(P-480) From Contact Coverage to Effective Coverage of Community Care for Patients with Severe Mental Disorders: A Real-world Investigation from Italy

Background: Assessment of rates of contact with mental healthcare (MHC) overestimates the expected health benefits of mental services. The concept of effective coverage introduces the dimension of service quality and captures improvements in health achieved by patients who receive care

Objectives: To measure the gap between contact and effective coverage of MHC

Methods: This study is based on healthcare databases from 4 Italian regions (Lombardy, Emilia-Romagna, Lazio, Sicily), including information as hospital diagnosis, drug prescriptions, outpatient visits, and specific diagnostic and therapeutic codes for patients receiving specialist MHC.
NHS beneficiaries, aged 18-40, who during 2013-2016 had a diagnosis of depression, schizophrenia, bipolar or personality disorder, were identified. Patents who received a prior diagnosis of mental disorder at any time or received two prescriptions of psychotropic drugs in the 2 years prior the index date, were excluded. Newly taken-into-care patients were followed-up until the 06/30/2018.
Emergency admissions to psychiatric wards during follow-up were recorded as outcome episodes and considered as surrogates of relapse. Exposure to two broad categories of MHC, specific therapies with psychotropic drugs (i.e., antidepressants, antipsychotics, mood stabilizers) and community care (generic care, psychosocial and psychotherapy interventions), were considered.
We used a self-controlled case series (SCCS) design, a within person approach to compare the rates of relapse while a patient was covered or uncovered by MHC.
To account for the possibility that exposure and relapse are time-correlated, estimates generated from the SCCS were compared with those generated from a self-controlled referent series (matched with the case series individuals for gender, age, date of mental diagnosis).
Conditional Poisson regression was used for estimating incidence rate ratios for both case (IRRc) and referent (IRRr) series. Dividing IRRc by IRRr, the time-trend adjusted IRRa was obtained.

Results: 11500 relapses occurred. Relapse risk was reduced during periods covered by (i) psychotherapy for patients with depression (IRR 0.67; 95% CI, 0.49-0.91), and bipolar disorder (0.64; 0.29-0.99); (ii) psychosocial interventions for those with depression (0.74; 0.56-0.98), schizophrenia (0.83; 0.68-0.99), and bipolar disorder (0.55; 0.36-0.84), (iii) pharmacotherapy for those with schizophrenia (0.58; 0.49 to 0.69), and bipolar disorder (0.59; 0.44 to 0.78). Coverage with generic MHC did not affect the relapse risk

Conclusions: This study ascertained the gap between utilization of MHC and effective coverage, showing that administrative data can usefully contribute to assessing the effectiveness of a mental health system