Pathways and obstacles to social recovery following the elimination of SARS-CoV-2 from Aotearoa New Zealand: a qualitative cross-sectional study

Abstract Background Many public health experts have claimed that elimination strategies of pandemic response allow ‘normal social life’ to resume. Recognizing that social connections and feelings of normality are important for public health, this study examines whether, and for whom, that goal is realized, and identifies obstacles that may inhibit its achievement. Methods Thematic analysis of narratives obtained via a qualitative cross-sectional survey of a community cohort in Aotearoa | New Zealand. Results A majority of participants reported that life after elimination was ‘more or less the same’ as before the pandemic. Some became more social. Nevertheless, a sizeable minority reported being less social, even many months after elimination. Key obstacles to social recovery included fears that the virus was circulating undetected and the enduring impact of lockdowns upon social relationships, personal habits and mental health. Within our sample, old age and underlying health conditions were both associated with a propensity to become less social. Conclusions Elimination strategies can successfully allow ‘normal social life’ to resume. However, this outcome is not guaranteed. People may encounter difficulties with re-establishing social connections in Zero-COVID settings. Measures designed to overcome such obstacles should be an integral part of elimination strategies.


Introduction
Since the onset of the COVID-19 pandemic, the respective merits of mitigation and elimination strategies have been hotly debated in public health circles. 1 In mitigation, the virus continues to circulate, albeit at reduced levels, due to non-pharmaceutical interventions (NPIs). Elimination, or 'Zero-COVID', involves initial deployment of stringent NPIs with a view to reducing community spread to zero; further outbreaks are then guarded against via strict border controls. Though vaccinations are playing an increasingly important role in international responses to COVID-19, the mitigation-elimination debate remains salient-both when vaccination rates remain low, 2 and when planning for future pandemics. 3 Notwithstanding technical questions regarding elimination's feasibility (which may prove difficult in developing countries or when faced with highly transmissible SARS-CoV-2 variants), proponents of each strategy are motivated by competing understandings of how best to protect the public good. Advocates of mitigation argue that the economic damage wrought by the stringent lockdowns and border closures necessary to secure Zero-COVID status would have long-term repercussions for public health. 1 Supporters of elimination counter that it minimizes COVID-19 fatalities, supports economic recovery and involves less overall restriction of civil liberties. 1,4 Elimination is also seen as having social and psychological benefits, often couched in the idiom of allowing 'normal social life' to resume'. 5,6 These benefits are themselves a public health matter, since a sense of 'normality' and a high quantity and quality of social relations are associated with improved physical and mental health. 7,8 Current evidence supports both the clinical and economic arguments in favour of elimination. 4,8,9 However, although research indicates that people living through mitigation strategies have fewer social contacts once 'lockdown' restrictions are lifted than they did pre-pandemic, 10 little is yet known about whether a 'normal' or satisfying social life returns after elimination. Measuring such attributes requires a qualitative, social constructionist approach, since feelings of 'normality' and 'satisfaction' principally inhere in subjective evaluations of a practice or situation compared to an imagined baseline of 'normal activity'; these evaluations are then expressed through narrative in a process of 'narrative sense-making'. 11 Generated by individuals whose yardsticks of evaluation are both constantly evolving and distinct from those of others (even as they are to some degree co-constructed), each narrative must be engaged with qualitatively, and on its own terms, whilst also remaining attentive to patterns across different cases. In this paper, we thus develop a thematic analysis of narratives describing life in a Zero-COVID setting-New Zealand between February and August 2021-to add greater nuance to existing claims regarding the extent and ease of social recovery following viral elimination.

Site selection
New Zealand has received international acclaim for the success of its elimination strategy. Elimination first occurred in May 2020, following a 49-day national lockdown. 12 On 8 June 2020, the country moved to 'Alert Level 1': activities could resume without restriction-although enhanced recordkeeping was recommended. Small community outbreaks in August 2020 and February 2021 led to short periods of enhanced restriction, mostly in Auckland. Nevertheless, until the Delta variant triggered a further nationwide lockdown in August 2021, New Zealand had enjoyed many months of freedom from COVID-19 restrictions. With vaccine rollout progressing slowly during this time (only 18% of the population were double vaccinated by the time of this study), 13 elimination was the cornerstone on which prospects of social recovery depended. Indeed, the New Zealand government cited 'get[ting] back to a sense of normality' as a reason to persist with its elimination strategy. 14 New Zealand is thus an instructive site in which to investigate how post-elimination social life has been experienced.

Research design
Following Cole and Knowles' argument that every 'exploration of an individual life-in-context brings us that much closer to understanding the complexities of lives in communities', 15 this study sought to document the range of possible experiences that people could have following elimination, using thematic analysis of respondents' narratives to identify key dynamics underpinning different behavioural pathways.
Recognizing the flexibility, scalability and richness of online surveys as a method of gathering qualitative data, 16 we advertised a self-administered online survey via nationwide Facebook and Instagram campaigns between 18 August and 25 August 2021, with bespoke campaigns targeted at men and younger age groups to heighten variation within the sample. These campaigns recruited 225 participants. The study was also advertised to a database of 1417 contacts who had participated in previous surveys, themselves recruited via campaigns intended to maximize variation. 815 of these contacts completed the survey (a response rate of 57.5%).
The final respondent pool contained a wide breadth of ages and regions of residence but, despite attempts to maximize variation, contained disproportionate numbers of women, New Zealand European/Pākehā people and university graduates-as is often the case with survey research in New Zealand. 17 Respondents were asked to comment on multiple aspects of New Zealand's pandemic response, to evaluate how much their lives had changed since before the pandemic, and to provide narrative elaborations (see Supplemental Material, Annex 1).

Analysis
Following an unstructured familiarization phase, respondents' answers to open-ended questions were independently coded by two researchers, using both anticipated themes and emergent themes that were discovered in the data and refined via discussion. Data were checked against an initial framework and refinements made as necessary. We also conducted a descriptive statistical analysis to examine whether certain codes were associated with respondent characteristics such as gender, age, ethnicity, education status, household size or medical vulnerability to COVID-19 (see Supplemental Material, Annex 2).

Results
966 of the 1040 survey respondents (92.9%) answered Q6.4-a 'tick-box' question about how their social life compared to life before the pandemic. 546 respondents (52.5% of the total) provided narrative elaborations. Thematic analysis of these narratives revealed three overarching patterns of social behaviour in the wake of elimination.
Returning to a pre-pandemic 'normal' 531 respondents indicated their friendships and social life had been 'more or less the same' over the previous six months (i.e. from February to August 2021) as before the pandemic. When elaborating on their answers, most attested that nothing had changed (Table 1 -Quote 1). They described level 1 as allowing a return to normality (Quote 2), and feeling grateful and 'lucky' that the New Zealand government had adopted an elimination strategy (Quotes 3 and 4). Some suggested the social gains of life at levels 1 and 2 justified the 'sacrifice' of lockdowns (Quote 5). Several mentioned that elimination had allowed them to feel 'safe' (Quotes 3 and 6), alleviating their feelings of 'fear' (Quote 7). One respondent, who had spent three months in the UK, which has adopted a mitigation strategy, contrasted the 'normality' of Zero-COVID New Zealand with the 'frightening' feeling of life in Britain (Quote 8). Only occasionally was the 'normality' of social life linked to a wilful blindness towards the pandemic (Quote 9). Interestingly, although most respondents were supportive of the vaccination programme elsewhere in the survey, none mentioned it contributing to their experiences of social recovery.
Eleven respondents, some of whom indicated that their social life had become 'a little different', described minor changes resulting from heightened awareness of health and hygiene as opposed to substantive shifts in social activity (Quotes 10 and 11). Nevertheless, several reports of a 'return to normality' were haunted by a sense of contingency, with respondents indicating that life might have been less normal if they had friends who were border workers (Quote 12), or if they were less adept at handling differences of opinion within their relationships (Quote 13). This indicates a recognition that the post-elimination context may put strain on certain relationships.
Becoming more social 95 respondents described intensified social activities following elimination. Some framed this as a response to lockdown, couched in the idiom of 'making up for lost time' (Table 2 -Quote 1). Others explained that the pandemic had revealed the fragility of social freedoms (Quote 2), and, indeed, human life (Quote 3), inspiring them to prioritize friendships and social activities more than previously (Quotes 4 and 5), and to appreciate their loved ones more (Quote 6).
In several cases, being 'locked down' had afforded opportunities for new friendships to arise in the local community, and these had persisted into the post-elimination period (Quote 7). Others indicated the pandemic had led them to forge closer relationships, allowing them to be more honest about their emotions (Quote 8), or rendering friendships more 'meaningful' (Quote 9).

Becoming less social
253 respondents presented narratives in which life after elimination was associated with a decline in the quantity or quality of their social relationships. In 24 cases, this was explicitly linked to the border closures integral to the elimination strategy (Table 3 -Quote 1). For the majority, however, their social lives within New Zealand had changed. They reported changes in activities ('socializing less' or spending less time in public places -Quotes 2 and 3), changes in character (becoming 'less social' -Quote 4), and an overall sense of their world 'having shrunk' (Quote 5).
For some respondents, these changes were linked to fear that SARS-CoV-2 might have entered New Zealand and be in circulation, despite announcements that community transmission had been eliminated. They worried that, by socializing, they might either contract COVID-19 (Quote 6) or pass it to others (Quote 7). Others described how the anxiety and stress they had experienced during the pandemic had triggered feelings of depression that then impeded them from undertaking social activities (Quote 8). Such anxieties could have knockon consequences for others, with a diminished social life sometimes arising from frequent cancellations (Quote 9).
In other cases, the change in social patterns was presented as a consequence of the 7-week lockdown in March-May 2020 (and, in some cases, subsequent local lockdowns). One respondent reported 'lockdown fatigue' (Quote 10), whereas others explained lockdown had 'habituated' them to staying at home (Quote 11). Some friendships had been strained during lockdown due to disagreements over rule-breaking (Quote 12), whereas other respondents felt that the lockdown had led people to 'withdraw into family life' at the expense of other relationships (Quote 13). Comparable dynamics were sometimes reported as having arisen even after lockdown: disagreements over vaccine uptake or the government's COVID response had caused some people to sever ties with friends (Quote 14), whereas those whose friends had a propensity to 'stay at home'-for whatever reason-noted that those relationships now felt thinner, with less to talk about (Quote 15). Not all respondents viewed such changes as negative. Some appreciated being able to focus on their 'most important' relationships (Quotes 3 and 16), the 'deeper' conversation afforded by smaller gatherings (Quote 17), or being able to 'indulge their introvert side' (Quote 18). For others, however, the loss of connection had fostered feelings of isolation (Quote 19) and deteriorating mental health (Quote 20).

Response distribution
The three patterns described above could be observed amongst respondents of all backgrounds. Descriptive statistical analysis (Supplemental Material, Annex 2) did not indicate any significant associations with ethnicity, education status or residence size. There were, however, notable associations with health status and age. Respondents with underlying conditions were more likely to have become less social and less likely to have become more social than those without such conditions. Similarly, those in younger age brackets were more likely to have become more social, and those in older age brackets more likely to have become less social-although this pattern may partly reflect the increased prevalence of underlying health conditions in older age groups. There was also a strong association with gender: women were more likely to report having become more social or having become less social, and men more likely to report continuity. This finding may reflect longstanding gender roles in Western societies, in which women have often been deemed responsible for thinking about and managing social relationships and thus potentially more inclined to detect and report changes in their social networks. 18

Main findings of this study
Our findings indicate that elimination strategies can indeed allow many people to regain a sense of 'normality' within their social lives, even inspiring them to cultivate social relationships more actively than ever before. Such positive outcomes, however, are not guaranteed: many respondents reported lower levels of social contact after the virus had been eliminated than before the pandemic. Moreover, becoming less social was associated with older age and underlying health conditions, suggesting that those most vulnerable to COVID-19 may be least able to achieve social recovery. As our thematic analysis reveals, the shrinkage of a social network is not always undesirable: it need not equate to 'loneliness', and may even be experienced as a relief. Nevertheless, research in New Zealand and beyond points to strong correlations between the number and quality of social relationships and overall physical and mental health. [19][20][21][22] There are also known psychological benefits associated with living in a world that feels 'normal'. 7 Enabling people to restore or expand their pre-pandemic social networks is thus a public health imperative.
Elimination strategies can be improved by anticipating and mitigating against common obstacles to social recovery that arise in post-elimination settings. Our study identified two. First is the ongoing fear of contagion, which is not necessarily eliminated with the virus-especially when a pandemic continues to rage internationally. In addressing such fear, policy makers must strike a delicate balance between promoting appropriate levels of caution (e.g. using contact tracing apps) and encouraging people to take advantage of hard-won freedoms. Public health messaging should champion adjustments made to make public venues COVID-secure (e.g. by increasing airflow), promote low-risk forms of social contact (e.g. meeting outdoors) and emphasize that reconnecting with others is itself a public health good, perhaps harnessing the tropes of 'kindness' and 'togetherness' that underpinned New Zealand's initial messaging around lockdown. 23 Second, the case of New Zealand shows how even relatively short lockdowns, when stringent enough to achieve elimination, can have long-term impacts on relationships, personal habits and mental health-and, by extension, social networks. Funding of mental health services (including Table 3 Becoming less social   Quote 1 Can't visit my best friend in Australia who has gone through a really tough time.
We've drifted apart a bit because of this.
Māori and Pākehā woman, 30s Quote 2 I seem to go out less to meet up with friends. European woman, 50s Quote 3 It has been harder to reconnect with others. My attendance/participation in usual activities (e.g. going to church) has changed, I don't feel so well connected. I'm wary of larger gatherings, not going to movies etc. as much as before.
Pākehā man, 50s systemic psychotherapies) should thus be increased; indeed, in New Zealand this has been a priority since well before the pandemic. 24 Healthcare providers could encourage volunteering and other forms of social prescribing. 25,26 Public health messaging should highlight the value of repairing interpersonal tensions that have arisen during the pandemic, disseminate advice on how to do this, and openly acknowledge that 'returning to normal' may need to be undertaken consciously and effortfully, rather than occurring automatically.

What is already known on this topic
Public health scholars, politicians and journalists advocating for elimination frequently claim it enables a 'return to normality' and facilitates social recovery, sometimes citing New Zealand as an example. 5,14,27,28 The evidence underpinning such claims is either anecdotal or based on thin, quantitative, measures of activity that overlook how postelimination life is subjectively experienced. 29,30 Few studies have addressed this latter concern. Research immediately following New Zealand's 2020 lockdown documented high levels of aversion to mixing with strangers, 31 uncertainty about the trajectory of the pandemic, 32 and worsening mental health. [33][34][35] However, little is known about how long such dynamics persist.

What this study adds
To our knowledge, this is the first study to examine how life in any elimination setting has been experienced more than 6 months after elimination, let alone to capture those experiences in people's own words. Although our study generally supports arguments in favour of elimination strategies, it adds nuance by identifying several obstacles that may impede social recovery. Foremost amongst these are ongoing anxieties that SARS-CoV-2 might be circulating undetected, and the long-term repercussions that lockdowns have had on people's mental health, inclinations to be sociable, and friendships. Additional obstacles include stigma against border workers, disagreements over the pandemic response, and hopelessness linked to cancelled plans and uncertain futures. Future research should investigate whether comparable patterns are observed in settings that have achieved elimination in different ways-such as Taiwan, which prioritized contact tracing over lockdowns. 36 Such research would usefully inform decisions over what kind of elimination strategy governments should aspire to. Future work should also examine perspectives missing from this study, the design of which precluded access to under-18s, those too economically disadvantaged to have internet access, and non-English speakers. In New Zealand, more research is needed on Māori and Pacific experiences, given the relatively small numbers participating in the survey, the long-standing health disparities and structural inequities affecting these groups, and their disproportionate vulnerability to COVID-19. 37,38 Limitations of this study Our survey's cross-sectional nature meant the narrative sensemaking we recorded was occurring at a specific moment in time. By happenstance, community outbreaks of the Delta variant led to a new lockdown just before our survey launched. Pessimism may have led some respondents to exaggerate what had been lost since before the pandemic and others to romanticize life at level 1.
Although thematic analysis allowed us to identify three post-elimination behavioural pathways and the causal logics underpinning them, we cannot be certain of their relative prevalence amongst New Zealand's public. Given the high proportion of respondents identifying as women (76.6%) and reporting underlying health conditions (31.7%), and a mean age (48.8) above the national average (37.2), the pathway of 'becoming less social' is probably overrepresented in our results. Nevertheless, although a statistically representative study would better delineate the scale of the challenges, the high volume of such narratives enabled us to identify a broad range of dynamics that can obstruct social recovery, and to suggest specific measures and messaging that could have better supported the public in their transition to postelimination life.

Conclusion
Where possible, governments should consider elimination strategies of pandemic response on the grounds that they can enable social recovery, as well as guarding against excess mortality and limiting economic damage. Nevertheless, pandemic planning must anticipate the challenges that people might encounter in transitioning back to a satisfying life. Pandemic control measures can strain social relationships in various ways and people may need support in repairing those relationships. Clear guidance on how to socialize safely, and the importance of doing so, is also crucial to ensure members of the public can safeguard access to social support and thereby protect their physical and mental health-and the health of others.

Supplementary data
Supplementary data are available at the Journal of Public Health online.