nutritional modulation in general en cl caloric restriction in particular consti- tute only one of many indications of an important role of nutrition in toxicity. In addition, nutrition is often claimed to be important in influencing the aging processes of humans and other animals (Porte, 1980~. Some published data do indicate that nutrition modulates many adverse effects of aging (Masoro, 1985~. Further- more, some nutritional regimens and dietary components appear to promote aging processes, or at least the occurrence of age- associated deterioration (Guigoz and Munro, 1985~. A substantial
ENVIRONMENT, AGING, AND THE AGED 81 data base exists on nutrition and aging, as well as on the role of nutrition in cancer, but strategies to determine the mechanism by which nutrition modifies these processes are needed. However, before considering the effect of nutritional modifica- tion of aging, we should clarify some of the concepts that underlie research in this field. In tumorigenesis, "bumps and lumpish can be counted fairly directly, but the situation for aging is more complex. Changes in longevity have been widely used as criteria for determining the impact of dietary modification on senescence. However, there are problems with that approach. For instance, during the last 140 years, life expectancy at birth in the United States has markedly increased (Hazzard, 1983~. The increase is probably not a result of factors influencing basic aging processes, but rather results primarily from protecting the population from premature death due to specific infectious diseases and injuries. A change in the life span of a species is a better criterion than life expectancy. A nutritional manipulation that increases life span probably does so by lowering the rate of aging. However, a nutritional manipulation that decreases life span might do so by speeding the aging process or through a number of other possible mechanisms, such as modifying responses to toxic reactions or speeding disease processes. A major problem in the use of longevity as a criterion to identify factors influencing the rate of aging is the length of time needed to make this assessment. Indeed, for practical purposes this criterion can be used only with short-lived animal models such as rodents. Therefore, there has been much effort toward developing reliable biomarkers of aging (Reff and Schneider, 19823. It has been difficult to be certain that a particular functional or morphologic measurement is a reliable biomarker of aging, be- cause it is not possible to define aging in the fundamental biologic terms that could serve as a primary standard. Changes in many functional activities and the occurrence of many disease processes are associated with aging, and it is important to learn about nu- tritional manipulations that influence these age-associated events. However, the influence on such age-associated events should not be taken as evidence that a nutritional manipulation has modulated the aging processes.
82 Food Restriction AGING IN TODAY'S ENVIRON~NT The only nutritional manipulation that has been shown to increase life span in mammals is food restriction, and it has been truly demonstrated only in laboratory rodents. The phenomenon was first demonstrated by McCay and Crowell (1934) and has been confirmed by many others (Barrows and Kokkonen, 1977~. Although the dietary protocols have varied widely, all successful approaches have reduced caloric intake by 20-60% (weindruch' 1985~. In the initial studies, food restriction was started at wean- ing; it was later found that life span also increased when food restriction began during adult life (Beauchene et al., 1986; Cheney et al., 1983; Goodrick et al., 1983; Stuchlikov et al., 1975; Wein- druch and Walford, 1982; Yu et al., 1985~. Indeed, Yu et al. (1985) found that food restriction begun at the age of 6 months (early in adult life) was as effective in extending the life span of rats as that begun at the age of 6 weeks (2 weeks after weaning). This effect of food restriction has not been shown in other mammals, but careful research has not been done on long-lived marnInals. Findings similar to those in rodents have been obtained in nonmammals: protozoa (Rudzinska, 1952), rotifers (FanestiT and Barrows, 1965), Dapinia (Ingle et al., 1937), Drosophila (Loeb and Northrop, 1917), and fish (Comfort, 1963~. Food restriction not only influences longevity, but delays or prevents many age-related functional changes (Table 5-1~. It also slows or prevents a spectrum of age-associated diseases (Table 5-2~. McCay et al. (1935) suggested that food restriction delayed the aging processes by slowing growth and development. The findings of Barrows and Roeder (1965) supported that view. Indeed, for many years, the view prevailed that retarding maturation and slowing and prolonging growth were at the basis of the effects of food restriction on longevity. However, since 1965, many studies have challenged the concept (Cheney et al.5 1983; Goodrick et al., 1983; Stuchlikov et al., 1975; Weindruch and Walford, 1982~. Moreover, Yu et al. (1985) found that food restriction in rats only from the age of 6 weeks to the age of 6 months (the period of rapid growth) was less effective in extending life span and retarding age-associated processes than food restriction begun at either the age of 6 weeks or the age of 6 months. Those findings have redirected views on the action of food restriction away from growth
ENVIRONMENT, AGING, AND THE AGED TABLE 5-1 Action of Food Restriction on Age-Associated Physiologic Changes 83 Age-Associated Physiologic Change Delayed or Partially or Completely Prevented by Food Restriction Species References Increase in serum cholesterol Increase in serum triglycerides Rat Liepa et al., 1980; Rat Increase in serum parathyroid hormone Increase in serum calcitonin Loss of response of adipocytes to insulin Loss of response of adipocytes to glucagon Loss of response of adipocytes to catecholamines Loss of neurotransmitter receptors and Rat related central nervous system functions Loss of soluble gamma cystallins from eye lens Lose of reproductive function Loss of skeletal muscle structure and function Loss of spontaneous locomotor activity Loss of immune function Rat Rat Rat Rat Mouse Rat Rat Masoro et al., 1983 Liepa et al.. 1980; Masoro et al., 1983 Kalu et al., 1984 Kalu et al., 1983 Rea~ren et al., 1983 Bertrand et al., 1980b; Voss et al., 1982 Yu et al., 1980 Joseph et al., 1983; Levin et al., 1981; London et al., 1985 Le~reille et al., 1984 Merry and Holehan, 1979; Merry and Holehan, 1985 McCarter et al., 1982 Rat Yu et al., 1985 Mouse Cheney et al., 1983; Fernandes et al., 1978; Weindruch et al., 1983 and development to retardation of aging processes in the mature animal. Berg and Simrns (1960) first proposed that food restriction could increase longevity by reducing body fat. Bertrand et al. (1980a) showed that food restriction decreased not only body fat but also the number of adipocytes in the fat depots. However, the group of rats fed ad libitum showed no correlation between body fat and length of life. Moreover, within the group of food-restricted rats, length of life was significantly and positively correlated with body fat. Thus, the lower fat mass of food-restricted rats does not appear to be involved in the life-prolonging action. On the basis of data from a study in which the duration of food restriction and the part of the life span involved were varied, Stuchlikov et al. (1975) reachect the same conclusion. And Harrison et al. (1984), who found that food-restricted obese (ob/ob) mice were fatter but
84 AGING IN TODAY'S ENVIRONMENT TABLE 5-2 Action of Food Restriction on Age-Associated Disease Processes Disease Process Retarded Totally or Partially by Food Restriction Species References Chronic nephropathy Rat Cardiomyopathy Gastric ulcers Osteodystrophy Metastatic calcification Neoplastic disease Autoimmune renal disease Mice Hypertension-related problems Rat Rat Rat Rat Rat Rat. mouse Berg and Simms, 1960; Bras and Ross, 1964; Nolen, 1972; Sexton and Kimball, 1941; Tucker et al., 1976; Yu et al., 1982; Maeda et al., 1985 Maeda et al., 1985 Maeda et al., 1985 Maeda et al., 1985 Maeda et al., 1985 Maeda et al., 1985; Sexton and Kimball, 1941; Tannenbaum, 1945; Ross and Bras, 1965; Silberberg and Silberberg, 1955; Cheney et al., 1983; Pollard et al., 1984 Fernandes et al., 1978; Lloyd, 1984 lived longer than lean littermates fed ad libitum, concluded that reducing body fat does not play a major role in the action of food restriction on longevity. The proposal by Sacher (1977) that food restriction reduces the metabolic rate and by so doing retards the aging processes has been widely embraced. Indeed, Harman (1981) linked the hypothesis to the free-radical theory of aging: if the rate of electron transport and thus oxygen use are reduced, the rate of generation of oxygen free radicals is reduced. Indeed, if these free radicals are important contributors to the aging process, then reducing their production should slow the aging process. Sacher based his proposal on published data and concepts. Early in this century, Rubner (1908a,b) proposed, on the basis of his studies with domestic animals of various sizes, that all species use a similar number of calories per unit of body mass per lifetime. Pear} (1928) generalized that line of thought and proposed that the higher the metabolic rate per unit of body mass, the higher the rate of aging and the shorter the life. Moreover, there is evidence that limiting food intake reduces the metabolic
ENVIRONMENT, AGING, AND THE AGED 85 rate per unit metabolic mass (Apfelbaum, 1978; Forsum et al., 1981~; however, these studies were of short duration (weeks in the case of rats and months in humans). Sacher supported his hypothesis directly with calculations based on published data of Ross (1969) that showed that food-restricted rats consumed the same number of calories per gram of body mass per lifetime as rats fed ad libitum. Recently, McCarter et al. (1985) directly measured the meta- bolic rate of rats fed ad libitum and of rats undergoing prolonged, life-extending food restriction. They found that food restriction that markedly increased life span and retarded a spectrum of aging processes did not decrease metabolic rate per unit of lean body mass or "metabolic mass." Thus, the hypothesis that food restriction slows the aging processes by lowering the metabolic rate should be discarded, because its effects on aging can occur without a decrease in metabolic rate. Masoro et al. (1982) challenged the classical view that food restriction acts by reducing the intake of calories or other nutrients per unit of metabolic mass. They found that lean body mass of food-restricted rats was decreased in proportion to the decrease in caloric intake; nutrient or caloric input per unit of lean body mass or metabolic mass did not decrease. Thus, food restriction cannot be linked to aging processes by a reduced input of calories or any other nutrient per unit of tissue mass. Rather, the total organism's response to food restric- tion must be involved. The most likely possibility is that food restriction influences the regulatory systems endocrine, neural, or both and that food restriction is linked to aging processes in tissues and organs through these regulatory systems. The nature of the specific regulatory systems involved remains to be defined. One possibility was suggested more than a decade ago by Everitt (1973), who proposed that food restriction decreases the secretion of an aging factor by the pituitary gland. He and his co-workers found that removing the pituitary (hypophysectomy) of rats (given cortisone replacement therapy only), like food re- striction, increased life expectancy, increased life span, inhibited the onset of renal and neoplastic disease, and retarded the aging of collagen (Everitt et al., 1980~. The hypothesis was further devel- oped by the following model (Everitt, 1982~: food restriction alters
86 AGING IN TODAY'S ENYIRON~NT neurotransmitter metabolism within the hypothalamus, which de- creases the secretion of a hypothalamic releasing hormone and thus results in decreased secretion of a pituitary hormone. Although the hypothesis is intriguing, it is not yet supported by a solid data base. Hypophysectomy inhibits food consumption, and that makes it difficult to know the extent to which the findings on aging are due to a direct endocrine influence on aging or are secondary to the decrease in food intake. If they are due directly to an endocrine factor, the nature of the hormone system remains to be defined. The hypothalamic-pituitary-adrenocortical system is a can- didate. The view that hyperadrenocorticism promotes the aging processes has a long history. Finch and Landfield (1985) reviewed the evidence on which this belief is based. A recent provocative idea is the glucocorticoid-cascade hypothesis of aging proposed by Sapolsky et al. (1986b), which was discussed in Chapter 3. Food restriction might influence aging by affecting the control systems that regulate plasma glucose concentrations and glucose homeostasis. Cerami (1985) proposed that glucose is a mediator of aging. The hypothesis was discussed in Chapter 3 and should be explored in regard to the mechanism of action of food restriction. Food restriction markedly retards the aging processes in ro- dents. It does not do so by reducing the input of a nutrient per unit of body weight. It probably acts via neural, endocrine, or neuroendocrine control systems. Almost no work has been done on the nature of the control systems involved. Although food re- striction probably does not act directly on most tissues and organs, it must influence their function via endocrine and neural signals in such a way as to retard the aging processes; the nature of the functional modification is not known. Support is accumulating for Lindell's (1982) view that food re- striction acts by maintaining gene expression. He postulated that enhanced gene expression plays an important role in maintaining cellular homeostasis. Richardson and Cheung (1982) expanded that view by stressing the importance of maintaining protein turnover for adequate cellular homeostasis. Richardson's group (Birchenall-Sparks et al., 1985; Ricketts et al., 1985) has provided evidence that food restriction retards the age-related decrease in the rate of protein synthesis. Lewis et al. (1985) showed that food restriction increased the turnover of body proteins during most of the life span. Surprisingly, the results of the work (Merry and
ENVIRONMENT' AGING, AND THE AGED 87 Holehan, 1985) on the RNA:DNA ratio and the protein:RNA ratio did not seem consistent with the turnover findings. Unlike the situation with restriction of all food elements or all macror~utrients, there is no conclusive evidence that increasing or decreasing the intake of individual specific nutrients (without changing caloric intake) influences the aging processes. This find- ing might be due not to the absence of such actions, but rather to the inadequacy of the research to date. Research in this field has been both scant and usually flawed. The major flaws have been that the effect of the dietary manipulation under study on life span was not determined and that the influence of dietary manipulation on total food and total caloric intake was not adequately assessed. Dietary Protein Intake Results of some studies (Goodrick, 1978; Leto et al., 1976) have indicated that decreasing the intake of dietary protein in- creases longevity. Results of other studies (Nakagawa and Masana, 1971; Ross and Bras, 1973) have not supported that view. The discrepancy might be related to a failure to measure food intake carefully or to report measurements fully. Davis et al. (1983) stud- ied food-restricted rats that were given diets with varied protein content and concluded that decreasing the dietary protein content decreased longevity. But Yu et al. (1985) found that decreasing the casein content of the diet of male Fischer 344 rats fed ad libi- tum from 21~o to 1.2.6% caused a small but significant increase in both life expectancy and life span. Caloric intake was the same in both groups of rats fed ad libitum. The rats on the protein-restricted diet had less severe chronic nephropathy than rats fed the 21% protein diet (Maeda et al., 1985), and that is probably why the rats fed the 12.6~o protein diet had increased longevity. The rats with the lower protein intake also had less severe cardiomyopathy and a lower incidence of calcification and degeneration of the skeletal muscle; but incidence of neoplastic disease was not influenced. Replacing the casein in the diet with soy protein, without reducing dietary protein content or changing caloric intake, also increased life expectancy and life span of male Fischer 344 rats (Twasaki et al., 1986~. The major reason for the increase in longevity was thought to be retardation in the age-related progression in severity of chronic nephropathy. It has been found that tryptophan-deficient diets increase
88 AGING IN TODAY'S ENVIRONMENT longevity in rats (Segall, 1979~. But they also reduce food intake, and that reduction rather than the tryptophan deficiency itself might be responsible for the increase in longevity Dietary Fat Intake Early studies showed that increasing dietary fat content de- creased the life span (French et al., 1953; Silberberg ant! Silber- berg, 1955~. High fat contents have also been found to result in the appearance of tumors at earlier ages and to increase the in- cidence of some kinds of tumors (Carroll, 1975; CIayson, 1975; Reddy et al., 1976~. Diets high in fat also accelerate the aging of collagen (Everitt et al., 1981; Hruza and Chvapil, 1962), decrease cell-mediated immunity, and promote autoimmune disease in NZB mice (Fernandes et al., 1972, 1973~. The issue not fully addressed in these studies is the extent to which the findings were due to fat itself, rather than to a change in caloric intake. Dietary Carbohydrate Intake Little research has been done on the effects of dietary carbohy- drates on aging processes. Durand et aI. (1968) fed male rats diets containing carbohydrate at 39~o; sucrose, glucose, or corn starch was the source of carbohydrate. The life expectancy of Wistar rats was not influenced by the source of carbohydrate, but the life expectancy of BHE rats was reduced when sucrose was the source. Life span was not reported. Dalderup and Visser (1969) reported that adding sucrose to a complete diet reduced life expectancy of male Wistar rats. They also found that the development of chronic nephropathy was accel- erated in the sucrose-fed rats. Data on life span were not reported. Recent studies by Shafrir and Adler (1984) showed that a diet containing sucrose reduces the length of life, in comparison with the same diet in which starch replaces sucrose. The spiny mouse was the animal model. The results of studies on the effects of sucrose are provocative. However, none generated detailed survival curves. The effects of sucrose in particular and carbohydrates in general on aging processes deserve further study.
ENYIRONAfENT, A GINO, AND THE A GED Dietary Vitamin Intake 89 Kokkonen and Barrows (1985) reported the results of a study in which male C57BL/63 mice were fed ad libitum diets that contained either the amount of vitamins recommended by the Na- tional Research Council, half the amount, or 4 times the amount. Life expectancy was significantly reduced when the dietary vita- min content was half that recommended, but not affected by the increase in dietary vitamin content. The interpretation of these re- sults is difficult because food intake knight not have been constant, and deficiency or malnutrition might have been the fundamental problem. The complexity of the issue is emphasized by the report of Kayser et al. (1972) that the life span of rats is lengthened by vitamin restriction. Porta et al. (1980) reported that high dietary vitamin E pro- Tonged the life of rats fed diets that were high in unsaturated fat, but the finding is difficult to interpret because food intake was decreased in rats so treated. Antioxidants, including vitamins, have often been found to increase life expectancy, but not life span, in roclents (Herman, 1978~. Moreover, the role of food in- take in increasing life expectancy has usually not been adequately addressed. Dietary Mineral Intake The effects of reducing mineral intake to the same extent as in life-prolonging food restriction have been investigated in male Fischer 344 rats. The results can be briefly summarized as follows. Restriction of mineral intake did not influence life expectancy or life span. The restriction of sodium, chloride, calcium, and phosphorus to the same extent (40~o) that they are restricted in a life-prolonging, food-restriction regimen did not influence life expectancy or life span of male Fischer 344 rats (Yu et al., 1982, 1985~. A marked reduction in the zinc intake of NZB and NZB/W mice, which normally die from autoimmune disease, retarded the development of autoimmune disease and increased longevity (Beach et al., 1981, 1982~. However, the increased longevity prom ably resulted from control of the disease and not from control of the aging processes. The relevance of these findings to the aging of most mice strains and other animals seems doubtful, but they
go AGING IN TODAY'S ENVIRONMENT do point to a need to explore each mineral element in more than one animal mode! in relation to aging processes. Special Dietary Requirements of the Aged Are the nutritional requirements of the American elderly met by the diets they consume? To answer this question, it is necessary to have a reference, or standard, that defines the nutritional needs of the elderly. The commonly used reference in the United States is the Recommended Dietary Allowances published periodically by the Committee on Dietary Allowances of the National Research Council's Food and Nutrition Board. The most recent revision was published in 1980 (National Research Council, 1980a). Unfor- tunately, there is a lack of data on the dietary needs of the elderly. Although seven age ranges from birth to 22 years are considered, only two include people over 22 specifically, 23-50 years and 50 and older. That people 50-60 years old have the same dietary requirements as those over 70 is questionable. The question is recognized in the recommended dietary allowances (RDAs) only for energy allowance, which is lower for people 76 and older. In addition to the lack of an adequate reference, many method- ologic problems and errors have occurred in the study of nutritional requirements and status of the elderly (Gerry and Hunt, 1986~. They include the use of unstandardized methods, errors in the estimation of nutrient consumption, errors in the analytic values published in food tables, and errors in assumptions about the bioavailability of nutrients in food sources. To circumvent these many pitfalls, at least in part, nutritional surveillance of the elderly should include assessment of nutrient concentrations in tissues, especially plasma and blood cells. Such analyses present other interpretational problems, of which the pri- mary one is that the relation of tissue concentrations of nutrients to functional impairment has not been well established, particularly if age is a factor in the consideration. A further problem in regard to the aged is that many have diseases that can influence nutritional needs and status (Flynn, 1984~. Although clinicians must deal with all aspects of nutri- tional requirements of aged patients, including those related to disease, the logical development of our understanding of the nu- tritional requirements of the aged must first focus on age alone
ENVIRONMENT, AGING, AND THE AGED 91 (i.e., age in the absence of disease). Similarly, many of the el- derly might not receive appropriate nutrients because of economic circumstances. Although knowledge of this problem is necessary to permit corrective action, it should not be confused with the nutritional status of the healthy elderly who are not constrained by economic factors, that is, it should not be confused with the . . · · - baSlC 310 Ogle lSSUe. What information do we have on the nutritional requirements and needs of elderly people who are free of disease and have adequate economic resources? Carry et al. (1982a) addressed this issue. The subjects were 138 men and 166 women over 60 years old, citizens of Albuquerque, New Mexico, free of major illnesses, and receiving no prescription medication. Most of the subjects were at the middle income level. With the possible exception of a few nutrients, the dietary intake of most subjects appeared to be adequate. Intakes of vitamins B6, Bit, D, and E, folic acid, calcium, and zinc might have been inadequate, inasmuch as one- fourth of the subjects received less than 75~o of the RDAs. Indeed, one-fourth of the subjects had vitamins B6, D, and E and folic acid intakes of less than 50% of the RDAs; in the case of women, that was also true of zinc. Although the intake of those substances was less than desir- able, there is no evidence that the subjects suffered deficiency problems. For example, in the case of vitamin BY and folic acid, none of the subjects had megaloblastic anemia. The low intake of vitamin B6 and folate knight not be real, in that for many foods reliable data on content are lacking. Clearly, much more extensive functional testing ~ required before intakes can be assessed rela- tive to RDAs with respect to the existence of nutritional adequacy. Not only was the protein intake of most subjects greater than the RDA, but 70~o of the protein intake was from animal sources. The influence of protein from different sources, such as vegetables, is not known. Calcium and vitamin D intakes were below the RDAs for most women; they were related to the low consumption of dairy products. The next step was to learn whether dietary intake, which for the most part seemed adequate, resulted in good nutritional sta- tus in the elderly subjects. In regard to vitamin C (Gerry et al., 1982c), 95.9~o of the population had a plasma ascorbic acid con- centration of over 0.4%0 mg/~l and thus were felt to be at low risk of developing clinical symptoms of hypovitaminosis C. However,
92 AGING IN TODAY'S ENVIRON~NT 2.2~o of the subjects had concentrations of 0.2-0.4 mg/~l and thus were at moderate risk, and l.9~o had concentrations less than 0.2 mg/~l and were at high risk. With an erythrocyte glutathione reductase activity coefficient of over 1.35 as an index of high risk of developing clinical symptoms of hypovitarninosis B2, only loo of the subjects were at risk (Gerry et al., 1982b). To consider the vita- min D status of the subjects further, plasma 25-hydroxyvitamin D concentration was measured, as was plasma alkaline phosphatase activity, the latter as an index of bone loss (Omdah} et al., 1982~. The mean plasma 25-hyciroxy~itamin D concentration of the el- derly subjects was 15.5 ng/mI, compared with 29.1 ng/m} in a younger control population. Within the elderly subjects, plasma 2~hydroxyvitamin D concentration showed a nadir in January and a zenith in September and was higher in men than women. Plasma alkaline phosphatase activity was inversely related to the plasma 25-hydroxyvitamin D concentration, which might mean that the elderly need moderate vitamin D supplementation and more sunlight. Further information of this type is needed to define more fully the nutritional status of these subjects and of other populations of healthy elderly people, with regard to the spectrum of nutrients. In particular, more clinical work should be carried out on such people to determine if there are signs of nutritional deficiencies. The effects of disease must ultimately be considered because there is a high prevalence of disease in the elderly. Indeed, disease in the elderly might result from or be the cause of malnutrition. An obvious example is the extent to which dental problems can limit food choices and eating practices. Thus, the major disease problems encountered in the aged should be assessed with regard to their nutritional aspects. Osteoporosis is common in the elderly, particularly in women. A number of factors appear to contribute to this problem, and dietary calcium intake could be one (Heany et al., 1982~. It has been suggested that postmenopausal women should have a calcium intake of about twice the RDA to prevent loss of bone mass. In fact, however, the calcium intake of most women over 44 appears to be only about 60~o of the RDA. Vitamin D intake appears to be well below the RDA; but it should be noted that the efficacy of (lietary calcium in retarding osteoporosis is complex and in debate (Riggs and Melton, 1986~. The role of the environment in
ENVIRONMENT, AGING, AND THE AGED 93 bone metabolism and Vitamin D nutrition is discussed further in Chapter 7. A major problem that appears in many at advanced ages is a decline in cognitive ability. There is suggestive evidence that at least in some people this might be related to inadequate vitamin nutrition (Goodwin et al., 1983~. However, much more work is needed to define the relationship between cognitive loss and nutri- tion and to determine the possible therapeutic use of nutrition. The immune system, including immune responsiveness and autoimmunity, deteriorates with age. These changes could be involved in many aspects of disease in the elderly (Hausman and Weksler, 1985~. In rodents, nutrition has been shown to modulate age-related changes in the immune system (Masoro, 1985~. No data show clearly that nutrition has such an action in humans, but it is a subject with great potential for the development of beneficial intervention for the aged, and it should be a major research focus of the future. There is evidence that the pathogenesis of age-related involu- tional changes of the kidney can be modulated by diet (Brenner et al., 1982~. High dietary protein accelerates progression of the lesions, and low-protein diets retard the process. Atherosclerosis is an age-related disease that has been studied extensively with regard to the role of nutrition (Bierman, 1985) a role that is still the subject of debate. Hypertension is an age-related problem that appears in part to be associated with nutrition. The most obvious association is with obesity, which is a risk factor in the development of hy- pertension. Weight reduction can lower arterial blood pressure in some hypertensive patients (Kennel, 1985~. Excessive sodium intake has long been linked to hypertension, but recent evidence has indicated that the extent of its role might have been overgen- eralized, inasmuch as only some hypertensive patients respond to changes in dietary sodium intake (Laragh and Pecker, 1983~. Re- cent findings have suggested that inadequate calcium intake might result in hypertension (McCarron, 1985~; although controversial, this association might be particularly important in relation to the aged because many have low dietary calcium intake. Adult-onset diabetes occurs increasingly with advancing age (Silverberg, 1984~. Obesity appears to play a role in its pathogen- esis, and weight reduction often helps to control it. Although cancer occurs in people of all ages, the incidence of
94 AGING IN TODAY'S ENVIRONMENT many types of cancer (e.g., cancer of the colon) is age-related (U~ ton, 1977~. There is some evidence that diets can prevent or delay the occurrence of cancer, but many more data are needed before firm conclusions can be drawn (NRC, 1982~. Cancer often causes malnutrition. Prevention of or treatment of this malnutrition can improve the quality of life of the cancer patient (Flynn, 1984~. The elderly consume many more drugs than the young (Vestal and Dawson, 1985~. Drugs affect nutrition by influencing di- etary intake, metabolism, and rates of elimination of nutrients (Roe, 1985~. In addition, the nutritional status of an elderly person can influence both the efficacy and the toxicity of drugs (Welling, 1985~. Thus, there should be increased consideration of drug-nutrient interactions by both medical practitioners and researchers. Many healthy old people with sufficient income have an ade- quate intake of most nutrients. However, major exceptions appear to be calcium and vitamin D (see Chapter 6 for further discus- sion). Lack of sufficient economic resources might well result in malnutrition in the elderly, but that would be true of any age group. An added problem with the aged is the high prevalence of a variety of diseases, many of which can be caused in part or influenced by nutrition and can also adversely affect nutritional status. The aged with such diseases need nutritional programs designed specifically for their particular disease processes. Many claims have been made about diets and dietary regimens in regard to life extension (Schneider and Reed, 1985~. A recent version (Walford, 1986) was based on the food-restriction data base for rodents coupled with an abundant intake of essential nutrients and other substances having some evidence of antiaging action. It should be underscored, however, that there is very little evidence that such diets influence the aging processes of humans. Moreover, there is no evidence that their long-term use will not adversely affect the health of humans. PHARMACEUTICALS Demographic Considerations The proportions of elderly people in developed countries have been rising steadily over the last several decades because of falling birth rates and medical, economic, and social factors that favor
ENVIRO~ENT, AGING, AND THE AGED 95 prolonged life. People over 65 years old make up 12~o of the Amer- ican population (more than 23 million people) and spend 2~25% of the national total for drugs and drug sundries, or about $3 bil- lion per year. It is projected that by the year 2030 that age group will contain more than 64 million people and will constitute 21~o or more of the population. Thus, by the year 2030 expenditures for drugs by the elderly in the United States might constitute 35- 45~o of the national total (Vestal, 1985~. In the United Kingdom, where the elderly make up 12~o of the population, they are al- reacly responsible for approximately 30% of expenditures for drug prescriptions (O'Malley et al., 1980~. Clearly, the needs of geri- atric patients will constitute an increasingly important aspect of medical care in the future. Patterns of Drug Use and Drug Prescribing Epidemiologic data are difficult to compare, not only within the United States, but also within and between other countries. The Nordic countries have used the "defined daily dose" (000) to lessen the difficulty of comparison (Bergman et al., 1980~. This method might be used in future studies in other countries as well. Nevertheless, as reviewed by Nolan and O'Malley (1987a), elderly patients receive more drugs than younger patients. Ambulatory Populations In a study of an ambulatory community-dwelling population in Albany, New York, Chien et al. (1978) found that the most commonly used medications were analgesics (used by 67~o of the population), cardiovascular preparations (34%), laxatives (31%), vitamins (29~o), antacids (26~o), and antianxiety agents (22%~. Of the 244 persons over 60 who were studied, 83%0 were taking two or more medications. Over-the-counter (OTC) preparations accounted for 40~o of the drug purchases; 60~o were prescribed. In a similar study in Washington, D.C., Guttmann (1978) found that the most frequently used classes of medications were cardiovascular preparations (used by 61%), sedatives and tranquil- izers (17%), antiarthritis preparations (12%), and gastrointestinal preparations (11%~. OTC drugs were used by 59~o of this sample of 447 subjects, 52~o using analgesics. Vitarn~ns and laxatives were used by To and Who, respectively, of this group.
96 AGING IN TODAY'S ENVIRONMENT A recent study in Seattle confirmed the extensive use of drugs by the elderly. Interviews of 183 independently living elderly res- idents of two urban high-rise apartment buildings revealed that 75~o used at least one prescription drug regularly and 82~o used at least one nonprescription drug regularly (Ostrum et al., 1985~. Studies in the United Kingdom have also shown greater drug use in the ambulatory elderly than in the young (Murdoch, 1980; Skegg et al., 1977~. The proportion of elderly in a general practice setting taking at least one prescription drug has ranged from 33~o in studies of patients over 65 to 80~o of males and 85% of females over 75 (Freer, 1985a; Murdoch, 1980~. Most studies have indicated that 50-75%0 of the elderly take at least one prescribed drug (Freer, 1985b; Murdoch, 1980; Shaw and Opit, 1976; Skegg et al., 1977; Tulloch, 1981~. Reports from New Zealand (Campbell et al., 1983), Canada (Skoll et al., 1979), Sweden (Boethius, 1977), and Spain (Mas et al., 1983) have contained data that confirm that drug use among the elderly is similar in many other developed countries. Because of the use of OTC preparations on an as-needed basis, different ways of classifying drugs, and different kinds of patient selection, the data are difficult to compare. However, nonnarcotic analgesics appear to be the most commonly consumed nonpre- scribed drugs, and cardiovascular preparations and psychoactive substances the most commonly prescribed drugs. Older patients are probably not more avid consumers of OTC drugs, however, than other segments of the population (Bush and Rabin, 1976~. Hospital Populations It is not surprising that older hospitalized patients receive more drugs than younger ones. That has been documented by studies in the United States (Borda et al., 1967), the United King- dom (Houston, 1979), and Finland (Sotaniemi and Palva, 1972~. However, it is a little surprising how different prescribing practices in different countries can be. A study by the Boston Collaborative Drug Surveillance Program showed that American hospital inpa- tients received an average of 9.1 different preparations, compared with 7.1 for patients in Canada, 6.3 for Israel, 6.8 for New Zealand, and 4.6 for Scotland. A detailed comparison between Scotland and the United States showed that patients in the United States were treated
ENVIRONMENT, A GINO, AND THE A GED 97 more intensively for diarrhea, dehydration, constipation, diabetes, and hypertension. Anxiety, pain, congestive heart failure, and anemia were treated similarly in the two countries. Differences in prescription patterns were especially obvious in the treatment of infections. Three antibiotics in Scotland and 10 in the United States were used to treat 75% of infections (Lawson and Jick, 1976~. Specific data on age were not included in the study, but it seems likely that the findings would apply to geriatric patients. Why prescribing habits vary so widely among countries, and even among regions of the same country, is not fully understood. The explanation undoubtedly lies in the approach to drug therapy taught in undergraduate and postgraduate medical institutions. Long-Term-Care Facilities Patients in chronic-care institutions commonly receive many drugs, tranquilizers and hypnotic-sedatives being very common (Institute of Medicine and National Research Council, 1985~. Once again, the comparison between the United States and the United Kingdom is interesting. Estimates of drug use by English nursing home residents are similar to those for elderly patients living in the community and range from 1.5 to 3.1 drugs per patient (Bruce, 1982; Clarke et al., 1981~. In contrast, an average of seven drugs are prescribed concurrently for American patients (Bergman et al., 1980; Rawlings and Frisk, 1975; Segal et al., 1979~. In both the United Kingdom and the United States, most institutionalized patients over 65 take at least one prescribed drug. Patients in a Tong-term-care facility in Boston received a mean of eight drugs in the first 10 days (Borda et al., 1967~. In an- other survey (Kalchthaler et al., 1977), psychotropic drugs were most commonly prescribed (61~o of patients), followed by diuretic and antihypertensive drugs (46%), antimicrobials (14°~7o), and car- diotonics (who). The contribution of consultant pharmacists to rational drug use in long-term-care facilities has been emphasized by Cooper and Bagwell (1978), who showed that over a period of a year the use of scheduled drugs was reduced by 19.4% and the average number of drugs per patient was reduced from 7.2 to 4.8. Psychoactive drugs might be prescribed more often to patients who are mentally normal and have minimal physical disabilities than to those who are more severely disabled (Ingman et al., 1975~. In a study whose results suggested misuse of antipsychotic drugs
98 AGING IN TODAY'S ENVIRONMENT in nursing homes, each resident was matched with an ambulatory person enrolled in Medicaid (Ray et al., 1980~. Among nursing- home patients, central nervous system (CNS) drugs were the most often prescribed medications, being given to 74% of patients; only 36~o of the ambulatory comparison group received CNS drugs. Nursing-home patients often received prescriptions from mul- tiple categories of CNS drugs: 34~o from two or more categories, To from three or more, and 1.6%0 from four. The most com- mon combinations were those of an antipsychotic and a sedative- hypnotic, most often thioridazine and flurazepam. The next most common combinations were those of a minor tranquilizer and a sedative-hypnotic, usually diazepam and chIoral hydrate. The three most commonly prescribed antipsychotic drugs were thioridazine, chlorpromazine, and haloperidol. The authors sug- gested that those drugs might be used to mold patients into the institutional routine. Medication Compliance in the Elderly Many patients have difficulty in taking medications as pre- scribed by their physicians. Blackwell (1972) reviewed over 50 studies and found that 25-50~o of all outpatients completely failed to take their medication. On the basis of careful review of drug histories in 178 chroni- cally ill ambulatory patients aged 60 or over in the General Med- ical Clinic at New York Hospital, Schwartz et al. (1962) found that 59~o made one or more medication errors and 26~o made potentially serious errors. Error-prone patients were more likely to make multiple than single mistakes. The average number of errors-was 2.6 per error-making patient. Omission of medication was the most frequent error, followed by lack of knowledge about medications, use of medications not prescribed by a physician, and errors in dosage, sequence, or timing. Almost identical data were obtained using similar techniques in a Seattle-area clinic (Neely and Patrick, 1968~. In a study of geriatric patients 10 days after hospitalization, Parkin et al. (1976) found that 66 of 130 patients deviated from the drug regimens prescribed at discharge. Noncomprehension or lack of a clear understanding of a regimen (in 46 patients) was actually a greater problem than noncompliance or failure to follow instruction (in 20 patients).
ENVIRONMENT, AGING, AND 1lHE AGED 99 Although medication errors seem to be prevalent among el- derly patients, studies that used objective measures of compliance have indicated that the elderly are not necessarily more prone to noncompliance than younger patients. That has been demon- strated in patients taking digoxin (Weintraub et al., 1973~. A recent example is the feasibility study for the multicenter clinical trials in the United States called the Systolic Hypertension in the Elderly Program (SHEP), which was conducted between 1980 and 1984. A final cohort of 551 persons over 65 years old with isolated systolic hypertension was enrolled. Compliance with therapy was evaluated by self-reporting, pill counts, and urinary assay for the presence of study drugs. More than 80~o of partici- pants in both treatment and placebo groups complied with their SHEP prescriptions (Smith et al., 1985~. Researchers must also allow for intelligent noncompliance by patients who appropriately alter their drug regimen to minimize side effects or to eliminate unnecessary treatment (Weintraub, 1984~. Additional research is needed to clarify the clinical impor- tance of such noncompliance in older patients. Adverse Drllg Reactions Adverse drug reactions might be the inevitable price of im- proved drug therapy for disease (Barr, 1955; Jick, 1974~. The possibility that these reactions are costly in terms of human illness and economics (Campbell et al., 1977; Melmon, 1971) is a mat- ter of controversy. It has been stated that adverse drug reactions constitute a formidable health problem with staggering economic consequences that one-seventh of all hospital days are devoted to the care of drug toxicity, at an estimated yearly cost of $3 billion (Melmon, 1971~. In support of that contention, previously cited studies from several countries indicate that 3-5~o of all hospital admissions are primarily for a drug reaction; that 1~30% of all hospitalized patients have an adverse drug reaction, doubling the duration of hospitalization; and that 30~o of these patients have a second reaction during their hospital stay. The lay press has asserted that 30,000 Americans die as a direct result of the use of drugs prescribed by doctors (Rensberger, 19763. In contrast, it has been suggested that drugs are remarkably nontoxic, considering their extensive use (Jick, 1974~. According
100 AGING IN TODAY'S ENVIRON to data from the Boston Collaborative Drug Surveillance Program, death from drugs occurs in less than 0.3~o of hospitalized medical patients in the United States, and for each course of drug ther- apy the frequency of adverse drug reactions IS only Who, or one in 20 treatments. Furthermore, although associated with discom- fort, most drug reactions such as nausea, drowsiness, diarrhea, vomiting, and rash are transient and of minor consequence to the patient. Others have rejected both positions, arguing that estunates of the magnitude of the problem of adverse drug reac- tions are characterized by incomplete, arbitrary, and uncontrolled data and by unreliable and inaccurate cost estimates (Karch and Lasagna, 1975). In the context of unresolved controversy, the issue of adverse reactions to drugs in the elderly must be considered. The extent to which age itself is a risk factor for untoward therapeutic events is uncertain and has been the subject of several critical reviews (Gardner and Club, 1970; Klein et al., 1981; Nolan and O'Malley, 1987b; Vestal et at., 1985). Not all the studies agree, but many have demonstrated a higher incidence of adverse drug reactions in hospitalized geriatric patients. In studies in which an age-related increase was observed, it ranged from 1.6-fold to 5-fold. Patients over 70 appear to be particularly vulnerable, with an incidence of about Who, compared with 3-10~o in patients under 30 (Hurwitz, 1969; Seid! et al., 19663. Studies performed in the United States (SeidI et al., 1966), North- ern Ireland (Hurwitz, 1969), New Zealand (Smidt and McQueen, 1972), Switzerland (Klein et al., 1976), and Israel (Levy et al., 1977) have consistently documented an association between age and an increased incidence of adverse drug reactions. The data on outpatients are limited. Two studies demon- strated an increase with age (Kellaway and McCrae, 1973; LumIey et al., 1986) ~d two demonstrated no effect of age (Klein et al., 1984; Muiroy, 1973~. In one of the studies that did not show an age effect, the results suggested that the elderly might be unable to distinguish drug side effects from symptoms of "old ages and unable to communicate adverse effects to their providers (Klein et al., 1984~. Limitations of the data on adverse drug reactions in the elderly have been enumerated (Klein et al., 1981; Nolan and O'Malley, 1987b; Vestal et al., 1985~. Most studies have involved hospitalized patients. True populations at risk, which would include patients
ENVIRONMENT, AGING, AND THE AGED 101 talking drugs in the community as well as patients in the hospital, frequently have not been studied. Patients have not been strat- ified according to severity of illness or concurrent drug therapy. Unmedicated control groups have not been included. Despite the lirn~tations, the available data indicate that sev- eral risk factors are associated with adverse drug reactions, in- cluding age, sex, drug dosage, previous drug reaction, impairment in hepatic or renal function, length of hospital stay, and multiple- drug therapy (Vestal et al., 1985~. In addition, age differences in physiology and pharmacology might help to explain the apparent propensity of the elderly to suffer adverse reactions to drugs. LIFESTYLE Deliberate Chemical Exposure Life-styTe is important in determining the types and quantities of chern~cal agents (other than pharmaceuticals) that are deliber- ately brought into contact with the human body. Some of the agents reflect cultural trends, others are more individual. The chemicals and their uses are diverse. Specialized com- ponents of the diet have become popular in the United States, although scientific data to justify their use are generally lack- ing. Some people are convinced of the merits of an exclusively or predominantly vegetarian diet; others use supplements that in- clude pure preparations of vitamins, amino acids, minerals, and other compounds, as well as many complex chemical mixtures represented by "unusual" plant and other materials or extracts. Whether such materials have any effect on the aging processes, ei- ther adverse or beneficial, generally has not been tested. Similarly, their toxic effects are usually unstudied. Another important source of deliberate human exposure to chemical agents comes from the use of various types of fragrances and cosmetic agents. These are often complex chemical mixtures whose major ant] minor components are untested beyond acute toxic potential. Studies involving repeated application to the skin of experimental animals have sometimes revealed dermal and neu- rologic damage. Deliberate chemical exposure of human subjects occurs in a number of other settings, many of which are influenced by individ- ual preferences and life-styI~for example, use of mouthwash and
102 AGING IN TODAY'S ENVIRONMENT oral deodorants, antiseborrheic shampoos, sunscreen and suntan lotions, and insect repellents. [earned Heiplessness Aging is associated with a Toss of control over homeostatic responses, mobility, and cognitive function, for example. In physi- ologic terms, the loss of control leads to a narrowing of homeostatic reserve and, in behavioral terms, to helplessness. The tendency of fondly and friends, as well as members of the health professions, to urge accommodation rather than resistance to progressive incur- sions on functional capacity leads to "learned helplessness." The acceptance of helplessness leads not only to greater helplessness, but also to increased susceptibility to disease. A dramatic example of the relationship between learned help- lessness and disease occurs in rats subjected to the unyoked electric- shock paradigm of Weiss (Sklar and Anisman, 1979; Visinta~ner et al., 1982~. Two rats are connected (yoked) in series to a random electric-shock generating system. Both rats have levers in their cages, but only one lever functions as a switch for the electric cir- cuit. When the functional switch is pressed, the electric shock to both rats is turned off. However, only the rat with the functional switch is in control; the other is subjected to "learned helpless- ness.~ A third rat is placed in the cage, but Is not subjected to electric shock. The effect of these environmental manipulations on susceptibility to disease is striking. For example, if the dose of cancer ceils that kills 50% of the rats not subjected to electric shock is given to rats that have learned helplessness, 80~o of the animals die. The same dose of cancer cells kills only 20%0 of the rats in control. It seems reasonable to conclude that some of the increased intrinsic susceptibility to disease that is manifested in the elderly could be related to environmental influences that result in learned helplessness among physiologically comprorn~sed and behaviorally impaired people. Atrophy of Disuse Life-style can inevitably affect aging processes and the aged
ENVIRONMENT, AGING, AND 17IE AGED 103 through occupation, recreational activities, exercise habits, eat- ing and drinking habits, tobacco smoking, exposure to stress- almost any behavior that creates contact between the person and the environment. Elements of life-style can increase or decrease risks of acquiring age-related degenerative diseases. They can also conceivably accelerate or delay physiologic and anatomic changes associated with the passage of years. Clear examples are the va- riety of age-related- diseases induced by toxic chern~cals in tobacco smoke and the decrease in risk of cardiovascular disease produced by regular exercise. Atrophy of disuse is a consequence of life-style and therefore belongs among factors responsible for extrinsic aging. Decline in the vigor, health, and well-being of the aging and the aged is tacitly assumed to be due to intrinsic aging, but these unfavorable changes result largely from atrophy of disuse. The debilitated state resulting from atrophy of disuse conceivably contributes to susceptibility to age-related diseases; however, such effects are difficult to dissect from the general complex of events that make up intrinsic and extrinsic aging. In a study of some 17,000 middle-aged and older Harvard alumni over several years, Paffenbarger et al. (1984) found a 49~o excess risk of coronary heart disease among people who led seden- ta~y life-styles (Butler and Lewis, 1986~. Spoor Pollutants The indoor environment is a subject of growing concern. Dur- ing the last decade, health experts and regulatory officials have rec- ognized that indoor exposures can have a greater effect on human health and comfort than outdoor pollutants. Americans spend up to 80~o of their time indoors in homes, in public buildings, in offices or other places of work, or in various modes of trans- portation. Those who are most susceptible to the health effects of pollutionthe very old, the very young, and the chronically ill might spend even more of their time indoors. No federal standards for indoor airapart from industrial workplace standards exist. Yet the concentrations of some pol- lutants in office buildings, homes, and public places are known to exceed their primary ambient-air quality standards. Modern energy-conservation measures that reduce ventilation can cause "sick buildings" and the related health effects.
104 AGING IN TODAY'S ENVIRONMENT Indoor pollutants of concern identified by the National Re- search Council (1981a) and in other reports include radon and its decay products, asbestos, tobacco smoke, formaldehyde and other organic compounds, the gases from combustion (principally nitro- gen dioxide and carbon monoxide), and microorganisms and aller- gens. There is little epidemiologic information on the health effects of chronic, low-level exposure to any of these pollutants except to- bacco smoke (National Research Council, 1986~. What data exist have been derived principally from occupational exposuresto uranium, pesticides, and asbestos, for example and most involve acute exposures. For some pollutants, human discomfort provides a useful in- dicator of environmental contamination. Discomfort gives imme- diate incentive to avoid or correct environmental deficiencies and can lead to closer investigation of their source. Unfortunately, the populations that are most susceptible to the effects of pollution knight be the least able to do anything about it. The elderly com- monly have diminished sensory perception and are less able to detect contaminants in their environment. People who seldom go outside into fresh air or who are chronically exposed to carbon dioxide or carbon monoxide might also have diminished sensory capacity. Temperature Conditions for thermal comfort appear to vary little, if at all, with such factors as geographic location, sex, ethnic background, and even age (ranger, 1972~. Aging merits some special con- sideration: the basal metabolic rate decreases progressively, but evaporative heat loss does also. These two changes tend to offset each other to a limited extent that is difficult to assess, because the elderly spend much more time than the young in sedentary activities. Furthermore, with the energy conservation practices now common indoors during winter, the elderly have a narrower temperature range, which limits their thermal resistance (Sacher, 1979~. Sensory adaptations in sedentary older persons might be se- verely diminished, and a person could fail to notice the symptoms of impending hypothermia until they became severe.
ENVIRONMENT, AGING, AND ME AGED Radon 105 The radioactive gas raclon and its decay products have drawn considerable attention in recent years after discoveries of very high indoor concentrations in some parts of the country. Exposure to high concentrations of radon and its progeny causes lung cancer. Most of the data on the health effects of radon progeny come from studies of underground miners whose occupational exposures were 10~1,000 times those of the general population. However, homes in some locales have recorded radon concentrations higher than those in mines. Data on Canadian uranium miners indicate an excess of lung cancers even in the group with the lowest cumu- lative dose a dose that would be in the same range as the lifetime cumulative close from many home exposures. A few epidemiologic studies of nonoccupational radon expo- sure have been done, but most have failed to control for smoking. However, it appears that both active smoking and passive smok- ing increase susceptibility to the effects of radon, because radon progeny adhere to respirable particulate matter in tobacco smoke. Nero et al. (1986) recently reviewed the associated risks. Formaldehyde There are many sources of formaldehyde exposure in the in- door environment. Particleboard, paneling, plywood, and urea- formaldehyde foam insulation can be important sources of va- pors. Because of their extensive use of particleboard, with its high surface-to-volume ratio, mobile homes have been found to have much higher concentrations of formaldehyde than conventional homes. In addition, cosmetic products, especially shampoos, are a source of skin exposure, and cigarette smoke contains 10-15 mg of formaldehyde per cigarette. Formaldehyde, a contaminant of concern to all populations, might be particularly important in the health and comfort of el- derly people and people with chronic respiratory problems. The present Occupational Safety and Health Administration standard for formaldehyde is 3 ppm. However, the National Institute for Occupational Safety and Health has recommended that occupa- tional exposure be reduced to the lowest feasible, to minimize the risk of cancer (U.S. NIOSH, 1981~. Potential carcinogenic- ity has been reviewed by the International Agency for Research
106 AGING IN TODAY'S ENVIRONMENT on Cancer (1982~. In addition, people have reported mild ear, nose, and throat discomfort and other symptoms at led than 0.5 ppm. Between 10 and 20% of the general population might be susceptible to the effects of formaldehyde and react acutely at any concentration. The principal effect of formaldehyde at low concentrations is irritation of the eyes and mucous membranes. Human eyes are very sensitive to airborne formaldehyde; they can be irritated by atmospheric concentrations of 0.0~0.5 ppm and in some cases respond to 0.01 ppm when formaldehyde ~ mixed with other pol- lutants. At relatively low concentrations (0.1 ppm), it can irritate upper airways; it more frequently does so at 1-11 ppm (NRC, 1981a). In some susceptible persons, an allergic reaction might occur at very low concentrations, causing bronchoconstriction and asthmatic symptoms. Combustion Products The two indoor combustion products that most often cause concern about health effects on those exposed are nitrogen dioxide (NO2) and carbon monoxide (CO). The amount of NO2 formed in unvented indoor combustion devices Is not usually sufficient to cause acute toxicity; however, NO2 concentrations equal to or greater than the current ambient-air quality standard of 0.05 ppm are not unusual in kitchens with gas cooking (NRC, 1981a). In animal studies, NO2 has been shown to produce transient and long- term damage to small bronchial airways and alveolar tissue (WHO, 1977~. In addition, changes in nonciliated cells, destruction of Type ~ epithelial cells, and proliferation of Type IT cells after relatively low exposures (2 ppm for 4 hours) suggest that chronic exposure couIcl lead to chronic bronchitis and the development of emphysema (Evans and Freeman, 1980~. Resistance was weakened in animals chronically exposed to NO2 at low concentrations with bacterial aerosol challenge (Gardner et al., 1979~. A 1981 study by Mostardi et al. compared the health of chil- dren attending two schools, one of which bordered an industry and had comparatively high concentrations of NO2 and sulfur dioxide. The study found a higher incidence of acute respiratory illness and significantly compromised pulmonary function among the children attending the school near the industry. That finding is important
ENVIRONMENT, A GINO, AND LIE A GED 107 because recent evidence has suggested that frequency and sever- ity of acute respiratory illness in childhood are related to chronic obstructive lung disease in adulthood. The second import ant product of combustion, CO, is found in cigarette smoke and fossil-fuel combustion effluents, and thus is found commonly indoors in a broad range of concentrations. The principal health effect of CO is impaired oxygen transport. CO is readily absorbed from inspired air and binds to hemoglobin with over 200 times the affinity of oxygen. Doses are cumulative, so exposure to CO at even Tow concentrations in the air can result in substantial carboxyhemogIobin concentrations in the blood. Al- though healthy people are not greatly affected by exposure to CO at low concentrations, those with limited cardiovascular reserve can be at greater risk because their blood has reduced oxygen- carrying capacity. In addition, CO has been implicated in the pathogenesis of atherosclerosis. Asbestos Asbestos has been widely used in public and private buildings, in insulation and fireproofing materials, in ornamental decoration and soundproofing, and on surfaces In public areas. An extensive program has recently been undertaken to identify and remove it from school buildings. The relationship between exposure to asbestos and Jung can- cer is well known. Asbestosis, mesothelioma, and gastrointestinal tract cancers have been found in excess among workers occupation- ally exposed to asbestos. Less is known about the health effects of exposures to asbestos and other fibrous materials in environments other than the workplace. Environmental Tobacco Smoke Tobacco smoke is a major source of pollution in the indoor environment. Nonsmokers absorb measurable amounts of CO and nicotine and can absorb small amounts of other constituents from environmental tobacco smoke (ETS). ETS contains several other known hazardous pollutants, but few studies have been done on exposure to them by this route. Many of the chemicals identified in cigarette smoke are irritat- ing to the nose, throat, and eyes. Studies of the long-term effects
108 AGING IN TODAY'S ENVIRONMENT of passive smoking have suggested that it can cause small-airway dysfunction, which ~ an early precursor of clinically important chronic obstructive lung disease. Exposure to ETS appears to cause some increase in systolic blood pressure, especially in chil- dren. As mentioned above, carboxyhemogiobin concentration can be increased by exposure to tobacco smoke, which would reduce the maximal exercise capacity in normal adults. Some people are allergic to ETS, although documentation on the numbers of these people is inadequate. Chronic exposure to tobacco smoke apparently can impair immune system function. Laboratory studies on mice have found that chronic tobacco-smoke exposure accelerated many of the im- munologic changes associated with aging, including marked mod- ifications of the responses associated with the T-lymphocyte arm of the immune system and of the systemic clearance mechanisms. Populations that might be more susceptible than others to the effects of ETS include children, people with coronary arterial disease, and people with chronic Jung disease. Patients with com- prom~sed heart or lung function suffer earlier onset of angina or clyspnea, respectively, after exposure to ETS. Pesticides A wide variety of pesticides are available for use in the home and in public and office buildings. They are assumed to be an important source of exposure to known hazardous substances. About 9 of 10 households in the United States use some type of pesticide in the house, garden, or yard; nearly 84~o of this use is in the house. A study conducted in South Carolina (a region of heavy pesticide use) by Kie! et al. (1969) found that one-third of the families that used pesticides applied them during each week of the year. A study by Savage et al. (1981) surveyed 10,000 households in different regions of the United States and found over 500 pesticide formulations being used. Many people did Lot know what pesticide they had used (e.g., Bug sprayed. Interviews with members of 8,254 households yielded the recording of use or storage of 1,756 containers of unknown insecticides.
