Reply to Letter to Editor by Ahmed HS, et al re: Possibility of Age and Prior Psychiatric Illnesses Affecting the Study Design

To the Editor:

We thank Ahmed (2022)¹ for highlighting features of our recent article in which we report and contrast the rates of cognitive dysfunction, emotional distress, and functional decline approximately 4 months after acute coronavirus disease 2019 (COVID-19) illness in those who did and did not require intensive care unit (ICU) admission.² Ahmed (2022)¹ pointed out that patients in our sample who received care in the ICU were older than non-ICU patients and proposed that this discrepancy may have contributed to the generally lower cognitive composite test scores observed among those requiring more intensive treatment. It was similarly noted that the ICU patient group seemed to exhibit less psychiatric distress than their non-ICU peers and that it remains unclear the extent to which either cognitive dysfunction or psychiatric symptoms were present prior to COVID-19 illness. A question was also raised as to the nature and interpretation of a self-report instrument.

Older adults appear at greater risk of more severe COVID-19 illness and are more likely to have pre-existing diminution in cognitive functioning due to normal and/or pathologic aging.³ We addressed this potential confound in our data by analyzing age-adjusted cognitive test performances that account for the contribution of age to cognitive test scores. We also employed statistical adjustments for this and other nonmedical characteristics that are frequently associated with cognitive test performances (i.e., sex, racial/ethnic minority status, educational attainment, and estimated IQ). Our finding that those receiving care in the ICU continued to demonstrate poorer global cognition after such adjustments suggests that more severe COVID-19 illness and/or factors associated with ICU admission, rather than systematic biases in our clinical groups, may underly the poorer cognitive outcomes observed in ICU survivors. Our findings that cognitive functioning was not associated with the severity of mental health symptoms or functional decline and that neuropsychiatric symptoms were frequent regardless of the need for ICU care are important, as they highlight the role of routine assessment of mental health symptoms after COVID-19 regardless of initial illness severity.

We noted that ours was a descriptive study of the neuropsychiatric features characterizing those adults who are likely to receive care in a post-COVID-19 pulmonary clinic. It was not designed to address the underlying etiologies for the neuropsychiatric symptoms we observed. However, we agree with Ahmed (2022)¹ that studies employing control groups well matched for pre-existing medical, cognitive, and mental health features are needed in order to more clearly identify the extent to which COVID-19 exerts an independent effect on neuropsychiatric outcomes, identify those at greatest risk, and to guide management. Recent data from hospitalized adults in Wuhan, China, demonstrate that older patients with severe illness, when compared with non-COVID-19 controls and those with less severe illness, are more likely to show persistent and progressive cognitive decline over the first year after the illness.⁴ Such individuals may represent a particularly at-risk group and may require ongoing neuropsychiatric specialty care. In contrast, clinical data from studies of patients seeking neuropsychological services at post-COVID-19 condition clinics in the United States⁵ suggest that such groups disproportionately had middle-aged (mean age <50 y) and female (84%) populations and report high rates of pre-existing psychiatric conditions (57%). In such samples, subjective cognitive dysfunction is more pronounced than the generally mild deficits observed in formal neuropsychological testing and tends to be associated with the severity of mood/anxiety symptoms.⁵ Those findings suggest that among clinically referred patient groups, there may be a primary role for psychologic and psychiatric interventions aimed at improving patient distress and functioning.

Finally, Ahmed (2022)¹ suggests that the Quick Dementia Rating Scale be employed as a measure of mental health symptoms and would more appropriately be interpreted as an index of cognitive ability. Here, we respectfully disagree as this instrument was employed to assess the extent of patient-reported decline in daily functioning attributable to subjective changes in cognition, mood, and ability to perform other activities of daily living. Thus, the Quick Dementia Rating Scale served as an adjunct to, rather than a direct measure of, cognition and psychiatric symptoms.