Medications play a unique role among the elderly because they can be major contributors to both increased functional capacity and increased disability. When used correctly, medications can be the single most important element of the therapeutic encounter. When used inappropriately, however, medications can generate considerable morbidity and even mortality. Striking changes, among the most dramatic seen in geriatric medicine, can occur in clinical status if needed drugs are provided to a patient who has not been receiving them or if medications with toxic side effects are removed.
The definition of disability related to medication (its presence or absence) presents several methodological problems. It is fairly easy to define some of the more straightforward complications of medication use (e.g., an acute allergic reaction), but these reactions are low on the list of drug side effects in the elderly, in terms of both frequency and clinical importance. Far more challenging is the definition and quantification of more relevant kinds of drug-induced morbidity, including falls and fractures, mental status changes, or, at the extreme end of the continuum, the broad category of functional, cognitive, and affective states that together are known as quality of life outcomes. Defining the disability-preventing actions of medications is equally difficult, which may tend to deter the consideration of these aspects. Both methodologically and conceptually, it is difficult to consider an event that did not occur—for example, a stroke in an older patient whose high blood pressure has been adequately controlled
over several decades. Nevertheless, a balanced look at the role medications play in the health of the elderly requires consideration of such events, even though they may be identifiable only by the traces they leave in morbidity and mortality rates, rather than as observable entities in their own right.
There are important physiologic reasons why the role of medication increases (as does its potential for good or ill effect) throughout the second 50 years of life. First, in this part of the life cycle, illness occurs with increasing frequency, and such illness is often amenable to drug therapy. Less well understood are the ways in which the effects of medications are magnified by the physiology of even normal aging. There is a well-documented decrease in renal function with advancing age, which increases the effect of medications (e.g., digoxin, cimetidine, aminoglycosides) that are excreted primarily through the kidney.39 Although there is a clear age-related decline on average as people age, more recent research has made it clear that there is great interindividual variability in the pace with which such declines occur.28 This finding has important implications for the effect of medications on individual elderly patients: the older the patient, the less able the physician will be to predict the optimal dose of a medication on the basis of clinical judgment and routine laboratory tests alone. An even greater controversy surrounds the role of aging in the impairment of hepatic function, the other major route of elimination of drugs. Certain hepatic metabolic functions appear to diminish with age; they certainly do so in the face of illnesses that are more common in the elderly than in younger patients (e.g., congestive heart failure), thus decreasing the margin of safety for many medications.48
Other changes that occur in the second half of life also have profound implications for drug effects in the elderly. There is an age-related increase in the proportion of body weight that is fat as opposed to muscle, which tends to increase the half-life and steady-state concentration of lipophilic medications (such as the benzodiazepines) but has the reverse effect on polar drugs (such as lithium). There is also an inverse relationship between age and weight in the current American population. However, data indicate that physicians do not correct for such changes in determining the dose of several commonly used medications in the elderly, which results in the prescription of more milligrams per kilogram in elderly patients than in younger ones.15
Furthermore, research on changes in receptor physiology with age suggests that receptors for many commonly used medications may actually become more sensitive with advancing age, thus intensifying
drug effects in the elderly at the very time that, because of the factors described above, the concentration of available drug is higher.42
A large number of studies, comprising thousands of patients in several countries, have shown that adverse drug reactions are an important cause of hospital admissions, accounting for between 3 and 8 percent of all admissions.16,27 The frequency of adverse effects certainly rises with the coming of old age, but there is some question as to how much of this increase is explained by the increasing prevalence of disease in this age group as opposed to the independent contribution of age-related changes in pharmacokinetics and pharmacodynamics.30 Nonetheless, from clinical, public health, and economic perspectives, it is certainly true that adverse effects of medications are an increasingly important contribution to hospitalization in the elderly, and debate has begun to focus on the preventability of such drug-related admissions. Are drug-induced illnesses more common in the elderly because they are inevitable consequences of the use of powerful therapeutic agents in an age group that needs them, or could more judicious use of therapies result in a reduction in the rates of adverse drug reactions? The latter possibility is addressed in a study of several hundred emergency admissions to a teaching hospital, in which preadmission outpatient records were reviewed to determine whether the admission could have been prevented. Investigators determined that, for drug-related admissions, fully 50 percent could have been prevented by more judicious prescribing from patients' ambulatory care physicians.12
The economic burden of drug-induced disability is difficult to measure with precision. Estimates can be made by considering the proportion of hospital days associated with adverse drug reactions (including those causing admission as well as additional days of hospitalization required by inpatient drug reactions) and adjusting this figure to reflect hospital days used by those over the age of 50. Using this method, which considers only hospital-related drug reactions, the cost of drug-induced illness is high.23,49 The cost would need to be reduced substantially (perhaps as much as by half) if one were to include only preventable drug toxicity.
Another important aspect of the economic burden of medication-related disability is the cost of care for illnesses that could have been prevented or ameliorated by the rational use of drug therapy. This expense is even more difficult to calculate, but it can be roughly estimated by considering, for example, the reduction in the rate of stroke that can be achieved with the management of high blood pressure in this age group and then estimating the prevalence of nontreatment of hypertensive patients aged 50 and older. The costs of ''preventable strokes that were not prevented" are substantial, bringing into perspective the contribution to total health care expenditures of medication under- as well as overuse.1,45
The Impairment/Disability/Handicap Sequence
Even in the most basic studies of drug effects, the elderly have been systematically neglected. For example, early studies of the treatment of essential hypertension contained few subjects over the age of 65, either because of perceived problems of accessibility or outright exclusion of such patients by the study design. It is only in the last few years that clinical trials specifically designed to address high blood pressure in the elderly have begun to appear.2 Data from these studies indicate that it is, indeed, advantageous to treat high blood pressure in older patients, at least up to age 80, and that, applied widely, this practice could prevent considerable morbidity, especially from stroke. These findings, however, have not yet begun to permeate the consciousness of many practicing physicians (see Chapter 3 for a more detailed discussion of high blood pressure).
Evidence on the efficacy/risk profile of most new drugs is gathered in premarketing testing that generally includes only modest numbers of truly elderly subjects, most of whom are in better health than the typical geriatric patient who receives such medications once they have been marketed. Elderly patients are often excluded from trials because they are more likely to experience complications from therapy (or from underlying diseases). The result of their underrepresentation in new drug trials is that the scientific basis used by physicians (or patients) to make rational assessments of benefit and risk is deficient for many of the medications used most commonly in the older age group.50 This problem is illustrated by the case of nonsteroidal anti-inflammatory drugs (NSAIDs) in the treatment of osteoarthritis. Degenerative joint disease is a major cause of disability in the second 50 years of life, and its medical management represents one of the most important ways in which drugs can enhance functional capacity in this age group. Yet even for
this common class of medications, major questions remain concerning risks and benefits as a function of age. First, not enough is known about the differences between the newer NSAIDs and aspirin in this regard, despite the enormous difference in cost. Tinnitus can serve as a sign of salicylate toxicity in the young, but age-associated loss of high-frequency hearing may blunt its occurrence in the old. What are the implications of this understanding for drug choice in the management of arthritis? Similarly, it is possible that the important age-related difference in NSAID toxicity may not be the frequency with which gastrointestinal bleeding occurs but the frequency with which it is fatal, which may be higher in the elderly.22 There is some provocative evidence on this point, but clearly, more data are needed.
Lipid-lowering therapy is another area of practice that lacks risk-benefit information in relation to the older patient. The reduction of serum cholesterol is seen by most authorities as an efficacious approach to the prevention of cardiac disease; yet there has been little study of the contributions of the new lipid-lowering drugs, particularly in the second half of life and despite the fact that the elderly are among the most prominent consumers of such agents. (Ironically, the old may be a cohort for whom such medication may be of considerably less importance.46) In addition to the methodological challenge of quantifying the relative benefit of prolonging life in older versus younger patients, there is also the thorny issue of quantifying the "good" derived from postponing or preventing cardiovascular disability in the two age groups. These difficulties are by no means unique to the evaluation of drug therapies and apply with equal vexation to such interventions as smoking cessation and dietary change. But intervention in the form of medication adds another level of complexity because of the higher frequency of "dis-benefits" that can occur with drugs, whether as adverse effects or greater costs. Considering only the former, even if benefits were relatively constant but the risk of adverse effects increased with age, there would probably be a point at which on average the medication would be as likely to do harm as good. Running in a completely opposite direction are speculations that the new class of lipid-lowering agents may prevent numerous forms of debility associated with the aging process, potentially turning the risk-benefit relationship around. Clearly, much more research is needed on the actual contribution of this important new class of drugs to health promotion and disability prevention in older patients.
Too few studies have considered the impact of drugs on functional capacity (its improvement or deterioration) in the second half of life. Typically, clinical endpoints reflect much more narrow
therapeutic goals (e.g., the reduction of blood pressure) instead of focusing on the ways in which functional status (e.g., cognition, affect, or sexual potency) is affected by antihypertensive therapy. Recently, a larger view has been taken of drug effects in the elderly, which are frequently described as quality of life endpoints.18 Within this framework, medications that initially appear to have comparable therapeutic efficacy in terms of a narrowly defined outcome may have vastly different effects on more broadly defined measures of health status. Specifically, antihypertensive therapies have different effects on cognition and mood, independent of their impact on blood pressure. Alternatively, quality of life outcomes may be the direct result of a drug's intended therapeutic effect, as when the successful treatment of incontinence has a positive impact on self-image and life satisfaction.37
Efforts to describe the outcomes of drug therapy in terms of a quantitative expression of such measures (such as the quality-adjusted life year, or QALY) have been few and far between. Considering the major methodologic and conceptual problems associated with this approach, it is not surprising and may in fact be beneficial. In the hands of some authors, the calculation of QALYs systematically devalues interventions that benefit the old because an intervention in the second half of life is likely to yield fewer "years saved." Each year of life may be further devalued by the presence of common conditions of old age, such as arthritis, or, in the schemes of some authors, by old age itself.3
Research on drug effects in the elderly suggests the complex ways in which aging and therapeutics can interact to affect health status. The frequency of adverse drug reactions (ADRs) increase with age,30 although it is less than clear whether this increase is an effect of senescence itself or of the higher frequency of co-existing illness and polypharmacy in the elderly—an important distinction. Also unclear at times is the relationship between the prescribing event and the ADR, which may be temporally and clinically separated. For example, there is considerable uncertainty regarding the increase in hip fractures associated with psychoactive drug use in the elderly,34 especially as other evidence suggests that medications can have a protective effect against this most disabling geriatric illness. (The use of estrogens or thiazides can reduce the frequency and severity of osteoporosis;32,38 likewise, estrogens may prevent some cardiovascular disability in postmenopausal women.14) Nonetheless, ADRs are clearly an important cause of hospitalization and disability, much of which may be prevented through more judicious prescribing.12
An often overlooked contributing factor to ADRs is the interaction of self-prescribed over-the-counter medications with prescribed medications. In addition to being the largest consumers of prescription drugs, the elderly also are the largest consumers of over-the-counter drugs. As more of these over-the-counter medications become available, the potential adverse interactions will also increase. Moreover, the direct adverse effects of these drugs themselves need further study.51
Mental impairment resulting from the use of psychoactive drugs, in both institutionalized elderly and younger patients past the age of 50, is an issue of some controversy. Clearly, some reduction in cognitive capacity can result from chronic use of benzodiazepines and neuroleptics in some patients, but more needs to be learned about the nature and magnitude of such changes.26,40 In one area of the mental health realm, however, age may provide protection from drug-induced disability. The depression associated with beta-blocker use appears to occur with less frequency in the old than in those who are younger, a phenomenon that may be related to the reduced sensitivity of adrenergic receptors with aging.6
PREVENTABILITY OF BURDEN
A large body of data indicate that there is considerable room for improvement in the way medications are used in the over-50 age group. Physicians are not as proficient as they might be in optimal prescribing for the elderly, a deficit reflected in actual prescribing practices11,33 and in surveys of physician knowledge.4,20 In addition, physician-patient communication is often problematic, both in the areas of history taking for therapeutic decision making and communication about drug effects, precautions, and compliance.17,19
Fortunately, a number of interventions have been developed to address these problems; some have even been field-tested in randomized controlled trials. Traditional educational methods using group lectures and mailed informational material appear to be of limited efficacy in changing prescribing practices;7,24,44 however, consistent, reproducible data indicate that in-person, face-to-face education provided by clinical educators (either pharmacists or other physicians) has proven effective in making physicians' prescribing decisions more precise. In several studies that track physician prescribing across six states, such interventions have resulted in a reduction in inappropriate prescribing that was both statistically and economically significant.9,41 A formal benefit-cost analysis of the largest of these studies has further shown that the dollars saved by a third-party reimbursement program
(Medicaid) actually exceeded the cost of running the program by a factor of 2 to 1.43 This approach has now been extended to nursing homes, in which an educational outreach effort (also known as "public interest detailing") resulted in a significant reduction in the excessive use of psychoactive medication in the six long-term care facilities studied.5 The intervention, which consisted of separate educational sessions with physicians, nurses, and aides, also resulted in an improvement in the cognitive status of residents in the experimental homes as measured by a detailed battery of neuropsychiatric and functional status tests.
Physicians should review, with their elderly patients, all medications being taken, prescription and non-prescription, for appropriateness, potential adverse interactions, and continued need at least every six months.
Periodic medication review should be reimbursed as a non-procedural activity.
Risk versus benefit: For many drug groups, it may not be appropriate to translate the risk and benefit properties of medications as determined in younger populations into care of the elderly. For example, although the efficacy of a drug may not diminish with age, the frequency and severity of the adverse effects it causes may well increase. Therefore, these drug characteristics should be reassessed in elderly patients de novo and not extrapolated from younger populations.
Impact on functional capacity: It is not enough to measure drug effects in older patients merely in relation to one immediate therapeutic goal (e.g., cessation of ventricular arrhythmias, decrease in intra-ocular pressure). Rather, the effects of drugs should be measured in relation to functional status and other less obvious endpoints, including central nervous system function, gait stability, and functional capacity of other organ systems.
Virtually all health professional education (including that of most physicians, nurses, and pharmacists) is inadequate in its treatment
of therapeutics, as well as in its presentation of materials in geriatrics.8 Considerably more information on the proper use of medications in the elderly should be systematically integrated into the curricula of health education programs, even as the knowledge base itself is extended in the manner described above.
Elderly patients themselves would benefit from additional education concerning their use of medications. Although some programs have been developed, more are needed, particularly to alert elderly patients and their families to the possibility of drug side effects (instead of emphasizing only patient responsibility for total compliance with prescribed regimens).29
There continues to be substantial controversy concerning drug testing in the elderly. Some fault the Food and Drug Administration for a lack of guidelines, which have been discussed since 1983 but never enacted as law; the agency and many in the pharmaceutical industry claim that discussion of these guidelines has resulted in voluntary compliance with their principles, making further regulatory action unnecessary.47 At a minimum, data are needed to clarify whether this is, indeed, the case. In addition, distinctions must be made between the needs of patients labeled as "old" because they are 60 or older (many of whom are 61 or 62) versus truly geriatric patients in their eighties or more, with at least some coexisting illness. If voluntary compliance appears to be inadequate, stronger regulatory measures may be necessary.
New drugs to be used primarily among the elderly should be studied prior to marketing to determine their effects on important geriatric endpoints (e.g., intellectual functioning and other aspects of functional capacity), as well as the impact on their labeled indication.
Given the inadequate training of most physicians in the optimal use of medications in the elderly, it is not surprising that evidence of poor prescribing abounds, ranging from the improper choice of a dose or of agents within a class, to failure to prescribe an indicated therapy (e.g., an antihypertensive) for a patient who needs it.31,36 In the coming decades, the increasing role of government and other payers in shaping clinical decision making will make it more acceptable to require demonstrations of competence in various areas of practice, including prescribing for the elderly, to maintain credentials or receive payment for services.10
more than it costs.43 This approach should now be tested in large-scale demonstration studies.
The use of existing, claims-based data sets of prescription information to study drug effects in the elderly should be increased.13,35 With such data sets, drug epidemiologists can provide surveillance to track adverse drug effects in key subpopulations. However, adequate funding has yet to be provided.
As developments in pharmacology and molecular biology produce ever more powerful (and costly) new additions to the therapeutic armamentarium, it will become increasingly important for the health care system to gain a more sophisticated understanding of drug effects defined broadly in the aging population. Consideration should be given to how drugs are tested for efficacy and risk, who participates in trials, what outcomes are studied, and how surveillance of drug effects on large populations is carried out, once the drugs are in widespread use.
1. Amery, A., Birkenhauger, W., Brixko, R., et al. Efficacy of antihypertensive drug treatment in patients over 60. Lancet 1986; 1:589.
2. Applegate, W. B. Hypertension in elderly patients. Annals of Internal Medicine 1989; 110:901.
3. Avorn, J. Benefit and cost analysis in geriatric care: Turning age discrimination into health policy. New England Journal of Medicine 1984; 310:1294-1301.
4. Avorn, J., Chen, M., and Hartley, R. Scientific versus commercial sources of influence on the prescribing behavior of physicians. American Journal of Medicine 1982; 73:4-8.
5. Avorn, J., Soumerai, S. B., et al. Clinical consequences of a reduction in psychoactive drug use in nursing homes. The Gerontologist, in press.
6. Avorn, J., Everitt, D. E., Weiss, S., et al. Increased antidepressant use in patients prescribed beta-blockers. Journal of the American Medical Association 1986; 255:357-360.
7. Avorn, J., Olins, N. J., and Trotter, M. Efficacy of printed medication information for the elderly: A randomized controlled trial. Harvard Medical School, unpublished manuscript.
8. Avorn, J. Pharmacology and geriatric practice: A case study in technology nontransfer. Clinical Pharmacology and Therapeutics 1987; 42:674-676.
9. Avorn, J., and Soumerai, S. B. Improving drug-therapy decisions through educational outreach. New England Journal of Medicine 1983; 308:1457-1463.
10. Avorn, J. Testimony before a joint hearing of the U.S. Senate and House of Representatives. U.S. Congressional Record, S. Hrg. 98-392. Washington, D.C.: U.S. Government Printing Office.
11. Beers, M., Avorn, J., Soumerai, S. B., et al. Medication use in nursing home residents. Journal of the American Medical Association 1988; 260:3016-3020.
12. Bigby, J., Dunn, J., Goldman, L., Adams, J. B., Jen, P., Landefeld, C. S., and Komaroff, A. L. Assessing the preventability of emergency hospital admissions. A method for evaluating the quality of medical care in a primary care facility. American Journal of Medicine 1987; 83(6):1031-1036.
13. Bright, R. A., Avorn, J., and Everitt, D. E. Medicaid database development for epidemiologic research. Journal of Clinical Epidemiology, in press.
14. Bush, T. L., Barrett-Connor, E., et al. Cardiovascular mortality and non-contraceptive use of estrogen. Circulation 1987; 75:1102.
15. Campion, E. W., deLabry, L. O., and Glynn, R. J. The effect of age on serum albumin in healthy males: Report from the Normative Aging Study . Journal of Gerontology 1988; 43:M18-M20.
16. Caranosos, C. J., Stewart, R. B., and Cluff, L. E. Drug-induced illness leading to hospitalization. Journal of the American Medical Association 1974; 228:713.
17. Carty, M., Avorn, J., Everitt, D. E., and Bright, R. A. Physician-patient communication and geriatric medication use. The Gerontologist 1985; 25:33.
18. Croog, S. H., Levine, S., Testa, M. A., et al. The effects of hypertensive therapy on the quality of life. New England Journal of Medicine 1986; 314:1657-1664.
19. Everitt, D., Avorn, J., and Baker, M. The prescribing of hypnotics: Therapeutic decisions based on inadequate history. The American Journal of Medicine, in press.
20. Ferry, M. E., Lamy, P. P., Becker, L. A., et al. Physicians' knowledge of prescribing for the elderly. Journal of the American. Geriatrics Society 1985; 33:616.
21. Greenblatt, D., Seller, E., and Shader, R. Drug disposition in old age. New England Journal of Medicine 1982; 306:1081-1087.
22. Griffin, M. R., Ray, W. A., and Schaffner, W. Nonsteroidal anti-inflammatory drug use and death from peptic ulcer in elderly persons. Annals of Internal Medicine 1988; 109:359-363.
23. Grymonpre, R. E., Mitenko, P. A., Sitar, D. S., Aoki, F. Y., and Montgomery, P. R. Drug-associated hospital admissions in older medical patients. Journal of the American Geriatrics Society 1988; 36:1092-1098.
24. Haynes, R. B., Davis, D. A., McKibbon, A., and Tugwell, P. A critical appraisal of the efficacy of continuing medical education. Journal of the American Medical Association 1984; 251:61-64.
25. Jaffe, M. E. The clinical investigation of drugs for use by the elderly: Industry initiatives. Clinical Pharmacology and Therapeutics 1987; 42:677-680.
26. Larson, E. B., Kukull, W. A., Buchner, D., et al. Adverse drug reactions associated with global cognitive impairment in elderly persons. Annals of Internal Medicine 1987; 107:169-173.
27. Levy, M., Kewitz, H., Altwein, W., et al. Hospital admissions due to drug reactions: A comparative study from Jerusalem and Berlin. European Journal of Clinical Pharmacology 1980; 17:25.
28. Lindeman, R. D., Tobin, J. D., and Shock, N. W. Longitudinal studies on the rate of decline in renal function with age. Journal of the American Geriatrics Society 1985; 33(4):278-285.
29. Morrow, D., Leirer, V., and Sheikh, J. Adherence and medication instructions: Review and recommendations. Journal of the American Geriatrics Society 1988; 36:1147-1160.
30. Nolan, L., and O'Malley, K. Prescribing for the elderly. Part I. Sensitivity of the elderly to adverse drug reactions. Journal of the American Geriatrics Society 1988; 36:142-149.
31. Nolan, L., and O'Malley, K. Prescribing for the elderly. Part II. Prescribing patterns: Differences due to age. Journal of the American Geriatrics Society 1988; 36:245-254.
32. Quigley, M. E., Martin, P. L., Burnier, A. M., and Brooks, P. Estrogen therapy arrests bone loss in elderly women. American Journal of Obstetrics and Gynecology 1987; 156:1516-1523.
33. Ray, W. A., Federspiel, C. F., and Schaffner, W. A study of antipsychotic drug use in nursing homes: Epidemiologic evidence suggesting misuse. American Journal of Public Health 1980; 70:485-491.
34. Ray, W. A., Griffin, M. R., Schafner, W., Baugh, D. K., and Melton, L. J. III. Psychotropic drug use and the risk of hip fracture. New England Journal of Medicine 1987; 316:363-369.
35. Ray, W. A., and Griffin, M. R. Use of Medicaid data for pharmacoepidemiology. American Journal of Epidemiology 1989; 129:837-849.
36. Reidenberg, M. M. Drug therapy in the elderly: The problem from the point of view of a clinical pharmacologist. Clinical Pharmacology and Therapeutics 1987; 42:677-680.
37. Resnick, N. M., and Yalla, S. V. Management of urinary incontinence in the elderly. New England Journal of Medicine 1985; 313(13):800-805.
38. Riggs, B. L., and Melton, L. J. (eds.), Osteoporosis. New York: Raven Press, 1988.
39. Rowe, J. Aging, renal function, and response to drugs. In: L. Jarvik (ed.), Clinical Pharmacology and the Aged Patient. New York: Raven Press, 1981.
40. Salzman, C. Clinical Geriatric Psychopharmacology. New York: McGraw-Hill, 1984.
41. Schaffner, W., Ray, W. A., Federspiel, C. F., and Miller, W. D. Improving antibiotic prescribing in office practice. Journal of the American Medical Association 1983; 250:1728.
42. Scott, J. C., and Stankski, D. R. Decreased fentanyl and alfentanil dose requirements with age: A simultaneous pharmacokinetic and pharmacodynamic evaluation. Journal of Pharmacological and Experimental Therapeutics 1987; 204:159-166.
43. Soumerai, S. B., and Avorn, J. Economic and policy analysis of university-based drug "detailing." Medical Care 1986; 24:313-331.
44. Soumerai, S. B., and Avorn, J. Efficacy and cost-containment in hospital pharmacotherapy: State of the art and future directions. Milbank Memorial Fund Quarterly 1984; 62(3):447-474.
45. Strandgaard, S., and Haunso, S. Why does anti-hypertensive treatment prevent stroke but not myocardial infarction? Lancet 1987; 2:658.
46. Taylor, W. C., Pass, T. M., Shepard, D. S., and Komaroff, A. L. Cholesterol reduction and life expectancy. A model incorporating multiple risk factors. Annals of Internal Medicine 1987; 106(4):605-614.
47. Temple, R. FDA discussion paper on testing of drugs in the elderly. Department of Health and Human Services, Public Health Service, Food and Drug Administration, Washington, D.C., September 1983.
48. Vestal, R., Norris, A., Tobin, J., Cohen, B., Shock, N., and Andres, R. Antipyrine metabolism in man: Influence of age, alcohol, caffeine, and smoking. Clinical Pharmacology and Therapeutics 1975; 18:425-432.
49. Williamson, J., and Chopin, J. M. Adverse reactions to prescribed drugs in the elderly: A multicentre investigation. Age and Ageing 1980; 9:73.
50. Wolfe, S. M., et al. Worst Pills, Best Pills. Washington, D.C.: Public Citizen, 1988.
51. Abrams, R. C., and Alexopoulous, G. S. Substance abuse in the elderly: Over the counter illegal drugs. Hospital and Community Psychiatry 1988; 39(8):822-823.