Can multisite clinical trial results change clinical practice? Use of long-acting injectable risperidone nationally in the Veterans Health Administration.

BACKGROUND: Multisite practical clinical trials evaluate treatments in real-world practice. A multisite randomized Veterans Health Administration (VHA) cooperative study (CSP#555) published in 2011 compared the first long-acting injectable (LAI) second-generation antipsychotic (SGA), Risperidone Consta®, in veterans with a diagnosis of schizophrenia or schizoaffective disorder, to oral antipsychotics, with unexpected null results for effectiveness and cost-effectiveness. Whether null results of this type could change VHA practice has not been studied. METHODS: A longitudinal observational analysis was used to evaluate the impact of the trial findings on VHA clinical practices. National administrative data compared new starts on LAI risperidone during the 4 years before the publication of CSP#555 in 2011 to new starts on LAI risperidone during the 4 years after. RESULTS: Among 119,565 Veterans with the indicated diagnoses treated with antipsychotics from 2007 to 2015, the number and proportion of new starts on LAI risperidone declined significantly following the study publication, as did the total number of annual users and drug expenditures. However, data from 2007 to 2010 showed the decline in new starts actually preceded the publication of CSP#555. This change was likely explained by the increase in new starts, total use, and expenditures on a newer medicine, LAI paliperidone, a 4-week LAI treatment, in the 2 years prior to the publication of CSP#555. CONCLUSIONS: The declining use of LAI risperidone likely primarily reflects the substitution of a longer-acting LAI SGA, paliperidone, that came to market 2 years before the study publication, a substitution that may have been reinforced by null CSP#555 study results for LAI risperidone.

Impact of multi-site clinical trial results on clinical practice: Use of risperidone to treat PTSD nationally in the veterans health administration.

Multi-site randomized effectiveness trials evaluate treatments under real-world conditions. Whether results change practice is under-studied. A 6-month 26-site Veterans Health Administration (VHA) cooperative study published in 2011 compared an oral second-generation antipsychotic, risperidone, to placebo for refractory PTSD with null results. National VHA administrative data compared new starts on risperidone during the 5 years before and after the year of publication. Among the 450,000-841,000 Veterans diagnosed with PTSD annually from 2006 to 2016 the proportion with new starts on risperidone declined every year before and after publication. No evidence of an effect of null study results on VHA clinical practice was observed.