Agenda-setting for Canadian caregivers: using media analysis of the maternity leave benefit to inform the compassionate care benefit

Since World War II, and particularly the second half of the 20^th century, there has been exponential growth in female participation in the labour market, essentially doubling between 1973 and 2002 [1]. In Canada, this long-standing trend has been the primary reason for increasing labour supply [7]. While the participation of women in labour markets peaked between the 1970s and 1980s, the trend continues to hold true to this day [8]. The increase participation in the labour market has been linked to several factors, including changes to family structure, economic pressures, and changes to social structures. Differences in women’s labour market participation across Canadian provinces have been attributed to variance in infertility rates and accessibility to day care, education and job opportunities. Accordingly, the management of dual roles of employee and caregiver - whether caring for dependent children and/or aging family/friends - has become of interest in the research and policy domains.

Women have traditionally been socialized to provide care for dependents, including elderly and younger family members; this gender role continues exist today. In fact, upwards of 85-90% of the care in Canada is provided by unpaid informal caregivers [9], and 80% of which are women [10]. Morris et al. [11] argues that “women family members were expected to supplement home care services without pay and at greater personal expenses in terms of their own health, incomes, benefits, career development and pension accumulation, while men were not under as much pressure…” (p. vi). Although more women are participating in the labour market, research indicates that their likelihood of becoming a caregiver has not diminished but, rather, has grown [12].

Although not exclusively a gendered issue, recent demographic and health service changes have created a new burden for female employees and caregivers. The number of terminally ill Canadians is expected to grow between two and four times from 2010 to 2030 [13], with the “baby boom” cohort nearing retirement and peaking between those years [14]. Many adults are being required to care for their aging parents in the community, often at home, while concurrently managing the responsibility of caring for their own children, a phenomenon that is expected to grow [15]. Known as the “third shift”, the caregiving shift is being negotiated alongside the first shift of paid labour, and a second shift of childrearing [16]. Social and moral attitudes have reinforced society’s expectations that family members will be end-of-life caregivers. The role of informal caregivers at end-of-life has also been heightened in response to financial pressure that has shifted health care from institutions into the community and private homes [13], together with desire to die at home [17]. These trends are reflective of the shifting neoliberalism we are experiencing in both culture and politics, all of which are proving to have a negative effect on the social construction of benefits, gender, and caregiving.

Despite the growing need for comprehensive policies, and the fact that quality end-of-life care has been identified as “the right of every Canadian”, end-of-life care has been described as a “patchwork quilt” [18]. As a result, the act of informal caregiving, and subsequent burden, has been described as an emerging public health issue. Although the rewards of caregiving are many, the costs of caregiving can be extensive; informal caregivers often report negative emotional, physical, social and psychological outcomes as a result of their caregiving duties [19‐22]. Demands on caregiving also often translate into financial pressures; the average out-of-pocket cost of caregiving expenditures is often over $6,000 within the last month of life [23]. These out-of-pocket costs are particularly troublesome due to caregivers reporting decreased workplace productivity as a result of their caregiving roles [24].

Welfare state responses have evolved in order to meet these growing demands on employed workers, and particularly women, in recent years. These programs have been described as “a set of government programs aimed at ensuring citizens’ welfare in the face of the contingencies of life…provid[ing] social insurance against the financial consequences of certain biological risks (illness, incapacity to work, childbirth, childrearing, old age)…”[25]. The MLB and the CCB are two examples of welfare state responses aimed at supporting Canadian laborers’ and, in particular, woman’s multiple roles. The MLB was implemented in 1971 in recognition of the increasing number of women becoming involved in the paid labour force, and the necessity to implement policies to support women and develop equity in the workforce. The MLB applies to the case of children where the CCB applies to the case of all ages, but most commonly, elder adults. In 2004, a similar leave, entitled the Compassionate Care Benefit (CCB), was implemented in recognition of the informal caregiving work being required in light of the growing palliative and end-of-life population. Informal caregivers caring for families or friends at end-of-life can seek paid leave from work through the CCB. Both the MLB (including Parental Leave) and the CCB are legislated nationally through Employment Insurance (EI); applicants need to make contributions to the EI scheme in order to be successful in their application [26]. Contributions to the EI scheme occur through workplace payroll deductions, irrespective of the employees work status as either full-time or part-time, for example. While the financial supports available from both the MLB and the CCB are legislated through national policy, the individual provinces have jurisdiction beyond what is stipulated by the federal government; these are defined by provincial labour laws which, for example, guarantee women the right to take the MLB. National legislation is the sole focus herein. Differences between individual provincial and workplace policies, which may vary widely from the national standards, are not explored. The unique eligibility requirements and program features of both the MLB and CCB are summarized in Table 1.

While the introduction of CCB marked a major step forward for supporting informal caregivers, low uptake rates suggests that the program has been unable to fulfill its mandate [27]. A 2008 EI tracking survey found that 49% of Canadians were not very aware, or not aware at all, of the CCB; this number has only dropped slightly from when the program was initially made available in 2004 [28]. The experience of the CCB contrasts with that of the MLB, which has shown higher uptake rates throughout the course of its evolution [29]. Canada’s MLB policy has received international acclaim as a result of its length and coverage, in particular its extension to fathers in the form of Parental Leave Benefits [30]. Of all those receiving EI in 2010, 29% took MLB or Parental Leave Benefits, while only 2% took CCB [26].

Reasons for the low uptake of CCB have been documented in earlier qualitative research [31]. In this work, the perspectives and expectations of front-line palliative care providers, caregivers and human resource personnel/employers was gathered; data analysis showed five key recommendations for improving the program and broadening its accessibility: (1) implementing a knowledge translation campaign to increase awareness; (2) simplifying and streamlining the application process; (3) eliminating the two-week unpaid waiting period; (4) increasing the length of the CCB, and; (5) increasing the financial remuneration available [31].

Due to their similarities in legislation and purpose, examining the evolution of the comparatively mature MLB may offer insights as to how the CCB may evolve in order to better meet the needs of informal caregivers. Identifying the processes through which changes to the MLB was implemented may suggest comparable mechanisms through which the CCB may develop. In order to do this, a review of the evolution of MLB and CCB policies was undertaken using Kingdon’s theoretical framework to evaluate policy change [6].

The MLB available to Canadians have changed drastically since its inception in 1971. The earliest form of the MLB available spanned 15 weeks and was offered only to women. It operated as a measure to address inequalities in the workforce, where women were being laid off as a result of being pregnant. The most notable changes to the program have resulted in increased flexibility for applicants, with an extension of the program in the form of Parental Leave Benefits to biological fathers, adoptive parents, and recently, extension of the benefits to self-employed parents. The length of leave has been expanded greatly throughout the years, from an initial length of 15 weeks to its current amount of upwards of 50 weeks of leave, if a combination of maternity and Parental Leave Benefits is taken. The application process has also evolved to recognize women who are taking MLB for a second child soon after the first; the number of qualifying hours in that case has been reduced. Table 3 illustrates key events in the evolution of the MLB in Canada.

Analysis of the media coverage of the MLB from 1960, since pre-inception, and through to 2010 was accompanied by categorization of the specific events described in the media coverage according to the three policy channels identified by Kingdon [6]; problems, processes, and political processes. The number of articles accessed through the initial media search (n = 2,698), was much greater than those saved and categorized according to the three streams (n = 163) defined by Kingdon, as summarized in Table 2[6]. In addition, the three policy channels (problems; processes and; political processes) were quantified. It is important to note that often an article may discuss examples of multiple channels through which a policy change was influenced (i.e. it may discuss both a problem, as well as a process scenario, or may be considered both a prominent event as well as crisis). As such, while the quantification and tracking of the types of events can serve as a mechanism for general comparison, the actual numbers are more subjective. Even so, differences in the prevalence of each of the channels were evident. Once again, Table 2 presents the quantitative results of the summative coding of the articles. Certain articles may have been coded in multiple categories, resulting in total numbers for each category not adding up to reflect the total number of articles reviewed.

As Kingdon describes, processes include the gradual accumulation of knowledge and/or perspective among specialists in a given policy area, as well as the generation of policy proposals (106 process-type channels were categorized in the MLB media coverage) [6]. In particular, advocacy on behalf of academics, specialists, and stakeholders generated the most momentum in shaping MLB policy. For example, in 1967 the Women’s Bureau of Canada (Department of Labour) ran a MLB survey, showing that Canada was significantly behind international equivalents and discussed options for implementing a like policy. Similarly, in 1981, Federal Employment Minister Lloyd Axworthy tabled a report suggesting alternate policies that extended leave benefits to adoptive parents, a feature which was implemented in 1984. The generation of policy proposals such as these clearly represents the process route of agenda-setting. The opening of a policy window through a process route was not only undertaken by influential organizations or political figures, but also by academics. For example, in 1987, the national media covered academic research exploring the dual responsibilities faced by female employees, suggesting that an increase in the leave length of the MLB was necessary to support women managing multiple roles.

Problems were slightly less common as motivators for agenda-setting in MLB evolution (75 problem-type channels were categorized in the MLB media coverage). Problems are defined by Kingdon as 'crises', 'prominent events', and 'changes in indicators’ that generate incentive to change policy [6]. For example, the Schachter vs. Canada case influenced the implementation of Parental Leave Benefits for biological fathers in 1990. The Bliss vs. Canada case in 1994 also played a hand in mounting public pressure to abolish the 'Magic Ten’ rule. Previously women were required to prove they had worked ten weeks prior to conception in order to be eligible for the MLB, on top of regular eligibility criteria. 'Changes in indicators’ were also explored in the media coverage, and were also often categorized as processes, as they also usually represented gradual accumulation of knowledge by specialists and academics. For example, a study in 1993 reported that very few men were applying for maternity leave, despite the availability of Parental Leave Benefits. Changes were eventually made in 2000 that increased the length of leave available, with the hope that more leave time would motivate parents to split up caregiving duties. In addition, the two-week waiting period was amended, requiring only one parent to meet this stipulation, rather than two.

Although there were significant differences in the type of route most often used to generate the opening of a policy window, additional trends were evident according to a temporal analysis of the evolution of the MLB. Processes were the most common type of route for agenda-setting prior to and during the earlier years of MLB legislation. For example, the first media discussions accessed about the MLB discussed Canada’s lack of benefits, and available research on international comparisons which could be used to generate a policy proposal. Problems became more common after the implementation of the MLB, and were influences by the involvement of female advocacy groups interested in minimizing inequality. Political processes, as by their definition, occur around governmental changes, an important contextual avenue which was not explicitly explored in this research but likely would demonstrate a temporal patterning in relation to national political changes.

The Compassionate Care Benefit (CCB), compared to the MLB, has had a relatively short evolutionary period. Since the initial legislation in 2004, the only policy change to the CCB occurred in 2006, when eligibility criteria changed to allow a more flexible definition of family member. Prior to this change, applicants were required to be a blood or law relation to the care recipient. New eligibility criteria allows the caregiver to be considered 'like family’ in response to the recognition that the care recipient should have flexibility in deciding who their more compatible caregiver is, and that the caregiver most suited to the role may be a neighbour or close friend, rather than a family member per se. In addition, the CCB has also recently been extended to self-employed workers, much like the MLB, to be available to applicants in 2011. Policy changes to the CCB have been summarized in Table 4.

At several points during its evolution, the MLB has expanded its coverage length to allow longer lengths of leave in response to social changes. Adoptive mothers were granted 15 weeks of leave originally, in understanding that the purpose of maternity leave is not just for the physical recuperation of the mother, but also for the newborn or child’s adjustment. In 1989 an additional 10 weeks of leave on top of the 15 weeks of leave were granted to allow biological fathers time to bond with the child and partake in caregiving. The additional 10 weeks of leave were made available to either natural mothers or fathers. Later, changes in 2000 extended the length of leave available again, in part due to the recognition that many men were not taking Parental Leave Benefits. The new regulations increased the length of Parental Leave Benefits to 35 weeks (from 10 weeks), making the total length of the MLB and Parental Leave Benefits together 50 weeks. Additionally, the waiting period was eliminated for one parent if both parents were applying.

In 1974 the MLB was made more flexible by allowing women to collect 15 weeks maternity leave at any point during a 26 week period, including eight weeks prior to, and up to 17 weeks after, birth. In contrast, the CCB only allows for six weeks of leave, and only prior to death. Regardless of how many weeks of leave the applicant has actually used, the leave ends following the death of the person being cared for; this is extremely limiting, given the period of bereavement and grieving that may follow end-of-life. As Hallenbeck states, “care does not end with death of the patient but continues through death pronouncement, family notification of the death, discussion of autopsy, and immediate bereavement support.” [37], p. 2265. Bereavement and grief has been reported to take between six months [38] to 24 months [39], although quantification is difficult [39]. Negative components of grief, including anger and depression, often do not peak until five or six months after death [39]. Indeed, attention to the care needs of the informal caregiver after the patient’s death should be built into the CCB, given that the caregiver is part of the unit of care [40, 41].

The Compassionate Care Benefit (CCB) has also recognized the need to allow for multiple caregivers, both by allowing multiple people to apply for and split the eight week leave and by changing its definition of eligible family members in 2006 to include those considered 'like family’. However, unlike the MLB, the CCB has not been extended; only eight weeks are available to be split between caregivers. Considering difficulties in prognostication at end-of-life, extending the length of the CCB needs to be considered. Prognostication at end-of-life is complex, as it includes not only estimating time until death, but also informs the disease trajectory, all of which impacts: care receiver symptoms; caregivers’ life worlds; financial costs, and; changing therapies [42]. Both the prediction of the disease trajectory, as well as prediction of time-of-death, are difficult, and have been reported as potential barriers to timely access to hospice palliative care [42, 43]. Underutilization of hospice palliative care services has been associated with poor prognostication and palliative care referral [44, 45]; this underutilization may impact those eligible for the CCB, given the limited time frame in which they are considered eligible, and given their self-reported difficulty in deciding when to apply in order to maximize their time spent with the patient [31]. By lengthening the amount of CCB leave available, these issues specific to prognostication and concerns about when to take the leave may be negated.

The Maternity Leave Benefit (MLB) has incorporated, during the course of evolution, another key feature which introduces more flexibility in its eligibility criteria; this would also be appropriate for the CCB. In 1979, and again in 2000, eligibility criteria were eased for women who had been on Maternity Leave in the previous year and who were applying for Maternity Leave again, allowing them to have worked fewer hours in order to qualify. While some research indicates that women do not increase their participation in the labour market in the year preceding birth, in order to be eligible for the MLB [1], other research indicates that availability of the Canadian MLB policy may influence a couples’ decision to pursue parenthood [1, 46, 47]. Similarly, fertility planning around maternity leave policies has shown strong correlation in international research [48, 49].

Informal caregivers at end-of-life may also be required to provide palliative care repeatedly in one year. We know, for example, that hospitalization or death may cause negative health outcomes in another who is closely connected [50]. Bereavement has been associated with increased risk of mortality, particularly in the first six months after a loved-one’s death and most common among spousal deaths [51]. Therefore, it is likely that caregivers may be called upon to provide palliative care more than once a year. Creating more flexible eligibility criteria for end-of-life caregivers, similar to the MLB, would provide more comprehensive coverage to Canadians.

Triangulating the results of tracking the evolution of the MLB, together with the temporal patterning of how policy windows are opened, generates some important information that can inform potential changes to the CCB. Considering that processes were the most common measures identified in this study, through which policy windows were opened and positive changes to MLB legislation was achieved, it follows that perhaps resources and energy should be directed via stakeholders, specialists and academics at processes to facilitate the opening up of policy windows.

The Board finds no that there is no compassion…the failure of the Commission and the Minister to act swiftly in these matters of Compassionate Care amendments has only served to exacerbate the suffering endured by families as they care for a dying family member…[52].

Having gathered national media attention and, together with the process-type stakeholder advocacy, this problem-type event prompted the later opening of a policy window to change the eligibility criteria to be more inclusive. Eligibility criteria now allow applicants from a variety of relationships to the care recipient to apply, including those considered 'like family’.

However, the CCB is nine years past implementation, yet very few problems have reached the mainstream media or generated momentum for policy change. One hypothesis for this is the general disinclination to discuss end-of-life care; Canada remains largely death-denying, with a greater focus on curative medicine than on comprehensive care [53]. Greater efforts to bring problems with hospice palliative care services, including the CCB, into the mainstream will force national attention on these issues, and perhaps generate more political will among government to add hospice palliative care and the CCB to their political agendas. These efforts may gain momentum given that a growing number of Canadians face the realities of providing end-of-life care, reflected in the aging of the 'baby boomer’ generation. More focus on the process and problem realms of agenda setting may, in turn, generate more impetus for taking up the cause of palliative care by politicians, thus setting up opportunities for agenda-setting. Further, there exist salient differences between the MLB and CCB. For example, the MLB has two long-term economic goals, those being: (1) the retention of women in the labour force, and; (2) producing healthier and well-adjusted babies; without the MLB, Canadians may not have babies. In contrast, the CCB shares only one of these goals: that being retaining women in the labour force.

While evidence about the influence of stakeholders and champions in moving policy forward has been explored in the literature, it is also noted that the process of opening policy windows is a complex and iterative process. As Kingdon notes, it is the coupling of one or more of the three streams that opens policy windows [6]. Interestingly, the slow temporal nature of policy change for both the MLB and CCB is perhaps indicative of the Canada’s policy system. Canada has been labeled as having an incremental policy system, where policy changes occur without an overarching framework, and where responsibility for changes in the health care system falls to both the provincial and federal level governments [54].

By implementing some of the policy recommendations identified, the CCB will better support the needs and expectations of informal caregivers and allow women, in particular, a better chance for success in balancing multiple roles and thereby enjoy better health and wellbeing [55]. There is no question that this issue will likely become a greater concern to women in the coming years, as the number of people requiring care at end-of-life continues to grow [18].

A number of limitations have been identified in this research, particularly with the application of Kingdon’s framework, as well as the methodological approach. As Kingdon discusses, tracing the origin of policy ideas is an iterative, never-ending process [6]. Thus, the setting of an arbitrary time point at which to start the literature search, while necessary, may have limited earlier events important to the evolution of the policies investigated. The use of The Globe and Mail newspaper as the sole source for media coverage of the MLB may limit the scope of the media coverage reviewed. For example, while The Globe and Mail typically represented national events, articles with provincial foci were often about Quebec and Ontario, leading to concern that events in other provinces may not be adequately reviewed. As the focus of this paper is national, this is of minimal concern, however there is the potential for events in other provinces, which may have influenced agenda-setting, to have been missed. Related to this is the exclusion of associated provincial policies. Further, the exclusion of a wide range of other sources, such as, for example: primary documents made by proponents of legislative change; other media sources, or; actual statements by politicians, other policy makers, or members of civil society not filtered through the media. In addition, as the search terms used for the whole time period were “maternity leave” and “maternity benefit” the search strategy did not take into account articles about employment insurance generally that may have affected the evolution of maternity leave. For example, changes to employment insurance according to economic pressures and political changes may have influenced changes in eligibility and coverage available to MLB applicants. Similarly, the search terms did not include “Parental Leave Benefits” or “paternity leave”, which may have reduced the number of relevant hits if the article did not use the term “maternity” as well. Finally, there are some limitations in choosing Kingdon’s framework; policy windows are only one part of the discourse on policy development and other frameworks and theories were not considered, although it is recognized that they could provide different perspectives [56]. Likewise, Kingdon’s framework does not account for the magnitude an event will have or distinguish significance between two or more different events [6]. For example, we cannot know the impact of one academic report versus another. Further, it is likely that, given the retrospective analysis, contextual factors such as the political landscape and the socio-economic climate of the time in which policy windows were opened, have not been fully accounted for in this analysis.

Both the MLB and CCB have, in fact, been the focus of attention recently, as governments examine the opportunity to separate both policies from the Employment Insurance program, and whether to devolve the responsibility to provincial levels [1]. Much of the interest of in doing so has developed in light of the separation of such policies from the federal government in the province of Quebec. Quebec has run its own family policy system separate from the national system since 2006, and has reported huge success in increasing support systems through the evolution of the provincial policies [1]. Indeed, while not examined, much of the media attention initially picked up through MLB and CCB reviews covered the separation of Quebec benefits. Similarly, workplace policies, and lobbying for enhanced workplace benefits was a continuous topic of interest in the media. For example, media coverage of the MLB in the 1980s focused largely on the Canadian postal workers’ fight for the MLB. It is likely their experiences may have shaped ensuing national policy. Closer examination of these events, and their potential for opening up policy windows, would be valuable. Further, linking the developmental changes experienced by both the MLB and CCB with actual changes in uptake would be most valuable. If this were accomplished, responses to programmatic changes could be known and used to inform future program revisions.