COVID-19 had a significant impact on the mental health of the affected population. Such multifactorial risk for a deterioration of mental health suggests the need to identify groups of patients with psychiatric vulnerability and to establish strategies of intervention based on scientific evidence.

The aim of the study was to identify psychiatric outcomes one year after recovery and possible associations between these and the clinical, anamnestic, and sociodemographic variables.

The Mini International Neuropsychiatric Interview was employed to assess current and lifetime mental illness in a cohort of 100 patients discharged between March and April 2020 from COVID-19 wards of the San Paolo Hospital in Milan, Italy. The Kendall rank correlation coefficient was administered to measure the ordinal association between clinical-demographic variables and the psychiatric diagnoses of patients. Bivariate correlation was used to explore the association between psychiatric outcomes and the sample characteristics.

Almost one third of subjects screened positive for a diagnosis of a new psychiatric disorder, and a novel onset of psychiatric morbidity did not differ significantly in patients with and without a positive history of mental illness (42 and 58%). New psychiatric disorders were grouped into stress reactions, anxiety-group disorders and mood disorders. Concerning demographic characteristics, advanced age represented a protective factor against the onset of new psychiatric disorders (rτ = -0,203, p =0,008). Despite a lower risk of contracting the infection, women in our cohort were more vulnerable to psychiatric post-Covid symptoms (rτ =0,190, p =0,029). The correlation between the onset of new psychiatric disorders and some pre-admission vulnerability factors, such as an overweight condition (rτ =0,185, p =0,026) and a positive medical history for cigarette smoking (rτ =0,203, p =0,026), were statistically significant. Moreover, subjects who reported taking a therapy to control the infection prior to hospitalization were more likely to receive a new psychiatric diagnose (rτ =0,269, p =0,005). Of note, variables related to the severity of hospitalization such as oxygenation intensity, days of hospitalization, or requirement of intensive car were not associated with new psychiatric diagnoses.

The onset of psychiatric disorders shows a relevant frequency in patients hospitalized for COVID-19, suggesting that mental health services should structure adequate screening and diagnosis methods. Three levels of intervention can also be expected to reduce the overall risk and burden of psychiatric morbidity: increasing awareness regarding modifiable risk factors; guaranteeing a minimal level of mental health support to patients hospitalized for COVID-19; providing personalized interventions with respect to gender and age groups.

None Declared