Neuropsychiatric symptoms cluster as primary drivers of Long COVID complexity: a South Texas retrospective cohort study

Abstract

Long COVID, previously known as Post-Acute Sequelae of SARS-CoV-2 (PASC), refers to prolonged symptoms or diagnosable conditions following COVID-19 infection. The neuropsychiatric profile of Long COVID patients remains ambiguous. This study aimed to assess neuropsychiatric symptoms in a retrospective cohort of Long COVID patients (N = 162) at a Rehabilitation Medicine clinic in South Texas. Clinical data from patient records were used to calculate a Symptom Score, and screening tools for stress/PTSD (PCL-5), depression (PHQ-9), anxiety (GAD-7), and quality of life (SWL) were employed to evaluate if Long COVID duration and severity could predict neuropsychiatric outcomes. The majority were female (71%) and Hispanics (53%) who presented for treatment of Long COVID symptoms during the study period, including fatigue (93%), coughing/shortness of breath (81%), fever (67%), anosmia (58%), ageusia (54%), and weight loss (56%). A minority of participants were hospitalized (N = 49) or required ventilator support (N = 5) during acute infection. There was a high burden of neuropsychiatric symptoms, including subjective cognitive impairment (79%), headache (74%), and insomnia (58%). Symptom Score (median = 9, IQR [8,11]) was significantly correlated with increased depression (PHQ-9; p < 0.05), anxiety (GAD-7; p < 0.05) and elevated stress/PTSD (PCL-5; p < 0.05) symptoms. Long COVID patients taking stimulants or mood stabilizers had higher GAD-7 (p < 0.031, p < 0.035) and PHQ-9 (p < 0.034, p < 0.009) scores but not PCL-5 scores. Importantly, duration of Long COVID symptomatology also did not predict PCL-5 scores. No patient factors (e.g., sex, age, BMI, ethnicity) mediated Symptom Score. Nonetheless, historically marginalized groups, such as women and Hispanics, have been disproportionately affected by COVID-19. This study is the first to utilize validated screening tools to determine the presence and severity of neuropsychiatric symptoms in Long COVID patients. These findings may guide clinical management and future research on Long COVID, especially in historically excluded populations.

Keywords: GAD-7; Long COVID; PCL-5; PHQ-9; PTSD; anxiety; depression; stress.

Figure 1

Summary of Symptom Score components and incidence (%) in analytical sample. We collected self-reported data on PASC symptomatology across all major systems as previously described (9, 11). Symptoms were ranked by incidence (% of patients who reported symptom) in our sample. We then calculated a PASC Symptom Score based on the total number of comorbid symptoms present in a patient based on the top 16 most common symptoms, summarized here. Of note, the most common PASC symptoms noted by our sample was fatigue (93%). There was a high burden of symptoms classically related to acute COVID-19 infection (coughing, shortness of breath [SOB], loss of smell and taste) as well as subjective neuropsychiatric symptoms (cognitive impairment = 79%, muscle/joint pain = 78%, insomnia/sleep disturbance = 58%).

Figure 2

Correlation of Symptom Score with measures of depression (PHQ-9), stress (PCL-5), and anxiety (GAD-7) symptoms. We assessed whether PASC Symptom Scores were correlated with the presence of depression, stress, or anxiety symptomatology as determined by validated screening tools. Spearman rank correlations were computed for each pairwise combination of PASC Symptom Score and PCL-5 or GAD-7 scores. PHQ-9, PCL-5, and GAD-7 scores were mutually associated. PASC Symptom Score was positively associated with PHQ-9 (0.218, p < 0.05), GAD-7 (r = 0.29, p < 0.05) and PCL-5 (r = 0.25, p < 0.05).

Figure 3

Correlation of individual PCL-5 items with Long COVID symptoms. Spearman rank correlations were calculated for the intersection of the score of each item, reflecting frequency of stress and PSTD-related symptoms by item of the PCL-5 and the PASC symptoms used to calculate PASC Symptom Score. We contextualize the PCL-5 item scores with their respective symptom clusters for Post-Traumatic Stress Disorder according to the DSM-5, designated to the left of each PCL-5 item: Cluster B (The traumatic event is persistently experienced), Cluster C (Avoidance of trauma-related stimuli after trauma), Cluster D (Negative thoughts or feelings began or worsened after the trauma), Cluster E (Trama-related arousal and reactivity that began or worsened after the trauma). Correlations are given for all findings with p < 0.05.

Figure 4

Correlation of depression (PHQ-9), anxiety (GAD-7), and stress (PCL-5) symptomatology with current medications. Spearman rank correlations were calculated for the intersection of HQ-9, GAD-7, and PCL-5 total scores with classes of current medications endorsed by self-report. Correlations are given for all findings with p < 0.05.