key: cord-0915623-l2ighvho authors: Cortis, Natasha; Smyth, Ciara; valentine, Kylie; Breckenridge, Jan; Cullen, Patricia title: Adapting Service Delivery during COVID-19: Experiences of Domestic Violence Practitioners date: 2021-07-27 journal: Br J Soc Work DOI: 10.1093/bjsw/bcab105 sha: 740a134f0dd543330f405836e5ea2b1ada2900bd doc_id: 915623 cord_uid: l2ighvho COVID-19 rapidly altered patterns of domestic and family violence, increasing the complexity of women’s needs, and presenting new barriers to service use. This article examines service responses in Australia, exploring practitioners' accounts of adapting service delivery models in the early months of the pandemic. Data from a qualitatively enriched online survey of practitioners (n = 100) show the ways services rapidly shifted to engage with clients via remote, technology-mediated modes, as physical distancing requirements triggered rapid expansion in the use of phone, email, video calls and messaging, and many face-to-face interventions temporarily ceased. Many practitioners and service managers found that remote service delivery improved accessibility and efficiency. Others expressed concerns about their capacity to assess risk without face-to-face contact, and were unsure whether new service modalities would meet the needs of all client groups and reflect best practice. Findings attest to practitioners' mixed experiences during this period of rapid service innovation and change, and underline the importance of monitoring emerging approaches to establish which service adaptations are effective for different groups of people, and to determine good practice for combining remote and face-to-face service options in the longer term. COVID-19 rapidly affected all aspects of social and economic life, generating new needs and risks and exacerbating inequalities. Evidence quickly accumulated showing the pandemic's gendered impacts, including women's increased domestic responsibility and mental load, escalation of violence against women at home, increased complexity of women's needs, rising service demand and barriers to service use (Alon et al., 2020; Kaukinen, 2020; Mahase, 2020; Raile et al. 2020; Sharma and Borah, 2020; Speed et al., 2020; Usher et al., 2020) . Even in Australia and New Zealand, where infections were comparatively low, public health measures, including temporary closures of businesses and schools and directives to stay home, hugely impacted individuals and families, and the service systems that support them. Social service providers needed to quickly reconfigure supports to clients and communities, as physical distancing rules made innovation an urgent imperative (Heinonen and Strandvik, 2020) . This article focuses on processes of innovation and adaptation in domestic and family violence services, drawing on practitioners' experiences of changing service models captured in an Australian survey in July 2020. The study objective was to understand service responses to COVID-19 and determine the ways frontline practitioners, managers and other staff were experiencing the pandemic; any responses they considered innovative or effective; and long-term implications for service delivery. Incorporating constructivist principles, the survey utilised a Teaser text During COVID-19, many domestic and family violence services shifted to remote methods of service delivery. A survey was conducted to examine Australian practitioners' experiences of adapting services during the pandemic. For many practitioners, technology-mediated service delivery was a new way of working, adopted out of necessity. Domestic violence work become more complex and challenging during the pandemic. Without face-to-face contact, it was more difficult for practitioners to assess risk. However, remote service delivery enabled services to be maintained and improved access for some client groups. Research should continue to monitor the impacts of technology-mediated service delivery and engage clients in assessing what works. hybrid design, containing some questions with pre-determined (closed) response options and multiple open-ended questions enabling participants to express experiences of service delivery through the pandemic using their own language and frames. Findings offer insight into practitioners' experiences of adapting service models to incorporate technology-mediated client engagement methods, in the context of the pandemic. Whilst digitally mediated methods were available prior, the use was not widespread and effectiveness unclear. In the context of COVID-19, many practitioners felt their service innovations vastly improved service accessibility, suggesting some adaptations will endure beyond the crisis. Others, however, were conflicted about the impacts of remote service delivery on their ability to effectively assess risk and support clients. This was evident in practitioners' comments but obscured in structured (closed) survey questions. As such, the findings both provide insight into processes of abrupt service transformations associated with COVID-19, and underline the importance of embedding qualitative approaches within online surveys to recognise and value practitioners' interpretations and narratives. In the early weeks of the outbreak and as governments around the world sought to respond, advocates sounded alarms that lockdown measures were likely to increase violence against women and girls, in a 'shadow pandemic' (Koshan et al., 2021; Morgan and Boxall, 2020; Pfitzner et al., 2020a; Sharma and Borah, 2020; UN Women, 2020) . Without an evidence base specific to the unfolding circumstances, concerns about increased violence were initially based on media reports and research showing tendencies for violence against women to increase following natural disasters and emergencies (Harville et al., 2011; Parkinson and Zara, 2013; International Rescue Committee, 2019; Parkinson, 2019) . American research, for example, showed increased domestic violence amongst the social crises following Hurricane Katrina (Schumacher et al., 2010) . In Australia, James et al. (2014) showed women's increased vulnerability to domestic violence during and following a major cyclone, with psychological abuse in particular increasing. However, these studies of disaster responses left the service implications of a pandemic, and physical distancing, in particular, unknown. Indeed, the circumstances of COVID-19 and associated service disruptions differ substantially from disasters destroying physical environments and dispersing communities (Emezue, 2020) . Importantly, the pandemic restricted human contact, prevented victims from accessing supports and made it difficult for practitioners to follow established service models. However, the pandemic did not destroy actual service infrastructure, leaving open possibilities for services to successfully adapt to new conditions of remote, physically distanced provision. Such adaptation was of critical importance, as social and economic stressors associated with the pandemic, along with lockdowns, provided fertile ground for escalating physical, emotional and financial abuse, and for tactics of surveillance, isolation and coercive control. Early information from Hubei Province in mainland China indicated increased calls to domestic violence services during February 2020 (Allen-Ebrahimian, 2020). Subsequent data from Spain showed a 23 per cent increase in intimate partner violence, with psychological violence especially prevalent (Arenas-Arroyo et al., 2020) . In late March, a UK survey of frontline services found three-quarters were operating with reduced capacity (SafeLives, 2020). Face-to-face appointments and group work were most frequently reduced or cancelled, followed by children's services and court work. Services experienced challenges with information technology (IT) and Internet access, and practitioners raised concerns about client privacy and confidentiality, working at home and isolation. In Australia, domestic violence was also found to increase in the early months of the pandemic. A large online survey of women showed the probability of violence was over 1.3 times higher for those isolated from family and friends, and 2.8 times higher amongst women experiencing financial stress . A survey of eighty frontline workers in New South Wales (Foster and Fletcher, 2020) provided early indications of the changes domestic violence services were making, including working at home. At that time, 71 per cent of respondents reported working remotely some or all of the time. Simultaneously, workers were changing service procedures and working with limited referral options, as other services closed or reduced capacity. Almost three-quarters ceased offering face-to-face services, and there was much uncertainty about capacity to pay for the technology, infrastructure and training needed as staff transitioned to work from home. Initially, services shifted from preventative, early intervention approaches to more crisis-driven, reactive responses, driven by victim's isolation with perpetrators (Pfitzner et al., 2020a) . Service innovations in the period involved increased use of voice and video calls, email, webchat and messaging services, to enable continued client engagement. Other innovations included using code words in text and phone communications to signal risks whilst perpetrators were present; encrypted web-based video calls to avoid downloading phone apps that perpetrators might find; and transporting women and children to safety using women-run ridesharing (Pfitzner et al., 2020a) . As practitioners quickly transformed services, declines in their mental well-being were observed, despite availability of support, supervision and debriefing (Pfitzner et al., 2020c) . Widespread adoption of remote service delivery by domestic and family violence services through this period was a major development, yet it proceeded with little research guidance. Scholarship on technology-mediated social service delivery, and practitioners' experiences of it, has been thin. Prior to the pandemic, communication technologies and helplines provided options for social work and therapeutic practice, but only a small literature explored the implications of using these with services users, and amongst colleagues, supervisors and collaborators (Byrne and Kirwan, 2019) . Telephone crisis helplines were well-recognised for enabling workers to develop rapport, listen non-judgmentally and offer advice and referral options, but technologies were typically used for oneoff, time-limited interventions, often as pathways to traditional services (Middleton et al. 2014; Pirkis et al., 2016) . Importantly, prior to the pandemic, using video calls and chat apps to provide support, information and counselling were not universally accepted. Bayles (2012) questioned whether video calls were acceptable for providing therapeutic support, given the loss of physical connection and difficulties observing and interpreting clients' non-verbal and physical cues without physical proximity. Byrne and Kirwan (2019) found that despite embracing multiple communication tools, social workers were aware about conducting relational work electronically. Social workers used text messaging primarily administratively, for setting up meetings or coordinating teams, but avoided remote counselling, talking about personal matters or exchanging confidential information via these means, given the potential for misinterpretation. They questioned whether technology should be considered an 'add-on' to existing practice, or a fundamentally different practice model which alters how people interact and relate, effectively depleting service quality (Byrne and Kirwan, 2019) . With debates about technology-mediated service delivery unresolved, COVID-19 provided a 'tipping point' (Chen et al., 2020) , with telehealth used by necessity not choice. The prolonged, recurrent nature of the pandemic means these arrangements are likely to endure, making it imperative to examine and monitor the processes and impacts of change. Developments during COVID-19 cast light on longstanding legacies in human services in which technology-mediated services are framed as 'substandard' compared with in-person services (Burgoyne and Cohn, 2020 ). Yet, tele-counselling may increase user satisfaction, and appropriateness for diverse populations and issues, in many cases producing similar outcomes as face-to-face therapy (Burgoyne and Cohn, 2020) . In the swift practice transformations following COVID-19, videoconferencing has been found to improve access, including for rural or isolated clients, clients with physical disability or urban clients lacking the time, transport or financial resources to access face-to-face services (Burgoyne and Cohn, 2020) . Integrating smartphones into practice has also been seen to improve practitioners' availability and attentiveness, reducing relational distance and enabling deeper engagement, especially with young people (Burgoyne and Cohn, 2020) . Chen et al. (2020) highlight benefits of increased volumes of users served, reduced logistical barriers, decreased no-shows, easier scheduling, protection for people with disabilities or health conditions, increased privacy by eliminating the need for physical attendance and increased access to support for those unable to leave home. Chen et al. (2020) , however, also point out some disadvantages of telehealth, including disruption due to glitches, difficulty reading non-verbal communication, loss of intimacy and privacy and increased disparities affecting people unable to use technology. For Banks et al. (2020) , technology use during COVID-19 raises ethical questions. They note that as it became too risky to undertake everyday service activities, such as home visits, meetings, accompanying people to court or appointments, or visiting residential settings, the expansion of remote service delivery made it more difficult for practitioners to collect the information needed to evaluate conditions affecting clients and detect potentially abusive relationships (Banks et al., 2020) . So far, few studies have focused specifically on domestic and family violence practitioners' experiences of technology-mediated service delivery during COVID-19. Digital mitigation strategies, including screening, risk awareness, education, referral and support to safely leave abusive relationships, are particularly important whilst usual supports cannot be delivered under lockdown conditions, and with many victims isolated with perpetrators (Emezue, 2020) . Some domestic violence survivors reportedly prefer technology-based interventions and online support for its practicality and confidentiality, although barriers exist, related to technology and connectivity, especially for people in rural communities, those with low incomes, and older adults (Emezue, 2020) . Australian survey research (Pfitzner et al., 2020b) found domestic violence practitioners considered service accessibility to improve, as virtual platforms offered efficient ways to reach more clients. We examine practitioners' perspectives in more detail, using research on their experience of adapting domestic and family violence services in mid-2020. To capture practitioners' experiences, a survey was developed, with the protocol for informed consent approved by the University of New South Wales Ethics Panel. The survey captured information via conventional, pre-determined structured survey questions informed by previous practitioner surveys (Cortis et al., 2018) . Closed questions captured participants' roles, locations and demographics; their employment and service characteristics; and whether overall requests for support, face-to-face contact, phone, online and video calls had either increased, stayed the same or decreased since the outbreak of the virus. Closed questions were also used to capture changes in supervision and support, whether risk management instruments had changed, and whether practitioners worked remotely. Survey design was qualitatively enriched by including multiple free-text questions. These enabled practitioners to use their own language and frames, without word limits, to describe changes arising from the pandemic in their contact with clients, risk assessment and safety planning, supervision and working arrangements, and what would help strengthen service capacity in the context of COVID-19. Questions elicited how COVID-19 impacted on need in the community and demand for service; how their contact with people affected by violence had changed in the context of COVID-19; and any changes in recognising signs of violence and assessing risk; or in their supervision and working arrangements. As similar themes relating to technology-mediated service delivery arose across the open-ended questions, analysis draws material from across the survey questions, using the data-set as a whole to understand practitioners' experiences and perceptions. This approach ensured comprehensiveness, because whilst some participants directly answered questions, others answered or elaborated later or contributed relevant experience before they were asked. The number of open-ended questions posed, and our treatment of responses as a cohesive data-set, meant the survey results contributed context-rich accounts via what may otherwise be experienced as a distanced, 'tick-box' method. This was especially important in the context of COVID-19, for enabling respondents to provide their own explanations of new experiences, which could not be assumed to the extent required by a fully structured survey with fixed response options. However, whilst it enabled qualitative input and analysis, the approach differed from a flexible, semi-structured interview or focus group method. Unlike a fully qualitative approach, the self-administered questionnaire contained structured questions presented in standard order (Braun et al., 2020) . Although respondents could freely contribute text, the approach precluded conversation, follow-up probing and field observation typical of fully qualitative methods (Gobo and Mauceri, 2014; Bazeley, 2018) . Notwithstanding, the approach strengthened the method, capturing richer, more complex accounts than conventional surveys. As practitioners are not connected in a single network or registry, the survey was distributed primarily via networks of not-for-profit community-based organisations that provide a range of specialist and non-specialist services to people affected by violence. Whilst statutory workers, such as those in hospitals and police were not excluded, recruitment targeted community sector practitioners. Respondents were more difficult than usual to reach due to remote working, and we used multiple methods to access them. Primarily, we contacted networks of relevant organisations via their representative (or umbrella) associations. Specifically, we distributed the online survey link to two state-based networks of specialist domestic violence organisations, one state-based women's health network and one national network of family services, requesting that they pass the survey invitation onto members via emails, newsletters or other channels. To promote national reach, we supplemented this by emailing links directly to seventy-five domestic violence services with publicly listed email addresses in the jurisdictions not covered by the networks. The survey was completed by 100 practitioners during July 2020. The vast majority (92 per cent) were women. Most of them (58 per cent) were located in New South Wales, and the same proportion was based in a capital city. Participants delivered a range of services. A third (33 per cent) said their service focused on family support and advocacy, 32 per cent on court support, 18 per cent on general counselling and 17 per cent on accommodation, including refugees and tenancy support. Twothirds (64 per cent) were frontline practitioners, including counsellors, caseworkers, social workers or family support workers; whilst 28 per cent said they were managers, including centre coordinators, team leaders or Chief Executive Officers (CEOs). Almost a quarter (23 per cent) said their highest educational qualification was in social work, whilst 41 per cent said other community services, welfare, counselling or youth work disciplines. The remainder was qualified in psychology, legal fields, nursing, social science, social policy or other disciplines. Closed survey responses were analysed using descriptive statistics, such as frequencies and categorical comparisons. As indicated above, rather than separately analysing comments on each open-ended question, we treated material as a cohesive data-set (Braun et al., 2020) . Themes were identified following a reading of all open-ended comments, to identify important issues and sub-issues relating to the changes to services in the context of COVID-19, and practitioners' range of experiences. This provided insight into the challenges faced by practitioners, offering more nuanced accounts than in closed survey questions by showing that practitioners felt highly conflicted about new arrangements and procedures for working with clients and communities. Whilst the survey did not ask about exact client numbers, responses attest to changes in patterns of demand. Some experienced initial drops in demand, often considered temporary. As one respondent explained: Initially it was slower than usual when everyone went into lockdownwomen just tried to cope. As the time went on it became harder and women when they had the chance were reaching out for support. [The] pace [is] increasing further now restrictions have lifted (CEO, Sydney). Most respondents, however, stated that their service experienced increased demand. When asked whether requests for support to their service had increased, stayed the same or decreased since the outbreak, around three in five (59 per cent) reported increases. When asked if complexity of need amongst people using their service had increased, stayed the same or decreased, almost three-quarters (73 per cent) reported increases, likely reflecting victims' isolation with family members using violence, the impacts of children home from school or childcare, impacts of job loss and the inability to draw on face-to-face support from friends or extended family. Many commented that people with no previous histories of violence or who had not previously sought support required service responses for the first time. In addition, closures or disruptions to services, including refugees operating at low capacity to comply with distancing requirements, impacted on other services, which faced reduced referral options. Services sought to improve accessibility, including via new offerings of online support and expanded contact hours, increasing demands on practitioners. One described change, for example: Clients were very wary with their husbands in the home 24/7 and only wanted voice messages, and they called us when their husband was not around. We were supporting clients outside of the working days 24/7 so there was a support person to speak to (frontline practitioner, Sydney). Other modes of contact with clients greatly increased. Where I would normally have one 60-minute (face-to-face) session a week with a young person or mother, these blew out to multiple 60-minute phone calls per week (frontline practitioner, regional Queensland). Additional challenges were evident. Practitioners explained that people sequestered with perpetrators found it more difficult to safely and discreetly seek help: Women have been finding it harder to make calls due to their abusive partner being in the house all the time. It is also hard for them to google to find out about services. There have been some zoom DV [domestic violence] groups, but women still living under the same roof [as their abuser] are not able to do these (Social worker in a child protection service, Sydney). Practitioners described extra tasks, including checking on clients more regularly, choosing suitable technology and apps, ensuring clients had useable devices, assisting clients with online communications and discussing safe use of technology. Practitioners also needed to ensure access for people with language barriers, those without technology, and parents supporting children at home with remote learning, who often lacked adequate technology or resource to keep children occupied. Practitioners scrambled to effectively respond to need during the pandemic, and to provide alternatives to face-to-face services. Almost all (93 per cent) reported that their service had changed their way of engaging with clients; 94 per cent reported their service changed how it worked with other services; and 91 per cent changed where staff do their work. Overwhelmingly, respondents considered the changes made in response to COVID-19 to be positive. Over a third (35 per cent) reported that changes made were 'very effective' and over half (59 per cent) reported they were 'effective'. Largely, this reflected frontline practitioners' intensive, creative efforts to support clients via phone, email, video calls or messaging apps rather than face-to-face. The growing importance of digital communications, and the adjustment to technologymediated client contact, recurred in practitioners' comments. Some described the range of adjustments practitioners had made, for example: Increase in staff time to prepare and reassure clients through daily check-ins to identify any key issues, check how they are going, updating on COVID-19 by not just texting but also virtual calls to explain what it means and wrap-around supports-going virtual-developing emails for all families, ensuring they have data and devices, know how to use, for the children, counseling for the mums, where to get educational resources, providing food packages, winter packages, etc. (Service manager, Sydney). One manager described how the creativity and resilience of her counselling team underpinned their success in quickly adapting to new circumstances: My counsellors were fantastic at adopting new counselling technology like zoom and phone overnight. They had minimal training and supported each other through a difficult time (Service manager, regional NSW). Through the survey, respondents reiterated that shifts to phone and online/remote delivery involved considerable extra work for practitioners, added stress at a time that practitioners were working away from their usual collegial supports, and took a toll on well-being: Making the group sessions work, including therapeutic activities for children and families over video conferencing has been a great deal of extra work and challenging -not ideal (Social worker, regional NSW). Our use of phone and video chat to meet with clients increased dramatically. In almost all cases it was successful, though extremely tiring for staff (Program coordinator, Sydney). Faced with the need to quickly change how they worked, half of respondents (50 per cent) said their overall level of contact with clients (through any means) increased since the outbreak, and only 14 per cent said their contact decreased (levels reportedly stayed the same for the remainder). Face-to-face contact decreased for a majority (80 per cent) and slightly more than this (85 per cent) reported increased contact by phone, whilst a third (34 per cent) reported increases in contact with client via online chat and 40 per cent said they increased contact via video calls. Consistent with the positive views captured in closed survey questions about the effectiveness of changes, some commented on the overall success of shifting to remote modes, with clients for the most part satisfied, for example: The service I work for stopped all face-to-face contact for all employees meaning we all transitioned to utilising phones and computers as the main pathway to communicate with clients. I believe it has been a successful transition whilst COVID 19 has been occurring. I am unsure if it is more impersonal for clients however the majority of clients have seemed happy with the service response even though they are not able to speak face-to-face with workers (Frontline practitioner, Brisbane). Others also described successful shift to remote methods, and in this case, transition back as restrictions lifted: We adapted quickly with remote working, up-learning to stay connected with each other and external bodies via Zoom meetings and teleconferencing; enhancing our policies and procedures for a pandemic; making our workplace safe for workers and clients. As the restrictions have lifted and we are seeing more women face-to-face and workers back in the office where social distancing is able to take place (CEO, Sydney). Several mentioned that clients appreciated their support, even when not face-to-face, or identified expanded access benefiting additional groups needing the service: I have been impressed by my organisation's and other's ability to utilise technology to continue to offer services and meet the needs of the community -we have been able to offer service to clients who would not normally be able to attend our Centre for group sessions (Social worker in a management role, regional NSW). I think for us the most positive thing to come out of COVID is that we are now offering clients a bigger range of options for contact. Previously it was almost all done by appointment in our office. We are now offering phone and video calling as a matter of course, which give the women a greater freedom. This also enables them to seek support in a way that is safest and easiest for them to manage (Program coordinator, women's health setting, regional NSW). A few highlighted how shifting to remote delivery was positive in that it gave them more time to spend directly with clients, and the clients generally responded well and accepted technology-mediated delivery in the context of a pandemic: Clients seem to respond well to being in the comfort of their own home for more effective telephone counselling. Despite video chat opportunities most clients have been content to talk on the phone which seems to give some increased anonymity for disclosing (Social worker, rural NSW). Other comments demonstrated that successful transition was achieved with enormous effort by practitioners. One manager, for example, explained that whilst the numbers of people using the service were unchanged, efforts to deliver the service increased substantially, as the transition to remote delivery created new tasks: What actually changed was the increase from low to high level of effort needed in particular ensuring clients were checked in daily, engaged to get feedback if all we were providing was enough and supporting the communication line-showing them how to use the devices/ and their email addresses. It was like going back to the basics for many, who have had no need for virtual or online methods of communication (Service Manager, Sydney). In contrast, larger numbers of practitioners commented that changes were less than ideal. Whilst phone calls and online support could help provide an immediate response, many practitioners commented that these were not suitable for all clients, with some responding better in person, or disengaging. For example: Some clients are really missing the face-to-face reassurance of speaking with someone physically as opposed to receiving/making phone calls. Many clients are grateful to receive a phone call and know that there is someone else thinking of them. For some clients, phone calls are problematic because it is a safety concern (Domestic and family violence worker, legal service, Sydney). Some clients increased contact via phone calls, and workers became more flexible in time when working at home. Some clients disengaged, particularly if they did not have strong internet connections/access to video conferencing or did not like using it (Social worker, regional NSW). In assessing shifts to remote delivery, many expressed concerns about whether practices adopted were best practice, and whether remote support would prove sustainable. Such comments depicted remote service delivery as a 'band-aid' rather than a longer-term solution. Others found that remote provision helped some groups more than the others, with people with language difficulties, older women or those lacking access to technology facing barriers: Engagement with younger women experiencing domestic violence has been effective using text messages, and has increased contact/support. Older women or women who live in 'black spot' mobile coverage areas have decreased contact. Older women tended to prefer face-to-face contact (Social worker, regional NSW). Particular concerns related to client safety, as it could be difficult for practitioners to ascertain whether perpetrators were present during service delivery. They explained, for example: supporting clients in DV relationships via zoom was somewhat problematic as they were isolating with their partners who were the perpetrator and were often present during zoom sessions. I had some safety concerns for clients to be able to discuss domestic violence in a safe place when partners were present. Clients were also aware of this and partners being present (Social worker, Canberra). Another described: increased difficulty connecting due to the women often not knowing when the partner and the children will be out of the house so she can talk in confidence. A lot of women calling while they go for a walk or driving somewhere (Social worker, Sydney). A male practitioner working directly with men who use violence was particularly critical of remote delivery, raising safety concerns: People who are isolating in the homes with the victims of their violence accessing telephone services feels like it increases risk. We're challenging cognitions that underpin the use of family and domestic violence and their partners are potentially just in the next room while they're escalating. It is not a ubiquitously safe space. It is a sham of service delivery, in modality, in supervision, and in outcomes for the families (Family support worker, Perth). Practice challenges in the context of COVID-19 included difficulties recognising signs of abuse and assessing risk. Practitioners noted it could be difficult to assess clients' circumstances and needs when their contact was limited to phone calls and they could not meet in usual service settings: I am used to engaging with women at court face-to-face. The opportunity to establish some rapport and give the women space and time to discuss their circumstances has diminished now we cannot see women face-to-face. Establishing rapport over the telephone can be very difficult (Manager, regional NSW). Not being able to meet face-to-face and have an opportunity to read someone's expressions/body language has made it difficult. Having open phone conversations regarding issues was not possible with perpetrator in the home (Network co-ordinator, regional NSW). Others noted that whilst much work was previously phone-based, they now felt less confident to identify violence and its impacts, without faceto-face appointments: A lot of our work is usually done on the phone so the rapport building and frequency of contact has not really changed allowing the assessment and identifying of signs of violence to continue. However with the reduction of face-to-face appointments we are less confident in being able to identify isolation tactics as well as any changes in children's behaviours / interactions that are not expressed by the parent (Manager, regional NSW). Many comments about practice challenges focused on difficulties of assessing risk without face-to-face contact. This was noted, for example, by a frontline worker providing specialist support to Aboriginal families, who felt uneasy assessing situations without physically visiting homes or meeting with clients. She explained: Text and calls have been effective. It's the non-verbal cues that are missing from home visits that prevents me from getting the full picture (Aboriginal specialist worker, regional NSW). Similarly, a family support worker who made home visits before the pandemic explained the difficulty of assessing risk without the usual pattern of personal contact: Difficult to recognise signs or assess risks as our work has been contained to home visiting with workers in the front yard, living with their abuser/s so unable to confidently communicate with us in person and many of our clients do not have internet access and irregular access to phone credit. Difficult to recognise signs and assess risks as we are not seeing young people and their families as often, most not at all, as we are not running our drop-in programs (Family support worker, Sydney). A multicultural worker explained that 'text messages, online and phone work best when mixed in with face-to-face contact' and that faceto-face contact facilitated clearer disclosures. She stated: Clients in my experience are less likely to disclose the scope of the violence if you are not seeing them face-to-face. I find when I am working with clients virtually it is more task orientated (Multicultural support worker, regional NSW). This was a source of anxiety among practitioners, who were concerned they would miss non-verbal cues when using remote, telehealth approaches. As such, they worked harder to fill the gaps in their knowledge: I am relying heavily on practice experience to recognize stilted conversations over the phone that is due to POI [Person of Interest] being present to the client. Due to COVID there has been limited opportunity to arrange code words or in-depth safety plans due to the possibility of POI monitoring. Occasionally the conversation is terminated abruptly (Social worker, regional NSW). When we see someone face-to-face we can access their body language. We can build more rapport and they seem to trust us to tell us more about their situation. It becomes more challenging online or by phone, not impossible, [we] just need to work harder (Practitioner, regional NSW). Recognising these challenges, many practitioners said changes were made to their risk assessment instruments. A little over half (52 per cent) reportedly used the same assessment tool as prior to COVID-19, whilst 40 per cent said changes had been made. Changes included adding questions about COVID-19 isolation and controlling behaviours, or recognising that children may be present during screening. Some described difficulties assessing risk whilst perpetrators supervised phone calls or blocked access, whilst others noted large volumes of missing information as practitioners could not properly observe clients and gauge circumstances. Safety planning was similarly difficult to perform remotely, increasing the complexity and strain of domestic violence work. One practitioner underlined the challenges of developing safety plans by telephone: I recognize the complexity of the issues, and I understand that in light of a pandemic, that changes would have to occur and I don't really have the answers on how something as fluid as safety planning could be done over the phone. Video calling, maybe, but text messages or phone calls is so much harder (Family support worker, Perth). Others noted safety planning was more complicated as women were under intense scrutiny at home and violence could escalate quickly. Enacting plans were difficult due to limited opportunities to access safe spaces, with lack of service availability, loss of income and clients feeling unable to go to family members' homes, including because it may put elderly parents at risk of the virus. One commented, for example, on how individual responsibility was placed on victims to manage risks on their own, noting that: Women themselves have had to learn how to be much more savvy around their personal decisions about how to keep safe and practice resistance within this context (Practitioner, legal context, Sydney). This article has contributed Australian domestic violence practitioners' accounts of adapting services during the initial phase of COVID-19. This was a unique period, distinct from disasters studied previously in the changes experienced in patterns of need, barriers to service use and provision. Although some technology-mediated services were delivered prior, these modes rapidly expanded early in the pandemic. Our survey captured the complexities and nuances of practitioners' experiences, to deepen insight into the ways service transformation was implemented and experienced. For most, technology-mediated service delivery constituted a new way of working, adopted out of necessity not choice, and without the reference points of practice guidelines, or an evidence base. Many practitioners encountered challenges in the process, most notably in accessing the comprehensive information needed to assess risk. However, adaptation of models enabled services to be maintained, and accessibility improved for many clients, although some groups continued to face barriers. Of course, the changes described occurred in the context of Australia's relatively successful containment of the virus, but nonetheless are likely to reflect experiences common across countries and social work contexts. Since the data were collected, Australia has experienced additional periods of local and state-wide lockdowns, and remote service models appear likely to endure. Services appear well-positioned to incorporate technology-mediated options into their practice mix for the longer term, but our findings underline the need to evaluate their implementation and impacts. Indeed, the services innovated out of necessity not choice and at an unprecedented, unplanned pace, means new practices in place are largely untested, yet their longer-term sustainability requires that service managers, practitioners and clients make informed, evidence-based decisions about the circumstances appropriate for remote, face-to-face or hybrid modalities. The pace and extent of change underline the importance of ongoing research to assess the effects of remote service delivery for different types of services and for different groups of practitioners and clients. Further research should delve deeper to explore innovations and impacts in particular contexts, to inform practice guidelines and risk assessment tools. Reliance on practitioners' accounts is a limitation of the study, and future research should draw on clients' accounts, to deepen understandings of their experiences of technology-mediated service delivery and any barriers. Clients' perspectives are essential to determining which changes are appropriate to maintain, and the ways remote options can work effectively alongside face-to-face work. Finally, whilst the qualitatively enriched survey generated rich insight of experiences as the pandemic unfolded, this should be considered among the suite of social work research methods, without replacing conventional qualitative approaches that allow for probing, clarification and participant interaction. China's domestic violence epidemic The impact of COVID-19 on gender equality Can't Leave you now! Intimate Partner Violence Under Forced Coexistence and Economic Uncertainty Practising ethically during COVID-19: Social work challenges and responses Adapting Service Delivery during COVID-19 1795 Is physical proximity essential to the psychoanalytic process? An exploration through the lens of Skype The prevalence of domestic violence among women during the COVID-19 pandemic The online survey as a qualitative research tool Lessons from the Transition to Relational Teletherapy During COVID-19 Relationship-based social work and electronic communication technologies: anticipation, adaptation and achievement COVID-19 and telepsychiatry: Early outpatient experiences and implications for the future National Survey of Workers in the Domestic, Family and Sexual Violence Sectors Impact of COVID-19 on Women and Children Experiencing Domestic and Family Violence and Frontline Domestic and Family Violence Services Constructing Survey Data: An Interactional Approach Experience of Hurricane Katrina and Reported Intimate Partner Violence Reframing service innovation: COVID-19 as a catalyst for imposed service innovation Rapid assessment on gender and violence against women and girls in the Ebola outbreak in Beni, DRC Responding to domestic violence in the wake of disasters: Exploring the workers' perceptions of the effects of cyclone yasi on women When stay-at-home orders leave victims unsafe at home: Exploring the risk and consequences of intimate partner violence during the COVID-19 pandemic COVID-19, the shadow pandemic, and access to justice for survivors of domestic violence Covid-19: EU states report 60% rise in emergency calls about domestic violence Systematic review of research into frequent callers to crisis helplines Social isolation, time spent at home, financial stress and domestic violence during the COVID-19 pandemic', Trends & Issues in Crime and Criminal Justice No. 609 Investigating the increase in domestic violence post disaster: An Australian case study The hidden disaster: Domestic violence in the aftermath of natural disaster Responding to the 'Shadow Pandemic': Practitioner Views on the Nature of and Responses to Violence against Women in Victoria, Australia during the COVID-19 Restrictions When Home Becomes the Workplace: Family Violence, Practitioner Wellbeing and Remote Service Delivery during COVID-19 Restrictions We are in a bubble that is set to burst. Why urgent support must be given to domestic violence workers Frequent callers to telephone helplines: New evidence and a new service model Women and the weight of a pandemic: A survey of four Western U.S. states early in the coronavirus outbreak Intimate partner violence and Hurricane Katrina: Predictors and associated mental health outcomes Covid-19 and domestic violence: An indirect path to social and economic crisis Stay home, stay safe, save lives? An analysis of the impact of COVID-19 on the ability of victims of gender-based violence to access justice Adapting Service Delivery during COVID-19 1797 Domestic abuse frontline service COVID-19 survey results Violence against women and girls-Data collection during COVID-19 Family violence and COVID-19: Increased vulnerability and reduced options for support The authors thank the organisations which generously assisted by distributing the survey to domestic violence workers, and the workers who participated. We also thank anonymous reviewers for helpful comments on previous drafts. The article draws on research which was funded by the COVID-19 Rapid Response Research Initiative at the University of New South Wales.