key: cord-0269604-1hd8jhpf authors: French, C. E.; Denford, S.; Brooks-Pollock, E.; Wehling, H.; Hickman, M. title: Low uptake of COVID-19 lateral flow testing among university students: a mixed methods evaluation date: 2021-07-22 journal: nan DOI: 10.1101/2021.07.20.21260836 sha: b06950aed61bb86b9562df04cee2a8a92685b1df doc_id: 269604 cord_uid: 1hd8jhpf Background: University populations offer a unique opportunity to quantify COVID-19 lateral flow testing (LFT) uptake. Methods: Mixed methods evaluation of LFT among University of Bristol students comprising an analysis of testing uptake using logistic regression analyses; a survey; and qualitative interviews to explore experiences of testing and subsequent behaviour. Results: 12,391 LFTs were conducted on 8025/36,054 (22.3%) students. Only one in 10 students had the recommended two tests. There were striking demographic disparities in uptake with those from ethnic minority groups having lower uptake (e.g. 3% of Chinese students were tested vs. 30.7% of White students), and variations by level and year of study (ranging from 5.3% to 33.7%), place of residence (29.0% to 35.6%) and faculty (15.2% to 32.8%). Barriers to engagement with testing included a lack of awareness, knowledge and understanding, and concerns about the accuracy and safety. Students understood limitations of LFTs but requested further information about test accuracy. Tests were used to inform behavioural decisions, often in combination with other information, such as the potential for exposure to the virus and perceptions of vulnerability. Conclusions: The low uptake of testing brings into question the role of mass LFT in university settings. Lateral flow testing (LFT) of asymptomatic people is an integral part of the UK's COVID-19 response. Since 9 th April 2021 everyone in England has been eligible to take a LFT twice weekly [1] [2] [3] [4] . There is an ongoing and polarised debate around mass testing to detect asymptomatic infections using this technology. Since approximately one third of people infected with SARS-CoV-2 have no symptoms, it is argued that identifying infections among this group so that they can isolate and their contacts be traced is key to controlling the pandemic [3, 4] . Although this policy was well received by some [5] [6] [7] others have raised concerns, particularly around test accuracy and the potential consequences of inaccurate results [8] [9] [10] [11] . While the accuracy of LFT is important, much less attention has been paid to levels of uptake of testing, which could pose a major barrier for the use and effectiveness of asymptomatic testing. In Autumn 2020, COVID cases were high among university students in the UK [12] . In November 2020 the government recommended LFT for university students, recommending that all students should have two negative tests before travelling home for the winter break [1, 13] . Evaluation of this testing strategy, including equity in testing uptake is crucial if testing continues to be used to control the pandemic in the future. University populations offer a unique opportunity to quantify testing uptake in a well-defined group of individuals. Our study aims to i) assess uptake of LFT among University of Bristol students, including demographic variations; ii) explore the acceptability and feasibility of asymptomatic testing and iii) to explore the barriers and facilitators to uptake and effective implementation of testing. We conducted a mixed methods evaluation of LFT among University of Bristol students who did not have COVID-19 symptoms; comprising a quantitative analysis of testing uptake data, a student survey and qualitative interviews. We analysed data on the uptake of LFT from 30 th November to 18 th December 2020. Students prebooked their tests online. On arrival at testing venues they were asked to swipe their university identity card. A list of all students enrolled at the university, held by student records, was matched with the date of any tests undertaken, as collected via card swipes at testing venues using student ID number. Information on location of students during the study period was not available. However, a sensitivity analysis was conducted by excluding students who were either enrolled on a distance learning course or completed a 'location of study' form indicating that they were likely not going to be on campus. The total number of positive results was recorded at testing sites but was not documented for individual students. Univariable and multivariable analyses were conducted using . CC-BY-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 July 22, 2021. ; https://doi.org/10.1101/2021.07.20.21260836 doi: medRxiv preprint logistic regression to explore demographic factors associated with being tested. All explanatory variables were included in the multivariable model a priori. Analyses were conducted in STATA 16.1. Participants were invited to complete a confidential online survey about their views of university testing (Supplement 1). A link to the survey was shared by the university communications team via social media (Facebook, Twitter and Instagram) and via the student newsletter. Informed consent was obtained. Frequencies and descriptive statistics are presented for closed survey questions. Free text answers were used to offer further insight into answers given to closed survey questions. We identified key barriers to engagement with testing using qualitative content analysis in three stages [14] [15] [16] survey responses were coded independently by two authors, codes were then categorised into a list of barriers and facilitators, and data assigned to each category. Volunteers who took part in the survey and provided consent to be contacted by the research team were invited to take part in an online interview. Participants were >18 years and a registered student at the university. We purposely sampled for diversity in key factors, including ethnicity, living arrangements, enrolled course, and whether or not they had taken a test at the university. Sample size was informed by the concept of 'information power' [17] , with continuous assessment of the data in relation to study objectives. Potential participants were provided with a study information sheet and given an opportunity to ask questions, informed of the voluntary nature of the study, and assured of the confidentiality of their data. As all interviews were conducted via the telephone or online, and audio recorded verbal consent was obtained. The semi-structured topic guide (Supplement 2) aimed to explore participants' views about testing, understanding and interpretation of test results, and impact on behaviour. Data from interviews were analysed using a thematic approach [18, 19] . Two researchers independently read and assigned codes to transcripts. Possible themes were identified and refined. Charts were developed for each theme, and relevant text from transcripts were copied verbatim. Charts were then used to compare data within and between individuals. Ethical approval was obtained from University of Bristol faculty ethics committee (Reference 115084). . CC-BY-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. (which was not certified by peer review) Demographic variations in testing uptake (Tables 1 and 2) Although the absolute percentage of students taking up testing was similar across genders (21.9% for men and 22.5% for women), women were more likely to be tested than men (adjusted odds ratio [ A sensitivity multivariable analysis excluding students who were likely not to have been on campus during the testing period (n=4,907, 13.6% of all students) did not alter the observed patterns in testing uptake. Odds ratios changed a little (all <10%) and were within the confidence intervals reported in Table 2 . A total of 436 students completed the survey, of which 328 (75%) had taken part in testing and 108 (25%) had not (Supplement 3). Among students who engaged in the university testing service and those who did not, the majority described their views of getting regular tests as either somewhat positive (31% and 31% respectively) or very positive (51% v 31%). Few participants described their views of testing as somewhat negative or very negative (18% of those who did not participate in testing vs. 5% of those who did: Table 3 ). . CC-BY-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. (which was not certified by peer review) The copyright holder for this preprint this version posted July 22, 2021. ; https://doi.org/10.1101/2021.07.20.21260836 doi: medRxiv preprint Most students understood that a negative test result meant that the person is probably not infectious (84% of those who had a test versus 75% of those who did not - Table 3 ). Only a minority of students in both groups thought a negative test means the person is definitely not infectious (6% of those engaging in testing v 12% of those who did not) or that they did not know (4% of those engaging in testing v 9% of those who did not). Approximately half of students engaging in testing reported that the level of contact with others had not changed in the seven days following the testing period (55%). 19% of students reported that close contact increased, and 17% reported that close contact had decreased following tests ( Table 3 ). Self-reported adherence to the guidance was similar between the groups, with 90% of those engaging in testing and 81% of those not engaging in testing reporting that they had been adherent to the guidance all or most of the time (Table 3) . A total of 108 comments were coded and used to identify barriers to engagement in testing (Table 4 ). Barriers were categorised as 1) Perceived lack of need or demand 2) Problems accessing the service 3) Safety concerns 4) Knowledge and understanding and 5) Lack of support for self-isolation. Twenty students were interviewed about testing. Fourteen reported that they had taken a test at the university in December 2020 and seven had not. Data is presented under three main themes 1) motives for engaging in testing 2) barriers to testing 3) and using test results to inform behavioural decisions. Three main motives for taking part in university testing procedures included 1) to reduce the risk of transmitting the virus 2) for information and 3) following recommendations and guidance. Students were willing to take tests to protect other people from the virus. Most students were more concerned about the risk to others than to themselves (Table 5 quote 1), and tests provided reassurance that they were not spreading the virus to others (quote 2). This was particularly important for those planning to relocate for the holidays (quote 3), those with vulnerable family members (quote 4), or those who considered themselves to have been at risk of exposure to the virus (quote 5). . CC-BY-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. (which was not certified by peer review) The copyright holder for this preprint this version posted July 22, 2021. ; https://doi.org/10.1101/2021.07.20.21260836 doi: medRxiv preprint Some students wanted to take tests for information (quote 6). Although these students were not necessarily planning to travel, they were keen to take tests for their own benefit (quote 7), or for their mental health (quote 8). Some students took tests because they were available (quote 9), and supported by the University (quote 10). In some cases, tests were required before the student could attend in person lectures (quote 11), or travel (quote 12). Barriers to uptake of testing include 1) lack of need 2) lack of awareness 3) access 4) risk of exposure at the testing site. Some students did not take a test because they did not think that tests were required or intended for them. For example, one student explained that she had not taken a test at the university because she was not planning to travel away from Bristol (quote 13). Other students were able and willing to isolate, and considered this preferable to testing (quote 14), or demonstrated a preference for PCR tests over LFT (quote 15). Some students had not taken a test because they had not been aware that testing was available (quote 16). Students thought that more could be done to promote awareness of testing, particularly among those who do not have a strong network of peers (quote 17). A number of practical barriers were described; including access issues (quote 18), and issues with the timing and location of test sites (quote 19). A potential barrier to engagement in testing was concerns of catching the virus at or on route to the testing centre (quote 20), particularly among those who had to travel long distances (quote 21). It was noted that cases of the virus were high among the student population, and some considered the risk of exposure to outweigh the benefits of getting tested (quote 22). Most students were very aware of the ongoing debate about the accuracy of LFTs, and reported having discussions with their friends, families, and in some cases, with the university about how accurate the tests were (Quote 23). Tests were considered just one piece of information from which to inform decisions (quote 24), often being used alongside other key indicators -such as whether or not the person had been in contact with someone with the virus, of if they had any symptoms . CC-BY-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 July 22, 2021. ; https://doi.org/10.1101/2021.07.20.21260836 doi: medRxiv preprint (quote 25). Some students reported that testing had reassured them that they had 'done everything they could' before travelling (quote 32). Despite limitations, tests were seen as 'good enough' to inform decisions (quote 26), and although students reported feeling somewhat reassured by negative test results (quote 29), they described being unlikely to drastically increase contact or to visit anyone considered to be vulnerable (quote 31). Activities were limited to those that were considered essential, such as shopping and exercise (quote 30) and it was recognised that any negative rest result was only "valid" for a limited time, and any subsequent contact was a potential risk (quote 27 and 28). There was an acknowledgement that receiving a negative test could increase close contact behaviour, but generally it was noted that students who were likely to break the rules would do so regardless of testing status (quote 33). Our research revealed that one in 10 students had the recommended two LFTs and highlighted demographic disparities in uptake by ethnic group, level of study and year group, and faculty. Data collected from survey and interview participants suggested that whilst students were generally positive about testing, key barriers to uptake remain. Our qualitative data revealed that many participants were motivated to take tests to protect those around them and avoid transmitting the virus to their friends and family. However, students reported a number of barriers to uptake; including a lack of awareness of the testing service, problems accessing the service, a lack of knowledge and understanding of testing procedures, and concerns about the accuracy and safety of testing. However, whilst overall uptake was low, many of those who did not take tests described a lack of need for tests, either because they were not travelling, were unlikely to have been exposed to the virus, were already isolating, or were tested elsewhere. Mass testing for COVID-19 is relatively new, and results of testing programmes are ongoing. Whilst our data revealed a low uptake, particularly among those form ethnic minority groups, this is very much in line with observations of testing uptake in other public health interventions such as home HIV testing [20] . Meanwhile, the mass COVID-19 LFT pilot conducted in Liverpool also reported a lower test uptake, as well as a higher positivity rate, among those from minority ethnic groups [21] . The very small number of positive tests during the study period precluded analyses on demographic variations in positivity, both due to a lack of power and the potential for deductive disclosure. Further research is urgently needed to explore barriers to testing among these populations and cocreate interventions to support the uptake of tests if and when required. . CC-BY-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 July 22, 2021. ; https://doi.org/10.1101/2021.07.20.21260836 doi: medRxiv preprint In line with findings from other universities, students engaging in testing were motivated to do so to protect those around them [22] . Students were well informed about the limitations of tests, often describing test results as just one piece of information, and using them with caution to inform their behaviour [23] . Many students had done their own research, had discussions with their friends, family, tutors and lecturers to maximise their knowledge of testing. This highlights the need for improved communications from universities to enable students to make their own informed decisions. Indeed, recent research that has shown basic and simple messages may not be suitable for communicating complex information about how to behave during the pandemic [24] and students are likely to appreciate having the opportunity to access information about the sensitivity and specificity of the tests. Despite concerns that testing would increase risky contact, we did not find evidence to support this. Students were well informed about the limitations of the tests, and used them with caution to inform behavioural decisions. A key strength of this research is the use of a mixed methods approach. Additionally, though some other universities have evaluated their LFT programmes [25, 26] we are not aware of any reporting data on testing uptake and exploring demographic variations in uptake among the whole student body. This is a unique strength of our work provides crucial information to inform future university testing strategies. Our work identified several ways in which engagement may be enhanced. As many students had not been aware of the testing service, a persuasive, targeted and personalised advertising campaign may increase uptake. To maximise engagement, advertisements should be cocreated with the intended recipients of campaign. Such a campaign should include encouragement from trusted sources, and emphasise the benefits of testing to encourage participation among those who may be apathetic. It would also need to reassure those who are anxious about accessing the testing services. A limitation of the analyses on testing uptake is that denominator was all students enrolled at the university. The university doesn't hold comprehensive and reliable information on which students were resident in Bristol during the testing period. However, in our sensitivity analysis in which students who were likely not to be in Bristol at the time of testing the findings were little altered. A key limitation of the survey and interview data is that participant recruitment occurred via social media, and it is likely key communities (e.g., those who do not engage with university managed social media accounts) were missed. . CC-BY-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 July 22, 2021. ; https://doi.org/10.1101/2021.07.20.21260836 doi: medRxiv preprint LFT continues to play an important and expanding role in the UK's COVID strategy [3, 4] . If regular LFT is considered appropriate and worthwhile going forwards then work is needed to monitor trends in testing uptake among student, and other, populations. Importantly, we need to strive for equity in access to and uptake of testing. Our findings should be used to inform the wider debate around the usefulness and appropriateness of the widespread use of LFT for asymptomatic people. . CC-BY-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. Ethics approval and consent to participate Ethical approval was provided by the University of Bristol -Ethical approval was obtained from University of Bristol faculty ethics committee (Reference 115084). All interview participants verbally consented to take part in the study. All participants provided verbal or written consent for data to be included in publications. The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request. . CC-BY-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 July 22, 2021. . CC-BY-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 July 22, 2021. 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 July 22, 2021. *An odds ratio of less than one indicates lower uptake of testing as compared with the reference group **Includes 153 pre-sessional students . CC-BY-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 July 22, 2021. 12 . CC-BY-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 July 22, 2021. . CC-BY-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 July 22, 2021. 7 Includes concerns about the lack of "My other main concern is the lack of mental health support for those . CC-BY-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 July 22, 2021. ; https://doi.org/10.1101/2021.07.20.21260836 doi: medRxiv preprint support for those who test positive isolating and/or following all guidelines" . CC-BY-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 July 22, 2021. ; https://doi.org/10.1101/2021.07.20.21260836 doi: medRxiv preprint . CC-BY-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 July 22, 2021. Quote 30 Obviously I wouldn't say get tested and go to parties because that's ridiculous but going to the shops and going on a walk and just going to places that you have to be" . CC-BY-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 July 22, 2021. ; https://doi.org/10.1101/2021.07.20.21260836 doi: medRxiv preprint Quote 31 and 32 "but then I was very aware that if I went into the supermarket then I could just easily have gone and got infected again so it was like yeah for now but [laugh] 'cause the wording was like at the time you took your test, you tested negative but reinforces like this is very temporary assessment of your situation but it's still better than like having no idea" "my confidence in [the negative test result] decreases with the more contacts I have with people or the more public places I got to or when I'm with people. My confidence decreases the more exposure I have to people" (female, Asian, tested) Quote 33 "I'm sure for some that it would but I'm sure for most that it wouldn't and I think the people who would probably act differently following one of those negative tests would probably act like that anyway. So I don't think, for the good impact it would have I think the negative impact would be very small"(female, white, tested) Recommendations for improving the service . CC-BY-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. (which was not certified by peer review) The copyright holder for this preprint this version posted July 22, 2021. "I know a lot of people who didn't bother to do the testing just 'cause they couldn't be bothered or anything -so there's definitely some ways that you could influence them, money probably or something like that"(male, white, tested) Quote 41 "If that could open up me going to the library or going to lectures or anything like that it's definitely a price worth taking" (female, mixed ethnicity, tested). . CC-BY-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. (which was not certified by peer review) The copyright holder for this preprint this version posted July 22, 2021. ; https://doi.org/10.1101/2021.07.20.21260836 doi: medRxiv preprint COVID-19 Response -Spring 2021 from More employers sign up to rapid testing to protect workforce -GOV Understanding laterla flow antigen testing for people without symptoms New campaign urges public to get tested twice a week Liverpool COVID-19 Community Testing Pilot. Interim Evaluation Report Executive summary: It's wrong not to test: The case for universal Clarifying the evidence on SARS-CoV-2 antigen rapid tests in public health responses to COVID-19. The Lancet Covid-19: government must urgently rethink lateral flow test roll out COVID-19: Lateral flow tests miss over half of cases, Liverpool pilot data show COVID-19: How the UK is using lateral flow tests in the pandemic Covid-19: Controversial rapid test policy divides doctors and scientists How has coronavirus (COVID-19) spread among students in England? 2020: Accessed Students urged to get COVID-19 tests ahead of the Christmas 'travel window Content analysis and thematic analysis: implications for conducting a qualitative descriptive study Theme development in qualitative content analysis and thematic analysis Journal of Nursing Education and Practice Three approaches to qualitative content analysis. Qualitative Health Research Sample Size in qualitative interview studies: guided by information power. Qualitative Health Research Reflecting on reflexive thematic analysis. Qualitative Research in Sport, Exercise and Health Using thematic analysis in psychology Acceptability of HIV self-sampling kits (TINY vial) among people of black African ethnicity in the UK: a qualitative study