O7.5. STANDARDIZED MORTALITY RATIOS IN EARLY- AND ADULT ONSET FOR THREE DISORDERS

BACKGROUND: Severe mental illness is associated with a reduced life expectancy as compared to the general population. Furthermore, in many studies, early-onset of a severe mental disorder has been associated with a worse prognosis compared to adult-onset. In the current study early-onset groups of patients with schizophrenia, bipolar disorder or depression are compared on standardized mortality ratios (SMRs) to adult-onset. METHODS: The study is a register-based cohort study of the Danish population with the use of the Danish Register of Causes of Death, The Danish Civil Registration System, Danish Patient Registry and the Psychiatric Research Register. The cohort consisted of all patients diagnosed with schizophrenia, bipolar disorder or depression who were alive and living in Denmark during part or all of the study period from January 1, 2000 to the end of 2014. Patient diagnosis was available from the Danish Patient Registry as well as the Danish Psychiatric Research Registry from 1965 until end of study. Patients were followed from inclusion within the study period until age of 40 years old where patients were censored, death or end of study, whichever came first. If patients had received more than one of the included diagnoses, we grouped patients according to a hierarchy with schizophrenia being ranked highest, then bipolar disorder and lowest depression. Early-onset (EO) was defined as a diagnosis prior to 18 years and for each psychiatric disorder and adult-onset (AO) as diagnosis between 18–40 years of age. The primary outcome was SMR for all-cause mortality. As secondary outcomes, we looked at SMRs for unnatural death between each diagnostic group and the general background population. Unnatural death was defined as death by suicide, violence or accident. Early-onset patients (EO) were compared to adult-onset (AO) patients on all outcomes defined above. RESULTS: The total population consisted of 4,661,271 persons (51.1% males), of which 27,753 were diagnosed with schizophrenia (62.1% males), 13,925 with bipolar disorder (40.5% males) and 107,963 with unipolar depression (either single episode or recurrent) (34.5% males). Compared to the general population, SMRs were > 1 in all groups, for both early-onset and adult-onset. Patients with early-onset schizophrenia had a SMR of 7.4 (95% CI: 4.7–11.7) as compared to adult-onset schizophrenia which had a SMR of 8.5 (95% CI: 7.6–9.6). Early-onset bipolar disorder patients had a SMR of 5.9 (95% CI: 2.2–15.7) compared to the patients with adult-onset bipolar disorder 6.0 (95% CI: 4.7–7.6). In patients with unipolar depression with early-onset a SMR of 2.7 (95% CI: 1.8–4.0) was shown as compared to a SMR of 3.7 (95% CI: 3.3–4.1) in patients with adult-onset unipolar depression. When investigating SMRs for unnatural death results were as follows: Schizophrenia EO = 11.0 (95% CI: 6.7–17.9), schizophrenia AO = 16.9 (95% CI: 15.2–18.6); bipolar EO = 9.5 (95% CI: 3.6–25.3), bipolar AO = 17.6 (95% CI: 14.4–21.5) and depression EO = 9.1 (95% CI: 6.3–13.3), depression AO = 9.6 (95% CI: 8.7–10.6). DISCUSSION: SMRs were > 1 for all three disorders for all-cause mortality as well as unnatural deaths. Overall, we did not show worse relative outcome between the investigated groups of early-onset versus adult-onset patients within the three patients groups. Numerically, adult-onset appeared to have higher SMRs than early-onset in schizophrenia and depression, and in all three groups for unnatural death, but the differences did not reach statistical significance.