Healthcare resource utilization after medium-term residential assessment for epilepsy and psychogenic nonepileptic seizures.

Introduction: Epilepsy and epilepsy mimics may lead to high healthcare resource utilization (HRU) including diagnostic resources. The William Quarrier Scottish Epilepsy Centre (SEC) provides medium-term residential assessment (MTRA; average length of stay: 28days) and treatment for complex presentations of epilepsy and related conditions (principally psychogenic nonepileptic seizures, PNES). We studied the effect of MTRA on HRU in a defined health board area in Scotland.

Methods: A retrospective audit of individuals admitted to the SEC from a defined health board area using SEC and health board medical records. Neurological HRU assessed included emergency department visits, hospital admissions, outpatient clinic appointments, and brain imaging prior to and post-MTRA. Healthcare resource utilization was also compared with individuals referred but not admitted to the SEC because of individual circumstances and choice.

Results: Seventy-three individuals (51 female, average age: 37.51; 22 men, average age: 43.72) were identified from three years of admissions (1st April 2010 to 31st March 2013). Final diagnosis was epilepsy (ES), 32; ES and psychogenic nonepileptic seizures (ES+PNES), 17; and PNES alone, 24. Twenty-two individuals were identified as a comparison group (8 men, 14 women; average age: 37.21 and 43.90, respectively). Total average contacts per patient per year (CPY) was significantly different pre- and post-MTRA (4.16 vs. 1.32; t(72)=6.11, p<.0001, d=.72). Comparison of HRU in the first year of baseline and last full year of follow-up showed a post-MTRA reduction in HRU for PNES of 92.28%, for ES of 46.81%, and for ES+PNES of 28.3%. During the course of follow-up, PNES CPY continued to drop (1.13 first year vs. 0.10 at 3years post-MTRA). For individuals with epilepsy (with or without PNES), HRU use dropped significantly in the year after admission, and these gains remained stable (total first vs. third postdischarge CPY, 1.74 vs. 1.29). The participants in the comparison group, who were not admitted, had no comparable drop across the study period and were using significantly more resources at each follow-up point than those in the admitted group (F (1, 48)=44.45, p<.01, ηp(2)=.49).

Conclusion: Medium-term residential assessment is associated with sustained reduction in HRU especially in patients with PNES. Overall HRU reduction was 68.27% following admission (d=.72). This suggests benefit from the MTRA model for people with complex presentations.