In examining the demographics of a state Medicaid population, we found that members of ethnic minorities, rural residents, women, and patients aged 45 years and older were more likely to receive first-generation antipsychotic medications than the newer, second-generation agents, which have a more favorable side-effect profile (Fehr et al., 2000).
These findings are noteworthy (all findings were statistically significant, p0.001) because older, first-generation antipsychotic medications have serious side effects, including tardive dyskinesia, for which African-Americans, the elderly and women seem to be most at risk (Arana, 2000).
Our study compared recipients of the first-generation drugs haloperidol (Haldol), perphenazine (Trilafon) and thioridazine (Mellaril) (n=12,012) with those who received the second-generation agents clozapine (Clozaril), olanzapine (Zyprexa), and risperidone (Risperdal) (n=10,157). Focus was placed on these drugs because they accounted for the largest proportion of antipsychotic prescriptions in this population. All study subjects resided in the same state and were enrolled in Medicaid from Jan. 1, 1996, through June 30, 1997.
Because African-Americans are twice as likely to develop tardive dyskinesia as Caucasians (Glazer, 2000a), we expected that this group would receive more second-generation agents. Instead, the data showed the opposite to be true-only 38% of African-Americans received the newer medications, compared with 50% of Caucasians (Figure 1).
Age also proved to be a factor (Figure 2). The percentage of patients receiving newer agents ranged from 51% to 58% for those up to 44 years of age, while for patients aged 45 to 65 years and older, only 36% to 43% received the second-generation medications. These disparities are significant because patients over age 50 who are initiating treatment with first-generation agents are three to five times more likely to develop tardive dyskinesia than younger patients (Woerner et al., 1998). Moreover, the risk rises with age and length of exposure to first-generation neuroleptic agents (Jeste, 2000).
In addition, the elderly tend to develop tardive dyskinesia sooner--with up to a 63% incidence after three years of cumulative exposure, compared to a 4% to 5% incidence in young, healthy adults--and while taking lower doses of medication (Conley, 2000; Jeste, 2000; Masand, 2000).
Gender disparities also existed, although they were not as pronounced. Of the patients in our study who received second-generation antipsychotics, 44% were females, compared with 48% of males. In general, females are considered to be at higher risk for tardive dyskinesia than males (Yassa and Jeste, 1992).
In our study, only 38% of rural residents received the newer agents, compared with nearly 47% of urban dwellers. We can only surmise the reasons for this disparity, but it may be due to urban residents' easier access to large, teaching hospitals and medical centers where new pharmaceutical agents may be adopted more quickly. It is also possible that, in the state we studied, some ethnic minorities and older patients may be more likely to live in rural areas.
The fact that this research study was limited to Medicaid recipients should not have affected the types of drugs prescribed, because prescriptions are generally 100% covered under Medicaid, with no copayment by the patient. In addition, the study drugs were all covered by the state Medicaid formulary, and only 5% of the prescription claims were for Medicaid HMO patients. We feel that this small percentage of HMO claims did not unduly weight the results toward the less expensive, first-generation agents.>
Advantages of Second-Generation Antipsychotics
According to Glazer (2000b), the improved side-effect profile of second-generation antipsychotic medications "points to a lower risk for development of tardive dyskinesia." Even more importantly, Glazer concluded, "The majority of cases of tardive dyskinesia are generally accepted to be iatrogenic in origin." Therefore, it is important to consider replacing first-generation antipsychotic treatment with the newer drugs whenever feasible.
"Since no treatment has proved effective," Saltz et al. (2000) claimed, prevention is the only strategy in the management of tardive dyskinesia, particularly for the elderly.
Although the newer agents have been designed to minimize extrapyramidal symptoms, they do have other side effects that must be considered. For example, despite its superior efficacy and minimal extrapyramidal side effects, clozapine is associated with agranulocytosis. Therefore, its use must be accompanied by a biweekly monitoring program (Miller, 2000).
Risperidone is reported to lack anticholinergic effects, but it is associated with an average weight gain of 2.10 kg after 10 weeks of treatment (Conley, 2000). A three-year study by Guitierrez-Esteinour and Grebb (as cited in Conley, 2000) reported a 0.3% annual incidence of tardive dyskinesia for patients taking a daily dose of 7.6 mg to 9.4 mg of risperidone for at least one year. Increased prolactin levels have also been reported for patients taking risperidone, but there appeared to be little correlation with adverse events. Compared with first-generation agents, olanzapine has only a few adverse effects, of which the most serious is increased appetite, often leading to weight gain (Conley and Meltzer, 2000).
We are aware of small-scale studies that indicate olanzapine and clozapine, but not risperidone, have been associated with new-onset diabetes. However, these studies may not have considered factors that may affect the outcomes, such as age discrepancies, dosage levels and duration of therapy.
Researchers at AdvancePCS are conducting a large-scale, test-control study, with a test group of 58,750 patients taking antipsychotics and a control group of 5.8 million patients, that does take these and other variables into account. Our preliminary findings do not support the conclusions of these earlier studies.
Because quetiapine (Seroquel) and ziprasidone (Geodon) are two of the newest second-generation antipsychotic drugs, they were not included in our study. However, quetiapine's efficacy is comparable to first-generation agents, and it is said to have a benign side-effect profile, with low rates of extrapyramidal symptoms and no reports of sustained hyperprolactinemia. Substantial weight gain, however, has been reported in up to 23% of patients (Garver, 2000).
Newer Antipsychotics: Cost Versus Efficacy
Although the newer drugs are much more expensive than their first-generation predecessors, evidence is accumulating that overall medical costs may decrease when the second-generation agents are prescribed (Buckley, 1999).
The opinion of well over 90% of the participants in the 1997 National Consensus Conference on the Pharmacoeconomics of Treatment-Resistant Schizophrenia was, "starting the most powerful new medicines as soon as possible, regardless of daily cost," is the way large mental health care delivery systems can save the most money when managing serious mental illness (National Consensus Conference on the Pharmacoeconomics of Treatment-Resistant Schizophrenia, 1999).
The same group of physicians and health care administrators also agreed, by a 75% majority that recidivism is the most important factor in mental illness service costs. The increased side effects of first-generation antipsychotics can adversely affect treatment compliance, and noncompliance may lead to recidivism.
According to Robert R. Conley, M.D., "The difference in recidivism rates between patients taking typical [first-generation] antipsychotics and those taking novel [second-generation] antipsychotics is remarkable. These improved outcomes are due largely to compliance. Persistent medication compliance is a phenomenon rarely witnessed with patients taking typical antipsychotics" (Arana et al., 2000a).
The newer antipsychotics, with their increased efficacy and improved side-effect profiles, have greatly improved the treatment outcomes for many patients. As a result, Noordsy et al. (2000) have called for "a reexamination of our expectations for the outcomes of treatment of psychotic disordersto incorporate the substantial advances in our field and to ensure that we do not unknowingly contribute to artificial ceilings that block patient recovery."
In a discussion of side effects caused by atypical antipsychotics (Arana et al., 2000b), Herbert Y. Meltzer, M.D., stated, "Ethical issues have been raised around simply exposing patients to the risk of tardive dyskinesia with the use of conventional [first-generation] antipsychotics. In a 60-year-old patient, there is a 50% risk of tardive dyskinesia in 3 years, and there is a 25% risk in a 20-year-old." In the same discussion, Prakash S. Masand, M.D., concurred, "The risk of tardive dyskinesia is probably the most important reason not to expose patients to conventional neuroleptics."
If this is true for the population at large, it may be even more compelling for demographic groups at higher-than-average risk for tardive dyskinesia-African-Americans, the elderly and women. Second-generation antipsychotics should be considered the treatment of choice for psychosis (Arana et al., 2000b).
The authors would like to thank David S. Hutchins, M.B.A., M.H.S.A., and William F. Signa Sr., B.S., for their work designing and executing this research study.
1.Arana GW (2000), An overview of side effects caused by typical antipsychotics. J Clin Psychiatry 61(suppl 8):5-11.
2.Arana GW, Blackburn GL, Conley RR et al. (2000a), An overview of side effects caused by typical antipsychotics. J Clin Psychiatry 61(suppl 8):12-13 [discussion].
3.Arana GW, Blackburn GL, Conley RR (2000b), Side effects of antipsychotic medications: physician's choice of medication and patient compliance. J Clin Psychiatry 61(suppl 8):64-66 [conclusions].
4.Buckley PF (1999), New antipsychotic agents: emerging clinical profiles. J Clin Psychiatry 60(suppl 1):12-17 [see discussion pp28-30].
5.Conley RR (2000), Risperidone side effects. J Clin Psychiatry 61(suppl 8):20-23 [see discussion pp24-24].
6.Conley RR, Meltzer HY (2000), Adverse events related to olanzapine. J Clin Psychiatry 61(suppl 8):26-29 [see discussion p30].
7.Fehr AM, Hutchins DS, Signa WF (2000), Disparate distribution of first-generation antipsychotic medications in a Medicaid population. Poster presented at the 128th Annual Meeting of the American Public Health Association. Boston, Nov. 13.
8.Garver DL (2000), Review of quetiapine side effects. J Clin Psychiatry 61(suppl 8):31-33 [see discussion pp34-35].
9.Glazer WM (2000a), Review of incidence studies of tardive dyskinesia associated with typical antipsychotics. J Clin Psychiatry 61(suppl 4):15-20.
10.Glazer WM (2000b), Expected incidence of tardive dyskinesia associated with atypical antipsychotics. J Clin Psychiatry 61(suppl 4):21-26.
11.Jeste DV (2000), Tardive dyskinesia in older patients. J Clin Psychiatry 61(suppl 4):27-32.