key: cord-0275238-xmuzcmbj authors: Gabet, S.; Thierry, B.; Wasfi, R.; De Groh, M.; Simonelli, G.; Hudon, C.; Lessard, L.; Dube, E.; Nasri, B.; Kestens, Y.; Moullec, G. title: How is the COVID-19 pandemic impacting our life, mental health, and well-being? Design and preliminary findings of the pan-Canadian longitudinal COHESION Study date: 2022-05-27 journal: nan DOI: 10.1101/2022.05.26.22275645 sha: 0144bcc2b0f0b45c24d35c3151a4741bca4b5651 doc_id: 275238 cord_uid: xmuzcmbj Abstract With the advent of the COVID-19 pandemic, in-person social interactions and opportunities for accessing resources that sustain health and well-being have drastically reduced. We therefore designed the pan-Canadian population-based prospective COVID-19: HEalth and Social Inequities across Neighbourhoods (COHESION) cohort to provide deeper understanding of how the COVID-19 pandemic context affects mental health and well-being, key determinants of health, and health inequities. This paper presents the design of the two-phase COHESION Study, and descriptive results from the first phase conducted between May 2020 and September 2021. During that period, the COHESION research platform collected monthly data linked to COVID-19 such as infection and vaccination status, perceptions and attitudes regarding pandemic-related measures, and information on participants physical and mental health, well-being, sleep, loneliness, resilience, substances use, living conditions, social interactions, activities, and mobility. The 1,268 people enrolled in the Phase 1 COHESION Study are for the most part from Ontario (47%) and Quebec (33%), aged 48 {+/-} 16 years [mean{+/-} standard deviation (SD)], and mainly women (78%), White (85%), with a university degree (63%), and living in large urban centers (70%). According to the 298 {+/-} 68 (mean {+/-} SD) prospective questionnaires completed each month in average, the first year of follow-up reveals significant temporal variations in standardized indexes of well-being, loneliness, anxiety, depression, and psychological distress. The COHESION Study will allow identifying trajectories of mental health and well-being while investigating their determinants and how these may vary by subgroup, over time, and across different provinces in Canada, in the unique context of the COVID-19 pandemic. 128 COHESION Phase 2 will be launched in May 2022. All Phase 1 participants will be invited to 129 participate in Phase 2, and additional recruitments (n = 10,000) will be done. After consenting to participate to the study, participants are invited to fill a short 131 eligibility questionnaire. Inclusion criteria are being aged 15 years or above, currently residing 132 in Canada, and reading or speaking English or French. Eligible participants are then invited to 133 complete a baseline questionnaire. The recruitment for COHESION Phase 1 was launched on May 11 2020. We used a 135 combination of methods that have proven successful from previous experience in recruiting 136 population-based research project samples [31] , including media communication (e.g., 137 newspaper articles, radio interviews), social media promotion (e.g., Facebook, Twitter, 138 Instagram, and LinkedIn), and outreach through partners' local networks (e.g., local health 139 authorities that relayed the study on their website or through their newsletters). For COHESION Phase 2, we will further use quota sampling at the health region level, 141 based on 2016's Census data (i.e., age composition, gender, income, educational attainment, 142 and ethnic background). We engaged Potloc Inc., a tech-enabled consumer research company 143 that conducts survey sampling through social networks (Facebook, Twitter, Instagram, 144 LinkedIn), for geo-targeting of respondents based on the sampling quotas. Potloc Inc.'s 145 algorithm will push sociodemographic and geographically targeted online ads to our study until 146 local quotas are attained (targets are monitored daily) and that 10,000 participants have thereby 147 been enrolled. . CC-BY 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. (which was not certified by peer review) 219 One of the specificities of COHESION is that it integrates, in baseline and follow-up 220 questionnaires of the two phases, the Visualization and Evaluation of Route Itineraries, Travel . CC-BY 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. 267 "Small", "medium" and "large" urban population centers correspond to areas embracing 268 between 1,000 and 29,999, between 30,000 and 99,999, and 100,000 and more inhabitants, 269 respectively, while "rural area" is a residual value gathering all areas located outside population 270 centers. . CC-BY 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. (which was not certified by peer review) (Fig 3) . Among these ones, 1,268 (54%) completed the 306 baseline questionnaire and have been enrolled in the COHESION Study, and among the latter, 307 557 (44%) opted for the Ethica mobile application in addition to the online self-administered 308 questionnaires. Although recruitment was continuously open during the Phase 1 timespan, the 309 major part of participants joined the study during the first months following the study launch: 310 832 (66%) recruitments after three months of follow-up, 1,136 (90%) after six months (Fig 4) . COHESION Phase 1 participants are distributed all across Canada; they live mainly in 312 Ontario (597, 47%) and Quebec (417, 33%) (Fig 5) , and 910 (72%) declared English as their 313 favorite language. They are in average 48 ± 16 (mean ± standard deviation, SD) and are mainly 314 women (78%), White (85%), born in Canada (85%), with university or post-graduate level 315 degree (63%), and in a relationship (67%) ( Table 2a) . The most part is owner (62%) and lives 316 in a house (66%), with their partner or family (74%), without children at home (72%), with 317 private outside space access (97%), with pets (56%), and in large urban centers (i.e., in areas 318 including more than 100,000 inhabitants); though participants show wide contrasts in terms of 319 home surroundings' greenness and neighborhood's material and social deprivation (Table 2b) . 320 Participants are, by decreasing order, employed (58%), retired (19%), unemployed (14%), on 321 leave or disabled (5%), or students (2%); they are mostly satisfied about their household annual 322 income (77%) (Table 2a) . Concerning health, they are 44% and 35% being affected by a 323 physical (i.e., heart disease, lung disease, cancer, high blood pressure, diabetes, severe obesity, 324 and/or autoimmune disease) or a mental (i.e., depressive disorder and/or anxiety disorder) 325 chronic disease, respectively. A monthly consumption of alcohol, cigarettes and/or vape, and . CC-BY 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 May 27, 2022. ; https://doi.org/10.1101/2022.05.26.22275645 doi: medRxiv preprint 326 cannabis was reported by 72%, 14%, and 18% of participants at baseline, respectively. Lastly, 327 they are 27% and 38% considering their selves and/or someone in their household at a high risk 328 of being infected by-or of complications of COVID-19, respectively. 331 Throughout the first year of follow-up, i.e., from May 2020 to July 2021, up to 17 follow-up 332 questionnaires were administered to participants (Fig 4) , and 758 (60%) participants completed 333 at least one (Fig 3) . Depending on their date of recruitment, participants were contacted between 334 one and seventeen times as part of the prospective follow-up waves (Fig 4) , and they filled out 335 in average 29% ± 36% (mean ± SD) of the follow-up questionnaires they received. Among the 336 first year of follow-up, in average 298 ± 68 participants completed the follow-up questionnaire 337 by wave (Fig 6, see S1 Table) . Attrition rate throughout the Phase 1 follow-up appeared statistically distinct in some 339 specific groups (Table 3 ). The older the participants were and the higher the education level 340 was, the less the attrition was (p <0.001 and p = 0.069, respectively); average age was 50 ± 16 341 and 45 ± 16 (mean ± SD) in people participating or not in the longitudinal follow-up, 342 respectively (p <0.001). Attrition appeared higher in employed people and lower in retired ones 343 (p <0.001) (Table 3) , and also higher in people living in family (p = 0.005), with children (p = 344 0.013), with pets (p = 0.001), in a house (p = 0.001), and not owner (p = 0.001) (see S2a Table) . Concerning mental health, the first year of follow-up of the Phase 1 showed statistically 372 significant changes between questionnaire waves regarding reporting of anxiety symptoms, 373 depression symptoms, and psychological distress (Fig 7b; see S4b Table) . Thus, GAD-7 median 374 score ranged between 4 (IQR: 1-7; wave 21/27) and 5 (IQR: 2-10; wave 21/15) and PHQ-9 375 median score varied from 4 (IQR: 1-7; wave 21/27) and 6 (IQR: 3-9.3; wave 20/30). For these . CC-BY 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. 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 May 27, 2022. ; https://doi.org/10.1101/2022.05.26.22275645 doi: medRxiv preprint 433 methodological efforts to minimize it, the risk of attrition can hardly be eliminated. One way to 434 account for the potential effect of residual attrition bias is through various analytical strategies, 435 including the use of inverse probability weights for trajectory analyses (e.g., growth curve 436 models) on mental health and well-being. One critical issue with our cohort that was launched in mid-spring 2020 is the lack of a 439 true pre-pandemic baseline. This can partly be circumvented with retrospective questions, but 457 Throughout phase 2 of COHESION, we will continue adapting our survey content with timely . CC-BY 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. (which was not certified by peer review) . CC-BY 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 May 27, 2022. ; https://doi.org/10.1101/2022.05.26.22275645 doi: medRxiv preprint 479 6. Conclusion 480 Real-time monitoring and evaluation of the unintended consequences on mental health and 481 health inequities of the pandemic is essential for shaping and adapting effective public health 482 policies and programs targeting contextual living conditions (e.g., pedestrianization of streets, 483 securing access to parks, housing renovation programs, permanent supportive housing 484 programs, neighborhood greening program). We will benefit from the support of the Uni-Cité 485 Collaboratory [64] which specializes in science to policy approaches -to equip research teams 486 and cities with tools to better incorporate science into urban public policy. Our study will 487 provide a comprehensive portrait of the key pathways of the COVID-19 impacts on mental 488 health and well-being across Canada. Our flexible infrastructure will ensure that we can adjust 489 to local needs and to the evolving situation from pandemic to post-pandemic recovery. CC-BY 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. (which was not certified by peer review) The copyright holder for this preprint this version posted May 27, 2022. Follow-up questionnaires of the COHESION Study are short questionnaires (15 min). For Phase 1, they were first offered biweekly until August 2020, then monthly. COVID-19-related 198 topics cover infection and/or vaccination status, perceived vulnerability, perception of and 199 compliance with mitigation measures Health-related questions focus on general health (SF-12 UCLA 3-item loneliness score) 202 [36], anxiety symptoms (GAD-7) [37], depression symptoms (PHQ-9 For Phase 2, follow-up questionnaires will be split in a core ('light') and an optional 209 complementary ('complete') sections, and will be offered every two months we administered to participants well-being, sleep credit, 352 loneliness, anxiety, depression and psychological distress -related standardized modules 5, 9, 353 13, 8, 8, and 11 times, respectively (Table 1) respectively) (Fig 7a; see S4a Table). The sleep duration appeared stable throughout 362 the first year of the Phase 1 follow-up (mean ± SD: from 7.9 ± 1.3 to 8.0 ± 1.4 hours; waves 363 21/23 and 20/30, respectively) wave 21/15), yet for the wave 20/52, corresponding to the festive season. Moreover, all UCLA 368 3-item score median values through the follow-up were significantly higher than the UCLA 3-369 item score median assessed retrospectively for the Government of Canada. COVID-19 daily epidemiology update COVID-19: Emergency Measures Tracker Bruine de Bruin W. Age Differences in COVID-19 Risk Perceptions and Mental 530 Health: Evidence From a National U.S. Survey Conducted in COVID-19: the gendered impacts of the outbreak. The 534 Lancet COVID-19: a potential public health problem for homeless 536 populations. The Lancet Public Health 538 A portrait of the early and differential mental health impacts of the COVID-19 539 pandemic in Canada: Findings from the first wave of a nationally representative cross-540 sectional survey The New York Times. The Workers Who Face the Greatest Coronavirus Risk Revealing the unequal burden of COVID-19 by income, 545 race/ethnicity, and household crowding: US county versus zip code analyses COVID-19, school closures, and child poverty: a social 549 crisis in the making. The Lancet Public Health COVID-19 and African Americans Ethnicity and 554 COVID-19: an urgent public health research priority. The Lancet Urban nature as a source of resilience 557 during social distancing amidst the coronavirus pandemic COVID-19 Growth Rate Decreases with Social Capital Keep socially (but not physically) connected and 562 carry on: Preventing suicide in the age of COVID-19 Preventing Suicide in 565 Rural Communities During the COVID-19 Pandemic Mental health survey shows Manitobans feeling more anxious due to 568 social isolation Why COVID-19 loneliness can be especially hard on teens The role of daily mobility in mental 571 health inequalities: The interactive influence of activity space and neighbourhood of 572 residence on depression Use of tracking technology to examine life-space 575 mobility among people with depression: a systematic review protocol Does access to neighbourhood green space promote a 578 healthy duration of sleep? Novel findings from a cross-sectional study of Public Health Agency of Canada. The Chief Public Health Officer's Report on the 581 State of Public Health in Canada 2017 -Designing Healthy Living Economic, and Health Inequality and COVID-19 Infection in the United States Incidence des conséquences économiques de la COVID-19 sur 587 les préoccupations sociales des Canadiens The impact of persistent poor housing 589 conditions on mental health: A longitudinal population-based study. Preventive 590 Medicine Housing characteristics and staying at home during the 592 COVID-19 pandemic Food security and Canada's agricultural system challenged by 594 COVID-19 Vulnerability to food insecurity since the COVID-19 lockdown Canadians who report lower self-perceived mental health during the 598 COVID-19 pandemic more likely to report increased use of cannabis, alcohol and 599 tobacco Recruiting 601 Participants for Population Health Intervention Research: Effectiveness and Costs of 602 Recruitment Methods for the INTERACT (INTErventions, Research, and Action in 603 Cities Team) study Short Form 12 Health Survey (SF-12) Measuring the dimension of psychological general well-being by the WHO-5. 608 In: Quality of life newsletter The Pittsburgh Sleep Quality Index (PSQI) Screening 612 for serious mental illness in the general population with the K6 screening scale: results 613 from the WHO World Mental Health (WMH) survey initiative UCLA Loneliness Scale (Version 3): Reliability, validity, and factor 616 structure A brief measure for assessing 619 generalized anxiety disorder: The GAD-7 The PHQ-9: Validity of a brief depression 622 severity measure The brief 625 resilience scale: Assessing the ability to bounce back Canadian Community Health Survey (CCHS) -Mental Health A simple method to assess exercise behavior in the community. 632 Canadian journal of applied sport sciences Canadian Income Survey -2019 Questionnaire Canadian 638 Community Health Survey, Cycle 2.2, Nutrition (2004) -Income-Related Household 639 Food Security in Canada Combining social network and activity space 641 data for health research: tools and methods. Health and Place An interactive 644 mapping tool to assess individual mobility patterns in neighborhood studies Name generators in surveys of personal networks Wave 1 results of 649 the INTerventions, Research, and Action in Cities Team (INTERACT) cohort study: 650 Examining spatio-temporal measures for urban environments and health. Health & 651 Place Ecological Momentary Assessment Using 653 Smartphone-Based Mobile Application for Affect and Stress Assessment Assessing Mood in Daily Life An Area-658 based Material and Social Deprivation Index for Public Health in Québec and Canada Population centre size class values by province and territory Google Earth 664 Engine: Planetary-scale geospatial analysis for everyone. Remote Sensing of 665 Environment USGS Landsat 8 TOA Reflectance (Orthorectified) -2013 to 667 2017 A pan-Canadian 670 measure of active living environments using open data Canadian Institute for Health Information. COVID-19 Intervention Timeline in 673 Predictors of attrition in a 677 longitudinal population-based study of aging Attrition in longitudinal studies: Who do you lose? 680 Australian and New Zealand Journal of Public Health Recruiting 685 Participants for Population Health Intervention Research: Effectiveness and Costs of 686 Recruitment Methods for a Cohort Study Experiments with methods to reduce attrition in 689 longitudinal surveys Problems and pitfalls of retrospective 692 survey questions in COVID-19 studies CIUSSS du Nord-de-l'île-de-Montréal | Portrait de l'établissement Home -Collaboratoire Uni-Cité Built environment and health. Urban Form and Accessibility . CC-BY 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 May 27, 2022.