[ROM, benchmark and risk adjustment in a private practice]

The benchmark methods of routine outcome monitoring (ROM)-data by the Dutch Mental Health Care Benchmark Foundation (SBG) have not been evaluated in independent research. No benchmark studies concerning private practices have been published. The Alliance of Quality in Mental Health Care (AkwaGGZ) has taken over the development of ROM; an excellent opportunity for improvement.
AIM: To replicate (parts of) the SBG studies in a solo private practice.
METHOD: Observational research on repeated measures with the SCL-90.
RESULTS: During more than 12 years nearly 1,200 SCL-90 questionnaires have been completed by 644 patients. The ROM-group (≥2 measures, n = 339) and non-ROM-group (n = 280) differed in patient and treatment characteristics, but no differences in pre-test were found. In the ROM-group 73% reliable improvement was found and 45% clinically recovered. For the purpose of benchmarking, the ROM-cohort was divided in three subgroups, representing three so-called providers. Benchmark with these providers revealed differences in progress and results. Post-test was predicted by pre-test, co-morbidity, low and medium education, which explained 33% of the variance. Adjustment for these variables increased the discrepancies between providers.
CONCLUSION: Applying ROM in clinical practice is, at patient level, a useful tool to manage treatment process and determine its termination. Within-therapist-benchmark of aggregated measurements is helpful in detecting patterns and pitfalls. The SBG studies contain methodological imperfections. Risk variables cannot be generalized over samples and risk adjustment should be avoided.