Setting

Participants were recruited from three NHS Trusts in the West Midlands, one acute teaching, one rural district general, and one Primary Care Trust.

Ethical Approval

NRES approval for this project was granted by Nottingham Research Ethics Committee 2 (Ref: 07/H0408/120), and R&D approval was gained from each participating Trust.

Recruitment

Recruitment took place using advertisements via internal e-mail, through managers and snowballing (using key participants to recruit others). All participants approached by managers were made aware that participation was voluntary, and managers would not be told who had eventually participated. All potential participants who contacted the research team, regardless of how they had heard about the study, were sent an information sheet, and those who were willing to participate were entered on a database. Participants for focus groups were purposively selected from this database according to job category with a view to getting as wide a range of HCW roles as possible, including a mix of age, gender and seniority. Signed informed consent was gained from all participants prior to participation.

Design

Ten focus groups were planned with volunteer HCWs (grouped according to homogeneity of role: ×2 ancillary and clerical (A1 & A2), ×2 nurses (N1 & N2), professions allied to medicine (P), junior doctors (below consultant – JD), consultants (C), general practitioners (GP), managers (M) and community based HCWs (CH). Failure to identify a suitable time for a consultant doctor group led to interviews (n = 5) being undertaken with this category. Both focus groups and interviews followed a standardised topic guide, developed over a series of meetings of the authors and of the study steering committee. Open questions were used, enabling the content of the discussions, but not the form, to be participant led. The topic guide started by asking what participants knew about pandemic influenza, followed by questions about training and the effect they anticipated pandemic influenza to have on their work. Participants were asked if they felt they would be able to carry on 'business as usual as far as possible' [2] in their professional life; how likely they were to report to work if fit and well; whether they thought HCWs had a duty to work, why, and how far this duty extended. The discussions ended by asking participants to discuss the things that would worry them about working during an influenza pandemic, and what things they thought their employers, the Government, or professional bodies and unions could do to make it easier for them to work.

Focus groups are a recognised qualitative tool, suitable for research that aims to explore how people think and feel, where the emphasis is on exploring people's opinions, experiences, wishes and concerns [21, 22] The dynamic, interactive, nature of focus groups means that participants are continually interacting with one another, exchanging views and perspectives, and this level of interaction enables the researcher to examine meanings, reasons and motivations "with a degree of complexity that is not typically available with other methods" (p16) [23] and to access the rich texture of normative influences on people's attitudes and behaviours [24]. The aim of a focus group is not to gather data that is statistically significant or generalisable to the wider population, but to gather information that can help us begin to understand why people may choose to act and behave in the way they do. For this reason, the results are not presented numerically, as in this context frequency has little meaning and detracts from the qualitative presentation style.

Data analysis

All focus groups/interviews were transcribed verbatim, and were formally reviewed by three of the authors (SG/JI/JMP). Transcripts were initially free-coded by JI on ATLAS-ti software according to content [25], and then organised into thematic units that were continually re-visited and revised. Analysis and data collection occurred simultaneously, with a constant comparative method [26] utilised to ensure an iterative approach to the interpretive project. Analytic induction [25] was also employed, with themes being identified, hypotheses generated and fed into subsequent groups/interviews and then revised in the light of new data.

Given the nature of qualitative enquiry, the analysis was not conducted numerically. 'Significance' in qualitative research, and particularly in a focus group, cannot be determined by the frequency with which a view or opinion is raised, but rather in the manner in which it is raised, discussed, and negotiated by the group. What is important in a focus group analysis is not how many people stated a particular view, but rather what general themes emerged out of the discussion [27]. Bowling [28] points out that " [q]ualitative research describes in words rather than numbers" (p352), and as such, we have presented our results by describing themes that tended to arise out of the group process, rather than counting the number of individuals who expressed a particular view. This is justified, first, because the number of participants (n = 64) is insufficient to find any statistical significance. Second, our analysis focussed on what was said (the quality of views) rather than how often it was said (the quantity of views). What general frequency indicators are provided are given solely to give the reader an impression of whether the views under discussion were majority or minority views, and we have indicated in which groups the themes under discussion arose.

A different kind of study is needed to determine the prevalence of the views reported, and the qualitative data reported in this article have been used to develop a large scale survey tool, the work for which is ongoing [29]. This article only discusses our qualitative data, and focuses on the variety of views and perspectives present in our research population.

The Duty to Work

Overall, participants seemed to feel a strong sense of duty to work regardless of the circumstances and displayed a general willingness to work during an influenza pandemic. This sense of duty was found across all categories of workers, but was justified in a variety of ways that can be brought under three headings:

A professional ethic (prominent in 4/9 focus groups and 5/5 interviews)

Some participants felt the duty to work is a professional duty that entails an obligation to work even in difficult and dangerous circumstances, because that is what they signed up for when they joined their profession. This kind of view was expressed most forcefully by hospital doctors and GPs, some of whom felt that this duty, which was developed during their training, was one owed both to their profession and to their patients. There were mixed views among participants about how far this obligation extended across different healthcare roles. Some doctors, for example, felt that the professional obligation did not apply so much to ancillary staff or nurses, though others felt that it applied to everyone, (see below).

A duty to help (prominent in 8/9 focus groups and 2/5 interviews)

Some participants across a range of jobs spoke about an obligation arising from a general 'duty to help'. Some, for example, claimed that being in a position to help another person was a sufficient motivator for helping them. Others expressed the view that if a person or a society is in need, that need gives everyone a moral incentive to take steps to meet that need. Notions of a duty to help were sometimes accompanied by references to the "Blitz/Dunkirk' spirit" of wartime Britain in the 1940s, and the belief that in the event of a social crisis we all have to pull together and do what we can for the common good. Here, the duty to help was constructed both as an individual and a social requirement to contribute to the common good, applying to us as persons and not only as HCWs. One key source of the duty to help for some HCWs was that they had specific skills that would make them particularly useful, and this seemed to make them feel they had a special responsibility where others might not, see below.

A simple work ethic and confederate loyalty (prominent in 6/9 focus groups and 3/5 interviews)

Many participants, in a variety of roles, felt that absenteeism if one was able to work (whatever one's employment, in or out of the health service) was generally wrong and not just wrong during a pandemic.

This simple work ethic (see below) may be related to or reinforced by, a complex sense of loyalty and obligation to workplace confederates, in this case colleagues and patients. This account tended to emerge from a belief that, as both colleagues and patients depend upon you doing your job, by refusing to work (when you are physically able) you are letting both your patients and your colleagues down in way that is morally unacceptable, for example

Not wanting to let patients or colleagues down

P4: I feel like that really I would be letting those patients down if I didn't come in because they're my raison d'être, that's why I'm here.

P1: It's not just your patients. It's your colleagues as well.

P5: Yeah.

P1: You're letting your team down as well as the patients.

P4: That's right.

***************************************************

N2/4: I think that's the other thing. It's camaraderie. You get like a good little team on a ward and half the time if you do feel a little bit sort of a little bit sick you think 'oh well I've got to go in because they'll struggle without me'.

Barriers to working

Although all participants tended to feel they have a duty to work during an influenza pandemic, there were nonetheless a number of perceived barriers to so doing. These tended to fall into one of two categories:

• perceived barriers to ability (prominent in 9/9 focus groups and 5/5 interviews)

• perceived barriers to willingness (prominent in 9/9 focus groups and 5/5 interviews)

These categories, however, were not clear cut in all cases, and the lines were most blurred when it came to concerns about childcare and family obligation. Some participants who had children regarded staying at home to look after them as a necessity that affected their ability to work, rather than a choice (see below). Childcare is not, however, obviously distinguishable from a barrier to willingness; for instance, where parents choose to look after their children themselves rather than rely on available others to do so for them. Choice, of course, is itself a nebulous concept. Where there is no externally available or accessible childcare, and where children are too young to care for themselves, there is a barrier to ability. However, what counts as 'accessible' or 'too young' and whether the available childcare is regarded as an adequate or acceptable alternative may owe as much to personal choice or preference as to inescapable circumstance.

It seems likely that the 'childcare' barrier is age and gender related. Women with young children tended to regard it as an insurmountable obstacle, with men raising the issue less often. Whether this gender difference was due to the participants' need to present themselves to their peers as participating in typical gender roles is unclear, although the discrepancy observed is predictable, and consistent with what we might expect. The group in which it was least prominent was the GP group, which was comprised largely of older men. The discussion was most prominent in the nursing groups, where women with young children were in the majority. The apparent lack of concern about this issue in the GP group may reflect gendered norms in the home, the age of the participants which meant that they were unlikely to have young children, or that GPs were better placed to afford reliable, private childcare. Similarly, the junior doctors did not discuss this issue: it was raised in passing only once. When they were questioned about why it had not been discussed, the answer was simply that they did not have children so it was not an issue.

For some, the duty to family was expressed as the simple claim that 'family comes first' which was taken for granted as an unassailable moral premise. If a child or a family member needed them they would not come into work pandemic or no pandemic: duties to families were more important than any duty to work. This was not so much a matter of weighing up competing obligations (to work and to family) but rather represented a pre-defined moral hierarchy acting as a barrier to ability (see below). For others, childcare functioned as a barrier to willingness, where a choice puts family before work.

Some barriers to ability were fairly concrete like being ill oneself and problems with transport (including lack of fuel). Participants recognised that they could not work if incapacitated by illness, and also anticipated that in the event of an influenza pandemic transport infrastructures might be affected, making it difficult to travel to work. Some participants thought that during an influenza pandemic people might be reluctant to use public transport for fear of becoming infected, leading to more people travelling to work in private cars. These same participants anticipated that Trusts would not have the parking space to accommodate additional demand, creating a further barrier for people who would otherwise be willing and able to work (see below). Insurmountable barriers to ability, however, exist at the extreme end of a continuum; for example where a person is literally too ill to get out of bed or function safely, where the person is infectious and poses a demonstrable risk to others, or where there is no fuel at all and the distance to work is too great to cover on foot or cycle. Further along this continuum, HCWs will be exercising choice about how to prioritise the different demands on their time and resources, and the greater the scope for preference or choice to be exercised, the more like a barrier to willingness the perceived obstacle becomes.

Child-care as a barrier to ability

N2/1: Well I've got two small children, small children are probably more likely to get these things so, the ones at school, if the schools are closed there's no way I'd be able to come in because you can't [Laughs] You can't just leave your kids. They're obviously a priority.

****************************************************

A1/5: you'd have other outside influences wouldn't you? If you had children, if the schools were closed down and you'd got no-one to look after your children you... what, what options would you have then? You know your first duty is to their care so you would have to think about them before you could come in. If you'd got nobody to look after your children or equally your elderly parents if all that starts.

****************************************************

P2: I was gonna say where, in my case there's one adult to look after a number of children at home and I'm their sole means of support. And so to me they need that support every day you know whether the hospital needs mine or not, I'm not. You know that's the kind of a theoretical benefit to the hospital. You know 'cause I work in preventative health. So it's easy for me to say my kids need me every day, and it's gonna be a different decision for everybody isn't it?

Negotiating risk and duty (prominent in 9/9 focus groups and 4/5 interviews)

Other individuals saw the duty to their family as one of many competing claims. Concern about taking the virus home and infecting one's family, for example, was not perceived as a barrier to ability when participants believed that they were in a position to negate or mitigate the risk by employing infection control measures, or minimising direct contact with family members by staying at the workplace or sleeping in the spare room. Some groups, however, particularly consultants and managers, felt that anyone who thought that absenteeism would reduce the risk to their family had failed to appreciate that 'pandemic' meant that the virus was endemic in the community (see below).

The risk to self (prominent in 9/9 focus groups and 2/5 interviews)

Despite disagreement on what the risk was and how it could be managed, the risk to family members of working was important to everyone whereas the risk to self seemed to be of less concern. Even when one GP broke the consensus that seemed to be emerging from her group by suggesting that if the risk was great she would not work, she made only passing reference to her own safety and quickly justified her position with reference to the safety of her family (see below). However, participants' apparent ambivalence to personal risk was to some extent belied by regular discussion of personal protective equipment (PPE). Whilst participants were reluctant to say outright that personal risk concerned them, many stressed the importance of being provided with effective PPE (for example see below).

Reciprocity (prominent in 8/9 focus groups and 1/5 interviews)

The belief that the relationship between HCW and employer is not reciprocal was one of the most significant barriers to willingness. Specifically, participants did not believe that the efforts of HCWs would be reciprocated or rewarded. This was expressed in a variety of different ways, including an expectation that HCWs would get no thanks or recognition for their efforts, the worry that any PPE provided would be the 'cheaper alternative' (see above) and concern that workers would receive little guidance or decision-making support (e.g. with respect to how resources should be allocated, how treatment should be allocated, or whether decisions made would receive the backing of the Trust).

The majority of participants said they had been given neither information about pandemic influenza, nor been made aware of what would be expected of them during such a crisis, and this gave many the impression that their employing Trust did not care about them or take their needs seriously. Lack of information was a key theme across all groups, with the majority finding the lack of information and engagement a demotivator to work, while clear information, guidance and support seemed to be important motivators (see below). The obvious exception was the management group, which included public health doctors, who were concerned about giving staff too much information, as they did not know if current information was accurate. They reasoned that as it was so difficult to get information through to the workforce, it was a waste of resources to attempt to do so, and possibly counter-productive, if the information given turned out to be inaccurate. Their preference was to disseminate information if, when and as it was needed (for example below). This was in direct contrast to the views expressed by the majority of other groups, who wanted information immediately and to be involved in the planning effort. The GPs were most similar to the managers; most seemed confident that they could cope with the pandemic if and when it occurred, and believed that the necessary information and guidance would be sent to them as and when appropriate, and that their role was one of implementation.