J Fotiades, Bronx VAMC/VISN 3 MIRECC; M Lasco, Bronx VAMC; D Kaplow, Brooklyn VAMC

Objectives: (1) To compare the LOS for SMI patients with non-SMI controls, admitted to a VA medical ward for either: Pneumonia, Congestive Heart Failure (CHF), Diabetes Mellitus (DM) exacerbation, Myocardial Infarction (MI), or Chronic Obstructive Pulmonary Disease (COPD).

(2) To determine which (if any) variables significantly increases patients' LOS for any of the above medical conditions.

Methods: The charts for veterans in VISN 3 facilities were reviewed between October 1, 1994 and September 30 1999, using a Vista routine. Patients routinely seen in mental health clinics with International Classification of Disease (ICD) codes of 295,296,297,298 were entered as the SMI group. Patients not seen in mental health without any of the above documented ICD codes were entered as the non-SMI control group. The LOS was calculated for both groups admitted for Pneumonia, CHF, DM, MI and COPD. A logistic regression was conducted controlling for age, race, medical facility, type of psychiatric diagnosis, admitting diagnosis and other medical co-morbid conditions.

Results: A total of 9876 charts were reviewed. There was no statistical significant difference in the LOS for SMI patients admitted for any of the above medical conditions, compared to non-SMI controls, in any of the facilities in VISN 3. The only variable found to make a statistical difference impacting the LOS in the entire sample was additional co-morbid medical conditions. Each additional diagnosis was found to add 3.42 days to these patients' LOS.

Conclusions: The LOS for SMI patients admitted for a medical problem to a medical ward appears to be no different compared to non-SMI patients, when controlling for age, race, psychiatric diagnosis or number of co-existing medical co-morbidities. The only variable shown to increase all patients LOS was the number of additional co-morbid medical conditions.

Impact: SMI patients are often thought to have worse medical outcomes from medical diseases due to their unhealthy lifestyles, poor organizational skills and often non-adherence to treatment plans. The results of this study suggest that when measuring one outcome, LOS, patients with SMI appear to be no different than those without a psychiatric illness. This finding has important implications to hospital administrators when considering such factors as the need to allocate new resources, to establish in-patient Medical-Psychiatric units.