Background: It is likely that many people with autism who are in long-term psychiatric hospitals are misdiagnosed with other psychiatric disorders. Four published studies have estimated that among adults who are receiving mental health treatment, between 0.6% and 5.3% have autism, but only a tenth of those with autism are diagnosed as such; instead, they were most commonly diagnosed with schizophrenia.

It can be very difficult to distinguish between adults with some kinds of schizophrenia and adults with autism. Over the last 30 years, the definition of schizophrenia has become much narrower and the definition of autism has become much broader. While the symptoms of autism are evident at a much earlier age than the symptoms of schizophrenia, the social deficits that are a core feature of autism are quite common among people with schizophrenia. Without a careful evaluation that includes a thorough developmental history, it can be difficult to tell these disorders apart.

It is important that people with autism are diagnosed as such. Making the proper diagnosis may affect the treatment they receive. For example, the wrong diagnosis may lead to use of inappropriate medications. In addition, people with autism often respond to lower dosages than people with schizophrenia. Adults with autism may respond well to structured behavioral interventions that improve social functioning.

Objective: The purpose of this study is to determine how many patients at one large psychiatric hospital have autism, what characteristics differentiate them from individuals with other psychiatric disorders, and whether there is a feasible way to easily determine whether adults have autism when they first come to the hospital.

Study Design: We plan to screen patients in one psychiatric hospital for autism using a validated measure called the Social Responsiveness Scale (SRS). The SRS will be administered by day shift nurses who have known the patient for at least 4 months. For all patients who score very high on this scale and a sample of those with lower scores, we will conduct chart reviews, family interviews, and clinical interviews to collect more information that will help us differentiate between autism and other disorders. A team of psychiatrists, psychologists, and other researchers then will review each case to determine whether the information we have collected suggests that the patient has autism. Finally, we will use various statistical methods to determine which patient characteristics distinguish between individuals with autism and other disorders.

How This Research Might Help People with Autism: Successful completion of the proposed study will result in at least four important outcomes: (1) determining the potential for missed diagnosis of autism in psychiatric hospitals; (2) characterizing autism in adults who have not been diagnosed at earlier stages of development; (3) developing a method for differentiating autism from other psychiatric disorders; and (4) neuro-behavioral measures that can be used to assess whether interventions are working.

The results of this study will ultimately benefit adults with autism who have been misdiagnosed, with correct diagnosis hopefully leading to more appropriate treatment. The measures we develop will be used as diagnostic and assessment tools for adults. There will be immediate benefit to study participants as well, especially if it turns out that they in fact have autism. Risks to these subjects are minimal. We anticipate presenting our results at conferences and submitting the results for publication within a year. The use of our findings in diagnosing adults with suspected autism could occur quickly.