According to previous studies, mental status in 1-year COVID-19 survivors might range from 6–43%. Longer-term psychological consequences in recovered COVID-19 subjects are unknown, so we analyzed longer-term quality of life and mental status in recovered COVID-19 subjects at 2 years after infection. Among 144 recovered COVID-19 subjects in the Taizhou region, 73 and 45 completed face-to-face follow-ups at the first year and second year after infection, respectively, with a 61.7% follow-up rate. The questionnaire, which was administered at both follow-ups, included questions about quality of life, psychological health, and post-traumatic stress disorder (PTSD). The Mann-Whitney U test was used to the differences of each scale between the first and second year. Among the 45 people who completed both follow-up visits, the incidence of psychological problems was 4.4% (2/45) in the first year, and no new psychological abnormalities were observed in the second year. Quality of life improved, while the General Health Questionnaire (GHQ-12) and Impact of Event Scale-Revised (IES-R) scores did not improve over time. The incidence of mental disorders was lower than those in previous studies. Multidisciplinary management for COVID-19 in this study hospital may have reduced the frequency to a certain extent. However, among those with mental health problems, such problems may exist for a long time, and long-term attention should be given to the psychological status of recovered COVID-19 subjects.

Severity prediction of COVID-19 remains one of the major clinical challenges for the ongoing pandemic. Here, we have recruited a 144 COVID-19 patient cohort consisting of training, validation, and internal test sets, longitudinally recorded 124 routine clinical and laboratory parameters, and built a machine learning model to predict the disease progression based on measurements from the first 12 days since the disease onset when no patient became severe. A panel of 11 routine clinical factors, including oxygenation index, basophil counts, aspartate aminotransferase, gender, magnesium, gamma glutamyl transpeptidase, platelet counts, activated partial thromboplastin time, oxygen saturation, body temperature and days after symptom onset, constructed a classifier for COVID-19 severity prediction, achieving accuracy of over 94%. Validation of the model in an independent cohort containing 25 patients achieved accuracy of 80%. The overall sensitivity, specificity, PPV and NPV were 0.70, 0.99, 0.93 and 0.93, respectively. Our model captured predictive dynamics of LDH and CK while their levels were in the normal range. This study presents a practical model for timely severity prediction and surveillance for COVID-19, which is freely available at webserver https://guomics.shinyapps.io/covidAI/ .