Becky McCall

May 02, 2012

May 2, 2012 (Florence, Italy) — Patients are less likely to discontinue long-acting injection (LAI) antipsychotics or clozapine than oral antipsychotic drugs, according to results presented here at the 3rd Biennial Schizophrenia International Research Society (SIRS) Conference.

Rune A. Kroken, MD, from Haukeland University Hospital in Bergen, Norway, and colleagues compared the rate of discontinuation of common antipsychotics in a real-world study.

They found an 83% lower risk for all-cause discontinuation of clozapine than for oral olanzapine (Zyprexa, Eli Lilly and Co.) treatment. "Patient adherence with clozapine and the LAI antipsychotics is much better than with oral olanzapine, which has been considered the most efficacious drug, according to the 2005 Clinical Antipsychotic Trials of Intervention Effectiveness (CATIE) study," Dr. Kroken said.

He explained that despite the availability of effective drugs, patients with chronic schizophrenia tend to discontinue their treatments, leading to hospital readmission. In this study, Dr. Kroken and his colleagues investigated the length of time from hospital discharge to discontinuation of antipsychotic drugs in patients with schizophrenia.

He was curious that little difference was found in the times to discontinuation for these drugs in randomized controlled trials (RCTs). "I wondered whether this was because of selective recruitment for RCTs," he told Medscape Medical News. "We wanted to conduct a complete cohort study with patients discharged from our hospital and follow them until they discontinued their treatment."

According to Dr. Kroken, many patients with schizophrenia discontinue their drugs because of a combination of lack of insight into their illness and the adverse effects that outweigh the beneficial effects. "Many treatments for schizophrenia cause weight gain and neurologic side effects. These patients are also very severely ill; most patients in this study were admitted involuntarily in the acute setting."

The researchers identified 396 patients with schizophrenia who had been discharged from 2005 to 2010. Of these, 289 patients received antipsychotic monotherapy. Data from these patients were analyzed by multivariate Cox regression. Time to discontinuation, categorized as clinician-decided, patient-decided, or unknown, was the primary end point. Oral olanzapine was used as the reference drug against which other treatments were compared.

Data were controlled for covariates such as age, sex, number of previous hospital admissions, treatment after discharge, and drug or alcohol issues. "A strength of the study is the use of time-dependent covariates," such as outpatient and inpatient treatment in community mental health centers and involuntary outpatient treatment, Dr. Kroken pointed out.

The researchers found that time to discontinuation of clozapine was much longer than time to discontinuation of olanzapine (adjusted hazard ratio [AHR], 0.17; P < .001 for all-cause discontinuation). "Clozapine gives a strong result for patient-decision, clinician-decision, and all-cause discontinuations," Dr. Kroken reported.

He noted that patient-decided discontinuations were probably the most important findings because of the problems with drug adherence in patients with schizophrenia. LAIs were at lower risk for discontinuation. Compared with olanzapine, patient-decided discontinuation of first-generation antipsychotic LAIs had an AHR of 0.35 (P = .013) and of second-generation antipsychotic LAIs had an AHR 0.26 (P = .015).

"We showed that patients adhere to clozapine and LAIs far more than oral olanzapine," he said. "If a clinician puts a patient on clozapine, they tend to stay on it because there isn't a better alternative, but I'm quite surprised by the number of patients who decide to stay on it."

Another reason patients might stay on clozapine is the close patient-clinician relationship related to the requirement for regular blood tests, suggested Dr. Kroken. "Perhaps all patients on antipsychotics should go to their doctor once a month and discuss how it is going. Many patients are put on antipsychotics and are reviewed once or twice a year."

Dr. Kroken said they saw differences between drugs that could not be seen in an RCT because of high attrition rates, selective recruitment, and associated problems. "I think these results reflect the real-world discontinuation rates."

Mode of administration also appears to play a role, according to Dr. Kroken.

"I think LAIs are good, especially for severely ill patients with schizophrenia. If it was another field of medicine, the news that it can be taken once a month would be welcome, but in psychiatry, there is some irrationality in thinking about LAIs. We need to use them more."

In another analysis, the researchers compared risperidone, a second-generation LAI, with 2 older LAIs commonly used in Norway — zuclopenthixol and perphenazine. They found that the risk for discontinuation of zuclopenthixol was lower than for risperidone. "This was contrary to our hypothesis," Dr. Kroken explained.

Commenting on the poster, Maxine Patel, MD, FRCPsych, clinical senior lecturer from the Institute of Psychiatry, King's College London, United Kingdom, said one reason this work is interesting is the fact that it was not an RCT.

"Although RCTs are generally taken to be the gold standard in research, they don't actually capture all the patients in the field. In particular, patients who find it difficult to participate in an RCT, such as those with poor adherence, are not included," explained Dr. Patel. "This is pertinent for research on antipsychotic LAIs because they are commonly prescribed for patients with suboptimal adherence. Naturalistic studies are the only way to consider what is truly going on in routine clinical practice."

Dr. Patel added that she is impressed with the rigorous methods used. She said the consecutively sampled cohort ensured minimal risk for sampling bias.

"What is also interesting is the way they've separated the clinician decision and patient decision to stop the drug," she added. "I have often wondered whether classifying medication discontinuation as a patient decision only occurs when the patient cannot persuade their clinician to agree with them, whereas a clinician decision to stop a drug is their own or when they've accepted a that patient will do it anyway," she noted.

Dr. Patel added that regardless of whether the LAI is a first- or second-generation antipsychotic, the patients were more likely to decide to continue the LAI, compared with olanzapine, or at least to take the medication longer.

"This has to do with the formulation; patients allow prescribing to continue for longer on LAIs than on orals. This might also be due to...seeing a clinician every 2 weeks," said Dr. Patel.

The study was funded by the Norwegian government. Dr. Kroken has been reimbursed by Eli Lilly, Janssen Cilag, Bristol-Myers Squibb, and Lundbeck for attending conferences. Dr. Patel has disclosed no relevant financial relationships.

3rd Biennial Schizophrenia International Research Society (SIRS) Conference: Poster 204. Presented April 17, 2012.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.