Risk factors for COVID-19 infection and poorer outcome

Patients with SSD are postulated to be at a higher risk of acquiring COVID-19 and having a poorer health outcome.^{Reference Kozloff, Mulsant, Stergiopoulos and Voineskos2,Reference Shinn and Viron4} The increased risk of poorer health outcomes are attributed to factors such as higher rates of disadvantageous lifestyle, residential instability and smaller social networks.^{Reference Maguire and Looi5,Reference Druss6}

Patients with schizophrenia and related disorders are more likely to be over represented in congregated living situations, which puts them at additional risk for COVID-19 infection. Moreover, deficits in cognition and judgement among people with SSD may interfere with implementation of recommended measures such as hand washing, physical distancing, etc.^{Reference Kozloff, Mulsant, Stergiopoulos and Voineskos2,Reference Shinn and Viron4,Reference Brown, Gray and Lo Monaco47} Patients with SSD are reported to be at potentially higher risk of developing respiratory infections, in particular in the presence of other underlying medical conditions and lifestyle risk factors.^{Reference Fonseca, Diniz, Mendonça, Malinowski, Mari and Gadelha7}

Furthermore, patients with SSD have high rates of smoking and medical comorbidities such as cardiovascular disease, diabetes and respiratory illnesses; all of which are known to be associated with poorer prognosis in people who acquire COVID-19 infection.^{Reference Kozloff, Mulsant, Stergiopoulos and Voineskos2,Reference Bartels, Baggett, Freudenreich and Bird8,Reference Brown, Gray and Lo Monaco47} This population is also reported to have higher rates of impaired lung function, and higher rates of intensive care unit (ICU) admissions and morbidities when admitted to hospital for pulmonary conditions.^{Reference Fonseca, Diniz, Mendonça, Malinowski, Mari and Gadelha7} Additionally, patients with SSD are generally more likely to experience disparities in accessing primary care, and are more likely to have undiagnosed or untreated underlying medical conditions.^{Reference Kozloff, Mulsant, Stergiopoulos and Voineskos2,Reference Kahl and Correll9} Other potential risk factors include those related to treatment of SSD itself, namely increased risk of hypersalivation and aspiration as a side-effect of treatment with some antipsychotics as well as, risk of toxicity during acute medical illness.^{Reference Kozloff, Mulsant, Stergiopoulos and Voineskos2}

Higher risk of coagulopathy is another consideration for patients with SSD who acquire COVID-19 infection.^{Reference Mongan, Cannon and Cotter10} COVID-19 has been associated with hypercoagulation, and current guidelines recommend routine use of thromboprophylaxis in patients with COVID-19 infection. At the same time, patients with SSD have been reported to have a two- to three-fold increased risk of deep vein thrombosis or pulmonary embolism,^{Reference Lin, Chung, Chen and Chien51} likely multifactorial in nature, including biological dysregulation of the illness itself, medications and lifestyle. As such, it is hypothesised that patients with SSD who contract COVID-19 may be at higher risk of thrombotic complications, considering that both illnesses increase such risk.^{Reference Mongan, Cannon and Cotter10}

In addition to health risks related to COVID-19 infection in patients with SSD, there is concern about worsening of underlying medical conditions as a result of lack of access during the pandemic.^{Reference Melamed, Hahn, Agarwal, Taylor, Mulsant and Selby11} In general, patients with SMI, including patients with SSD, have high rates of chronic medical comorbidities and higher rates of undertreated conditions compared with the general population. Restrictions on access to ‘non-urgent’ care during the COVID-19 pandemic may result in further disruption to physical care and/or worsening of the untreated underlying medical condition in this patient population.^{Reference Melamed, Hahn, Agarwal, Taylor, Mulsant and Selby11}

Strategies to increase health support for patients with SSD

A number of strategies were suggested to help increase health support for patients with SMI, including those with SSD. Some of the proposed strategies included providing up-to-date information tailored toward this particular patient population.^{Reference Maguire and Looi5,Reference Druss6} Moreover, additional support to maintain healthy lifestyle habits and continuation of required care for underlying medical conditions were recommended.^{Reference Druss6}

Other authors suggested an individualised approach to patient education, to increase adherence with infection control measures.^{Reference Brown, Gray and Lo Monaco47}

It is also recommended that mental health services provide clear direction and remote contact channels to ensure continuity of care and reduce the risk of exposure. Continuous support for adherence with medication and the consideration of medication delivery are also suggested.^{Reference Fonseca, Diniz, Mendonça, Malinowski, Mari and Gadelha7}

Impact of the pandemic and restrictions

There is an increasing concern about the impact of the current pandemic on the mental health of the general population, and in particular, on patients with pre-existing mental illnesses. It is hypothesised that a number of factors may lead to worsening of psychosis in patients with SSD; these factors include fear and stress caused by the pandemic, infection with the virus, distress and isolation among those who acquire the virus, and treatment of the viral illness with steroids and other agents.^{Reference Kozloff, Mulsant, Stergiopoulos and Voineskos2}

Furthermore, the societal restrictions of the pandemic may have a negative impact on patients with SSD. In general, patients with SSD have smaller social networks, and therefore, further restricting their already tenuous support may be associated with negative consequences. Moreover, social isolation has been associated with poorer quality of life and paranoia,^{Reference Hamada and Fan12} as well as being a risk factor for suicide among patients with SSD. A potential increase in substance use during social isolation has also been presumed,^{Reference Hamada and Fan12} which could lead to further deterioration in this patient population. As such, maintaining connectedness through virtual means, as well as continuation of essential community services, is paramount for this patient population.^{Reference Kozloff, Mulsant, Stergiopoulos and Voineskos2}

Another consideration is the economic burden of the pandemic on this patient population. Patients with SSD are more likely to have jobs without sick-leave or benefits, and are more vulnerable to loss of employment.^{Reference Shinn and Viron4}

In light of reports from the Swine flu outbreak, Hamada and Fan^{Reference Hamada and Fan12} propose potential worsening of symptoms of obsessive–compulsive disorder, a common comorbidity in patients with SSD. Fears regarding acquiring infection may result in increased obsessive thoughts, which could add to functional limitations associated with SSD.

The ‘infodemic’, a surfeit of information about COVID-19 infection, has also been argued to potentially cause exacerbation of psychotic symptoms.^{Reference Hamada and Fan12} Reduction of media exposure has been suggested by some authors, with the rationale that overexposure to broadcasts of stressful situations has been associated with negative mental health outcomes.^{Reference Fonseca, Diniz, Mendonça, Malinowski, Mari and Gadelha7}

Iasevoli et al^{Reference Iasevoli, Fornaro, D'Urso, Galletta, Casella and Paternoster13} interviewed 205 patients with SMI 1 month after quarantine in Italy, and compared this group to 205 controls and 51 relatives of patients. Their results showed patients with SMI were four times more likely to perceive higher stress related to the pandemic than controls, and had two to three times higher risk of significant anxiety and depressive symptoms. They reported similar outcome when they separated patients with schizophrenia from those with mood disorders.

Changes in care delivery

Uninterrupted access to psychiatric care is imperative for patients with SMI, including those with SSD, to prevent decompensation.^{Reference Hategan and Abdurrahman14} During the COVID-19 pandemic, many out-patient services have been replaced by virtual means, such as telephone check-ins and telepsychiatry visits. On the other hand, many services have been cancelled, such as day programmes and peer support groups.^{Reference Shinn and Viron4,Reference Hamada and Fan12} Although the nature of the pandemic and health risks of COVID-19 infection require such measures, the impact of changes in mental healthcare delivery in this patient population has yet to be examined.^{Reference Kozloff, Mulsant, Stergiopoulos and Voineskos2}

Torous and Keshavan^{Reference Torous and Keshavan15} discussed the need for mental health support for patients with SSD (and patients with SMI in general) during the pandemic, and suggested the use of smartphones as an effective way to implement telepsychiatry. They discussed the potential for implementation of apps in the clinical care of patients with SSD, including creating personalised care plans, monitoring symptoms and relapse prediction.

In addition to changes to out-patient services, access to some of the evidence-based community services, such as assertive community treatment and case management, might become limited or replaced by virtual methods.^{Reference Kozloff, Mulsant, Stergiopoulos and Voineskos2}

Like many other services, hospital-based and in-patient psychiatric care have undergone changes during the COVID-19 pandemic. To date, not much is known about the potential impact of these necessary changes. For instance, following major outbreaks in the in-patient psychiatric units of Wuhan and South Korean hospitals, psychiatric wards have embraced preventative measures, such as prohibition of visitors and external food/clothing,^{Reference Kozloff, Mulsant, Stergiopoulos and Voineskos2} or total lockdown of the in-patient units.

Other care delivery changes include laboratory requirements, monitoring of patients taking clozapine and access to long-acting injectable (LAI) antipsychotic medications for patients with SSD. Although clozapine monitoring guidelines have proposed exceptions to the regular blood monitoring requirements (see the guideline section), patients who are in the first year of treatment with clozapine still require weekly or biweekly blood work, which may put them at increased risk of exposure. During the pandemic, regulatory agencies such as the USA Food and Drug Administration have implemented clozapine-monitoring programmes, allowing for a ‘grace period’ of up to 56 days for clozapine to be dispensed without blood work. However, when contacted, one pharmacy in a small community indicated they would only dispense up to 2 weeks of clozapine if a patient was late for blood work.^{Reference Leung, Wittenberger and Schak16} The authors highlighted the need for a collaborative approach to ensure continued care for patients taking clozapine.

At the same time, restricted access to hospitals under the advice of health authorities may create a challenge for patients who are on LAI antipsychotics. Ifteni et al^{Reference Ifteni, Dima and Teodorescu17} reported on data from a psychiatric hospital in Romania. Their results revealed a 49% reduction in the use of risperidone LAI antipsychotics, and a 70–90% reduction of aripiprazole, paliperidone and olanzapine LAI medication use in March 2020, compared with the 3-month period before the outbreak. Of concern is the increased risk of non-adherence and relapse subsequent to switching an LAI medication to oral treatment.

Nichols et al^{Reference Nichols, Gannon and Conlogue18} reported on changes implemented in their clozapine clinic, which serves 184 patients, during the pandemic. They shifted in-person nursing assessments to telepsychiatry visits, started to deliver clozapine for curb-side pick-up or through the mail, and implemented blood work exemptions (up to 3 months) as per USA Food and Drug Administration recommendations. These strategies significantly reduced the in-person visits, and therefore the risk of transmission of COVID-19.

Grover et al^{Reference Grover, Mishra, Chakrabarti, Mehra and Sahoo19} reported on the use of telecommunication to monitor patients receiving clozapine in an under-resourced area in India. These telephone appointments included information on symptoms of COVID-19, when to present to an emergency department and education regarding continuation of clozapine and regular blood work monitoring. Through these telephone calls, they were able to reach 205 out of 227 patients on clozapine. Of those contacted, 81.5% were in touch with their treating physician. Overall, 96.6% of patients stayed on the same dose of clozapine; however, about one fourth had difficulty obtaining clozapine because of a lack of availability at their local pharmacy. Only 24.4% were able to have blood work done in the previous month. The authors highlighted the importance of telephone monitoring of patients on clozapine in under-resourced areas. At the same time, this study also draws attention to the disproportionate impact of the pandemic on patients with SMI, including those with SSD, and in particular, those in lower socioeconomic areas.

Kopelovich et al^{Reference Kopelovich, Monroe-DeVita and Buck20} described changes in healthcare delivery across the USA, and made suggestions to better align these changes with current standards of practice. They discussed the use of multimodal service delivery, including a combination of telehealth, in-person clinics, community outreach, other digital interventions and warm lines. Warm lines were described as support lines for patients with no family or other sources of support. Such a multimodal service plan will provide flexibility based on clients’ needs and barriers to care. The authors also recommended development of COVID-19-compatible advance directives. Further recommendations included assessment through telehealth visits and encouraging clinicians to adapt to the patient population. For instance, patients with psychosis may not be able to tolerate a lengthy visit over telehealth, and in such cases, shorter but more frequent visits are suggested. For patients with SMI, and in particular patients with SSD, it is better for the first assessment to be done in person, considering the importance of building rapport, which could be challenging over telehealth, especially for patients with paranoia or disorganisation. They also recommended a careful risk–benefit assessment when deciding on the mode of care delivery for clients at elevated safety risks (i.e. suicide/violence). In certain cases, prioritisation for in-person visits is essential for management of psychiatric morbidity. Other recommendations include consideration of telepsychotherapy to address anxiety and insomnia, management of substance use, and addressing primary care and medication needs.

Case reports

To date, a number of case reports have been published on both new onset of psychotic symptoms and worsening of pre-existing psychosis in the context of COVID-19 infection. Of the ten case reports that we found, two studies each reported on a case with no previous psychiatric history and a brief psychotic episode triggered by the stress of the pandemic.^{Reference Huarcaya-Victoria, Herrera and Castillo21,Reference Zulkifli, Sulekha and Guan22} There were three case reports with new-onset non-reactive psychosis: one in a patient self-medicated with chloroquine;^{Reference Akkaoui, Lejoyeux and Geoffroy23} one in a patient who had been recently admitted to hospital for COVID-19 infection and received treatment including hydroxychloroquine and prednisone;^{Reference Correa-Palacio, Hernandez-Huerta, Gómez-Arnau, Loeck and Caballero30} and one in a patient who was admitted for COVID-19 infection and developed psychosis requiring a second admission to the hospital.^{Reference Majadas, Pérez, Casado-Espada, Zambrana, Bullón and Roncero31} Three studies reported a relapse of psychosis triggered by the stress of the pandemic^{Reference Fischer, Coogan, Faltraco and Thome35,Reference Ovejero, Baca-García and Barrigón36} or quarantine.^{Reference Sanchez-Alonso, Ovejero, Barrigon and Baca-Garcia37} In one of these cases, the patient had incorporated COVID-19 into their delusional system, and developed symptomatic COVID-19 infection during their hospital stay;^{Reference Ovejero, Baca-García and Barrigón36} and in another case, the patient was found to have a positive COVID-19 test result.^{Reference Sanchez-Alonso, Ovejero, Barrigon and Baca-Garcia37} The two remaining case reports were also about patients with pre-existing psychotic illness and COVID-19 infection: one presented with relapse of psychosis and delirium^{Reference Palomar-Ciria, del Valle, Hernández-Las Heras and Martínez-Gallardo39} and one presented with clozapine toxicity.^{Reference Cranshaw and Harikumar40}

In a Letter to the Editor, de Burgos-Berdud et al^{Reference de Burgos-Berdud, Valdés-Florido and López-Díaz24} reported on four studies that described reactive psychosis or psychotic symptoms triggered by the COVID-19 pandemic in a total of eight healthcare workers. They highlighted the need for providing support for healthcare workers during the pandemic.

Case series of reactive psychosis or psychotic relapses

Martin et al^{Reference Martin, González, Pérez, Sesmero, González and Mendoza32} reported on a case series from Spain (with an unknown number of cases) that had no previous psychiatric history and presented with COVID-19 infection and acute psychosis. They described subacute onset (less than a week) and fast recovery (within 2 weeks) in response to low-dose antipsychotic treatment.

In Italy, Finatti et al^{Reference Finatti, Pigato, Pavan, Toffanin and Favaro25} described three cases without any previous psychiatric history, who presented with brief psychotic disorder in the context of social isolation during the COVID-19 pandemic.

In another case series study from Spain, Valdés-Florido et al^{Reference Valdés-Florido, López-Díaz and Palermo-Zeballos26} reported on four cases of brief psychotic disorder during the first 2 weeks of national quarantine. Two of the cases presented with serious suicidal behaviour. The authors discussed a potential rise in cases of reactive psychosis as a result of the COVID-19 pandemic.

Shanbour et al^{Reference Shanbour, Khalid and Fana27} presented three cases with first-episode psychosis and paranoid delusions explicitly about COVID-19. One individual had delusions about the cell phone tower beside his apartment spreading COVID-19, another individual believed she was pregnant and that her negative pregnancy tests were because of COVID-19 infection, and another individual had delusions about being God's son and that he had predicted the COVID-19 pandemic 3 years prior. The authors highlighted the impact of the pandemic, the quarantine and infodemics on the precipitation of psychosis in vulnerable individuals.

Chandra et al^{Reference Chandra, Shiva, Nagendrappa, Ganjekar and Thippeswamy28} reported on two cases from India presenting with psychosis related to COVID-19. Both cases were precipitated by fear of acquiring COVID-19, on a background of personal vulnerabilities interfacing with sociocultural issues. The authors discussed COVID-19 as a social illness resulting in high stress, in particular in vulnerable individuals.

Ferrando et al^{Reference Ferrando, Klepacz, Lynch, Tavakkoli, Dornbush and Baharani33} reported on three cases, all of individuals aged 30–39 years, who presented with new-onset psychosis and were found to have a COVID-19 infection that was otherwise asymptomatic. They had elevated inflammatory markers, and particularly C-reactive protein. None of these cases were overtly preoccupied or worried about COVID-19 infection. All cases were treated with low-dose antipsychotics. The authors discussed potential immune-mediated psychopathology in these cases, and drew attention to ‘COVID-19 psychosis’ as a potential complication of COVID-19, requiring further exploration.

In Spain, Parra et al^{Reference Parra, Juanes and Losada34} reported on a retrospective study of ten patients with COVID-19 infection who developed psychotic symptoms. They excluded delirium cases. The mean age was 54.1 years. Only one case had no physical manifestation of COVID-19 infection, and their presentation was purely psychiatric. In 80% of cases, the onset of psychotic symptoms occurred 2 weeks after the initial symptoms of COVID-19, and these symptoms resolved in less than 2 weeks, with low-dose antipsychotics. The most frequent symptoms included structured delusions and confusion, although the latter could be explained in the context of medical illness (80% had bilateral pneumonia) and ICU treatment (50% of this sample). The authors discussed possible mechanisms of COVID-induced psychosis, including direct central nervous system involvement, indirect effect through neuroinflammation and metabolic disturbance, and iatrogenic disease via medications for treatment of COVID-19 and its complications.

Case series of patients with SSD with suspected COVID-19 infection

Liu et al^{Reference Liu, Lin, Jiang, Li, Zhong and Su38} reported on 21 in-patients with schizophrenia who were transferred to an isolation unit of a psychiatric hospital for suspected COVID-19 infection. The comparison group included 30 in-patients with schizophrenia without any COVID-related symptoms. Only one patient had positive COVID-19 test results, but another 11 patients met the diagnostic criteria for clinically confirmed cases. The Positive and Negative Syndrome Scale score was similar in both groups. However, compared with controls, patients with suspected COVID-19 infection scored significantly higher on depression, anxiety, perceived stress and sleep quality scales. About half of the patients with suspected COVID-19 (52.4%) required an increase in dose or addition of medications for management of their psychiatric symptoms.

Psychotic experiences in students

Hajdúk et al^{Reference Hajdúk, Dančík and Januška29} assessed psychotic experiences among students from Comenius University in Bratislava, using surveys 1 year before and during the initial phase of the COVID-19 pandemic. There were no significant changes in the scores across the study time points. However, they found a significant relationship between positive psychotic experiences and negative affectivity, in particular depressive symptoms. The authors discussed that the lack of change in psychotic experiences in this study could be explained by milder severity of the pandemic in the country or by some protective factors such as increased contact with family and higher social cohesion. Furthermore, this study was conducted at earlier phases of the pandemic and the impact on psychosis may take longer to manifest.

Case–control studies

In a nationwide study in Korea, Ji et al^{Reference Ji, Huh and Kang41} assessed comorbidities in patients who tested positive for COVID-19 compared with those with negative results. The study included 219 961 individuals who were tested for COVID-19 infection up until 15 May 2020, using reimbursement data and the KCD-7 diagnostic codes. Use of respiratory support was considered as a proxy measure for severity of COVID-19 infection. In patients with COVID-19 infection, comorbidities with significant association included diabetes (odds ratio range 1.2), osteoporosis (odds ratio range 1.1), rheumatoid arthritis (odds ratio range 1.2), substance use (odds ratio range 1.3) and schizophrenia (odds ratio range 1.6). Comorbidities associated with severe COVID-19 were all medical (i.e. diabetes, end-stage renal disease, etc.) and non-psychiatric. The authors concluded that the higher rate of comorbidity with schizophrenia is likely a result of outbreaks in psychiatric units or group homes rather than the vulnerability caused by the illness itself.