The specific aim of this study is to determine whether the new, long-acting, form of risperidone, Risperdal Consta, improves the ability of schizophrenia patients to benefit from skills training. The hypothesis guiding this study is that Risperdal Consta, by improving verbal memory, will improve the ability to benefit from skills training interventions among schizophrenia patients. The primary objective of this study is to compare patients on Risperdal Consta to patients on other atypical antipsychotic medications in terms of their ability to benefit from skills training interventions. A secondary objective of this study is to determine whether patients taking Risperdal Consta improve in other areas of cognitive functioning and social functioning.

The study will be conducted at the University of Illinois at Chicago and New York Presbyterian Hospital-Westchester Division. Patients with schizophrenia or schizoaffective disorder who are considered by their clinicians to not be clinically stable, or to be stable but still symptomatic, will be referred to the study. After obtaining informed consent, patients will be randomized either to the Risperdal Consta or control conditions. Patients in the control condition will remain on their oral antipsychotic medication for the duration of the study (note - patients currently taking oral risperidone will not be eligible for the study). Patients randomized to the Consta condition will take oral risperidone for 3 days, and if there are no adverse effects, they will begin receiving injections of Consta every 2 weeks over the next 6 months. During the first month, they will remain on oral risperidone, which will be tapered by the end of the month along with other antipsychotic medications other than Consta. At baseline, and every 4 weeks for 6 months, all patients will complete a computerized cognitive testing battery (CogTest), and be interviewed about social functioning and symptomatology. During months 4-6 of study participation, all patients will participate in a social skills group, the UCLA Basic Conversation Skills Module. Before and after this group, patients will be interviewed and participate in role-play based assessments regarding skills to be taught in the group. Patients will also receive physical exams, including and ECG, at baseline, and be monitored for side effects and adverse events every 2 weeks. The hypothesis driving this study is that patients in the Consta group will demonstrate greater improvement in verbal learning, skill acquisition in the training group, and improved social functioning, over the course of the 6 months. This hypothesis is based on prior data demonstrating that oral risperidone improves verbal memory, which is a mediator of skill acquisition in structured skills training groups. If this hypothesis is supported, it would be an important demonstration that a novel medication preparation (injectable Risperdal Consta) can, when combined with psychiatric rehabilitation, improve functional outcomes for patients participating in rehabilitation.

Efficacy Measures:

1. The primary efficacy measure for this study will be the Micro-Module Learning Tests (MMLT) (Silverstein et al., in press). This is a set of 7 psychometrically equivalent tests that assess responsiveness to the key components of skills training. Each test has 3 parts: 1) a verbal learning component wherein the patient is read material from an actual skills training group and then asked questions about the material; 2) a modeling component in which the patient views a model performing a specific behavior on a videotape, and is then asked questions about what was viewed; and 3) a role-play component in which the patient is asked to demonstrate, via a role-play, the behavior that the model demonstrated in the prior section. The MMLT is a dynamic assessment measure in that, in addition to measuring what people can recall/perform after an initial presentation of information, it allows for a determination of learning potential after repeated exposure to the information. With the MMLT, each time a question is initially answered incorrectly, a briefer version of the material, that highlights the most relevant information is presented. If an incorrect answer is given at this point, an even more focused version of the item is presented. With this scoring system, responsiveness to skills training as it occurs in the real world (where repetition of material is built into the teaching method) can be assessed more accurately than when using a static assessment (where information is only presented once).
2. A second primary efficacy measure will be the Comprehensive Module Test for the Basic Conversation Skills Module. This measure is administered before and after the group.
3. A secondary efficacy measure is the CogTest computerized cognitive battery. This battery was developed by Tonmoy Sharma, M.D. and colleagues, and has been used with over 2000 schizophrenia patients in studies of risperidone or Consta to date.
4. Another secondary efficacy measure is the Social Functioning Scale (SFS) (Birchwood et al., 1990). This is a 74 item scale that is rated by the patient. The SFS has seven subscales: 1) social engagement/withdrawal; 2) interpersonal communication; 3) independence-performance of activities of daily living; 4) independence-competence to perform activities of daily living; 5) frequency of engagement in recreational activities; 6) frequency of participation in social activities; and 7) employment.
5. A third secondary efficacy measure will be the observational measure of attentiveness in groups that was developed as part of the PI's current NIMH grant on cognitive rehabilitation in schizophrenia. This measure will allow us to determine: 1) if Risperdal Consta is improving a real-world measure of attention; and 2) whether there are any relationships between skill acquisition, medication, and attention.
6. The final efficacy measure will be the Positive and Negative Syndrome Scale, which will be used to measure symptoms.

Adverse Events: Side effects and adverse events will be assessed using a semi-structured interview given by a research nurse or psychiatrist. There are 40 specific side effects in 6 categories (general, neurological, cardiovascular, autonomic, urologic, muscular skeletal). Adverse events will be rated on an intensity scale of 0 to 4 and a relationship to medication of 0 to 5. Side effects will be counted as adverse events for scores of 3 or 4 on intensity (moderate or severe) and scores of 3-5 for relationship (possible to definite). These adverse events will then be aggregated into one of the 6 groups for analysis across groups.

Serious adverse events will include those that require medical intervention such as additional monitoring by the subject's primary physician or medical hospitalization. All significant adverse events (such as those listed above or any side effect with an intensity level of 4) will be reviewed with the treating psychiatrist and the independent medical reviewers. Each site will have two independent medical reviewers (one internist, one psychiatrist) to examine any serious or significant adverse events and determine if the subject should be withdrawn from the study. The two reviewers will not be directly connected to the research group and will not be investigators in the protocol.

In addition to the assessment of side effects, other scales will be used to measure side effects commonly associated with antipsychotic medications. Each month, a blinded research assistant will perform an Abnormal Involuntary Movement Scale (AIMS, to assess for tardive dyskinesia), Barnes Akathisia Scale, and Simpson-Angus Scale (to assess for dystonia and parkinsonian side effects). Movement disorders can affect the ability of patients to perform on cognitive tests and in skills training groups. Patients with akathisia often have reduced attention and reduced ability to concentrate in groups. Patients with parkinsonian symptoms can be slower on manual tasks and may have slower mental abilities.

Laboratory Assessment (to be done at baseline and last month of participation):

1. Complete Blood Count (CBC) to determine major hematological indices as some antipsychotic medications affect the white count.
2. Complete Chemistries including electrolytes, BUN, creatinine, liver function tests, and glucose. Some antipsychotic medications may increase the risk of diabetes and some medications may cause an increase in liver enzymes.
3. Lipid profile including cholesterol, triglycerides, LDL, and HDL. Some antipsychotic medications cause an increase in lipids particularly triglycerides
4. Prolactin: some antipsychotics may increase prolactin leading to menstrual irregularities, sexual dysfunction, and possibly osteoporosis. Months 0, 3, 6, only.
5. Thyroid stimulating hormone (TSH): thyroid functioning can affect cognition. Patients with a history of hypothyroidism will be allowed in the study if they are on stable dose of replacement and the TSH is in the normal range. Months 0, 3, 6, only.
6. Glycohemoglobin (HgA1c): to measure the overall level of glucose over the prior 2-3 months. Months 0, 3, 6, only.

Laboratory Assessment (monthly)

1. For women of childbearing potential, urine screen for pregnancy.

Inclusion Criteria:

1. Schizophrenia or schizoaffective disorder
2. Ages 18-55
3. Ability to give informed consent
4. Good general physical health or stable chronic medical conditions
5. Ability to be on a single antipsychotic medication
6. History of inattentiveness in psychosocial treatment settings
7. Poor social skills
8. Must be receiving a single antipsychotic medication
9. Must be clinically stable - defined by the absence of prominent delusions, hallucinations, disorganized speech, or grossly disorganized or catatonic behavior. However, there must also be continuing evidence of impairment, as indicated by the presence of negative symptoms or two or more symptoms of schizophrenia listed in Criterion A of the DSM-IV, in attenuated form (e.g., odd beliefs, unusual perceptual experiences).

Exclusion Criteria:

1. Inability to give informed consent
2. Substance dependency in the past 6 months
3. Diagnosis of dementia
4. Significant head injury or other brain injury leading to cognitive impairment
5. Mental retardation (premorbid IQ < 65)
6. Pregnant or nursing
7. Allergy or other significant adverse reaction to risperidone
8. Contraindication to Risperdal Consta as only antipsychotic
9. Currently taking risperidone