key: cord-0429082-r0o2djrs authors: Mehra, K.; Markoulakis, R.; Kodeeswaran, S.; Redelmeier, D.; Sinyor, M.; MacKillop, J.; Cheung, A.; Levitt, E. E.; Addison, T.; Levitt, A. J. title: The impact of mental health and substance use issues on COVID-19 vaccine readiness: a cross sectional community-based survey in Ontario, Canada date: 2021-09-03 journal: nan DOI: 10.1101/2021.08.30.21262844 sha: e74951c44ba83a13b1a41c71b1eab7ef957cf7fb doc_id: 429082 cord_uid: r0o2djrs Background: COVID-19 vaccines have been approved for use in Canada since December 2020. However, data about factors associated with vaccine hesitancy and the impact of mental health and/or substance use (MHSU) issues on vaccine uptake are currently not available. The goal of this study was to explore factors, particularly MHSU factors, that impact COVID-19 vaccination intentions in Ontario, Canada. Methods: A community-based cross-sectional survey with recruitment based on age, gender, and geographical location (to ensure a representative population of Ontario), was conducted in February 2021. Multinomial logistic regression was used to test the relationship between COVID-19 vaccination status and plans and sociodemographic background, social support, anxiety about contracting COVID-19, and MHSU concerns. Results: Of the total sample of 2528 respondents, 1932 (76.4%) were vaccine ready, 381 (15.1%) were hesitant, and 181 (7.1%) were resistant. Significant independent predictors of vaccine hesitancy compared with vaccine readiness included younger age (OR=2.11, 95%CI=1.62-2.74), female gender (OR=1.36, 95%CI=1.06-1.74), Black ethnicity (OR=2.11, 95%CI=1.19-3.75), lower education (OR=1.69, 95%CI=1.30-2.20), lower SES status (OR=.88, 95%CI=.84-.93), lower anxiety about self or someone close contracting COVID-19 (OR=2.06, 95%CI=1.50-2.82), and lower depression score (OR=.90, 95%CI=.82-.98). Significant independent predictors of vaccine resistance compared with readiness included younger age (OR=1.72, 95%CI=1.19-2.50), female gender (OR=1.57, 95%CI=1.10-2.24), being married (OR=1.50, 95%CI=1.04-2.16), lower SES (OR=.80, 95%CI=.74-.86), lower satisfaction with social support (OR=.78, 95%CI=.70-.88), lower anxiety about contracting COVID-19 (OR=7.51, 95%CI=5.18-10.91), and lower depression score (OR=.85, 95%CI=.76-.96). Interpretation: COVID-19 vaccination intention is affected by sociodemographic factors, anxiety about contracting COVID-19, and select mental health issues. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted September 3, 2021. ; https://doi.org/10.1101/2021.08.30.21262844 doi: medRxiv preprint Identifying predictors of COVID-19 vaccination hesitancy and resistance is crucial to implementing a successful vaccination campaign. In April 2020, a study conducted in the United States by Fisher et al. 1 found younger age, Black race, lower education, and prior missed vaccinations were independent predictors of COVID-19 vaccination hesitancy. Similarly, a study conducted in the United Kingdom and Ireland found that age, gender, ethnicity, geographical location, socioeconomic status (SES), and political affiliation were associated with COVID-19 vaccination hesitancy. 2 A recent survey in Ontario, Canada, 3 found 17.2% participants were unwilling to receive the COVID-19 vaccine, and that females and those with lower education were more likely to be unwilling.; however, data on the impact of MHSU issues on COVID-19 vaccination acceptance in Canada are currently unavailable. 7 Compared with the general population, people living with mental health and substance use (MHSU) issues have been shown to be at higher risk of contracting, and to have higher morbidity and mortality from, COVID-19 infection. 4, 5 Adverse medical outcomes from contracting COVID-19 are more substantial for people with severe mental illness than for those with less severe mental illness. 6 Thus, the goal of the current study was to understand the predictors of COVID-19 vaccination hesitancy and resistance in a representative population of the province of Ontario, Canada, with a focus on the impact of MHSU issues on these intentions. The current study involved a community-based, cross-sectional survey of 2528 Ontarians, age 18 years or older, recruited as a provincially representative sample through a respondent panel: Delvinia's AskingCanadians. Individuals registered with AskingCanadians were randomly . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted September 3, 2021. ; https://doi.org/10.1101/2021.08.30.21262844 doi: medRxiv preprint sent a unique survey link. Interlocking quotas, based on known age, gender, and regional population (Toronto, Southwestern, Eastern, Central, and Northern) proportions were used. 6, 7 The aim of the interlocking quotas for age and gender was to ensure provincial representativeness based on a difference less than five percentage points or within five percent of the mean. Although regional population quotas were utilized, lower density regions outside of the largest city, Toronto, were oversampled to ensure adequate sample across all five regions of Ontario, based on the power needed to detect the known prevalence of depression in each region, within a 5% margin of error. 8 The study was approved by the Sunnybrook Health Sciences Centre Research Ethics Board. Participants gave informed consent prior to participating. The survey comprised 63 items and was open from February 22 to March 15, 2021. Of those participants who were eligible and provided consent, the survey completion rate was 79%. Participants were asked about COVID-19 vaccination intention with responses reflecting whether or not they planned to get COVID-19 vaccination once available/eligible, had received the first or both doses of the vaccination, had heard about the vaccination but were undecided, had heard about it and did not plan to be vaccinated, and had not heard about the vaccination. Sociodemographic data and COVID-19 related information. Age, gender, ethnicity, geographical location, education level, marital status, SES, 9 and living situation information was collected. Satisfaction with social support since the onset of the pandemic (e.g., friends, family, community, co-workers, pets, etc.) was assessed via seven-point bipolar Likert scale (1=extremely satisfied/7=extremely dissatisfied). COVID-19 exposure risk was assessed using . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted September 3, 2021. ; https://doi.org/10.1101/2021.08.30.21262844 doi: medRxiv preprint 6 an original 12-item checklist, with responses reflecting higher risk (being tested and diagnosed with COVID-19, being told they had COVID-19 by a professional, someone they lived with diagnosed with COVID-19, someone else close to them diagnosed with COVID-19, someone close to them passed away as a result of COVID-19) and lower risk (suspecting they had COVID-19 but not diagnosed, tested for COVID-19 but negative, someone they lived with was tested for COVID-19, someone close to them was tested for COVID-19, self-quarantine due to travel, healthcare worker dealing with COVID-19 patients, confirmed COVID-19 cases at places they visit). Fear of self and/or someone close contracting COVID-19 was assessed on two separate items (but later grouped together for analysis) via five-point Likert scale (1=extremely worried/5=not worried at all). 10 This measure has good test-retest reliability and is clinically useful in Canadian samples. 11 The Cronbach's alpha, in the current study, for subscales with more than 1 item ranged from .69 to .85 and for the overall scale (20 items) was 0.94. Substance use was assessed by the World Health Organization Alcohol, Smoking, and Substance Involvement Screening Test (ASSIST v3.0). 12 Substances assessed included tobacco, alcohol, cannabis, and opioids. This scale categorizes participants into low, moderate, and high risk categories for substance use disorder and has good to excellent internal validity. 13 Statistical analysis was conducted using SPSS 26.0. Vaccination intention was divided into those who: planned to receive the vaccination or had received the first/both doses (vaccine . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted September 3, 2021. ; https://doi.org/10.1101/2021.08.30.21262844 doi: medRxiv preprint ready), were undecided (vaccine hesitant), or did not plan to receive the vaccination (vaccine resistant). Univariate analyses were conducted using chi-square tests and ANOVA to compare vaccine ready, hesitant, and resistant based on sociodemographic variables, COVID-19-related variables, and MHSU variables. Multinomial logistic regression was then conducted to compare vaccine hesitant or resistant to vaccine ready participants. Variables included in the logistic regression were those that were significant in the univariate analyses. In the multinomial logistic regression which compared vaccine ready participants with both vaccine hesitant or resistant participants, the independent variables included were age (median split, since continuous variable did not have a linear relationship with its logit), gender (women vs. men), ethnicity (Non-Caucasian vs. Caucasian), geographical location (GTA vs. outside GTA), education (completed college/university vs. not completed college/university), marital status (married vs. unmarried), and anxiety about self and/or someone close contracting COVID-19 (not worried vs. worried, since the continuous variable did not have a linear relationship with its logit). Scores on the depression, anxiety, suicidal ideation, psychosis, and repetitive thoughts and behaviors subscales from the DSM-5 CCSM, social support, and SES since onset of pandemic were entered as continuous variables. Tobacco, cannabis, and opioid use were entered as categorical variables (no/low risk vs moderate/high risk). The mean age of the study sample was 48.26 years (within 5% of the provincial average age of adults 18 years or older of 48.14 years.) 14 The gender distribution was 49.8% women and 48.9% men (within 5% of the provincial distribution of 51.2% women and 48.8% men.) 15 Out of 2528 participants, 1819 (72.0%) planned to get COVID-19 vaccination when available to them, 58 (2.3%) had received the first dose of the vaccination, 55 (2.2%) had . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted September 3, 2021. were less satisfied with social support as compared with vaccine ready participants. Those who . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted September 3, 2021. ; https://doi.org/10.1101/2021.08.30.21262844 doi: medRxiv preprint 9 were vaccine hesitant ( 2 =17.376, p<.001) and vaccine resistant ( 2 =176.384, p<.001) were significantly more likely to be worried about self and/or someone close contracting COVID-19 (Table 2 ). There was no significant difference between vaccine hesitancy, resistance, and readiness based on risk of COVID-19 (Table 2) . Univariate tests for vaccination intention: Mental health issues. Using ANOVA, there were significant differences between the mean scores on depression, suicidal ideation, anxiety, psychosis, and repetitive thoughts and behaviors sub-scales based on vaccination intention (Table 3) Univariate tests for vaccination intention: Substance use issues. There was a significant difference between vaccine ready, hesitant, and resistant participants in the proportion with moderate/high risk for substance use disorder (Table 4 ). Those who were vaccine hesitant were significantly more likely to be associated with moderate/high risk of tobacco use disorder ( 2 =6.513, p=.011), cannabis use disorder ( 2 =6.679, p=.010), and opioid use disorder ( 2 =4.871, p=.027) compared to vaccine ready; and vaccine resistance was significantly more likely to be associated with moderate/high risk of tobacco use disorder ( 2 = 8.906, p=.003), cannabis use disorder ( 2 = 9.926, p=.002), and opioid use disorder ( 2 = 13.532, p<.001) compared to vaccine readiness. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted September 3, 2021. This study found that higher scores on the depression measure independently differentiated participants who were vaccine ready compared to those who were hesitant or resistant; a novel finding, to the best of our knowledge. Depressed participants may be more introspective or concerned with their health, leading to higher readiness for vaccination. Alternatively, participants with depression may see vaccination as a concrete way to remove stressors from their lives. In addition, depressed participants may experience cognitive distortions contributing to negative interpretations of the pandemic, 16 contributing to vaccine readiness. A study conducted by Palgi et al. 17 found that among those who had already received the first dose of the vaccination, vaccine hesitancy was related to higher levels of depression. Our results do not support that finding, which may reflect a difference in the sampling, the tools . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted September 3, 2021. ; https://doi.org/10.1101/2021.08.30.21262844 doi: medRxiv preprint used to assess mental health issues, or geographic or cultural differences. Additionally, in the current study, no other MHSU issues, including anxiety disorders, significantly differentiated those who were vaccine hesitant or resistant from those who were vaccine ready. This finding suggests it is not reasonable to ascribe vaccine hesitancy or resistance to any large degree to most MHSU issues. However, depression was found to be associated with vaccine readiness and this relationship warrants further exploration. Age, gender, ethnicity, education, marital status, and SES were found to be significant independent predictors of vaccine hesitancy or resistance, with younger participants, females, and those who identified with lower SES since the onset of the pandemic significantly more likely to be vaccine hesitant and resistant, similar to previous studies in Canada and other countries. [1] [2] [3] 18 Lower level of education and identifying with Black ethnicity significantly distinguished between vaccine readiness and vaccine hesitancy, but not between vaccine readiness and resistance. Of note, marital status was the only other variable that distinguished the hesitant and resistant participants, such that unmarried participants were less likely to be vaccine resistant. Exploring the possible reasons for being undecided but not unwilling are out of the scope of the current study and have been explored in other studies. 2 Nonetheless, this finding might suggest that more intensive education and appropriate information targeting these groups may enhance the proportion of the population who might be vaccine ready in the future. Those participants who were vaccine resistant were significantly less satisfied with their social support since the onset of the pandemic than those who were vaccine ready. It is possible that people with fewer social interactions and lower satisfaction with those interactions may have less communication about COVID-19 vaccination through friends, family members, co-workers, community groups, etc. and therefore less information about the potential benefits of . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted September 3, 2021. ; https://doi.org/10.1101/2021.08.30.21262844 doi: medRxiv preprint vaccination. Furthermore, people who are happy with their social interactions may have the desire to continue or to expand their interactions, leading them to be more receptive to vaccination. Similar to previous findings, participants who were vaccine ready were more worried about themselves or someone close to them contracting COVID-19. 18 Thus, consistent with other evidence, relationships may play a crucial role in COVID-19 vaccination intention, in that those with close relationships may feel a sense of social duty or moral obligation to protect those around them, 19 or the fear of contracting the COVID-19 infection may lend to vaccine readiness. 18, 20 Limitations. This study was limited to Ontario, and may not be generalizable outside the province. Furthermore, the sample was randomly selected from participants registered with a respondent panel, and the survey was only available online and in the English language. Although this group may not be representative of the entire population of Ontario, the sample had similar characteristics to the population of Ontario. 14, 15 Moreover, the vaccine readiness of individuals of a small proportion of the population may change over time. 21 This cross-sectional design was also only able to identify associations between variables, and did not allow us to determine whether these findings have been stable as the vaccine rollout has progressed. For example, the spread of the COVID-19 Delta variant, after the current survey was conducted, may affect vaccine readiness. While other studies have shown stability in vaccine readiness over time, 22, 23 study of factors influencing vaccine readiness over time is warranted. Furthermore, the DSM-5 CCSM has been validated for use as a screening questionnaire, and is not meant to reflect severity or diagnosis of mental illness. Nonetheless, the internal reliability of the severity measure was adequate based on the excellent Cronbach's alpha. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted September 3, 2021. ; https://doi.org/10.1101/2021.08.30.21262844 doi: medRxiv preprint 13 This study provides data to decision makers to take into account as they work to enhance vaccine uptake and to ensure that certain populations are provided targeted and helpful information to make an informed choice about COVID-19 vaccination. Data access Data presented in this manuscript is available upon reasonable request. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted September 3, 2021. ; https://doi.org/10.1101/2021.08.30.21262844 doi: medRxiv preprint . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted September 3, 2021. Note: * = Only women and men were included in the analysis, other genders were excluded. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted September 3, 2021. ; https://doi.org/10.1101/2021.08.30.21262844 doi: medRxiv preprint . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted September 3, 2021. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted September 3, 2021. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted September 3, 2021. ; https://doi.org/10.1101/2021.08.30.21262844 doi: medRxiv preprint Attitudes Toward a Potential SARS-CoV-2 Vaccine : A Survey of U.S. Adults Psychological characteristics associated with COVID-19 vaccine hesitancy and resistance in Ireland and the United Kingdom COVID-19 vaccine perceptions and differenced by sex, age, and education in 1367 community adults in Ontario Increased risk of COVID-19 infection and mortality in people with mental disorders: analysis from electronic health records in the United States COVID-19 risk and outcomes in patients with substance use disorders: analyses from electronic health records in the United States Ontario Demographic Quarterly: Highlights of first Highlights: Factsheet 2: Age and Gender Profiles of Ontario International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity Relationship of subjective and objective social status with psychological and physiological functioning: Preliminary data in healthy white women The DSM-5 Self-Rated Level 1 Cross-Cutting Symptom Measure as a Screening Tool American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders (DSM-5®) The Alcohol, Smoking and Substance Involvement Screening Test (ASSIST): Development, reliability and feasibility Validation of the alcohol, smoking and substance involvement screening test (ASSIST) Table 17-10-0005-01 Population estimates on July 1st Statistics Canada. Ontario [Province] and Canada [Country] (table). Census Profile Mental Well-Being During Pandemic: The Role of Cognitive Biases and Emotion Regulation Strategies in Risk Perception and Affective Response to COVID-19. Front Psychiatry No psychological vaccination: Vaccine . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder doi: medRxiv preprint hesitancy is associated with negative psychiatric outcomes among Israelis who received COVID-19 vaccination Vaccine hesitancy: the next challenge in the fight against COVID-19 COVID-19 vaccination readiness among nurse faculty and student nurses COVID-19 vaccine hesitancy: Race/ethnicity, trust, and fear COVID-19 and vaccine hesitancy: A longitudinal study The changing acceptance of COVID-19 vaccination in different epidemic phases in China: A longitudinal study International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity The copyright holder for this preprint this version posted September 3, 2021. ; https://doi.org/10.1101/2021.08.30.21262844 doi: medRxiv preprint