Uncertainty and agency in COVID-19 hotel quarantine in Australia

Mandatory 14-day hotel COVID-19 quarantine was introduced for international arrivals into Australia in late March 2020, with no precedent and little time to prepare. This public health initiative was a key factor in Australia's relatively low COVID-19 burden in the first 18 months of the pandemic. We conducted an empirical bioethics study exploring the experience of people who had quarantined in hotels in Australia. We used in depth interviews to develop an understanding of context and normative analysis to consider whether the way the program is conducted is ethically justifiable. 58 people participated; they had been in hotel quarantine in different parts of Australia in the period March 2020–January 2021. Participants faced considerable uncertainty while in quarantine and many experienced this as burdensome. Some uncertainty resulted from not being given information about key aspects of quarantine, some from rules that changed frequently or were otherwise inconsistent, some from being physically isolated. Lack of information and uncertainty contributed to diminished agency. Communication efforts made by individual hotels was well received. Earlier ethics literature about quarantine does not take into account the context our participants described, where the hotel and supervision arrangements were central to the experience. We argue that more suitable arrangements must be made if quarantine is to be an ongoing proposition.


18
Australia has had a low burden of COVID-19 morbidity and mortality relative to other countries. 19 From the first reports of infection in the region in the first days of 2020 until the second half of 2021 20 the state and territory governments pursued a strong COVID suppression strategy, keeping infection 21 out of the country and eliminating rare outbreaks via contact tracing and public health restrictions. 22 Central to this approach was the adoption of strict border control measures. These included travel 23 restrictions and border closures and, from March 28 2020, mandatory 14-day hotel quarantine 24 (MHQ) for international arrivals. The use of hotels for quarantine was a response to the urgency of 25 the need (there was less than 24 hours between the policy announcement and it taking effect)(1), 26 lack of existing suitable state-and territory-run facilities, as well as the economic boost to a sector in 27 which demand had disappeared overnight. 28 As of late 2021, SARS-CoV-2 was circulating in the community and border restrictions eased in 29 Australia's most populous and most highly vaccinated states but remained in place in other states 30 and territories. Domestic travellers crossing state and territory borders have also been required to 31 quarantine in hotels when there have been outbreaks in the jurisdictions of their travel origin. Hotel 32 quarantine for international arrivals has been administered and funded by state and territory 33 J o u r n a l P r e -p r o o f governments under federal international border rules, but with no federal management oversight. A 34 fixed charge to travellers for quarantine expenses was introduced in July 2020. This fee is around 35 3000 AUD per person, but varies according to jurisdiction and the number of people traveling 36 together. (e.g.(2)). 37 Australia was among the first countries to introduce mandatory hotel quarantine for international 38 arrivals and, as with so much of the early response to the SARS-CoV-2 pandemic, decisions had to be 39 Quarantine in the Australian context 45 Australia has long used border restrictions (or "border protection") to produce a particular social and 46 political national vision, from its White Australia policy (3) to ongoing mandatory 47 detention of asylum seekers.(4) A pattern of focusing on 'security' through physically barring 48 perceived threats from the island nation has fed into a 'Fortress Australia' narrative, one that carries 49 with it a particular kind of exclusion-based nationalism. In this vein, quarantine and border 50 restrictions have been used to respond to the threat of disease since Australia's colonial era. In one 51 interpretation, Alison Bashford describes the imaging of Australia as an island state where 'island' 52 stands for 'immunity', and national 'healthiness' is tied to keeping undesirable others out.(5) 53 Quarantine is a bounded space that delineates (potentially) diseased from healthy, dirty from clean, 54 risky from safe.(6) It would be disingenuous to talk about Australia's supervised COVID quarantine 55 program without at least acknowledging these historical precedents. (7)  Ethics and quarantine in the literature 71 Many countries' public health laws allow for quarantine to be used to restrict the movement of 72 people with an increased likelihood of exposure to an infectious disease (e.g. Ebola, plague, 73 infectious tuberculosis). Prior to the SARS-CoV-2 pandemic, academic writing on the ethics of 74 quarantine focussed on its use within communities to respond to acute outbreaks, rather than as a 75 means of keeping infection out of communities (or countries) altogether. Quarantine is a restrictive 76 public health measure to be invoked justifiably only in situations of extremely heightened risk. The 77 policy in Australia of designating quarantine in a hotel, and charging for it, requires justification given 78 there are other alternatives available e.g. quarantine in one's own home. 79 Much of the pre-SARS-CoV-2 pandemic quarantine literature is framed around human rights and 80 prioritises the infringement of liberty as a key concern. The Siracusa Principles(8) state that a 81 government must use the least restrictive means to achieve desired outcomes in (public health) 82 emergency quarantine situations. In the context of quarantine, Wynia practically interpreted the 83 principles thus: "[D]on't use involuntary quarantine or surveillance devices such as bracelets if 84 voluntary measures will work; don't restrict someone to one room if an entire house is available; 85 don't preclude visitors if personal protective equipment is effective; and don't cut someone off from 86 their work if they can do it from inside quarantine."(9) Upshur added three additional principles to 87 least restrictive means: there must be a clear threat of harm, reciprocity must be upheld, and the 88 program and process must be transparent.(10) To these, Gostin et al introduced the idea of 89 compensating people for e.g. lost wages, but argued in the context of SARS in 2003 that while 90 quarantine is a "severe deprivation of liberty" it is justifiable.(11) Giubilini and colleagues argued 91 that people have a moral responsibility to submit to state enforced quarantine, on the grounds that 92 it generated a small sacrifice for some for the greater good of those around them, but that the 93 burden must be minimised by measures of reciprocity.(12) COVID quarantines have involved much 94 larger populations than earlier modern epidemics, including SARS e.g. (13, 14) In early 2020, Gostin 95 wrote that such measures seemed legally and ethically inconceivable in the United States (15) and 96 that imposition of quarantine that is not risk-based would be unlikely to be legal. Interviews took place over 3 distinct temporal phases. Each phase was triggered by a new or 133 changed MHQ policy. During the first phase we interviewed 30 people: 15 who had completed  day quarantine at home, in the weeks before hotel quarantine was mandated (in March 2020), and 135 15 people who had experienced mandatory hotel quarantine in its early days (March 28 -May 136 2020). In the second phase we interviewed 11 people who had been quarantined in two hotels in 137 Victoria which were subsequently identified as the sources of COVID-19 infection that 'leaked into 138 the community' and deemed unfit for purpose (April -June 2020) . (16, p13) Finally, in Phase 3 we 139 interviewed a further 32 individuals who completed mandatory hotel quarantine later in 2020, when 140 they were required to pay for quarantine (August 2020 -January 2021). This paper reports only 141 those experiences of mandatory quarantine in hotels or the designated quarantine facility in the 142 Northern Territory so includes interview data from 58 participants. Where quotes from participants 143 are used, the state and date refer to the quarantine location and time. 144 Participants 145 Participants were aged 19-75. All of the participants were Australian citizens or permanent residents 146 returning to Australia from abroad. The Phase One cohort was made up largely of two groups: 147 people whose Australian employers had required them to return to Australia, or; people who had 148 moved overseas in early 2020 and did not have well-established lives abroad when the pandemic hit. 149 Phase Two was made up mostly of tourists or longer-term travellers who had been 'stuck' overseas 150 for a reasonably short period. Phase Three was mixed but included people who had travelled for 151 family reasons, and people who had lived abroad for a long time whose circumstances had forced 152 their return. Table 1 summarises participant and quarantine details. We do not claim a 153 representative sample of people who had experienced hotel quarantine; we consider it likely that 154 the people who saw the Facebook recruitment advertisement had strong views about the 155 experience because they were motivated to join a group about the topic. 156 The SARS-CoV-2 pandemic was the context for this study, both in subject and timing. At the 159 beginning of the study the interviewers and many of the participants were subject to movement and 160 mixing restrictions as Australian states and territories attempted to curb the spread of COVID-19 in 161 the (non-returned traveller) community. Such mass lockdown had not been anticipated when the 162 study was conceived, and rules governing quarantine and restriction of movement were in flux. 163 Beyond a broad anticipation that MHQ would be difficult for some, we did not approach the study 164 with any particular expectations about what we would find. We heard very similar information from 165 participants across place and time, but the expressed impact of quarantine experiences varied 166 considerably. Some of the interviews were difficult and participants became emotional; conducting 167 interviews by Zoom meant that efforts to provide the type of embodied care that would ordinarily 168 accompany sensitive interviews -a cup of tea, passing a tissue -had to be modified. A small number 169 of participants needed assistance with technology and sought this from other people in their homes 170 (partners, grandchildren). The authors reflected on how conducting interviews in lockdown and via 171 Zoom might differ from previous research involvement; experiences from and reflections on this 172 study contributed to a paper published elsewhere. (20) It is also worth noting here that many 173 participants expressed that they wanted us to do something meaningful with their experiences, that 174 they hoped change might occur as a result of talking with us. The interviewers and study 175 documentation were very careful to manage this expectation, but it is an indication that the issue of 176 MHQ is morally and normatively important to people who participated in our study. 177 We used a reflexive thematic analysis approach to the interview study.(21) Our initial research 179 question was very broad: how did people experience mandatory (hotel) quarantine in Australia 180 during the COVID-19 pandemic? BH and JW regularly met to discuss interviews, and both 181 J o u r n a l P r e -p r o o f contributed to a rolling memo used to describe impressions and patterns. We both coded all 182 interviews and initially coded inductively using broad brush topics. As we continued to conduct 183 interviews we added and refined codes; as codes were further developed we recoded earlier 184 interviews for consistency. We developed themes based on patterns of meaning we identified 185 through coding and discussion. Working collaboratively was not for the purpose of agreeing themes 186 as such; underlying reflexive thematic analysis is the role of the researcher in thematic development. 187 Rather, we used each other's close understanding of the data to refine the themes that each 188 researcher had identified as central to participants' experiences of MHQ. 189 The normative analysis of the empirical findings was the result of reading and discussion. We There has been considerable inconsistency in hotel quarantine conditions. They vary from state to 213 state, facility to facility, and day to day. Some inconsistency is built into the quarantine process as a 214 J o u r n a l P r e -p r o o f matter of normalised practice, some the result of a lack of care, and other instances are likely the 215 result of using facilities that were not developed and staff who were not trained with quarantine in 216 mind. Inconsistency led to considerable uncertainty. That uncertainty was at the heart of many 217 descriptions of quarantine was initially something of a surprise. MHQ is, on the face of it, the 218 epitome of certainty: people are taken to a hotel room, they are provided with necessities, and 219 released after 14 days. Yet lack of information and resulting destabilising uncertainty was at the 220 heart of many of the more difficult accounts of MHQ. In this section we describe how information 221 was withheld, or perceived as such, and the impact of uncertainty on participants. 222 Withholding of information, actual and perceived 223 Uncertainty began as travellers left the airport via a bus which would take them to their 224 accommodation. Most participants reported that neither the authorities nor the bus driver would 225 tell travellers which hotel they were to be quarantining in. Participants who were familiar with the 226 city of their arrival reported trying to make sense of the route they were taking in order to predict 227 where they would spend the next 14 days; some said it became a game among bus passengers. 228 Many reported being upset or anxious on the bus journey from the airport to their destination, 229 particularly those whose expectations of different hotel conditions had been shaped by information This experience was repeated by participants over time and place, raising the question as to whether 240 information was withheld as part of routine procedure. 241 In phases one and two of this study, COVID testing was available on request for quarantined people 242 who reported symptoms. By phase three, two tests were mandatory (one near the beginning of the 243 stay and one on day 10-12, depending on the location of quarantine). In another example of 244 withheld information, many phase three participants said that they did not receive negative COVID-245 19 test results and that they were told that no news was good news. Participants reported finding 246 J o u r n a l P r e -p r o o f themselves spending an indeterminate amount of time post-test waiting for a phone call or a knock 247 at the door that would herald an unknown fate. 248

They say to you once you've had the test, "if we don't call you then you're fine". 249
That where they would have to go, how much longer they would have to be quarantined, whether or not 256 they would be allowed to keep their belongings -exacerbated anxiety surrounding the test 257 reporting policy. Renee, who had quarantined at a designated quarantine facility, described seeing 258 someone being "carted off", she assumed because of a positive covid test, and said "we didn't know 259 where they got taken to … that was a bit traumatic. So that's the dystopian fear, when the golf cart 260 comes, it's like a Handmaid's Tale." (NT, Jan 2021). For some participants, these uncertainties 261 compounded feelings of anxiety, distress and having lost control. 262 Unclear, incorrect, inconsistent information 263 Other uncertainties appeared due to a lack of policy or coordination or poorly trained staff. While 264 MHQ was set up under emergency circumstances, processes to ensure that quarantined travellers 265 received sufficient reliable information had not been prioritised by the time we finished interviewing 266 in January 2021. We heard the same frustrations over time, remarkably similar tales in April 2020 as 267 in January 2021. With few exceptions, for example, participants reported not knowing when they 268 were allowed to leave quarantine. This was variously because they were not told, or were given 269 unclear or conflicting information. There was confusion over when the 14- what was next that really did my head in." (Lloyd, NSW, April 2020) 294 One participant suggested that not giving a clear date and time for leaving quarantine was a 295 deliberate strategy, to absolve agencies in charge from blame or liability if travellers made onward 296 travel arrangements that they were unable to take up due to changes in quarantine requirements. 297 Rules were inconsistent across hotels, states, and time.(19) They were inconsistently applied by 298 whomever the participant spoke to on the telephone, or whoever was on shift that day. There were 299 exceptions. In the third phase of interviews, some participants reported experiences that reflected 300 an effort on the part of individual hotels to communicate effectively and consistently with returned 301 travellers. For the most part, however, this did not happen. More commonly, participants expressed 302 considerable frustration -it's like you're at the will of whoever is pulling the strings at that moment. 303 (Isobel, VIC, June 2020) -and anger -I got more and more worked-up like, ' happening to others in the same situation was reportedly very stressful for many of the participants 311 who were quarantined alone. They were not isolated from electronic communication; all of the 312 people we interviewed had a smartphone or tablet they could use to be in contact with others at any 313 time. However, participants could not see others in hotel quarantine. They seldom heard others in 314 quarantine, a situation described as "eerie". Those who did report hearing others described 315 distressing situations (yelling in anger, screaming in fear, calling for help). Because they did not know 316 what was happening, they worried. Some reported seeking reassurance from hotel workers that 317 others, strangers whose outburst they had heard, were alright. 318 "We heard screaming from the room across the hall from us. It was just, it was really 319

hysterical, scary screaming […] Begging for help. And um, the security guard wasn't letting 320
her come out, um, and wasn't going in to help. Um, and was just telling her to wait for, to 321 wait for the nurse to come up, and um, she was just, oh, it was just a horrible sound. She 322 was just screaming saying "he's losing consciousness, please, please help". And um, I wanted 323 to go out there and help but it was the security guard who wouldn't let us out. That was 324 really traumatic actually." (Kay, NSW, Dec 2020) 325 326 Some also described a panopticon-style assumption that they were always watched. They could not 327 see the corridors outside their rooms but assumed the 24/7 presence of quarantine enforcers that Having information written down also made a positive difference. While not foolproof, written 340 information was described as more secure and more likely to be correct. Some hotels also made 341 greater efforts and set up WhatsApp or Facebook groups to facilitate the transparent sharing of 342 information between and among the hotel and people in quarantine; this was appreciated by 343 participants whether or not they chose to actively participate in the groups. 344 Uncertainty led to a loss of power 345 Powerlessness was created or exacerbated by insufficient and inconsistent information and rules 346 that were described as inexplicable or irrational. Participants' reported experiences of uncertainty 347 are about more than just not knowing. They felt that the apparent withholding over information was 348 a (sometimes deliberate) ploy to deprive them of agency and power. Rhoda, for example, said "So 349 this is, this is how you were played. How to intimidate and yeah, put you off balance. It was constant. term -"not a true citizen", "dehumanising", "zoo animal", "criminal", "prisoner", "inmate", "just a 353 number". Participants who described a loss of agency tended to describe an initial period of 354 resistance -shouting in the corridor, crying or getting angry with someone on the phone -but as 355 time progressed, they moved towards acceptance. Patricia summarises a typical scenario: "At that 356 stage I just went, "It's just not worth it." I got really upset the day before […], and I just thought, "It's, 357 you just have to accept it." And so we just did everything that they told us to do and, and didn't really 358 have any more outbursts or cries on the phone, or anything. It was what it was." (March 2020, NSW). 359 Participants reported a realisation that they could not be their usual selves in MHQ because, for this 360 cohort, their usual selves had relatively high levels of control over their day to day lives. As Maeve 361 (NSW, Dec 2020) put it, "because you didn't have much information you didn't feel like you were you 362 anyway". This realisation led, for most, to a high level of compliance. They stopped asking questions 363 and accepted that nothing they did could change their current circumstances. Even the participants 364 who had expressed the most anger and frustration at the situation described a capitulation by the 365 end of the quarantine period. 366 Participants' reports evoked quarantine not as 'luxury' hotel surrounds but as a carceral space. 367 Comparisons with prison and detention were common. This was undoubtedly compounded by the 368 spatial dimension of MHQ -the large majority of participants were not allowed to leave their rooms 369 at all in the 14-day period and most did not have opening windows -but was attributed to an 370 agency deficit produced by a lack of information and the interactions with the system. Bernadette 371 sums this up: Yes, you're in a hotel, and the bed is comfortable and all that, but the psychology and 372 the way that people are interacting with you, it feels the same as being in detention. (NSW, Dec 373

2020) 374
Lack of/withholding information was sometimes described as a deliberate strategy, and sometimes 375 as the result of chaotic organisational and staffing measures. Participants reported a connection 376 between not knowing what was happening to them with powerlessness and a loss of agency. We 377 saw a pattern of angry, frustrated or upset resistance followed by resignation, acceptance and 378 compliance. In the next section we explore the relationship between information withholding and 379 power, and whether induced powerlessness fits with the ethics argument that quarantine is 380 experienced as a "small cost" that is morally obligated for the greater good. Federalism commonly sees blame-and cost-shifting between Australian states and territories and 477 the Commonwealth on many issues, and interjurisdictional cultural and political differences are 478 often played out to the detriment of its citizens. There have been official recommendations to 479 standardise conditions in MHQ (32), a move that would make expectations and communications 480 easier and that could minimise the autonomy-limiting experiences found in our study. Improving the 481 quality of information and its delivery to people in quarantine could go part-way to easing the 482 burdens they experience. Mental and emotional load in quarantine must be included in 483 considerations of reciprocity and restriction; to ignore these burdens lessens the ethical justifiability 484 of MHQ now and in the future. 485

486
People who participated in interviews may be more likely to have done so because they have had a 487 meaningful experience, either positive or negative, a situation that applies to interview studies more 488 broadly. We are also aware that we are reporting the experiences of a subset of Australians overseas 489 who can afford to return -airfares during the study period were very expensive and scarce and 490 quarantine is paid by the traveller. For many, quarantine is a liminal space that comes at the end of a 491 long and stressful period of time and may well be the beginning of another. We cannot say that the 492 experiences we report are typical, though we conducted a large number of interviews and heard 493 strikingly similar stories across phase, place and time. 494

495
A review of Australia's hotel quarantine system published in late 2020 begins: "Hotel quarantine is 496 difficult to endure, particularly for vulnerable people."(32) Australia's current Prime Minister has said 497 that quarantine in some form will be a long-term proposition. Continuing to use hotel rooms for 498 quarantine indefinitely is difficult to justify, for practical and ethical reasons. The use of hotels for 499 quarantine has been criticised on efficacy grounds, with a number of infection leaks being attributed 500 to air conditioning systems that allow communication between rooms and common spaces. We have 501 found that hotel accommodation, even in 5-star hotels, needs to be supplemented by clearer 502 provision of information to avoid producing unnecessary harms that create impositions and add to 503 burdens of newly arrived travellers. The very limited literature about experiences of MHQ in 504 Australia and other countries suggests that the difficulties described by this cohort are not 505 unique. (33) 506 In the context of Australia, where relaxing mandated hotel quarantine would likely have led to the 507 introduction of (absent at the time) community spread of COVID-19, we consider that some form of 508 mandatory supervised quarantine in an assigned location is likely justifiable. As vaccination levels 509 increase in Australia, this calculus is changing. 510