A Novel Survey Platform in the Age of COVID-19 to Increase Accuracy and Adoptability While Reducing Selection Bias

Abstract Given the ongoing COVID-19 pandemic, secure and distanced data collection platforms are critical for reaching vulnerable populations. Commonly used electronic data collection systems lack a myriad of critical features, including a modern technology stack, new data encryption and security standards, study workflows, and reporting algorithms. Moreover, these systems do not have multilingual mapping functionalities of survey and consent forms. All of these components ultimately increase selection bias while simultaneously reducing the security and quality of the response data. In order to directly address the aforementioned issues, we have developed a multilingual and highly secure data management platform. Our application is built using stable, tested, and modular programming frameworks and design patterns targeted at accommodating intricately complex structures of polyglot mapping, large volume of data, encryption and granular user authorization. The statistical accuracy along with the multilingual mapping are the core highlights of this system. The multilingual function of this platform has the ability to eliminate selection biases while creating a well-balanced cross-section of society. Modern survey design workflows and validation checks ultimately prevent data loss and help reduce data collection errors. The platform design was initiated in April 1, 2020 and has been pilot tested for use in multilingual populations. The currently active application version of the system is capable of supporting in-person and telephone interviews, emailing survey links to every registered participant, building family tree architecture, and online consent management. This platform also has built-in report functionality. Additional features are being explored to improve study coordination and monitoring.

care, and dementia care (AL) communities.As part of a broader, ongoing study assessing states' AL regulations regarding medical and mental health care for residents with Alzheimer's and related dementias (ADRD), stakeholders across the United States were recruited in July and August 2020 for semi-structured interviews to provide their perspectives on AL policies and practices specific to COVID-19 response.Stakeholders (n=32) consisted of state healthcare and trade association representatives, clinical practitioners, operators, researchers, and dementia care experts experienced in AL-related operational, healthcare, and regulatory affairs.Using thematic analysis, we describe several emerging topics regarding the opportunities, challenges, and innovations of responding to COVID-19 within the unique context of AL.States' public health responses to COVID-19 lacked an understanding of the broader long-term care system, especially AL's scope and purpose, workforce, capacity to implement infection control practices and policies, and unintended consequences of social isolation for older adults, specifically residents living in dementia care units.Despite these challenges, stakeholders described opportunities to expand telehealth infrastructure, communication and collaboration across states and among operators, and several innovations to mitigate the effects of social isolation.It is imperative for policymakers to understand the nuances of the AL context and design regulations and public health responses grounded in a whole-person perspective and in partnership with operators during, and beyond, pandemic circumstances.

"WHEN YOU ARE WORKING IN THIS ENVIR-ONMENT, YOU'RE MORE LIKELY TO GET SICK": MAPPING CARE RELATIONSHIPS IN LTC
Andreina Marquez de la Plata Gregor, 1 Katie Aubrecht, 1 Tamara Daly, 2 Ivy Bourgeault, 3 Susan Braedley, 4 Prince Owusu, 4 Pat Armstrong, 2 and Hugh Armstrong, 4 1.St. Francis Xavier University,Antigonish,Nova Scotia,Canada,2. York University,Toronto,Ontario,Canada,3. University of Ottawa,Ottawa,Ontario,Canada,4. Carleton University,Ottawa,Ontario,Canada The pandemic has shone a light on problems within the long-term care (LTC) sector.As was true prior to COVID-19, many of the present issues in LTC can be traced to challenging working conditions, such as persistent understaffing of care workers.Working short-staffed means rushing through care, while only satisfying the most basic bodily needs of the resident.This presentation shares early findings from a thematic analysis of interviews conducted with seven care workers as part of the "Mapping Care Relationships" stream of the Seniors -Adding Life to Years (SALTY) project, a pan-Canadian research program that maps how promising approaches to care relationships are organized and experienced in LTC.The purpose of the analysis was to understand how short-staffing is affecting the formation and preservation of meaningful staff-resident relations, and what the impact is on quality of care.Two overarching themes emerged: 1) a relationship between time and work-place illness, injury and violence; 2) a relationship between care worker autonomy and resident quality of care.When working conditions do not support workers in voicing and/or addressing challenges they experience in the workplace, whether this results from understaffing or hierarchical power structures, care workers' ability to deliver even basic care is jeopardized, and resident and worker health and wellness are placed at risk.Themes are discussed in the context of COVID-19 in light of responses to outbreaks in LTC that have reduced the availability of care workers, family visitors and volunteers, and emphasized topdown and even militarized approaches to care management.Antonius Skipper, 1 and Daniel Rose 2 , 1 In March 2020, COVID-19 was declared a pandemic and frequently presented as a virus primarily affecting older adults.News headlines led with statements such as, "Coronavirus deaths are so far mostly older men" (Ramzy, 2020).Although later determined inaccurate, this perspective contributed to openly ageist views and exchanges from people around the world.On the social media platform of Twitter, #BoomerRemover was used as a hashtag to express views related to older adults, and particularly baby boomers, as the primary targets of COVID-19.This study uses qualitative methods to analyze the messages of Twitter users that discuss COVID-19 with the use of the hashtag #BoomerRemover.A total of 1,875 tweets posted in English and including the hashtag "#BoomerRemover" from March 16, 2020 to March 30, 2020 were analyzed.Analytic methods employed an open coding procedure consistent with grounded theory and Numeric Content Analysis (Marks, 2015).Salient themes include: (1) COVID-19 is Politically Driven (2) There's a Real Intergenerational Divide, (3) Young People are Dying Too, and (4) #BoomerRemover is Simply Disrespectful.Findings suggest that only about a fourth of #BoomerRemover tweets could be considered ageist, and the large majority of tweets using the hashtag were related to politics and elections.Further, several of those using the #BoomerRemover hashtag to defend older adults were inadvertently causing it to remain relevant (trend) as a Twitter topic.This study recognizes the importance of considering Twitter -primarily composed of young adults -as a place where intergenerational attitudes vis-à-vis COVID-19 may be expressed.

A NOVEL SURVEY PLATFORM IN THE AGE OF COVID-19 TO INCREASE ACCURACY AND ADOPTABILITY WHILE REDUCING SELECTION BIAS.
Dinesh Mendhe, Stephanie Bergren, and XinQi Dong, Rutgers University, New Brunswick, New Jersey, United States Given the ongoing COVID-19 pandemic, secure and distanced data collection platforms are critical for reaching vulnerable populations.Commonly used electronic data collection systems lack a myriad of critical features, including a modern technology stack, new data encryption and security standards, study workflows, and reporting algorithms.Moreover, these systems do not have multilingual mapping functionalities of survey and consent forms.All of these components ultimately increase selection bias while simultaneously reducing the security and quality of the response data.In order to directly address the aforementioned issues, we have developed a multilingual and highly secure data management platform.Our application is built using stable, tested, and modular programming frameworks and design patterns targeted at accommodating intricately complex structures of polyglot mapping, large volume of data, encryption and granular user authorization.The statistical accuracy along with the multilingual mapping are the core highlights of this system.The multilingual function of this platform has the ability to eliminate selection biases while creating a well-balanced cross-section of society.Modern survey design workflows and validation checks ultimately prevent data loss and help reduce data collection errors.The platform design was initiated in April 1, 2020 and has been pilot tested for use in multilingual populations.The currently active application version of the system is capable of supporting in-person and telephone interviews, emailing survey links to every registered participant, building family tree architecture, and online consent management.This platform also has built-in report functionality.Additional features are being explored to improve study coordination and monitoring.

A SURVEY OF OLDER ADULTS' PERSPECTIVES OF IN-PERSON AND VIRTUAL PARKINSON'S-SPECIFIC EXERCISE CLASSES
Holly Bennett, Jennifer Vincenzo, and Chris Oholendt,

University of Arkansas for Medical Sciences, Little Rock, Arkansas, United States
Due to COVD-19, many health/wellness programs transitioned from in-person to virtual.This mixed methods study aims to explore older adults with Parkinson's disease (PD) perceptions of in-person versus virtual Parkinson's-specific exercise classes.Attitudes, perceptions, and experiences were determined through focus groups (n=9; Male=4; aged 75 years) among older adults with PD and an online survey (n=23; Male=14; aged 74 years).Eighteen respondents attended both in-person and virtually (n=18; Male=9).Four respondents only attended in-person, citing reasons such as difficulty with computer access to virtual classes, limited internet, easier accessibility to in-person classes, and physical injury preventing attendance to any classes.Respondents who participated in both delivery methods preferred virtual classes.Time, convenience, comfort at home, and not having to navigate transportation barriers supported participants' preference for virtual classes.The majority of respondents indicated their fatigue and mental health were either unchanged or improved.Eighty-nine percent reported improved mobility since attending either class; specifically, in balance (n=8), flexibility (n=7), and coordination (n=3).Older adults with PD who attended both classes had minimal difficulty with computer usage and accessing the virtual program with only one participant reporting difficulty transitioning from in-person to virtual classes.Seventy percent stated they would enjoy a combination of on-site and virtual programming.Eighty-nine percent and seventy-seven perfect felt the virtual-based program was safe and beneficial, respectively.Participants who transitioned from an in-person to a virtual exercise program for people with Parkinson's disease felt the program was safe, effective, and improved or prevented declines in their mobility.

AGE DIFFERENCES IN PREFERENCES FOR FEAR-ENHANCING VERSUS FEAR-REDUCING NEWS IN A GLOBAL PANDEMIC
Anthony Villalba, Jennifer Stanley, Jennifer Turner, and Michael Vale, The University of Akron, Akron, Ohio, United States Older adults (OA) prefer positive over negative information in a lab setting, compared to young adults (i.e., positivity effects; YA).The extent to which OA avoid negative events or information relevant for their health and safety is not clear.We first investigated age differences in preferences for fear-enhancing versus fear-reducing news articles during the Ebola Outbreak of 2014.We built upon this pilot study to further investigate this research question during the COVID-19 pandemic.In this study, 164 YA (18-30 years) and 171 OA (60-80 years) responded to an online survey about their preferences, feelings, and behaviors related to the COVID-19 pandemic across 13-days during the initial peak of the pandemic in the United States.Both YA and OA preferred to read positive over negative news about the coronavirus, but OA were even more likely than YA to prefer the positive news article.No age differences in the fear of contraction were found, but OA engaged in more health-protective behaviors compared to YA.Additionally, media engagement was related to fear for both age groups, with social media engagement, specifically, emerging as a key moderating factor for protective behavior change.Although OA may not fear or seek out negative information related to a health concern; they still engage in more protective health behaviors compared to YA.In this study, positivity effects are shown to exist within a health-related event, but OA appeared to still attend to enough negative information about COVID-19 to avoid impairing their health protective behaviors.

AGE DIFFERENCES IN THE CONSUMPTION AND AVOIDANCE OF COVID-19 INFORMATION
Stephanie Deng, Julia Nolte, and Corinna Loeckenhoff, Cornell University, Ithaca, New York, United States Staying informed about COVID-19 is crucial to maintaining public health.Although older adults are at increased risk of complications, recent data (Global Web Index, 2020) suggest that they are less likely to seek out information about COVID-19.This is consistent with prior evidence indicating that information seeking is negatively associated with age (Mata & Nunes, 2010).However, it remains unclear whether older adults merely fail to seek information or intentionally avoid information.In response, we examined whether age is associated with general information seeking and deliberate information avoidance in the wake of the COVID-19 pandemic.Based on previous work indicating age-related shifts in motivational priorities (Carstensen, 2006) we also examined whether avoidance motives differ by age.In a pre-registered online study, an adult lifespan sample (N=500, Mage=49.90, 51% female, 67% non-Hispanic White) completed self-report measures of media consumption, information avoidance, and avoidance motives with respect to the COVID-19 pandemic.In addition, we measured behavioral information avoidance by