Medications and Falls in the Elderly:A Review of the Evidence and

Practical ConsiderationsElsaris Z. Riefkohl, PharmD, Heather L. Bieber, PharmD, Mark B. Burlingame, PharmD, BCPS,

and David T. Lowenthal, MD, PhD

ABSTRACTThe syndrome of falls is a common and an often serious

problem in elderly people. The cause of falls is multifactorial,and medication use can be a significant contributor. Theobservational nature of the research in this field makes itdifficult to draw conclusions regarding medication use and fallsin clinical practice, although psychotropic drugs appear to bemost strongly associated with falls. Medication assessment isan important part of evaluating elderly patients at risk forfalling.

INTRODUCTIONFalling is a common health problem for institutionalized

and community-dwelling elderly people and is a significantcontributor to morbidity and mortality. A review of prospectivestudies of community-based elderly populations reports anannual mean of 34% (range, 30%–60%) of older adults who fall,with a mean incidence rate of 0.7 falls per person per year.1 Forinstitutionalized elderly people, the same review reports anannual mean of 43% (range, 16%–75%) of residents who fall, withan incidence rate of 1.6 falls per person per year in long-termcare facilities and 1.4 falls per resident per year in hospitalizedelderly patients. The rate of falls also rises with advancingage.2

Although most falls in older adults do not result in severeinjury, 5% to 10% of falls do result in a serious outcome, suchas fracture, head injury, or laceration.1 Furthermore, acci-dents are the fifth leading cause of death in older people, withfalls contributing to two thirds of these deaths.1 Among peo-ple who fall and have fractures, those with hip fracture face themost serious consequences; approximately 20% of victims diewithin a year, and another 20% are institutionalized for the firsttime.3

The cause of falling can vary, and many risk factors havebeen identified. One risk factor is the use of certain types ofmedications, especially when used in combination. Publishedguidelines recommend a medication review as part of a com-prehensive approach to the prevention of falls in the elderly.4,5

In this article, we discuss the published evidence examin-ing the relationship between medication use and falls to pro-vide a practical approach to the clinical evaluation of drugtherapy in elderly patients at risk for falling.

MEDICATIONS AND THE RISK OF FALLSSome of the intended and unintended pharmacological

effects of drug therapy (e.g., sedation, psychomotor impair-ment, cognitive changes, dizziness, and orthostatic hypo-tension) might be expected to increase the risk of falls.Numerous published studies have sought to establish anassociation between medication use and risk of falling.

Meta-analysesLeipzig et al. published two meta-analyses in 1999 in an

attempt to clarify some of the issues surrounding medicationuse and falls.6,7 In the first study, the authors identified 40trials, between 1966 and 1996, that evaluated the associationbetween the use of sedative/hypnotic agents, antidepressants,neuroleptics, and psychotropic drugs and the risk of falls inpeople who were 60 years of age and older.6 None of the stud-ies was a randomized, controlled trial, which is a significantlimitation in the literature on this topic. When the authorsanalyzed the pooled data and calculated the odds ratios, theyfound a significant relationship between the use of psycho-tropic drugs—as a group as well as for the various classes ofpsychotropic drugs—and one or more falls.

In the second study, the authors identified 29 trials, in thesame time period, that evaluated the association between theuse of several classes of cardiovascular or analgesic drugsand falls in people 60 years of age and older.7 Again, none ofthese studies was a randomized, controlled trial. The authorsfound a significant relationship between the risk of falls and theuse of type-IA antiarrhythmic agents, digoxin, and diureticsonly but not between falls and angiotensin-converting enzymeinhibitors, calcium-channel blockers, beta blockers, centrallyacting antihypertensive agents, and nitrates. No significantassociation was found between the use of any of the analgesicdrug classes analyzed and the risk of falls.

According to these results of the meta-analyses, psycho-tropic drugs had the strongest association with falls, withcardiac and analgesic drugs having little or no correlation. Theauthors also found that patients using three or more medi-cations appeared to be at an increased risk for recurrent falls.6,7

BenzodiazepinesBecause of the potential for prolonged central nervous sys-

tem side effects, it is generally recommended that long-actingbenzodiazepines be avoided in older people. Therefore, one

Dr. Riefkohl and Dr. Bieber are Clinical Pharmacists and Dr.Burlingame is Clinical and Education Program Manager at the NorthFlorida/South Georgia Veterans Health System in Gainesville, Florida.Dr. Lowenthal is Director Emeritus of the Geriatric Research, Educa-tion, and Clinical Center at the North Florida/South Georgia VeteransHealth System and Professor of Medicine, Pharmacology, and ExerciseScience at the University of Florida’s College of Medicine in Gainesville.

724 P&T® • November 2003 • Vol. 28 No. 11

might assume that the use of these agents would pose a greaterrisk of falling than would the use of the short-acting benzo-diazepines. However, results of studies that examined fallsand both types of benzodiazepines have been variable.

Ray et al. studied nursing-home residents in order to quan-tify the rate of falls among those who took benzodiazepines andto see the variations with regard to the elimination half-livesof the drugs.8 Current benzodiazepine users experienced a rateof falls that was 44% greater than the rate for non-users. Incohort members for whom benzodiazepine use was new, therate of falls was greatest in the first seven days after the initialtherapy and remained elevated after the first 30 days of ther-apy. The rate of falls increased with longer elimination half-lives, although users of very-short-acting benzodiazepines(less than 12 hours) experienced an increased rate of fallsduring the night but not during the day.

Some studies of hospitalized patients have suggested that agreater risk of falling is associated with the use of short-actingbenzodiazepines than with long-acting benzodiazepines. Inone of these studies, patients taking lorazepam (Ativan®,Wyeth-Ayerst) and alprazolam (Xanax®, Pharmacia) had ahigher rate of falls per dose dispensed than those takingdiazepam (Valium®, Roche).9 In another study of hospitalizedpatients, the use of short-acting and very-short-acting benzo-diazepines was positively associated with falls, whereas the useof long-acting benzodiazepines was not.10

Ensrud et al., conducting a prospective, multicenter cohortstudy in community-dwelling women 65 years of age and older,concluded that the use of both short-acting and long-actingbenzodiazepines was associated with frequent falls, comparedwith the rate for those not taking these drugs; however, theconfidence interval (CI) for both classes of benzodiazepinesoverlapped (1.0).11

Studies examining exposure to benzodiazepines and therisk of hip fracture—rather than the risk of falls—have sug-gested that short-acting benzodiazepines do not confer a safetyadvantage over long-acting benzodiazepines.12,13 One of thesestudies also indicated an increased fracture risk for those tak-ing higher doses of all benzodiazepines as well as a greater riskfor patients in the early stages of their benzodiazepine therapyand also for patients who have taken benzodiazepines forlonger than a month.13

On the basis of these results, it is difficult to make practicedecisions about the risk of falling and the relationship to long-acting versus short-acting benzodiazepines. Perhaps cliniciansshould weigh the advantages and disadvantages of all cate-gories of benzodiazepines when making treatment decisionsfor elderly people at risk for falls.

AntidepressantsBecause the use of selective serotonin reuptake inhibitors

(SSRIs) is generally preferable to the use of tricyclic anti-depressants (TCAs) in older patients, one might postulate thatSSRIs are associated with a reduced risk of falls compared withTCAs. However, published data examining different classes ofantidepressants and falls indicate that older people who useSSRIs might not be any safer than those who use TCAs.

Two studies, one in the nursing-home setting and the otherin the community setting, suggest a higher risk of falls for users

of SSRIs than for users of TCAs.11,14 Another nursing-homestudy documented a significantly greater rate of falls in usersof both TCAs and SSRIs.15 Furthermore, the fall rate increasedwith increasing doses for each class of antidepressant. Thesedata suggest that elderly people who begin therapy with anyantidepressant should be considered at risk for falling.

Antihypertensive DrugsAntihypertensive agents as a class also receive a great deal

of attention from clinicians who evaluate patients at risk forfalls. Some practitioners are concerned that the blood pres-sure–lowering effects of these drugs might contribute to ortho-stasis and dizziness. Of course, acute treatment of hyper-tension can result in orthostatic hypotension through severalmechanisms; however, chronic therapy is rarely associatedwith orthostasis. In fact, it is possible that antihypertensivedrugs might improve the cerebral and systemic vascularresponses to hypotensive stress and may reduce postprandialdeclines in blood pressure.16 Furthermore, an increased riskof falls has not been associated with the use of antihyper-tensive drugs.7,17–19

So far, it appears that the use of chronic antihypertensivetherapy should not be a major concern, although individualcases of documented antihypertensive-induced orthostasisshould not be ignored.

AnticonvulsantsThe results of the prospective cohort study in elderly women

by Ensrud et al. (see the earlier discussion of benzodiazepinesand antidepressants), also noted a significantly greater risk offalls in patients using anticonvulsant agents.11 The authorsconcluded that community-dwelling elderly women who wereusing medications that affect the central nervous systemexperienced an increased risk of falling; however, narcoticuse was not associated with an increased risk.

Summary of EvidenceAlthough the varied results from the research make it dif-

ficult to derive conclusions regarding medications and falls inthe elderly, we can generalize as follows:

1. Psychotropic agents are associated with an increased riskof falling, and caution is warranted with all of these drugs.Anticonvulsants, class IA antiarrhythmics, and digoxin(e.g., Digitek®, Bertek; Lanoxin®, GlaxoSmithKline)might also be associated with an increased risk of falls.

2. The chronic use of antihypertensive drugs is probably notassociated with an increased risk of falls.

3. Potentially important pharmacotherapeutic factors, otherthan the drug or the drug class, include the use of multi-ple medications and the dosages taken.

INTERVENTION STUDIESMost intervention studies on falls provide limited or no

details about the medications taken by the study population.Perhaps one of the best-known studies in which a medicationreview was specifically identified as part of a multifacetedintervention was a trial conducted by Tinetti and colleagues ina community-based population.20 In this study, sedatives were

Medications and Falls in the Elderly

Vol. 28 No. 11 • November 2003 • P&T® 725

withdrawn and the number of medications was decreased.The intervention group demonstrated a relative risk reductionof 31% for falls.

Other studies targeting the reduced use of psychotropic,21–23

cardiovascular,23,24 and analgesic drugs23 have also reportedsuccess in decreasing the risk of falls in older people. Wereported the use of medications as a major contributing factorto the risk of falling in 25% of patients evaluated in our fallsclinic; the most frequent recommendations about medicationuse involved the class of psychotropic drugs.25

PRACTICAL CONSIDERATIONSTo evaluate the medication regimens of older patients who

might be at risk for falls, clinicians should combine an evi-dence-based approach with a knowledge of the potential effectsof drug therapy. The evaluation should also be performed inthe context of:26

• the patient’s history of falls.• an appropriate laboratory assessment.• a gait and balance assessment.• a comprehensive physical evaluation, including:

� a vision examination.� a measurement of postural blood pressure.� a targeted neurologic, musculoskeletal, and

cardiovascular examination.

Medication HistoryBefore a medication regimen can be assessed, clinicians

must have an accurate listing of all drugs used. Therefore, avitally important aspect of the medication evaluation is themedication history.

Patients in the community setting often take prescribeddrugs in a way that differs from the instructions that are listedin the medication profile in the medical record. It is helpfulwhen ambulatory patients bring their medications with themto clinic visits.

Clinicians should take care to ask patients about over-the-counter drugs and herbal products they are taking, becausepatients do not always think of these as “medications.”

Clinicans should also ask about adverse reactions to previ-ous medications and about perceived side effects of theircurrent prescriptions.

Inquiring about alcohol consumption is also important,because alcohol can contribute to the risk of falls.

As part of the medication history, physicians should askpatients whether certain drugs, particularly analgesics, havebeen effective in alleviating their symptoms. In our experience,many patients who have used drugs such as propoxyphene(e.g., Darvon®, NeoSan) and gabapentin (Neurontin®, Pfizer)for pain report minimal or no positive effects, but they continueto take such medications because a health care provider hasprescribed them. Furthermore, ineffective analgesia mayinhibit patient par ticipation in any prescribed physicalrehabilitation programs, such as balance and strength-trainingexercises.

Minimizing the Use of Drugs Associated with the Riskof Falls

Published practice guidelines recommend the following:

Patients who have fallen should have their medicationsreviewed and altered or stopped as appropriate in light of theirrisk of future falls. Particular attention to medication reduc-tion should be given to older persons taking four or moremedications and to those taking psychotropic drugs.4

Although psychotropic drugs (e.g., benzodiazepines, anti-depressants, antipsychotic agents) may be the initial focus ofa medication review, other drugs and drug classes may haveintended or unintended pharmacological effects that canincrease the risk of falling and thus might also be considered“high risk” in some cases.

Table 1 lists drugs and drug classes for clinicians to considerwhen they are assessing drug regimens in patients at risk forfalls. Others to be targeted for elimination or a reduction indose include the following:

• drugs that are ineffective• drugs that are thought to be causing adverse effects• drugs for which a therapeutic duplication might exist• drugs for which an indication is not known or

documented• drugs for which the dose seems to be high for an older

person

Medications and Falls in the Elderly

726 P&T® • November 2003 • Vol. 28 No. 11

Antidepressants† ‡Antipsychotics†Benzodiazepines†AntihypertensivesAntihistamines§Anticonvulsants†Nonsteroidal anti-inflammatory drugsCorticosteroidsMuscle relaxantsNarcotic analgesicsAntiarrhythmics (type IA†)Digoxin†NitratesHypoglycemicsAntiparkinson drugsHistamine H2-receptor blockers

* Not all drugs and drug classes listed have been associated with fallsin published research.Therefore, this list should be used in the contextof a comprehensive clinical assessment for each individual patient.

† Published research suggests an association between the use of thisdrug or drug class and an increased risk of falling.

‡ Includes selective serotonin reuptake inhibitors (SSRIs).§ Especially sedating antihistamines, such as diphenhydramine HCl

(e.g., Benadryl®, Pfizer) and hydroxyzine (e.g.,Atarax®, Pfizer).

Table 1 Drugs and Drug Classes to Consider in Evaluating Elderly Patients with an Increased Risk of Falling*

continued on page 733

Assessing Patients for Untreated ConditionsIn the overall patient assessment, clinicians should consider

instituting drug therapy when it might help to improve func-tional status, as in patients with untreated or poorly controlledconditions, including Parkinson’s disease, benign prostatichyperplasia, pain, and depression.

Deficiencies of some vitamins (e.g., B1227 and D28) and

hormones (e.g., testosterone deficiency in males29 and hypo-thyroidism30) are more common in elderly people and maycontribute to symptoms that interfere with functional status.Clinicians should consider screening for vitamin deficiencies,and supplementation may be needed in selected individuals.Patients with osteopenia and osteoporosis should receivecalcium and vitamin D as well as anti-resorption drugs whennecessary.

SUMMARYFalling is a common health problem in the elderly popula-

tion. The use of medications is one of the many different fac-tors that can contribute to balance problems and the risk offalls. Clinicians should make decisions about drug therapybased upon published research about falls and a knowledge ofdesirable and undesirable drug effects. Integral practice-basedcomponents of medication assessment include taking thepatient’s medication history, minimizing the use of high-riskdrugs, and managing uncontrolled diseases and disorders. Athorough assessment of medication regimens and skillful med-ication management can reduce the risk of falls in elderlypeople.

REFERENCES1. Rubenstein LZ, Josephson KR. The epidemiology of falls and

syncope. Clin Geriatr Med 2002;18:141–158.2. Campbell AJ, Borrie MJ, Spears GF. Risk factors for falls in a com-

munity-based prospective study of people 70 years and older.J Gerontol 1989;44:M112–M117.

3. Cumming RG. Falls and fractures in the elderly. Drugs Aging1998;12:43–53.

4. American Geriatrics Society, British Geriatrics Society, and Amer-ican Academy of Orthopaedic Surgeons Panel on Falls Preven-tion. Guideline for the prevention of falls in older persons. J AmGeriatr Soc 2001;49:664–672.

5. Rubenstein LZ, Powers CM, MacLean CH. Quality indicators forthe management and prevention of falls and mobility problems invulnerable elders. Ann Intern Med 2001;135:686–693.

6. Leipzig RM, Cumming RG, Tinetti ME. Drugs and falls in olderpeople: A systematic review and meta-analysis, Part I. Psycho-tropic drugs. J Am Geriatr Soc 1999;47:30–39.

7. Leipzig RM, Cumming RG, Tinetti ME. Drugs and falls in olderpeople: A systematic review and meta-analysis, Part II. Cardiacand analgesic drugs. J Am Geriatr Soc 1999;47:40–50.

8. Ray WA, Thapa PB, Gideon P. Benzodiazepines and the risk of fallsin nursing home residents. J Am Geriatr Soc 2000;48:682–685.

9. Mendelson WB. The use of sedative/hypnotic medication and its

correlation with falling down in the hospital. Sleep 1996;19:698–701.

10. Passaro A, Volpato S, Romagnoni F, et al. Benzodiazepines withdifferent half-lives and falling in a hospitalized population: TheGIFA study. J Clin Epidemiol 2000;53:1222–1229.

11. Ensrud KE, Blackwell TL, Mangione CM, et al. Central nervoussystem: Active medications and risk for falls in older women. J AmGeriatri Soc 2002;50:1629–1637.

12. Pierfitte C, Macouillard G, Thicoipe M, et al. Benzodiazepines andhip fractures in elderly people: Case-control study. BMJ 2001;322:704–708.

13. Wang PS, Bohn RL, Glynn RJ, et al. Hazardous benzodiazepineregimens in the elderly: Effects of half-life, dosage, and durationon risk of hip fracture. Am J Psychiatry 2001;158:892–898.

14. Ruthazer R, Lipsitz LA. Antidepressants and falls among elderlypeople in long-term care. Am J Public Health 1993;83:746–749.

15. Thapa PB, Gideon P, Cost TW, et al. Antidepressants and the riskof falls among nursing home residents. N Engl J Med 1998;339:875–882.

16. Mukai S, Lipsitz LA. Orthostatic hypotension. Clin Geriatr Med2002;18:253–268.

17. Heitterachi E, Lord SR, Meyerkort P, et al. Blood pressure chan-ges on upright tilting predict falls in older people. Age Ageing2002:31:181–186.

18. Liu BA, Topper AK, Reeves RA, et al. Falls among older peo-ple: Relationship to medication use and orthostatic hypo-tension. J Am Geriatr Soc 1995;43:1141–1145.

19. Curb JD, Applegate WB, Vogt TM, et al. Antihypertensive ther-apy and falls and fractures in the systolic hypertension in theelderly program. J Am Geriatr Soc 1993;41:SA15.

20. Tinetti ME, Baker DI, McAvay G, et al. A multifactorial inter-vention to reduce the risk of falling among elderly people livingin the community. N Engl J Med 1994;331(13):821–827.

21. Ray WA, Taylor JA, Meador KG, et al. A randomized trial of aconsultation service to reduce falls in nursing homes. JAMA 1997;278:557–562.

22. Campbell AJ, Robertson MC, Gardner MM, et al. Psychotropicmedication withdrawal and a home-based exercise program toprevent falls: A randomized, controlled trial. J Am Geriatr Soc1999;47:850–853.

23. Haumschild MJ, Karfonta TL, Haumschild MS, et al. Clinical andeconomic outcomes of a fall-focused pharmaceutical interventionprogram. Am J Health Syst Pharm 2003;60:1029–1032.

24. Alsop K, Mac Mahon M. Withdrawing cardiovascular medicationsat a syncope clinic. Postgrad Med J 2001;77:403–405.

25. Machuca SF, Jamison GT, Burlingame MB, et al. Medicationassessment by a multidisciplinary consultation service in elderly,ambulatory veterans at risk for falls (Abstract). Pharmacotherapy1999;19:514.

26. Tinetti ME. Preventing falls in elderly persons. N Engl J Med 2003;348:42–49.

27. Lindenbaum J, Rosenberg IH, Wilson PWF, et al. Prevalence ofcobalamin deficiency in the Framingham elderly population.Am J Clin Nutr 1994;60:2–11.

28. Dhesi JK, Moniz C, Close JCT, et al. A rationale for vitamin Dprescribing in a falls clinic population. Age Ageing 2002;31:267–271.

29. Basaria S, Dobs AS. Hypogonadism and androgen replacementtherapy in elderly men. Am J Med 2001;110:563–572.

30. Danese MD, Powe NR, Sawin CT. Screening for mild thyroidfailure at the periodic health examination: A decision and cost-effectiveness analysis. JAMA 1996;276:285–292.

Medications and Falls in the Elderly

Vol. 28 No. 11 • November 2003 • P&T® 733

continued from page 726