Unheard voices, unmapped terrain: care work in long-term residential care for older people in Canada and Sweden

Abstract

This article aims to contribute to comparative welfare state research by analysing the everyday work life of long-term care facility workers in Canada and Sweden. The study’s empirical base was a survey of fixed and open-ended questions; this article presents results from a subset of respondents (Care Aides and Assistant Nurses) working in facilities in three Canadian provinces (n=557) and across Sweden (n=292). The workers’ experiences were linked to the broader economic and organisational contexts of residential care in the two jurisdictions. We found a high degree of country-specific differentiation of work organisation: Canada follows a model of highly differentiated task–oriented work, whereas Sweden represents an integrated relational care work model. Reflecting differences in the vertical division of labour, the Canadian Care Aides had more demanding working conditions than their Swedish colleagues. The consequences of these models for care workers, for elderly people and for their families are discussed.

Introduction

The ‘care work’ literature focussed at the service delivery level invokes expressions like ‘labour of love’ and ‘caring for and caring about’, which highlight that it is work, it involves a relationship, and it is physical and often arduous hands-on work (e.g. Graham, 1983; Ungerson, 1983). Care work involves what Twigg (2000) referred to as body work, social care for the mind and spirit and care for the space, including its cleaning and maintenance. In both its paid and unpaid forms, care work has been characterised as a combination of intellectual, emotional and manual labour (James, 1992) or in Leira’s (1994) words – loving, thinking and doing. For overviews and theoretical considerations on care work, see for example, England (2005), Fine (2007) and Leira & Saraceno (2002).

Like other areas of paid care work, such as home care, work in residential long-term care homes is often viewed through the lens of tasks, lacking an important connection to the affective and relational dimensions of the work. The interdisciplinary care work literature addressed some of the concrete practices constituting the care relationship – what Dyck, Kontos, Angus and McKeever (2005) referred to as the ‘micro-politics of care negotiation’ – by shedding important light on care as highly gendered and racialised work. In this stream of research, however, country-level or welfare regime comparative work is the exception.

In most mainstream comparative welfare-state research the paid care workers who actually carry out the care services are hardly visible. There is no articulation of care workers’ working conditions, and the encounter between the care worker and the recipient is rarely linked to the broader social, political and economic context. Therefore, while paid care work in most countries is under obvious pressure for transformation due to shifting steering and economic principles, the content and the strains of the daily work have been neglected. Thus, very little is known about whether there are national (or welfare regime specific) differences in the employment conditions and the workdays of care workers. This knowledge gap is problematic because it tends to render invisible a large, women-dominated sector of the workforce. It also means that comparative work in the area of long-term residential care is relatively unmapped terrain.

Aim of the article

Using the voices of workers, we focused on the everyday realities of the care work set in the relations between care workers and residents in the context of long-term residential care for older people in Canada and Sweden. The article describes some of the similarities and differences in the conditions of care work in these respective jurisdictions and analyses how work organisation affects workers. Thus, the article aims to broaden the agenda of comparative welfare state research by contributing care work research that links concrete practices within a gendered political economy of health perspective (Armstrong, Armstrong & Coburn, 2001). It also aims to contribute to debates about how care and dependency are situated in the welfare state. We argue that an on-the ground analysis of the relations between long-term care workers and residents reveals how each society cares for its most vulnerable and is thus symbolic of how care is valued by each society. Good working conditions – care for the worker – is the precursor to good care for residents (Armstrong & Daly, 2004), so an analysis of relations between care workers and residents – the micro-politics of care – is more broadly illustrative of how welfare states are addressing issues related to ageing and the need for care.

Method

Two research teams worked in tandem on this comparative study of long-term care in Scandinavia (Sweden, Denmark, Finland and Norway) and Canada (Ontario, Nova Scotia and Manitoba). (For a closer description of the study, see the Appendix.) A largely identical survey was distributed in the respective jurisdictions to unionised workers in long-term care in 2005 (Scandinavia) and in 2006 (Canada). The survey instrument included open and closed questions exploring employment qualifications, content of work, working conditions, health and safety, work and family life issues, and the health acuity of residents. The open-ended questions allowed workers the opportunity to comment on different aspects of work using their own words.¹ A thematic content analysis of these responses is reported together with an analysis of the workers’ workdays and experienced workloads.

This article is based on a subset of the survey respondents: 557 unionised care workers in residential care for older people in Canada (142 Assistant Nurses, i.e. licensed or registered practical nurses, and 415 Care Aides, i.e. personal support workers, health care aides or nurse’s aides); and of 292 unionised care workers in residential care for older people in Sweden (203 Assistant Nurses [in Swedish, undersköterska] and 89 Care Aides [vårdbiträde]). In Canada, the group we label assistant nurses are licensed or registered. When we choose not to use a prefix indicating regulation, it is simply an adaptation to the Swedish case where an ‘undersköterska’ is not more regulated than a ‘vårdbiträde’.

Long-term care is primarily a unionised work environment in the studied jurisdictions, with about eight in ten workers belonging to a union. However, it is important to stress that the study reflects responses of unionised workers who typically have more job security and longer work experience than their non-unionised counterparts. Further, as the Canadian data are for three provinces only, they may not necessarily reflect the situation in other provinces.

Long-term facility care in context: resources, facilities, residents and workers

Both Canada and Sweden are rich countries where a larger proportion of the Gross Domestic Product (GDP) was spent on residential eldercare than the average for the Organisation for Economic Co-operation and Development (OECD) countries: 2.07 per cent in Sweden and 1.06 per cent in Canada compared with a 0.88 per cent average in the OECD countries (OECD, 2005: 26).² However, Sweden has one of the world’s oldest populations with 5.0 per cent who are 80 years and older compared with 3.0 per cent in Canada and 3.1 per cent for the OECD average (OECD, 2005:101). In relation to the proportion of older persons in the population Sweden spent 46 percent more than the OECD average, while Canada spent 24 percent more.

In Canada, almost half of the residents live in commercially-owned facilities, about one-quarter in charitable facilities and another quarter in public facilities (Statistics Canada, 2007). In Sweden, the majority of eldercare services are publicly delivered, but an increasing proportion (17% in 2009) of the residents live in privately-run facilities (Swedish National Board of Health and Welfare [NBHW], 2010). At the time of the study, the average residential care facility in Sweden housed 34 residents; nearly all the residents had a private room or small apartment, usually with a private bathroom (>95 %) and with private cooking facilities (75 %) (NBHW, 2008).

An average residential care facility in Canada housed 96 residents and 70 per cent of the residents lived in facilities with more than 100 beds (Statistics Canada, 2007). The size and design of the Canadian facilities make them more hospital-like than home-like (Armstrong & Bannerjee, 2009; Gnaedinger, 2003). The resident populations in the two countries were similar: most of the residents had multiple health conditions, including dementia or other cognitive impairments, and most were women.

The workforce in both jurisdictions is highly gendered, with women comprising more than 90 per cent of the direct-care staff. The Canadian survey sample consisted of 15 per cent Registered Nurses (RN), 15 per cent Assistant Nurses (AN), 44 per cent Care Aides (CA), 18 per cent support workers (including dietary, housekeeping, laundry and maintenance) and 8 per cent other occupational groups. The composition of the Swedish residential care workforce was quite different. According to a large survey from eight Swedish municipalities from 2003 (Gustafsson & Szebehely, 2005), slightly less than 6 per cent of the residential care workforce were RN, 54 per cent were AN, 33 per cent were CA, 3 per cent managers or physio- or occupational therapists and 3 per cent were support workers (mainly janitors, housekeeping and kitchen staff).³

Although there was a larger proportion of RNs in the Canadian than in the Swedish care workforce, which makes Sweden’s sector appear less professionalised, Table 1 depicts a higher training level among Swedish care workers.⁴ Just over three-quarters of the Swedish care workers have had at least one year of formal training compared with 21 per cent of the Canadian care workers. This is a partial reflection of the fact that the majority of the Canadian care workers in the study were care aides (CA), while the majority of the Swedish care workers were assistant nurses (AN). But a larger proportion of Swedish AN (91%) than Canadian AN (40%) had at least one year of formal training, and also more Swedish than Canadian CA had one year training or more (41% compared with 15%). On the other hand, clearly more Swedish than Canadian CA had little or no formal training (43% compared with 9%).

Table 1.

Residential care workers in Canada and Sweden by years of training

	All care workers		Canada		Sweden
	Canada (n=557)	Sweden (n= 292)	Assistant nurses (n=142)	Care aides (n=415)	Assistant nurses (n=203)	Care aides (n=89)
None or less than 6 months	8.1	13.1	4.4	9.4	0.5	42.5
½–1 year*	70.7	11.1	55.6	76.0	8.9	16.1
1 year+	21.2	75.8	40.0	14.6	90.6	41.4

Note: Figures marked in italics indicate a statistically significant difference (p<0.05).

^*In the Canadian case 1/2–1 year training corresponds to a ‘health related college certificate completed in one year or less’ while 1 year+ training corresponds to a ‘health related diploma from college completed in more than 1 year’ or (in a few cases) a ‘health related university degree’.

Most care workers in our survey had significant job experience: almost two-thirds of the workers in Canada and slightly more of them in Sweden have worked in this capacity for a decade or more. Only a few care workers had neither experience nor formal training, suggesting that many had acquired qualifications and become skilled at their jobs through experience rather than through formal education (cf. Armstrong & Daly, 2004).

A large proportion of the direct care workers we surveyed in Sweden (65%) and Canada (41%) worked part-time. In both countries, part-time work was much more common among care workers than it is among the female workforce in general (Statistics Canada, 2004; Statistics Sweden, 2008). This casualisation is to some extent involuntary, since 22 per cent of the Swedish and 45 per cent of Canadian part-time workers worked shorter hours than they preferred. Another aspect of involuntary part-time was reflected in the proportion of the workforce holding more than one job – 7 per cent of the care workers in Sweden compared with 17 per cent in Canada – suggesting that there was both a need and a desire for full-time jobs, particularly in Canada.

Comparing long-term care workdays

Professional mix and division of care work

There were two important country-specific differences in the organisation of the care work based on inter-country variation in the professional/occupational mix and in the division of care work between the different types of workers. Professional/occupational mix refers to the relative number of each type of worker in the care workforce. The division of care work refers to the allocation of tasks that constitute care work, and the ways those tasks are divided between groups in the care work force on the basis of skill, education or job classification.

Compared with Canada, in Sweden a much larger share of the care workforce were direct care workers (CA and AN), and there were fewer RN as well as fewer support workers such as dietary and laundry workers. Further, among the care workers Sweden had more AN and fewer CA. These differences reflected the different ways care work was organised in terms of the tasks performed by the different types of workers.

With the exception of body work – including help with dressing, bathing, toileting or changing incontinence products – and serving food, which were ordinary parts of most surveyed care workers’ jobs irrespective of their country or occupational category, we noted some important differences depending on country and professional/occupational mix. In some task areas, Swedish and Canadian care workers had different workdays, see Table 2.

Table 2.

Care workers’ ordinary work tasks in Canada and Sweden.

	All Care workers		Canada		Sweden
% reporting doing the following ordinary work tasks at least monthly	Canada (n=557)	Sweden (n= 292)	Assistant nurses (n=142)	Care aides (n=415)	Assistant nurses (n=203)	Care aides (n=89)
Clean resident’s room/apartment	53.4	82.2	39.5	58.0	80.5	86.4
Serve food	86.4	86.6	73.0	90.6	86.4	87.2
Help with personal hygiene	95.3	100.0	81.4	99.8	100.0	100.0
Sit down for a coffee and a chat with a resident	46.6	64.4	46.5	46.7	65.3	62.4
Follow a resident on an errand outside facility	7.3	36.3	2.4	8.9	36.1	36.8
Administrative tasks	47.7	82.5	86.2	34.5	90.4	63.4
Contact with other health care providers	18.5	41.5	60.2	4.3	42.1	40.2
Hand out pills from a dispenser	23.3	91.3	86.5	3.1	91.0	92.0
Give an injection	19.2	38.8	74.2	1.3	43.6	27.6
% reporting that 5> tasks listed above are part of their ordinary work	26.8	81.7	67.9	13.5	84.2	76.1

Note: Figures marked in italics indicate a statistically significant difference (p<0.05).

For instance, Swedish workers more often reported being engaged in relational aspects of work, measured by how often they had a cup of coffee and a chat with a resident or how often they accompanied a resident on an errand outside the facility. Cleaning was also more commonly part of the workday among Swedish care workers than among their Canadian counterparts.⁵ With respect to cleaning residents’ rooms/apartments, more than 80 per cent of Swedish care workers (CA as well as AN) reported cleaning rooms, while around 40 per cent of Canadian AN and 58 per cent of Canadian CA did so, reflecting how in many Canadian facilities housekeepers were hired to clean. One-quarter of the Swedish care workers also cooked meals for the residents (not reported in Table 2; the question was not asked in the Canadian survey because meals are always communally prepared). Thus, Swedish care workers (both AN and CA) carried out a variety of tasks including cleaning, cooking, laundry, but also recreation and other ‘social activities’, which in the Canadian context were to a large extent carried out by other occupational groups such as dietary and activation workers and professional therapists.

Unlike in Canada where the division of labour between AN and CA is more pronounced, owing in part to governmental regulation of certain care acts by professional groups such as nurses, Swedish care workers collectively engaged in most of the care acts independent of their occupational position or educational background. With one exception (having a cup of coffee with a resident), there was a substantial difference between Canadian AN and CA workdays: what is considered professional work (e.g. medical and administrative tasks) were far more common among AN, while household tasks (e.g. cleaning and serving food) and body work (e.g. help with personal hygiene) were more commonly CA duties. In Sweden, the two groups of workers had workdays that were more similar than the Canadians, with the exception that, compared with Swedish CA, Swedish AN more commonly, though not exclusively, administered and gave (insulin) injections.⁶

The Swedish long-term facility care work appeared less task- and job-specific and more relational when compared with that done in Canada, where the division of labour between the two types of care workers was more ‘boundaried’, regulated and formalised. As a result, Canadian care aides experienced the least varied, least relational and most task-oriented work of our surveyed workers. While between 68 and 84 per cent of the Canadian and Swedish AN and the Swedish CA reported doing at least five of the nine tasks listed in Table 3 as part of their workday, only 14 per cent of the Canadian CA reported having such a varied workday.

Table 3.

Workload and experiences of work, residential care workers in Canada and Sweden

	Care workers		Canada		Sweden
	Canada	Sweden	Assistant nurses	Care aides	Assistant nurses	Care aides
Number of residents helped per workday during the week (average)	19.9	8.8	32.7	15.2	8.8	8.8
Carry, lift or pull heavy loads or persons (%)i	77.4	66.3	59.2	83.3	69.5	59.1
Too much to do (%) ii	57.8	40.9	50.4	60.3	42.9	36.4
Work short staffed due to sickness, vacation or vacancy (%) i	48.3	12.2	55.2	46.0	12.0	12.8
Experience physical violence from a resident or resident’s family (%) i	38.2	5.8	24.6	43.0	5.9	5.7
Can affect the planning of each day’s work (%)ii	24.1	48.1	24.8	23.8	48.3	47.7
Feel physically tired after a working day (%)iii	60.4	27.8	53.2	62.9	28.6	26.1
Feel mentally exhausted after a working day (%)iii	43.0	14.5	41.4	43.5	15.8	11.4

Notes: Figures marked in italics indicate a statistically significant difference (p<0.05).

ⁱMore or less every day. Other response alternatives: Every week, Every month, Less often, Never.

ⁱⁱ All or most of the time. Other response alternatives: Sometimes, Rarely, Never.

ⁱⁱⁱ Almost always. Other response alternatives: Often, Sometimes, Rarely, Never.

However, there were also contradictions inherent in the Canadian workdays, as evinced by two simultaneous but opposing trends. On the one hand, work in Canadian long-term care is very task-specific. Workers reported having a ‘bath girl’ who came in to help get caught up on the mandated baths (e.g. twice weekly in Ontario), or the ‘diaper girl’ who was responsible for doling out supplies including incontinence products. In this way the work was more Taylorist and assembly line, reflected in the comment from this worker: ‘I work a 4-hour bath shift to bath, dress, trim nails etc for 7 residents a day plus other duties. It is so fast that they are getting a ‘car wash’ to fit them all in.’ (CA) Residents were processed through their bath, one after the other, or allowed a pre-allotted number of diapers, after which care aides negotiated with managers for more by making a strong enough case on behalf of the resident such as arguing that the resident was sick or had an accident (Daly, Banerjee at al. 2011).

At the same time, there is evidence that the CA role was expanding with the downloading of tasks that these workers associated with other jobs. One-third of Canadian care workers’ responses to the open-ended question ‘Are there tasks that you have to do that you think you should not have to do?’ reflected a concern over the division of labour between direct care workers (i.e. AN, CA) and support workers (e.g. dietary and cleaning staff). This expansion of the scope of Canadian care aide’s role was troubling established divisions of labour, as reflected in the following statement:

(…) putting away residents’ personal clothing was once a laundry task. We also put away linens – laundry once did this too. Dietary used to clear tables and give out juices at snack time. Housekeeping used to clean, now they only pretend to clean – we have to clean faecal matter and blood – their job description doesn’t permit cleaning of bodily fluids (CA)

Many Canadian respondents indicated how they wanted to maintain very strict care work boundaries: ‘Making and changing beds. Should be done by housekeeping, not a nursing task’ (CA); ‘Putting clothes away. I think laundry should do this job’ (CA); ‘Clean fridges, wash chairs in dining area, should be kitchen’s and housecleaning’s job, not health care aides’ (CA); ‘Serving – dietary should do meals and we should only help feed.’ (AN). A few mentioned that others should do more qualified nursing tasks: ‘Application of treatment creams – RN or RPN [registered practical nurse] should do this in their job.’ (CA).

Unlike Canadians, few Swedish care workers wrote about division of labour issues. However, they indicated a preference not to do cleaning much more frequently than did their Canadian colleagues; it was mentioned by more than half of Swedish AN and CA who answered the question and by only one of ten Canadian care workers. This difference may reflect the fact that cleaning of common areas and residents’ rooms is part of the direct care workers’ daily work in Sweden, but often a task for housekeeping staff in Canada.

Among the Canadian care workers, one-quarter identified serving food as a task they did not want to do; this was not mentioned by any of the Swedish care workers. Frequently, the Canadian care workers used a hygiene argument to protest against serving food: ‘Looking after residents’ personal hygiene needs and then serving food isn’t hygienic!’ (AN); ‘Handling food. Not very sanitary after handling urine and faeces. Then you have to serve meals.’ (CA). That Swedish care workers did not mention an aversion to food serving for hygienic reasons suggests that they did not regard the combination of ‘dirty’ and ‘clean’ care work to be a problem in the same way as did many of their Canadian colleagues. Furthermore, in the Swedish context, there were fewer division of labour boundaries in that there was not a specific occupational group tasked with taking sole responsibility for either the personal care or the housekeeping work, and there were also many homes where residents were expected to participate as much as possible in the meal preparation and other household activities as part of the official Swedish residential care ideology, especially in smaller units that catered to elderly people with dementia (Malmberg & Zarit, 1993). When compared with Canada, Sweden’s smaller sized facilities, a less boundary laden division of labour, and an ideological and policy approach that privileges social care more than health care (Szebehely, 2009c), likely impact not only the workday of the care workers, but also the way the workers perceive their work tasks. When we compared the mix of workers, the tasks performed by the different groups of workers and the workers’ comments on the division of labour, our data indicated that Canada uses a model of highly differentiated work while Sweden represents an integrated work model.

Workloads

Workload involves both the intensity of the work and how much time workers have to carry out their duties. Staffing intensity is considered to be one of the most important factors affecting working conditions and quality of care (for a systematic review see Murphy, 2006; see also Eaton, 2000; Harrington et al., 2000). In Sweden, the staff to resident ratio in 2008 was almost one (0.98) FTE staff per resident (NBHW 2008). Using hours per resident in a 24-hour period, Statistics Canada (2007) reported a Canadian average of 4.7 total staff hours (including support workers) per resident for each 24-hour period.⁷ Due to the different ways of reporting national level staffing data, we must be cautious in our comparisons; however, Swedish residential care seems to have considerably higher staffing levels. These results are aligned with OECD spending levels reported above. Importantly, while aggregate data used to calculate the intensity ratio point to inter-country variation, these data miss the more nuanced ways in which work is organised and performed that may also impact intensity.

We found that there are substantial differences when comparing Canadian and Swedish workloads (Table 3). For instance, our survey data showed that Canadian care workers were, on average, responsible for more than twice as many residents per worker compared with Swedish care workers, raising questions about the amount of time available to care, and drawing attention to the significantly higher levels of work stress among the Canadian care workers shown in Table 3. Furthermore, our survey showed that there was a higher average number of residents cared for by each Canadian AN (33 residents) compared with each Canadian CA (15 residents), a finding which corresponded those regarding the division of labour, since AN were responsible for handing out medicines to residents in their unit (ward, floor or similar), whereas CA were responsible for more of the hands-on feeding and body work and would thus care for fewer residents overall but must do so in a way that was more time-consuming and intensive. Thus, the higher average numbers of residents cared for per day in Canada reflected the division of labour but also the larger size of the facilities and what appeared to be much lower staffing levels.

Having a too heavy workload and lacking time to properly care or being time-squeezed was reported in many of the open-ended comments by workers in both countries. Workload was mentioned by six out of ten Canadian care workers when we asked about reasons they might consider quitting their job: ‘So much work to do in so little time. Too many residents to look after, I feel I am doing a lousy job …’ (CA); ‘When my shift is through, I go home and unfortunately I have nothing left to give to my own family, I’m too tired, stressed!’ (CA). In responding to the same question, Swedish workers also indicated that their workload was heavy. Almost half responded with comments indicating that they worried about negative impacts on their own health: ‘Don’t think I will manage the workload with heavy lifting, dressing etc. Have been on long-term sick leave’ (AN). ‘I will not be able to manage the speed and workload, neither physically nor mentally’ (CA).

Many care workers in both countries also stressed the importance of relational care work and articulated wanting more time to talk and listen to the residents: ‘Sitting and chatting to our residents, listening to their concerns and fears and also to their stories, history of their life’ (CA, Canada); ‘Time to stay close to those who are anxious and unhappy. Let the residents take the time they need to feel good and to do their things in peace and quiet’ (AN, Sweden). Care workers valued social relations with residents and perceived that they lacked the time to care properly. But despite these similarities, the differences between the countries were striking in that more than half of the Canadian CA expressed a wish for more time to do basic personal care. The responses reflected on the heavy workload and severe time constraints they faced, but also on the routinised and task-based organisation of their work. As one Canadian CA noted: ‘getting residents ready for the day – bathing, feeding all. There is not enough time in the day. 45 mins. to get 12 residents for breakfast!!! How do you think that works?’ Another noted how she wanted to be able to: ‘…care for residents (feed) so they can eat hot meals. Toilet [them] every 2 hours not when [I am] able, have social activities with residents with time allowing, allow resident to have more than 10–15 min. baths, from start to finish’ (CA). Many argued that they felt bad about rushing the residents and wanted time to talk and listen while doing personal care – ‘to be more social and not rushed while caring for residents. Don’t like the feeling of assembly line care’ (CA) – while others stressed their own workload, and many mentioned both. The reference to assembly line care in the Canadian open-ended responses mirrored the survey results which showed Canadian CA’s responsibility for a greater number of residents and their less varied workday when compared with their peers in Sweden. In contrast, only a few of the Swedish care workers mentioned that they wanted more time for personal care, and those who did often mentioned wanting time for grooming rather than ‘basic body work’: ‘… put rollers in their hair. Fix their nails. A lot of small things that you never have time for’ (AN).

A majority of more than 60 per cent of the Swedish care workers stressed that they wanted more time to take the residents out from the facility for a short walk or to engage in recreation or rehabilitation, or more time to talk and to listen: ‘…go for a walk with the residents. It should be natural to get out in the fresh air once a day for everyone’ (CA); ‘To just be able to sit down and talk in peace and quiet. Have the time to go for a walk. To simply read a newspaper or play games, or just sing together’ (AN); ‘… to have time to go out for a walk every day with those who want to, bake a cake for the afternoon coffee with the residents, and cook the meals from scratch…’ (CA). In only a few instances did Canadian care workers mention wanting more time to ‘go out’, and a similar minority mentioned wanting more time to participate in residents’ recreation. The differences in the responses seemed to reflect inter-country workload variation and the fact that Canadian care workers frequently stressed that they were unable to meet even the minimum essential care needs. But the different response patterns may also reflect the different division of labour and the more health care-oriented focus in the Canadian setting where social activities like going for a walk are usually not a stipulated part of the CA or AN job, and the lack of comments about it reflected how workers may not think that their work organisation or model of care could be different.

Discussion and conclusions

There are similarities in residential care in Canada and Sweden: the workforce is almost entirely female and the majority of the residents are quite aged women who have substantial care needs. However, the resources, the composition of the workforce and the size of the facilities differ quite markedly. Sweden spent more money per capita and had a higher staff-to-resident ratio. Swedish care happened in smaller facilities where almost all of the residents have private rooms, while in Canada facilities are bigger and accommodation is frequently shared.

While the gendered context of the work is similar, we found that workers in each jurisdiction had different workdays in terms of the breadth of tasks that constituted care work, differing professional divisions of labour, and significant inter-country differences in workload, not only in terms of staffing intensity but also in terms of what is done and not done. Firstly, Canadian care work is more task-oriented than relational, resulting from work that does not account for social care needs, the high degree of task specialisation, which creates ‘assembly line’ care, and having too many residents and too little time to care for them. In contrast, Swedish care work is characterised by a more holistic social care model where each care worker is responsible for most aspects of care for a smaller number of residents.

Secondly, there were inter-country differences in the division of care work between AN and CA, which is reflected in the similarities between the Swedish CA and AN responses regarding their experience of their work contrasted with the Canadian CA who reported more demanding working conditions than the AN. Both groups of Canadian care workers, but in particular CA, were worse off than their Swedish colleagues. Not being able to provide care according to one’s own standards of ‘decent care’ has been strongly related to mental exhaustion (Gustafsson & Szebehely 2005) and the inability to provide quality care has been shown to be a primary reason residential care workers leave their jobs (Bowers et al., 2003). The more equal sharing of responsibilities among the Swedish care workers may moderate some of the negative consequences of a heavy workload, which is heavy both in the sense that there is a lot of lifting and because there is a great deal to accomplish in a day (or night).

In many countries, care workers have been exposed to fatigue, pain and work related injuries more frequently than most other occupational groups (Armstrong & Daly, 2004; Chappell & Novak, 1992; Ross, Carswell & Dalziel, 2002). Yet, the difference between Canada and Sweden is significant, even though care workers in both countries carry out physically heavy tasks every day. The Canadian care workers seemed to have much more demanding working conditions than their Swedish colleagues: each worker cares for many more residents each day, they work short staffed more frequently, they more often reported having too much to do, they far more frequently experience violence from residents, and they have less control over their workdays.

There is clear evidence that a combination of high demands and low control in one’s daily work is a particularly unhealthy combination (Karasek & Theorell, 1990), and the Canadian care workers are worse off in this respect. Not surprisingly, then, the Canadian care workers more frequently reported feeing physically and mentally exhausted after completing their workday.

Though having better conditions than in Canada, Swedish care workers are not without problems, and when compared with other Swedish workers, they have more physically demanding work, experience higher rates of backpain, of fatigue, of exposure to workplace violence, and of work related illnesses/injuries (Gustafsson & Szebehely 2009b; Bäckman, 2001; Wikman et al., 2010).

Concluding remarks

In conclusion, when considering the model of care work employed in each jurisdiction, we found that Canada follows a highly differentiated task-oriented care work model, whereas Sweden represents an integrated relational care work model. In addition, smaller care units and what appears to be a higher staffing ratio in Swedish long-term care likely improves care workers’ abilities to provide better care, thus improving living conditions for residents and working conditions for care workers.

This research injects the voices of long-term facility care workers into comparative welfare state research. Previous care research has shown the importance of relational caring in meeting the individual situation and changing needs of care recipients. This micro-politics of care negotiation is dependent upon linking practices at the micro level to meso and macro level contexts, which affect work organisation and the conditions of care. Further research must address how the macro structures affect organisation and conditions on the ground.

Thus, there are multiple reasons for policy-oriented welfare researchers to examine closely the everyday life of care workers in residential care, in conjunction with different models of care work organisation; there are important consequences for elderly residents, their families and for the care workers.

In both countries, many care workers expressed their desire to provide good care, but resource constraints, the size of the facilities and the way that work is organized hampered their intentions. This is particularly the case amongst Canadian care workers. If care workers with high ideals can be recruited and are willing to remain working, and if the organization of the services leaves enough time and space for workers to provide good quality care, only then can residential care be a welfare resource for all three parties involved. However, it is unrealistic to expect that care workers will continue to accept the poor working conditions, elevated health risks and the low pay associated with care work as it is currently organised. And it is indefensible from a social justice perspective to expect that the rewards they get from relating to and with the residents will compensate for low pay, exhaustion and health problems caused by scarce resources and un-caring organisational principles. This expectation is particularly important for policy makers to consider, since many of the care workers expressed their desire to increase the relational aspects of their work, but felt that they were unable to do so in practice. The real policy challenge is to make better care possible by improving working conditions.

Appendix

The Scandinavian study was financed by the Swedish Council for Working Life and Social Research and led by Szebehely, Stockholm University; the Canadian study was financed by the Canadian Institutes for Health Research and led by Armstrong, York University, Toronto. This article reports data from Canada and Sweden only; for analyses based on the larger study, see Elstad and Vabø (2008) and Armstrong, Banerjee, Szebehely, Amstrong, Daly & LaFrance (2009b). The Canadian sample and data collection was handled by the York Institute for Social Research (ISR). In Sweden, the survey was sent to a random sample of the unionised workers at their home addresses provided to the researchers by the Municipal workers’ union (Kommunal). The response rate was 67 per cent. In Canada it was not possibe to obtain worker’s direct addresses and the survey was thus distributed to a random sample of unionised residential care facilities in Ontario, Manitoba and Nova Scotia; 81 facilities were contacted and 71 participated (88%). Five major health care unions (Canadian Auto Workers, Canadian Federation of Nurses Unions, Canadian Union of Public Employees, National Union of Public and General Employees, Service Employees International Union) provided contacts and a union representative at each work place was asked to distribute the survey to staff members at the facility. Due to this procedure it is not possible to calculate the response rate. The completed questionnaires were mailed back to the ISR by respondents to ensure anonymity and independence in answers. The Canadian survey findings were verified with nine structured focus group discussions to provide a depth of understanding to the context of the work.