The aim of this study was to assess the effect of risperidone long-acting injection (RLAI) on hospitalisation using a mirror image analysis.

Data on the number of admissions and number of days in hospital were analysed during the two-year period before starting RLAI and the duration of therapy in 56 patients.

Admission rates and length of hospitalisation both decreased for approximately 60% of patients during RLAI therapy.

The time in hospital was significantly lower (p=0.002) during RLAI therapy compared to the previous two years. During the pre-RLAI phase, 59% of patients spent more than one month per year in hospital; in the RLAI phase this was reduced to 23%.

The median number of admissions was also lower, but although 55% of patients had no admissions there were also several patients with multiple admissions during RLAI therapy. During the pre-RLAI phase, no patients had an admission rate above 1.5 admissions per year, but in the post phase 10 patients (18%) had admission rates above 1.5. These results are partially explained by the fact that hospitalisation tended to be shorter during RLAI than in the previous 2 years (average 75 days per admission pre-RLAI vs. 34 days during RLAI).

As there was no control group the results should be interpreted cautiously. Any differences could be caused by other confounding factors which changed with time, and there was wide variation between patients.

In summary, RLAI significantly reduced hospitalisation and this may result in substantial cost savings.