key: cord-0274733-r4lvr7xx authors: Bonner, C.; Batcup, C.; Ayre, J.; Pickles, K.; Cvejic, E.; Copp, T.; Cornell, S.; Dodd, R.; Isautier, J.; Nickel, B.; McCaffery, K. J. title: Behavioural barriers to COVID-19 testing in Australia: Two national surveys to identify barriers and estimate prevalence by health literacy level date: 2021-08-28 journal: nan DOI: 10.1101/2021.08.26.21262649 sha: e01e60429f489757a882da4ea8335b7f6e407df8 doc_id: 274733 cord_uid: r4lvr7xx Background: COVID-19 testing and contact tracing has been crucial in Australia's prevention strategy. However, testing for COVID-19 is far from optimal, and behavioural barriers are unknown. Study 1 aimed to identify the range of barriers to testing. Study 2 aimed to estimate prevalence in a nationally relevant sample to target interventions. Methods: Study 1: National longitudinal COVID-19 survey from April-November 2020. Testing barriers were included in the June survey (n=1369). Open responses were coded using the COM-B framework (capability-opportunity-motivation). Study 2: Barriers from Study 1 were presented to a new nationally representative sample in November to estimate prevalence (n=2869). Barrier prevalence was analysed by health literacy level using Chi square tests. Results: Study 1: 49% strongly agreed to get tested for symptoms, and 69% would self-isolate. Concern about pain was the top barrier from a provided list (11%), but 32 additional barriers were identified from open responses and coded to the COM-B framework. Study 2: The most prevalent barriers were motivation issues (e.g. don't believe symptoms are COVID-19: 28%, few local cases: 18%). Capability issues were also common (e.g. not sure symptoms are bad enough: 19%, not sure whether symptoms need testing: 15%). Many barriers were more prevalent amongst people with low compared to high health literacy, including motivation (preference to self isolate: 21% vs 12%, pain: 15% vs 9%) and capability (not sure symptom needs testing: 12% vs 8%, not sure how to test:11% vs 4%). Conclusion: Even in a health system with free and widespread access to COVID-19 testing, motivation and capability barriers were prevalent issues, particularly for people with lower health literacy. This study highlights the important of diagnosing behaviour barriers to target public health interventions for COVID-19 and future pandemics. 4 to lack of supply or staff resource issues 7 . There may also be issues of delivering tests and transporting samples for remote areas 8 . Finally, there may be issues with inadequate communication and low community knowledge about which symptoms require testing and the process to follow 7 . In terms of opportunity barriers, Australia is fortunate to have efficient and free testing widely available, although this can vary by location. Testing clinics have been set up all around the country, which includes drive through options to minimise contact with others, and results are generally sent by text message within 2 days 9,10 . However, the process can take longer and is unpredictable where demand increases in new outbreak areas such as Victoria in July-September 2020 [11] [12] [13] . For capability barriers, public health messages have focused on basic processes but we know certain groups are less likely to understand and act on COVID-19 prevention advice. For example in Australia, people with lower health literacy and those who speak a language other than English were less likely to be able to identify COVID-19 symptoms and prevention measures 14 . Younger people, men and those with less education were more likely to agree with COVID-19 misinformation about prevention, management and cure 15 . These groups likely need different communication strategies to ensure everyone understands the message and takes appropriate action. 5 months and even years to come 18 . Better understanding of testing barriers is a pressing issue for COVID-19, but also essential to improve the management of future pandemics. We conducted two Australian studies to address a major gap in our understanding about how to improve COVID-19 prevention. Study 1 aimed to identify the range of barriers to testing. Study 2 aimed to estimate the prevalence of barriers to COVID-19 testing, to target interventions to the most important issues. The Sydney Health Literacy Lab (SHeLL) has been conducting a national longitudinal survey in Australia since April 2020 14, 19 . The original sample was recruited via an online market research panel, supplemented with social media advertising (n=4326). The social media users were followed-up monthly from April-July. A list of possible testing barriers provided to the authors by the NSW Department of Health was included in the June 2020 survey (n=1369), along with intentions to test and self-isolate if symptomatic. These were worded as: "over the next 4 weeks, I plan to get tested if I have COVID-19 symptoms (cough, sore throat, fever)" and "over the next 4 weeks, I plan to stay home if I have COVID-19 symptoms (cough, sore throat, fever)"; with 1-7 response options from strongly disagree to strongly agree. Open responses to "other" test barriers were also collected. Descriptive analyses are reported as percentages for the survey results. The test barriers identified from the survey responses were categorised using the COM-B framework 4 by two researchers who have been trained in the use of this framework (CB and CAB). CAB conducted the initial coding, and this was reviewed by CB with discussion of issues that could be coded in different ways until consensus was reached. All barriers could be categorised within the COM-B components. . 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. (which was not certified by peer review) The copyright holder for this preprint this version posted August 28, 2021. ; https://doi.org/10.1101/2021.08.26.21262649 doi: medRxiv preprint All barriers identified in Study 1 were presented to a new nationally representative sample in November 2020, recruited through online market research company Dynata with quotas for age, gender, education and state (n=2034). Participants were asked 'Below are some of the reasons why people don't want to get tested, or can't get tested, for COVID-19. Please select the reasons that apply to you. Pick as many or as few as you want. If I get COVID-19 symptoms (signs of having COVID-19), I might not get tested because...'. They were also presented with the option 'None of these apply to me'. Participants were then prompted to rank their selected barriers in order of importance ('Now please re-order the answers you gave to the last question, from the most important reason at the top to the least important. Don't worry about it being exactly right, this will just help us understand broadly which reasons are more important to people.'). We included all barriers identified in study 1 except physical opportunity barriers, as these could not be addressed through a messaging intervention which was the intended use of the results. Comparisons of barrier prevalence between low and high health literacy groups were based on independent Chi square tests. Health literacy was measured using Chew et al.'s validated Health Literacy Screener 20 . The sample demographics are shown in Table 1 . Study 1 had under-representation of male gender (32%) and lower education (14.5%), with a relatively small sample for low health literacy (8.9%). Quota sampling in Study 2 resulted in a more representative sample for male gender (49.5% male) and lower education (34.6%), which increased the sample for low health literacy (16.5%). . 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. (which was not certified by peer review) The copyright holder for this preprint this version posted August 28, 2021. ; https://doi.org/10.1101/2021.08.26.21262649 doi: medRxiv preprint The results of Study 1 survey questions are reported in Table 2 . Around half (49.1%) of people strongly agreed they would get tested if they had COVID-19 symptoms (cough, fever, sore throat), but most people agreed to some extent that they would get tested if they had symptoms (83.9%). For self-isolation, 69% strongly agreed they would stay home if they had symptoms. The most common barriers selected from the list provided were that testing is painful (11.2%), not knowing how to get tested (7.1%), and worry about getting infected at the testing centre (5.9%). Many participants (9.9%) indicated other reasons, with 136 open responses that included many additional barriers to testing than those provided in the survey question. Table 3 maps all . 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. Table 3 136 (9.9%) I don't know how, when and where to get tested 98 (7.1%) I'm worried I will get infected with COVID-19 at the testing clinic 81 (5.9%) I'll forget to get tested 33 . 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 August 28, 2021. ; https://doi.org/10.1101/2021.08.26.21262649 doi: medRxiv preprint I'm not sure my symptoms are bad enough "Believing that I don't meet the threshold eg when is a cough a cough or a runny nose with spicy food a runny nose? Otherwise I would get tested." I'm not sure this symptom is one that needs testing "I am not sure whether I would be eligible to be tested" The testing centres are hard to access -too far away from me "Hard to get to where I can get tested" The opening hours of the testing centres don't suit me "Being able to get to the testing clinic within opening hours when caring for children" I don't want to take public transport "Public transport required to get to testing centre is inappropriate" I will need to take time off work "hard to find the time with increased workload" I don't have enough time to get tested "Just tooooo busy -work during COVID" I don't have childcare "Depends on if I can get childcare and can find the time" If I get tested, it will impact me financially "My husbands test took 10 days to get results. Which meant he could not work. He is self employed which means he also did not get paid and lost 8 days of income" Social opportunity I am worried what others will think of me having a test/being positive "I will certainly feel worried about what others think of me, especially people who lives in the same household" . 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 August 28, 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 August 28, 2021. When we compared participants with low versus high health literacy, we found very similar issues arose for the top 10 barriers, covering reflective motivation and psychological capability. However, there were significant differences in the proportion of selections for the two groups. People with low health literacy were more likely to select certain capability issues: I'm not sure how to get tested, I'm not sure if this symptom needs testing; and motivation issues: I would prefer to self-isolate instead, I think the test is painful, I'm worried about spreading my illness, I'm worried I will catch COVID-19, I don't want to hear that I'm positive, I don't trust people who are asking me to take a test (Table 5) . . 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 August 28, 2021. ; https://doi.org/10.1101/2021.08.26.21262649 doi: medRxiv preprint . 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 August 28, 2021. . 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 August 28, 2021. . 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 August 28, 2021. Note: Comparison of proportions using independent chi-squared tests. * indicates significance at 0.05 level. . 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 August 28, 2021. These results provide new insights into behavioural barriers to COVID-19 testing, which is central to understanding and controlling the pandemic. Study 1 identified a wide range of barriers to COVID-19 testing that have not previously been described in the COVID-19 literature. These covered all three behavioural drivers in the COM-B model (capability, opportunity and motivation). Study 2 found that the motivation and capability barriers were far more prevalent than opportunity barriers in Australia. The top six most frequently selected barriers were the same as the most important barriers, covering psychological capability and reflective motivation. These issues were around misunderstanding COVID-19 symptoms, preferring to self-isolate rather than get a painful test, and believing there are few local cases. Many of these issues were more prevalent amongst people with lower health literacy. The large range of barriers identified in this study indicate that testing is a complex issue, so it requires a multifaceted approach to improve uptake depending on key barriers in different communities. The most prevalent category of barriers in this study was motivation (e.g. don't believe symptoms are COVID-19), which is likely to influence both testing intentions and behaviours. To address low intentions to test we need to identify and address the specific and wide-ranging concerns that individuals have. For some people, online symptom checkers could be a channel to provide tailored information, where users could be prompted for key concerns to link them to the most relevant information 21,22 , as opposed to searching for this in long 'Frequently Asked Questions' pages on government health websites. However, symptom checkers are not always reliable 23 , and may not be accessed by some groups such as culturally and linguistically diverse communities. Public health messaging could be better targeted to key motivation issues in different communities to address this. Our findings suggest a need to make it clear that self isolation means 14 days not just a few days while you have symptoms, so this is not seen as a good reason to avoid a painful test. We could also appeal to community benefits of testing, as this may be more motivating than individual risk in areas with low rates of COVID-19 24 . To address gaps between testing intentions and actual behaviour, people could be supported to . 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 August 28, 2021. ; plan in advance for the inconvenience of testing and self isolation; for example, managers could have a plan worked out in advance with their employees for how they will notify work and change shifts or work from home. Setting goals in advance has been shown to change various healthrelated behaviours 25, 26 , and this strategy has been applied in Australia to help communities plan for bushfire management 27 . For capability barriers (e.g. not sure which symptoms need testing), we know that government information about COVID-19 in Australia and in many countries is difficult to understand for the average person based on readability analyses 28,29 , and that communities with different language needs are not being adequately addressed 30 . We have also identified that certain groups are more susceptible to misinformation, such as younger people and men 19 . Addressing different information needs in the community may require using less traditional communication channels 31 and different spokespeople to reach these groups, such as social media influencers 32 . Although the Australian testing system is free and widely accessible, there are still important opportunity barriers particularly around financial losses and work expectations that require system changes. The state of Victoria addressed this issue during an outbreak in 2020 by providing financial support for those who were self isolating following a COVID-19 test, to support casual workers and others without access to sick pay 33 . There are also social opportunity issues where stigma and perceived judgment by others could prevent individuals from getting tested, which we have seen in other infectious disease conditions such as HIV 34 . The study occurred in a context where risk of COVID-19 has been relatively low compared to many other countries worldwide. Testing behaviour has been a key part of identifying and controlling emerging COVID-19 outbreaks. This will likely become more relevant in countries where vaccine rollout enables the current large number of new cases to be contained. Our findings from June 2020 are very similar to September results from national flu tracking data, which show 51% of people who experienced COVID-19 symptoms did not get tested 35 . We found 51% of people did not "strongly agree" they would get tested, but most people agreed to some . 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 August 28, 2021. ; extent and addressing perceived barriers may improve this overall positive orientation to COVID-19 testing. Another survey conducted in Australia in August 2020 found much higher rates of testing avoidance, and reported 85% of respondents who had symptoms had not been tested 36 . The effect of vaccine rollout on testing intentions in Australia remains to be seen, with only a small proportion of the population currently vaccinated and none at the time of data collection 37 . We have registered a trial to test a tailored intervention to address the capability and motivation issues identified in this study 38 (ACTRN12620001355965), but broader system approaches are also needed to address opportunity issues. This may include working with the media to avoid the identification and stigmatisation of individuals with positive COVID-19 test results, which has repeatedly happened where a single case is linked to a new outbreak 39 . Our findings provide a starting point for other surveys to identify and address local testing barriers. Media reports and anecdotal data from frontline health professionals may be a useful way to quickly identify emerging local issues that could supplement our list of testing barriers to make it more relevant to local communities. Our initial study only provided a limited list of testing barriers proposed by the NSW Department of Health, but this sample was not representative and did not cover a large number of "other" responses. We addressed this in our second study with a nationally representative sample and a full list of barriers. However, even the second study was not representative of all community groups, particularly those from culturally and linguistically diverse backgrounds which has been identified as a key area of need in Australia. We are currently conducting a separate survey with these groups using interpreters to conduct the survey in preferred languages, as a partnership with Western Sydney Local health District. The findings may not be generalisable to other countries, particularly where COVID-19 prevalence is too high for a test and trace approach to be feasible, and where opportunity issues such as cost or physical access to testing is a problem. . 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. (which was not certified by peer review) The copyright holder for this preprint this version posted August 28, 2021. ; However, understanding COVID-19 testing barriers can help all countries better prepare for the next pandemic where a test-trace-isolate system can be used. This paper uncovers a wide range of barriers to COVID-19 testing that had not previously been identified in the COVID-19 prevention literature. The prevalence estimates show that motivation and capability issues need to be addressed in Australia in order to increase testing behaviour, particularly for groups with lower health literacy. Interventions to improve testing uptake need to identify key barriers for target populations, and specifically address these using evidence-based behaviour change techniques. Our findings support broader advice from international experts in behaviour change, highlighting the importance of diagnosing behavioural barriers in order to increase adherence with COVID-19 prevention behaviours 2 . . 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. (which was not certified by peer review) The copyright holder for this preprint this version posted August 28, 2021. ; Behavioural, environmental, social, and systems interventions against covid-19 Applying principles of behaviour change to reduce SARS-CoV-2 transmission COVID-19 (Coronavirus) statistics. NSW Health The behaviour change wheel: A new method for characterising and designing behaviour change interventions Obstacles to COVID-19 control in east Africa Barriers to COVID-19 Testing and Treatment: Immigrants without Health Coverage in the United States Lack of Health Literacy a Barrier to Grasping COVID-19 Barriers to mass testing for COVID-19 in Africa Parklands Christian College hit by coronavirus scare to remain closed another week as test result delays stretch patience Coronavirus swabs trucked to Melbourne for testing causes delay in results No new coronavirus cases in SA, as Health Minister blames test result delays on IT glitch Disparities in COVID-19 related knowledge, attitudes, beliefs and behaviours by health literacy COVID-19 Misinformation Trends in Australia: Prospective Longitudinal National Survey Don't get it or don't spread it? Comparing self-interested versus prosocially framed COVID-19 prevention messaging. PsyArXiv Prepr Don't be a spreader' to prevent the COVID-19 pandemic COVID-19: Beliefs in misinformation in the Australian community. medRxiv Prepr Serv Heal Sci Brief report: Screening items to identify patients with limited health literacy skills COVID-19) interactive self-assessment. Department of Health and Human Services Victoria Will online symptom checkers improve health care in Australia? Contextualising COVID-19 prevention behaviour over time in Australia: Patterns and long-term predictors from April to July 2020 in an online social media sample Implementation Intentions and Goal Achievement: A Meta-analysis of Effects and Processes Does planning how to cope with anticipated Netw open Multilingual Australia is missing out on vital COVID-19 information. No wonder local councils and businesses are stepping in. The Conversation Young Australians avoid COVID-19 news so traditional health messaging doesn't work -ABC News Young men are more likely to believe COVID-19 myths. So how do we actually reach them? The Conversation Department of Health and Human Services Victoria | $450 Coronavirus (COVID-19) Test Isolation Payment. Victoria State Government Concerns stigma and fear might put people off from being tested for COVID-19 -ABC News SCRUB COVID-19 Survey Wave 7: How the health and vaccine-strategy-right-20210415-p57jey.html 38. ANZCTR -Registration Queensland's coronavirus controversy: past pandemics show us public shaming could harm public health. The Conversation We thank the participants of the longitudinal COVID-19 survey for their ongoing participation in this research. This study was not specifically funded, but in-kind support was provided by authors with research fellowships.