COVID-19 Survivors Are Still in Need of Neuropsychiatric Support Two Years after Infection

COVID-19 survivors have been reported to be at risk of long-term neuropsychiatric sequalae; however, prospective evidence in this regard is lacking. We prospectively assessed the occurrence of mental-health-domain-related symptoms over a 24-month period following COVID-19 onset in a cohort of 230 patients. Of them, 36.1% were still presenting with at least one symptom 24 months later. Across the study period, a significant reduction in overall symptoms from the onset was observed (p < 0.001); however, symptom prevalence was unchanged between the 12- and 24-month follow-ups across most symptomatic domains. At the 24-month follow-up, mental-health-domain-related symptoms only were higher than at the onset and were the most frequently reported symptoms. Dyspnea at the onset predicted both symptoms of psychiatric disorders (OR = 3.26, 95% CI = 1.22–8.70, and p = 0.019) and a lack of concentration and focus (OR = 3.17, 95% CI = 1.40–7.16, and p = 0.005) 24 months post-infection, with the number of comorbidities at the onset also predicting the occurrence of a lack of concentration and focus (OR = 1.52, 95% CI = 1.12–2.08, and p = 0.008). The findings of this study may have important public health implications, as they underlie the fact that COVID-19 survivors are still in need of neuropsychiatric support two years after infection.


Introduction
The coronavirus disease 19 (COVID-19) pandemic has had indirect detrimental effects on the general population's mental health due to restrictions and social disruptions [1,2]. Available evidence also indicates direct effects of COVID-19, in terms of the persistence or onset of multiple symptoms, with the most common being fatigue, shortness of breath, and cognitive dysfunction. Of relevance, neuropsychiatric symptoms have also been reported, possibly reflecting phenomena occurring in much later phases among COVID-19 survivors [3]. Such conditions, referred to as 'long COVID' or 'post-COVID (syndrome)' [4], carry a significant burden for individuals' wellbeing [5], and their longer-term effects are largely unknown. While predictors of poor mental health in the post-acute phase of the pandemic have been better elucidated [6,7], whether some patients are at an increased risk of presenting with neuropsychiatric symptoms long after infection remains to be investigated. This is of paramount importance, considering that mental health difficulties are widely expressed among the general population, and the associated enduring cognitive dysfunction, sleep problems, and somatic complaints may drastically affect people's ability to function in everyday life [3].

Study Design and Participants
This study investigated the neuropsychiatric symptom trajectory among COVID-19 survivors during the 24 months following the acute COVID-19 phase, also exploring potential sociodemographic and clinical predictors of symptom occurrence. Study details, including methods and recruitment procedures, have been reported before [8]. Briefly, between 1 March and 30 May 2020, a cohort of consecutive COVID-19 adult patients, aged 18 years or older, was recruited at the university hospital of Udine, Italy, a tertiary referral hospital of about 1000 beds offering healthcare services to a population of about 350,000 people, which was identified as a regional hub for COVID-19 patients. Both inpatients and outpatients were considered eligible for inclusion in the study if they presented with a COVID-19 diagnosis.

Assessment
Employing a longitudinal design, information on sociodemographic, clinical, and laboratory data was collected both at baseline and at subsequent follow-up assessments up to 24 months after the disease onset. COVID-19 diagnosis was based on a positive nucleic acid amplification test (NAAT) for SARS-CoV-2 in respiratory tract specimens (confirmed diagnosis) or either laboratory or imaging findings suggestive of infection and/or positive serology (suspected diagnosis). Based on clinical presentation, as assessed with a COVID-19 disease severity scale, patients were also classified as a function of the disease severity, from asymptomatic to critical disease. Finally, patients were also classified based on their disease management, depending on whether they were deemed to receive intensive care unit (ICU), hospital ward, or outpatient-based support [8]. This report focuses on the 24-month follow-up assessment. As was performed for the 12-month follow-up, information was collected over the phone by trained nurses via a standardized questionnaire. To offer a comprehensive overview of mental health difficulties among COVID-19 survivors, in line with investigations carried at the 12-month follow-up, information was collected with reference to common psychiatric disorders (i.e., depression, anxiety, and insomnia), cognitive impairment (i.e., a lack of concentration and focus), and somatic distress (i.e., fatigue) [8].

Statistical Analyses
Data analysis was performed by STATA 18, reporting absolute values and percentages in addition to comparing categorical variables across the time points (COVID-19 onset, 12-and 24-month follow-ups) via Cochran's Q test for three time points or the McNemar test for two time points. Finally, after dichotomizing COVID-19 survivors based on whether they presented with post-COVID-19 neuropsychiatric symptoms, multivariable logistic regressions explored potential predictors of suffering from such symptoms based on statistical significance in univariable analysis, estimating the odds ratios (OR; 95% confidence interval, CI).

Acute COVID-19 Presentation
Information on study sample characteristics in terms of acute COVID-19 presentation has been reported before [8]. Most COVID-19 patients had a symptomatic disease (92.6%) of mild severity (67.7%), with one in four patients presenting with moderate to critically severe COVID-19 (24.9%). Almost one-third of the sample required admission to hospital (28.7%), with a relatively low proportion of patients needing intensive care unit (ICU) support (5.2%). COVID-19 patients had a median in-hospital stay of 7 days (IQR 4-10). n, number; N, number as a denominator;ˆ, pulmonary disease: asthma, chronic obstructive pulmonary disease; *, cardiovascular disease: heart failure, ischemic heart disease, tachyarrhythmias, valvular heart disease, and venous thromboembolism; and • , depression, anxiety.

Symptomatic Course over the 24-Month Follow-Up
One in three patients (36.1%) were still presenting with at least one symptom at the 24-month follow-up. Although a significant overall symptom reduction from the onset was observed (p < 0.001), symptom prevalence appeared to be substantially unchanged between the 12-and the 24-month follow-ups. Similar stable patterns were observed for most of the single symptomatic domains where, if a symptom reduction had been observed between the onset and the 12-month follow-up [3], no further reduction occurred over the subsequent 12 months. Additionally, in line with observations carried at the 12-month follow-up [3], neuropsychiatric symptoms were higher than at onset and were still the most frequently reported symptoms at the 24-month follow-up (symptoms of psychiatric disorders, 9.6%; a lack of concentration and focus, 25.2%; and fatigue, 14.4%). Additionally, a lack of concentration and focus was the only symptom that increased substantially even over the second year of observation, although just approaching significance (p = 0.071; Table 2).

Multivariable Logistic Models
Multivariable logistic models revealed that dyspnea at onset, but not care intensity, predicted symptoms of both psychiatric disorders (OR = 3.26, 95% CI = 1.22-8.70, and p = 0.019) and a lack of concentration and focus (OR = 3.17, 95% CI = 1.40-7.16, and p = 0.005) 24 months post-infection (Tables 3 and 4). A significant association was also found for the number of comorbidities patients had at onset and the occurrence of a lack of concentration and focus at the 24-month follow-up (OR = 1.52, 95% CI = 1.12-2.08, and p = 0.008; Table 4).

Discussion
COVID-19-induced neuropsychiatric symptoms may have plateaued 24 months post-infection, but they have not reduced. Rather, cognitive difficulties seem to have increased, with one out of four patients complaining about them at the 24-month follow-up. Such a pattern seems to be unique to neuropsychiatric symptoms, urging for studies specifically investigating predictors of post-COVID psychiatric syndrome to sustain the greatest possible recovery. Results indicate that patients with pre-existing vulnerabilities, as indicated by the number of medical comorbidities at COVID-19 onset, and those suffering from severe COVID-19, as suggested by the occurrence of dyspnea, possibly accounting for higher care intensity in the acute phase, are particularly susceptible to presenting with long-lasting neuropsychiatric symptoms.
The limitations of this study include the absence of a standardized approach to diagnose and treat post-COVID-19 neuropsychiatric symptoms, possibly jeopardizing comparability across studies. Additionally, as the study design does not allow disentangling the detrimental indirect effect of the pandemic, its contribution to the observed mental health aftereffects cannot be completely ruled out. Moreover, whether the detected effects are specific to COVID-19 or would have occurred among patients presenting with other infectious diseases remains to be tested. The strengths of the current study include the large sample, the longer follow-up when compared to most of the available evidence, and the extensive data gathering.
In conclusion, evidence from epidemiological and clinical studies provides converging and convincing proof that COVID-19 may result in neuropsychiatric sequalae. Consequently, the focus of research must move from investigating whether post-COVID neuropsychiatric syndrome exists to understanding the mechanisms involved and who is the most susceptible. In particular, future studies will have to focus on the inflammatory responses that COVID-19 may trigger in the brain, provoking damages, and subsequent symptom onset [9,10].  Informed Consent Statement: Informed consent was obtained from all subjects before data collection. Data Availability Statement: Data available on request due to restrictions, e.g., privacy or ethical.