Impact of comorbidities on people with and without cancer early in the COVID-19 pandemic: An observational study

Abstract

Background: The present study sought to investigate how comorbidity burden influences cancer survivors' quality of life (QoL) and the challenges/adaptations during the coronavirus disease 2019 (COVID) pandemic, and to examine how appraisal processes are related to this impact.

Methods: This cross-sectional study, administered in spring/summer 2020, compared cancer survivors to a general-population comparison sample. QoL was assessed with standardized tools. COVID-specific questions included selected items compiled by the US National Institutes of Health, and cognitive appraisal processes were assessed using the QoL Appraisal Profile_v2 Short-Form. Principal components analysis reduced the number of comparisons. Multivariate analysis of covariance investigated group differences in QoL, COVID-specific variables, and cognitive-appraisal processes. Linear regression investigated group differences in COVID-specific variables as a function of cognitive-appraisal processes, QoL, demographic covariates, and their interactions.

Results: Cancer survivors fared substantially better than non-cancer participants in QoL and cognitive functioning when they had no other comorbidities, but substantially worse on QoL when they had three or more comorbidities. Cancer survivors with no comorbidities were less likely to feel worried about COVID, less likely to engage in self-protection, and prioritized engaging in problem-focused and prosocial actions compared to non-cancer participants. Conversely, cancer survivors confronted with multiple comorbidities exhibited more proactive self-protection and experienced more anxiety about the pandemic.

Conclusion: The impact of having multiple comorbidities in the context of cancer is associated with notable differences in social determinants of health, QoL outcomes, COVID-specific challenges/adaptations, and appraisal of QoL. These findings provide an empirical basis for implementing appraisal-based coping interventions.

Keywords: COVID; cancer survivor; comparison; quality of life; social determinants of health.

FIGURE 1

Number of comorbidities for cancer versus non‐cancer groups. Cancer survivors reported more comorbidities than non‐cancer participants across all groupings overall (χ ² = 77.7, df = 3, p < 0.0001), and particularly in the 2, and 3+ groups (11% and 84% more).

FIGURE 2

(A, B) Comparing the QoL composite and Neuro‐QoL Applied Cognition scores for cancer versus non‐cancer groups as a function of comorbidity level. As number of comorbidities increased, there was a downward trend in QoL and worsened cognitive functioning (F _{Wilk's Lambda} = 8.6, df = 6, p < 0.0001). Significant interactions between comorbidity level and cancer group for both the QoL and Applied Cognition outcomes (F _multivariate = 5.0 and 3.2, p = 0.002 and 0.023, respectively) revealed group differences on QoL among those with zero comorbidities and three or more comorbidities (post hoc comparison p = 0.006 and 0.019, respectively); and on Applied Cognition among those with zero comorbidities (post hoc comparison p = 0.002). Asterisks highlight significant interaction effects (**p < 0.01, *p < 0.05).

FIGURE 3

(A–D) Comparing COVID‐specific composite scores for cancer versus non‐cancer groups as a function of comorbidity level. Cancer survivors differed from non‐cancer participants on the COVID‐specific composite scores both overall (F _{Wilk's Lambda} = 14.6, df = 4, p < 0.0001) and as a function of comorbidity level (F _{Wilk's Lambda} = 2.6, df = 12, p = 0.002). Cancer survivors reported less growth/social support overall (F _{Between‐subjects} = 14.8, p < 0.0001), compared to non‐cancer participants. There was a significant interaction between cancer group and comorbidity level for the COVID protection and worry and covid altruism and coping composite scores (F _{Between‐subjects} = 4.2 and 5.4, p = 0.006 and 0.001, respectively). Post hoc comparisons revealed significant group differences on COVID protection and worry among those with zero comorbidities and three or more comorbidities (p = 0.025 and 0.039, respectively), and on COVID altruism and coping among those with zero and two comorbidities (p < 0.0001 and 0.05, respectively). Asterisks highlight significant interaction effects (**p < 0.01, *p < 0.05).

FIGURE 4

(A–F) Comparing appraisal composite scores for cancer versus non‐cancer groups as a function of comorbidity level. Cancer survivors differed from non‐cancer participants on the appraisal composite scores (F _{Wilk's Lambda} = 3.3, df = 6, p = 0.003) as did comorbidity levels (F _{Wilk's Lambda} = 2.5, df = 18, p < 0.0001). Cancer survivors generally engaged in fewer negative‐emphasis appraisals than non‐cancer participants (F _{Between‐subjects} = 12.6, p < 0.0001, E). As number of comorbidities increased, problem goals and health goals also increased overall (F _{Between‐subjects} = 3.5 and 5.4, p = 0.014 and 0.001, respectively, B, C).

FIGURE 5

(A, B) Radar charts showing explained variance in the three sets of models for cancer versus non‐cancer comparison groups. This components‐of‐variance finding suggests that the impact of demographic variables, comorbidity status, and QoL on COVID‐specific variables can be explained because one appraises differently.