Drug Therapy in the Elderly:

March 1, 2004
Mark A. Stratton, PharmD
Mark A. Stratton, PharmD

Sheryl Guiterres, PharmD
Sheryl Guiterres, PharmD

Robert Salinas, MD
Robert Salinas, MD

Numerous factors put elderly patients at risk for adverse drug events. On average, they take at least 6 medications a day, which increases the likelihood of drug-drug interactions. In addition, many drugs that are safe and effective in younger patients are inappropriate for older persons because of age-related changes and comorbid conditions that affect absorption, distribution, metabolism, and elimination. First-pass metabolism decreases with age, which may increase systemic absorption of some oral nitrates, ß-blockers, estrogens, and calcium channel blockers. The age-related rise in body fat increases the volume of distribution of lipid-soluble compounds, such as diazepam, and prolongs clearance. About two thirds of elderly persons have impaired kidney function; in these patients, the dosage of renally excreted drugs-such as digoxin-needs to be reduced. Other strategies for avoiding adverse drug events are detailed here.

There are currently about 35 million persons in this country over age 65-about 13% to 14% of the population. By 2020, it is projected that 1 in every 5 persons will be over age 65.1 The "oldest old"-those over age 85-constitute the most rapidly growing subsegment of this population; their numbers are expected to triple during the next 30 years.

In view of these estimates, it is not surprising that the management of medications in the elderly population is an area of considerable clinical importance. For example, of all deaths caused by adverse reactions to medications in 1986, 51% occurred in persons over age 60.2 In addition, 39% of all persons hospitalized because of an adverse reaction to medications were over age 60.2 Up to 100,000 patients die each year in US hospitals because of medical errors, and half of these deaths are medication-related.3

The toll of medication errors is considerable-not only in terms of morbidity and mortality but also in costs to the health care system. Data from the 1990s show that the rate of hospitalization for medication-related problems is 3 times higher in older persons.4 Medication errors account for an estimated 12% to 16% of all hospitalizations among persons over age 65.4 The estimated cost to the health care system of inappropriate drugs and the consequences of their use approaches $200 billion per year, half of which is incurred by the elderly.5,6

Here we discuss the factors that contribute to what former Secretary of Health and Human Services Lewis Sullivan termed "America's other drug problem." We review some of the challenges of prescribing for elderly patients-and offer recommendations for avoiding or minimizing adverse drug reactions.


Persons older than 65 years use 31% of all prescription drugs and purchase more than 40% of the over-the-counter (OTC) drugs sold in the United States.5,6 Studies of elderly persons who live independently show that they take an average of 6 different medications per day; those living in a long-term-care facility take an average of 8 medications.7,8 Polypharmacy heightens the risk of significant drug-drug interactions and other adverse events, particularly in frail elderly persons.

The use of medications that are inappropriate for older persons also increases the risk of adverse reactions. Beers and colleagues9,10 developed criteria for categorizing medications according to their risk in the elderly. An estimated 20% to 25% of elderly persons take at least 1 drug considered inappropriate based on the Beers criteria.11


Compliance, or adherence to a drug regimen, becomes problematic for elderly persons when the number of medications or the frequency of administration increases. However, the primary reason that older patients do not adhere to a regimen is the escalating cost of drugs. The cost of prescription drugs has increased more than two and a half times since 1993. A study found that 18% of older Californians either did not fill a prescription or skipped doses because of the high cost of medications; this percentage increased to 29% among those with no prescription drug coverage.12


Increasing age can significantly affect drug absorption, distribution, metabolism, and elimination. Other factors-including frailty and comorbid conditions, such as congestive heart failure (CHF)-also affect drug kinetics in older persons.

The kinetic changes that affect younger elderly persons are likely to continue as a patient ages and to become more clinically relevant. A relatively new concept is the impact of frailty on kinetics: as a patient becomes more frail, the kinetic changes associated with aging become more significant.

A summary of the kinetic changes associated with age and disease is provided in Table 1. However, bear in mind that the heterogeneity of the aging population makes generalizations about kinetic changes in older persons problematic. In addition, current knowledge concerning aging and pharmacokinetic changes has largely been derived from data in patients between the ages of 65 and 74 years. There are far fewer data for those between 75 and 85 years and little or none for those older than 85 years.

Absorption. Drug absorption changes in elderly persons are less clinically important than changes in the other pharmacokinetic parameters. First-pass metabolism decreases with age, which may increase systemic absorption of some oral nitrates, β-blockers, estrogens, and calcium channel blockers. For example, the effective dose of the oral nitrate isosorbide dinitrate is likely to be much lower in an 85-year-old than in a 55-year-old.13 The higher dose may produce such undesirable effects as significant orthostatic changes in blood pressure, with a subsequent increased risk of falls and fractures. This effect is not seen with isosorbide mononitrate.

Disease and age combine to alter absorption of oral medications. For example, in elderly patients with worsening CHF, bowel edema may interfere with the absorption of many medications. The slowed rate of furosemide absorption can lead to a diminished clinical effect.14 Substitute parenteral furosemide if this phenomenon occurs. Another alternative is oral torsemide, which is more predictably absorbed than furosemide in patients with worsening CHF.15

Distribution. With increasing age, the lean-to-fat ratio of body mass decreases, as does total body water.16 These changes significantly reduce the volume of distribution for water-soluble compounds and for compounds distributed only in lean tissues, such as digoxin and lithium. The therapeutic range of lithium for elderly patients is approximately half the standard range for younger persons.17

The age-related rise in body fat increases the volume of distribution of lipid-soluble compounds. For example, the volume of distribution of diazepam increases 2- to 3-fold from age 20 to age 80.18,19 Consequently, the clearance of this and other lipid-soluble psychoactive compounds is markedly prolonged, which increases the likelihood of serious adverse consequences.

Although plasma protein concentrations do not normally decline with age to a degree sufficient to alter pharmacokinetics, reduced food intake or catabolic disease states may lead to clinically relevant decreases.20 This becomes especially important with compounds that are normally highly protein-bound, such as warfarin and phenytoin. Use lower dosages of these agents in patients with hypoalbuminemia. Warfarin dosage changes are best made in small increments. The therapeutic range for phenytoin may be 5 to 10 µg/mL rather than 10 to 20 µg/mL.

Metabolism. Liver mass decreases by 25% to 35% and liver blood flow decreases by as much as 40% by the age of 90 years.21 A number of hepatic metabolic pathways diminish with age. This is the case with many phase I reactions, which include reduction, oxidation, hydroxylation, and demethylation. Many of these pathways are performed through the multiple cytochrome P-450 mixed function oxidase systems. About half of these pathways are impaired in elderly persons.22 Agents that use these pathways include the benzodiazepines diazepam, chlordiazepoxide, flurazepam, and alprazolam.

Studies have demonstrated that the half-life of diazepam and its metabolites increases 2- to 3-fold in elderly persons.18,19 The clinical effect of this metabolic change is an increased incidence of falls, fractures, and automobile accidents.23,24 It is therefore best to avoid using these compounds (and others that follow this metabolic pathway) in elderly patients.

Phase II reactions-which include conjugation, acetylation, glucuronidation, and sulfation-do not diminish with age. If benzodiazepine therapy is necessary, 3 acceptable alternatives are oxazepam, lorazepam, and triazolam. These agents are shorter-acting, and their metabolism is not affected by age.Use the lowest possible dose for the shortest period.

Elimination. It was once thought that renal function declined by approximately 10% per decade after the age of 50 years. We now know that, in the absence of hypertension and diabetes, 35% of elderly persons retain normal renal function until late in life; however, this means that most elderly persons have reduced renal function.25 The dosage of many compounds that depend solely on the kidneys for elimination-such as the aminoglycosides, vancomycin, and digoxin-therefore requires downward adjustment.

A frequently used formula for estimating creatinine clearance is ([140 − age] × lean body weight [kg]/(72 × serum creatinine level [mg/dL]).26 (For women, the total is multiplied by 0.85.) However, the formula loses its specificity with advancing age and loss of body mass.


With age, sensitivity to the therapeutic as well as the toxic effects of many medications-especially those that act centrally-increases. The effect of aging is thought to lead to decreased density and affinity of receptors, while the postreceptor effect is mixed.27 The effects summarized in Table 2 are more common and more intense in frail and very old patients.


Antihypertensives. Because elderly persons are sensitive to centrally mediated side effects, peripherally acting antihypertensives appear to be the agents of choice in those with hypertension. Angiotensin-converting enzyme (ACE) inhibitors are well tolerated. The calcium channel blockers are especially well matched to the pathology of hypertension in the elderly (with the exception of short-acting nifedipine, which should be avoided). In patients with CHF and hypertension, an ACE inhibitor can be used for both conditions. In patients with hypertension and angina, a calcium channel blocker can be used for both conditions. Such dual functions help reduce the number of medications.

If a centrally acting agent, such as clonidine, is necessary, start therapy at the lowest possible dosage. If a β-blocker is being considered, use the least lipophilic agent to minimize centrally mediated side effects. For example, atenolol is preferable to propranolol.

The benefits of diuretics were recently reaffirmed in the Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT), which showed that these agents were superior to ACE inhibitors and calcium channel blockers in preventing major forms of cardiovascular disease, including myocardial infarction.28 The recommended daily dose of hydrochlorothiazide is 12.5 mg, with a maximum of 25 mg/d. Dosages higher than these produce no further reductions in blood pressure and result in greater electrolyte disturbances. Hydrochlorothiazide is much less costly than ACE inhibitors or calcium channel blockers.

Antiarrhythmics. The pharmacokinetics of numerous antiarrhythmics are altered in elderly persons, either as a result of age alone or of age and pathology combined.16,29 The pharmacologic effect of an antiarrhythmic agent should be judged by its electrocardiographic consequences and not necessarily by serum levels. The therapeutic range in elderly patients may need to be redefined.

Digitalis glycosides. Because of decreased lean mass and reduced renal function in elderly persons, the digoxin dosage should be lower than in younger patients. Digoxin is sometimes used inappropriately in the absence of atrial fibrillation or a definitive diagnosis of CHF and thus has been discontinued in many elderly patients without ill effects.

In elderly patients with CHF of recent onset, the predominant pathology is likely to be diastolic dysfunction, which responds better to afterload reduction than to inotropic agents. Digoxin is contraindicated in such patients and therapy with other agents should be considered whenever possible.

Anticoagulants. Patients older than 65 years are more likely than younger ones to bleed as a result of anticoagulant therapy and to experience catastrophic consequences as a result.29,30 Bear the following in mind when considering the use of warfarin in elderly persons:

  • Evaluate the patient's plasma protein status. If the total protein or albumin concentration is low, initiate warfarin at a lower dosage.
  • Use the most current guidelines for anticoagulants.
  • Evaluate the need for therapy frequently and use warfarin for the shortest possible time.

Educate patients about the increased risk of bleeding with warfarin as well as the importance of compliance. Be sure they understand which types of bleeding they should notify you about. Discuss dietary considerations as well as the potential for interactions with other prescription and OTC medications.

Anxiolytics and sedatives. Select a short-acting agent that is metabolized efficiently and has a good safety record. Among the preferred agents are oxazepam, lorazepam, and triazolam. Use the lowest possible dosage, frequently evaluate the need for therapy, and closely monitor the patient's quality of life and mental status. Consider nondrug alternatives; for example, meditation and biofeedback relieve anxiety in many patients. Suggest to patients with insomnia that they closely examine their sleep hygiene habits and eliminate practices (such as caffeine use and daytime naps) that might contribute to insomnia. An appropriate sleep environment and regular physical activity help promote good sleep patterns.

Antipsychotic agents. The most appropriate antipsychotic agents are those that have minimal sedative and anticholinergic side effects. The newer atypical antipsychotics, such as risperidone, quetiapine, and olanzapine, meet these criteria. Of the three, risperidone has the most extrapyramidal side effects; quetiapine and olanzapine are associated with greater anticholinergic effects and weight gain.31,32 In the opinions of Richelson31 andMeyer,32 risperidone is the antipsychotic of choice in elderly patients because it has fewer extrapyramidal side effects than the older antipsychotics, such as haloperidol and thioridazine (which are therefore not recommended).Initiate therapy at the lowest possible dose, and monitor carefully for efficacy and toxicity. Only limited data are available about the long-term use of the newest atypical antipsychotics, aripiprazole and ziprasidone; thus, their role in the elderly remains to be defined.

Antidepressants. The idealantidepressant for an elderly patient is minimally sedating, has few cardiotoxic and anticholinergic effects, and is associated with minimal orthostatic changes in blood pressure. The antidepressants of choice are the selective serotonin reuptake inhibitors (SSRIs) sertraline, paroxetine, citalopram, and escitalopram and the mixed agents venlafaxine and bupropion.33 Fluoxetine is not recommended because of its propensity to produce anorexia, anxiety, and insomnia and because of its prolonged half-life.34 Trazodone and nefazodone are acceptable alternatives, especially in depression with concomitant insomnia, but their half-life is prolonged in the elderly. The dosages of antidepressants effective in elderly patients are typically one sixth to one third of those used in younger persons.

Analgesics. Because pain perception diminishes with age, elderly persons may be quite sensitive to the analgesic effects of narcotics and therefore require lower dosages. When a narcotic is needed, start with a relatively low-potency agent, such as codeine. However, this agent is associated with a high incidence of nausea, vomiting, and constipation. Meperidine is unacceptable not only because of its poor oral bioavailability but also because of the accumulation of the active metabolite normeperidine, which may produce psychosis in elderly persons.35 Propoxyphene is best avoided for the same reason. For severe chronic pain, an agent that contains oxycodone can be tried. An alternative narcotic-like agent, tramadol, may have a prolonged half-life in elderly persons. With any narcotic analgesic, begin with the lowest dosage and monitor carefully.

Elderly persons are more likely than younger ones to experience gastritis and GI bleeding from NSAID use. The COX-2 inhibitors are associated with a lower incidence of GI bleeding and are preferred for patients who have a history of peptic ulcer disease or GI problems stemming from NSAID therapy.36,37

Numerous advising bodies-including the American Geriatric Society-recommend acetaminophen for the management of mild to moderate pain in older persons.38 The daily dose should not exceed 4 g; lower dosages are advisable for those who have liver enzyme abnormalities. The maximum daily dose for patients who drink alcohol daily should probably not exceed 2 g. Monitor these patients for hepatotoxicity.

Antidiabetic agents. In elderly patients, glucose counterregulation is impaired and the presentation of hypoglycemia may be altered (eg, patients may present with somnolence or confusion). Thus, consider starting an oral antidiabetic agent athalf the usual dose. It is probably best to avoid the first-generation sulfonylureas (especially chlorpropamide) because of impaired elimination and a higher incidence of hypoglycemia. Glyburide has a longer elimination half-life than glipizide and also is associated with a higher incidence of hypoglycemic episodes in elderly patients.39

Insulin glargine, an alternative insulin preparation, has a smooth dose-response curve and thus may be less likely than regular insulin to cause hypoglycemia.

Antiulcer and reflux medications. The H2 antagonistcimetidine is excreted chiefly by the kidneys. Thus, elderly persons with diminished renal function may experience more CNS side effects at standard dosages. In addition, because elderly persons often take a large number of medications (many of which could interact with cimetidine), ranitidine or famotidine is a better choice, at half the usual adult dosage. The proton pump inhibitors omeprazole, esomeprazole, and pantoprazole appear to be equally safe for elderly persons.

Nonprescription drugs. Particular caution is warranted with OTC analgesics and antihistamines. Patient education is most important, especially because many prescription medications are being changed to OTC status. Elderly patients may take alternative agents, herbs, and other supplements, many of which contain active pharmacologic substances. Because of the effects of aging, patients may experience adverse effects with these products.


Some practical recommendations that may help avert adverse drug reactions in elderly patients include the following:

  • Consider adverse changes in quality of life to be the result of medication until proved otherwise.
  • Avoid treating side effects of drug therapy with additional medications whenever possible.
  • For patients with poor vision, helpful measures include dark print on a flat background and larger print on patient education materials and prescription labels, verbal explanations, and color coding for the tops of prescription bottles. These measures can be implemented through close collaboration between the clinician and the pharmacist.
  • Ask patients with hearing loss for suggestions on the best ways to communicate with them.
  • Enlist family members to help cognitively impaired patients with prescription directions.
  • Encourage patients to be active participants in their health care and to raise questions or concerns about their medications.
  • Encourage patients to patronize a single pharmacy to allow the pharmacist to monitor both their prescription and nonprescription medication use.

Additional recommendations are provided in Table 3.40



1. Tomita DK, Kennedy DL, Baum C, et al. Drug Utilization in the United States-1988: Tenth Annual Review. Bethesda, Md: US Dept of Health and Human Services (PB90-150921); 1989:1-44.

2. Knapp DE, Tomita DK. Second Annual Adverse Drug/Biologic Reaction Report: 1986. Bethesda, Md: Office of Epidemiology and Biostatistics, Center for Drugs and Biologics, Food and Drug Administration; 1987.

3. Kohn LT, Corrigan JM, Donaldson M, eds. To Err Is Human: Building a Safer Health System. Washington, DC: National Academy Press; 1999.

4. Col N, Fanale JE, Kronholm P. The role of medication non-compliance and adverse drug reactions in hospitalizations of the elderly. Arch Intern Med. 1990;150:841-845.

5. Lamy PP. New dimensions and opportunities. Drug Intell Clin Pharm. 1985;19:399-402.

6. Shaughnessy AF. Common drug interactions in the elderly. Emerg Med. 1992;24:21-32.

7. Honig PK, Cantilena LR. Polypharmacy. Pharmacokinetic perspectives. Clin Pharmacokinet. 1994;26: 85-90.

8. McCrea JB, Ranelli PL, Boyce, EG et al. Preliminary study of autonomy as a factor influencing medication-taking by elderly patients. Am J Hosp Pharm. 1993;50:296-298.

9. Beers MH, Ouslander JG, Rollingher I, et al. Explicit criteria for determining inappropriate medication use in nursing homes. Arch Intern Med. 1991;151: 1825-1832.

10. Fick DM, Cooper JW, Wade WE, et al. Updating the Beers criteria for potentially inappropriate medication use in older adults: results of a US consensus panel of experts. Arch Intern Med. 2003;163: 2716-2724.

11. Zhan C, Sangl J, Bierman AS, et al. Potentially inappropriate medication use in the community-dwelling elderly. JAMA. 2001;286:2823-2839.

12. Neuman T, Kitchman M, McMeans T, et al. California Seniors and Prescription Drugs. Boston: The Henry J. Kaiser Family Foundation and Tufts-New England Medical Center; 2002.

13. Noyes MA, Lucas DS, Stratton MA. Principles of geriatric pharmacotherapy. J Ger Drug Ther. 1996; 10:5-34.

14. Vargo D, Kramer WG, Black PK, et al. Bioavailability, pharmacokinetics, and pharmacodynamics of torsemide and furosemide in patient with congestive heart failure. Clin Pharmacol Ther. 1995;57: 601-609.

15. Murray MD, Deer MM, Ferguson JA, et al. Open-label randomized trial of torsemide compared with furosemide therapy in patients with heart failure. Am J Med. 2001;111:513-520.

16. Parker BM, Cusack BJ, Vestal RE. Pharmacokinetic optimization of drug therapy in elderly patients. Drugs Aging. 1995;7:10-18.

17. Sproule BA, Hardy BG, Shulman KI. Differential pharmacokinetics of lithium in elderly patients. Drugs Aging. 2000;16:165-177.

18. Divoll M, Greenblatt DJ, Ochs HR, Shader RI. Absolute bioavailability of oral and intramuscular diazepam: effects of age and sex. Anesth Analg. 1983; 62:1-8.

19. Greenblatt DJ, Allen MD, Harmatz JS, et al. Diazepam disposition determinants. Clin Pharmacol Ther. 1980;27:301-312.

20. Greenblatt DJ, Sellers EM, Shader RI. Drug disposition in old age. N Engl J Med. 1982;306: 1081-1087.

21. Dawling S, Crome P. Clinical pharmacokinetics in the elderly: an update. Clin Pharmacokinet. 1989;17:236-263.

22. Kinirons MT, Crome P. Clinical pharmacokinetic considerations in the elderly: an update. Clin Pharmacokinet. 1997;33:302-312.

23. Wang PS, Bohn RL, Glynn RJ, et al. Hazardous benzodiazepine regimens in the elderly: effects of half-life, dosage, and duration on risk of hip fracture. Am J Psychiatry. 2001;158:892-898.

24. Hemmelgarn B, Suissa S, Huang A, et al. Benzodiazepine use and the risk of motor vehicle crash in the elderly. JAMA. 1997;278:27-31.

25. Lindeman RD, Tobin J, Shock NW. Longitudinal studies on the rate of decline in renal function with age. J Am Geriatr Soc. 1985;33:278-285.

26. Cockcroft DW, Gault MH. Prediction of creatinine clearance from serum creatinine. Nephron. 1976; 16:31-41.

27. Feely J, Coakley D. Altered pharmacodynamics in the elderly. Clin Geriatr Med. 1990;6:269-283.

28. The ALLHAT Officers and Coordinators for the ALLHAT Collaborative Research Group. Major outcomes in high-risk hypertensive patients randomized to angiotensin-converting enzyme inhibitor or calcium channel blocker vs diuretic: the Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT). JAMA. 2002;288: 2891-2997.

29. Nielson C. Pharmacological considerations in critical care of the elderly. Clin Geriatr Med. 1994;10: 71-89.

30. Hammerlein, A, Derendorf H, Lowenthal DT. Pharmacokinetic and pharmacodynamic changes in the elderly. Clin Pharmacokinet. 1998;35: 49-64.

31. Richelson E. Receptor pharmacology of neuroleptics: relation to clinical effect. J Clin Psychiatry. 1999;60(suppl 10):5-14.

32. Meyer JM. Effect of atypical antipsychotics on weight and serum lipid levels. J Clin Psychiatry. 2001;62(suppl 27):27-34.

33. Gareri P, Falconi U, DeFazio P, et al. Conventional and new antidepressant drugs in the elderly. Prog Neurobiol. 2000;61:353-396.

34. Anderson IM. Meta-analytical studies on new antidepressants. Br Med Bull. 2001;57:161-178.

35. Gloth FM 3rd. Pain management in older adults: prevention and treatment. J Am Geriatr Soc. 2001;49:188-199.

36. Simon LS, Weaver AL, Graham DY, et al. Anti-inflammatory and upper gastrointestinal effects of celexocib in rheumatoid arthritis. JAMA. 1999;282: 1921-1928.

37. Langman MJ, Jensen DM, Watson DJ. Adverse upper gastrointestinal effects of rofecoxib compared with NSAIDs. JAMA. 1999;282:1929-1933.

38. AGS Panel on Chronic Pain in Older Persons. The management of chronic pain in older persons. J Am Geriatr Soc. 1998;46:635-651.

39. Shorr RI, Ray WA, Daugherty JR, Griffin MR. Individual sulfonylureas and serious hypoglycemia in older people. J Am Geriatr Soc. 1996;44:751-755.

40. Garnett WR, Barr WH. Geriatric pharmacokinetics. Upjohn Monograph Series. 1984:1-27.

41. Castleden CM, George CF, Marcer D, Hallett C. Increased sensitivity to nitrazepam in old age. Br Med J. 1977;1:10-12.

42. Agostini JV, Leo-Summers LS, Inouye SK. Cognitive and other adverse effects of diphenhydramine use in hospitalized older patients. Arch Intern Med. 2001;161:2091-2097.

43. Wang Y, Chan CLY, Holden JE, et al. Age-dependent decline in dopamine D1 receptors in human brain: a PET study. Synapse. 1998;30:56-61.

44. Swift CG. Pharmacodynamics: changes in homeostatic mechanisms, receptor and target organ sensitivity in the elderly. Br Med Bull. 1990;46:36-52.