Multi-disciplinary collaborative consensus guidance statement on the assessment and treatment of mental health symptoms in patients with post-acute sequelae of SARS-CoV-2 infection (PASC)

Post-acute sequelae of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection (PASC), also called Long COVID (coronavirus disease), is the experience of new or worsened signs, symptoms, or conditions that develop after resolution of the acute phase of a COVID-19 infection. Although some people with COVID-19 recover well, others have persisting symptoms.1 PASC can manifest as a wide-ranging constellation of disabling sequelae, including mental health conditions.2 Anxiety and depression have been reported as the second and third most common symptoms of PASC, respectively.3 Furthermore, in a narrative review summarizing neuropsychiatric dimensions of PASC, anxiety, post-traumatic stress disorder (PTSD), and depression were among the most reported symptoms of both ongoing symptomatic COVID-19 and PASC. Pooled prevalence of each was: anxiety 19.1% (95% confidence interval [CI], 13.3%–26.8%), PTSD 15.7% (95% CI, 9.9%–24.1%), and depression 12.9% (95% CI, 7.5%–21.5%).4 Despite the high prevalence and often disabling impact of PASC and PASC-related mental health symptoms and the emerging data that PASC can persist for months or years,5 limited guidance currently exists regarding the assessment and treatment of mental health conditions in patients with PASC. Addressing mental health symptoms in the setting of PASC involves several unique complexities, including addressing stigma that may interfere with appropriate diagnosis(es) and treatment, insufficient availability of mental health professionals, and differentiating mental and physical health diagnoses. Many patients have described being questioned about their PASC-related symptoms in a way that feels dismissive of their experience and/or mistakenly attributed to an underlying mental health condition.6 Although the effects of the COVID-19 pandemic on individuals and society can exacerbate direct sequelae from SARS-CoV-2 infection, increasingly evidence supports that new mental health symptoms can also be a component of PASC or caused by SARS- CoV-2 infection, and pre-existing mental health conditions can be exacerbated by PASC, yet mental health conditions are not, in and of themselves, the overall cause of PASC.7-9 Another challenge in addressing mental health disorders is the national and global shortage of mental health professionals.10 This challenge is further intensified because clinicians who are not mental health specialists often express discomfort and a perception of insufficient training to discuss mental health with their patients. Nevertheless, most patients report that they want clinicians who are not mental health specialists to broach the topic and acknowledge the interplay between mental and physical health.11-13 Finally, studies have found that PASC can manifest as symptoms that are not due to a mental health condition, but that can mimic and/or be exacerbated by a mental health disorder. These symptoms include fatigue, dysautonomia, disordered sleep, and cognitive dysfunction,14, 15 which can also interfere with a patient's ability to fully participate in first-line treatment recommendations. The goal of this consensus statement is to present practical guidance for clinicians who treat patients with PASC. Specifically, this statement addresses the assessment and initial treatment of PASC-related mental health symptoms including depression, anxiety disorders (including panic), and PTSD. People with PASC have also reported new or worsening suicidal ideation, psychosis, obsessive compulsive disorder (OCD), and pandemic-related grief and survivor's remorse8, 16-18; however, specific focus on these conditions is outside the scope of this statement. The recommendations in this statement are applicable to all patients with PASC who are experiencing mental health symptoms, regardless of the time course. Of note, this guidance statement reflects the current evidence base and related recommendations from an expert panel of health care professionals who regularly care for people with PASC. The recommendations should not preclude clinical judgment and must be applied in the context of each specific patient, with adjustments for patient preferences, comorbidities, and other factors. The American Academy of Physical Medicine and Rehabilitation (AAPM&R) Multi-Disciplinary PASC Collaborative (PASC Collaborative) was convened to address the pressing need for guidance in the care of patients with PASC. PASC Collaborative members include experts across a variety of clinical disciplines and specialties including PM&R, neurology, neuropsychiatry, neuropsychology, rehabilitation psychology, and primary care. The PASC Collaborative is following an iterative, modified Delphi process to achieve consensus on assessment and treatment recommendations that have been presented as a series of consensus guidance statements regarding the most common PASC symptoms. A full description of the modified Delphi methodology has been published previously.19 These recommendations are informed by experts from established PASC centers who have experience managing patients with the full range of PASC-related symptoms.20-26 Whenever possible, existing evidence related to mental health assessment and treatment specifically for people with PASC has been incorporated into this guidance statement, and this body of evidence is somewhat limited and evolving. As needed, recommendations have also been guided by applying general, yet well-established, mental health assessment and treatment approaches to unique considerations in people with PASC. The consensus guidance statements include an intentional focus on health equity because disparities in care are a critical factor in contributing to widespread disparities in clinical outcomes, and they must be considered and addressed. The PASC Collaborative's goal is to broaden the understanding of current patient care practices, not only to guide clinical management, but also to identify critical gaps and opportunities for future research. According to the original definition by the U.S. Centers for Disease Control and Prevention (CDC), “Long COVID (PASC) is broadly defined as signs, symptoms, and conditions that continue or develop after initial COVID-19 or SARS-CoV-2 infection. The signs, symptoms, and conditions are present 4 weeks or more after the initial phase of infection; may be multisystemic; and may present with a relapsing–remitting pattern and progression or worsening over time, with the possibility of severe and life-threatening events even months or years after infection. The CDC uses the 4-week timeframe in describing post-COVID conditions to emphasize the importance of initial clinical evaluation and supportive care during the initial 4 to 12 weeks after acute COVID-19.27 Based on patient feedback during our consensus process, we also advocate for improving access to beneficial interventions by facilitating early evaluation, diagnosis, and management of new or worsened symptoms that are experienced after COVID-19 infection. For the purposes of this guidance statement, we recommend assessment for PASC if symptoms are not improving within 1 month after acute symptom onset. Emotional and mood fluctuations are experienced as part of normal everyday life, but persistent, distressing disturbances in emotional regulation, behavior, and/or cognition can be indicative of a mental health disorder. According to the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5) depressive disorders are characterized primarily by persistent sadness, emptiness, irritability, and/or a loss of pleasure or interest in activities (anhedonia). They are often associated with other secondary symptoms including poor concentration, psychomotor agitation or retardation, fatigue or loss of energy, sleep difficulty (insomnia or hypersomnia), feelings of low self-worth or excessive guilt, hopelessness about the future, changes in appetite or weight, and/or thoughts about dying or suicide. Anxiety disorders are characterized by excessive fear, worry, apprehension, or dread that is out of proportion to the perceived threat. PTSD is characterized by intrusive distressing memories, dreams, dissociative reactions (flashbacks), prolonged psychological distress, and/or physiological reactions that occur after exposure to an actual or threatened traumatic event.28 The COVID-19 pandemic resulted in a 28% increase in cases of major depressive disorder (MDD) and a 26% increase in cases of anxiety disorders worldwide in 2020.29, 30 It was also estimated to have caused 137 additional disability-adjusted life years (DALYs) per 100,000 people due to MDD and 116 additional DALYs per 100,000 people due to anxiety disorders, and the incidence of new mental health symptoms experienced after SARS-CoV-2 infection has been higher than after influenza or other respiratory infections.15, 31, 32 Both women and people from underrepresented race and ethnic minority groups have been reported to have a higher incidence of PASC, regardless of the type(s) of PASC sequelae.33, 34 Women also have a higher prevalence of pre-existing anxiety and autoimmune diseases than men, and both conditions are risk factors for developing PASC35 and new or worsening mental health conditions. People who are at increased risk for specifically developing PASC-related mental health disorders include women, older adults, and potentially Black/African American individuals and people from low socioeconomic strata.36, 37 In addition to the direct effect of SARS-CoV-2 infection on mental health, the cumulative burden of chronic stress (allostatic load) has grown due to widespread pandemic-related stressors and concerns such as fear of COVID-19 infection and death, short- and long-term financial uncertainty, quarantine-associated social isolation, familial stress, and survivor's guilt.38 These stressors have been further exacerbated by pre-existing stigma surrounding mental health disorders, disruptions in the health care system which resulted in diminished access to mental health services, and frequent dismissal of patients' reported mental health symptoms by the health care community, particularly early in the pandemic and for those without formal laboratory evidence of SARS-CoV-2 infection.39 Moreover, pre-existing mental health disorders have been associated with worse clinical outcomes related to SARS-CoV-2 infection, including an increased risk of severe acute illness, hospitalization, death, PASC, and severe depression and anxiety symptoms in the setting of PASC.39-44 Although the pathophysiology of SARS-CoV-2 infection is still under investigation, it is surmised that the cytokine storm created by the virus may affect mental health by initiating a central inflammatory response, reducing serotonin production which attenuates downstream monoamine neurotransmission, and increasing brain glutamate and upregulation of N-methyl-d-aspartate (NMDA) receptors.7, 8 In aggregate, these changes produce damaging neuronal excitotoxicity with subsequent synaptic pruning and neuronal loss. This loss is compounded by a cytokine-mediated reduction in brain trophic factors. The cumulative effects of these disruptions may be the induction of new-onset or re-exacerbation of pre-existing neuropsychiatric conditions.8, 9 This role of cytokines in the context of PASC is similar to the pathophysiology of other pro-inflammatory conditions. That is, cytokines are appropriately generated in response to a peripheral infection or other process, and when they cross into the brain, they can cause normal, evolutionarily conserved, protective physiological symptoms of illness allowing the body to cope with the infection, such as fever, lethargy, and malaise.45 However, when activation of the peripheral immune response and pro-inflammatory cytokines is sustained beyond a normal physiological duration (e.g., in the setting of autoimmune disease, systemic infection, cancer, and/or PASC), the resulting immune signaling to the brain can lead to an exacerbation and the development of depressive symptoms in vulnerable individuals.46 In addition, mental health symptoms specifically in the setting of PASC may also be related to other biologic factors that are caused directly by SARS-CoV-2 infection, including persistent viral reservoirs, microvascular blood clots due to endothelial dysfunction, an altered gut microbiome, and autoimmunity.47-50 Many patients with PASC have reported encountering stigmatization, disbelief, and assumptions that the whole of the PASC syndrome is due to their psychological state,51 despite evidence to the contrary.50 This experience has often led to fractured relationships between health professionals and their patients, which has inherently limited the degree to which clinicians have been able to offer supportive interventions to these patients. We encourage clinicians who treat PASC to model their approach based on best practices for the treatment of patients with myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS). Specifically, PASC clinicians should take a supportive, open, nonjudgmental stance; they should consider the multifaceted biological aspects to the complex syndrome and recommend medical and mental health–related treatments when a patient truly requires both types of interventions.52 This can be especially challenging because patients with PASC who are experiencing depression, anxiety, and/or suicidal thoughts may underreport their mental health symptoms due to concerns that their clinician will attribute all their other PASC symptoms to their mental health. This phenomenon may contribute to the trend that these patients are more likely to report receiving a significantly below-optimal level of care.53 Therefore, it is helpful for clinicians to reassure patients and let them know that although not all patients with PASC experience mental health symptoms, they are one of the well-documented manifestations of PASC and have the potential to be severe.53 Intentional and effective assessment of mental health symptoms in patients with PASC can be challenging. Some patients have experienced a delay in accurate diagnosis and appropriate management because their PASC-related symptoms have been dismissed by health care professionals as purely psychosomatic manifestations.6 In light of this context, we suggest that clinicians introduce the topic of mental health by validating that mental health conditions are not, in and of themselves, the cause of PASC.50 However, mental health symptoms can be worsened by, can mimic, and can worsen, coexisting PASC symptoms, which is why assessment of mental health symptoms is essential.2, 54 For example, generalized fatigue and post-exertional malaise as symptoms of depression can both mimic and exacerbate one another, and they can result in similar impairments to a patient's function.55 Although beyond the scope of this statement, the PASC Collaborative developed and published guidance on cognitive and physical fatigue.20, 22 Clinicians should take great care to identify the underlying driver(s) of functional impairment to guide appropriate treatment. When possible, a thorough assessment of a patient's mental health–related symptoms can also be facilitated by collateral information from one or more close friends or family members. Collateral history may especially be helpful when a patient has impaired memory, poor awareness regarding how they are functioning and/or how long their level of function has been impaired, or otherwise would have difficulty independently reporting symptoms (e.g., for sleep-related symptoms such as snoring, dozing off during the daytime, frequent nighttime awakenings, etc.).22, 26 Loved ones can contribute important information regarding the time course and severity of observed changes to the patient's mood, coping abilities, energy level, appetite, sleep, daily function, and participation in usual activities (Table 1). Patients should be screened for signs and symptoms of new or worsening depression, anxiety, or PTSD using validated tools and instruments, as appropriate. Note: Clinicians should consider suicide risk screening for patients who screen positive for mental health conditions and should have an established suicide plan in place prior to screening. Although validated questionnaires can be helpful for efficient evaluation, if a clinician implements one or more mental health screening questionnaires, it is important for the clinician to understand whether the questionnaire assesses somatic symptoms. For instance, the Patient Health Questionnaire (PHQ-9) assesses somatic symptoms such as sleep dysfunction, fatigue, difficulty concentrating, and appetite because they can be manifestations of depressive disorders.56 Nevertheless, these symptoms can also be present due to other PASC-related conditions such as autonomic dysfunction,20-26 so the PHQ-9 can overpredict the presence or severity of depressive symptoms in patients with multiple manifestations of PASC.57 In contrast, the Patient-Reported Outcomes Measurement Information System (PROMIS) Depression scale does not inquire about somatic symptoms to reduce confounding in patients who have comorbidities (such as PASC) that affect physical health.58 Other measures that may be used include the PHQ-2 (for depression screening), Generalized Anxiety Disorder 7 (GAD-7) (for anxiety screening), and Hospital Anxiety and Depression Scale (HADS) (for depression and anxiety screening). Regardless of the validated tool(s) used, a follow-up interview is important to clarify whether the reported mental health symptoms present across settings versus only manifest in select settings (e.g., school/work, home, social contexts), which may guide diagnosis and treatment plans. Common screening tools for PTSD include the PCL-5 (PTSD Checklist for DSM-5) and IES (Impact of Events Scale). However, confirmation of the diagnosis may require a referral to a specialist for more extensive testing. Suspicion for PTSD should especially be raised in patients who are survivors of critical illness, experienced care in an intensive care unit (ICU),59 had a confirmed SARS-CoV-2 infection, are frontline health care workers, and/or were quarantined. PTSD can manifest as flashback-like dissociative reactions; efforts to avoid trauma-related thoughts, feelings, places, and/or people; persistent negative cognition and mood; and symptoms of hyperarousal, such as anxiety, sleep difficulties, and irritability.60-63 The context of delivering a mental health screening tool can affect patients' responses. Although screening questionnaires are often given to patients prior to a clinical encounter and before any clinical context is provided, some patients report answering mental health–related questions more thoughtfully and honestly when: (1) a clinical team member describes the relevance and purpose of the screening tool as it is introduced to them, and (2) the clinician acknowledges the patient's responses on the tool and explains how the results impact treatment recommendations.11 Of note, some clinicians express discomfort in assessing suicide ideation and depressed mood due to a lack of familiarity of immediate next steps if a patient expresses active suicide ideation. The American Psychological Association provides continuing education on assessing and intervening with patients at risk of suicide (https://www.apa.org/monitor/2022/06/continuing-education-intervene-suicide), and Columbia University, through the Lighthouse Project, provides training and disseminates the Columbia Protocol—also known as the Columbia-Suicide Severity Rating Scale (C-SSRS) at https://cssrs.columbia.edu/. Clinicians will need to establish what their plan will be for patients who endorse suicide ideation. A plan may include immediate consultation with an in-clinic mental health specialist such as a social worker or psychologist in an interdisciplinary setting, or contacting a crisis worker, suicide hotline (988 or National Alliance on Mental Illness [NAMI] hotline), or mental health crisis center for more detailed inquiry with the patient and necessary treatment or intervention to maintain safety and care of the patient. If a patient expresses active suicidal ideation, they should not be left unsupervised. It is important to assure the patient of the importance of acknowledging their level of distress and providing immediate care, which may also include transfer or referral to a hospital emergency department and/or emergency psychiatric hospital. Urgent care of suicide ideation is equally as important as other physical health crisis situations and should be managed with equal immediacy. When selecting appropriate screening tools, administration approaches, and collateral history informants, it is essential to consider a patient's chronological/mental age, educational history, and social environments. For example, in patients with a low reading level (in the language in which the screening tool is being administered), screening tools may need to be read aloud rather than self-administered. In these cases, patients may be more sensitive to the perceived social acceptability of their reported symptoms, which may introduce bias into the results of the screening tool. Clinicians who manage patients with neurodevelopmental disorders should take an especially judicious approach to selecting assessment measurements and collateral history informants. Assessment of people with neurodevelopmental and/or sensorimotor disorders may require proxy screening inventories that are focused largely on observed behaviors rather than subjective experiences, and when possible, multiple informants should be used to characterize the patient's clinical course (e.g., symptom onset, functional changes). Nevertheless, to the extent possible based on the patient's comprehension and language abilities, both self and proxy (e.g., caregiver, teacher, etc.) assessments should be incorporated into the diagnostic evaluation. Literature regarding people with intellectual disabilities has suggested mixed agreement in self- and caregiver-report across health-related factors, with better correlation for reported subjective health than for subjective distress, physical activity engagement, and social support.64 It is also important to consider identity factors such as race, ethnicity and sexual orientation and gender identity, and clinicians should make a conscious effort to understand their own related implicit (unconscious) biases. Clinicians must also be aware that cultural differences exist in how mental health symptoms are expressed and/or described such that patients from certain cultures may be particularly hesitant to endorse mental health symptoms due to a concern for social acceptability.65 As appropriate, diagnostic tests should be reviewed and/or obtained to evaluate for medical conditions that can masquerade as symptoms of a mental health disorder. For example, a patient's tachycardia and palpitations can be attributed mistakenly to anxiety when they are manifestations of cardiovascular and/or autonomic dysfunction.23, 24 A higher risk of having a medical condition that presents as, or contributes to, mental health symptoms has been reported in older adults, people with pre-existing medical conditions, people with a prior psychiatric history, and people from lower socioeconomic strata.66 There is considerable overlap in the symptoms of neurologic and psychiatric disorders that commonly manifest in PASC. Commonly overlapping symptoms include pain, disordered sleep, fatigue, apathy, cognitive changes, and mood changes. Major categories of neurologic disease that can mimic psychiatric conditions and contribute to mental health symptoms include neurodegenerative disorders (e.g., Parkinson disease, Huntington's disease, Alzheimer's disease), vascular disorders, inflammatory/infectious/autoimmune disorders, trauma, metabolic/endocrine disturbances, nutritional deficiencies, structural disorders, and primary sleep disorders. Post-intensive care syndrome (PICS) can also manifest as neuropsychiatric symptoms and can coexist with PASC, although movement disorders are more common in PICS than in PASC.67 Factors that may suggest an underlying (or coexisting) neurologic disorder as opposed to an isolated psychiatric disorder include: features atypical for the psychiatric disorder, an unusual age of symptom onset, paroxysmal symptom onset, normal functioning prior to symptom onset, recent substance abuse/intoxication leading to encephalopathy, prescription medicine overuse, and/or symptoms that are resistant or respond unusually to conventional treatment.66 The DSM-5 requires that formal diagnosis of a mental health disorder considers a careful clinical history to assess for these factors. Because emotional wellness is bidirectionally related to a person's ability to engage in daily tasks, it is also important to consider the effects of mental health symptoms on a person's function, and vice versa. A functional history should explore the patient's baseline and current ability to engage in daily self-care activities and instrumental activities of daily living (e.g., bathing, dressing, grooming, ambulating with or without use of adaptive equipment, cooking, cleaning, shopping, attending to household chores, and paying for needs/bills). PASC-related physical and mental health symptoms can also limit a patient's ability to attend and fully participate in work, school, social, and family life activities, and these limitations can, in turn, impact mood, adaptive coping, and satisfaction with life.68 Decreased engagement in these activities can result in added stress, financial strain, and a feeling of isolation, and patients may feel pressured to engage in these activities even if participation worsens their other PASC-related symptoms such as post-exertional malaise. Understanding this patient-specific interplay is an important prerequisite to developing a patient-centered treatment plan that addresses interrelated physical impairments, mood symptoms, and coping skills, regardless of whether symptoms were premorbid versus new-onset with PASC. Clinicians should also inquire about potential triggers that exacerbate a patient's mental health symptoms. Reported potential symptom triggers include, but are not limited to, certain foods, menstrual cycles,69 pre-existing medical conditions,70 prescribed and over-the-counter medications, alcohol and drug use or misuse,71 physical or social stress, and psychosocial factors such as being uninsured or underinsured, being unemployed or underemployed, being socially isolated, having limited social support (e.g., people to assist with transportation, translation, or other needs), and having concerns regarding personal safety, food insecurity, and stable housing.72-74 These psychosocial stressors are more prevalent in people with PASC.72,73 Furthermore, mental health assessment should also consider the patient-specific impact of the COVID-19 pandemic itself, such as: (1) social stress from prolonged social distancing, family illness, and/or disruptions in school, childcare, and/or work, or (2) traumatic stress due to the frequent media coverage of life-threatening illness and/or direct exposure to other people who had severe SARS-CoV-2 infection.75 The clinical course for mental health symptoms in the setting of PASC varies widely from patient to patient. Symptoms may be more severe and prolonged in patients who had a pre-existing mental health condition and/or a more severe acute course of COVID-19.76, 77 Available evidence suggests that the general principles of treating mental health symptoms in the setting of PASC should generally mirror the treatment principles for those who do not have PASC (see Table 2).78, 79 Patients' symptom constellations, pre-existing and/or PASC-related comorbidities, degree of functional impairment, and the extent to which their quality of life is impacted should guide the treatment approach. For patients with symptoms that are impacting everyday life/function and are outside the scope of practice of the treating clinician, consider referral to an appropriate specialist (examples include: social work, therapist, counselor, neuropsychologist, clinical psychologist, and/or psychiatrist) with formal expertise in psychological/psychiatric assessment and treatment. Note: Clinicians should discuss with the patient that mental health symptoms can be worsened by (and can worsen) other coexisting PASC symptoms. It is important to note that the treatment recommendations provided in this guidance statement should not preclude clinical judgment and must be applied in the context of each specific patient, with adjustments for patient preferences, comorbidities, and other factors. In addition, the PASC Collaborative recognizes that patients with PASC typically present with a cluster of symptoms that cross multiple body systems, such as fatigue and dysautonomia, which might limit their ability to fully participate in some of these assessment and treatment recommendations. These co-existing conditions must always be considered when suggesting a test or treatment such as physical rehabilitation. As with any treatment plan, clinicians treating patients with PASC are encouraged to discuss the unknowns of PASC treatments and prognosis, as well as the benefits and risks of any treatment approach. If a patient screens positive for a psychiatric disorder using one of the aforementioned screening tools or in clinical interview, clinicians with adequate training and comfort with psychopharmacologic management may provide in-office care as a means of offering immediate intervention. However, referrals to a mental health specialist (e.g., psychiatrist, psychologist, neuropsychologist, social worker, mental health counselor, etc.) for further evaluation