Rena Conti1, Jesper Akesson2, Elisa Weiss3, Maria Sae-Hau3, Marialanna Lee3, Gabriela Gracia4, Brian Connell3, Lucy Culp3, Robert Metcalfe1,2
1 Boston University, Boston, MA
2 The Behaviouralist, London, UK
3 The Leukemia & Lymphoma Society, Rye Brook, NY
4 Independent Researcher
Cancer patients, and specifically those with blood cancer, are among the most vulnerable to disruptions in care and poor outcomes associated with COVID-19 infection.1,2,3 January 2021 CDC guidelines prioritize cancer patients in the ‘1c’ vaccine group based on this evidence.4 National clinical practice guidelines5 and noted physician advocacy groups6 recommend that blood cancer patients get vaccinated, suggesting that the expected benefits outweigh the risks. However, we do not know the extent to which cancer patients demonstrate vaccine hesitancy.*
We conducted a nationwide online survey of blood cancer patients and survivors about their experiences and attitudes related to COVID-19 and COVID-19 vaccines, using a convenience sample of blood cancer patients and survivors with valid email addresses who are in The Leukemia & Lymphoma Society’s constituent database. In total, 6,516 patients and survivors responded to the survey, for a response rate of nearly 7%.
The main outcome is vaccine attitudes, specifically respondents’ answers to the question: ‘Imagine that your doctor offers you a COVID-19 vaccine for free in January 2021. How likely are you to choose to get the vaccine?’ scored on a five-point Likert scale (very unlikely, unlikely, neither likely nor unlikely, likely, and very likely). Rationales for answers to this question were recorded and categorized. We also asked respondents: 1) whether and how COVID-19 impacted their medical care, finances, and social support; 2) about their COVID-19-related beliefs and engagement with protective health-related behaviors (e.g., wearing a mask or washing their hands frequently); and 3) about their demographic characteristics. The survey was fielded between December 1-21, 2020.
We present mean responses to the main outcome. We conduct t-tests of differences in proportions to evaluate how vaccine attitudes are related to other health beliefs and behaviors. We also conduct three Linear Probability Model (LPM) regressions to correlate participants’ vaccine attitudes with other observable variables (e.g., their demographic characteristics and whether they responded to the survey before or after the Pfizer-BioNTech and Moderna vaccines had been emergency authorized in the US [beginning 10 December 2020]).8,9 The outcomes for these regressions are three binary variables. The first variable takes the value of one if the respondent is ‘very unlikely’ or ‘unlikely’ to take a vaccine, and zero otherwise. The second variable takes the value of one if the respondent is ‘neither likely nor unlikely’ to take a vaccine, and zero otherwise. The third variable takes the value of one if the respondent is ‘likely’ or ‘very likely’ to take a vaccine, and zero otherwise. We verify the robustness of the regression results by estimating a multinomial logit model and by re-weighting the data so that the population better reflects the overall population of blood cancer patients in the US (by re-weighting on race/ethnicity, gender, and age). We only report results that are statistically significant on a 5% level. We use the statistical software Stata (version 16) to conduct the data analysis. The Institutional Review Board of Boston University approved this study.
The average age of respondents is 64, 59.8% are female, and 70% have an Associate degree or more (Table 1). Further, 86.5% identify as White or Caucasian, 6.5% identify as Black or African American, and 4.7% identify as Hispanic or Latino/a.
We find that 17% of respondents indicate that they are unlikely or very unlikely to take a vaccine (Figure 1). Among those who are vaccine hesitant (i.e., those that state that they are ‘unlikely’ or ‘very unlikely’ to get vaccinated), the two most common reasons (both 54%) are concerns regarding side effects and a belief that the vaccines “will not be/have not been tested properly” (Figure 2). Relative to other respondents, those who are vaccine hesitant are significantly (14.6 percentage points) more likely to say that they do not think they would get hospitalized if they contracted COVID-19 (Supplement 1). They are also significantly (5 percentage points) more likely to say that they do not think they will contract COVID-19.
Vaccine hesitant respondents are also significantly less likely to engage in protective health behaviors (Supplement 2). They are, for example, 3.8 percentage points less likely to say that they have worn a face mask and 1.6 percentage points more likely to say that they have taken no other protective measures.
Using LPM regression, we find that vaccine hesitancy is significantly predicted by age; the older participants are, the more likely they are to say they will take a vaccine if it becomes available (Supplement 3). Vaccine hesitancy is less articulated among respondents who identify as white and more articulated among women (7.5 percentage points). Those who reside in rural or suburban areas are also more likely (on average, 6.2 and 9.4 percentage points respectively) to be vaccine hesitant compared to respondents residing in urban areas. Moreover, the one third of participants who responded after COVID-19 vaccines had been emergency authorized were significantly more likely (3.7 percentage points) to say that they would get vaccinated than those surveyed before authorization.
None of the aforementioned results change appreciably when re-weighting the data (for age, gender, and race/ethnicity) or using alternative statistical specifications.
It is unknown whether these results generalize to wider groups of cancer patients or racial and ethnic populations underrepresented in the survey population or those with lower levels of education (although our re-weighted analysis suggests that the results do generalize). It is also not clear how stated intentions to take a COVID-19 vaccine translate into actual vaccination behavior.10-12
We conducted the largest and most timely survey of cancer patient and survivor attitudes, experiences, and behaviors related to COVID-19. We found that a little less than one in five cancer patient and survivors report vaccine hesitancy. The most frequent rationales for vaccine hesitancy are consistent with the fact that the currently available vaccines were not tested on these patient types (cancer patients and survivors were excluded from the trials), but inconsistent with recent recommendations of public health and cancer experts who believe the benefits of vaccination outweigh the risks.13
Our finding regarding the prevalence of vaccine hesitancy can be compared to that of the Kaiser Family Foundation (KFF) survey fielded coincidently at about the same time in December 2020, which reported lower rates of vaccine hesitancy. KFF found that 71% of the general population would definitely or probably get vaccinated if it were determined to be safe by scientists and available for free vs. 70% among our sample respondents. KFF also found that among households in which someone had a serious health condition (defined to include cancer), 78% would definitely or probably get vaccinated.14 While it may be surprising that our figure is slightly lower than these estimates (given the more significant risks of morbidity and mortality among blood cancer patients who contract COVID-19), the greater degree of hesitancy among blood cancer patients may also reflect greater underlying uncertainty regarding not only the safety but also the efficacy of COVID-19 vaccines for this specific population. In fact, although national guidelines and physician groups recommend that blood cancer patients be vaccinated, they note that the efficacy (i.e., the degree to which cancer patients develop immunity when vaccinated) is unknown, and there is concern that patients with a blood cancer will have a lower immune response than the general public.5
We also found that participants became more positive toward taking a COVID-19 vaccine once the FDA had approved the Pfizer-BioNTech and Moderna vaccines. This suggests that hesitancy may be influenced by emerging information dissemination, government action and vaccine availability, transforming the hypothetical opportunity of vaccination to a real one.
Patients, physicians and policymakers need data on the effectiveness of COVID-19 vaccines in cancer patients to inform practice. To overcome vaccine hesitancy, clear, consistent and culturally appropriate messaging targeting cancer patients, emphasizing the risks of COVID-19 infection and the benefits of vaccination may be needed. It is also important to increase diversity of participants in COVID-19 vaccine trials, so that participants are from communities most impacted by the virus.15 Guidelines might also consider the benefits of offering vaccinations to caregivers of cancer patients to reduce COVID-19 transmission and improve access to care.
*Vaccine hesitancy, according to the World Health Organization, is the “delay in acceptance or refusal of safe vaccines despite availability of vaccination services.”7