Mental Health and Adherence to COVID-19 Protective Behaviors among Cancer Patients during the COVID-19 Pandemic: An International, Multinational Cross-Sectional Study

Full reference: Kassianos, A.P., Georgiou, A., Kyprianidou, M., Lamnisios, D., Lubenko, J., Presti, G., Squatrito, V., Constantinou, M., Nicolaou, C., Papacostas, S., et al. (2021). Mental health and adherence to COVID-19 protective behaviors among cancer patients during the COVID-19 pandemic: An international, multinational cross-sectional study. Cancers, 13, 6294. https://doi.org/10.3390/cancers13246294

Study type: International empirical cross-sectional study

Context and Objectives

This study addresses an important research gap regarding the impact of the COVID-19 pandemic on mental health and protective behaviors in cancer patients. Although it has been extensively documented that cancer patients represent a high-risk population for COVID-19 with elevated mortality rates (10-30% in single-center studies, and 25.6% in pooled analyses), little information exists about how this population manages the psychological impact of the virus and adherence to preventive behaviors. The study was motivated by reports of adverse COVID-19-related outcomes in cancer patients, including severe complications and increases in mortality. The researchers specifically sought to understand: (1) the level of adherence to COVID-19 protective behaviors in cancer patients compared to non-cancer participants, and whether factors associated with adherence differed between groups; (2) the impact of the pandemic and lockdown measures on mental health and emotional status of cancer patients compared to non-cancer participants, and whether factors associated with elevated stress differed between groups; and (3) whether cancer patients perceived COVID-19 as a bigger threat than cancer itself and whether this perception affected their psychological outcomes.

Method

Participants

The study enrolled 9,565 individuals from 78 countries between April 7 and June 7, 2020, during the first wave of the COVID-19 pandemic when the World Health Organization had declared a state of emergency in most participating countries. Of these total participants, 264 (2.8%) reported being cancer patients. The median age of cancer patients was 51.5 years (interquartile range 40-60) compared to 34 years (26-45) for non-cancer participants (p < 0.01). The majority of cancer patients were female (79.9%), compared to 77.6% of females in the non-cancer group. Cancer patients reported worse financial losses (44% versus 25%, p = 0.02) and higher stress levels (median 15, interquartile range 10-21) compared to non-cancer men (median 12, interquartile range 9-17, p = 0.03). Participants with cancer resided in 30 different countries. Regarding cancer types, the majority had breast cancer (28%), followed by female reproductive system cancers (22%). Approximately one third (31%) reported receiving adjuvant therapy at the time of survey, while 12% received treatment for active disease, and the remaining 57% were not receiving active treatment.

Study Design

A population-based, anonymous, web-based cross-sectional study design was employed. A 20-minute online survey was distributed through 33 participating universities, sent to students and academic staff, and through local press, social media, professional networks, local health centers, social institutions, and professional group email lists in participating countries. Data collection occurred over a two-month period during which a COVID-19 emergency state was declared in most participating countries.

Instruments

Multiple validated and well-established measures were utilized. Perceived stress was assessed using the Perceived Stress Scale (PSS). Depressive symptomatology was measured with the Multidimensional State Boredom Scale (MSBS). Wellbeing was evaluated with the Mental Health Continuum Short Form for Adults (MHCSF) and the Positive and Negative Affect Scale (PANAS). Psychological flexibility was measured with the Psychological Flexibility Scale (Psy-Flex). Social support was evaluated using the Oslo Social Support Scale (OSS), and family functioning with the Brief Assessment of Family Functioning (BAFF). Prosocial behavior was measured with the Prosocial Scale for Adults (PSA). Mindfulness was assessed with the Cognitive and Affective Mindfulness Scale (CAMS). To evaluate beliefs about COVID-19, a modified version of a questionnaire measuring Health Belief Model parameters of perceived susceptibility and perceived severity was used. COVID-19 protective behaviors (isolating, keeping distance, hand washing) were assessed through questions about adherence to recommended national guidelines, using Likert-type scales ranging 0-10. Specific questions were included for cancer patients regarding cancer type, whether they received anti-cancer therapy, and whether therapy was adjuvant (preventive) or for active disease. Participants were also asked about their feelings toward cancer and COVID-19 as health priorities.

Data analysis

Descriptive analysis was conducted presenting means ± standard deviation (SD) for continuous variables with normal distribution, and median and interquartile range (IQR) for measures not following normal distribution. Pearson's chi-square test was employed to detect differences between variables on cancer diagnosis versus non-cancer, with the variable of current priority (cancer versus COVID-19). T-test and Kolmogorov-Smirnov tests were applied to detect differences between cancer and non-cancer participants, using predefined cut-off values for interpreting effect sizes. For the protective behaviors hypothesis and mental health hypothesis, multivariate linear regression modeling was performed to evaluate the significance of different psychological and behavioral parameters on stress and the level of physical distancing, adjusting for different sociodemographic and socioeconomic characteristics, and psychological flexibility. Statistical analysis was conducted using STATA 14.0 with statistical significance set at α = 0.05.

Results

Comparisons between Cancer Patients and Non-Cancer Participants

Cancer patients reported significantly greater adherence to COVID-19 protective behaviors compared to non-cancer participants. On the behavior of maintaining physical distance when going out, cancer patients achieved a median of 10 on 0-10 scale (IQR 9-10) versus 9 (8-10) in non-cancer participants (p < 0.01, d = 0.19 small effect size). Regarding self-isolating and limiting unnecessary travel according to national guidelines, both groups showed median of 10 (9-10) with no significant differences (p = 0.05, d = 0.03). For regularly washing hands with water and soap, again no differences were observed (p = 0.04, d = 0.17). However, self-efficacy in protective behaviors was significantly higher in cancer patients (median 6.6, IQR 5.9-7) versus non-cancer participants (median 6.2, IQR 5.6-7, p < 0.01, d = 0.20 small effect size). A higher proportion of cancer patients (67%) reported having the intention to follow national recommendations for social distancing in the following week compared to non-cancer participants (57%, p < 0.01).

Coping Styles, Social Support, and Family Functioning

No significant differences were identified between the two groups regarding coping with stressors including using emotional support, humor, and self-blame. However, a higher proportion of cancer patients (31%) reported greater social support compared to non-cancer participants (23%), although with very small effect size (p < 0.01, d = 0.03). No significant differences were observed in family functioning between both groups.

Mental Health Outcomes

Cancer patients reported significantly lower stress (median 15, IQR 10-20) compared to non-cancer participants (median 17, IQR 12-22) with medium effect size (p < 0.01, d = 0.27). There were differences in the three dimensions of wellbeing: eudaimonic psychological wellbeing was higher in cancer patients (median 23, IQR 18-26) versus non-cancer participants (median 21, IQR 15-24, p < 0.01, d = 0.24). On all dimensions, cancer patients reported higher wellbeing levels, with 50% reporting flourishing wellbeing status and 41.9% with moderate mental health status. In contrast, non-cancer participants showed lower flourishing proportion (39.6%) and higher languishing proportion (10.2%). Cancer patients reported higher levels of positive affect (median 30, IQR 25-37) versus non-cancer participants (median 29, IQR 23-35, p = 0.01, d = 0.21 small effect size), reflecting greater propensity to experience positive emotions. They reported lower levels of negative affect (median 26, IQR 20-33) versus non-cancer participants (median 28, IQR 25-38, p < 0.01, d = 0.19), indicating lower tendency to experience the world in a negative way. Cancer patients demonstrated being more psychologically flexible (p < 0.01, d = 0.37 large effect size), holding their thoughts lightly and taking action guided by what is important to them even in challenging situations. Additionally, they reported being more mindful compared to non-cancer participants (p = 0.04, d = 0.19).

Predictors of Stress and Adherence to Protective Behaviors

Among cancer patients, the multivariate model adjusted for all predictors identified that it was more likely they adhered to maintaining physical distance when feeling less susceptible to COVID-19, perceiving the disease as more severe, having financial losses due to the pandemic, and being more psychologically flexible (model 5). In a second multivariate model adjusted for all predictors, females reported higher stress levels compared to males. Higher stress levels were associated with greater perceived susceptibility to COVID-19, inability to obtain basic supplies, and lower psychological flexibility (final model 5).

Comparisons According to Current Health Priority

Participants with cancer were investigated according to whether they reported cancer as their primary health concern (cancer priority group, n = 52) or COVID-19 as their top priority (COVID-19 priority group, n = 191). Impressively, 80.3% of cancer patients reported that COVID-19 was their top priority at that time compared to 19.7% who continued to view cancer as their top priority.

Regarding COVID-19 protective behaviors, no statistically significant differences were observed between the cancer priority group (56%) and the COVID-19 priority group (73%) on protective behaviors and self-efficacy. However, it was more likely that those in the COVID-19 priority group (78%) reported that following recommended national guidelines was important to them compared to the cancer priority group (56%, table 5). Specifically, 62.1% of the COVID-19 priority group and 51.9% of the cancer priority group reported keeping distance from other people when going out all the time, respectively. Additionally, 73.2% of the COVID-19 priority group reported self-isolating and limiting unnecessary travel according to national guidelines while the corresponding percentage for the cancer priority group was 55.8%. Finally, a similar percentage in both the cancer priority group (65.3%) and the COVID-19 priority group (61.5%) reported washing their hands regularly with water and soap.

In terms of mental health outcomes, no significant differences were observed between the two groups on perceived stress, mindfulness, prosocial behavior, boredom, mental wellbeing, and psychological flexibility (table 5). Those in the cancer priority group reported higher positive affect (median 32, IQR 28.5-38) compared to those in the COVID-19 priority group (median 30, IQR 24-37, p < 0.03, d = 0.34 large effect size) suggesting greater likelihood of experiencing positive emotions and interacting with others positively despite challenges in their lives.

Discussion and Conclusions

This study provided a unique perspective on the impact of the COVID-19 pandemic on the mental health of cancer patients through an international population-based cross-sectional study that included 264 cancer participants and 9,301 non-cancer participants from 78 countries. The main findings contradict the study's initial hypothesis in several important aspects.

Contrary to expectations, cancer patients demonstrated greater adherence to COVID-19 protective behaviors compared to non-cancer participants. This greater adherence likely reflects their understanding of their increased vulnerability as a high-risk population, their prior experience dealing with health-related threats stemming from their cancer diagnoses, and their ability to maintain psychological flexibility in the face of challenging and unpredictable situations. The researchers suggest that cancer patients' prior experience managing uncertainty related to their diagnoses may have contributed to their more effective adaptation to the COVID-19 crisis. Data collected during the first wave when no COVID-19 vaccine was available may have specifically affected how individuals perceived their susceptibility to the disease.

Contrary to the second hypothesis, this study found that cancer patients experienced less stress and had better eudaimonic psychological wellbeing, superior coping, higher positive affect, lower negative affect, and greater psychological flexibility compared to those without cancer. This interesting finding may be related to individuals who have experienced cancer developing abilities to better cope with new health threats compared to those who have never experienced such threats. The stress levels experienced by cancer patients in this study were similar to those of frontline healthcare workers found in a recent study. This interesting finding suggests that cancer patients may have adapted their perspective to accept their experiences and engage in what is important to them even in challenging situations that become difficult. Half of the cancer patients in the study were off cancer treatment and may have been long-term survivors, which may partially explain lower perceived stress levels, higher level of acceptance, and higher levels of social support and intention to help others in need. These individuals may as a result of their cancer experience have become better adapted and functioning well, resilient and adjustable, prepared to deal with what would otherwise be a worldwide crisis.

Regarding beliefs about COVID-19, this study found that cancer patients considered themselves more susceptible to COVID-19 and were more afraid of it compared to cancer at the time of the first study (first wave). Among these cancer patients, more than 80% considered COVID-19 as their main priority without important differences in behavioral adherence to protective behaviors among those whose priority was cancer versus COVID-19. This study demonstrated that cancer patients worldwide felt threatened by COVID-19. Cancer patients felt more susceptible to being infected with COVID-19 which they considered a severe disease compared to non-cancer participants. The majority of these patients (65%) reported that COVID-19 scared them more and that COVID-19 could harm them more than anything else (71%). It could be assumed that this relates to reports in early days of the pandemic suggesting that the prevalence of COVID-19 in cancer patients was high and that cancer patients comprised a high-risk population for COVID-19. Indeed, frequency, duration, and diversity of media exposure are related to COVID-19-associated fear in the general population, and cancer patients are likely to have less screen time compared to non-cancer participants, and therefore likely had less exposure to multimedia and scary information compared to non-cancer participants.

Three factors were identified consistently associated with both perceived stress and adherence to protective behaviors: the Health Belief Model parameters of perceived susceptibility and perceived severity and the Acceptance and Commitment Therapy construct of psychological flexibility. COVID-19 is a context where people's attitudes toward the disease strongly influence how they feel and behave, and to understand how the Health Belief Model can be used to understand health behaviors, it is important that the literature describes how it can be used to understand health behaviors. Psychological flexibility is comprised of inter- and intra-personal skills that refer to behaviors relating to recognizing and adapting to situations via making changes around what is most meaningful and functional for the person. There are abundant bodies of evidence demonstrating that individuals who exhibit psychological flexibility adapt better to significant life stressors.

Importance and Contribution

This study provides important evidence that cancer patients managed the first unpredictable wave of the COVID-19 pandemic well in an efficient manner and with a positive attitude towards otherwise frightening unknown situations. Although these good results should not lead to a dismissal of the specific needs of cancer patients and how the virus affects the cancer population, research results suggest that dedicated cancer population specific and longitudinal studies are warranted to ensure adequate medical and psychological care for these patients. Future research should examine longitudinally how the pandemic has affected the cancer population, especially in the long term. As the virus progressed and reports of patients being unable to attend follow-up appointments and diagnostic procedures being postponed, these may bear lasting psychological effects. All these need to be considered so that cancer patients receive adequate medical and psychological care.

---