Psychosocial perspectives among cancer patients during the coronavirus disease 2019 (COVID‐19) crisis: An observational longitudinal study

Abstract Background The coronavirus disease 2019 (COVID‐19) crisis and consequent changes in medical practice have engendered feelings of distress in diverse populations, potentially adversely affecting the psychological well‐being of cancer patients. Aim The purpose of this observational longitudinal study was to evaluate psychosocial perspectives among patients with cancer on intravenous treatment during the COVID‐19 pandemic. Methods and results The study recruited 164 cancer patients undergoing intravenous anti‐neoplastic therapy in a tertiary cancer center. Psychosocial indices were assessed at two points in time, corresponding with the beginning of the first wave of COVID‐19 pandemic in Israel (March 2020) and the time of easing of restrictions implemented to curtail spread of infection (May 2020). At Time 1 (T1), elevated COVID‐19 distress levels (score 1 and 2 on 5‐point scale) were observed in 44% of patients, and associated with pre‐existing hypertension and lung disease in multivariate analyses but no demographic or cancer related factors. At Time 2 (T2), 10% had elevated anxiety and 24% depression as indicated by Hospital Anxiety and Depression Scale (HADS‐A/D). COVID‐19 distress at T1 was related to higher levels of HADS‐A at T2 (Spearman 0.33 p < .01), but not HADS‐D. Patients with breast cancer expressed greater COVID‐19 distress compared with other cancer types (p < .01), while both HADS‐A and HADS‐D were highest for patients with GI cancer. Patient report of loneliness and decreased support from relatives were factors associated with HADS‐A (p = .03 and p < .01, respectively), while HADS‐D was not similarly related to the factors evaluated. Conclusion Patients with cancer undergoing intravenous treatment may be vulnerable to acute adverse psychological ramifications of COVID‐19, specifically exhibiting high levels of anxiety. These appear unrelated to patient age or disease stage. Those with underlying comorbidities, breast cancer or reduced social support may be at higher risk.

with underlying comorbidities, breast cancer or reduced social support may be at higher risk.  19) has engendered in many a state of uncertainty and panic. Unfavorable mental health has been observed among healthcare workers, 1,2 infected patients 3 and the general population. 4 Early on, oncological patients were presumed to be at increased risk for COVID-19 infection, leading to modification of official treatment guidelines and preferred therapeutic modalities 5,6 ; these changes were based on limited evidence, 7,8 and at the potential risk of detriment to patient outcomes. 9,10 Cancer patients exhibit higher rates of anxiety and depression than those without cancer during ordinary times. 11,12 In the midst of the pandemic, the psychological impact on cancer patients has magnified, carrying both a dread of infection, as well as a concern for maintaining optimal treatment deliveryand demand for emotional assistance has soared. 5 Many have adopted a self-imposed lockdown. It has been shown previously that psychological distress can affect cancer progression and patient prognosis by modulating intricate bio-behavioral processes. 13,14 Social isolation and loneliness have also been associated with mortality. 15 On the other hand, a resilience despite the inherent distress of a cancer diagnosis has been identified in patient subgroups as a positive coping mechanism. 16 The COVID-19 pandemic reached Israel in February 2020, and enforcement of social distancing rules began in March 2020. On March 19, a state of national emergency was declared and a national lockdown prohibited citizens from leaving their homes unless for activities deemed essential. A gradual easing of restrictions began in early May. A second wave of COVID-19 infection was announced during July. Located in the northern city of Haifa, Rambam Medical Center (RMC) is the tertiary referral center for 12 district hospitals, serving approximately 20% of the total nationwide population. 17 Throughout the pandemic, cancer treatment services were maintained with minimal modifications, and patients were permitted to visit the hospital as usual.
The main objective of our study was to evaluate the psychosocial consequences of COVID-19 on patients with cancer undergoing intravenous treatment in the oncology center during the peak of the crisis in Israel (March-May 2020). The cohort was longitudinally followed over the course of a month; first subjects completed a survey of demographics and general distress associated with COVID-19, and ultimately, a validated assay of anxiety and depression. We evaluated indices of psychological well-being and emotional and functional burden, considered the existence of more vulnerable subgroups of patients, and investigated a possible association between scores at different points in time.

| Design and procedure
The present study included patients with cancer receiving intravenous treatment administered at the ambulatory unit of the oncology center or inpatient service, within RMC. Inclusion criteria included patients of age 18 or more, with a cancer diagnosis within the past year, undergoing intravenous oncological treatment in RMC, and capable of reading and writing in the national language, Hebrew. Those with hematological malignancies were excluded, as treated in a separate department in our institution, as were patients who were not on active intravenous treatment (oral drugs, radiation or other local therapy, surgery or best supportive care). All participants signed a written informed consent. During the study period of March-May 2020, subjects completed two questionnaires for sociodemographic information as well as emotional and functional burden data. The study protocol was approved by the Institutional Ethics Committee.

| Measures
Two questionnaires were administrated at baseline (Time 1, T1) and 1 month later (Time 2, T2) to evaluate participants' psychological state and psychosocial parameters.

| COVID-19 distress
The T1 questionnaire available in Data S1 (Supplement 1) included demographic data, clinical variables and co-morbidities, and a general assay of distress in relation to COVID-19, using a scale from 1 (high distress) to 5 (low distress). Subjects were instructed to answer in relation to COVID-19 specifically, and not a general state. A single item measure, this was meant as a general survey to determine how wary patients were of the new COVID-19 situation.

| Anxiety and depression symptoms
At T2, after the national lockdown and period of peak of infection, and with the results of T1, it was decided to distribute a validated questionnaire to assess and quantify patient psychological indices for comparison.
The Hospital Anxiety and Depression Scale (HADS) questionnaire was chosen. HADS is a 14-item validated screening tool with two subscales for anxiety (HADS-A) and depression (HADS-D). For each subscale, a cutoff HADS score of equal to or greater than 8 is considered suggestive for anxiety or depression. 18 Designed for patients with organic diseases, HADS excludes somatic symptoms of emotional distress that may be caused by cancer, resulting in high efficacy in clinical practice, 19 and moreover was found especially useful for screening patients in the context of COVID-19. 20 For the current data set, Cronbach's alpha was 0.76 (95% CI: 0.71, 0.82).

| Emotional and function burden due to the COVID-19 pandemic
The T2 questionnaire included a series of questions designed to assess additional factors of emotional and functional burden. These were developed by the study team psychologist and social worker (available in Data S1). Parameters were analyzed as single items and a composite score was not computed.

| Statistical analysis
Statistical analyses were performed using the R statistical software. 21 All tests were 2-sided with a significance level of α = 0.05. A multivariate rank regression model was used to analyze the association between general COVID-19-associated distress and demographic and co-morbidity data, where "general COVID-19 distress score" (ranking 1-5) was the dependent variable (Table 1). Association between "general COVID-19 distress score" at T1 and HADS-A and HADS-D at T2 was assessed using Spearman's rank correlation. Univariate Fisher and Mann-Whitney tests correlated HADS-A and D scores with key domains of emotional and functional burden.

| Participants
This study recruited 164 patients with cancer aged 23-90 undergoing intravenous treatment at the RHCC oncology center. Within the patient cohort, 60% had metastatic disease and 40% had loco-regional cancer. Most common malignancies were breast (26%), lung (25%) and gastrointestinal cancer (25%), followed by genitourinary, head and T A B L E 1 Rank regression with dependent variable "general anxiety score" (ranked [1][2][3][4][5] (Figure 2A,B). Of groups with over 15 subjects, HADS-A was highest among breast cancer patients, and HADS-D in those with GI cancer, however HADS-D was lowest among patients with breast cancer ( Figure 3 and Table 4). Despite small sample size, a largely abnormal mean depression score of over 11 was observed only in patients with sarcoma. Anxiety (HADS≥8) was not associated with gender,   Figure 4B).

| DISCUSSION
The mental health outcomes in frontline healthcare workers, 1 and depressive symptoms were even observed among school-age children. 23 As the pandemic spread to Europe, similar studies exposed mental health issues in Spanish populations, 24 healthcare workers in Italy, 2 and F I G U R E 4 Spearman rank correlation between COVID-19 distress at T1 and HADS scores at T2. Patients with higher initial COVID-19 distress scores at T1, later manifested greater anxiety at T2 (A). Similar results were not demonstrated for depression; at T2, depression levels centered homogeneously around 6 and were not associated with COVID-19 distress at T1 (B) more. A review of 62 studies from 17 countries found high mental burden among medical staff and general public, however distress was most evident among patients. 25 Studies generally delineate psychological indices from a snapshot in time, rather than a longitudinal follow-up more likely to mitigate the chance of an acute surge in distress that abated over time.
The present study aimed to longitudinally evaluate psychosocial perspectives of patients with cancer undergoing intravenous treatment during the time of peak COVID-19 infection, and found initially elevated COVID-19 distress levels that predicted patient anxiety after 1 month, as well as loneliness and feelings of decreased social support. Nevertheless, similar correlations were not found for depression.
Compared to the general population, depression is known to be relatively more common among cancer patients, especially those with metastatic disease. In the present study, depression levels were but not shown here. Patients of younger age and with localized disease have been shown to manifest resilience to psychological distress in spite of a cancer diagnosis. 16 Here, age and stage did not affect distress, suggesting that general parameters of prognosis do not allay COVIDstress response. The lack of difference between cancer type and stage may be due to patient receipt of intravenous treatment, which may be a factor in determining vulnerability. This analysis evaluated patients receiving intravenous treatment, which may define a group of patients with cancer who are more aware of reductions in immune system function and possible susceptibility to COVID-19 infection and complications, as opposed to those treated with surgery, receiving radiotherapy or other local treatment, oral medication at home or none at all. Finally, social support is considered an important contributor to improving well-being and reducing distress in cancer patients both in ordinary and extraordinary times. 16 In international databases, a steep decline in oncology-related procedures and visits has been observed. 28 anxiety as a unique COVID-19 related issue and providing further distinction between cancer-anxiety and COVID-19-anxiety. It can be assumed that actual levels may in fact be higher as patients who did not arrive for treatment due to self-isolation were not represented.
Large interventional controlled trials are needed to support findings of patient distress and determine ramifications on patient treatment or outcome.
While in some parts of the world the pandemic is abating, in other parts, peak levels are being experienced, and in still others, harbingers of additional waves are emergingregardless, oncology teams should anticipate emotional distress among patients and prepare in advance.
Availability of accurate information and the use of precautionary measures as hand-washing and wearing a mask, have been shown to correlate with lower COVID-19-related anxiety and depression. 29 Evidence-based clear guidelines and detailed treatment plans should be conveyed and implemented. Active interventions should be pursued, such as psychological counseling, interactive virtual meetings, encouragement of maintaining healthy diet and exercise. 5 Meaningful guidance and multidisciplinary support should aim to reduce feelings of hopelessness and lack of control. In conclusion, sufficient attention should be reserved for the psychological care of cancer patients concerning the current pandemic, with the aim of alleviating anxiety especially at vulnerable points in time, in order to minimize risk without compromising care.

ACKNOWLEDGMENT
This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.