Out of 300 million persons in the United States, about one-half of one percent, or 1.5 million, have a diagnosis of schizophrenia. Schizophrenia begins in young adulthood, and often is chronic and disabling for the remainder of the life course], which is shorter than for the general population by as much as 25 years. The costs of schizophrenia in the United States are estimated to be between $30 and $60 billion dollars annually. Treatment for schizophrenia is only marginally successful: in the Clinical Antipsychotic Trials of Intervention Effectiveness (CATIE), for example, the medication prescribed at the beginning of the trial was stopped or changed in nearly 75% by the completion of the trial 18 months later. The medications have limited effect on negative symptoms or cognitive impairments of schizophrenia, and many have severe and permanent side effects. The basic hypothesis underlying treatment for schizophrenia has not changed for more than half a century. New treatments are needed.

Much accumulating evidence suggests that sensitivity to gluten may be related to symptoms or etiology in schizophrenia and that gluten free diets may lead to significant symptom resolution, but only in patients who are known to have antibodies to gluten.

Gluten sensitivity may be more common than thought and stems from a different etiology and symptom presentation than Celiac Disease. The investigators analysis of the CATIE sample show that about 23% of persons with schizophrenia (compared to 3% of healthy controls) have Gluten Sensitivity (about 300,000 persons in the United States) through the identification of gliadin positive antibodies in their blood. The investigators hypothesize that people with this biomarker could have robust symptom improvements with the removal of the antigen from the diet (gluten). If only half of people with schizophrenia and these antibodies were to substantially benefit from removal of gluten from the diet, as in the case studies and with certain subjects in the clinical trials, this would provide a new transformative treatment option for an identifiable subpopulation of people with schizophrenia and would be of enormous benefit to patients, families and society. Another benefit to the public's health from this study will be enhanced knowledge of the etiology of schizophrenia, including possible linkages between neuropsychiatric disease and immune system activation, and identification of novel, immune-linked treatment targets.

The results of this research could lead to screening for Anti-Gliadin Antibodies early in life or at the first episode of schizophrenia, as recommended by some already. Screening involves financial and emotional costs, and better evidence is needed before this recommendation can be justified. Moreover, a new treatment paradigm of removing gluten from the diet by means of gluten blocking medications (already in early study) could advance treatment significantly.

This study will test the efficacy, in a pilot fashion, of 20 participants in a double blind five week randomized placebo controlled gluten free diet vs identical diet with gluten in gliadin-positive individuals with schizophrenia. Approximately equal numbers will receive the addition of gluten, or non-gluten starch, in identical form (given as flour in food). The investigators plan to develop mechanisms and procedures to locate, screen, and recruit subjects into the inpatient intervention study, retain them during the inpatient phase. Once admitted baseline assessments may take approximately a few days but will be mostly completed in the first week prior to the 5 week randomization, thus patients may stay longer than 5 weeks. At the end of the double blind trial the investigators will prepare for discharge and then test the feasibility of successfully maintaining gluten free diets after the intervention phase is complete, for at least two months.

Inclusion Criteria:

1. DSM-IV diagnosis of schizophrenia or schizoaffective disorder
2. positive for antibodies to gliadin (determined by positive assay in screening protocol)
3. BPRS total score ≥29
4. Age 18- 45 years
5. Same antipsychotic for at least 4 weeks
6. Ability to consent determined by a score of 10 or greater on the Evaluation to Sign Consent.

Exclusion Criteria:

1. Persons already on gluten free diets
2. Pregnant or lactating females
3. Organic brain disorder or mental retardation
4. Medical condition whose pathology or treatment could alter the presentation or treatment of schizophrenia or significantly increase the risk associated with the proposed treatment protocol
5. Meets DSM-IV criteria for alcohol or substance abuse (other than nicotine) within the last month
6. Gluten ataxia, assessed by the International Cooperative Ataxia Rating Scale (ICARS)[15]
7. Inability to provide informed consent.