The incidence of polypharmacy is on the rise, and with the increase comes a greater risk of drug-drug reactions. One survey estimated that patients seeing a psychiatrist may be six times more likely to receive multiple psychotropic medicines compared to patients seen by a primary care physician. This article provides an overview of the extent of polypharmacy, the factors driving the phenomenon and issues clinicians should consider when treating patients who are already taking medicines for other illnesses.
Multiple medication use is the rule rather than exception in modern therapeutics. Factors affecting the recent increase in utilization of medications include the growth of third-party insurance coverage for drugs; increased marketing efforts to promote new medications to prescribers and directly to consumers; and clinical guidelines recommending long-term treatment for chronic conditions such as high cholesterol, acid reflux disease, heart disease, diabetes, asthma and clinical depression.
This issue is particularly important for those treating patients with psychiatric illness. According to the data from the 1989 National Ambulatory Medical Care Survey, patients seen by a psychiatrist were six times more likely to receive multiple psychotropic medications, as compared with those seen by a primary care physician (Nichol et al., 1995). A recent pharmacoepidemiology study found Veterans Affairs Administration outpatients on antidepressants were on more medications than age-matched and gender controls not on antidepressants (Preskorn, 2005).
Multiple medication use poses questions for clinicians: What is the extent of polypharmacy in general practice and psychiatry? What factors impact the increase in polypharmacy seen today? What should the clinician consider when treating a patient taking two or more medications concurrently?
The Incidence of Polypharmacy
According to the annual report, "Health, United States, 2004," prescription drug use is rising among people of all ages, and increases with age (National Center for Health Statistics, 2004). Almost half of those over 18 in the United States are taking at least one prescription medicine, and one in six is taking three or more. These percentages are substantially higher for Americans 65 or older (Table).
The reason is, at least in part, the increasing ability to treat and even prevent chronic illnesses. As a result, patients accumulate drugs as they age and are on them for years, during which time other drugs can be added and stopped. Prescription drug use is on the rise, and there are an increasing number of medications available for the chronic treatment of hypertension, diabetes and hyperlipidemia, the three leading causes of death in United States (National Center for Health Statistics, 2004).
The National Center for Health Statistics (2004) reported a number of data regarding the increasing incidence of polypharmacy. Among its findings: The use of multiple psychoactive drugs is an increasingly common practice in both primary care and psychiatric practice; the use of antidepressants in the U.S. adult population almost tripled between 1988 to 1994 and 1999 to 2000; 10% of women age 18 or older and 4% of men now take antidepressants; three times as many white adults in the United States took antidepressants as did blacks or Mexican Americans in the most recent year for which data are available. There is also an increased use of specific psychoactive medications in younger patients, and use is also increasing among younger patients including psychoactive medication for attention-deficit/hyperactivity disorder and antidepressants (National Center for Health Statistics, 2004).
Factors Affecting Polypharmacy
A number of risk factors in primary care and psychiatric practice may lead to the increasing incidence of multiple medication use.
Comorbid medical and psychiatric conditions. Patients with psychiatric disorders may have significant comorbidity with medical conditions including HIV infection, heart disease, diabetes, cancer, fibromyalgia, irritable bowel syndrome, liver disease, renal failure and many more. So overall, psychiatric illness increases a person's risk of developing a number of general medical conditions (Goldman, 2000). The Figure shows the prevalence of major depression in specific medically ill populations.
Patients with mood and anxiety disorders have also been found to utilize more health care services than those without these disorders and thus to have higher incidence and greater complexity of concomitant medication usage. Patients with substance abuse or dependence are another group at increased risk for drug-drug interactions (DDIs) resulting from polypharmacy (George and Krystal, 2000).
Increase in number of approved psychoactive drugs. With more drugs approved for use by the U.S. Food and Drug Administration, there is more potential for drug interactions. Frye et al. (2000) found a substantial increase in polypsychopharmacy from 1974 to 1996. Since 1990, the FDA has approved more than 20 new psychotropics, including antidepressants, antipsychotics, and medications for bipolar disorder, dementia, insomnia, sexual dysfunction, narcolepsy and ADHD (Table 2).
Increase in use of nonprescription drugs. Many well-educated patients increasingly turn to alternative medicine because they find such therapies more congruent with personal values and holistic views of health care (Astin, 1998). A wealth of such alternative medicines is available, including prescription drugs, over-the-counter (OTC) medicines, herbal preparations, illicit substances and dietary supplements. Adverse drug reactions stemming from concurrent use of such alternative medications with prescription drugs can present in almost any way clinically imaginable, from a sudden, catastrophic adverse event to a mild increase in nuisance side effects. Drug-drug interactions can also present as a loss of efficacy on a previously effective drug regimen, withdrawal symptoms, emergence of a new illness or worsening of an existing one.
For example, ibuprofen, an OTC analgesic, can cause serious and even life-threatening elevations in lithium (Eskalith, Lithobid) levels by affecting its rate of tubular reabsorption (Ragheb et al., 1980). St. John's wort (Hypericum perforatum) is a substantial inducer of CYP3A and can accelerate the clearance of oral contraceptives, potentially rendering them ineffective (Hall et al., 2003). Smoking can induce the metabolism of drugs such as clozapine (Clozaril), which are normally cleared by smoking-induced CYP1A2 (Hasegawa et al., 1993). In addition, a case report describes clozapine-induced seizures in a man when he gave up smoking (McCarthy, 1994).
Increased Internet availability. It is now common for patients to buy medications on the Internet for general medical and psychiatric disorders. In 2004, the FDA reiterated its warning against online drug purchases after it purchased so-called generic versions of sildenafil (Viagra), atorvastatin (Lipitor) and zolpidem (Ambien) from a Canadian-advertised Web site. The FDA reported that all three were "fake, substandard and potentially dangerous" (Rados, 2004). Clinicians should discourage this method of obtaining medications until approved by the FDA.
Syndromic nature of psychiatric disorders. Common psychiatric conditions such as mood, psychotic and anxiety disorders usually have multiple signs and symptoms involving more than one bodily system, treatment of which can lead to multiple medication use. Nichol et al. (1995) found that patients diagnosed with mania were four times more likely to receive multiple psychotropic medications, and those diagnosed with schizophrenia were three times more likely. There is also a substantial overlap in the diagnostic and ancillary features outlined in DSM-IV for bipolar disorder, schizoaffective disorder and borderline personality disorder (Preskorn and Baker, 2002). A patient with borderline personality disorder may be treated with lithium for mood lability, an antidepressant for depressive symptoms, an anxiolytic for anxiety symptoms and an antipsychotic for brief psychotic symptoms.
Treatment of side effects. Treatment for a primary illness or disorder sometimes produces side effects that must be treated by the use of adjunctive pharmacotherapy. For example, benztropine (Cogentin) may be used to treat the extrapyramidal symptoms of antipsychotics. Antinausea medication may be given to a patient receiving cancer chemotherapy. Increasingly common is the treatment of metabolic syndrome (i.e., hyperglycemia, hyperlipidemia, diabetes and hypertension) resulting from treatment with some atypical antipsychotics (Toalson et al., 2004).
Increased prescription medicine coverage. Private health insurance covered almost 50% of prescription drug costs in 2002, as compared to 25% in 1990 (National Center for Health Statistics, 2004). When cost is not an issue, use goes up.
(While the percentage of prescription drugs has increased, the amount of shared costs shifted to patients through co-payments to third-party health insurance providers has also seen a steady increase. The jury is still out on how this impacts the cost of care. For example, many formularies disallow newer medications, while providing increased coverage for older drugs. At least in the case of antipsychotics, the side-effect burden of older drugs makes their use less preferred. Furthermore, some studies have shown that even a modest increase in co-payments can have a significant impact on patient use of prescription medications. The interested reader may want to see Nelson et al.  and Stuart and Zacker --Ed.)
Lack of communication. In the absence of a primary provider to coordinate drug therapy, multiple doctors often prescribe prescription drugs to the same patient (Colley and Lucas, 1993; Rolland and Verdaris, 2003).
Decreased use of behavioral and social techniques. Primary care physicians may never have been trained in psychosocial intervention for patients suffering from psychiatric disorders and frequently they do not have time to do them. Mintz et al. (2004) found that psychiatrists practicing under utilization-management protocols of managed care plans may also not have time to discuss behavioral techniques with their patients (for example, encouraging proper sleep hygiene in patients complaining of insomnia) and often instead prescribe a medication. Once a medication is prescribed, the psychiatrist may be reluctant to take their patient off the medications, especially if they are doing well (Mintz et al., 2004).
Aging. Older patients take more medications in general, and this situation is compounded if patients have a psychiatric illness because adverse DDIs involving these medications can mimic psychiatric symptoms. For example, geriatric patients frequently are exposed to a number of highly anticholinergic drugs including antispasmodics, tricyclic antidepressants (e.g., amitriptyline), antihistamines, anti-Parkinson's drugs, antipsychotics and OTC cold medicines, which if used in combination with other drugs can produce delirium (Beresin, 1988; Cadieux, 1989). Antihypertensives are the most likely to induce a depression in the elderly. Common offenders are methyldopa (Aldomet, Amodopa), propranolol (Inderal), clonidine (Catapress), guanethidine (Ismelin) and reserpine (Serpalan, Serpasil) (Marsh, 1997). The akathisia of antipsychotic agents or the selective serotonin reuptake inhibitors can resemble anxiety in the elderly (Marsh, 1997). Other commonly prescribed drugs that can produce anxiety symptoms in elderly patients are bronchodilators, calcium-channel blockers, steroids, OTC preparations with adrenergic-like agents and antihistaminic agents (Cadieux, 1989; Marsh, 1997).
Polypharmacy is associated with increased health risks and costs (Preskorn, 2005; Silkey et al., 2005), and can result in potential pharmacokinetically and/or pharmacodynamically mediated DDIs. These, in turn, set the stage for increased vulnerability to side effects, increased number of prescription drugs used, compliance problems, increased patient morbidity and mortality, and increased health care costs. Clinicians must keep in mind a number of factors.
Potential DDIs. A DDI is a measurable change in magnitude, nature or duration of the action of one drug as a result of the presence of another drug (Khan and Preskorn, 2003). Drug-drug interactions are not limited to or determined by therapeutic class. Individual patient response to a given combination of drugs may be influenced by three main factors: pharmacodynamics, pharmacokinetics and biological variability between patients. Pharmacodynamics refers to a drug's affinity for the site of action in the body, or its biochemical and physiological effect and mechanism of action. Pharmackinetics refers to the concentration of the drug at the site of action, or its action in the body over a period of time, including the processes of absorption, distribution, localization in tissues, biotransformation and excretion. Biological variability refers to the fact that a given drug may not produce identical effects in different patients and may be influenced by such factors as the patient's genotype, age, environment and the presence of disease. The presence of other drugs can quantitatively or qualitatively alter the patient's response to a medication by altering its pharmacodynamic or pharmacokinetic properties. The clinician should weigh these factors carefully when deciding on the treatment for their patient, including what drug to prescribe and at what doses.
Intestinal motility. Drug absorption, distribution, metabolism and excretion can be altered by changes in gastric pH or gastrointestinal motility, including gastric emptying. Examples include pyloric stenosis, gastroparesis secondary to diabetes mellitus, systemic sclerosis, postvagotomy states, celiac disease, Whipple's disease, Crohn's disease and therapy with anticholinergic agents. Also gastric and intestinal surgeries and/or radiation will affect absorption of drugs. Likewise, lower gastric acidity produced by proton pump inhibitors like omeprazole (Prilosec), or by histamine H2 receptor blockers like cimetidine (Tagamet), or by antacids can increase absorption of weak bases such as tricyclic antidepressants, benzodiazepines and some antipsychotics (Janicak et al., 2001; Khan and Preskorn, 2003).
Liver disease. Most psychoactive medications require oxidative metabolism as a necessary step in their eventual elimination, which principally occurs in the liver. For this reason, diseases that lead to hepatic insufficiency (like cirrhosis, viral infections, collagen vascular diseases and metabolic disorders) can significantly reduce the clearance of drugs. This effect can be compounded by the concurrent use of multiple medications. For example, diazepam (Valium) requires extensive biotransformation as a necessary step in its elimination, but can persist in the body in case of liver disease (Janicak et al., 2001). On the other hand, the clearance of drugs that undergo only glucuronidation is not affected by even significant liver damage. For example, all of the 3-hydroxybenzodiazepines (clonazepam [Klonopin], lorazepam [Ativan], oxazepam [Serax] and temazepam [Restoril]) can be readily cleared by patients with normal renal function (Kahn and Preskorn, 2003).
Liver diseases can alter plasma proteins that may significantly affect free drug concentration and distribution and hence concentration at the site of action, thus altering the magnitude of the drug's effect. As the majority of psychiatric medications are highly protein bound, protein binding can affect therapeutic drug monitoring results.
Renal disease. While most drugs require oxidative metabolism to be converted into polar metabolites as a necessary step in their elimination, the final route of elimination is via the kidneys. Classic examples of psychiatric drugs cleared principally, if not exclusively, via the kidneys include lithium and gabapentin (Neurontin), according to the PDR (2004). Slow dose titration and careful monitoring of blood levels may be necessary when prescribing medications dependent on renal filtration in patients with significant renal disease.
Medications used for physical/medical conditions can also affect psychiatric medications. For example, ritonavir (Norvir), a protease inhibitor used in the treatment of HIV and AIDS, is a significant inhibitor of CYP3A4. This means it can increase levels of certain heterocyclic antidepressants (e.g., imipramine [Tofranil], clomipramine [Anafranil] and amitriptyline), triazolobenzodiazepines (e.g., alprazolam [Xanax] and triazolam [Halcion]), zolpidem (Ambien), methadone and clozapine if taken concurrently (Flexner, 1998). Anti-migraine drugs (e.g., sumatriptan [Imitrex], zolmitriptan [Zomig] and rizatriptan [Maxalt]) are all 5-HT1 agonists and can produce serotonin syndrome if used in combination with SSRIs like sertraline (Zoloft), and monoamine oxidase inhibitors like phenelzine (Nardil). Itraconazole (Sporanox), an antifungal agent, is also a potent inhibitor of CYP3A4 and can increase levels of buspirone (BuSpar) and haloperidol (Haldol) (Goldman, 2000). Rifampin (Rifadin, Rimactane), an anti-tuberculosis agent, induces CYP3A4 and CYP1A2, which can lead to decreased levels of clozapine, buspirone and zaleplon (Sonata) when taken concurrently with any one of them (Goldman, 2000).
Older patients.Elderly patients are particularly susceptible to the effects of polypharmacy as aging is associated with diminution in the functional capacity of many organ systems including the cardiovascular system, the liver, the kidneys and the brain. The latter includes mental slowing due the development of numerous neuronal degenerative diseases, particularly including Alzheimer's disease. For these reasons clinicians should be familiar with the physical status of elderly patients before prescribing any psychoactive medications.
Noncompliance. Noncompliance is common when the patient is taking one medication and even more prevalent when several medications are taken concurrently. Col et al. (1990) found that the complexity of medication regimen (as reflected in greater number of doses per day and greater number of both scheduled and as needed medications) was directly related with an increased risk of noncompliance, contributed to a more frequent need for hospitalization.
Fatal outcomes. Fatal drug reactions can result from the combined ingestion of tranylcypromine (Parnatte) and imipramine, resulting in hyperpyrexia, excruciating headache, palpitations, muscle rigidity, opisthotonus, cyanosis and convulsions (Otte et al., 2003). Combination of MAOIs and meperidine (Demerol) can result in excitement, muscle rigidity, hyperpyrexia, flushing, sweating, unconsciousness and respiratory depression (Meyer and Halfin, 1981; Miller, 2004). Another fatal combination can be MAOIs and SSRIs, e.g., fluoxetine (Prozac). It can result in the development of serotonin syndrome with chills, confusion, sedation, exhaustion, hyperpyrexia, tachycardia, muscle rigidity and coma (Feighner et al., 1990). Pimozide (Orap) is partly metabolized through CYP3A and clarithromycin (Biaxin) substantially inhibits this enzyme, resulting in serious cardiotoxicity.
Health care cost. Increased costs can result from the costs of each individual drug in the patient's regimen and from DDIs. These DDIs can present in a multitude of ways ranging from nuisance to more serious adverse effects, which can mimic the development of a new disease or the worsening of an existing disease. DDIs can of course result in lawsuits, which further increase health care costs. Johnson and Bootman (1995) found that morbidity and mortality related to drug therapy in ambulatory patients in the United States costs $76.6 billion annually.
Although prescription drugs comprise only one-tenth of total national medical expenses, they remain the fastest growing expenditure (National Center for Health Statistics, 2004). In 2002, the overall cost of prescription medications and medical supplies rose above 5%, but increased use of prescription drugs pushed total expenditure for prescription medications up 15.3%. The United States spent 14.9% of its gross domestic product on health care in 2002, up from 14.1% in 2001. Only two other countries, Switzerland and Germany, spent as much as 11% of their gross domestic produce in health care in 2001.
Patients with psychiatric illnesses may be at higher risk for the development of certain medical problems because of the physiologic derangements of their psychiatric conditions, or as a result of the somatic treatment used to help them, hence at risk for drug-drug reactions from taking several medications concurrently. Psychopharmacological treatment of psychiatric disorders can be successfully and safely accomplished in the medically ill taking multiple medications when the psychiatrist is cognizant of potential difficulties stemming from both their different medical illnesses and the medications the patient is taking for the treatments of those illnesses by following the principles outlined in (Table 3). Each patient should be assessed for potential disease-related pharmacokinetic and pharmacodynamic changes altering drug distribution, metabolism and clearance, as well as possible drug interactions and vulnerability to the side effects.
As the use of multiple drugs increases and drug expenditures rise, trade-offs between the cost and benefits of medications are becoming major clinical and policy issues. As a result of all of these variables, there is a growing and urgent need for further research to evaluate the potential risks associated with polypharmacy, such as developing drug classifications or groupings based on pharmacodynamic and pharmacokinetic characteristics. Research can identify the most prevalent drug regimens and evaluate their potential to interact. There is a clear need to implement online drug screening or computerized drug alert systems. Expert consensus guidelines regarding specific combinations can be developed.
Dr. Khan is assistant professor in the department of psychiatry and behavioral sciences at the University of Kansas School of Medicine in Wichita and director of the resident outpatient clinic.
Dr. Preskorn is professor and chair for the department of psychiatry and behavioral sciences at the University of Kansas School of Medicine in Wichita. He is also CEO and president of the Clinical Research Institute, which is affiliated with the University of Kansas School of Medicine.
Astin JA (1998), Why patients use alternative medicine: results of a national study. JAMA 279(19):1548-1553 [see comments].
Beresin EV (1988), Delirium in the elderly. J Geriatr Psychiatry Neurol 1(3):127-143.
Cadieux RJ (1989), Drug interactions in the elderly. How multiple drug use increases risk. Postgrad Med 86(8):179-186.
Col N, Fanale JE, Kronholm P (1990), The role of medication noncompliance and adverse drug reactions in hospitalizations of the elderly. Arch Intern Med 150(4):841-845.
Colley CA, Lucas LM (1993), Polypharmacy: the cure becomes the disease. J Gen Intern Med 8(5):278-283.
Feighner JP, Boyer WF, Tyler DL, Neborsky RU (1990), Adverse consequences of fluoxetine-MAOI combination therapy. J Clin Psychiatry 51(6): 222-225 [see comment].
Flexner C (1998), HIV-protease inhibitors. New Engl J Med 338(18):1281-1292 [see comment].
Frye MA, Ketter TA, Leverich GS et al. (2000), The increasing use of polypharmacotherapy for refractory mood disorders: 22 years of study. J Clin Psychiatry 61(1):9-15.
George TP, Krystal JH (2000), Comorbidity of psychiatric and substance abuse disorders. Curr Opin Psychiatry 13(3):327-331.
Goldman LS (2000), Comorbid medical illness in psychiatric patients. Curr Psychiatry Rep 2(3):256-263.
Hall SD, Wang Z, Huang SM et al. (2003), The interaction between St John's wort and an oral contraceptive. Clin Pharmacol Ther 74(6):525-535.
Hasegawa M, Gutierrez-Esteinou R, Way L, Meltzer HY (1993), Relationship between clinical efficacy and clozapine concentrations in plasma in schizophrenia: effect of smoking. J Clin Psychopharmacology 13(6):383-390.
Janicak PG, Davis JM, Preskorn SH, Ayd FJ (2001), Principles and Practice of Psychopharmacotherapy, 3rd ed. Philadelphia: Lippincott Williams & Wilkins, pp51-68.
Johnson JA, Bootman JL (1995), Drug-related morbidity and mortality. A cost-of-illness model. Arch Intern Med 155(18):1949-1956 [see comment].
Khan AY, Preskorn S (2003), Handbook of Medical Psychiatry. New York: Dekker, pp933-944.
Marsh CM (1997), Psychiatric presentations of medical illness. Psychiatr Clin North Am 20(1):181-204.
McCarthy RH (1994), Seizures following smoking cessation in a clozapine responder. Pharmacopsychiatry 27(5):210-221.
Meyer D, Halfin V (1981), Toxicity secondary to meperidine in patients on monoamine oxidase inhibitors: a case report and critical review. J Clin Psychopharmacol 1(5):319-321.
Miller J (2004), Managing antidepression overdoses. Emerg Med Serv 33(10):113-119.
Mintz DC, Marcus SC, Druss BG et al. (2004), Association of utilization management and treatment plan modifications among practicing U.S. psychiatrists. Am J Psychiatry 161(6):1103-1109.
National Center for Health Statistics (2004), Health, United States, 2004. With chartbook on trends in the health of Americans. Washington, D.C.: U.S. Government Printing Office. Available at: www.cdc.gov/nchs/hus.htm. Accessed Aug. 19, 2005.
Nelson AA Jr, Reeder CE, Dickson WM (1984), The effect of a Medicaid drug copayment program on the utilization and cost of prescription services. Med Care 22(8):724-736.
Nichol MB, Stimmel GL, Lange SC (1995), Factors predicting the use of multiple psychotropic medications. J Clin Psychiatry 56(2):60-66.
Otte W, Birkenhager TK, van den Broek WW (2003), Fatal interaction between tranylcypromine and imipramine. Eur Psychiatry 18(5):264-265.
PDR (2004), Montvale, N.J.: Medical Economics Co. Inc. Preskorn SH (2005), Multiple medication use in patients seen in the veteran affairs healthcare system: so what? J Psychiatr Pract 11(1):46-50.
Preskorn SH, Baker B (2002), The overlap of DSM-IV syndromes: potential implications for the practice of polypsychopharmacology, psychiatric drug development, and the human genome project. J Psychiatr Pract 8(3):170-177.
Rados C (2004), FDA reiterates warning against online drug buying. FDA Consumer Magazine. Available at: www.fda.gov/fdac/features/2004/504_drug.html. Accessed Aug. 22, 2005.
Ragheb M, Ban TA, Buchanan D, Frolich JC (1980), Interaction of indomethacin and ibuprofen with lithium in manic patients under a steady-state lithium level. J Clin Psychiatry 41(11):397-398.
Rolland P, Verdaris M (2003), Case report: multiple physicians, polypharmacy, and patient counseling. J Am Pharm Assoc (Wash DC) 43(5):550 [letter].
Silkey B, Preskorn SH, Golbeck A et al. (2005), Complexity of medication use in the Veterans Affairs healthcare system: Part II. Antidepressant use among younger and older outpatients. J Psychiatr Pract 11(1):16-26.
Stuart B, Zacker C (1999), Who bears the burden of Medicaid drug copayment policies? Health Affairs 18(2):201-212.
Toalson P, Ahmed S, Hardy T, Kabinoff G (2004), The metabolic syndrome in patients with severe mental illnesses. Prim Care Companion J Clin Psychiatry 6(4):152-157.
Levy RH, Thummel KE, Trager WF et al. (2000), Metabolic Drug Interactions. Philadelphia: Lippincott Williams & Wilkins.