Surrounded by a sea of uninsured patients, a host of bureaucratic complexities, underfunded budgets, and a dearth of empirical evidence and best-practice guidelines,1,2 clinicians' decision making can be, to put it mildly, bizarre. Nowhere in the behavioral health field is this picture as apparent as in the case of long-acting (depot) formulations of antipsychotic medications.
Injectable depot formulations are used in the treatment of the symptoms of schizophrenia and other psychotic disorders. The available depot formulations include two older-generation agents, fluphenazine decanoate (Prolixin, 1967) and haloperidol decanoate (Haldol, 1988), and one second-generation antipsychotic, risperidone (Risperdal Consta, 2006). Injections are administered every two, three, or four weeks and are preferable over oral formulations for some patients because it is easier to receive one or two injections a month than to have to remember to take pills or liquids every day. With injections, clinicians can be certain that the patient is receiving the exact amount of medication prescribed. Such certainty can help assure adherence to the medication.
Nonadherence to antipsychotic medication has received increasing attention since the 1980s.3 It generally is agreed that roughly half of patients with schizophrenia do not adhere to their prescribed medication regimen.4,5 Studies have documented substantial cost increases associated with nonadherence among these patients. For example, the yearly cost in 1993 U.S. dollars of rehospitalization due to nonadherence was estimated to be $800 million.6
While there is no direct evidence that depot formulations by themselves improve adherence rates in comparison to oral formulations, it generally is accepted that the depot formulations offer a distinct advantage in the ability to document and observe nonadherent behavior.3,7 If the patient does not appear for the injection, he clearly is not following the treatment plan, and such clarity allows the clinician to modify the treatment plan and align with the patient to improve adherence rates.
So it would seem clinically logical to select nonadherent patients for depot formulations. If approximately 50% of the patients in this population are nonadherent, it would stand to reason that about 50% of antipsychotic prescriptions would be for depot formulations. But the paradox is that “depot antipsychotics play a relatively minor role in the treatment of schizophrenia,”8 and American clinicians appear to use these formulations less often than many non-U.S. counterparts.9 Indeed, a recent prospective, observational study of the treatment for schizophrenia in the United States conducted by Shi et al found that only 26% of patients (569/2,186) were treated with depot formulations of typical antipsychotics at least once during the designated three-year period.10
Why the Paradox?
Clues explaining the underutilization of depot formulations are emerging. Heres et al observed that in England, the presumed resistance among clinicians to the long-acting first-generation medications was explained by concerns about extrapyramidal (parkinsonian) side effects.8 Puzzled that the introduction of the long-acting form of risperidone, with less risk for extrapyramidal side effects, did not affect the frequency of use of depot formulations, these investigators surveyed 350 psychiatrists attending a national conference. Interestingly, the main factor for opposing both first- and second-generation antipsychotic depot formulations was the clinicians' assumption that their patients were, in fact, treatment adherent. Yet as the researchers point out, this assumption is “not supported by the current evidence.”
The study by Shi et al10 mentioned earlier is the most comprehensive study ever performed to characterize the patients selected for depot formulations. The study describes differences in patient characteristics and antipsychotic use patterns between those who received the two first-generation antipsychotic depot formulations and those who received oral formulations. Patients who received the depot formulations were more likely, in comparison to those on oral formulations, to be younger, male, African American, less educated, and more clinically and socially dysfunctional. Interestingly, during the one-year follow-up period, the adherence behavior (as measured by the medication possession ratio) of the patients treated with depot formulations improved.
Heres et al's and Shi et al's studies go a long way to explain the “bizarre” nature of the clinical decision-making process for long-acting antipsychotic formulations. Neither study shows any indication that clinicians use “nonadherence” as the selection criterion for depot formulations. In fact, Heres et al's study suggests that clinicians overestimate the degree to which their patients adhere to their medications, and none of the factors associated with depot use in Shi et al's study has been specifically associated with nonadherent behavior.